How Microgravity Affects the Biology of Living Systems

BioMed Research International

How Microgravity Affects the Biology of Living Systems

Guest Editors Mariano Bizzarri Monica Monici and Jack J W A van Loon

How Microgravity Affects the Biology ofLiving Systems



Guest Editors Mariano Bizzarri Monica Moniciand Jack J W A van Loon

Copyright copy 2015 Hindawi Publishing Corporation All rights reserved

This is a special issue published in ldquoBioMed Research Internationalrdquo All articles are open access articles distributed under the CreativeCommons Attribution License which permits unrestricted use distribution and reproduction in any medium provided the originalwork is properly cited

Contents

HowMicrogravity Affects the Biology of Living Systems Mariano Bizzarri Monica Moniciand Jack J W A van LoonVolume 2015 Article ID 863075 4 pages

Simulated Microgravity Critical Review on the Use of Random Positioning Machines for MammalianCell Culture Simon L Wuest Stephane Richard Sascha Kopp Daniela Grimm and Marcel EgliVolume 2015 Article ID 971474 8 pages

Regulation of ICAM-1 in Cells of the MonocyteMacrophage System in Microgravity Katrin PaulsenSvantje Tauber Claudia Dumrese Gesine Bradacs Dana M Simmet Nadine Golz Swantje HauschildChristiane Raig Stephanie Engeli Annett Gutewort Eva Hurlimann Josefine Biskup Felix UnverdorbenGabriela Rieder Daniel Hofmanner Lisa Mutschler Sonja Krammer Isabell Buttron Claudia PhilpotAndreas Huge Hartwin Lier Ines Barz Frank Engelmann Liliana E Layer Cora S Thiel and Oliver UllrichVolume 2015 Article ID 538786 18 pages

Genes Required for Survival in Microgravity Revealed by Genome-Wide Yeast Deletion CollectionsCultured during Spaceflight Corey Nislow Anna Y Lee Patricia L Allen Guri Giaever Andrew SmithMarinella Gebbia Louis S Stodieck Jeffrey S Hammond Holly H Birdsall and Timothy G HammondVolume 2015 Article ID 976458 10 pages

RhoGTPases as Key Players in Mammalian Cell Adaptation to Microgravity Fiona LouisChristophe Deroanne Betty Nusgens Laurence Vico and Alain GuignandonVolume 2015 Article ID 747693 17 pages

A Tissue Retrieval and Postharvest Processing Regimen for Rodent Reproductive Tissues Compatiblewith Long-Term Storage on the International Space Station and Postflight Biospecimen SharingProgram Vijayalaxmi Gupta Lesya Holets-Bondar Katherine F Roby George Enders and Joseph S TashVolume 2015 Article ID 475935 12 pages

Large Artery Remodeling and Dynamics following Simulated Microgravity by Prolonged Head-DownTilt Bed Rest in Humans Carlo Palombo Carmela Morizzo Martino Baluci Daniela Lucini Stefano RicciGianni Biolo Piero Tortoli and Michaela KozakovaVolume 2015 Article ID 342565 7 pages

Space Flight Effects on Antioxidant Molecules in Dry TardigradesThe TARDIKISS ExperimentAngela Maria Rizzo Tiziana Altiero Paola Antonia Corsetto Gigliola Montorfano Roberto Guidettiand Lorena RebecchiVolume 2015 Article ID 167642 7 pages

Identification of Reference Genes in HumanMyelomonocytic Cells for Gene Expression Studies inAltered Gravity Cora S Thiel Swantje Hauschild Svantje Tauber Katrin Paulsen Christiane RaigArnold Raem Josefine Biskup Annett Gutewort Eva Hurlimann Felix Unverdorben Isabell ButtronBeatrice Lauber Claudia Philpot Hartwin Lier Frank Engelmann Liliana E Layer and Oliver UllrichVolume 2015 Article ID 363575 20 pages

AWhole-GenomeMicroarray Study of Arabidopsis thaliana Semisolid Callus Cultures Exposed toMicrogravity and Nonmicrogravity Related Spaceflight Conditions for 5 Days on Board of Shenzhou 8Svenja Fengler Ina Spirer Maren Neef Margret Ecke Kay Nieselt and Rudiger HamppVolume 2015 Article ID 547495 15 pages

RCCS Bioreactor-Based Modelled Microgravity Induces Significant Changes on In Vitro 3D NeuroglialCell Cultures Caterina Morabito Nathalie Steimberg Giovanna Mazzoleni Simone GuarnieriGiorgio Fano-Illic and Maria A MariggioVolume 2015 Article ID 754283 14 pages

The Impact of Microgravity and Hypergravity on Endothelial Cells Jeanette A M MaierFrancesca Cialdai Monica Monici and Lucia MorbidelliVolume 2015 Article ID 434803 13 pages

A Functional Interplay between 5-Lipoxygenase and 120583-Calpain Affects Survival and Cytokine Profile ofHuman Jurkat T Lymphocyte Exposed to Simulated Microgravity Valeria Gasperi Cinzia RapinoNatalia Battista Monica Bari Nicolina Mastrangelo Silvia Angeletti Enrico Dainese and MauroMaccarroneVolume 2014 Article ID 782390 10 pages

HowMicrogravity Changes Galectin-3 inThyroid Follicles Elisabetta Albi Francesco CurcioAndrea Lazzarini Alessandro Floridi Samuela Cataldi Remo Lazzarini Elisabetta Loreti Ivana Ferriand Francesco Saverio Ambesi-ImpiombatoVolume 2014 Article ID 652863 5 pages

The Influence of Simulated Microgravity on Purinergic Signaling Is Different between IndividualCulture and Endothelial and Smooth Muscle Cell Coculture Yu Zhang Patrick Lau Andreas PanskyMatthias Kassack Ruth Hemmersbach and Edda TobiaschVolume 2014 Article ID 413708 11 pages

Human Locomotion under Reduced Gravity Conditions Biomechanical and NeurophysiologicalConsiderations Francesca Sylos-Labini Francesco Lacquaniti and Yuri P IvanenkoVolume 2014 Article ID 547242 12 pages

Conditioned Media fromMicrovascular Endothelial Cells Cultured in Simulated Microgravity InhibitOsteoblast Activity Alessandra Cazzaniga Sara Castiglioni and Jeanette A M MaierVolume 2014 Article ID 857934 9 pages

Phenotypic Switch Induced by Simulated Microgravity on MDA-MB-231 Breast Cancer CellsMaria Grazia Masiello Alessandra Cucina Sara Proietti Alessandro Palombo Pierpaolo ColucciaFabrizio DrsquoAnselmi Simona Dinicola Alessia Pasqualato Veronica Morini and Mariano BizzarriVolume 2014 Article ID 652434 12 pages

Oxidative Stress and NO Signalling in the Root Apex as an Early Response to Changes in GravityConditions Sergio Mugnai Camilla Pandolfi Elisa Masi Elisa Azzarello Emanuela MonettiDiego Comparini Boris Voigt Dieter Volkmann and Stefano MancusoVolume 2014 Article ID 834134 10 pages

Cytoskeleton Modifications and Autophagy Induction in TCam-2 Seminoma Cells Exposed toSimulated Microgravity Francesca Ferranti Maria Caruso Marcella Cammarota Maria Grazia MasielloKatia Corano Scheri Cinzia Fabrizi Lorenzo Fumagalli Chiara Schiraldi Alessandra CucinaAngela Catizone and Giulia RicciVolume 2014 Article ID 904396 14 pages

Gravity Affects the Closure of the Traps inDionaea muscipula Camilla Pandolfi Elisa Masi Boris VoigtSergio Mugnai Dieter Volkmann and Stefano MancusoVolume 2014 Article ID 964203 5 pages

The Impact of Simulated and Real Microgravity on Bone Cells and Mesenchymal Stem CellsClaudia Ulbrich Markus Wehland Jessica Pietsch Ganna Aleshcheva Petra Wise Jack van LoonNils Magnusson Manfred Infanger Jirka Grosse Christoph Eilles Alamelu Sundaresan and Daniela GrimmVolume 2014 Article ID 928507 15 pages

Multisensory Integration and Internal Models for Sensing Gravity Effects in PrimatesFrancesco Lacquaniti Gianfranco Bosco Silvio Gravano Iole Indovina Barbara La Scaleia Vincenzo Maffeiand Myrka ZagoVolume 2014 Article ID 615854 10 pages

Integration Analysis of MicroRNA and mRNA Expression Profiles in Human Peripheral BloodLymphocytes Cultured in Modeled Microgravity C Girardi C De Pitta S Casara E CaluraC Romualdi L Celotti and M MognatoVolume 2014 Article ID 296747 16 pages

EditorialHow Microgravity Affects the Biology of Living Systems

Mariano Bizzarri1 Monica Monici2 and Jack J W A van Loon3

1Department of Experimental Medicine Systems Biology Group University La Sapienza 00161 Rome Italy2ASAcampus Joint Laboratory ASA Research Division Department of Experimental and Clinical Biomedical Sciences University ofFlorence 50121 Florence Italy3Department of Oral and Maxillofacial SurgeryOral Pathology VU-University Medical Center 1081 HZ Amsterdam Netherlands

Correspondence should be addressed to Mariano Bizzarri marianobizzarriuniroma1it

Received 20 November 2014 Accepted 20 November 2014

Copyright copy 2015 Mariano Bizzarri et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Gravity has constantly influenced both physical and biologi-cal phenomena throughout Earthrsquos history The gravitationalfield has played a major role in shaping evolution whenlife moved from water to land even if for a while it hasbeen generally deemed to influence natural selection onlyby limiting the range of acceptable body sizes accordingto Galileirsquos principle Indeed to counteract gravity livingorganisms would need to develop systems to provide cellmembrane rigidity fluid flow regulation and appropriatestructural support for locomotion However gravity mayinfluence in a more deep and subtle fashion the way the cellsbehave and build themselves

The first empirical experiments mostly done by Russianscientists in the 60s were unable to unravel major changesafter exposure tomicrogravity thus nurturing the false notionfor which near weightlessness does not get any appreciableeffects on living organisms [1 2] However as fundamentalinvestigations began in the space environment it became evi-dent that biological properties change as gravitational force isdiminished underscoring the relationship between physicalforce and biological function Cells exposed to microgravitycan indeed be profoundly affected by the physical changesthat occur in this unique environment which include theloss of gravity-dependent convection negligible hydrody-namic shear and lack of sedimentation [3ndash5] Cell-substrateadhesions as well as cell-to-cell junctions are consequentlyprofoundly affected at Earthrsquos gravity impairing multicellularaggregates and tissue formation while such structures can bemore easily sustained for days or months in microgravity [6]These modifications eventually lead to a significant change

in the way the cell mechanosensor apparatus responds to awide array of environmental and internal biophysical stresses[7] As a consequence enzymatic genetic and epigeneticpathways change in concert leading to several modificationsin cells and tissues shape function and behavior [8 9]Fruitful insights about the involvement of several molecularpathways during microgravity exposure are reported in thisissue by the studies of V Gasperi et al (unravelling newpathways involved in immune function impairment duringspaceflight) and E Albi et al (overexpression of Galectin-3in thyroid follicles due to microgravity-induced membraneremodelling) Namely a sophisticated analysis of mRNAexpression in human blood lymphocytes carried out byC Girardi et al confirmed that microgravity induces ageneralized inhibition of proliferation and a contemporaryincrease in apoptosis rate

Indeedmdashand unfortunatelymdashnear weightlessness dra-matically impairs biological functions and thereby contrarytowhatwas previously thought [2] cells cannot be consideredldquoblindrdquo with respect to gravity

The microgravity space environment may result in achallenging threat for living beings as aptly documented bythe paper from C Nislow et al showing that spaceflighthas subtle but significant effects on core cellular processesincluding growth control via RNA and ribosomal biogen-esis metabolism modification and decay pathways It isnoteworthy that despite the fact that some reference-genesremain stable during microgravity exposure several othersinvestigated in the study of C S Thiel et al change quitedramatically thus reinforcing the concept that exposure to

Hindawi Publishing CorporationBioMed Research InternationalVolume 2015 Article ID 863075 4 pageshttpdxdoiorg1011552015863075

2 BioMed Research International

near weightlessness may have a profound impact on livingprocesses Namely it seems that genes involved in ROSdetoxification are especially impaired in such conditionas reported by the paper from S Fengler et al thereforesuggesting how relevant could be the role sustained by theredox status in counteracting at least some downstreamconsequences of microgravity Yet as reported in the articleof S Mugnai et al both nitric oxide and ROS are likely toplay a previously unrecognized role as messengers during thegravitropic response inmany root tips Relevance of oxidativeprocesses during microgravity exposure was also reported bythe study of A M Rizzo et al in which a significant increasein oxidative stress has been observed in tardigrades exposedto spaceflight

Cells may ldquosenserdquo changes in the microgravitational fieldthrough (a) an indirect mechanism (mainly based on themodification of physical properties of their microenviron-ment) (b) the development of specialized structures forthe mechanical perception and transduction of gravitationalforces (like the cytoskeleton) and (c) changes in the dynamicsof enzymes kinetics or protein network self-assembly It isworth noting that the latter two processes are dramaticallyaffected by nonequilibrium dynamics Nonlinear dynamicalprocesses far from equilibrium involve an appropriate com-bination of reaction and diffusion and the pattern arisingfrom those interactions is tightly influenced by evenminimalchanges in reactant concentrations or modification in thestrength of the morphogenetic field [10] Processes of thiskind are called Turing or dissipative structures given that aconsumption of energy is required to drive and maintain thesystem far from equilibrium That prerequisite is needed inorder to allow the system to promptly change its configura-tion according to the systemrsquos needs In turn the dissipativeenergy provides the thermodynamic driving force for theself-organization processes Some experimental evidence hasalready been provided that change of the gravitational fieldmay significantly affect some nonlinear reactions occurringwithin cells and tissues [11 12] Herein a further confirmationis provided by the article of M G Masiello et al in whichthe near weightlessness condition is shown to drive thesystems towards different attractor states thus enabling cellsto acquire new and unexpected phenotypes in the courseof a true phase transition [13] According to such resultsgravity seems to be an ldquoinescapablerdquo constraint that obligesliving beings to adopt only a few configurations amongmanyothers By ldquoremovingrdquo the gravitational field living structureswill be free to recover more degrees of freedom thusacquiring new phenotypes and new functionspropertiesThat statement raises several crucial questions Some of theseentail fundamentals of theoretical biology as they questionthe gene-centered paradigm according to which biologicalbehavior can be explained by solely genetic mechanisms [14]

What are the mechanism(s) through which microgravitymay so profoundly modify cell function and structureSeveral studies included in this issue deal with that topiccalling into the question the pivotal role sustained by thecytoskeleton in mediating several microgravity-based effects

A common outcome in nearly all cell types exposed tomicrogravity is indeed the alteration of cytoskeletal elements

actin microfilaments and microtubules [15 16] Disorga-nization of basic cellular architecture can affect activitiesranging from cell signalling and migration to cell cycling andapoptosis In this issue K Paulsen and colleagues investigatedhow surface expression of ICAM-1 protein and expressionof ICAM-1 mRNA in cells of the monocytemacrophagesystem change inmicrogravity Given that ICAMproteins areessential for cell-to-cell adhesion as well as for cytoskeletonproper functioning such results outline the involvement ofthe cytoskeleton system in mediating at least some effectsdue to microgravity That statement is further reinforced bythe paper from F Louis et al in which dramatic decreasein RhoGTPases activity has been documented RhoGTPasesrepresent a unique hub for integration of biochemical andmechanical signals As such they are probably very rapidlyinvolved in a cellrsquos adaptation to microgravity-related con-ditions Additionally RhoGTPases activity is tightly andmechanistically bound to alterations of the cytoskeletonadhesion and fibrillogenesis as well as to an enhancementof ROS delivery As a result RhoGTPases may be consideredtrue mechanosensitive switches responsible for cytoskeletaldynamics and cells commitment Relevant modification ofthe cytoskeleton architecture and microtubule organizationin testicular cells has been also reported in the study byF Ferranti et al where a significant correlation betweencytoskeleton abnormalities induced by simulated micrograv-ity and enhanced autophagy was recorded Yet cytoskeletonchanges affect different cell types including endothelial cellsIn the paper of J Maier et al it is shown that endothelial cellsare highly sensitive to gravitational stress as microgravityleads to changes in the production and expression of vasoac-tive and inflammatory mediators and adhesion moleculeswhich mainly results from changes in the remodelling of thecytoskeleton and the distribution of caveolae In addition bykeeping in mind that the cytoskeleton dynamics is a funda-mental player in cell proliferation and migration it is notsurprising that microgravity significantly affects the flytrapclosure a process involving not only the actin dynamics butalso the ion channels and aquaporin activities as evidencedin the article from C Pandolfi et al

Cytoskeleton changes have also profound consequenceson both cell shape and tissue modelling Simulated nearweightlessness in human volunteers is associated with asignificant change in arterial geometry flow stiffness andshear rate as documented by C Palombo et al Microgravityis acting on endothelial cells also through modulation of P2-receptor and the release of several cytokines as reportedby the study from Y Zhang et al Given that P2-receptorartificial ligands are applied as drugs it is reasonable toassume that they might be promising candidates againstthe cardiovascular deconditioning the astronauts experienceduring spaceflight

Overall the alterations occurring in microgravity haveundoubtedly significant backwashes on the physiologicalhomeostasis of the whole organism Such aspect is high-lighted by two papers from the group of F Lacquaniti et aldealingwith the effects of nearweightlessness on nervous sys-tem function Gravity is indeed crucial for spatial perceptionpostural equilibrium and movement generation The brain

BioMed Research International 3

may deal with the gravitational field by integrating a widearray of different signals thus enabling the system to triggerthe most appropriate response F Lacquaniti et al providecompelling evidence that this ability depends on the fact thatgravity effects are stored in brain regions which integratevisual vestibular and neck proprioceptive signals where thenervous system combines this information with an internalmodel of gravity effects The second study evidenced inturn the beneficial effect of the neurophysiologic adaptationto near weightlessness and how knowledge acquired onthis field may even enhance the development of innovativetechnologies for gait rehabilitation

Research on microgravity and hypergravity effectivelyadvances our knowledge on physiology and biochemistrythus providing valuable data and models for the understand-ing for some important human diseases Moreover space-based research has played and presumably will continuouslyplay an important role in reformulating the theoreticalframework in biology and physiology and may serve as anovel paradigm for innovation Namely microgravity-relatedresearch fostered the development of new tools-like forculturing cells in three dimensions It is now well understoodthat 3D growth environments that facilitate unrestricted cell-cell interactions are mandatory for defining the biology ofcancer cells and tissues including tumour formation tumourmicroenvironment and tumour progression [17 18] Indeedthree-dimensional culture in real and simulatedmicrogravityallows a more precise appreciation of the role the biophysicalconstraints play in shaping cell phenotypes and functions Inturn such devices may help in improving tissue-engineeringtechniques Experimental models of cellstissues culturesin both simulated and real microgravity need howeverto be further improved in order to obtain more reliableand reproducible data and to minimize the impact of con-founding factors Such studies may indeed provide valu-able information about modulations in signal transductioncell adhesion or extracellular matrix induced by alteredgravity conditions These systems also facilitate the analysisof the impact of growth factors hormones or drugs onthese tissue-like constructs in order to better address issueslike pharmacokinetics and pharmacodynamics Paradigmaticexamples of such studies are reported in this issue by thearticles of several groups (C Ulbrich et al C Morabitoet al V Gupta et al) some of which (S L Wuest et al)critically reviewed the reliability of available technical tools(like the Random Positioning Machine) These facilities mayalso allow investigating developmental and organogenesisprocesses

The motivation for this focussed issue of the BiomedResearch International Journal is to take stock of the state ofresearch and identify possible areas for future developmentThere is an urgent need for this as the last comprehensivecollection of studies devoted to space biomedicine researchdates back to the 90s [19]

As editors we have collected an eclectic mix of arti-cles provided by research groups fully involved in spacebiomedicine research and actively participating in studiescarried out both on the International Space Station andon the ground by means of different techniques enabling

performing conditions of simulated near weightlessness andincreased gravityThis is not a ldquoone view fits allrdquo approach It israther one to ldquolet a hundred flowers bloomrdquo Yet they providea fruitful overview on what is going to come from spacebiomedicine research Overall studies reported in the issuedemonstrated how relevant physical cues may be in shapingbiological phenotypes and function influencing so in depthmolecular and genetic pathways It is regrettable to noticethat such influences have been for so long overlooked bythe scientificmainstream [20 21] Furthermore microgravitystudies forced us to develop new technological solutions andmore appropriate experimental models Thereby knowledgegathered in space research has offered an invaluable supportin understanding both human physiology and pathologyfostering technological innovation and the development ofpriceless medical and experimental devices

This is why it has been argued that the ultimate reason forhuman space exploration is precisely to enable us to discoverourselves Undoubtedly the microgravity and space relatedresearch present an unlimited horizon for investigation anddiscovery Controlled studies conducted in microgravity canadvance our knowledge providing amazing and unforeseeninsights into the biological mechanism underlying physiol-ogy as well as many relevant diseases like cancer [22]

Mariano BizzarriMonica Monici

Jack J W A van Loon

References

[1] P O Montgomery Jr J E Cook R C Reynolds et al ldquoTheresponse of single human cells to zero gravityrdquo In Vitro vol 14no 2 pp 165ndash173 1978

[2] M G Tairbekov G P Parfyonov E Y Shepelev and F VSushkov ldquoExperimental and theoretical analysis of the influenceof gravity at the cellular level a reviewrdquo Advances in SpaceResearch vol 3 no 9 pp 153ndash158 1983

[3] J J W van Loon ldquoThe gravity environment in Space exper-imentsrdquo in Biology in Space and Life on Earth Effects ofSpaceflight on Biological Systems E Brinckmann Ed pp 17ndash32Wiley-VCH 2007

[4] T G Hammond and J M Hammond ldquoOptimized suspen-sion culture the rotating-wall vesselrdquo American Journal ofPhysiologymdashRenal Physiology vol 281 no 1 pp F12ndashF25 2001

[5] P Todd ldquoGravity-dependent phenomena at the scale of thesingle cellrdquo ASGSB Bulletin vol 2 pp 95ndash113 1989

[6] L E Freed R Langer I Martin N R Pellis and G Vunjak-Novakovic ldquoTissue engineering of cartilage in spacerdquo Proceed-ings of the National Academy of Sciences of the United States ofAmerica vol 94 no 25 pp 13885ndash13890 1997

[7] J Klein-Nulend R G Bacabac J P Veldhuijzen and J J WA Van Loon ldquoMicrogravity and bone cell mechanosensitivityrdquoAdvances in Space Research vol 32 no 8 pp 1551ndash1559 2003

[8] S J PardoM J PatelMC Sykes et al ldquoSimulatedmicrogravityusing the Random Positioning Machine inhibits differentiationand alters gene expression profiles of 2T3 preosteoblastsrdquoAmerican Journal of PhysiologymdashCell Physiology vol 288 no6 pp C1211ndashC1221 2005


[9] M Monici F Fusi M Paglierani et al ldquoModeled gravitationalunloading triggers differentiation and apoptosis in preosteo-clastic cellsrdquo Journal of Cellular Biochemistry vol 98 no 1 pp65ndash80 2006

[10] G Nicolis and I Prigogine ldquoIntroductionrdquo in Self-Organizationin Nonequilibrium Systems FromDissipative Structures to OrderThrough Fluctuations John Wiley amp Sons New York NY USA1977

[11] P J Stiles and D F Fletcher ldquoThe effect of gravity on the rateof a simple liquid-state reaction in a small unstirred cylindricalvesselrdquo Physical Chemistry Chemical Physics vol 3 no 9 pp1617ndash1621 2001

[12] C Papaseit N Pochon and J Tabony ldquoMicrotubule self-organization is gravity-dependentrdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 97 no15 pp 8364ndash8368 2000

[13] M Bizzarri and A Giuliani ldquoRepreseenting cancer cell trajec-tories in a phase-space diagram switching cellular states bybiological phase transitionsrdquo in Applied Statistics for NetworkBiology Methods in Systems Biology M Dehmer F Emmert-Streib A Graber and A Salvador Eds pp 377ndash403 WileyNew York NY USA 2011

[14] M Bizzarri A Cucina A Palombo and M Grazia MasielloldquoGravity sensing by cellsMechanisms and theoretical groundsrdquoRendiconti Lincei vol 25 no 1 pp 29ndash38 2014

[15] M L Lewis ldquoThe cytoskeleton in spaceflown cells an overviewrdquoGravitational and Space Biology Bulletin vol 17 pp 1ndash11 2004

[16] D Vorselen W H Roos F C MacKintosh G J L Wuite andJ J W A van Loon ldquoThe role of the cytoskeleton in sensingchanges in gravity by nonspecialized cellsrdquo FASEB Journal vol28 no 2 pp 536ndash547 2014

[17] D Grimm M Wehland J Pietsch et al ldquoGrowing tissuesin real and simulated microgravity new methods for tissueengineeringrdquo Tissue Engineering Part B Reviews vol 20 no 6pp 555ndash566 2014

[18] G R Souza J R Molina R M Raphael et al ldquoThree-dimensional tissue culture based on magnetic cell levitationrdquoNature Nanotechnology vol 5 no 4 pp 291ndash296 2010

[19] D Schmitt ldquoWorkshop purpose and structurerdquo The FASEBJournal vol 13 supplement S1 no 9001 1999

[20] M Bizzarri A Pasqualato A Cucina and V Pasta ldquoPhysicalforces and non linear dynamics mould fractal cell shapeQuantitative morphological parameters and cell phenotyperdquoHistology and Histopathology vol 28 no 2 pp 155ndash174 2013

[21] M Monici and J van Loon Cell Mechanochemistry Bio-logical Systems and Factors Inducing Mechanical Stress Suchas Light Pressure and Gravity Trivandrum Research Sign-postTransword Research Network 2010

[22] J L Becker and G R Souza ldquoUsing space-based investigationsto informcancer research onEarthrdquoNature ReviewsCancer vol13 no 5 pp 315ndash327 2013

Review ArticleSimulated Microgravity Critical Review on the Use ofRandom Positioning Machines for Mammalian Cell Culture

Simon L Wuest1 Steacutephane Richard1 Sascha Kopp2 Daniela Grimm2 and Marcel Egli1

1 Lucerne University of Applied Sciences and Arts School of Engineering and ArchitectureCC Aerospace Biomedical Science and Technology Space Biology Group Lucerne University of Applied Sciences and ArtsSeestraszlige 41 6052 Hergiswil Switzerland

2 Institute of Biomedicine Pharmacology Aarhus University Wilhelm Meyers Alle 4 8000 Aarhus C Denmark

Correspondence should be addressed to Marcel Egli marceleglihsluch

Received 15 May 2014 Revised 12 September 2014 Accepted 6 October 2014

Academic Editor Jack J W A Van Loon

Copyright copy 2015 Simon L Wuest et alThis is an open access article distributed under the Creative CommonsAttribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Random Positioning Machines (RPMs) have been used since many years as a ground-based model to simulate microgravity Inthis review we discuss several aspects of the RPM Recent technological development has expanded the operative range of theRPM substantially New possibilities of live cell imaging and partial gravity simulations for example are of particular interestFor obtaining valuable and reliable results from RPM experiments the appropriate use of the RPM is of utmost importanceThe simulation of microgravity requires that the RPMrsquos rotation is faster than the biological process under study but not sofast that undesired side effects appear It remains a legitimate question however whether the RPM can accurately and reliablysimulate microgravity conditions comparable to real microgravity in space We attempt to answer this question by mathematicallyanalyzing the forces working on the samples while they are mounted on the operating RPM and by comparing data obtained underreal microgravity in space and simulated microgravity on the RPM In conclusion and after taking the mentioned constraintsinto consideration we are convinced that simulated microgravity experiments on the RPM are a valid alternative for conductingexaminations on the influence of the force of gravity in a fast and straightforward approach

1 Introduction

Gravity is an omnipresent force on Earth and all livingorganisms have evolved under the influence of constantgravity Some organisms have learned to take advantage ofthe force of gravity by using it as a reference for orientationThe condition ofmicrogravity (or nearweightlessness) and itseffects on living organisms on the other hand have alwayspresented a fascinating scenario in biology and medicineWith the first manned space flights it became clear thatthe human organism reacts with a series of adaptations tomicrogravity Interestingly some of the symptoms observedin space such as wasting muscle mass and decreasing bonedensity are typically diagnosed in the elderly as well [1ndash3]This is one important factor that fostered scientistsrsquo interestin doing space research

Numerous studies on mammalian organisms for exam-ple have demonstrated that the absence of gravity has severe

effects not only on a systemic level but also on a cellular levelShort-term effects of microgravity (on the order of seconds)can be studied on research platforms such as drop towers orairplanes that fly in parabolic maneuvers In contrast long-term effects can only be studied on board sounding rockets(on the order of minutes) and space vehicles in flight Due tothe extensive preparation effort safety constraints and rareflight opportunities however access to space experimentsis limited For many years the random positioning machine(RPM) besides other tools has been successfully usedto simulate microgravity for screening studies pre- andpostflight experiments and hardware testing The principleof the RPM (a specialized two-axis form of the clinostat)is based on gravity vector averaging to zero [4] The typicalRPM system comprises two gimbal-mounted frames whichare each driven by independent motors Through dedicatedalgorithms the samples placed on the inner frame are con-stantly reoriented such that the gravity vector is distributed



in all directions over time Thus from the samplersquos point ofview the constantly reorienting gravity vectorrsquos trajectoryaveraged over time shall converge toward zero However 1 g isalways acting on the sample at any given instant It is assumedthat the gravity vector needs to point in a specific directionfor a minimal period of time in order to allow biologicalsystems like cells to adapt to the gravity vector But if thegravity vector constantly changes its orientation the cellswill lose the sense of direction and thus experience a statesimilar to microgravity (removed gravity vector) Thereforethe rotation of the frames shall be faster than the biologicalprocess studied [5] However the rotation cannot be too fastas centrifugal forces will become effective [6] Therefore theRPM is typically used to examine slow processes which areobserved at least on the timescale of hours It remains a legiti-mate question whether the RPM can reliably simulate micro-gravity In this review we attempt to provide an answer to thatquestion by comparing data of mammalian cells obtainedat real microgravity in space and at simulated microgravitygenerated by using the RPM In the first part however asummary is provided on the latest technical development aswell as new applications of the RPM

2 RPM Development and Technology

21 RPM Systems Todayrsquos RPMs were introduced byJapanese plant researchers for conducting their particularstudies [7 8] Later on a similar machine was developed inThe Netherlands (Dutch Space) [5] Although both systemswere commercialized [6] their range of use for doing space-related experiments was limited For instance scientificstudies with mammalian cells that are very sensitive totemperature fluctuations were difficult to carry out becauseof a missing temperature control unit Thus these kinds ofexperiments had to be operated in a temperature-controlledroom (eg a growth chamber) One approach to overcomethis limitation was to miniaturize the RPM to fit into anordinary cell culture incubator (max size 50 times 50 times 50 cm)that offers precisely controlled temperatures (also referred toas desktop RPM) [4] Through this RPM modification theinstallation of large climate chambers around RPMs becameunnecessary We have recently reported another approach toupgrading the RPM by installing a commercial CO

2incu-

bator onto the rotating frames This RPM called ldquorandompositioning incubatorrdquo (RPI) [9 10] has the advantage ofbeing independent of large laboratory incubators (Figure 1)Furthermore the closed chamber of the incubator isolates theenvironment of the culture flasks and thus prevents exposureof biological samples to vapor and wear from the machineryfor example [10] Besides differences in the design of thethree RPM types (regular RPM small desktop RPM andRPI) there appear to be slightly different concepts of how toaverage the gravity vectorThe algorithm implemented on theJapanese RPM (referred to as a regular RPM) lets the RPMrun with random rotational speeds and changes the velocityafter two possible predefined periods (eg 30 or 60 s) [8]TheDutch systems (referred to as regular and desktop RPMs)rotate with random speeds that are varied at random timepoints [6] In contrast our RPI rotates with constant velocity

Figure 1 Randompositioning incubator (RPI) featuring a fully inte-grated CO

2incubator (developed by the Institute for Automation

University of Applied Science Northwestern Switzerland)

but the rotation direction is inverted at random time pointsThe transition from forward to backward takes place at apredefined rotational acceleration [10] All three algorithmsemployed by the different RPM types are reported to bereliable in averaging gravity To our current knowledge thesealgorithms are equivalent from a biological point of view

22 Live Cell Imaging on the RPM Microscopy is a commonanalytical tool used in cell biology Even though microscopesare used on clinostats (rotation around one horizontal axis)[11 12] until recently live cell imaging was not successful onan operating RPM To date most of the optical microscopytechniques applied under simulated microgravity conditionshave been realized in the field of physical sciences For suchexperiments microscopes with a low numerical apertureand poor imaging performances were used because of theirintrinsic robustness to environmental disturbances such asvibrations In life science however high magnification isneeded to detect modifications at the cellular or subcellu-lar level Because most of the ground-based microgravityresearch platforms are not vibration-free high-performancemicroscopy has not been applicable Thus studies involvingcell imaging have been conducted in ground laboratories afterchemical fixation of the cell in microgravity This approachimplies a series of static shots which cannot truly reveal thedynamic processes and labile cellular events occurring in cellsin response to microgravity exposure

Until recently there was no system available that allowedhigh-quality real-time images taken at cellular or subcellularlevel under (real or simulated) microgravity The break-through came with the use of a digital holographic micro-scope (DHM) that we have combined with an epifluorescentmicroscope In this dual-mode microscope the two imagingmodes (DHM and fluorescent) operate sequentially TheDHM is an innovative interferometric microscope that is lesssensitive to vibrations The technological advantages of theDHMwhich comprise continuous and fast digital autofocus-ingwith a short exposure time allowhigh-resolution imaging[13ndash16] We tested the DHM on the RPM as well as duringparabolic flights and in both cases we obtained good data


[13ndash16] For instance we followed reorganization of the actincytoskeleton and fluctuations of the intracellular calciumconcentration under simulated microgravity (unpublisheddata)

3 Partial Earth Gravity Load

During past years RPM development was focused on theimprovement of the hardware We have also been workingon an upgrade of the software responsible for controllingthe motion of the RPM Three different algorithms wereintroduced recently that simulated partial Earth gravity (0to 06 g) allowing simulation of moon- or Mars-like gravityconditions [9] All algorithms are adaptations of the randomwalk algorithm originally designed to simulate microgravity[10] As described above simulated microgravity is achievedby rotating both frames with constant velocity and invertingthe rotation direction at random times Partial gravity isachieved by altering the randomwalk in a way that the Earthrsquosgravity vector is not completely randomized anymore andpoints (from the samplersquos point of view) for a prolongedtime in a specific direction In one algorithm version thisis accomplished by slowing down the rotational velocitywhile the gravity vector (considered in the sample frame)is pointing downwards Otherwise the frames rotate withthe predefined velocity The ratio of the two velocities finallydetermines the mean gravity (gravity vector averaged overtime) In the other two algorithm versions the random walkis interleaved with static intervals in which the frames standstill in a predefined orientation However the timing of thesestatic intervals (start point and duration) is handled differ-ently In one case the timing is flexible and adjusted online asthe experiment runs In the other case the timing is strictlyperiodic and predefined before the experiment starts [9]

All three algorithms were tested on suspended humanT cells and adherent mice myoblasts Chemically activatedT cells showed a decreased activation rate that correlatesstrongly to the decreasing simulated mean gravity values[9] The results were similar for all algorithms The adheredmyoblast (C2C12 cell line) showed a decreased proliferationrate with decreasing mean gravity [9] Interestingly thiseffect is algorithm dependent The correlation between meangravity and proliferation was reduced or disappeared in thetwo algorithms involving static intervals [9] Ideally thesetypes of partial gravity experiments are carried out in spaceby using a centrifuge To our knowledge no comparablespace experiments have been conducted so far except duringparticular parabolic flight campaigns of the European SpaceAgency (ESA) Therefore a direct comparison to spaceis not possible at this time However these experimentsdemonstrate that simulation of partial gravity opens a newfield of scientific questions that attracts other research groupsDutch Space was attracted by the new topic as well andthus recently introduced a modified desktop RPM (pre-sented at the ELGRA meeting 2013) allowing partial gravitysimulations Partial Earth gravity enabling RPMs increasethe application range substantially allowing investigation ofthe influence of gravitymdashlike on the moon or Mars forexamplemdashon cells and small organisms at affordable cost

Figure 2 Mouse myoblasts (C2C12 cell line) were cultured untilnear confluence and subsequently exposed to a frequently passingair bubble The culture chamber filled with medium was swingingupside down such that the intentional air bubble frequently passedthe same trajectory The sample was fixed and stained for actin(green) and DNA (blue) thereafter The cells in the trajectory of theair bubble got detached from the substrate (dark central area) whilecells in the unaffected area kept proliferating (lateral green areas)Interestingly detached cells could reattach to the opposite side of theculture chamber Measuring bar 200120583m (Due to the limited field ofview this image has been stitched together from five images)

These results may help to estimate the biological response ofcells or even whole organisms when exposed to the gravityloads of other planets or moons

4 RPM Use and Experiment Quality

41 Cultivation Method of Mammalian Cells In order toobtain comparable data it is important to standardize cellculture methods One of the most important aspects of doingso is a stable cultivating environment When cultivatingcells on the moving RPM additional aspects have to beconsidered such as avoiding air bubbles in the culturechambers [4] Experiments have shown that an air bubblepassing by adherent cells at the same trajectory repetitively (asthe culture chamber moves in a ldquoswinging motionrdquo) the cellscan detach from the substrate (Figure 2 unpublished obser-vation) Interestingly these cells often reattach at the oppositeside of the culture chamber wall Using air- and gas-tight cul-ture chambers on the RPM has the advantage of being moreindependent of the culture environmentHowever a gas-tightculture chamber requires a culture medium that does notrequire CO

2for pH buffering which reduces the overall cul-

tivation period in which the culture flasks do not have to bemanipulated Gas-tight chambers in turn can cause problemswhen cultivating gas-producing cells such as yeast cells

42 Artifacts through Kinematic Rotation In addition toa standardized cultivation method artifacts caused by thekinematic rotation need to be considered While the Earthrsquosgravity vector is distributed in a way that the mean gravityconverges to zero over time the accelerations caused by theRPMrsquos kinematics are not well controlled In order to avoidartifacts the rotational velocity the samplersquos distance to thecenter of rotation and the rotational acceleration (duringvelocity transitions) have to be chosen appropriately Sincethere has been no systematic study on acceptable limitsscientists have to rely on their common sense The followingconsiderations can be used as guidelines For explanatoryreasons we also refer here to the somewhat simpler caseof clinorotation around one axis Clinorotation and therelated rotating wall vessel (RWV) bioreactor are alternativemethods commonly used in many laboratories to simulate


0 5 10 15 20 25 30

0

002

004

006

008

01

012

014

016

018

Peak centrifugal acceleration worst case

Radius from center of rotation (cm)

Cen

trifu

gal a

ccel

erat

ion

(g)

90

80

70

60

50

40

30

Rota

tiona

l vel

ocity

(deg

s)

Figure 3 The worst-case peak centrifugal acceleration on an RPMdepending on the distance to the center of rotation (119886pc asymp 241sdot120596

2 sdot119903)For example a moderate rotational velocity of 60 degs (cyan line)and a distance of 10 cm from the center of rotation (vertical dashedline) results in a peak centrifugal acceleration of approximately003 g (horizontal dashed line)

microgravity on the ground These methods simulate micro-gravity by rotating samples around a horizontal axis (Select-ing the appropriate rotation velocity for suspended cells inclinostat experiments has been discussed elsewhere [17])

To minimize centrifugal acceleration the rotationalvelocity and the samplersquos distance to the center of rotationshould be set as low as the experiment allows As mentionedearlier the rotation shall be clearly faster than the biologicalprocesses investigated [5] For mammalian cell experimentsmany scientists have used a rotational velocity of 60 degs[4] In the case of chemically activated T cells (as discussedfurther below) we could also create a microgravity-likeenvironment with a rotational velocity of 40 degs [10] Forrotation around one axis as in a clinostat or centrifuge thecentrifugal acceleration (in ms2) is time independent and iscomputed by 119886

119888= 1205962 sdot 119903 where 120596 is the rotation velocity (in

rads) and 119903 is the distance from the center of rotation (inmeters) For rotations around two perpendicular axes as isthe case for RPMs the centrifugal acceleration becomes timedependentThus the centrifugal acceleration depends now onthe two rotation velocities the samplersquos position in space andtime It is no longer trivial tomake a statement on the effectivecentrifugal acceleration at the samples within the cultivationchamber For the simplified case where both velocities areequal and constant the centrifugal acceleration becomesperiodically oscillating By focusing on a worst-case scenarioin terms of centrifugal acceleration the analysis provideseasy equations in such a scenario the peak centrifugalacceleration (inms2) can be approximated to 119886pc asymp 241sdot120596

2 sdot119903

(Figure 3) where 120596 is the rotation velocity of both frames(in rads) and 119903 is the distance from the center of rotation(in meters) As the equation indicates all cells are ideallyplaced at the center of rotation Therefore the scientist is

0 5 10 15 20 25 300

0005

001

0015

002

0025

003

0035

004

0045

Tangential acceleration (velocity transition) worst case


Tang

entia

l acc

eler

atio

n (g

)

40

30

20

10

5

Rota

tiona

l acc

eler

atio

n (d

egs2)

Figure 4 The worst-case tangential acceleration depending onthe distance from the center of rotation (119886

119905= 2 sdot 120572 sdot 119903) For a

smooth velocity transition of for example 10 degs2 (green line)and 10 cm distance from the center of rotation (vertical dashedline) a tangential acceleration of approximately 0004 g is expected(horizontal dashed line)

responsible for compactly placing the samples around thecenter of rotation By using the distance to the center ofrotation from the sample farthest away from this point (worstcase) the largest expected centrifugal acceleration can beestimated For a moderate velocity of typically 60 degs [4]and a moderate distance from the center of rotation (eg10 cm) the centrifugal acceleration is in the order of 10minus2 gSuch small forces are detectable by some specialized cells[18] Since at any instance in time the Earthrsquos gravity vector(which is averaged to zero over time) is present as well thecentrifugal acceleration is two orders of magnitude smallerand we therefore consider it to be negligible In additionthe transitions of the framesrsquo rotational velocities introduceadditional accelerations and thus should be smooth byselecting a small rotational acceleration For the clinostatthis tangential acceleration (in ms2) is 119886

119905= 120572 sdot 119903 where

120572 is the rotational acceleration (in rads2) For the RPMthe tangential acceleration becomes 119886

119905= 2 sdot 120572 sdot 119903 in

the worst case when both frames accelerate simultaneously(Figure 4) For a smooth velocity transition of 10 degs2 anda moderate distance from the center of rotation (eg 10 cm)the tangential acceleration is well below 10minus2 g

Besides these parasitic accelerations rotation introducesfluid motion in the culture flask leading to shear forcesand enhanced convection (Figure 5)This condition is unlikespace conditions where no convection is present Thereforethe nutrition supply on the RPM is enhanced as comparedto static or space experiments In order to avoid additionalmechanical stimulation such as shear stress a moderaterotational velocity needs to be chosen and the velocitytransitions have to be smooth [19] Because the behavior offluid motion has not been fully elucidated yet the acceptablelimits for rotation velocity and acceleration are not clarified


0 s

(a)

12 s

(b)

24 s

(c)

40 s

(d)

Figure 5 The RPM rotation introduces fluid motion in the culture flask leading to shear forces and enhanced convection Therefore amoderate rotational velocity needs to be chosen and the velocity transitions have to be smooth in order to minimize the introduction ofadditional mechanical stimulation of the samples In this numerical illustration the fluid motion is shown if both frames rotate at 60 degsThis results in a periodic motion of 6 seconds The four images indicate snapshots of the velocity at 0 s (a) 12 s (b) 22 s (c) and 4 s (d)

However the values provided above are a good starting pointand have been successfully used in previous experiments[9 10]

5 Experiment Reporting

As new and innovative technologies expand the range ofpossible experiments it is becoming important to documentthe used hardware precisely In accordance with good labo-ratory practice (GLP) any researcher who is using RPMs orclinostats should follow the ldquoBonnCriteriardquo In this documentit is stated that ldquoExperimental reporting should include theproperties of the culture vessel culture media and carrierbeads These should also include dimensions and rotationspeed of vessels chemical consistency including density andviscosity of media size density and porosity of beads sizedensity and porosity of cells whether cells are motile or

non-motile density of beads with cells attached as well astime of rotation nature of controls operating temperatureand gas content [20]rdquo As described above improper use ofthe RPM can introduce additional forces leading to unwantedmechanical stimulation of the sample cells Interpretingresults from such experiments could lead to wrong conclu-sions and could thus jeopardize a whole study

6 RPM Application in MammalianCell Culture

61 Can the RPM Reproduce Microgravity Conditions Des-pite the long history of RPM usage the difference betweensimulated and real microgravity in space shall be criticallyexamined when interpreting experimental results Particu-larly for adhered cells the rotation generated by the RPMcould provide an unwanted source of mechanical stimuli


[6] Unfortunately only a few researches have systematicallycompared experiments performed in a real microgravityenvironment and on an RPM Most of these comparativestudies have been done on leukocytes for which the RPMshowed good agreement with space experiments it is wellknown that T lymphocytes fail to activate in microgravityafter being exposed to the activator ConA [21]This effect wasreproduced numerous times on an RPM [9 10 22 23] Simi-larly Villa and colleagues have shown slower proliferation ofthe human leukemic myelomonocytic cell line U937 exposedto simulated microgravity on the RPM [24] The samephenomenon was previously observed on a space shuttleexperiment [25] In a study on cell mobility under micro-gravity with the human leukemic monocytemacrophage cellline the RPM predicted real microgravity results Monocytelocomotion ability was clearly reduced in real as well as insimulatedmicrogravityThe authors suggest that this is linkedto changes in the cytoskeletal structures since they observedreduced density of actin filaments and disruption of the 120573-tubulin architecture [26 27] Furthermore peripheral bloodmononuclear cells cultured for 48 hours onboard the Inter-national Space Station (ISS) showed remarkably increasedapoptotic hallmarks which could also be reproduced undersimulated microgravity [28]

In recent years two investigators directly compared theresults from RPM experiments to results obtained in spaceconditions performed simultaneously in the first experi-ment primary porcine chondrocytes from articular cartilagewere flown for 16 days aboard the ISS Cells exposed tomicrogravity showed higher collagen III ratio and reducedaggrecanversican ratio at the mRNA level In addition celldensity was significantly reduced and the extracellular mat-rix straining was weaker on the ISS samplesThe samples thatwere simultaneously exposed to simulated microgravity onan RPM generally showed results that were similar to thoseof the space samples but not as prominent [29] In the secondexperiment cells from the human thyroid carcinoma cellline FTC-133 were flown aboard the Shenzhou-8 spacecraftand fixed after 10 days in space Cells exposed to spaceflightappeared to form three-dimensional tumor spheroids whilethe inflight 1 g controls remained in two-dimensional mono-layers The FTC-133 cells exposed to simulated microgravityon the RPM also formed three-dimensional spheroids eventhough the spheroids appeared to be smaller than thoseformed in space [30] In addition EGF and CTGF geneexpressionwas upregulated in both real and simulatedmicro-gravity Interestingly EGF expression was lower and CTGFexpression was higher in the RPM samples than the spacesamples [30] The reason the RPM sample showed inter-mediate effects between the 1 g control and the space samplesis not clear at this point Since the RPM can only be used forslow processes one possible speculation is that some of theunderlying molecular processes might be too fast for RPM-simulated microgravity

In conclusion the RPM has been shown to mimic micro-gravity responses reliably for several but not all experimentalconditions Particularly for leukocytes several effects seenin space were reproduced on the RPM Particular stud-ies designed to investigate differences in cellular responses

between space samples and samples exposed to simulatedmicrogravity elucidated an underestimation or overestima-tion of simulated versus real microgravity Overall the RPMgenerally seems to underestimate the spaceflight effectsTherefore results from RPM experiments need to be inter-preted with caution and if possible more directly comparedto experiments under realmicrogravity in order to fully assesstheir capability to support gravitational biology studies

62 Novel Applications of the RPM The exact mechanismsby which mechanical stimuli initiate cellular modificationshave still not been fully elucidated [31]This is the motivationof mechanobiologists to expose cells to various mechani-cal stimuli such as distinct patterns of shear flow tensilestretch or mechanical compression at various parametriccombinations of magnitude duration or frequency [31] TheRPM can be regarded as an additional mechanical devicefor reducing the long-term effects of the mechanical force ofgravity Due to the constant reorientation of samples on theRPM gravity-dependent intracellular responses will not betriggered anymore Thus one can say that the RPM generatesa state of a mechanically unloaded environment in which thelonger-lasting impact of gravity can be studied

Monolayer (two-dimensional) cell cultures have beensuccessfully used for many decades allowing a better under-standing of many cellular and molecular processes Theyactually represent an important source of information priorto animal experimentation Despite numerous advantagesthe monolayer model cannot simulate organs or tissuesrealistically Therefore three-dimensional cell culturing hasemerged over the last decades as an alternative to mimicbetter tissue-like organization with the idea of closing thegap of uncertainty between tissue-like and monolayer cellculture The RPM in that context appears as an alternativeapproach to generating a three-dimensional culture [32]The random repositioning of the cells around the gravityvector over time allows constant redistribution of gravityforces which thus leads to the formation of cell aggregatesthat can form microspheroids (Figure 6) [32ndash34] Spheroidsorganized as multilayers are closer to in vivo tissue situationthan monolayer cells [32] Such samples are therefore moreaccurate as a model integrating the three-dimensional realsurroundings of a cell in an in vivo tissue Thus spheroidstructures open a new field of applications such as testsystems for drug therapies or diagnosis [35] The spheroidstructure is actually a good model to screen for penetrationcharacteristics of drugs or antibodies through tissue

7 Conclusion

Several RPMs have evolved during the past years that featuredifferent designs functions and motion patterns They allhave reliably proven to simulate microgravity conditionsDevelopments to RPMhardware and software have expandedthe experimental possibilities substantially The successfuloperation of digital holographic microscopy (DHM) on theRPM and the implementation of partial gravity algorithmshave opened new fields in gravitational research particularlyin mechanobiology


Figure 6Thyrocytes cultured for seven days on the RPM organizedto spheroid structures (arrow)

In order to obtain reliable and comparable data theappropriate use of the RPM and application of standardizedcultivation methods are of central importance The RPM hasbeen established as a reliable tool supporting ground-basedmicrogravity studies Effects seen in real microgravity werereproducedwith good agreement on RPMs Some RPM stud-ies however also showed cellular effects that were betweenthose of the real microgravity and 1 g ground control resultsThe RPM is furthermore an ideal tool for preliminary micro-gravity tests screening studies inwhich simulatedmicrograv-ity effects are checked on various organisms and hardwaretesting Particularly for suggesting live science experimentsfor the conduction under real microgravity in space thepresentation of preliminary data showing modificationsunder simulated microgravity is becoming very importantAdvances in RPM engineering and live science qualify theRPM as an interesting tool for novel applications such asthree-dimensional cell culturing as well as tissue engineering

Conflict of Interests

The authors have no conflict of interests regarding thepublication of this paper

Acknowledgments

The authors thank their coworkers at the CC AerospaceBiomedical Science and Technology and especially NicoleWittkopf for the support and critical discussions Further-more they would like to thank Adrian Koller and MarianaReyes Perez from the CC of Mechanical Systems LucerneSchool of Engineering and Architecture for interesting dis-cussions and close collaboration Special thanks also go toJorg Sekler and his coworkers at the Institute for AutomationUniversity of Applied Science Northwestern Switzerland forthe fruitful collaboration

References

[1] R H Fitts S W Trappe D L Costill et al ldquoProlongedspace flight-induced alterations in the structure and function of

human skeletalmuscle fibresrdquo Journal of Physiology vol 588 no18 pp 3567ndash3592 2010

[2] D A Riley J L W Bain J L Thompson et al ldquoDecreased thinfilament density and length in human atrophic soleus musclefibers after spaceflightrdquo Journal of Applied Physiology vol 88no 2 pp 567ndash572 2000

[3] S W Trappe T A Trappe G A Lee J J Widrick D L Costilland R H Fitts ldquoComparison of a space shuttle flight (STS-78)and bed rest on human muscle functionrdquo Journal of AppliedPhysiology vol 91 no 1 pp 57ndash64 2001

[4] A G Borst and J J W A van Loon ldquoTechnology and develop-ments for the randompositioningmachine RPMrdquoMicrogravityScience and Technology vol 21 no 4 pp 287ndash292 2009

[5] D Mesland ldquoNovel ground-based facilities for research in theeffects of weightrdquo ESAMicrogravity News vol 9 pp 5ndash10 1996

[6] J J W A van Loon ldquoSome history and use of the random posi-tioning machine RPM in gravity related researchrdquoAdvances inSpace Research vol 39 no 7 pp 1161ndash1165 2007

[7] T Hoson S Kamisaka YMasuda andM Yamashita ldquoChangesin plant growth processes under microgravity conditions simu-lated by a three-dimensional clinostatrdquoThe Botanical MagazineTokyo vol 105 no 1 pp 53ndash70 1992

[8] T Hoson S Kamisaka Y Masuda M Yamashita and BBuchen ldquoEvaluation of the three-dimensional clinostat as asimulator of weightlessnessrdquo Planta vol 203 no 1 pp S187ndashS197 1997

[9] T Benavides Damm I Walther S L Wuest J Sekler and MEgli ldquoCell cultivation under different gravitational loads usinga novel random positioning incubatorrdquo Biotechnology and Bio-engineering vol 111 no 6 pp 1180ndash1190 2014

[10] S L Wuest S Richard I Walther et al ldquoA novel micrograv-ity simulator applicable for three-dimensional cell culturingrdquoMicrogravity Science and Technology vol 26 no 2 pp 77ndash882014

[11] M Cogoli ldquoThe fast rotating clinostat a history of its use ingravitational biology and a comparison of ground-based andflight experiment resultsrdquo ASGSB Bulletin vol 5 no 2 pp 59ndash67 1992

[12] R Hemmersbach M von der Wiesche and D Seibt ldquoGround-based experimental platforms in gravitational biology andhuman physiologyrdquo Signal Transduction vol 6 no 6 pp 381ndash387 2006

[13] M F Toy J Kuhn S Richard J ParentM Egli andCDepeurs-inge ldquoAccelerated autofocusing of off-axis holograms usingcritical samplingrdquo Optics Letters vol 37 no 24 pp 5094ndash50962012

[14] C Pache J Kuhn K Westphal et al ldquoDigital holographicmicroscopy real-time monitoring of cytoarchitectural alter-ations during simulated microgravityrdquo Journal of BiomedicalOptics vol 15 no 2 Article ID 026021 2010

[15] M F Toy S Richard J Kuhn A Franco-Obregon M Egliand C Depeursinge ldquoEnhanced robustness digital holographicmicroscopy for demanding environment of space biologyrdquoBiomedical Optics Express vol 3 no 2 pp 313ndash326 2012

[16] M F Toy C Pache J Parent J Kuhn M Egli and CDepeursinge ldquoDual-mode digital holographic and fluorescencemicroscopy for the study of morphological changes in cellsunder simulated microgravityrdquo inThree-Dimensional and Mul-tidimensional Microscopy Image Acquisition and ProcessingXVII pp 7570ndash7573 2010


[17] D M Klaus P Todd and A Schatz ldquoFunctional weightlessnessduring clinorotation of cell suspensionsrdquo Advances in SpaceResearch vol 21 no 8-9 pp 1315ndash1318 1998

[18] D Driss-Ecole V Legue E Carnero-Diaz and G PerballdquoGravisensitivity and automorphogenesis of lentil seedlingroots grown on board the International Space Stationrdquo Physi-ologia Plantarum vol 134 no 1 pp 191ndash201 2008

[19] C A Leguy R Delfos M J B M Pourquie et al ldquoFluidmotionformicrogravity simulations in a randompositioningmachinerdquoGravitational and Space Biology Bulletin vol 25 no 1 2011

[20] T Hammond and P Allen ldquoThe Bonn criteria minimal experi-mental parameter reporting for clinostat and random position-ing machine experiments with cells and tissuesrdquo MicrogravityScience and Technology vol 23 no 2 pp 271ndash275 2011

[21] M Cogoli-Greuter ldquoThe lymphocyte storymdashan overview ofselected highlights on the in vitro activation of human lympho-cytes in spacerdquoMicrogravity Science and Technology vol 25 no6 pp 343ndash352 2014

[22] M Schwarzenberg P Pippia M A Meloni G Cossu MCogoli-Greuter and A Cogoli ldquoSignal transduction in T lym-phocytesmdasha comparison of the data from space the free fallmachine and the random positioning machinerdquo Advances inSpace Research vol 24 no 6 pp 793ndash800 1999

[23] I Walther P Pippia M A Meloni F Turrini F Mannu and ACogoli ldquoSimulated microgravity inhibits the genetic expressionof interleukin-2 and its receptor in mitogen-activated T lym-phocytesrdquo FEBS Letters vol 436 no 1 pp 115ndash118 1998

[24] A Villa S Versari J A Maier and S Bradamante ldquoCell behav-ior in simulated microgravity a comparison of results obtainedwith RWV and RPMrdquo Gravitational and Space Biology Bulletinvol 18 no 2 pp 89ndash90 2005

[25] J P Hatton F Gaubert M L Lewis et al ldquoThe kinetics oftranslocation and cellular quantity of protein kinaseC in humanleukocytes aremodified during spaceflightrdquoTheFASEB Journalvol 13 supplement pp S23ndashS33 1999

[26] M A Meloni G Galleri G Pani A Saba P Pippia and MCogoli-Greuter ldquoSpace flight affects motility and cytoskeletalstructures in human monocyte cell line J-111rdquo Cytoskeleton vol68 no 2 pp 125ndash137 2011

[27] M A Meloni G Galleri P Pippia and M Cogoli-GreuterldquoCytoskeleton changes and impaired motility of monocytes atmodelled low gravityrdquo Protoplasma vol 229 no 2-4 pp 243ndash249 2006

[28] N Battista M A Meloni M Bari et al ldquo5-Lipoxygenase-dependent apoptosis of human lymphocytes in the interna-tional space station data from the ROALD experimentrdquo TheFASEB Journal vol 26 no 5 pp 1791ndash1798 2012

[29] V Stamenkovic G Keller D Nesic A Cogoli and S P GroganldquoNeocartilage formation in 1 g simulated and microgravityenvironments implications for tissue engineeringrdquo Tissue Engi-neeringmdashPart A vol 16 no 5 pp 1729ndash1736 2010

[30] J Pietsch X Ma M Wehland et al ldquoSpheroid formation ofhuman thyroid cancer cells in an automated culturing systemduring the Shenzhou-8 SpacemissionrdquoBiomaterials vol 34 no31 pp 7694ndash7705 2013

[31] J Wang D Lu D Mao and M Long ldquoMechanomics anemerging field between biology and biomechanicsrdquo Protein ampCell vol 5 no 7 pp 518ndash531 2014

[32] J Pietsch A Sickmann G Weber et al ldquoA proteomic approachto analysing spheroid formation of two human thyroid cell linescultured on a random positioning machinerdquo PROTEOMICSvol 11 no 10 pp 2095ndash2104 2011

[33] D Grimm J Bauer C Ulbrich et al ldquoDifferent responsivenessof endothelial cells to vascular endothelial growth factor andbasic fibroblast growth factor added to culture media undergravity and simulated microgravityrdquo Tissue Engineering Part Avol 16 no 5 pp 1559ndash1573 2010

[34] C Ulbrich ldquoCharacterization of human chondrocytes exposedto simulated microgravityrdquo Cellular Physiology and Biochem-istry vol 25 no 4-5 pp 551ndash560 2010

[35] A Ivascu and M Kubbies ldquoRapid generation of single-tumorspheroids for high-throughput cell function and toxicity analy-sisrdquo Journal of Biomolecular Screening vol 11 no 8 pp 922ndash9322006

Research ArticleRegulation of ICAM-1 in Cells of the MonocyteMacrophageSystem in Microgravity

Katrin Paulsen1 Svantje Tauber12 Claudia Dumrese13 Gesine Bradacs1

Dana M Simmet12 Nadine Goumllz1 Swantje Hauschild12 Christiane Raig1

Stephanie Engeli1 Annett Gutewort12 Eva Huumlrlimann1 Josefine Biskup1

Felix Unverdorben2 Gabriela Rieder1 Daniel Hofmaumlnner1 Lisa Mutschler1

Sonja Krammer1 Isabell Buttron1 Claudia Philpot4 Andreas Huge5 Hartwin Lier6

Ines Barz7 Frank Engelmann67 Liliana E Layer1 Cora S Thiel12 and Oliver Ullrich1289

1 Institute of Anatomy Faculty of Medicine University of Zurich Winterthurerstraszlig 190 8057 Zurich Switzerland2Department of Machine Design Engineering Design and Product Development Institute of Mechanical EngineeringOtto-von-Guericke-University Magdeburg Universitatsplatz 2 39106 Magdeburg Germany3Flow Cytometry Facility University of Zurich Winterthurerstraszlig 190 8057 Zurich Switzerland4German Aerospace Center Space Agency Konigswinterer Straszlige 522-524 53227 Bonn Germany5Integrated Functional Genomics (IFG) University of Muenster Roentgenstraszlig 21 48149 Muenster Germany6KEK GmbH Kemberger Straszlige 5 06905 Bad Schmiedeberg Germany7University of Applied Science Jena Carl-Zeiss-Promenade 2 07745 Jena Germany8Zurich Center for Integrative Human Physiology (ZIHP) University of Zurich Winterthurerstraszlig 190 8057 Zurich Switzerland9Study Group ldquoMagdeburger Arbeitsgemeinschaft fur Forschung unter Raumfahrt- und Schwerelosigkeitsbedingungenrdquo (MARS)Otto-von-Guericke-University Magdeburg Universitatsplatz 2 39106 Magdeburg Germany

Correspondence should be addressed to Oliver Ullrich oliverullrichuzhch



Copyright copy 2015 Katrin Paulsen et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Cells of the immune system are highly sensitive to altered gravity and themonocyte as well as themacrophage function is proven tobe impaired under microgravity conditions In our study we investigated the surface expression of ICAM-1 protein and expressionof ICAM-1 mRNA in cells of the monocytemacrophage system in microgravity during clinostat parabolic flight sounding rocketand orbital experiments In murine BV-2 microglial cells we detected a downregulation of ICAM-1 expression in clinorotationexperiments and a rapid and reversible downregulation in the microgravity phase of parabolic flight experiments In contrastICAM-1 expression increased inmacrophage-like differentiated humanU937 cells during themicrogravity phase of parabolic flightsand in long-termmicrogravity provided by a 2D clinostat or during the orbital SIMBOXShenzhou-8 mission In nondifferentiatedU937 cells no effect of microgravity on ICAM-1 expression could be observed during parabolic flight experiments We concludethat disturbed immune function in microgravity could be a consequence of ICAM-1 modulation in the monocytemacrophagesystem which in turn could have a strong impact on the interaction with T lymphocytes and cell migration Thus ICAM-1 can beconsidered as a rapid-reacting and sustained gravity-regulated molecule in mammalian cells

1 Introduction

Several limiting factors for human health and performancein microgravity have been clearly identified arising fromthe immune system and substantial research activities are

required in order to provide the basic information for appro-priate integrated risk management The gravity-sensitivenature of cells of the immune system renders them anideal biological model in search for general gravity-sensitivemechanisms to understandhow the architecture and function



of human cells are related to the gravitational force andtherefore adapted to life on Earth Cells of the immune systemare highly sensitive to altered gravity (for review see [1ndash4]) T lymphocytes as well as monocytes and macrophagesare impaired severely in their functions under microgravityconditions [2ndash4] T cell activation is severely disturbed undermicrogravity conditions as shown in the blood of astronautsduring and after space flight [5] and in numerous in vitroexperiments (reviewed by [6]) In monocytes the secretionof the cytokines IL-1 IL-6 TNF-alpha and IL-10 is alteredunder microgravity conditions [7 8] Substantial changes ingene expression of monocytes and in gene induction associ-ated with the differentiation of monocytes into macrophageshave been observed [8]

Migration and adhesion of immune competent cells atareas of infection inflammation or structural disordersare indispensable for the immune response [9] For theseprocesses the communication and connection between cellsare essential The integrins of the LeuCAM family (LFA-1 and MAC-1) and their ligands the intercellular adhesionmolecules (ICAMs) are receptors that mediate the attach-ment between cells (cell-cell contact) and of cells and theextracellular matrix (cell-matrix contact) [10] ICAMs aretransmembrane proteins that are expressed on epithelial cellsendothelial cells and cells of the immune system includingT cells and macrophages Binding of ICAM-1 (CD54) toreceptors on endothelia of blood vessels enables leucocytesto attach and migrate through the endothelia to sites ofinflammation [11] Later on in the immune reaction closeand strong interaction between ICAM-1 and LFA-1 is indis-pensable for the immunological synapse formation betweenT cells and antigen-presenting cells such as monocytes[12]

ICAM-1 expression is known to be upregulated duringmechanical stress [13] in a long-term microgravity envi-ronment [14] in the NASA-developed Rotary Cell CultureSystems (RCCS) as well as during short-term microgravityin parabolic flights [15] in endothelial cells While thesestudies show gravity sensitivity of ICAM-1 in endothelialcells less is known about the effects of microgravity on cellsof the 2 monocytemacrophage system (MMS) Thereforein this study we investigate whether the ICAM-1 surfaceexpression is regulated by altered gravity in these cell typesThe MMS belongs to the innate immune system and rep-resents the bodyrsquos first line of defense The innate immunesystem is characterized by a fast but unspecific immunereaction and it activates the adaptive immune responseThisactivation occurs through interaction of antigen-presentingcells (APCs)mdashdendritic cells and macrophages [16]mdashwithT lymphocytes Macrophages are relatively long-lived carrya variety of surface receptors and reside in many tissuesincluding the gastrointestinal tract the respiratory tract theliver the spleen bones and connective tissues [17]Microglialcells are the brain-resident macrophage population whichcrucially controls and regulates immune reactions inside thecentral nervous system (CNS)

In our study we investigated the surface expression ofICAM-1 protein and expression of ICAM-1 mRNA in cellsof the monocytemacrophage system in microgravity As cell

models we used primary cells (macrophages T cells) as wellas cell lines (U937 myelomonocytic cells macrophage-likedifferentiated U937 cells and BV-2 microglial cells) We con-ducted experiments with different durations of microgravityin clinorotation parabolic flight sounding rocket and orbitalflight experiments

2 Methods

21 U937 Cell Culture and Macrophage-Like DifferentiationU937 cells (ATCC CRL-15932) are a human monocyticcell line that preserves the main monoblastic character-istics of monocytes including the ability to differentiateinto a macrophage-like phenotype U937 cells were culturedin RPMI 1640 medium with or without 20mM HEPES(Biochrom Berlin Germany) supplemented with 10 fetalcalf serum (FCS Biochrom) or 10 human serum (HSBiowest) 2mMglutamine (PromoCell) 100UmLpenicillinand 100 120583gmL streptomycin (Gibco) Subcultivation wasdone at a cell density of 1 times 106 cellsmL Stimulationand differentiation were performed by adding 25 nM phor-bolmyristylacetate (PMA) (Sigma-Aldrich) in dimethylsul-foxid (DMSO 01) (Sigma-Aldrich) at a cell density of05times 106 cellsmLDifferentiationmediumwas supplementedwith 10 HS 2mM glutamine 100UmL penicillin and100 120583gmL streptomycin Cells were differentiated on poly-carbonate (Makrolon) slides (SIMBOXShenzhou-8) or incell culture flasks (parabolic flights and clinorotation exper-iments) for 72 h at 37∘C with 5 CO

2into a macrophage-

like phenotype U937macrophage-like cells were detached byMacrophage Detachment Solution DXF (PromoCell) follow-ing manufacturerrsquos protocol After detaching cells were filledimmediately into Nutrimix bags (B BraunMelsungen) (1-2 times107 cells in 10mL medium each bag) for parabolic flight orin serological pipettes (1mL cell suspension with 025ndash05 times106mL) for clinorotation experiments [18 19]

22 Primary Human Macrophages Human primary M2macrophages (PromoCell) were cultivated with M2-Macrophage Generation Medium DXF (PromoCell) Cellswere detached by Macrophage Detachment Solution DXF(PromoCell) following the manufacturerrsquos protocol Afterdetaching cells were filled immediately into Nutrimix bags(B Braun Melsungen) (2 times 106 cells in 10mL medium eachbag) for parabolic flight or in serological pipettes (1mLcell suspension with 025ndash05 times 106mL) for clinorotationexperiment [18 19]

23 BV-2 Microglia Cell Culture Since primary microglia arenot available in quantities required for parabolic flight exper-iments we used the murine cell line BV-2 whose functionresembles that of tissue macrophages and which share manyproperties with both peripheral macrophages andmonocytes[20] BV-2 microglial cells were cultured in DMEMmedium(Biochrom Berlin Germany) supplemented with 10 FCSand without antibiotics 72 hours before parabolic flightexperiment the cells were set on only 2 FCS (serumstarved) for the transport and were supplemented again


(a) (b)

(c) (d)

Figure 1 Technology for cell culture experiments in different microgravity research platforms (a) Fast-rotating two-dimensional (2D)clinostat manufactured by the German Aerospace Center (DLR Cologne Germany) was used to provide simulated microgravity Underthe chosen experimental conditions (60 rpm 4mm pipette diameter) a maximal residual acceleration of 4 times 10minus3 g is achieved at the outerradius of the pipette and decreases towards the center (b) Experimental hardware structure which consists of an incubator rack to store thecell containers temporarily before the experiment at 37∘C (left) an experimental rack in which all active aggregates are accommodated andwhere the living cells are handled during altered gravity (right) and a cooling rack to temporarily store all cell containers after the injection ofthe stopfixation liquid at 4∘C until landing (front) (c) Payload of TEXUS-49 sounding rocket tempered and vacuum-resistant container withexperiment syringe systems (d) Plunger unit EUE for SIMBOX (Science in Microgravity Box) incubator system support structure (housingmade of PEEK) which includes three culture chambers and six supply units two for each culture compartment Each culture chamberrepresents an independent loop The culture chambers filled with medium are closed on the top of the housing by means of polycarbonatespecimen window slides where the adherent cells are attached beforehand The housing is tightened by silicon sealing and covered by analuminum plate (cover) fixed with screws

with 10 FCS after arrival Before transport to Bordeaux-Merignac airport BV-2 cells were transferred into 200mLNutrimix bags (B Braun Melsungen Melsungen Germany)at a density of 3 times 106 cells in 15mL medium

24 Experiments in Simulated Microgravity (2D Clinorota-tion) A fast-rotating two-dimensional (2D) clinostat man-ufactured by the German Aerospace Center (DLR CologneGermany) was used to provide simulated microgravity(Figure 1(a)) The principle of clinorotation-induced micro-gravity is the rotation of a cell suspension in a serologicalpipette perpendicular to the Earthrsquos gravityThemicrogravityproduced is an averaging of the gravity vector if the clinostatrotates with 40ndash100 rpm Under the chosen experimental

conditions (60 rpm 4mmpipette diameter) amaximal resid-ual acceleration of 4 times 10minus3 g is achieved at the outer radiusof the serological pipette and decreases towards the centerThe clinostat device was placed in an incubator providingconstant 37∘C Fifteen serological pipettes rotated at the sametime with 60 rpm 1 g controls were placed at the ground plateof the clinostat without rotation but the same environmentalconditions are as the 120583g samples The density of the cellsuspension was 05 times 106mL (U937 macrophage-like cells)025ndash05 times 106mL (human primary macrophages) or 075 times106mL (BV-2 microglial cells) in 1mL volume each Theduration of the filling procedure was not longer than 10minfor all 30 serological pipettes Cells were cultured in theserological pipettes for 24 hndash96 h After clinorotation cells


were fixed by the addition of 500 120583L of 3 PFA (Sigma-Aldrich)2 sucrose (Sigma-Aldrich) solution for 30minwashed with PBS and analyzed after immunocytochemicalstaining by flow cytometry

25 Parabolic Flights asMicrogravity Research Platform Dur-ing a parabolicmaneuver an aircraft is weightless by flying ona Keplerian trajectory described as an unpropelled body inideally frictionless space subjected to a centrally symmetricgravitational field [21] During this free-fall trajectory theresultant of all forces acting on the aircraft other thangravity is zeroed During a flight campaign which normallyconsists of three individual flights 31 parabolas are flownon each flight with 93 parabolas in total On each parabolathere is a period of increased gravity (18 g) which lastsfor 20 seconds immediately prior to and following the 20-second period of reduced gravity (acceleration in 119909- 119910-and 119911-axis was below 2 times 10minus3 g at all times during the 120583gparabola Figure 1(a)) During the parabolic flight maneuverthe aircraft gradually pulls up its nose and starts climbingat an angle of approximately 45 degrees This phase lastsfor about 20 seconds during which the aircraft experiencesan acceleration of around 18 g The engine thrust is thenreduced to the minimum required to compensate for air-drag and the aircraft is then in a free-fall condition lastingapproximately for 20 seconds during which weightlessness isachieved At the end of this phase the aircraft must pull outthe parabolic arc a maneuver which gives rise to another 20second period to 18 g on the aircraft after which it returnsto normal level flight attitude Special designated flight areaswere above the Atlantic Ocean and the Mediterranean SeaExperiments were conducted during the 10th and 19th DLRparabolic flight campaigns of the German Aerospace Center(ldquoDeutsches Zentrum fur Luft- und Raumfahrtrdquo DLR) inBordeaux France The campaign used the only large aircraftthat is licensed in Europe to perform parabolic flights forresearch purposes the Airbus A300 ZERO-GThis aircraft isa specially configured test aircraft operated by NOVESPACE(Bordeaux France) according to the standing orders ofNOVESPACE (A300 ZERO-G Rules and Guidelines RG-2001-1 RG-2008-1 RG-2008-2 RG-2009-1 andRG-2009-02)and the CEV (Centre drsquoessai en vol)

26 In-Flight Hardware for Parabolic Flight Experiments Acustom-made hardware meeting the requirements for exper-iments with human cell culture on board the Airbus A300ZERO-G was developed in collaboration with KEK GmbHGermany (Figure 1(b)) The system has already been usedsuccessfully for cell culture experiments during 9 parabolicflight campaigns [18 19]The system consists of double-sealedcell containers holding the cells of themonocyte-macrophagesystem and three experimentalmodules that supply storage ofsamples before the experiment half-automated performanceof the experiment and storage of the processed samplesThe first module holds the cell containers at 365∘C in ahanging position From there containers are transferred intothe second module manually In this module cells were fixedby the addition of fixation reagent upon triggering Triggering

was done manually at defined time intervals (20 sec) afterthe onset of the gravitational condition of interest Thethird experimental module served as in-flight storage for thefixated samples at 4∘C Three samples could be processedin parallel Sample exchange required approximately oneminute of a defined procedure by three trained persons

27 Procedures during Parabolic Flight Experiments Trans-port of in-flight cell culture bags in in-flight-configurationand of fixed samples after the parabolic flight was providedby the Swiss Air Force from Zurich to Bordeaux during eachflight day of the 13th DLR Parabolic Flight Campaign or bytrain during each flight day of the 19th DLR Parabolic FlightCampaign After arrival at the flight location on the eveningbefore the flight cells were incubated overnight at 37∘C andhandled very carefully in order to avoid any mechanical ortemperature cell stress All steps of the entire cell preparationand transport procedure had been tested extensively withrespect to cell viability and function beforehand All proce-dures during the parabolic flight campaign had been testedseveral times and highly standardized following an extensiveand detailed standard protocol During the campaign allprocedures were documented and double-checked In-flight120583g and 1 g control experiments were performed in 200mLNutrimix bags [18 19] used as in-flight cell culture bags con-taining 3times 107 cells in 15mLDuring the onset of 120583g or during1 g (in-flight control experiments) 10 ng PMAmL (with 001residual DMSO) or 10 ng TNF-120572mL or plain cell culturemediumwere added to the cells After 20 sec of 120583g or 1 g cellswere fixed by addition of 1 formaldehyde (Sigma-Aldrich)(for cytometry analysis) or lysed by RLT buffer (Qiagen)(for RNA analysis) and cooled immediately (4∘C) during theremaining flight Experiments were performed at least threetimes during independent flights and separate flight daysAfter the flight fixed cells were transported to the laboratorieson the same day harvested and subjected to analysis

28 TEXUS-49 Sounding Rocket Experiment For theTEXUS-49 campaign at ESRANGE (Kiruna Sweden) U937cells were cultivated in the fully installed laboratories on siteCells were seeded with a density of 02 times 106 cellsmL and themedium was exchanged every 48 hours as described aboveOn the launch day cells were visually inspected harvestedcounted and pooled to a concentration of 5 times 107 cellsmL05mL of this cell suspension was filled into a sterile 3mLplastic syringe shortly before the launch Additionallyone syringe was filled with 03mL of cell culture mediumand another one with 1mL Trizol LS (Life TechnologiesGermany) The three syringes were mounted on a plasticblock with a tubing system connecting them This unit wasfinally integrated into the automatically operated experimentsystem (Figure 1(c)) In total 35 of these experiment unitswere prepared and kept at 37∘C until the integration intothe payload of the rocket During the experimental runfirst the 03mL of medium as a potential placeholder foran activation solution and then the 1mL of Trizol LS wereinjected to the cell suspension at defined time points to lysethe cells and preserve the current status of differential gene


expression Injections were performed at 75 sec after launchto monitor a so-called baseline (BL) directly before the 120583gphase and at 375 sec after launch at the end of the 120583g phaseA group of 1 g ground controls was treated immediatelyafter the 120583g sample group TEXUS-49 consisted of a VSB-30engine (S-30 solid rocket stage with an S-31 second stage)and of the payload The rocket was launched on March 292011 at 0601 am from the ESRANGE Space Center nearKiruna Sweden During the ballistic suborbital flight analtitude of 268 km and 378 sec of microgravity with a qualityof 10minus5 g were achieved

29 SIMBOX Incubator System with Plunger ExperimentInsert SIMBOX (Astrium GmbH Friedrichshafen Ger-many Kayser Italia Livorno Italy) is a programmable space-qualified incubator for biological research in space equippedwith a 1 g in-flight centrifuge for 1 g control experiments Theincubator allows for fully automatic execution of biologicalexperiments with limited use of commands during orbitalflight in a controlled thermal environment The SIMBOXincubator (internal volume 34 liters dimensions 461 times 551 times273mm empty mass 16 kg fully integrated mass 345 kgmax power 130W) accommodates 40 experiment uniqueequipment (EUEs) with 24 EUEs on the 120583g-platform and16 EUEs on the 1 g-centrifuge The plunger experimentinsert (Figure 1(d)) was developed by Astrium GmbH andis described in the Astrium Space Biology Product Catalog[22] It allows medium exchange and chemical fixation ofadherent cell cultures There are two plungers which can befilled with any liquid and automatically activated to injectit into the experimental volume The EUEs consisted of asupport structure (housing made of PEEK) which includesthree culture chambers (CCs) and six supply units (SUplungers) two per culture compartment Each CC has twoSUs and represents an independent loop The CCs are closedon the top of the housing by Specimen Slides (SS) made ofpolycarbonate on which the adherent cells were attachedThe chamber (covered by the window slide) contained themedium The housing is tightened by silicon sealing andcovered with an aluminum plate (cover) which is fixedwith screws The container lid of the Biorack standardtype I container is mounted onto the housing The Biorackstandard is based on the accommodation of various EUEsinto experiment containers which provide the interface tofacilities and support infrastructure [22] The plunger unit isqualified for an unmanned capsulemission and for use on theInternational Space Station (ISS)

The unmanned Shenzhou-8 spacecraft was launched onOctober 31 2011 at 2158 UTC (November 1 2011 0558LT) on board of a Long March 2F (CZ-2F) rocket from theJiuquan Satellite Launch Center (JSLC) in Inner MongoliaOn November 17 the capsule was autonomously deorbitedand landed at 1238 UTC (2038 LT) around 500 km northof Beijing The SIMBOX was recovered immediately andtransported by helicopter and jet aircraft to the PITCBeijing Total early retrieval time was 6 hours On arrivalat the PITC the SIMBOX was opened and the EUEswere removed and inspected The samples were recovered

and stored in cold (4∘C) PBS until arrival in Zurich foranalysis

210 SIMBOX Experiment Execution Medium was changedbefore integration of the slides into the EUEs Insidethe EUEs the slides were bedded in 05mL fully CO

2

saturated RPMI 1640 medium with 10 HS 2mM glu-tamine 100UmL penicillin 100 120583gmL streptomycin and250 ngmL amphotericin B (PromoCell) Bellow 1 3 and 5 ofthe EUEs [23] were filled with RPMI 1640 medium 10 HS100UmL penicillin 100120583gmL streptomycin 250 ngmLamphotericin B 2mM glutamine 1 PFA and 06 sucroseBellow 2 4 and 6 of the EUEs [23] were filled withPBS 100UmL penicillin 100 120583gmL streptomycin and250 ngmL amphotericin B Two EUEs (6 chambers) wereprepared for the 120583g-position and one EUE (3 chambers) wasprepared for the 1 g position The gravity vector of the 1 gposition was perpendicular to the surface of sample slides (119911-axis) During the unpowered transport from the laboratory tothe spacecraft and installation in the spacecraft (total time 3 h06min) the temperature was always above 21∘C Shenzhou-8launch was on October 31 2011 2158 UTC and the spacecraftattained orbit at 2208 UTC The SIMBOX timeline startedat 2234 UTC Active temperature control was set to 23∘CThe centrifuge speed for the 1 g reference centrifuge was7440 rpm Plungers 1 3 and 5 of all three EUEs wereactivated between 1205000 and 1205520 (hoursminsec)of the timeline sequences of 40 seconds Plungers 2 4 and6 of all three EUSs were activated between 1225000 and1225522 (hoursminsec) of the timeline sequences of 40seconds Human macrophage-like U937 cells were cultivatedfor 5 days inside the SIMBOX hardware on board of theShenzhou-8 spacecraft in 120583g and 1 g conditions fixed with1 PFAsucrose solution for 2 h and stored in PBS on boardat 23∘C until landing After landing the polycarbonate slideswere removed washed and then stored in PBS at 4∘C for2 weeks until analysis The ground control experiment wasexecuted analogously to the flight scenario Details about theexperiment were published previously [23]

211 Quantification of ICAM-1 by Flow Cytometry Surfaceexpression of ICAM-1 on BV-2 microglial cells and U937monocytic and macrophage-like cells as well as primaryhuman macrophages was analyzed by flow cytometry Cellswere collected from the Nutrimix bags (parabolic flight)or standardized serological pipettes (clinorotation) fixatedin PFAsucrose solution After the washing procedure (PBSwithout CaMg Biochrom) cells were stained with ICAM-1monoclonal antibody (BV2 Invitrogen FITC labeled U937and primarymacrophages cell signaling PE labeled) Analy-sis was performed using a flow cytometer (FACSCanto II BDBiosciences Heidelberg Germany) collecting at least 20000cells per sampleMean fluorescence intensity Ratio (MFI)wascalculated as MFI of sampleMFI of isotype control

212 ICAM-1 Analysis in BV-2 Cells from Parabolic FlightExperiments Cells were quadruple stained for ICAM-1apoptosis (TUNEL) cell delineation (HCS cell mask) and


DNA (DAPI) In brief cells were cytospinned onto glassslides washed 3x with PBS permeabilized for 1min with01 Triton-X 100 (Sigma-Aldrich Buchs Switzerland) andwashed again 3x with PBS and incubated with the TUNELlabeling mix (Boehringer Mannheim Germany) accordingto themanufacturerrsquos instructions For TUNEL staining rho-damine coupled dUTP was used Subsequently to overnightincubation cells were washed again 3x with PBS blockedwith 05 BSA and stained with FITC labeled ICAM-1antibody (BD Pharmingen San Jose USA) at a concentrationof 005mgmL for 2 h After additional washing cells werestained entirely with HCS cell mask deep red cytoplasmicand nuclear stain (Invitrogen Basel Switzerland) using adilution of 1 20000 and nuclei were labeled with DAPI(Invitrogen) at 1 120583gmL for 10min Labeled cells were imagedusing a Leica microscope DMI 6000 and LAS AF software(Leica Microsystems Wetzlar Germany) For automatedimaging the unified random sampling module was utilized63 randomized images of each sample were recorded and atleast 500 single cells from 3 independent experiments from3 different parabolas were analyzed From each image cellswere identified according to the following criteria nucleusof a predefined size and brightness being TUNEL negativeand containingHCS staining over a certain threshold Surfacecalculation of these cells was performed with Imaris andautomated for all images using batch coordinator (BitplaneAG Zurich Switzerland) Therefore the mean intensity ofthe ICAM-1 signal was analyzed in living cells exclusivelyand binned into ICAM-1 intensity categories of 50 graylevels Statistical analysis was carried out using GraphPadPrism software (GraphPad Software Inc La Jolla USA) andStudentrsquos 119905-test was applied for all analyzed data

213 ICAM-1 Analysis in Differentiated U937 Cells fromthe SIMBOX Experiment Polycarbonate slides were cut bya water jet method into 16 T-shaped pieces Each piecewas stained individually In order to differentiate betweendead (necroticapoptotic) and living cells before fixationslides were stained with CellMask-deep red plasma mem-brane stain (Invitrogen) andTUNEL reagent (Fluorescein-12-dUTP Roche) In addition cells were labeled with differentmono- and polyclonal primary antibodies directed againstthe cytoskeleton components and immunological relevantsurface molecules (reported in [17]) and ICAM-1 in con-centrations according to the manufacturersrsquo protocols Afterblocking with 1 BSA in PBS for 1 h primary antibodies weredetected by species specific secondary antibodies used in adilution of 1 1000 in 05 BSA in PBS Secondary antibodieswere labeled with Alexa-Fluor405 or Alexa-Fluor568 (Invit-rogen) Slide pieces were analyzed by confocal laser scanningmicroscopy (Leica SP5) Only cells positive for CellMask andnegative for TUNEL were subjected to further analysis sincethese represent the living cell population in the experimentDigital image analysis was performed using Imaris software(Bitplane)

214 RNA Isolation from the Parabolic Flight ExperimentsAfter the return of the aircraft and transport of

the samples to the on-site laboratory facilities the containerswere disassembled the Nutrimix bags were gently agitatedand the lysed cell solution from each bag was filled into aT75 straight neck cell culture flask The cell solution wasvortexed for 10 sec and passed four times through a Oslash08 times 120mm needle (B Braun Melsungen Germany) fittedto a 50mL syringe 50mL of absolute ethanol was addedand precipitates were resuspended by vigorous shakingA valve and a sterile connective piece were placed on aQiavac 24 plus vacuum system (Qiagen Germany) and anRNA maxi column (Qiagen Germany) was attached to theconnective piece A vacuum of minus200mbar was adjusted andthe column was loaded with the lysed cell suspension Thenthe valve was closed and the column was centrifuged at4000 g for 3min 15mL of buffer RW1 (Qiagen Germany)was applied for washing membrane bound RNA Aftercentrifugation at 4000 g for 7min the flow was discardedand two washing steps with 10mL RPE buffer (QiagenGermany) followed each with centrifugation at 4000 g for3min and 10min respectively The column bound RNA waseluted by application of 600 120583L of RNase-free water (QiagenGermany) incubation for 1min at room temperature andcentrifugation for 4min at 4000 g The elution step wasrepeated with the first eluate The RNA was transported atapproximately minus150∘C in a Cryo Express dry shipper (CX-100 Taylor-Wharton USA) prepared with liquid nitrogenand stored at minus80∘C until the processing of the RNA for themicroarray analysis

215 RNA Isolation during the TEXUS-49 Sounding RocketCampaign Directly after landing localization and recoveryof the payload the experiment modules were dismantledand handed over to the scientists The cell suspension wassheared three times with a 20G needle (B BraunMelsungenGermany) and distributed in two 20mL tubes 01mL ofchloroform (Sigma-Aldrich Germany) was added and thesolution was vortexed for 15 sec and incubated for 5minat room temperature before a 15min centrifugation step at11000 g and 4∘C The upper phase of both 20mL tubes wastransferred into a 15mL tube and 4mL of RLT buffer and3mL of absolute ethanol were added and mixed 4mL ofthis solution was pipetted on an RNAMidi column (QiagenGermany) and centrifuged for 30 sec at 3000 g at roomtemperatureTheflowwas discarded and the residual 4mLofRNA solution was loaded on the column and centrifuged for5min at 3000 g at room temperatureThen the columns werewashed twice with 25mL of RPE buffer and centrifuged for2min and 5min respectively at 3000 g and room tempera-ture The RNA was eluted by the addition of 250120583L RNasefree water (Qiagen Germany) to the column incubationfor 1min at room temperature and centrifugation for 3minat 3000 g and room temperature The eluate was loadedagain onto the column followed by a 1min incubation andcentrifugation for 5min at 3000 g and room temperatureThe isolated RNAwas transferred into sterile Cryo-tubes andstored until the return transport at approximately minus150∘Cin a Cryo Express dry shipper (CX-100 Taylor-WhartonUSA) preparedwith liquid nitrogen After arrival in the home


laboratory samples were stored at minus80∘C until processing theRNA for the microarray analysis

216 RNA Processing and Microarray Analysis for ParabolicFlight and TEXUS-49 Sounding Rocket Campaign SamplesRNA quantity and purity were analyzed spectrophotomet-rically using a Nanodrop 1000 (Thermo Scientific) IsolatedRNA samples were all of high quality with 260280 nmratios between 19 and 21 The RNA integrity was measuredusing an Agilent 2100 Bioanalyzer (Agilent TechnologiesUSA) Only RNA with an RNA Integrity Number (RIN) gt87 was used for the following microarray analysis 400 ngtotal RNA was applied to Cy3-labeling with the ldquoLow RNAInput Linear Amplification Kit PLUS One-Colorrdquo (AgilentTech-nologies) and hybridized for 175 h to a NimbleGenexpression microarray (12 times 135000 features) employing theldquoGene Expression Hybridization Kitrdquo (Agilent TechnologiesUSA) Afterwards arrays were washed and scanned by theMicro Array Scanner G2505B (Agilent Technologies USA)

The image files of the scanner were analyzed with theNimbleScan Software 26 using the RobustMulti-ArrayAnal-ysis (RMA) with the default parameters RMA a probe-levelsummarization method identifies probes that are outliersin the overall behavior of the expression measured for agiven gene The contribution of outlier probes is reducedin the reported gene expression level which has beendemonstrated to improve the sensitivity and reproducibilityof microarray results In addition to screening outlier probesNimbleScan softwarersquos implementation of RMAused quantilenormalization and background correction The normalizedmicroarray data were analyzed using Partek Genomics Suite66 Statistical analysis was performed using the one-wayANOVA and the false discovery rate (FDR) for multipletesting corrections Further the coefficient of variation (CV)expressed in percent was calculated also known as ldquorelativevariabilityrdquo which equals the standard deviation divided bythe mean Genes of interest were identified and the log2values of the measured fluorescent intensities returned bythe Partek software were back calculated to linear valuesThen means of all values of the same gene generated bydifferent probes were calculated if at least three values existedexcluding outliers Subsequently standard deviations werecalculated for the means and an unpaired 119905-test with Welchcorrection was performed to test statistical significance

217 Pathway Enrichment Analysis The pathway enrichmentanalysis was performed using Partek Genomics Suite 66 andthe KEGG human pathway library [24 25]The119875 values werecalculated by the Fisher exact test Enrichment analysis wasapplied on the genes showing differential expression with 119875values of lt005 and fold change gt+15 or lt minus15

218 Statistical Analysis Data are expressed as median oras median plusmn SE Groups contain the analysis of 200ndash1000cells (SIMBOX shown in box-plots) or data of three inde-pendent experiments with 1ndash5 samples (119899 = 3ndash15 shown incolumns) Data were analyzed by one-way ANOVA followedby Wilcoxon or unpaired 119905-test using GraphPad Prism 5

lowast119875 lt 01 was considered to be significant lowastlowast119875 lt 005 assignificant and lowastlowastlowast119875 lt 001 as very significant

3 Results

31 Clinorotation of Downregulated ICAM-1 Expression inBV-2 Microglial Cells First we analyzed ICAM-1 expressionin BV-2 microglial cells after 24 h clinorotation (60 rpm4mm pipette diameter maximal residual acceleration of 4 times10minus3 g at the outer radius of the pipette) The clinostat devicewas placed in an incubator which provides constant 37∘CFifteen serological pipettes rotated at the same time with60 rpm 1 g controls were placed at the ground plate of theclinostat without rotation but with the same environmentcondition like 120583g samples A 1 g control group of BV-2 cellswas filled into 1mL serological pipettes in the same way asthe clinorotation cell group but was not clinorotated Anothercontrol group was kept at regular cell culture conditionsin the incubator (37∘C 5 CO

2) Cells were subsequently

stained for cell surface ICAM-1 apoptosis (TUNEL) celldelineation (HCS CellMask) and DNA (DAPI) (Figure 2)ICAM-1 expression analysis by flow cytometry revealed twodistinct subtypes of cells in the clinorotated group (120583ggroup) compared to the 1 g control group and the incubatorcontrol group consisting of only one subtype respectively(Figure 2(a)) The first of the two subtypes was small andstronger granulated (subtype 1) than the second subtypewhich appears taller but less granulated (subtype 2) Apop-totic cells were excluded from the analysis by TUNELstaining Subtype 1 could possibly represent an activated stateSubtype 1 was found in the 120583g group as well as in the 1 gcontrol group whereas the incubator control did virtuallynot contain this subtype Subtype 2 was represented in allthree cell groups 120583g 1 g control and incubator control cellgroup However it was primarily present in the 120583g andin the incubator control group and less present in the 1 gcontrol groupThe population distribution within cell groupsis illustrated in Figure 2(b) showing the relative cell numbersof each population in each cell group Since the incubatorcontrol group consisted almost exclusively of cells in subtype2 this number was nearly 100 whereas subtype 1 was closeto 0 The 120583g group had almost as many cells in subtype2 as in subtype 1 with a slight predominance in subtype 2In Figure 2(c) the mean fluorescence intensity of the cellsubtypes in the different cell groups was depicted Whilethe ICAM-1 expression in the incubator control group wasstable in both subtypes (2158 plusmn 2344 RFU versus 2082 plusmn171 RFU) and the 120583g cell group displayed significantly lessexpression of ICAM-1 in subtype 1 compared to subtype 2ICAM-1 expression was significantly reduced in the 120583g groupcompared to the 1 g control and the incubator control groupCells in the 1 g control group exhibited a similar ICAM-1expression distribution as cells from the 120583g group The meanfluorescence intensities between subtype 2 of different groupsdid not change dramatically except a significant differencebetween the 1 g control group and the incubator controlgroup In summary we suppose that ICAM-1 expression wasdownregulated in microglia cells in simulated microgravity


Incubator control group

Subtype 1Subtype 1

Subtype 2

Subtype 1

Subtype 2 Subtype 2

0

0

SSC-

A

FSC-A

120583g group 1g control group250K

200K

200K

150K

150K

100K

100K

50K

0

SSC-

A

250K

200K

150K

100K

50K

0

SSC-

A

250K

200K

150K

100K

50K

50K 250K 0

FSC-A200K150K100K50K 250K 0

FSC-A200K150K100K50K 250K

(a)

200

150

100

50Cel

ls in

pop

ulat

ion

()

120583g groupIncubator control group

lowastlowast

lowastlowast

lowastlowast

lowastlowast

1 2

Subtype

1g control group

(b)

5000

4000

3000

2000

1000Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

lowastlowast

lowastlowast

lowastlowastlowast

lowastlowastlowast

lowastlowast

120583g group 1g control groupIncubator control group

All 1 2

Subtype

(c)

Figure 2 Cytometry analysis of ICAM-1 expression in BV-2 microglial cells in simulated microgravity (2D clinorotation) BV-2 microglialcells were exposed to either clinorotation (120583g) placed in the clinostat but not rotated (1 g control group) or cultured under standard cellculture conditions (incubator control) for 24 h Cells were stained for ICAM-1 surface expression and analyzed by flow cytometry The levelof ICAM-1 surface expression is represented by the mean fluorescent intensity assessed by flow cytometry (a) In forwardsideward scatterdetection mode of flow cytometry two gates were set to separate two subtypes of BV-2 microglial cells that appeared different in size andgranulation (subtypes 1 and 2 in dot plots) (b) Distribution of BV-2 microglial cells in subtypes 1 and 2 after exposure to different gravityconditions (c) Quantification of ICAM-1 expression after exposure to different gravity conditions within subtypes 1 and 2 Data are given asmedian plusmn SE (lowast119875 lt 01 lowastlowast119875 lt 005 lowastlowastlowast119875 lt 001 119899 = 3 according to one-way ANOVA followed by Wilcoxon or unpaired 119905-test)

32 Rapid and Reversible Downregulation of ICAM-1 SurfaceExpression in BV-2 Microglial Cells in Real Microgravity Inthe next step we investigated the cell surface expression ofICAM-1 in real microgravity provided by parabolic flightsin murine BV-2 microglial cells During parabolic flightexperiments cells were activated at the onset of 120583g (or during1 g for in-flight control experiments) by the addition of PMAor TNF-120572 or not activated by the addition of medium onlyAfter a 20 sec period of altered gravity cells were fixed by theaddition of formaldehyde

During the 13th DLR parabolic flight campaign we alsoaddressed the issue that during parabolic flight experi-ments cells are generally subjected to irregular stress bycell preparation and handling and by the in-flight situationitself This combination of interference factors always leadsto a significant degree of damaged or dead cells whichcould affect the experiment results and mask a possiblemicrogravity-related effect even under presence of internal

controls For this reason we developed an automated analysismethod which allows for the specific analysis of alive andmorphologically intact cells at the moment of fixation

Experiments from different parabolas (1 g and 120583g resp)and different flights were analyzed The experiments wereperformed in a sequence of three consecutive 120583g and 1 gphases A quadruple fluorescent staining was performedusing TUNEL (rhodamine) for detection of apoptotic cellsDAPI for the nuclei high content screening (HCS) CellMaskdeep red for the delineation of cells and FITC-labeled antiICAM-1 antibody for identification of cell surface expressionof ICAM-1 Cells were imaged with a widefield microscope(Leica Microsystems Wetzlar Germany) using the uniformrandom sampling module and identified by an iso-surfacecalculation (Imaris Bitplane AG Zurich Switzerland) Thisquadruple staining allowed the exclusion of apoptotic cellsin a highly reliable fashion An example of an apoptotic celland a living cell is depicted in Figure 3(a)Themean intensity


(A) (B) (C) (D) (E) (F)

(a)

Con100

80

60

40

20

0

0 250 500 750 1000

Rela

tive f

requ

ency

MIF of ICAM-1

(b)

lowastlowastlowast

1600

1400

1200

1000

800

600

400

200

0

ICA

M-1

inte

nsity

120583g 1g

(c)

lowastlowastlowast

120583g 1g

700

600

500

400

300

200

100

0

MIF

ICA

M-1

(d)

PMA100

80

60

40

20

0

0 250 500 750 1000

Rela

tive f

requ

ency

MIF of ICAM-1

(e)

lowastlowastlowast

120583g 1g

1000

800

600

400

200

0

ICA

M-1

inte

nsity

(f)

lowastlowastlowast

120583g 1g

400

300

200

100

0M

IF IC

AM

-1

(g)

TNF-120572100

80

60

40

20

0

0 250 500 750 1000

Rela

tive f

requ

ency

MIF of ICAM-1

120583g1g

(h)

120583g 1g TNF-120572

1000

800

600

400

200

0

ICA

M-1

inte

nsity

(i)

120583g 1g TNF-120572

300

200

100

0

MIF

ICA

M-1

(j)

Figure 3 ICAM-1 surface expression reacts rapidly and reversibly to microgravity (a) Microscopy of ICAM-1 TUNEL HCS CellMask andDAPI including surface calculation for HCS In order to identify nuclei cells were stained with DAPI (A) Apoptotic cells were identifiedby TUNEL reaction (B) and HCS CellMask label (C) which can be retained to a higher extend in nonapoptotic cells ICAM-1 intensity isdepicted in (D) A merge of TUNEL DAPI and ICAM-1 (E) shows an apoptotic cell (998819) and a living cell (rarr ) The automated calculation ofan iso-surface is exclusively done for living cells using the HCS CellMask channel as shown in the merge with TUNEL DAPI and ICAM-1(F) (b)ndash(j) BV-2 microglial cells were treated with PMA ((e) (f) and (g)) or TNF-120572 ((h) (i) and (j)) at the onset of microgravity or duringthe 1 g in-flight control phase or left untreated ((b) (c) and (d)) Cells were fixed in flight after 20 sec normogravity (1 g) (-e-) or 20 secmicrogravity (120583g) (-I-) Cells were stained imaged and analyzed as described above The mean intensity of the ICAM-1 signal was binnedintomean intensity fluorescence (MIF) categories and the number of cells (frequency) is plotted against these intensity categories ((b) (e) and(h)) ICAM-1 fluorescence intensity of all analyzed cells ((c) (f) and (i)) is depicted for normogravity (triangles) and microgravity (squares)Mean ICAM-1 fluorescence intensity of all analyzed cells ((d) (g) and (j)) was pooled for normogravity (black bar) and microgravity (openbar) For automated imaging the unified random sampling module was utilized and 63 randomized images of each sample were recordedand at least 500 single cells from 3 independent experiments from 3 different parabolas were analyzed Mean intensity and SEM are shownand studentrsquos 119905-test showed highly significant difference of the fluorescence values of lowastlowastlowast119875 lt 00001 119899 = 3


120583g group1g control group

3000

2000

1000

Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

lowastlowastlowast

lowastlowastlowast

lowastlowastlowast

lowastlowast

1 3 5

Time (d)

Differentiated U937 cells

(a)


3000

2000

1000Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

lowastlowastlowast

lowast

lowast

lowastlowast

1 3 5

Time (d)

Primary macrophages

(b)

Figure 4 Cytometry analysis of ICAM-1 expression in macrophage-like differentiated U937 cells or primary human macrophages insimulated microgravity (2D clinorotation) Macrophage-like differentiated U937 cells (a) or primary human macrophages (b) were exposedto either clinorotation (120583g) placed in the clinostat but not rotated (1 g control group) or cultured under standard cell culture conditions(incubator control) Cells were stained for ICAM-1 surface expression and analyzed by flow cytometryThe level of ICAM-1 surface expressionis represented by the mean fluorescent intensity assessed by flow cytometry (a) Quantification of ICAM-1 expression in macrophage-likedifferentiated U937 cells after exposure to different gravity conditions for 1 h 3 h or 5 h 119899 = 6 (b) Quantification of ICAM-1 expression inprimary humanmacrophages after exposure to different gravity conditions for 1 h 3 h or 5 h 119899 = 3 Data are given as median plusmn SE (lowast119875 lt 01lowastlowast119875 lt 005 lowastlowastlowast119875 lt 001 according to one-way ANOVA followed by Wilcoxon or unpaired 119905-test)

of the ICAM-1 signal was analyzed in nondamaged andnonapoptotic cells only and binned into intensity categoriesThe relative frequency of these cells was plotted against thefluorescence intensity (Figures 3(b) 3(e) and 3(h))

We found a rapid and reversible downregulation ofICAM-1 on the surface of BV-2 microglial cells after 20 secof microgravity apparent by the frequency of cells expressingICAM-1 in various intensities (Figures 3(b) and 3(c)) andthe mean of ICAM-1 expression intensity (Figure 3(d)) beingonly 70 in microgravity compared to normogravity In thepresence of PMA ICAM-1 expression was upregulated (Fig-ures 3(e) 3(f) and 3(g)) whereas the presence of the proin-flammatory cytokine TNF-120572 abrogated the microgravity-induced ICAM-1 downregulation (Figures 3(h) 3(i) and3(j)) Statistical analysis of all pooled data revealed downregu-lation of ICAM-1 expression in unstimulated microglia uponmicrogravity to be highly significantly different with 119875 lt0001 (Figure 3(d)) Changes of ICAM-1 expression in PMAstimulated cells were as well highly significant (Figure 3(g))whereas TNF-120572 stimulation slightly ameliorated the gravity-dependent changes in ICAM-1 expression (Figure 3(j))Thuswe found a rapid and reversible disappearance of ICAM-1protein from the cell surface in microgravity

33 Increase of ICAM-1 Expression in U937 HumanMacrophage-Like Cells and Human Primary Macrophagesin Simulated Microgravity To corroborate the relevanceof the results obtained with murine BV-2 microglial cellswe investigated a human macrophage-like cell systemTherefore human monocytic U937 cells were differentiated

into macrophage-like cells [23] and humanM2 macrophageswere differentiated from blood mononuclear cells Beforethe experiment differentiated macrophage-like cells weredetached resuspended in fresh medium and filled into1mL standardized serological pipettes for the clinostatClinorotation was performed for 1 d 3 d and 5 d The 1 gcontrol group of differentiated U937 cells was filled into 1mLserological pipettes in the same way as the clinorotation cellgroup but was not rotated Cells were subsequently fixed andstained for cell surface ICAM-1 and apoptosis (TUNEL) toexclude apoptotic cells from the analysis and subjected toflow cytometry We detected a highly significant increaseof ICAM-1 expression in the clinorotated cells (120583g group)compared to the nonrotated cells (1 g control group) after 1 dand 5 d in differentiated U937 cells (Figure 4(a)) and primarymacrophages (Figure 4(b)) However this increase recededafter 3 and 5 days of clinorotation Therefore we concludethat ICAM-1 expression is increased in human macrophagesafter 1 and 5 days of simulated microgravity

34 Increased ICAM-1 Expression in Differentiated U937Cells in Real Microgravity during Parabolic Flight Duringparabolic flight experiments we investigated rapid effectsof real microgravity on nondifferentiated and differenti-ated U937 cells and on primary human M2-differentiatedmacrophages Nondifferentiated and differentiated myelom-onocytic U937 cells were cultured and seeded into Nutrimixbags as described During the parabolicmaneuvers cells wereactivated at the onset of 120583g or during 1 g for in-flight controlexperiments by the addition of PMA with medium in the


case of nondifferentiated U937 cells or not activated by theaddition of medium only in the case of differentiated U937cells and primary macrophages After 20 sec microgravitycells were fixed by the addition of paraformaldehyde A groupof ground control cells was left in Nutrimix bags in thelaboratory incubator and activated and fixed after landing inthe same experimental equipment Experiments from differ-ent parabolas (1 g and 120583g resp) and different flights wereanalyzed A quadruple fluorescent staining was performedusing TUNEL (rhodamine) for detection of apoptotic cellsDAPI for the nuclei high content screening (HCS) CellMaskdeep red for the delineation of cells and FITC-labeled antiICAM-1 antibody for identification of cell surface expressionof ICAM-1

Nondifferentiated U937 Cells Differentiation of U937 mono-cytic cells into macrophage-like cells significantly increasedthe cell surface expression of ICAM-1 (Figure 5(a)) Non-differentiated U937 did not demonstrate differential expres-sion of ICAM-1 in microgravity neither in PMA-stimulatedmyelomonocytic U937 cells nor in non-stimulated cellsany significant alteration of ICAM-1 expression could bedetected in comparison between microgravity and 1 g condi-tions (Figure 5(b)) The only significant difference could beobserved in nonstimulated U937 cells between the groundcontrol group the 120583g group and the 1 g control groupDifferences between 1 g ground and 1 g in-flight controls canbe attributed to the flight itself (eg vibrations handling ofcell containers) and not to an altered gravity

Differentiated U937 Cells In contrast to nondifferentiatedU937 cells macrophage-like U937 cells displayed a highlysignificant gravity-dependent change in ICAM-1 expression(Figure 5(c)) In flight cell surface ICAM-1 was reduced dras-tically compared to the ground control In the microgravitygroup ICAM-1 expression was enhanced This finding isconsistent with our experiments in simulated microgravityWe suppose that ICAM-1 is upregulated in differentiatedmacrophage-like cells in microgravity

PrimaryMacrophages For the analysis of primary humanM2macrophages double fluorescence staining was performedusing TUNEL (rhodamine) for detection of apoptotic cellsand FITC-labeled anti ICAM-1 antibody for identificationof cell surface expression of ICAM-1 Unfortunately the 1 gincubator control was lost during the experiment proce-dures Between the 1 g in-flight control and the microgravitygroup no differences in ICAM-1 surface expression couldbe detected in primary human macrophages However dueto the technical problems and low detected expression levelscompared to primary macrophages in clinostat experiments(see Figure 4(b)) the informative value of these results maybe limited and it is planned to repeat the parabolic flightexperiment with primary human macrophages

35 Increased ICAM-1 Expression in Differentiated U937 Cellsduring Long-Term Microgravity in the SIMBOX ExperimentDuring the SIMBOX (Science in Microgravity Box) mis-sion on Shenzhou-8 we investigatedmicrogravity-associated

long-term alterations inmacrophage-like differentiated U937cells and analyzed the effect of long-term microgravityon the cytoskeleton and immunologically relevant surfacemolecules [23] Human U937 cells were differentiated intoa macrophage-like phenotype and exposed to micrograv-ity or 1 g on a reference centrifuge on orbit for 5 daysThe unmanned Shenzhou-8 spacecraft was launched with aLong March 2F (CZ-2F) rocket from the Jiuquan SatelliteLaunch Center (JSLC) and landed after a 17-day missionAfter on-orbit fixation the samples were analyzed withimmunocytochemical staining and confocalmicroscopy afterlanding Double fluorescent staining was performed usingHCS CellMask deep red for the delineation of cells andFITC-labeled anti ICAM-1 antibody for identification ofcell surface expression of ICAM-1 Cells were analyzed asdescribed above We detected a significant higher expressionof ICAM-1 in long-term microgravity in comparison tothe in-flight 1 g control group (Figure 6) Similar to theparabolic flight experiments incubation of the macrophage-like differentiated U937 cells in the experiment hardwarecaused a significant downregulation of ICAM-1 expressionThus it can be excluded that the microgravity effects onICAM-1 were caused by the experiment system itself

36 No Influence of Altered Gravity on ICAM-1 mRNA RNAsamples were analyzed for their quantity and quality and fur-ther processed for the microarray hybridization on 12 times 135 KRoche NimbleGen arrays Data from 46 single microarrays(19th DLR PFC 8x 120583g 6x HW 8x 1 g and 6x 18 g TEXUS-49 7x120583g 6xHW and 5xBL)were collected normalized andfurther analyzed The data tables were screened for ICAM-1values and mean fluorescence intensities including standarddeviations were calculated for all samples of one conditionICAM-1 shows stable expression for all gravity conditionsduring the 19th DLR PFC and the TEXUS-49 campaignas well as for the HW controls (Figure 7) indicating thatmicrogravity and hypergravity conditions did not have aninfluence on mRNA ICAM-1 level in the range of 20 secondsuntil 6 minutes

37 PathwayAnalysis Reveals an Influence of RealMicrogravityon the Natural Killer Cell Mediated Cytotoxicity of MonocyticU937 Cells Due to the wealth of data microarray analysisprovides we were able to perform a GeneSet enrichmentanalysis to identify any affected pathways or biologicalnetworks in connection with ICAM-1 (see Supplement 1 inthe Supplementary Material available online at httpdxdoiorg1011552015538786) For the experiments performedduring the 19th DLR PFC with monocytic U937 cellswe identified one significantly influenced ICAM-1 relatedpathway during 20 sec of microgravity compared to thein-flight 1 g control namely the natural killer cell mediatedcytotoxicity (enrichment 119875 value 00203328) For theexperiments performed on TEXUS-49 during 6min ofmicrogravity with monocytic U937 cells we found twoweakly altered pathways Specifically the NF-kappa Bsignaling pathway (enrichment 119875 value 00632651) andthe Epstein-Barr virus infection (enrichment 119875 value


8000

6000

4000

2000

Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

Mon U937 Mac U937


lowastlowastlowast

(a)


4000

3000

2000

1000

Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

+minus +minus +minus

PMA (10120583M)

lowastlowastlowast

lowastlowastlowast

U937 cells

(b)


8000

6000

4000

Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

lowastlowastlowast

lowastlowastlowast

lowastlowastlowast


(c)


Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

500

400

300

200

100

Primary macrophages

(d)

Figure 5 ICAM-1 expression in U937 cells macrophage-like differentiated U937 cells and primary human macrophages in different gravityconditions during parabolic flight experiment ICAM-1 expression was assessed by flow cytometry and fluorescent microscopy followingimmunocytochemical staining Cells were cultured under standard cell culture conditions (incubator control) or exposed to different gravityconditions during the 19th DLR parabolic flight campaign U937 cells were fixed either after PMA-activation in microgravity (120583g group) orin 1 g (1 g control group) Differentiated U937 and primary macrophages were fixed after the microgravity phases (120583g group) or after the 1 gphases before and after the120583g phase (1 g control group)The level of ICAM-1 surface expression is represented by themeanfluorescent intensityassessed by flow cytometry (a) ICAM-1 surface expression in myelomonocytic U937 cells (mon U937) and macrophage-like differentiatedU937 cells (max U937) under standard cell culture conditions (b) ICAM-1 surface expression in U937 cells with and without activationby PMA in different gravity conditions (c) ICAM-1 surface expression in macrophage-like differentiated U937 cells in different gravityconditions (d) ICAM-1 surface expression in primary macrophages in different gravity conditions Data are given as median plusmn SE (lowast119875 lt 01lowastlowast119875 lt 005 and lowastlowastlowast119875 lt 001 according to one-way ANOVA followed by Wilcoxon or unpaired 119905-test)

00641782) appeared sensitive to microgravity compared tobaseline

4 Discussion

In our study we investigated the surface expression ofICAM-1 protein and expression of ICAM-1 mRNA in cells

of the monocytemacrophage system in microgravity dur-ing clinostat parabolic flight sounding rocket and orbitalexperiments In murine BV-2 microglial cells we founda downregulation of ICAM-1 expression in clinorotationexperiments and a rapid and reversible downregulation inthe microgravity phase of parabolic flight experiments Incontrast ICAM-1 expression increased in macrophage-like



1000

800

600

400

200

Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

lowastlowastlowast

lowastlowastlowast

(a)

CD54

conOverlay

10120583m

10120583m 10120583m 10120583m

10120583m10120583m

1g 120583g

(b)

Figure 6 ICAM-1 expression in macrophage-like differentiated U937 cells after long-term exposure to microgravity during theSIMBOXShenzhou-8 mission Cells were cultured under standard cell culture conditions (incubator control) or exposed to different gravityconditions during the SIMBOXShenzhou-8 mission Differentiated U937 cells were fixed in microgravity (120583g group) or in 1 g (1 g controlgroup) after 5 days Only CellMask-positive and TUNEL-negative cells were analyzed (a) Each group represents analysis of the meanfluorescence of 200ndash1000 individual cells from one recovered slide Data are expressed as the median of mean single cell fluorescenceintensities with the smallest observation (sample minimum) lower quartile median upper quartile and largest observation (samplemaximum) Statistical analysis was performed with GraphPad Prism 5 Wilcoxon test lowast119875 lt 005 lowastlowast119875 lt 001 and lowastlowastlowast119875 lt 0001 (b) Standardcell culture control (con) 1 g hardware control (1 g) and the microgravity sample (120583g)

10000

8000

6000

4000

2000

0

Fluo

resc

ence

inte

nsity

19th DLR PFC ICAM-1

HW 1g 18 g 120583g

(a)

TEXUS-49 ICAM-110000

8000

6000

4000

2000

0

HW BL 120583g

Fluo

resc

ence

inte

nsity

(b)

Figure 7 Influence of altered gravity during parabolic flight and sounding rocket flight on ICAM-1 mRNA expression levels (a) ICAM-1mRNA expression levels are demonstrated for samples of the 19th DLR parabolic flight campaign after 1 g (light gray) 18 g (dark gray) 120583g(black) and hardware ground controls (HW striped) exposure and (b) for samples of the TEXUS-49 campaign after launch and acceleration(BL dark gray) 120583g (black) and hardware ground controls (HW striped) ICAM-1 fluorescence intensities do not show any significantdifferences for all compared conditions in both experimental setups The number of analyzed arrays 19th DLR PFC 1 g (119899 = 8) 18 g (119899 = 6)120583g (119899 = 8) and HW (119899 = 6) TEXUS-49 HW (119899 = 6) 120583g (119899 = 7) BL (119899 = 5)

differentiated human U937 cells during the micrograv-ity phase of parabolic flights and in long-term micro-gravity provided by a 2D clinostat or during the orbitalSIMBOXShenzhou-8 mission

In nondifferentiated U937 cells no effect of microgravityon ICAM-1 expression could be observed during parabolicflight experiments A summarizing table which presents anoverview about the cell types tested the platforms usedthe experiment durations analysis method the number ofexperiments and the detected effects on ICAM-1 is demon-strated in Table 1 In our study and according to previous

investigations [26] we detected effects of the experimentalhardware whichwere controlled by the appropriate hardwarecontrol experiments

In clinostat experiments subtype of BV-2 microglial cellsappeared in the FACS analysis of all clinostat samples (120583g or1 g controls) but not at all in the incubator controlsThis sub-type consisted of smaller cells and we suppose an ldquoactivatedrdquophenotype of BV-2 microglial cells It is well established thatmicroglial form and function are linked and that cells cancycle reversibly from a simple rounded (activated and amoe-boid) to a complex branched form (ramified and resting) [27]


Table1Re

gulatio

nofICAM-1in

cells

ofthem

onocytemacroph

agesystem

inmicrogravityO

verviewabou

tcelltypesm

icrogravity

platform

sthee

xperim

entd

urationsanalysis

metho

ds

numbero

fexp

erim

ents

andeffectson

ICAM-1

Experim

ental

platform

Celltypes

Experim

entalgroup

sAnalysis

metho

dNum

bero

freplicates

Timeo

f120583gexpo

sal

Regu

latio

nof

ICAM-1

120583g1g

1gin

cubatorc

ontro

l

Clinorotation

BV2

microgliacells

120583ggrou

p1g

controlgroup

FACS

(2000

0events

sample)

119873=3

24h

darr

Prim

ary

macroph

ages

120583ggrou

p1g

controlgroup

FACS

(5000

events

sample)

119873=3

1ndash5d

uarr

U937

macroph

ages

120583ggrou

p1g

controlgroup

FACS

(1000

0events

sample)

119873=6

1ndash5d

uarr

Parabo

licflight

BV2

microgliacells

120583ggrou

p1g

controlgroup

Con

focalm

icroscop

yrand

omsamplingmod

ule

(gt500cellssam

ple)

119873=3(w

o)

uarr

119873=3(w

ithPM

A)

20sec

darr

119873=3(w

ithTN

F-a)

mdash

U937

macroph

ages

120583ggrou

p1g

controlgroup

Groun

dcontrol

FACS

(1000

0events

sample)

119873=9

20sec

uarrdarr

119873=9

119873=6

U937

mon

ocytes

120583ggrou

p1g

controlgroup

Groun

dcontrol

FACS

(1000

0events

sample)

119873=9(w

oPM

A)

20sec

mdashmdash

119873=5(w

ithPM

A)

119873=12

(wo

PMA)

119873=6(w

ithPM

A)

119873=2(w

oPM

A)

119873=1(with

PMA)

120583ggrou

p1g

controlgroup

18gcontrolgroup

Hardw

arec

ontro

l

Microarray

119873=8

20sec

mdashmdash

119873=8

119873=6

119873=6

Prim

ary

macroph

ages

120583ggrou

p1g

controlgroup

FACS

(1000

0events

sample)

119873=9

20sec

mdash119873

=9

SIMBO

XShenzhou

8U937

Macroph

ages

120583ggrou

p1g

controlgroup

Incubatorc

ontro

l

Con

focalm

icroscop

y(300

cellssam

ple)

(650

cellssam

ple)

(100

0cellssam

ple)

119873=1

5duarr

darr

TEXU

S-49

U937

mon

ocytes

120583ggrou

pbaselin

egroup

Hardw

arec

ontro

lMicroarray

119873=7

10min

mdashmdash

119873=5

119873=6


Thus we assume that microgravity activates microglias cellsbut downregulates ICAM-1 expression (Figure 2) Controlexperiments revealed no influence of the serological pipetteincubation system on the ICAM-1 expression compared toldquonormalrdquo cell culture conditions between 1 and 5 d (data notshown)

In this study we also developed a method that alloweda randomized screening of only those cells that were aliveat the fixation time point after the parabola (Figure 3)This is of particular importance because of the damagecaused to all cells subjected to a flight experiment Untilnow a method for analyzing only viable and nondamagedcells obtained from flight experiments was lacking Dueto the very limited number of samples during the exper-iment with BV-2 cells during the 13th DLR PFC FACSanalysis could not be utilized We therefore developed amicroscopy based method to analyze exclusively the livingcell portion Samples were imaged using the uniform ran-dom sampling module of Leica LAS AF software in orderto fulfill all statistically necessary criteria of randomizedsampling Surface calculation of cells negative for TUNELlabel under a certain threshold and positive for HCS Cell-Mask allowed the exclusion of all apoptotic cells The meanICAM-1 intensity value of each analyzed cell was taken intoaccount

Modulation of the expression of surface adhesionmolecules such as ICAM-1 has been reported as the conse-quence of long-term microgravity [28 29] In our study wefound that ICAM-1 surface expression responds to gravitychanges in BV-2 microglial cells within 20 seconds Therapid and reversible changes of ICAM-1 on the cell surfacesuggest a direct gravity-sensitive effect on the membranecompartment or on protein folding whereas transcriptionalor proteolytic processes are rather unlikely as they wouldbe too slow Interestingly ICAM-1 cell surface expressionin microgravity was upregulated in macrophage-like dif-ferentiated human U937 cells (Figures 4 5 and 7) butdownregulated in murine BV-2 microglial cells (Figure 2) Inprimary humanmacrophages no clear conclusion is possiblebecause of the very low fluorescence levels in the analysisof parabolic flight samples (Figure 5) However the clinostatexperiments with primary human macrophages (Figure 4)suggest an upregulation of ICAM-1 in microgravity Thedifferent ICAM-1 regulation between macrophage-like dif-ferentiated human U937 and murine BV-2 microglial cellsin microgravity could be the consequence of the differentspecies (murine and human) or differentmolecular and func-tional features of peripheral macrophages and CNS macro-phages

In our study we detected no effect of microgravity onICAM-1 mRNA expression neither in a parabolic flightexperiment nor during the sounding rocket experiment(Figure 7) In a previous study also no effect of simulatedmicrogravity on ICAM-1 mRNA expression in endothelialcells could be found [15] However performing pathwayanalyses on ICAM-1 related pathways we identified thenatural killer cell mediated cytotoxicity being influencedsignificantly after 20 sec of microgravity After 6min ofmicrogravity this effect appeared to be reversed and we

found the NF-kappa B signaling pathway and the Epstein-Barr virus infection close to significant alteration Thisobservation is in line with findings in astronauts after long-term space missions where latent viruses persisting in adormant state after primary infection were reactivated [3031] Therefore we hypothesize that these two pathways maybe stronger affected over a longer period of microgravity Aswe were not able to find an influence on the natural killercell mediated cytotoxicity after 6min of microgravity wesuppose this is one of the short-term reversible processesthat can recover after an adaptation phase to micrograv-ity

Related to the regulation of surface ICAM-1 expressioninternalization and receptor recycling of ICAM-1 are highlydynamic processes [32 33] and linked to cytoskeletal function[34 35]

Multiple investigators have reported that this complexnetwork of fibers is sensitive to environmental factors suchas microgravity and altered gravitational forces [36ndash38]Several studies demonstrated modifications of the actin andmicrotubule cytoskeleton in real and simulated microgravityin lymphocytes astrocytes neurons glial cells mesenchymalstem cells and thyroid carcinoma cells [36ndash41] Morphologi-cal differences of both themicrotubule and actin componentsof the cytoskeleton have been observed in cells grown inreal and simulatedmicrogravity [39ndash43] During space flightactin reorganization in response to the gravity level andabnormal assembly of actin stress fibers has been reported[44ndash46]

We conclude that disturbed immune function in micro-gravity could be a consequence of ICAM-1 modulation inthe monocytemacrophage system which in turn could havea strong impact on the cellsrsquo interaction with T lympho-cytes and migration An experiment under real microgravityconditions on board of the ISS was conducted by Italianand Swiss investigators to test the hypothesis that lack ofinteraction might be the reason for the loss of activity ofT cells in microgravity [14] The investigation consistedof analyzing the cap formation of the adhesion proteinsLFA-1 on T cells and ICAM-1 on monocytes The datashowed that LFA-1ICAM-1 interactions occur in space butare dependent on activation time they show differences innumber arrangement and fluorescence intensityThus LFA-1 and ICAM-1 adhesion proteins seem to be sensitive to realmicrogravity without being altered in their interaction Lossof functional ICAM-1 in the brain-resident microglial cellsbears the risk of a significant impairment of the CNS immunesystem Indeed reactivation and shed of varicella-zoster virus(VZV) have been reported in astronauts [30 31] a viruswhich becomes latent in the nervous system after primaryinfection but is reactivated frequently in immune suppressedindividuals

In conclusion we found that ICAM-1 can be downreg-ulated rapidly and reversibly in BV-2 microglial cells andupregulated in macrophage-like differentiated U937 cells inresponse to microgravity In both cell types long-term effectsup to several days could be detected Thus ICAM-1 canbe considered as a rapid-reacting and sustained gravity-regulated molecule in mammalian cells


Abbreviations

BL BaselineCC Cell culture controlDLR German Aerospace CenterESA European Space AgencyESRANGE European Space and Sounding Rocket RangeEUE Experiment Unique EquipmentFACS Fluorescence activated cell sortingFCS Fetal calf serumGC Ground controlHW HardwareICAM-1 Intercellular adhesion molecule 1JSLC Jiuquan Satellite Launch CenterLFA-1 Lymphocyte function-associated antigen 1MAC-1 Myelomonocytic leukocyte integrin CD11b120583g MicrogravityMMS Monocyte-macrophage-systemPBMC Peripheral blood mononuclear cellPEEK Polyether ether ketonePFA ParaformaldehydePFC Parabolic flight campaignPITC Payload integration and test centerPMA 12-O-Tetradecanoylphorbol-13-acetateRCCS Rotary Cell Culture SystemsRFI Relative fluorescence intensityRIN RNA integrity numberRPM Random positioning machineSSC Swedish Space CorporationSIMBOX Science in Microgravity BoxTUNEL TdT-mediated dUTP-biotin nick end labeling


The authors declare that they have no competing interests

Authorsrsquo Contribution

Oliver Ullrich developed the study idea concept and theoverall study design in addition to planning coordinatingand supervising the study Katrin Paulsen and Oliver Ullrichwrote the paper Svantje Tauber Cora S Thiel and Liliana ELayer contributed to the paper Oliver Ullrich Cora S ThielGesine Bradacs Sonja Krammer Josefine Biskup GabrielaRieder Lisa Mutschler Daniel Hofmanner Isabell Buttronand Hartwin Lier performed the experiments during the13th DLR parabolic flight campaign Liliana E Layer CoraS Thiel Oliver Ullrich Svantje Tauber Swantje HauschildClaudia Philpot Annett Gutewort Eva Hurlimann Jose-fine Biskup and Hartwin Lier performed the experimentsduring the 19th DLR parabolic flight campaign AndreasHuge performed the pathway analysis Svantje Tauber CoraS Thiel and Oliver Ullrich performed the experimentsduring the TEXUS-49 mission and Katrin Paulsen SvantjeTauber Dana M Simmet Oliver Ullrich Eva Hurlimannand Nadine Golz during the SIMBOXShenzhou-8 missionKatrin Paulsen was responsible for the sample analysis fromthe 13th DLR and 19th DLR parabolic flight campaigns andthe SIMBOXShenzhou-8 mission and Cora S Thiel was

responsible for sample analysis from the TEXUS-49 missionNadine Golz and Svantje Tauber contributed to the sampleanalysis Frank Engelmann contributed to and supervised thetechnical procedures during the 13th and 19th DLR parabolicflight campaigns

Acknowledgments

The authors gratefully acknowledge financial support by theGerman Aerospace Center DLR (Grant nos 50WB0912 and50WB1219) They also gratefully acknowledge the support of(in alphabetic order) Markus Braun Miriam Christen Gio-vanni Colacicco Ulrike Friedrich Andre Hilliger AndreasHuge Schirin Ibrahim Otfried Joop Andre Melik ShirinMilani Brice Moutett Marianne Ott Irina Rau Frank RuhliChen Sang Burkhard Schmitz Andreas Schutte JohannaStahn Marc Studer Susanne Wolf and Fengyuan ZhuangThey would like to thank the Swiss Air Force for theoutstanding support of their study by providing daily airtransports of in-flight cell samples from Zurich to Bordeauxand back during the 13th DLR parabolic flight campaign Itsreliable and rapid transport system guaranteed the recoveryof flown samples in an outstanding quality for analysis

References

[1] G Sonnenfeld ldquoThe immune system in space andmicrogravityrdquoMedicine and Science in Sports and Exercise vol 34 no 12 pp2021ndash2027 2002

[2] O Ullrich K Huber and K Lang ldquoSignal transduction in cellsof the immune system in microgravityrdquo Cell Communicationand Signaling vol 6 article 9 2008

[3] N Gueguinou C Huin-Schohn M Bascove et al ldquoCouldspaceflight-associated immune system weakening preclude theexpansion of human presence beyond Earthrsquos orbitrdquo Journal ofLeukocyte Biology vol 86 no 5 pp 1027ndash1038 2009

[4] O Ullrich and C S Thiel ldquoGravitational force triggeredstress in cells of the immune systemrdquo in Stress Challengesand Immunity in Space From Mechanisms to Monitoring andPreventive Strategies A Chouker Ed chapter 14 pp 187ndash202Springer Berlin Germany 2012

[5] I V Konstantinova E N Antropova V I Legenrsquokov and VD Zazhireı ldquoReactivity of lymphoid blood cells in the crew ofldquoSoiuz-6rdquo ldquoSoiuz-7rdquo and ldquoSoiuz-8rdquo spacecraft before and afterflightrdquo Kosmicheskaia Biologiia i Aviakosmicheskaia Meditsinavol 7 no 6 pp 35ndash40 1973

[6] S Hauschild S Tauber B Lauber C SThiel L E Layer andOUllrich ldquoT cell regulation in microgravitymdashthe current knowl-edge from in vitro experiments conducted in space parabolicflights and ground-based facilitiesrdquo Acta Astronautica vol 104no 1 pp 365ndash377 2014

[7] B Crucian R Stowe H Quiriarte D Pierson and C SamsldquoMonocyte phenotype and cytokine production profiles aredysregulated by short-duration spaceflightrdquo Aviation Spaceand Environmental Medicine vol 82 no 9 pp 857ndash862 2011

[8] M Hughes-Fulford T Chang and C F Li ldquoEffect of gravity onmonocyte differentiationrdquo in Proceedings of the 10th ESA LifeSciences Symposium29th Annual ISGP Meeting24th AnnualASGSB MeetingELGRA Symposium ldquoLife in Space for Life onEarthrdquo Angers France June 2008


[9] B Bechler A Cogoli M Cogoli-Creuter O Muller E Hun-zinger and S B Criswell ldquoActivation of microcarrier-attachedlymphocytes in microgravityrdquo Biotechnology and Bioengineer-ing vol 40 no 8 pp 991ndash996 1992

[10] J-L Wautier H Setiadi D Vilette D Weill and M-P WautierldquoLeukocyte adhesion to endothelial cellsrdquo Biorheology vol 27no 3-4 pp 425ndash432 1990

[11] W A Muller ldquoMechanisms of leukocyte transendothelialmigrationrdquoAnnual Review of PathologyMechanisms of Diseasevol 6 pp 323ndash344 2011

[12] A Grakoui S K Bromley C Sumen et al ldquoThe immunologicalsynapse a molecular machine controlling T cell activationrdquoScience vol 285 no 5425 pp 221ndash227 1999

[13] T Nagel N Resnick W J Atkinson C F Dewey Jr and M AGimbrone Jr ldquoShear stress selectively upregulates intercellularadhesion molecule-1 expression in cultured human vascularendothelial cellsrdquo The Journal of Clinical Investigation vol 94no 2 pp 885ndash889 1994

[14] M A Meloni G Galleri G Pani A Saba P Pippia and MCogoli-Greuter ldquoEffects of real microgravity aboard interna-tional space station onmonocytes motility and interaction withT-lymphocytesrdquo in Proceedings of the 10th ESA Life SciencesSymposium29th Annual ISGP Meeting24th Annual ASGSBMeetingELGRA Symposium ldquoLife in Space for Life on EarthrdquoAngers France 2008

[15] Y Zhang C Sang K Paulsen et al ldquoICAM-1 expression andorganization in human endothelial cells is sensitive to gravityrdquoActa Astronautica vol 67 no 9-10 pp 1073ndash1080 2010

[16] K P Murphy Janewayrsquos Immunology Garland Science Taylor ampFrancis Group LLC New York NY USA 8th edition 2012

[17] B Alberts A Johnson J Lewis M Raff K Roberts and PWalter Molecular Biology of the Cell Garland Science Tayloramp Francis Group LLC New York NY USA 5th edition 2008

[18] C S Thiel K Paulsen G Bradacs et al ldquoRapid alterationsof cell cycle control proteins in human T lymphocytes inmicrogravityrdquo Cell Communication and Signaling vol 10 no 1article 1 2012

[19] K Paulsen C Thiel J Timm et al ldquoMicrogravity-inducedalterations in signal transduction in cells of the immunesystemrdquo Acta Astronautica vol 67 no 9-10 pp 1116ndash1125 2010

[20] E Blasi R Barluzzi V Bocchini R Mazzolla and F BistonildquoImmortalization of murine microglial cells by a v-raf v-myccarrying retrovirusrdquo Journal of Neuroimmunology vol 27 no2-3 pp 229ndash237 1990

[21] S J Gerathewohl Ed Zero-G Devices and WeightlessnessSimulators Report for the Armed Forces-NAS-NRC Committeeon Bioastronautics Panel on Acceleration National ResearchCouncil Publication 781 National Academy of Sciences Wash-ington DC USA 1961

[22] U Kuebler ldquoSIMBOX Plungerrdquo in Astrium Space Biology Prod-uct Catalog chapter 718 pp 153ndash157 AstriumSpaceTransporta-tion Department of New Business Friedrichshafen Germany2012

[23] K Paulsen S Tauber N Goelz et al ldquoSevere disruption of thecytoskeleton and immunologically relevant surface moleculesin a human macrophageal cell line in microgravitymdashresultsof an in vitro experiment on board of the Shenzhou-8 spacemissionrdquo Acta Astronautica vol 94 no 1 pp 277ndash292 2014

[24] M Kanehisa S Goto Y Sato M Kawashima M Furumichiand M Tanabe ldquoData information knowledge and principleback to metabolism in KEGGrdquo Nucleic Acids Research vol 42no 1 pp D199ndashD205 2014

[25] M Kanehisa and S Goto ldquoKEGG kyoto encyclopedia of genesand genomesrdquo Nucleic Acids Research vol 28 no 1 pp 27ndash302000

[26] S Tauber S Hauschild C Crescio et al ldquoSignal transductionin primary human T lymphocytes in altered gravitymdashresults ofthe MASER-12 suborbital space flight missionrdquo Cell Communi-cation and Signaling vol 11 no 1 article 32 2013

[27] A Karperien H Ahammer and H F Jelinek ldquoQuantitatingthe subtleties of microglial morphology with fractal analysisrdquoFrontiers in Cellular Neuroscience vol 7 no 3 2013

[28] L Buravkova Y Romanov M Rykova O Grigorieva andN Merzlikina ldquoCell-to-cell interactions in changed gravityground-based and flight experimentsrdquo Acta Astronautica vol57 no 2-8 pp 67ndash74 2005

[29] Y A Romanov L B Buravkova M P Rikova E N AntropovaN N Savchenko and N V Kabaeva ldquoExpression of celladhesion molecules and lymphocyte-endothelium interactionunder simulated hypogravity in vitrordquo Journal of GravitationalPhysiology vol 8 no 1 pp 5ndash8 2001

[30] R J Cohrs S K Mehta D S Schmid D H Gilden and DL Pierson ldquoAsymptomatic reactivation and shed of infectiousvaricella zoster virus in astronautsrdquo Journal of Medical Virologyvol 80 no 6 pp 1116ndash1122 2008

[31] S K Mehta R J Cohrs B Forghani G Zerbe D H Gildenand D L Pierson ldquoStress-induced subclinical reactivation ofvaricella zoster virus in astronautsrdquo Journal of Medical Virologyvol 72 no 1 pp 174ndash179 2004

[32] S Muro R Wiewrodt A Thomas et al ldquoA novel endocyticpathway induced by clustering endothelial ICAM-1 or PECAM-1rdquo Journal of Cell Science vol 116 no 8 pp 1599ndash1609 2003

[33] S Muro C Gajewski M Koval and V RMuzykantov ldquoICAM-1 recycling in endothelial cells A novel pathway for sustainedintracellular delivery and prolonged effects of drugsrdquoBlood vol105 no 2 pp 650ndash658 2005

[34] O Carpen P Pallai D E Staunton and T A SpringerldquoAssociation of intercellular adhesion molecule-1 (ICAM-1)with actin-containing cytoskeleton and 120572-actininrdquo Journal ofCell Biology vol 118 no 5 pp 1223ndash1234 1992

[35] E VandenBerg M D Reid J D Edwards and H W DavisldquoThe role of the cytoskeleton in cellular adhesion moleculeexpression in tumor necrosis factor-stimulated endothelialcellsrdquo Journal of Cellular Biochemistry vol 91 no 5 pp 926ndash937 2004

[36] H Schatten M L Lewis and A Chakrabarti ldquoSpaceflight andclinorotation cause cytoskeleton andmitochondria changes andincreases in apoptosis in cultured cellsrdquo Acta Astronautica vol49 no 3-10 pp 399ndash418 2001


[38] S J Crawford-Young ldquoEffects of microgravity on cell cytoskele-ton and embryogenesisrdquo International Journal of DevelopmentalBiology vol 50 no 2-3 pp 183ndash191 2006

[39] B M Uva M A Masini M Sturla et al ldquoClinorotation-induced weightlessness influences the cytoskeleton of glial cellsin culturerdquo Brain Research vol 934 no 2 pp 132ndash139 2002

[40] B M Uva F Strollo F Ricci M Pastorino J I Mason andM A Masini ldquoMorpho-functional alterations in testicular andnervous cells submitted to modelled microgravityrdquo Journal ofEndocrinological Investigation vol 28 no 11 pp 84ndash91 2005


[41] M Infanger P Kossmehl M Shakibaei et al ldquoSimulatedweightlessness changes the cytoskeleton and extracellularmatrix proteins in papillary thyroid carcinoma cellsrdquo Cell andTissue Research vol 324 no 2 pp 267ndash277 2006

[42] V E Meyers M Zayzafoon J T Douglas and J M McDonaldldquoRhoA and cytoskeletal disruption mediate reduced osteoblas-togenesis and enhanced adipogenesis of human mesenchymalstem cells in modeled microgravityrdquo Journal of Bone andMineral Research vol 20 no 10 pp 1858ndash1866 2005

[43] M L Lewis J L Reynolds L A Cubano J P Hatton B DesalesLawless and E H Piepmeier ldquoSpaceflight alters microtubulesand increases apoptosis in human lymphocytes (Jurkat)rdquo TheFASEB Journal vol 12 no 11 pp 1007ndash1018 1998

[44] M Hughes-Fulford ldquoReview of the biological effects of weight-lessness on the human endocrine systemrdquo Receptor vol 3 no3 pp 145ndash154 1993

[45] R Gruener R Roberts and R Reitstetter ldquoReduced receptoraggregation and altered cytoskeleton in cultured myocytes afterspace-flightrdquo Uchu Seibutsu Kagaku vol 8 no 2 pp 79ndash931994

[46] M Hughes-Fulford ldquoFunction of the cytoskeleton in gravisens-ing during spaceflightrdquo Advances in Space Research vol 32 no8 pp 1585ndash1593 2003

Research ArticleGenes Required for Survival in MicrogravityRevealed by Genome-Wide Yeast Deletion CollectionsCultured during Spaceflight

Corey Nislow1 Anna Y Lee2 Patricia L Allen3 Guri Giaever1

Andrew Smith2 Marinella Gebbia2 Louis S Stodieck4 Jeffrey S Hammond5

Holly H Birdsall67 and Timothy G Hammond3689

1Faculty of Pharmaceutical Sciences The University of British Columbia Vancouver BC Canada V6T 1Z32Donnelly CCBR University of Toronto Toronto ON Canada M5S 3E13Durham VAMedical Center Research amp Development Service Durham NC 27705 USA4Bioserve Space Technologies University of Colorado Boulder CO 80309 USA5The Institute for Medical Research Durham NC 27705 USA6Department of Veterans Affairs Office of Research and Development Washington DC 20420 USA7Departments of Otorhinolaryngology Immunology and Psychiatry Baylor College of Medicine Houston TX 77030 USA8Nephrology Division Department of Internal Medicine Duke University School of Medicine Durham NC 27705 USA9Nephrology Section Department of Internal Medicine George Washington University School of MedicineWashington DC 20052 USA

Correspondence should be addressed to Timothy G Hammond grumpy70115yahoocom



Copyright copy 2015 Corey Nislow et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Spaceflight is a unique environment with profound effects on biological systems including tissue redistribution andmusculoskeletalstresses However the more subtle biological effects of spaceflight on cells and organisms are difficult to measure in a systematicunbiasedmanner Here we test the utility of the molecularly barcoded yeast deletion collection to provide a quantitative assessmentof the effects of microgravity on a model organism We developed robust hardware to screen in parallel the complete collectionof sim4800 homozygous and sim5900 heterozygous (including sim1100 single-copy deletions of essential genes) yeast deletion strainseach carrying unique DNA that acts as strain identifiers We compared strain fitness for the homozygous and heterozygous yeastdeletion collections grown in spaceflight and ground as well as plus and minus hyperosmolar sodium chloride providing a secondadditive stressor The genome-wide sensitivity profiles obtained from these treatments were then queried for their similarity to acompendium of drugs whose effects on the yeast collection have been previously reported We found that the effects of spaceflighthave high concordance with the effects of DNA-damaging agents and changes in redox state suggesting mechanisms by whichspaceflight may negatively affect cell fitness

1 Introduction

Physical effects of microgravity during spaceflight can oftenbe described by equations that allow their quantification[1 2] For example microgravity has well-defined effects onsedimentation in association with reduced terminal velocityand shear in suspension culture [3] and reduced gravity-dependent convection of gases [4] Biological effects of

spaceflight on cells and organisms on the other hand aremuch harder to define [1 2] For example spaceflight alsoentails radiation exposure which has been studied in diversesystems but whose effects are not fully understood [5] Whatis needed is a robust unbiased quantifiable system that isrelevant for translation to ground-based applications andthat is able to clearly distinguish spaceflight effects It isour premise that yeast deletion collections are ideally suited



for this type of analysis as yeast can be precisely controlledgenetically and readily grown under spaceflight conditionsBiological responses of yeast strains during spaceflight canbe quantified and compared to well-established databasesof ground-based stressors and the comparisons can revealfeatures that are unique to microgravity as well as featuresthat are sharedwith ground-based perturbationsWhile yeastcannot completely reflect the complexities of mammaliancells organized into tissues the high degree of homologyshared with human (sim70 of all essential yeast genes havea significant human homolog) provides hypotheses for themechanism of many responses of interest [6 7]

Previous studies have attempted to identify isolate andoffset the various physical factors changing during spaceflightto demonstrate their effects in an iterative fashion [8ndash10]Earlier studies on the effects of space radiation on yeast failedto find any change in point mutation rates DNA replicationandor repair heritable damage or colony morphology [58 11 12] However those studies were limited by assaysensitivity Here we applied the yeast deletion collection asa biological reporter to understand the metabolic pathwaysaffecting survival during culture in spaceflight In this fash-ion we are able to make genome-wide comparisons and testfor concordance against an extensive library of more than3200 physical and pharmacological stressors [13]

Yeast is the first and to date only organism for whicha complete genome-wide knockdown collection is availableThis collection is comprised of a genome-wide set of strainswhere each strain carries a precise deletion of a single gene[14] Assembled over a four-year period by a consortium of35 laboratories this collection has been used by hundredsof laboratories to test thousands of different environmentalstressors to define the genes required for survival in thoseconditions (see [15] for review) The molecular barcodespresent in each strain allow the yeast deletion collection tobe grown as a pool in the presence or absence of the stressorof interest after which the relative abundance of each strainis subsequently quantified [16] Strains carrying a deletion ofa gene required for survival in the presence of the stressorgrow more slowly and thus exhibit a fitness defect reflectedby their reduced abundance at the end of the culture periodIn this manner all genes required for growth can be readilyidentified in a single experiment revealing the genes andassociated pathways affected by the stressor

To identify the metabolic and genomic pathways affectedby spaceflight the homozygous and heterozygous yeastdeletion collections were grown in spaceflight and groundcontrol conditions with and without hyperosmolar sodiumchloride providing a second stressor In spaceflight alonethe homozygous deletion collection revealed the importanceof processes linked to mitochondria while the heterozygouscollection highlighted genes involved in regulating transla-tion and ribosomal RNA transport Both homozygous andheterozygous collections highlighted DNA repair With theaddition of NaCl the homozygous collection also revealedthe importance of RNA-related processes including ribosomeassembly and biogenesis andmRNA processing and decay aswell as modification of tRNAs Moreover the NaCl additionhighlighted replication processes more clearly (compared to

the homozygous collection without NaCl) suggesting thatspaceflight has measurable effects on these core and evolu-tionarily conserved processes With the heterozygous collec-tion the addition of NaCl led to the identification of a nuclearpore organization gene potentially providing additionalinsight into how RNA transport is affected by spaceflightTaken together the deletion collections identified severalbiological processes associatedwith spaceflight and the addi-tional hyperosmolar stress emphasized the importance ofrelated processes

In a follow-up analysis we queried the effects of space-flight against a database of drug effects on yeast to search forthose that are most concordant thereby suggesting similarmechanisms of perturbation Not only do the effects ofspaceflight have relatively high concordance with the effectsof DNA-damaging agents but also there is tight agreementamongst multiple therapeutic agents in this drug classproviding additional support for these findings

2 Materials and Methods

21 Overall Design The Opticell Processing Moduledescribed below was used to perform a series of sim21 genera-tion pooled growth experiments on two yeast deletioncollections (i) sim4800 homozygous strains and (ii) sim5900heterozygous strains (including sim1100 single-copy deletionsof essential genes) each carrying unique DNA barcodes thatact as strain identifiers Experiments were performed in bothrich media and rich media supplemented with 05M NaClto assess the additional effect of osmotic stress on survivalThe samples flew sortie on space shuttle mission STS-135to the International Space Station (ISS) Parallel controlexperiments were performed in static 1G terrestrial controlsin the Orbital Environmental Simulator at Kennedy SpaceCenter to match temperature humidity air compositionand volatile organic compounds Ground controls wereconducted in a 24-hour asynchronous fashion to allowmatching of the experimental timelines on ISS as relayedthrough air-to-ground communication by the flight crew Atthe end of the growth period the fitness of each strain in eachexperimental pool was assessed as described [17] Brieflygenomic DNA was extracted from each sample the barcodesin each pool were amplified by PCR and the abundance ofeach barcode was quantified by next generation sequencingA barcode count reflects the abundance of the correspondingstrain at the end of the experiment that is a quantification ofthe relative requirement of the deleted gene for growth in thetested condition In total the experiment results in a countfor each gene resulting in a gene list rank ordered by theirimportance for growth in the tested condition

22 Yeast Deletion Pool Construction The yeast deletion col-lections were stored as individual strains in YPD containing7 DMSO at minus80∘C in 96-well plates The plates werethawed mixed and robotically pinned onto YPD agar platesas an array of 384 strains After two days of growth at 30∘Ccolonies were consolidated (four plates of 384 to one plateof 1536 colonies) and robotically pinned in triplicate Cellswere grown in 30∘C for 2-3 days until colonies formed Slow


growing strainswere grown separately for 2-3 additional daysAll plates were then flooded with 5ndash7mL of media scrapedand pooled in YPD + 7 DMSO to a final concentration ofOD600

= 084 and frozen atminus80∘Cuntil use as described [17]

23 Construction of Opticell Culture System and SpaceflightExperiment In this study we designed the Opticell Pro-cessing Module or OPM (Figure 1) that was capable ofmaintaining the yeast deletion collection as a pool grownin liquid culture for at least 20 generations in microgravityThe hardware comprised a liquid-sealed system of growthchambers (Opticells) that allowed for gas exchange acrosspolystyrene membranes Each OPM consisted of three NuncOpticells held together with a common manifold and valvesystem that is autoclaved and attached with watertight O-ring seals A 3mL syringe connected to the manifold witha Luer fitting is used to transfer liquid between chambersandmixwithout breaking sterility andwithminimal operatorinterventionThe valve on the manifold has four settings thatconnect the syringe to the following port locations 1 Offposition 2 Opticell A 3 Opticell B or 4 Opticell C TheOPM allows propagation of each deletion collection for acombined sim21 generations of growth when three chambersare used and the inoculum and transfer volumes are 05mL

To perform a growth assay in the OPM each of thethree chambers was prefilled with 7mL of sterile growthmedia Deletion collection aliquots were preloaded intoeach syringe and shipped to Kennedy Space Center frozenin media containing 7 DMSO (vv) as a cryoprotectantDuring final integration at Kennedy Space Center the OPMswere prechilled to 4∘C Each deletion collection aliquotwas thawed attached to an OPM manifold injected andmixed into chamber A Cultures were maintained at 4∘C andflown to the International Space Station (ISS) The growthexperiment was initiated on orbit by warming the OPMs to30∘C After 16ndash24 h at 30∘C a 05mL sample was removedfrom chamber A using the same syringe and inoculated intochamber B The process was repeated 16ndash24 hours later toinoculate 05mL of sample from chamber B into chamberC After an additional 16ndash24 hours the OPMs were cooledback down to 4∘C to greatly reduce any further growthand preserve the samples for return to Earth and postflightanalysis Exponential yeast growth leads to early depletionof growth media nutrients and significant retardation offurther growth well before 16ndash24 hours Growth is limitedby media volume and strain distribution within the yeastdeletion library reaches a steady state within that Opticell

24 Next Generation Sequencing The flight samples thatreturned from the ISS were handled in parallel with theground control set The OPMwas disassembled into its threeOpticells and the entire contents were transferred to a storagetube using a blunt needle connected to a 20mL syringe OnemL of each sample (at a final OD

600of 10-20) was pro-

cessed to extract genomic DNA Purified deletion pool DNAwas amplified in two separate PCR reactions as described[17] and the amplicons purified prior to sequencing on anIllumina HiSeq2000 Each purified amplicon library wassequenced to a minimum depth of 500 countsstrainsample

Syringe with yeast

3-way stopcock

Opticell A

Opticell B

Opticell C

Figure 1 The Opticell Processing Module (OPM) designed forpropagation of each deletion collection for sim21 generations ofgrowth The OPM comprises three commercially available opticallyclear chambers (Opticells Nunc) that are joined by a manifold andscaffold that can be autoclaved and assembled rapidlyThemanifoldcontains a multiway valve unit which mates to each Opticell or toan the off position using O-ring seals The opposite side of the valvecontains a Luer fitting into which a standard 3cc syringe is attachedTo perform a growth assay in theOPM each of the three chambers isfilled with 7mL of sterile growth media Deletion pools are loadedinto the inoculation syringe and then injected into Chamber A ofthe OPM precooled to 4∘C Growth is initiated by warming theunit to 30∘C After 16ndash24 h 05mL is removed fromChamber A andinjected and mixed into Chamber B using the same syringe This isrepeated to continue multigenerational growth in Chamber C

as described [18] Duplicate experiments were performed forall conditions to ensure that at least one complete time coursewas collected for each pool (heterozygote and homozygote)and each condition Due to failures in sample processingseveral time points were not recovered or did not meetour in-house quality metrics (eg if sequence countsstrainwere below threshold values) Accordingly we focused onevaluating each experimental condition using singleton dataas described in Table 1

25 Data Analysis All computational analyses were per-formed using 119877 [19] unless otherwise indicated

251 Normalization of Sequence Counts Sequence countsfor each strain in each experiment were quantified and nor-malized according to [18] Briefly each 20-mer barcode wasamplified with primers comprised of the common barcodeprimers in addition to the sequences required for clusterformation on the Illumina flow cell Formultiplexed Illuminasequencing 5-mer tag sequences were incorporated into eachprimer between the Illumina and barcode primer sequencesThis multiplexing tag allowed postsequencing assignment ofeach amplicon to a particular experiment Results for the7-generation time point of the heterozygous deletion poolgrown in spaceflight without NaCl did not pass our quality


Table 1 Experimental samples collected and available for analysis

Condition Zygosity GenerationsGround Homozygous 7 14 21Ground + 05M NaCl Homozygous 7 14 21Flight Homozygous 7 14 21Flight + 05M NaCl Homozygous 7 14 21Ground Heterozygous 14 21lowast

Ground + 05M NaCl Heterozygous 7 14 21Flight Heterozygous 14 21lowast

Flight + 05M NaCl Heterozygous 7 14 21lowast7-generation samples from the indicated condition were not available foranalysis due to failures in sample processing or failure to meet in-housequality metrics as described in Methods

control and consequently that time point was omitted fromanalyses of the heterozygous pool without NaCl All othercounts were mean-normalized between experiments suchthat each experiment had the same mean count We addedten pseudocounts to all sequence tag tallies (and thus allsubsequent gene tallies) to prevent division by zero duringdata analyses (see Table S1 and Table S2 for mean normalizedcounts in Supplementary Materials available online at httpdxdoiorg1011552014976458)

252 Barcode Selection for Each Strain For each strain weused signal from only the upstream or the downstreambarcode (relative to the deletion site) First we assumed thatbarcode counts close to zero represent backgroundnoise (egpossibly due to incorrect mapping of reads to barcodes) Wethus selected a background threshold (bgThreshold = 100see Parameter Selection) assuming that counts below it donot accurately reflect strain abundance Then for each timecourse we filtered out barcodes where the average (normal-ized) count for the first 14 generations (the earliest time pointwith usable data in all experiments) was below bgThresholdThis filtering removed all barcodes for sim650ndash3200 strains(depending on the time course) and these strains wereomitted from subsequent analyses see Supplementary TablesS3 (homozygous strains) and S4 (heterozygous strains)

For time courses that only had two time points (andthus an inefficient number to compute fits) strains thatstill had two barcodes after filtering were represented bytheir upstream barcodes due to their overall better behaviorobserved in a previous study [20] For other time courseslinear fits (with and without the time logged) were computedfor each remaining barcode We defined the best fit as the fitwith the lowest residual sum of squares (RSS) and used the119865-test to compute a 119875 value estimating the significance withwhich the fitted model is better than the null model (of a flatline at the average count value)TheBenjamini andHochbergmethod was used to correct the 119875 for multiple comparisonsand generate FDR values [21] Strains with two remainingbarcodes were represented by the barcode with the higher 1198772(ameasure of the amount of variation in the data explained bythe fitted model) because they manifest less noise and betterfit the data

253 Parameter Selection The selected normalizationmethod (tested mean and quartile normalization) andbgThreshold (tested 50 100 and 150) is the combination thatresulted in the most significant enrichment of slow growingstrains identified in the heterozygous deletion pool sampledevery two generations for 20 generations (data not shown)with slow growers identified in a previous study [22] Brieflywe defined slow growers as those exhibiting sizable decreasesin abundance over time The significance of the decrease wasestimated with FDR values (see Barcode Selection for EachStrain) and the magnitude was estimated with ΔAUC =(⟨area under the growth curve⟩ minus ⟨area under the flat growthcurve⟩)(119905max minus 1199050) where the flat growth curve is fixed at the1199050abundance level and the area under a curve is estimated

using the trapezoid method Also if at some time point theabundance of a strain is less than or equal to bgThreshold andremains at negligible levels for the rest of the time course weidentified the strain as slow growing

254 Identification of Significant Fitness Defects in TimePoint Comparisons To identify strains that exhibited sig-nificant fitness defects at a later time point (14 or 21 gen-erations) compared to the first time point (7 generations)normalized counts less than bgThreshold were first forcedto equal bgThreshold Then for each strain we comput-ed log

2ratio = log

2(abundance

7Gabundance14G21G) whereabundanceyG is the count of the strain at 119910 generationsFor a given time point robust 119885 scores were computedfrom the set of log

2ratios for example 119885

119894= (log

2ratio119894minus

⟨log2ratio median⟩)⟨log

2ratio MAD⟩ for strain 119894 Each 119885

119894

was then used to obtain 119875119894from the standard normal

distribution and we assume that strains with low 119875 valuesare outliers in the distribution of log

2ratios Moreover strains

with counts above bgThreshold at the first time point andcounts equivalent to bgThreshold at the later time point ofinterest are defined as having dropped out Taken togetherwe define strains with significant fitness defects at a specifictime point as strains with log

2ratio ge 1 and 119875 le 0001 andor

strains that dropped out (Table S5 Table S6)

255 Spaceflight versus Ground Comparisons For compar-isons involving specific time points we identified the set ofstrains that exhibited significant fitness defects (relative tothe first time point) in the flight condition but not in theground conditionThis set is then further restricted to the setof strains with useable data in both conditions

256 GeneOntology (GO) Enrichment Analysis Weobtainedgene ontology (GO) annotations of yeast genes from theSaccharomyces Genome Database (downloaded on May 262012) GO biological processes that were too specific (con-taining less than five genes) or too general (containing greaterthan 300 genes) were excluded from the analysis

Given a query set of genes (eg genes deleted from a setof (flight-ground) strains) we used the hypergeometric test toobtain a119875 value estimating the significancewithwhich the setis enrichedwith genes annotated to a given biological processrelative to a gene universe defined as the set of genes withusable data for both flight and ground conditions Due to a


limited number of significantly enriched processes followingcorrection for multiple comparisons (FDR le 01) here wereport significantly enriched processes prior to the correction(119875 le 001)

We visualized GO enrichment results with enrichmentmaps shown in Figures 2 and 3 that were generated using anapproach similar to the Enrichment Map Cytoscape Pluginv11 [23 24] In contrast to the plugin the nodes in eachmap were clustered with MCL (inflation = 2) using theoverlap coefficient computed by the plugin as the similaritymetric (coefficients less than 05 were set to zero) Nodes inthe same cluster were assigned the same node color and acluster label was determined based on common themes inthe processes within the cluster Moreover the size of a nodewas made to be proportional to the significance with whichthe corresponding process is enriched [minuslog

10(119875)] Each bar

plot summarizes the genes that contribute the most to theenrichment of processes with the same node color as the plotborder Specifically a plot shows the flight-associated genesthat are annotated to the largest number of relevant processes(if more than 10 genes only the top 10 are shown) For eachgene the bar length is proportional to a fitness defectmeasure(ie log

2ratio)

The enrichment maps also combine two sets of enrich-ment results with the processes enriched in one set shownwith circle nodes the processes enriched in the second setshown with square nodes and the processes enriched in bothsets shown with diamond nodes

257 Similarity between Flight-Associated Genes and Com-pound-Associated Genes We previously treated pools ofyeast deletion strains with sim3200 compounds separately[13] Each compound was subsequently associated with aset of genes deleted from strains that exhibited significantfitness defects induced by the compound Like sets of flight-associated genes in this study sets of compound-associatedgenes were assessed for enrichment of genes annotated tospecific biological processes (as described above) resultingin an ldquoenrichment profilerdquo for each condition of interest Ineach profile each process is associated with a 119875 measuringthe significance of enrichment Similarity between a pairof enrichment profiles was computed by concordance ofminuslog10(119875) across all processes common to both profiles

where concordance is like Pearson correlation except thatscale is not ignored Compounds with enrichment profilesthat are most similar to a given flight enrichment profile mayinduce cellular responses that aremost similar to the responseinduced by flight

3 Results and Discussion

Because we cannot distinguish the individual parameters thatinclude flight lack of gravity and increased radiation for thepurposes of this paper these are referred to collectively asldquospaceflightrdquo throughout the text To measure the effects ofspaceflight on the rate of yeast growth in the Opticell weinoculated 05mL of a yeast deletion pool into 7mL of YPDresulting in a starting OD

600of sim006mL and incubated at

30∘C Following growth forsim24 hr (sim7 generations) 05mLof

the saturated culturewas inoculated into the second chamberThis process was repeated for the final growth phase in thethird chamber Population doubling time was sim100min inmicrogravity compared tosim90min in ground-based controlsEach sample was grown for seven generationsOpticell for atotal of 21 generations (Table 1) Doubling times were back-calculated using the OD

600of the samples collected at each

time pointThe morphology of Opticell-grown yeast in spaceflight

was indistinguishable from static controls when observedby light microscopy for example budding pattern overallshape and size were not detectably different in the twoconditions On scanning electron microscopy there werebudding polarity and ruffling changes in every field but therewere no consistent differences (data not shown)

We assessed the yeast deletion collection samples forpatterns of strain sensitivity in the followingmanner barcodecounts for each strain in each sample were measured andnormalized as described inMethodsThe counts were used torank each strain in each sample in order of their importancefor growth Four different samples were available from bothspaceflight and ground cultures (1) homozygous deletion col-lection in YPD (2) homozygous deletion collection in YPDplus 05MNaCl (3) heterozygous deletion collection in YPDand (4) heterozygous deletion collection in YPD plus 05MNaCl Each culturewas sampled at three different time points7 generations 14 generations and 21 generations and shownin Table 1 Samples from ground controls were compared tothe corresponding samples grown inmicrogravity on the ISS

We analyzed changes in strain abundance by comparingeach time point to a later time point Using this approachallowed us to capture those strains that became depleted inany seven-generation interval Strains with sizable decreasesin measured abundance or with abundances that drop tobackground levels (and remain there) were identified asexhibiting fitness defects (FDs) Moreover strains with flight-specific FDs were identified by subtracting the strains withFDs in the ground condition

For the purposes of our gene ontology (GO httpamigogeneontologyorg) enrichment analysis we considered thehomozygous and heterozygous data separately Based on awealth of published data [14 15] the homozygous nonessen-tial deletion collection tends to reveal a similar set of genesinvolved in pathways required for resistance tosurvival inmultiple environmental conditions whereas the heterozy-gous collection of all strains tends to be more specificidentifying essential proteins uniquely required for growth ina specific condition [13]

For the homozygous deletion collection strains thatwere depleted from the pool specifically in spaceflight con-ditions are significantly enriched for genes in biologicalprocesses related to different aspects of RNA metabolismand catabolism including ribosome biogenesis regulationof ribosomal protein transcription cytoplasmic RNA trans-lation rRNA processing tRNA modification and mRNAdecay (Table 2 Figure 2 and Table S5) We also found thatprocesses related to DNA integrity were required for survivalin spaceflight In particular the linked processes of DNArepair and DNA recombination and replication as well as


00 04 08 12

GIR2RPL27ARPS12RPL20ARPL21A

FD score

xx

xx

x

14G

00 10 20

PEX19

MON1AFG3SAM37VPS9MMM1VPS3

FD score

xxx

xx

x

21G00 10 20

TRM7SIT4NCS2ELP6TGS1ELP2URM1

FD score

xxx

21G

00 10 20

MGM1MMM1RPO41MRPL8GGC1

FD score

x

x

21G

00 10 20

RPS11ARPS17ARPS11BSSF1RPS27B

FD score

xx

xx

21G

00 10 20

SRB2HMO1

FD score

x 21G

00 10 20

BOI1BEM4SLG1

FD score

xx

21G

00 10 20

PEX19VPS9VPS3MDM31

FD score

x 21G

00 10 20

MON1

FD score

x 21G

00 10 20

SNF2MUS81MMS1IRC19RAD54RAD27RAD51ARP8

FD score

xxx

xx

x

21G

00 10 20

NOP12RPL35A

FD score

xx

21G

00 10 20

RPS9BRPS11A

RPS17ASLX9RPS11BRPS16BHCR1RPS21ARPS27B

FD score

xx

xx

x

21G00 10 20

SNF2PHO2NHP10RAD54ARP8

FD score

xx

x

21G

Ribosomeassembly

DNA repairrecombination

Cytoplasmictranslation

rRNA processing

Rho proteinsignal

transductionOrganelle inheritance

Regulation ofribosomal proteingene transcription

Mitochondrialgenome maintenance

Chromatin remodeling

Maturation oflarge subunit rRNA

Vesicle docking

tRNA modification Protein localizationin mitochondrion

SAC3

TOF1

BUD21

VAM3

SAC3LOA1

TOM5

Figure 2 Biological processes enriched amongst genes associatedwith flight-specific fitness defects at different time points in the homozygousdeletion series Each node represents a significantly enriched gene ontology (GO) biological process (hypergeometric test 119875 le 001) A circlenode indicates enrichment at 14 generations compared to 7 generations (the first time point) a square node indicates enrichment at 21generations compared to 7 generations and a diamond node indicates enrichment at both 14 generations and 21 generations (see Methods)Node size is proportional to the significance of enrichment [minuslog

10(119875)] Node color indicates processes that share genes (see Methods) and

summary labels are shown for nodes of the same color Edges indicate ge 50 gene overlap between connected processes width is proportionalto the degree of overlap Each bar plot provides fitness defect (FD) scores for genes that contribute to the enrichment of processeswith the samenode color as the plot border Specifically the length of a bar is proportional to the log

2(abundance

7Gabundance14G21G) where abundanceyGrepresents the abundance of the corresponding gene deletion strain at 119910 generations (see Methods) An ldquo119909rdquo on the bar indicates that theabundance of the strain lowers to background level at later time point

chromatin remodeling were all required for resistance to theeffects of spaceflight Finally these DNA repair requirementsextend to the mitochondria which by virtue of its smallgenome is hypersensitive to DNA damage Consistent withthis we found that genes required for both mitochondrialmaintenance and proper protein localization to the mito-chondria were enriched in the homozygous samples

The enrichment of these particular processes is consistentwith a general induction of DNA damage which in turn

perturbs RNA biogenesis [25] Interestingly we have previ-ously observed this phenomenon with a class of therapeuticsthat act as nucleotide analogs such as 5-fluorouridine andfluorocytosine (described in detail below) Additionally it isparticularly noteworthy that although we do see evidenceof a requirement for RNA and DNA processing genesin spaceflight alone the requirement is exacerbated whenspaceflight is combined with the additional hyperosmoticstress imposed by the addition of 05M NaCl (Table S6)


00 10 20

SPE1HYP2

FD score

x

00 10 20

EPT1SPE1

FD score

xx

14G

14G

00 05 10 15

EPT1KAP95PSD1

FD score

xxx

21G

00 05 10 15

KAP95NUP57BIK1POM33

FD score

xxxx

00 05 10 15

MSH3

FD score

xx

21G

21G

00 05 10 15

RAP1PLC1YKU80

FD score

xx

x

21G

Modifiedamino acidbiosynthesis

Phosphatidylcholinemetabolism

Biogenic aminemetabolism

Nuclear poreorganization DNA catabolism

Proteinlocalization

to chromosome

MAG1

Figure 3 Biological processes enriched amongst genes associated with flight-specific fitness defects in the presence of NaCl at differenttime points in heterozygous deletion samples Each node represents a significantly enriched gene ontology (GO) biological process(hypergeometric test 119875 le 001) Nodes edges and plots are as specified for Figure 2

Table 2 Effects of spaceflight on yeast genome responses identifiedwith the homozygous deletion series

General pathway GO biological process

RNA metabolismand catabolism

(i) Ribosome biogenesis(ii) Regulation of ribosomal proteintranscription(iii) Cytoplasmic RNA translation(iv) rRNA processing(v) tRNA modification(vi) mRNA decay

DNA integrity

(i) DNA repair(ii) Recombination and replication(iii) Chromatin remodeling(iv) Mitochondrial maintenance(v) Proper protein localization to themitochondria

We speculate that the added salt stress potentiates the DNA-damaging effects of spaceflight via the induction of reactiveoxygen species (ROS)The ability of salt stress to induce ROSand subsequent DNA damage has been previously reported[26] and in particular the yeast mitochondria appears to behypersensitive to this type of stress consistent with its smallgenome being susceptible to the effects of DNA damage [27]Furthermore mitochondrial protein abundance has beenshown to rapidly increase upon osmotic shock and therefore

the enrichment for mitochondrial protein localization weobserve may reflect this requirement

To gain further insight into the pathways that modulatethe response to bothmicrogravity stress and combined space-flight and salt stress we used the GO enrichment profiles toquery a database of over 3200 distinct drug treatments of theyeast deletion collections [13] Specifically we quantified thesimilarity between the GO enrichments by computing theconcordance of minuslog

10(119875) between any two profiles where

119875 measures the significance of enrichment of a single GOcategory These concordance values are similar to Pearsoncorrelation values that is values closer to one indicate greatersimilarity between profiles except that high concordance alsorequires the scale of values to be similar between the profilesWhen calculating concordance we focused on GO biologicalprocess enrichment profiles (Table S7)

One of the strong concordances was observed with 5-fluorouridine (042) an FDA-approved anticancer drug thatis thought to also act by two mechanisms (i) inhibitingthymidylate synthetase and (ii) through metabolism intocytotoxic ribonucleotides and deoxyribonucleotides that canbe incorporated into DNA and RNA (Table 3) [14] Inaddition to being incorporated in DNA and RNA we andothers have shown that the drug has been shown to inhibit theessential ribonuclease activity of the exosome complex [28]Similarly carmofur a derivative of 5-fluorouracil displaysa concordance of 034 A similar concordance is seen with


Table 3 Concordance between drug effects and spaceflight effectson yeast genome responses identified with the homozygous deletionseries (+NaCl)

Drug (concordance) Biological function

5-Fluorouridine (042)5-Fluorouracil (036)Carmofur (034)5-Fluorocytosine (035)

Pyrimidine analogs that inhibitthymidylate synthase and aremetabolized into cytotoxicribonucleotides anddeoxyribonucleotides that can beincorporated into DNA and RNA

8-Methoxypsoralen(032) DNA-damaging agent

Diallyl disulfide (04)Increased glutathione-S-transferasechanges redox state by bindingelectrophilic toxins

5-fluorocytosine (5-FC) whose activity is identical to 5-fluorouracil (5-FU) Finally 8-methoxypsoralen a DNA-damaging agent that upon photoactivation conjugates andforms covalent bonds with DNA shows a congruence of 032This compound causes the formation of bothmonofunctional(addition to a single strand ofDNA) and bifunctional adducts(crosslinking of psoralen to both strands of DNA) thatultimately result in cell death

We also found high concordance to the diallyl disulfideprofile (040) an agent that has been demonstrated to beefficient for detoxification of a variety of cells Diallyl disulfideand related garlic derivatives have been shown to signifi-cantly increase the production of the enzyme glutathione S-transferase (GST) which binds electrophilic toxins in thecell Overloading the cell with inhibitory doses of diallyldisulfide reveals genes required for survival in the presenceof increased reactive oxygen species (ROS) [29]

In the case of the heterozygous collection we foundsignificant GO enrichments for the following categories lipidmetabolism DNA catabolism and regulation of transla-tion and posttranslational modification (specifically proteinphosphorylation) (Figure 3) As expected (based on previousstudies of the heterozygous collection) both the number ofgenes associated with FDs and the number of enriched cate-gories are considerably smaller than those derived from thehomozygous collection [30] This likely reflects two relatedphenomena first genes that when deleted in heterozygotesare sensitive to spaceflight encode proteins that participate inthe pathways identified in the homozygous collection wherethe fitness defect is stronger because the gene is completelyabsent Second none of these heterozygote strains encode adirect target of the perturbation

Interestingly when we searched for drug profiles withhigh concordance with the spaceflight profiles derived fromthe heterozygous collection we detected modest concor-dance with two human chemotherapeutics mitoxantrone(concordance = 019) and Epirubicin (congruence 0142)Both of these agents damage DNA by intercalating intothe DNA double helix and also by stabilizing the cleavablecomplex that is the substrate of topoisomerase II [31ndash33]

4 Conclusions

The experiments presented here represent a proof of prin-ciple for conducting full genome environmental screens inspaceflight using robust hardware that can recapitulate a fullautomation suite with environmental control in the space of asmall suitcaseThe performance of this platform is significantfor spaceflight studies and promises to enable terrestrialexperiments in extreme environments that will have directapplication to microbial bioprocessing for manufacturingalternative fuel development and basic research The resultsfrom these experiments suggest that spaceflight has subtle butsignificant effects on core cellular processes including growthcontrol via RNA and ribosomal biogenesis metabolismmodification and decay pathways Furthermore significantroles for DNA repair and replication response to pH signal-ing control of gene expression and mitochondrial functionwere observedThe yeast chemogenetic analysis of spaceflightsamples presented here strongly implicates DNA and RNAdamage as the major ground-based analogs of spaceflightstress Given the unique and substantial radiation exposurein space this is consistent with major radiation-mediatedeffects Unfortunately a 1 g control on ISS that might haveallowed better discrimination between the contributions ofspace radiation versus the effects of microgravity on yeastresponses was not available to us at this time Current on-going experiments are designed explore these effects anddissect them from other potentially confounding variablesThe high concordance to the profile induced by diallyl disul-fide suggests increased glutathione S-transferase binding ofelectrophilic toxins increased reactive oxygen species andchange in redox stateThese pathways which are required forsurvival in spaceflight can guide future experiments in twofundamental ways firstly by suggesting environmental mod-ifications that can bolster cellular and organismal integrity byavoiding further stress to these pathways and secondly byidentifying drug stresses that can exacerbate these pathwayrequirements in an effort to control pathological cell growthin the case of proliferative diseases

Abbreviations

AUC Area under the curveBP Biologic processesDMSO Dimethyl sulfoxideFD Fitness defectFDR False discovery rateGO Gene ontologyGST Glutathione S-transferaseISS International Space StationMAD Median absolute deviationMCL Markov Clustering AlgorithmOD Optical densityOPM Opticell Processing ModulePCR Polymerase chain reactionROS Reactive oxygen speciesRSS Residual sum of squaresYPD Yeast peptone dextrose



None of the authors have any commercial associations thatmight create a conflict of interests

Acknowledgments

These studies were supported by NASA Grant noNNX10AP01G The authors thank NASA for spaceflightaccess under the auspices of the International Space StationNational Lab Pathfinder program This material is the resultof work supported with resources and the use of facilities atthe Durhan Veterans Affairs Medical Center and the Officeof Research and Development Department of VeteransAffairs Veterans Health Administration Sequencing wasperformed in part at UBCSeq Vancouver Contents do notrepresent the views of the Department of Veterans Affairs orthe United States of America

References

[1] R Herranz R Anken J Boonstra et al ldquoGround-basedfacilities for simulation of microgravity organism-specific rec-ommendations for their use and recommended terminologyrdquoAstrobiology vol 13 no 1 pp 1ndash17 2013

[2] J J van Loon E H T E Folgering C V C Bouten J PVeldhuijzen and T H Smit ldquoInertial shear forces and the useof centrifuges in gravity research What is the proper controlrdquoJournal of Biomechanical Engineering vol 125 no 3 pp 342ndash346 2003

[3] T G Hammond and J M Hammond ldquoOptimized suspensionculture the rotating-wall vesselrdquo American Journal of Physiol-ogy Renal Physiology vol 281 no 1 pp F12ndashF25 2001

[4] M R Benoit R B Brown P Todd E S Nelson and D MKlaus ldquoBuoyant plumes from solute gradients generated bynon-motile Escherichia colirdquo Physical Biology vol 5 no 4Article ID 046007 2008

[5] J Kiefer and H D Pross ldquoSpace radiation effects and micro-gravityrdquoMutation Research vol 430 no 2 pp 299ndash305 1999

[6] D Botstein S A Chervitz and J M Cherry ldquoYeast as a modelorganismrdquo Science vol 277 no 5330 pp 1259ndash1260 1997

[7] D Botstein and G R Fink ldquoYeast an experimental organismfor 21st century biologyrdquo Genetics vol 189 no 3 pp 695ndash7042011

[8] P Todd ldquoOverview of the spaceflight radiation environmentand its impact on cell biology experimentsrdquo Journal of Gravi-tational Physiology vol 11 no 1 pp 11ndash16 2004

[9] J J W A van Loon ldquoMicro-gravity and mechanomicsrdquo Gravi-tational and Space Biology vol 20 no 2 pp 3ndash18 2007

[10] M Hughes-Fulford ldquoTo infinity and beyond Human space-flight and life sciencerdquo FASEB Journal vol 25 no 9 pp 2858ndash2864 2011

[11] T Fukuda K Fukuda A Takahashi et al ldquoAnalysis of deletionmutations of the rpsL gene in the yeast Saccharomyces cerevisiaedetected after long-term flight on the Russian space stationMirrdquoMutation Research Genetic Toxicology and EnvironmentalMutagenesis vol 470 no 2 pp 125ndash132 2000

[12] A Takahashi K Ohnishi S Takahashi et al ldquoThe effectsof microgravity on induced mutation in Escherichia coli andSaccharomyces cerevisiaerdquo Advances in Space Research vol 28no 4 pp 555ndash561 2001

[13] A Y Lee R P StOnge M J Proctor et al ldquoMapping thecellular response to small molecules using chemogenomicfitness signaturesrdquo Science vol 344 no 6186 pp 208ndash211 2014

[14] G Giaever P Flaherty J Kumm et al ldquoChemogenomic profil-ing identifying the functional interactions of small moleculesin yeastrdquo Proceedings of the National Academy of Sciences of theUnited States of America vol 101 no 3 pp 793ndash798 2004

[15] G Giaever and C Nislow ldquoThe yeast deletion collection adecade of functional genomicsrdquoGenetics vol 197 no 2 pp 451ndash465 2014

[16] T Roemer J Davies G Giaever and C Nislow ldquoBugs drugsand chemical genomicsrdquo Nature Chemical Biology vol 8 no 1pp 46ndash56 2012

[17] S E Pierce R W Davis C Nislow and G Giaever ldquoGenome-wide analysis of barcoded Saccharomyces cerevisiae gene-deletion mutants in pooled culturesrdquo Nature Protocols vol 2no 11 pp 2958ndash2974 2007

[18] A M Smith L E Heisler J Mellor et al ldquoQuantitativephenotyping via deep barcode sequencingrdquo Genome Researchvol 19 no 10 pp 1836ndash1842 2009

[19] Development Core Team R A Language and Environment forStatistical Computing R Foundation for Statistical ComputingVienna Austria 2011

[20] A C Douglas A M Smith S Sharifpoor et al ldquoFunctionalanalysis with a barcoder yeast gene overexpression systemrdquo G3vol 2 no 10 pp 1279ndash1289 2012

[21] Y Benjamini and Y Hochberg ldquoControlling the false discoveryrate a practical and powerful approach to multiple testingrdquoJournal of the Royal Statistical Society Series B vol 57 no 1 pp289ndash300 1995

[22] A M Deutschbauer D F Jaramillo M Proctor et al ldquoMecha-nisms of haploinsufficiency revealed by genome-wide profilingin yeastrdquo Genetics vol 169 no 4 pp 1915ndash1925 2005

[23] D Merico R Isserlin O Stueker A Emili and G DBader ldquoEnrichment map a network-basedmethod for gene-setenrichment visualization and interpretationrdquo PLoS ONE vol 5no 11 Article ID e13984 2010

[24] M E Smoot K Ono J Ruscheinski P-L Wang and T IdekerldquoCytoscape 28 new features for data integration and networkvisualizationrdquo Bioinformatics vol 27 no 3 pp 431ndash432 2011

[25] S Llanos and M Serrano ldquoDepletion of ribosomal protein L37occurs in response to DNA damage and activates p53 throughthe L11MDM2pathwayrdquoCell Cycle vol 9 no 19 pp 4005ndash40122010

[26] G F Ribeiro M Corte-Real and B Johansson ldquoCharacteriza-tion of DNA damage in yeast apoptosis induced by hydrogenperoxide acetic acid and hyperosmotic shockrdquo MolecularBiology of the Cell vol 17 no 10 pp 4584ndash4591 2006

[27] N A Doudican B Song G S Shadel and P W DoetschldquoOxidative DNA damage causes mitochondrial genomic insta-bility in Saccharomyces cerevisiaerdquoMolecular and Cellular Biol-ogy vol 25 no 12 pp 5196ndash5204 2005

[28] F Fang J Hoskins and J S Butler ldquo5-fluorouracil enhancesexosome-dependent accumulation of polyadenylated rRNAsrdquoMolecular and Cellular Biology vol 24 no 24 pp 10766ndash107762004

[29] Y-T Lin J-S Yang S-Y Lin et al ldquoDiallyl disulfide (DADS)induces apoptosis in human cervical cancer Ca Ski cells viareactive oxygen species and Ca2+-dependent mitochondria-dependent pathwayrdquo Anticancer Research vol 28 no 5 pp2791ndash2799 2008


[30] M E Hillenmeyer E Fung J Wildenhain et al ldquoThe chemicalgenomic portrait of yeast uncovering a phenotype for all genesrdquoScience vol 320 no 5874 pp 362ndash365 2008

[31] L Capolongo G Belvedere and M DrsquoIncalci ldquoDNA dam-age and cytotoxicity of mitoxantrone and doxorubicin indoxorubicin-sensitive and -resistant human colon carcinomacellsrdquo Cancer Chemotherapy and Pharmacology vol 25 no 6pp 430ndash434 1990

[32] B Bellosillo D Colomer G Pons and J Gil ldquoMitoxantronea topoisomerase II inhibitor induces apoptosis of B-chroniclymphocytic leukaemia cellsrdquo British Journal of Haematologyvol 100 no 1 pp 142ndash146 1998

[33] P Vejpongsa and E T H Yeh ldquoTopoisomerase 2120573 a promisingmolecular target for primary prevention of anthracycline-induced cardiotoxicityrdquo Clinical Pharmacology and Therapeu-tics vol 95 no 1 pp 45ndash52 2014

Review ArticleRhoGTPases as Key Players in MammalianCell Adaptation to Microgravity

Fiona Louis1 Christophe Deroanne2 Betty Nusgens2

Laurence Vico1 and Alain Guignandon1

1 INSERM U1059 Laboratoire de Biologie du Tissu Osseux Universite Jean Monnet 42023 Saint-Etienne Cedex France2Laboratoire de Biologie des Tissus Conjonctifs GIGA Universite de Liege 4000 Sart Tilman Belgium

Correspondence should be addressed to Alain Guignandon guignanduniv-st-etiennefr

Received 25 April 2014 Revised 14 August 2014 Accepted 9 September 2014

Academic Editor Monica Monici

Copyright copy 2015 Fiona Louis et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

A growing number of studies are revealing that cells reorganize their cytoskeleton when exposed to conditions of microgravityMost if not all of the structural changes observed on flown cells can be explained by modulation of RhoGTPases which aremechanosensitive switches responsible for cytoskeletal dynamics control This review identifies general principles defining cellsensitivity to gravitational stresses We discuss what is known about changes in cell shape nucleus and focal adhesions and try toestablish the relationship with specific RhoGTPase activities We conclude by considering the potential relevance of live imagingof RhoGTPase activity or cytoskeletal structures in order to enhance our understanding of cell adaptation to microgravity-relatedconditions

1 Introduction

Microgravity has been demonstrated to have profound effectson both cellular and molecular levels including changes incell morphology [1 2] alterations of proliferation growthor differentiation [3 4] modification of gene expression [5ndash7] and changes in signal transduction cascades [5 8] Singleundifferentiated cells in vitro respond to altered conditionsof gravity but not all sensors and upstream regulators areknown which limits our understanding of cell sensitivity tomicrogravity-related conditions and evenmore tomicrograv-ity per se

There are numerous observations strengthening the ideathat cytoskeletal structures and cell surface receptors con-nected to them play an important role in the regulation ofthe differentiation potential of stem cells [9] As changes ofshape and of the inner cytoskeletal architecture are com-mon cell responses under conditions of real or simulatedmicrogravity [2] the idea of cytoskeletal involvement in thecellular response to microgravity seems obvious Moreoverstem cells or multipotent cells are recognized as being sen-sitive to mechanical stresses which are known to influence

cell commitment [10 11] The idea that not only terminallydifferentiated cells but also multipotent cells are sensitive tomicrogravity explains why even limited effects on cell com-mitment could have dramatic consequences Small GTPasesof the Rho family are known to control several aspects of celldynamics (vesicular transport traffic cytoskeleton turnover)[12 13] and appear to be the key players when trying togain a better understanding of the effects of microgravity ondifferentiated and multipotent cells

This review first attempts to highlight the fact that struc-tures involved inmechanotransduction pathways are respon-sible for adaptation to microgravity it will be explained thatstructural changes observed in cells exposed to real and sim-ulatedmicrogravitymay result from specific RhoGTPase reg-ulationsThen the degree to which the effects ofmicrogravityare important controllers of multipotent cell commitmentwill be discussed highlighting the critical role of RhoGT-Pases in these regulations The monitoring of RhoGTPaseactivities in conditions of microgravity is still a challenge asit is a dynamic process that controls other highly dynamicprocesses such as actin polymerization or focal adhesionturnover In order to decipher cell adaptation in conditions



of microgravity the community is in need of a live imagingtechnology like the one from Pache et al [15] but that can beset up in flightWe are conscious of all the difficulties of usingForster resonance energy transfer- (FRET-) based biosensorsdedicated to RhoA (Ras homolog gene family member A)and Rac1 (Ras-related C3 botulinum toxin substrate 1) twoimportant actors of this GTPases family under conditionsof microgravity and we are convinced that research groupsthat are successful with these types of sensors will providevery exciting results that will eliminate many confoundingfactors related to conditions of microgravity such as launchvibrations We predict that many specific GAP and GEF(resp RhoGTPases inhibitors and stimulators) will turn outto be key players in cell adaptation to microgravity-relatedconditions in the future

2 Mechanotransductors as Gravity Sensors

Discussions of whether an in vitro single cell or a cell pop-ulation can sense changes in the gravitational field are verycontroversial The currently most unknown research areainvolves the mechanism by which the physical event of g-force susception (by invagination sedimentation or buoy-ancy) becomes the biological process of g-force percep-tion Despite this an enormous body of experimental dataundoubtedly indicates that several types of cultured cellsare sensitive to gravity [16 17] If in fact cells do not fall(collapse) it is because they are supported in some wayThis support takes the form of a mechanical stress set upby the intermolecular forces in response to the distortionproduced by gravity In conditions where gravity is limited(microgravity) (such as those found in an orbiting vehicle)there is thus no distortion produced and consequently thereis no (limited) mechanical stress

It seems that undifferentiated cells have structural ele-ments that may play the role of ldquogravitational sensorsrdquo andldquosenserdquo the intensity of a mechanical tension and that severalintracellular processes can depend on the value of the gravita-tional forceTheoretical considerations suggest that the forcesinvolved are too small to trigger any response to the changedenvironment Several research teams think that these effectsaremostly caused by changes at the tissue and organ level [17]and that such environmental changes are stronger and morediverse [18] (eg lung heart and kidney become larger whilespleen or pancreas get smaller in rats [19]) In conclusiongravitational effects have been considered significant for cellswith a diameter of no less than 10 120583m[20]Thusmicrogravityseems to alter mammalian cells as compared to bacterial cellswhich are normally too small

Actors in the mechanotransduction chain represent keyelements involved in microgravity adaptation Nature pro-vides clear examples of defined mechanoreceptors in eukary-otes such as the statoliths in plants and the otoliths of theinner ear in most species of vertebrates Similar specializedcells of the sense organs detect pressure (touch) and vibra-tions and communicate these physical stimulations to thenerves of the afferent pathway up to the brain

It thus seems that undifferentiated mammalian cells doindeed have structural elements that may play the role of

a ldquogravitational sensorrdquo and ldquosenserdquo the intensity of amechan-ical tension and that many intracellular processes (adhesionproliferation survival contractility migration extracellularmatrix (ECM) architecture gene expression etc) can dependon the intensity of the gravitational forceThe identification ofcell structures capable of acting as gravisensors in in vitro cellsstill remains a problemThe general view of mechanosensingis that the overall cell is sensitive and is not a particularelement

In our opinion the most significant element (primummovens) that may impact on cytoskeletal dynamics undermicrogravity is the displacement of the nucleus The locationof the nucleus is probably dictated by a tension equilibriumbetween the cyto- and nucleoskeletons and we can imaginethat these tensions are constantly changing (in response tosignals) and that the nucleus probably oscillates continuously[21] A microgravity environment may influence the oscil-lating behavior of the nucleus [22] and then trigger a seriesof mechanical adjustments that may modulate cell shapeand structures as well as functions by way of transcriptionactivities

In response to changes in nucleus location cytoskeletalstructures and integrinsmight be solicited for cell adaptationThe cytoskeleton is a network of three interconnected systemsof filaments the actin microfilaments the microtubules andthe intermediate filaments They condition the shape of thecells and the major mechanical functions such as adhesionpolarization directional migration as well as proliferationsurvival or apoptosis gene expression and architecturalorganization of their supporting scaffold [12]

Experiments in real and simulated conditions of micro-gravity have shown that cytoskeletal modulations can occurquickly after variations in gravity have taken place Numer-ousarticles have reported on changes within 30min of theonset of a microgravity simulation affecting from focal adhe-sions to signal transduction Nevertheless cell response canbe observed only after few seconds following gravitationalchanges for example in parabolic flight experiments Afteronly 22 seconds of microgravity ML-1 thyroid cancer cellsshowed no sign of apoptosis or necrosis but the F-actin andcytokeratin cytoskeleton was altered [23] Endothelial cellsalso demonstrated no signs of death (after 31 parabolas of 22seconds) but had a cytoplasmic rearrangement and an alter-ation of cytoskeleton gene expressions [24] Concerningmes-enchymal stem cellsmorphologic characteristics of apoptosiscells (cell shrinkage membrane blebbing nuclear chromatincondensation etc) and decreased cell viability (rate of apop-tosis up to 5695) were reported 12 h after parabolic flightexperiment The F-actin stress fibers and microtubules weredisrupted and the expression of p53 (mRNA and proteinlevels) was upregulated [25] So gravity-induced response ofcells can occur very early within seconds

The reorganization of the cytoskeleton is believed togovern the modifications in size and shape of cells and nucleias well as the patterning number and maturation of focaladhesions The structures of the cytoskeleton nuclei andintegrins may claim to varying degrees to fulfill the role ofgravisensors [26]


The most likely candidates to assume the role of thesestructures are various elements of the cytoskeleton thenucleus intracellular organelles and also certain cell surfacereceptors (integrins) which interact both with cytoskeletalstructures and the extracellular matrix These structures areable to sense constraints and deformations in the matrixwhich are caused either by a gravitational or mechanicalfield and convert this signal into intracellular messengerswhich then give rise to a cellular response to the changes ingravity [21 27] It is also noteworthy that the cytoskeletonand integrins are not the primary sensors but react inresponse to their regulatory proteins (controllers of polymer-izationdestabilization agent)

Numerous cellular processes are controlled by gravity forexample calcium signaling mechanotransduction ligand-receptors interactions and cell-cell communications whichare all linked [28] During these mechanisms cell density isimportant because force transmission is greatest at cell-celland cell-substrate focal contacts where signaling moleculesare concentrated or clustered (ie integrin clustering) [17]Indeed transmission of forces from outside the cell throughcell-matrix and cell-cell contacts appears to control thematu-ration or disassembly of these adhesions which rearrange theorganization and contractile activity of the cytoskeleton Thecytoskeletal tensions formed at adhesions mediate mechan-ical signalling [29] Thus vinculin phosphorylation deter-mines whether cadherins transmit force and can producebiologically distinct functions [30]

In microgravity gravity-induced breakage of cell-celladhesions is reduced So cell-cell interaction was shown tobe promoted in absence of gravity [31] Cell adhesion proteinexpression specifically proteins found in tight junctions andadherens junctions was upregulated resulting in enhancedcell-cell contact between cells (endothelial cells [32]) Alsoincreased levels of E-cadherin were observed in 3D tumorconstructs cultured in simulated microgravity [33]

In osteoblasts a downregulation of cell-cell adhesion pro-teins such as catenin is observed [34] and also a reduc-tion in adhesion proteins such as vinculin and extracellularmatrix proteins such as fibronectin [35] To explain this phe-nomenon Levenberg et al showed that there is an autoreg-ulatory pathway that is activated by the presence of cell-cellor cell-substrate adhesion sites So when cell-cell adhesionis enhanced cell-matrix adhesion is decreased [36] Theseadhesion processes are also dependent on Ca2+ signalingpathways such as cell-cell adhesion via E-cadherinThis Ca2+dependence is through activation of the protein kinase C(PKC) second messenger system as well as activation ofphospholipase C (PLC) which in turn activates a signalingcascade resulting in the release of intracellular Ca2+ [37]This release of intracellular calcium facilitating the bindingof cadherins and 120573-catenin to the actin filaments comprisingthe cytoskeleton resulted in increased strength of cell-cellcontacts [38]

And several teams actually found a calcium release in vas-cular smooth muscle cells after 14 days of hindlimb unload-ing [39] and a downregulation of Calcium channel after

Microgravity

Cytoskeleton disruptionnuclear shape changesion channel activation

Rho-GTPases

Celladhesion

ECM

CSKdynamic

Integrated cell responses cell survival proliferation cell differentiation and stem cell commitment

ROS

Intracelltension

Wnt120573-catsignaling

Cell-cellcontact

Figure 1 Central role of the RhoGTPases in the integrated responseof mammalian cell to microgravity-related conditions A growingnumber of studies are revealing that cells reorganize their cytoskele-ton modulate intracellular tension and initiate nuclear shapeschanges when exposed to conditions ofmicrogravityMost if not allof the structural changes observed on flown cells can be explained bymodulation of RhoGTPases which are mechanosensitive switchesRhoGTPases are known for cytoskeletal dynamics control never-theless they are also involved in many other aspects as discussed inthis review We identify general principles dependent on RhoGT-Pases and define cell sensitivity to gravitational stresses such asoxidative stress intracellular tension cell-cell and cell-ECM adhe-sions and Wnt120573-catenin pathways We will try to establish thatintegrated cellular responses in microgravity are related to specificRhoGTPase activities

28 days [40] Also a reduction in intracellular calcium con-centration is observed after 2 days of simulated microgravityin chondrocytes [41] as well as in neurons [42] Moreoverin neutrophils PKC pathway is inhibited under microgravityleading to a decrease in intracellular concentration of Ca2+[43]

All the structural changes observed in cells subjectedto microgravity-related conditions are dictatedcontrolled bydynamic molecular switches of the GTPase family (Figure 1)Small RhoGTPases mainly control the regulation of intra-cellular traffic and are responsible for cytoskeletal dynamics[44]


RhoA

Rac1

Stabilisation microtubulesLinear actin polymerization in filopodia

Fibrillar actin stabilization

Actinmyosin fibers polymerization

Actinmyosin fibers depolymerization

Branching actin polymerization in lamellipodia+ end elongation microtubules

Fibrillogenesis

Migration

GEF

GAP

GDI

Sensing adhesion

cell-cell interactions

shape and polarity

Cell membrane

ECM

Integrins

Figure 2 RhoGTPase actions on the cytoskeleton and cell dynamics (modified from [14]) Integrins are necessary for translating themechan-ical properties of the extracellular environment into intracellular RhoGTPase-signaling pathways RhoA influences filopodia formation andfocal adhesion assembly and maturation in addition to controlling stress fiber formation and intracellular tension Rac1 primarily controlsactin assembly and formation of lamellipodia to ensure cell migration Fibrillogenesis is controlled positively by RhoA and negatively byRac1 Both RhoA and Rac1 are controlled by specific activators (GEF) and inhibitors (GAP GDI) Cell adaptation to mechanicalgravitationalchallenges triggers activation of pathways integrated by RhoGTPases

3 RhoGTPases MechanosensitiveMolecular Switches

RhoGTPases found in all eukaryotic cells are key regulatorymolecules which link surface receptors to the organizationand turnover of the cytoskeleton govern the formation ofcell-matrix adhesions and uphold the transcriptional controlof gene expression cell survival and proliferation [45] Theyare members of the Ras superfamily of small GTP-bindingproteins and are divided into three major classes RhoARac1 and Cdc42 GTPases are molecular switches that usea simple biochemical strategy to control complex cellularprocesses They switch between two conformational states aguanosine triphosphate- (GTP-) bound (ldquoactiverdquo) state andanother (ldquoinactiverdquo) state related to guanosine diphosphate(GDP) In their inactive forms RhoGTPases are sequestratedin the cytoplasm while upon signaling identified by integrinsand growth factor receptors they switch to their activeforms and translocate to the cell membrane [46] Therethey activate distinct and specific effector molecules whichin turn regulate the organization of the cytoskeleton and cell-matrix adhesions thus controlling cellular activities such asadhesion and also affect cell proliferation and the expressionof specific genes (Figure 2) [12] The cycle between theactive and inactive forms is under the direct control ofthree groups of regulatory proteins The guanine nucleotideexchange factors (GEFs) catalyze the exchange of GDP forGTP to activate Rho proteins The Rho proteins are thendeactivated by GTPase-activating proteins (GAPs) whichincrease the intrinsic GTPase activity of the Rho proteinleading to the hydrolysis of GTP to GDP The third groupof proteins involved in the cycle of Rho signaling is guaninedissociation inhibitors (RhoGDI) which hide the isoprenyl

groups of GTPases an action that promotes the sequestrationof inactive GTPases in the cytosol The RhoGDIs also inhibitthe release of GDP from the GTPase and contribute tothe maintenance of GTPases in an inactive state The Rhoprotein cycle is stimulated by agonists acting through Gprotein-coupled receptors (GPCRs) tyrosine kinase recep-tors cytokine receptor activation and mechanical stressesthat mainly govern the activity of the GEFs [47] The bestknown actions of the RhoGTPases onmechanical parametersof the cytoskeleton can be underscored by the expression ofconstitutively active RhoA and Rac1 in cell lines These mod-elsshow that RhoAactivation leads to better cell spreading butlower mechanical properties while Rac1 activation inducesmechanotransduction [48] As we assume that exposureto gravitational stress is a mechanical stimulation Rac1might be rapidly induced in microgravity-related conditionsThese results reveal the importance of RhoGTPases onmechanosensing cell shape adaptation or signal transduc-tionWewill summarize below the different controls they canhave on cellular mechanisms and metabolism

4 RhoGTPases Control Cytoskeleton Dynamic

In microgravity a qualitative and quantitative analysis of thestructures of F-actin 120573-tubulin and vinculin has revealeda higher density of filamentous actin and a decreased orga-nization in stress fibers Exposing mesenchymal stem cells(MSCs) to low gravity affected the distribution of the differentfilaments and more specifically led to a significant reductionof the F-actin fibers [49 50] extended filopodia increasedperinuclear distribution and decreased density [15 51]Moreover other research groups have found evidence of anaccumulation of actin at the cell border [52 53] This loss of


stress fibers is accompanied by an increase in monomeric G-actin content within the cells The preceded alterations maybe explained by a preferential reduction of RhoA activity

Indeed the activation of RhoAor Rac1 leads to the assem-bly of contractile actinmyosin filaments protrusive actin-rich lamellipodia and protrusive actin-rich filopodia whichin turn give rise to both the formation (actin polymerization)and the organization (filament bundling) of actin filamentsThus a number of studies (eg [54]) have shown that Rhokinase (ROCK) modulates the nonmuscle myosin II (NMM-II) activity by phosphorylation Another protein cofilinregulates actin polymerization and filament elongation Itsphosphorylation leads to inactivation and occurs primarilythrough LIM kinases (LIMK) which are activated by Rac1-dependent kinases Moreover LIMK-dependent phosphory-lation of cofilin can also be induced by RhoA acting throughits target ROCK whichmay be an important event in the sta-bilization of actinmyosin filaments [55] Microgravity leadsto an alteration of the actin cytoskeleton and consequentlyto a decrease of integrin signaling that may be caused by theinhibition of RhoA activity The absence of gravity increasestheG-actin form which reduces cofilin phosphorylation andis consistent with a decrease in focal adhesions and thus stressfibers [56]

Finally if a constitutively active RhoA is overexpresseda recovery stress of the fibers is enabled similar to what canbe observed under normal gravity and integrin signaling isrestored as shown in MSCs [57]

Microtubules play critical roles in eukaryotic cells Theyare key structural elements of the mitotic spindle apparatusduringmitosis and interphase and serve as tracks uponwhichmotor proteins transport vesicles and other componentsmove throughout the cell [58] Several studies have men-tioned perinuclear clustering in the microtubular networkduringmicrogravity [50 59] Also the loss of the radial struc-ture of microtubules has been observed after long stretches oftime (4 h) in microgravity [60]

Microgravity has also been proposed to influence micro-tubules by affecting the self-organization of filaments Accord-ing to the theory on self-organization and in a series of invitro studies with a change in gravity direction [61 62] andmicrogravity [61] it was clearly shown that microtubule self-organization is sensitive to the direction and the magnitudeof gravity which may explain the results obtained undermicrogravity Furthermore the observed disorganization ofmicrotubules may lead to a reduced rate of chromosome seg-regation during mitosis while alterations of actin microfila-ments and focal adhesions may also slow down cytokinesisand thus cell proliferation

RhoGTPases regulate microtubule dynamics in differ-ent ways Rac1 can phosphorylate at Ser16 of the microtu-bule plus-end-binding proteins (stathmins) which occurs inresponse to a number of extracellular stimuli [63] The effectof RhoA on microtubule dynamics is likely to be context-dependent For instance in migrating fibroblasts RhoA pro-motes the formation of stabilized microtubules Also micro-tubules play a major role in defining cell shape and polaritythrough the specific interaction of their plus-ends with pro-teins at the cell cortex This plus-end capture of microtubules

has been attributed to a number of plus-end-binding pro-teins whose activities are influenced by RhoGTPases [12]Altogether results onmicrotubules observed in conditions ofmicrogravity may be explained by an alteration of the RhoAand Rac1 activities

Microgravity has also had an impact on intermediatefilaments which after 12min in microgravity appeared aslarge bundles and aggregates in the vimentin network thatis the most distributed of all intermediate filament proteins[64] ROCK phosphorylates intermediate filament proteinsspecifically at the cleavage furrow during cytokinesis Thiscleavage furrow-specific phosphorylation plays an importantrole in the breakdown of local intermediate filaments andenables an efficient separation of intermediate filament net-works [65] In fact RhoA and Rac1 induce phosphorylationand reorganization of vimentin through kinases such asRhoA-associated protein kinase 2 (ROCK2) p21-activatedkinase (PAK) Src kinase (family of nonreceptor tyrosinekinases) and tyrosine kinases [66]

Concerning lamins which are nuclear intermediate fil-aments Uva et al showed DNA fragmentation in glial cellsafter 30min of microgravity and explained the phenomenonby caspases causing lamina to collapse and chromatin to con-dense [67] Proteins linking nucleoskeleton and cytoskeletoncomplexes (LINC) thus connecting lamina to the cytoskele-ton have been foundWhen it comes to laminopathymodelsin which this LINC complex is disrupted they lead mostlyto RhoA inhibition and lowered cytoplasmic elasticity whileactin and focal adhesion structures are mildly affected [68]Changes in nuclear structures that we identified earlier as animportant initiator ofmicrogravity effects [22] might explainthe RhoA activity inhibition and changes in cell tensionevoked under microgravity

Rac1 was shown to accumulate in the nuclear envelope inaddition to being expressed in the nucleoplasm and seemedto have the same pattern as that reported for lamin B [69]This Rac1 accumulation was proven to promote cell divisionIn microgravity the altered proliferation observed by Dai etal orDammet al [70 71] is controversial since Yuge et al [72]rather found an increased proliferation in human mesenchy-mal stem cellsWe thus suggest based on our results obtainedon rat osteosarcoma [73] that the lower proliferation mightbe explained by a reduced Rac1 activity in conditions ofmicrogravity

5 RhoGTPases as Regulators of Cell Adhesionand Matrix Remodeling

Integrins are transmembrane receptors that mediate theattachment between a cell and its surroundings such as othercells or the ECM In signal transduction integrins conveyinformation about the chemical composition andmechanicalstatus of the environment into the cell Therefore in additionto transmitting mechanical forces they are involved in cellsignaling and the regulation of cell cycles shapes andmotility [74]

Among the ligands of integrins can be mentionedfibronectin vitronectin collagen and lamininThen adapter


proteins such as talin and vinculin link the cytoskeleton tointegrins which attach the cell to the substrate forming afocal adhesion A variety of signaling proteins are associatedwith focal adhesions including focal adhesion kinase (FAK)which is an important mediator of signaling at these centersForces are also transmitted to the substrate at these sitesIn fibroblasts local forces correlate with the area of focaladhesions and actomyosin contractility blocking results in arapid disruption of focal adhesions [75]

In conditions of microgravity a reduced focal adhesion-related area (frequently reported [35 76]) can be explained bythe lower tension applied to the cytoskeleton This situationcan be associated with an inactivation of RhoA and as aresult by decreased fibrillogenesis (fibronectin collagen) dra-matically limiting integrin signaling The proof of a reducedintegrin signaling is thatMSCs have been observed to displaychanges in the expression levels of collagen-specific integrinsafter 7 days of cultivation in a rotational bioreactor [77] Infact activated expression of the 1205722-integrin has been seenduring the course ofMSC differentiation to osteogenesis [53]In addition Loesberg et al found a downregulation of 1205721 1205731and 1205733 integrins after 48 h of simulated microgravity [78]1205731 integrin has been shown to be important formediating

the response of MSCs to mechanical stimulation [79] Uponapplication of fluid shear stress an increase in alkaline phos-phatase (ALP) activity and expression of osteogenic markersis observed along with the activation of FAK and extra-cellular signal-regulated kinase 12 (ERK12) But when 1205731integrins are blocked FAK and ERK12 activation becomesinhibited [79] Phosphorylation of FAKhas also been demon-strated to be important for osteogenic differentiation ofhuman MSCs in response to tension [80] In microgravity-related conditions the limitation of integrin signaling can bea plausible explanation for the reduced osteogenesis

In addition limitation of the integrin-mediated responsecan also reduce important negative regulatory pathwaysThus growth of preadipocytes on a fibronectin matrix inhib-its adipocyte differentiation and this effect is overcome whenactin filaments are disrupted and promotes a rounding-upof cells [81] However 1205731 in association with 1205725 binds tofibronectin and Liu et al [82] reported the presence of anexpression switch from 1205725 to 1205726 at the growth arrest stageof differentiation which is consistent with an ECM changeobserved during adipogenesis This switch is necessary inorder to go from proliferation to differentiation of preadi-pocytes and can be explained by integrins 12057261205731 that bindto laminin and can thus interfere with chromatin and generegulation

These two integrins 1205725 and 1205726 are coordinately regulatedby cyclic adenosine monophosphate (cAMP) InterestinglycAMP has been shown to be activated in microgravity [83ndash85] RhoA and cAMP have antagonistic roles in regulatingcellular morphology [86] Thus the excessive productionof cAMP in microgravity may explain the limitation of RhoAactivation during adipogenesis followed by the integrinswitch of 1205725 to 1205726 to promote adipogenesis Also it is wellestablished that cAMP enhances the expression of bothCCAAT-enhancer-binding proteins (CEBP) 120572 and 120573 [87 88]

and initiates adipogenesis via the transcription factor CREB(cAMP response element binding protein) [89]

Concerning Rac1 cell adhesion to fibronectin (1205725 inte-grin) but not to laminin (1205726 integrin) is particularly effi-cient in activating Rac1 [90] leading to osteogenesis via 120573-cateninWnt pathways [91] In microgravity fibrillogenesis israpidly limited [92 93] which explains the delay or absence ofosteogenesis in multipotent cells The extracellular domainsof cadherins and 120573-catenin provide a link to 120572-catenin andthe actin cytoskeleton [94] Upon tyrosine phosphorylation120573-catenin also plays a significant role in signaling whentranslocated to the nucleus to regulate cell proliferation [95]

Noritake et al [96] have explained the increase in Rac1during osteogenesis until subconfluence cell adhesions accu-mulate E-cadherins at the sites of cell-cell contacts whichinduce Rac1 and thus actin-meshwork formation and 120573-catenin leading to osteogenesis In fact before E-cadherin-mediated cell-cell adhesion is establishedGDP-Rac1 is seques-tered in the cytosol by Rho GDIWhen E-cadherins accumu-late GDP-Rac1 is converted to GTP-Rac1 through the actionof a GEF and is targeted to the plasma membrane releasing120573-catenin linked to E-cadherin which can go to the nucleus[97]

In addition cell-to-cell physical contact via N-cadherinalso plays a crucial role in regulating osteoblastic activitysuch as alkaline phosphatase activity and 120573-catenin signaling[98 99] Consequently reduced cell-cell adhesion observedin microgravity due to limited proliferation may induce adecrease in Rac1 action and osteogenesis

Moreover it has been largely described that matrix rigid-ity affects osteogenesis MSCs grown on collagen-I stiff gels(linking to 1205721 or 1205722-1205731 integrins) have demonstrated acti-vated osteogenesis whereas softer collagen-I gels primeMSCs for a myogenic lineage [100] However cytoskeletonand the dynamicmechanical balance that exists between cellsand their ECM support appear as major players in severalmechanotransduction pathways [74] Microgravity decreasesthe expression of collagen I [101ndash103] induces matrix met-alloproteinases (MMP) production and reduces the levelof fibrillar collagen Thus it could be expected that alteredconditions of gravity may change the mechanical propertiesof ECM (ie the stiffness) Several studies for exampleMcBeath et al or Shih et al [104 105] have shown that oste-ogenic differentiation becomes increased on stiffer matri-ces as evident by type-I collagen osteocalcin Runx2 geneexpressions ROCK FAK and ERK12 induction and alizarinred S staining for mineralization Consequently FAK affectsosteogenic differentiation through ERK12 whereas RhoAand ROCK regulate both FAK and ERK12 [105]

In microgravity an initial modification of cytoskeletaldynamics might be at the origin of the following vicious cir-cle remodeling of a cytoskeleton is associated with a reducedinternal tension (contractility) leading to the dispersion ofFA With such a reduction in FA the cell tension cannotbe restored and fibrillogenesis might be limited Matrixdeposition limitation and MMP activation (Rac1 dependentprocess [106 107]) may further reduce the matrix stiffnessthus amplifying the dispersion of FAand reducing cell tensionand fibrillogenesis After a short exposure (from minutes to


hours) to microgravity-related conditions (before fibrilloge-nesis MMP production) the matrix stiffness is not modifiedWe can thus speculate that the ability of the cells to detect thestiff matrix they are normally seeded on has become rapidlyimpaired Mechanical information is normally conveyed byECM and cells by FA adaptation following tensegrity prin-ciples (equilibrium of internal and external tension) [21] inmicrogravity it seems that the displacement of the nucleus(sensitive to G) conveys the mechanical stimulus and from atensegrity perspective the cell adapts to the reduced tensionby lowering the ECM tension (interruption of fibrillogen-esis and MMP production) The short-term adaptation ofthe cell to microgravity that we have described up to nowseems to be characterized by a rapid reduction of RhoA andan increased Rac1 activity Altogether these studies revealedthat the control of cytoskeleton remodeling by RhoGTPasesimpacts on cell adhesion signaling limiting internal cel-lular tension as well as ECM fibrillogenesis and triggersMMP production thus limiting cell-matrix adhesion andsurvival

6 RhoGTPases in Stem Cell Commitment

In simulated microgravity cellular morphology is drasticallychanged after 7 days The MSCs appear rounder and lessfirmly attached to their substrate than under conditions ofnormal gravity Rather they are very spread out and display afibroblastic morphology [53]

Since the work by McBeath et al we know that theshape of a cell affects its differentiation potential [104] ThusMSCs that have been allowed to adhere over a larger areaare able to differentiate towards the osteogenic lineage whilecells adhering to a smaller area are restricted to the adi-pogenic lineage These impacts on lineage commitment bymesenchymal stem cells seem to be regulated by shape-induced changes in the RhoGTPase activity and cytoskeletaltension [108] Yao et al [109] showed that the cell shape itselfis an inherent cue to regulate stem cell differentiation bothwith and without external chemical induction factors Thusaccording to McBeath et al [104] expressing dominant-negative RhoA causes MSCs to become adipocytes whileconstitutively active RhoA induces osteoblastic or myocyticdifferentiation [110 111]

Concerning Rac1 it has been shown to promote cell adhe-sion and spreading and thereby to prevent the cell shapechange and the establishment of the cortical actin structurenecessary for adipocyte formation [109] Adhering cells arecharacterized by an elaborate network of stress fibers andfocal adhesions and are thus more prone to adopt a fibroblas-tic cell shape reflecting cytoskeleton tension [112 113] whichseems to be altered in conditions of microgravity

The cell shape may also depend on the available substratearea and hence the cell density However if cellular growth isreduced in microgravity the cell density will also be alteredGao et al [110] found that levels of RhoA activity did notvary substantially but that the Rac1 activity was significantlyhigher in well-spread cells during early differentiation than inhigh-density cells

They also demonstrated that Rac1 is necessary for osteo-genesis and that constitutively active Rac1 inhibited adipoge-nesis even if it is important for adipose commitment Liuet al [82] showed that an increase in preadipocyte densityinhibited the RhoA activity and that a downregulation of theRhoA-ROCK pathway was required for both adipose lineagecommitment and maturation [104 111] An increased celldensity thus appeared to be critically important

GTPases have also been shown to act in the cell cyclemitosis and cytokinesis RhoGTPases influence the cyclin-dependent kinase (cdk) activity during the G1-Phase of thecell cycle Thus RhoGTPases control the organization ofthe microtubule and actin fibers during cell cycling Aninhibition of RhoA or Rac1 blocks the G1 progression in avariety of mammalian cell types [114 115] Also Rac1 (but notRhoA) stimulates cyclin D1 transcription mediated by NF-120581B (nuclear factor kappa-light-chain-enhancer of activated Bcells) [116 117]Thus the necessity to downmodulate the Rac1activity in adipogenesis is that Rac1may prolong proliferationof preadipocytes which is consistent with the reported effectsof Rac1 on cyclin D1 [90 118 119] In fact Rac1 accumulates inthe nucleus during theG2phase of the cell cycle andpromotescell division [69] Concerning the cell division itself it hasbeen shown that actinmyosin filaments under the controlof ROCK are required at the cortex to allow positioning ofthe centrosomes [120] RhoA also plays a crucial role in thecontractile ring function [121]

Microgravity affects the growth proliferation and dif-ferentiation of osteoblasts Since the inhibition of RhoAobserved undermicrogravity blocksG1 progression [114 115]this may explain the altered proliferation and differentiationof osteoblastic cells and increased adipogenesis as summa-rized in Figure 3

Furthermore several cytoskeletal components includingRac1 GTPase activating protein 1 (Rac-GAP1) and Tropo-modulin 1 segregate asymmetrically during stem cell divi-sion and overexpression of these proteins may enhanceMSCcommitment as already proven with asymmetrical divisionsof hematopoietic stem cells to progenitor cells [122]

7 RhoGTPases and Wnt120573-CateninSignaling Crosstalk

Three Wnt signaling pathways have been characterized thecanonical Wnt pathway the noncanonical planar cell polar-ity pathway and the noncanonical Wntcalcium pathwayThe canonical Wnt pathway leads to regulation of genetranscription the noncanonical planar cell polarity pathwayregulates the cytoskeleton via a RhoGTPase regulation thatis responsible for the shape of the cell and the noncanonicalWntcalcium pathway regulates calcium inside the cell [123]

Mellor et al found that Wnt signaling was inhibited inconditions of microgravity [124] and mouse osteoblasts sub-jected to simulated microgravity were found to have lowerlevels of several components of the Wnt120573-catenin signalingpathway This may indicate even indirectly the activation ofan adipogenic program under microgravity [125] MoreoverWan et al [126] recently demonstrated a changed RhoA and


Microgravity

cAMP

RhoA

Rac1

E-cadherins N-cadherins

120573-Catenin

FAK ERK12

OsteoblastogenesisAdipogenesis

On MSC

12057251205731

12057261205731

CEBP120572et 120573

Figure 3 Role of AMPc on RhoGTPases activities and commitment of multipotent cells Microgravity affects the growth proliferation anddifferentiation ofmultipotent cells by increasingAMPc production AMPc contributes to cytoskeleton reorganization as it regulates negativelyRhoA activity Limitation of osteoblastogenesis might be linked to the ability of microgravity to reduce RhoA and Rac1 activities RhoAand Rac1 activations support osteoblasts differentiation for their respective role in ERK activation and beta-catenin nuclear translocationSustained adipogenesis observed in microgravity-related condition might be linked to ability of AMPc to trigger integrin a5b1a6b1 switchSignaling through a6b1 integrin is known to support adipogenesis A direct activation of adipogenic transcription factors (cEBPs) by AMPchas been also described

120573-catenin signaling after 1 and 25 h respectively in clinoro-tated osteoblasts They revealed that both the RhoA activityand the TCFLEF (T-cell factor-1 and lymphoid enhancingfactor-1) activity a 120573-catenin recruiter were downregulatedby unloading However the inhibition of 120573-catenin signalingblocked the unloading-inducedRhoA suppression and dom-inant negative RhoA inhibited the TCFLEF suppressionrevealing a regulation loop between 120573-catenin RhoA andTCFLEF Furthermore while 120573-catenin signaling seemedto be required for microgravity regulation of RhoA thisresponse was not mediated by the actin cytoskeleton or intra-cellular tension [126] The same was observed for Rac1120573-catenin signaling [91]

The Wnt canonical pathway involves the translocationof 120573-catenin to the nucleus and 120573-catenin has been shownto promote osteogenic differentiation in early osteoblastprogenitors in vivo [127] In contrast other studies have sug-gested that canonical Wnt signaling may actually promotestem cell renewal and inhibit osteogenic differentiation ofosteoprogenitor cells in vivo [128] as well as promoting stemcell renewal in human MSCs derived from bone marrow[129] Arnsdorf and colleagues [130] investigated the role ofnoncanonical Wnt signaling in mechanically induced osteo-genic differentiation of MSCs Exposure of MSCs to oscilla-tory fluid flow resulted in a translocation of 120573-catenin [131]and an upregulation of Wnt5a which is capable of inducingboth canonical and noncanonical pathways Wnt5a is alsonecessary for the flow-induced activation of RhoA Howeverthe inhibition of Wnt5a did not affect the 120573-catenin translo-cation which may instead be regulated by cadherin-catenin

signaling In addition Santos et al [132] showed that theactivation of the RhoAROCK pathway by Wnt5a induceda downregulation of adipogenic markers It was furtherreported that RhoA could also be activated by Wnt3a one ofthe canonical Wnt family members [133] and that an inhi-bition of intracellular 120573-catenin decreased the RhoA activity[134]

Kim et al [135] also found that Wnt signaling regulatedthe MSC differentiation into cardiomyocyte-like cells with aconcomitant downregulation of RhoA and upregulation ofRac1 Concerning Rac1 it was shown to be a critical regulatorin shear stress-driven 120573-catenin signaling in osteoblasts [91]and constitutively active Rac1 mutant caused a significantenhancement of the TCFLEF activity

These studies demonstrate that Wnt signaling is impor-tant for mechanically induced differentiation through RhoAor Rac1 pathways So in conditions of microgravity reducedRhoA cell shape and migratory behaviors can be explainedby Wnt and 120573-catenin signaling Finally RhoGTPases areregulated by Wnt signaling but in return 120573-catenin location(translocation) is dependent on RhoGTPases This intricateinterplay between both regulatory elements makes themparticularly important for the interpretation of microgravityeffects

8 RhoGTPases and Oxidative Stress

One of the first targets of Rac1 to be identified was p67phoxan essential structural component of the NADPH oxidasecomplex [136] Since then Rac1 has been reported to promote


reactive oxygen species (ROS) production in many cellsand to mediate the activity of the Nox family [137 138]Consequently Rac1 activation leads to the generation ofROS enabling adipogenesis commitment [139] and reducingosteoblastogenesis [140 141] Moreover GTPases act on theantioxidant master gene Nrf2 (nuclear factor-like 2) whichactivates a protective adaptive response to oxidative stressthrough transcriptional activation of antioxidant defensegenes [142]

RhoA is involved in Nrf2 phosphorylation which is nec-essary for its activation [143] Nrf2 is a transcription factorfor Hace1 (HECT domain and ankyrin repeat containing E3ubiquitin protein ligase 1) and Hace1 binds and ubiquitylatesRac1 when the latter is associated with NADPH oxidase thusblocking ROS generation byNOX [143 144] So RhoA activa-tion may limit ROS production and adipogenesis while Rac1activation may support it However several research groupshave reported that ROS causes RhoA activation [145 146]while Nimnual et al demonstrated that Rac1-mediated ROSproduction results in the downregulation of theRhoAactivity[147] This is also required for Rac1-induced formation ofmembrane ruffles and integrin-mediated cell spreading TheGTPase regulation by oxidative cell status thus still remainsunclear

In line with these papers several research groups suchas Versari et al have found increased oxidative stress duringspace flight due to microgravity [148 149] and cosmic radia-tions [150] As RhoA is decreased in microgravity this couldexplain the increased production of reactive oxygen speciesAccording to this paper we can assume that Rac1 activitiesare increased in microgravity An upregulated Rac1 activityfits well with enhanced ROS production and improved adi-pogenesis

However a higher Rac1 activity is also consistent witha higher ability for cell migration [151 152] Neverthelessresults of migration in space are controversial Bone marrowcells from rats and human embryonic brain cells show a facil-itated cell migration [153 154] while bone marrow CD34+cells have a lower migration potential in simulated micro-gravity [155] We can interpret the apparent discrepancies inmigration results based on the time spent in microgravityfor short-term exposure (from minutes to hours) there areseveral reasons to believe that RhoA is decreased and Rac1increased in line with their reciprocal inhibition [156] butfor longer exposure (from hours to days) the Rac1-inducedROS productionmay increase RhoA activation [145 146] andreduce the Rac1 activity limiting migration capabilities Themissing information in microgravity is related to the lack ofmeasurements of specific RhoGTPase activities

9 RhoGTPases Activities Monitoringin Microgravity

Meyers et al showed a reduction in active RhoA (minus88(plusmn2)) and a decrease in phosphorylation of cofilin after7 days in microgravity in addition to the absence of stressfibers [56] If overexpression of active RhoA is carried outthis enables a recovery of stress fibers and restored integrin

signals similar to those observed in normal gravity in MSC[57] In simulated microgravity a decrease in RhoA activitywas also observed after 72 h [157 158] Unfortunately nothingis known about Rac1 activity Zayzafoon et al thus proposedamodel in which the cytoskeleton is actually not the first sen-sor but a secondary step affected by a gravity-sensitive sensorIn this model it is the RhoA inactivation that is followedby cytoskeletal changes and transduction at FAs [57] whichexplains the alterations on MSC differentiations observedin microgravity To our knowledge our team is the first toperform RhoA and Rac1 monitoring during osteogenesisand adipogenesis in simulatedmicrogravity using embryonicmesenchymal stem cells C3H10T12 multipotent cells werecultured in modeled microgravity using NASArsquos rotatingwall vessels (RWV) or in control cultures under conditionsof earth gravity for up to 8 days seeded on collagen-coatedmicrobeads (Cytodex 3 Sigma) The results presented inFigure 4 show significant decreases in both RhoA and Rac1after long-term exposure to simulated microgravity To ourknowledge no comparison can bemade with data obtained inreal microgravity unfortunately Regardless of the limitationof themodel when it comes to simulatedmicrogravity-relatedconditions these results clearly showed that downregulationsof RhoA and Rac1 were compatible with enhanced adipoge-nesis and limited osteogenesis

As preservation of active RhoGTPases in flight conditionmight be challenging the recent validation of biosensorsfor imaging of active RhoA Rac1 and Cdc42 represents animportant step in understanding cell responses tomicrograv-ity Despite the critical role of RhoGTPases that we describein this review there is a dramatic lack of data concerning themonitoring of their activities during exposure to micrograv-ity particularly in real microgravity These data are of crucialimportance since cell adaptation is a dynamic process weneed to use available technologies such as fused fluorescentproteins and biosensors dedicated to following RhoGTPaseactivities in order to decipher cell adaptation in conditions ofmicrogravity On ground experiments extensive biochemicaland profiling studies on mechanotransduction pathways canbe performed In an automated spaceflight the use of biosen-sors specific to molecules integrating many pathways such asRhoGTPases should be presented as a simplified and inte-grated view of cell mechanics The community is in needof a live imaging data (already validated on ground [159])that can be now used in flight conditions We believe thatgroups that are successful in providing this type of integrateddata will surprise our community whose thinking is limitedby analysis of fixed images of cells and the monitoring ofindividual parameters

10 Conclusion

RhoGTPases represent a unique hub for integration of bio-chemical and mechanical signals As such they are probablyvery rapidly involved in a cellrsquos adaptation to microgravity-related conditions Published data describing this adaptationhave reported on alterations of the cytoskeleton adhesionand fibrillogenesis as well as an enhancement of the ROS


RhoA active assay-AD media

D6 120583G D6 1G0

10

20

30

40

Time (days)

( o

f pos

itive

cont

rol120583

g A

DN

)

lowast

(a)

RhoA active assay-OB media

D6 120583G D6 1G0

20

40

60

80

100P = 005

Time (days)

( o

f pos

itive

cont

rol120583

g A

DN

)

(b)

Rac1 active assay-AD media

D6 120583G D6 1G0

20

40

60

80

Time (days)

( o

f pos

itive

cont

rol120583

g A

DN

)

lowastlowast

(c)

Rac1 active assay-OB media

D6 120583G D6 1G0

100

200

300P = 005

Time (days)

( o

f pos

itive

cont

rol120583

g A

DN

)

(d)

Figure 4 RhoA and Rac1 activities are downregulated after 6 days of culture in simulated-microgravity conditions Cultures were performedwith C3H10T12 (multipotent embryonic cells) on collagen-coated microbeads (Cytodex 3 Sigma) for adipogenic induction and on Cytodex3 beads coated with apatite minerals complexed to collagen for an osteogenic oneThe adipogenic media contained 1 120583Mof rosiglitazone andthe osteogenic media 5mgmL of L-ascorbic acid 120573-glycerophosphate at 10minus3M and retinoic acid at 10minus5M in 120572MEMMicrobeads with cellswere cultured for 2 days in 90mm petri dishes (untreated for culture) with 10mL of proliferation media (120572MEM) after which the cells wereswitched 2 days in differentiated media and finally left for 6 days in a NASA rotating wall vessel (RWV) In parallel controls were realized byculturing beads in petri dishes RhoA and Rac1 active assays were performed with specific G-LISA kits (cytoskeleton) The positive controlswere pure active proteins of RhoA and Rac1 provided with the kit The results are expressed as percentage of the positive controls they showstandard error of themean (SEM) of samples extracted from three independent experiments and are compared with Studentrsquos statistical 119905-test

production and migration that can be explained by the spe-cific regulation of RhoGTPases To summarize the literaturewe can speculate that after a short exposure of a cell tomicrogravity the RhoA activity is depressed and the Rac1activity increased For long-term exposure osteogenesis hasbeen reported to be impaired and adipogenesis promotedThese changes in multipotent cell commitment fit nicelywith prolonged depressed activities of both RhoA and Rac1(Figure 5)

As we are convinced that focal adhesion and F-actinfibers are not the primary sensors of microgravity-relatedsignals (but rather transducers or effectors of the response)many specific GAP and GEF (resp RhoGTPase inhibitorsand stimulators) will emerge as new players in the adaptation

of cells to microgravity-related conditions What are themechanisms that explain the activation or inhibition of theseGTPases regulators As we try to establish that mechanosen-sors are involved in cell adaptation to microgravity we canpredict that critical players identified in these extreme con-ditions will in return be recognized in the mechanobiologyfield

Abbreviations

ALP Alkaline phosphataseCEBP CCAAT-enhancer-binding proteinscAMP Cyclic adenosine monophosphateCREB cAMP response element-binding protein


ECM

RhoATension Rac1

Migration

Earth gravity

(a)

ECM (Stiff)

RhoATension Rac1Smallernucleus

Migration

Microgravity

Short term(minutes to hours)

ROS

TranscriptionFocal

adhesions

(b)

ECM

RhoA Rac1

Migration

Focal adhesions

(Soft)

Anoikis

Fibrillo-genesis

Microgravity

Long term(hours to days) Osteogenesis

myogenesis Adipogenesis

Actin stress fibersMicrotubules

Intermediate filamentsPerinuclear actin

(c)

Figure 5 Proposedmodels describing the regulations of RhoA andRac1 activities in space-related conditions On EarthMSCs are well spreadand exhibit a tensed cytoskeleton in particular of microtubules intermediate filaments and actin stress fibers associated with stable focaladhesions within the extracellular matrixThese elements are controlled by GTPases RhoA and Rac1 We hypothesize that during short-termexposure to microgravity RhoAmight be inhibited to allow cytoskeleton reorganization in respect to the newmechanical status Cell tensionreduction might be mandatory during this adaptation At the same time Rac1 is activated to control peripheral actin polymerization andinduces ROS production All these events lead rapidly to a rounder cell shape with disorganization of microtubules stress fibers intermediatefilaments and focal adhesions Transcription may be also altered as nucleus shape is changed In these conditions cell is still able to migrateWhen exposure to microgravity is prolonged both RhoA and Rac1 might be inhibited explaining decreases in osteogenesis and myogenesisand enhancement of adipogenesis of MSCs In addition RhoA inhibition limits fibrillogenesis (a tension-dependent process) extracellularmatrix is not properly synthesized and lost its mechanical properties appearing softer for MSCs reinforcing adipogenesis At that timemigration is inhibited consistent with cytoskeleton alterations and Rac1 decrease MSCs become very round have low adhesion and mayinitiate anoikis

CSK CytoskeletonECM Extracellular matrixERK12 Extracellular signal-regulated kinase 12FAK Focal adhesion kinaseFRET Forster resonance energy transferGAPs GTPase-activating proteinsGDIs Guanine dissociation inhibitorsGDP Guanosine diphosphate

GEFs Guanine nucleotide exchange factorsGPCR G protein-coupled receptorGTP Guanosine triphosphateHace1 HECT domain and ankyrin repeat

containing E3 ubiquitin protein ligase 1LIMK LIM kinasesLINC Proteins linking nucleoskeleton and

cytoskeleton complexes


MMPs Matrix metalloproteinasesMSC Mesenchymal stem cellNF-120581B Nuclear factor

kappa-light-chain-enhancer of activatedB cells

NMM-II Nonmuscle myosin IINrf2 Nuclear factor (erythroid-derived 2-)

like 2PAK p21-activated kinaseRac1 Ras-related C3 botulinum toxin

substrate 1RhoA Ras homolog gene family member AROCK Rho kinaseROCK2 RhoA-associated protein kinase 2ROS Reactive oxygen speciesRWV Rotating wall vesselsSEM Standard error of the meanSrc family kinase Family of nonreceptor tyrosine kinasesTCFLEF T-cell factor-1 (Tcf-1) and lymphoid

enhancing factor-1 (Lef-1)


The authors declare that there is no conflict of interestsregarding to the publication of this paper

Acknowledgment

Thestudywas partially funded by the European SpaceAgency(Microgravity Application ProgramMAP ldquoERISTOrdquo) (Euro-pean Research in Space and Terrestrial Osteoporosis Con-tract no 1423200NLSH) and the French Centre NationaldrsquoEtudes Spatiales (CNES)

References

[1] P J Rijken R P de GrootW Briegleb et al ldquoEpidermal growthfactor-induced cell rounding is sensitive to simulated micro-gravityrdquo Aviation Space amp Environmental Medicine vol 62 no1 pp 32ndash36 1991


[3] A Cogoli ldquoFundamentals of space biology research on cellsanimals and plants in spacerdquo inCell Biology G Clement andKSlenzka Eds pp 121ndash170 Springer New York NY USA 2006

[4] D Grimm P Wise M Lebert P Richter and S Baatout ldquoHowand why does the proteome respond to microgravityrdquo ExpertReview of Proteomics vol 8 no 1 pp 13ndash27 2011

[5] R P de Groot P J Rijken J Boonstra A J Verkleij S W deLaat andW Kruijer ldquoEpidermal growth factor-induced expres-sion of c-fos is influenced by altered gravity conditionsrdquo Avia-tion Space and Environmental Medicine vol 62 no 1 pp 37ndash401991

[6] T G Hammond F C Lewis T J Goodwin et al ldquoGene expres-sion in spacerdquo Nature Medicine vol 5 no 4 p 359 1999

[7] Y Liu and EWang ldquoTranscriptional analysis of normal humanfibroblast responses tomicrogravity stressrdquoGenomics Proteom-ics and Bioinformatics vol 6 no 1 pp 29ndash41 2008


[9] P S Mathieu and E G Loboa ldquoCytoskeletal and focal adhesioninfluences on mesenchymal stem cell shape mechanical prop-erties and differentiation down osteogenic adipogenic andchondrogenic pathwaysrdquo Tissue EngineeringmdashPart B Reviewsvol 18 no 6 pp 436ndash444 2012

[10] J C Chen and C R Jacobs ldquoMechanically induced osteogeniclineage commitment of stem cellsrdquo Stem Cell Research andTherapy vol 4 article 107 no 5 2013

[11] J Eyckmans G L Lin and C S Chen ldquoAdhesive and mechan-ical regulation of mesenchymal stem cell differentiation inhuman bonemarrow and periosteum-derived progenitor cellsrdquoBiology Open vol 1 no 11 pp 1058ndash1068 2012

[12] A B Jaffe and A Hall ldquoRho GTPases biochemistry andbiologyrdquo Annual Review of Cell and Developmental Biology vol21 pp 247ndash269 2005

[13] A Hall ldquoG proteins and small GTpases distant relatives keep intouchrdquo Science vol 280 no 5372 pp 2074ndash2075 1998

[14] M Raftopoulou and A Hall ldquoCell migration rho GTPases leadthe wayrdquoDevelopmental Biology vol 265 no 1 pp 23ndash32 2004

[15] C Pache J Kuhn K Westphal et al ldquoDigital holographicmicroscopy real-time monitoring of cytoarchitectural altera-tions during simulated microgravityrdquo Journal of BiomedicalOptics vol 15 no 2 Article ID 026021 2010

[16] L B Buravkova ldquoProblems of the gravitational physiology of acellrdquo Human Physiology vol 36 no 7 pp 746ndash753 2010

[17] D Ingber ldquoHow cells (might) sense microgravityrdquo The FASEBJournal vol 13 pp S3ndashS15 1999

[18] I D Pestov ldquoFundamentals of gravitational biologyrdquo Kosmich-eskaia Biologiia i Meditsina vol 2 no 1 9 pages 1997

[19] M Miyake M Yamasaki A Hazama S Nielsen and TShimizu ldquoEffects of microgravity on organ development of theneonatal ratrdquo Uchu Seibutsu Kagaku vol 18 no 3 pp 126ndash1272004

[20] E C Pollard ldquoTheoretical studies on living systems in theabsence ofmechanical stressrdquo Journal ofTheoretical Biology vol8 no 1 pp 113ndash123 1965

[21] D E Ingber N Wang and D Stamenovic ldquoTensegrity cellularbiophysics and the mechanics of living systemsrdquo Reports onProgress in Physics vol 77 no 4 2014

[22] R G Bacabac T H Smit J J W A van Loon B Z DoulabiM Helder and J Klein-Nulend ldquoBone cell responses to high-frequency vibration stress does the nucleus oscillate within thecytoplasmrdquo The FASEB Journal vol 20 no 7 pp 858ndash8642006

[23] C Ulbrich J Pietsch J Grosse et al ldquoDifferential gene reg-ulation under altered gravity conditions in follicular thyroidcancer cells relationship between the extracellular matrix andthe cytoskeletonrdquo Cellular Physiology and Biochemistry vol 28no 2 pp 185ndash198 2011

[24] J Grosse M Wehland J Pietsch et al ldquoShort-term weight-lessness produced by parabolic flight maneuvers altered geneexpression patterns in human endothelial cellsrdquo The FASEBJournal vol 26 no 2 pp 639ndash655 2012

[25] R Meng H-Y Xu S-M Di et al ldquoHuman mesenchymal stemcells are sensitive to abnormal gravity and exhibit classic apop-totic featuresrdquo Acta Biochimica et Biophysica Sinica vol 43 no2 pp 133ndash142 2011


[26] M G Tairbekov ldquoMolekulyarnye i kletochnye osnovy gravitat-sionnoi chuvstvitelrsquonosti (Molecular andCellular Fundamentalsof Gravitational Sensitivity)rdquo 2002

[27] C A Lambert C M Lapiere and B V Nusgens ldquoBiology ofadherent cells in microgravityrdquo in Biology in Space and Life onEarth R N Enno Brinckmann Ed pp 123ndash155 Wiley-VCHNew York NY USA 2007

[28] T B Damm and M Egli ldquoCalciumrsquos role in mechanotransduc-tion during muscle developmentrdquo Cellular Physiology and Bio-chemistry vol 33 no 2 pp 249ndash272 2014

[29] C S Chen J Tan and J Tien ldquoMechanotransduction at cell-matrix and cell-cell contactsrdquo Annual Review of BiomedicalEngineering vol 6 pp 275ndash302 2004

[30] J L Bays X Peng C E Tolbert et al ldquoVinculin phosphory-lation differentially regulates mechanotransduction at cell-celland cell-matrix adhesionsrdquo Journal of Cell Biology vol 205 no2 pp 251ndash263 2014

[31] N L Cowger E Benes P L Allen and T G HammondldquoExpression of renal cell proteinmarkers is dependent on initialmechanical culture conditionsrdquo Journal of Applied Physiologyvol 92 no 2 pp 691ndash700 2002

[32] G L Sanford D Ellerson C Melhado-Gardner A E Sroufeand S Harris-Hooker ldquoThree-dimensional growth of endothe-lial cells in the microgravity-based rotating wall vessel bioreac-torrdquo In Vitro Cellular amp Developmental Biology Animal vol 38no 9 pp 493ndash504 2002

[33] M IngramG B Techy R Saroufeem et al ldquoThree-dimensionalgrowth patterns of various human tumor cell lines in simulatedmicrogravity of a NASA bioreactorrdquo In Vitro Cellular amp Devel-opmental BiologymdashAnimal vol 33 no 6 pp 459ndash466 1997

[34] H L Nichols N Zhang and X Wen ldquoProteomics and genom-ics of microgravityrdquo Physiological Genomics vol 26 no 3 pp163ndash171 2006

[35] A Guignandon M H Lafage-Proust Y Usson et al ldquoCellcycling determines integrin-mediated adhesion in osteoblasticROS 1728 cells exposed to space-related conditionsrdquo TheFASEB journal vol 15 no 11 pp 2036ndash2038 2001

[36] S Levenberg B-Z Katz K M Yamada and B Geiger ldquoLong-range and selective autoregulation of cell-cell or cell-matrixadhesions by cadherin or integrin ligandsrdquo Journal of CellScience vol 111 no 3 pp 347ndash357 1998

[37] J A Felix V V Chaban M L Woodruff and E R DirksenldquoMechanical stimulation initiates intercellular Ca2+ signaling inintact tracheal epitheliummaintained under normal gravity andsimulated microgravityrdquo American Journal of Respiratory Celland Molecular Biology vol 18 no 5 pp 602ndash610 1998

[38] K S Ko P D Arora V Bhide A Chen and C A McCullochldquoCell-cell adhesion in human fibroblasts requires calcium sig-nalingrdquo Journal of Cell Science vol 114 part 6 pp 1155ndash11672001

[39] P N Colleran B J Behnke M K Wilkerson A J Donatoand M D Delp ldquoSimulated microgravity alters rat mesentericartery vasoconstrictor dynamics through an intracellular Ca2+releasemechanismrdquoAmerican Journal of Physiology RegulatoryIntegrative and Comparative Physiology vol 294 no 5 ppR1577ndashR1585 2008

[40] M-J Xie Y-G Ma F Gao et al ldquoActivation of BKCa channel isassociated with increased apoptosis of cerebrovascular smoothmuscle cells in simulated microgravity ratsrdquo American Journalof Physiology Cell Physiology vol 298 no 6 pp C1489ndashC15002010

[41] X Li S Yang S Li P Jiang and Z Lin ldquoEffects of simulatedmicrogravity on the alkaline phosphatase activity and intracel-lular calcium concentration of cultured chondrocytesrdquo ChineseScience Bulletin vol 44 no 3 pp 218ndash221 1999

[42] K Meissner J R Piqueira and W Hanke ldquoFluorescent anddispersion experiments on biological membranes under micro-gravityrdquo Journal of Gravitational Physiology vol 11 no 2 ppP195ndashP196 2004

[43] A Sundaresan D Risin and N R Pellis ldquoLoss of signal trans-duction and inhibition of lymphocyte locomotion in a ground-basedmodel ofmicrogravityrdquo InVitro CellularampDevelopmentalBiologymdashAnimal vol 38 no 2 pp 118ndash122 2002

[44] A Hall ldquoRho GTpases and the actin cytoskeletonrdquo Science vol279 no 5350 pp 509ndash514 1998

[45] S Etienne-Manneville and A Hall ldquoRho GTPases in cell biol-ogyrdquo Nature vol 420 no 6916 pp 629ndash635 2002

[46] A J Ridley and A Hall ldquoThe small GTP-binding protein rhoregulates the assembly of focal adhesions and actin stress fibersin response to growth factorsrdquo Cell vol 70 no 3 pp 389ndash3991992

[47] M Chiariello J P Vaque P Crespo and J S GutkindldquoActivation of Ras and Rho GTPases and MAP Kinases by G-protein-coupled receptorsrdquo Methods in Molecular Biology vol661 pp 137ndash150 2010

[48] S Servotte Z Zhang C A Lambert et al ldquoEstablishmentof stable human fibroblast cell lines constitutively expressingactive Rho-GTPasesrdquo Protoplasma vol 229 no 2ndash4 pp 215ndash220 2006

[49] MHughes-Fulford andM L Lewis ldquoEffects ofmicrogravity onosteoblast growth activationrdquo Experimental Cell Research vol224 no 1 pp 103ndash109 1996


[51] S I M Carlsson M T S Bertilaccio E Ballabio and J AMMaier ldquoEndothelial stress by gravitational unloading effectson cell growth and cytoskeletal organizationrdquo Biochimica etBiophysica ActamdashMolecular Cell Research vol 1642 no 3 pp173ndash179 2003

[52] L B Buravkova P M Gershovich J G Gershovich and A IGrigorrsquoev ldquoMechanisms of gravitational sensitivity of osteogenicprecursor cellsrdquo Acta Naturae vol 2 no 1 pp 28ndash36 2010

[53] V E Meyers M Zayzafoon S R Gonda W E Gathingsand J M McDonald ldquoModeled microgravity disrupts colla-gen Iintegrin signaling during osteoblastic differentiation ofhuman mesenchymal stem cellsrdquo Journal of Cellular Biochem-istry vol 93 no 4 pp 697ndash707 2004

[54] F Matsumura ldquoRegulation of myosin II during cytokinesis inhigher eukaryotesrdquo Trends in Cell Biology vol 15 no 7 pp 371ndash377 2005

[55] K Ohashi K Nagata M Maekawa T Ishizaki S Narumiyaand K Mizuno ldquoRho-associated kinase ROCK activates LIM-kinase 1 by phosphorylation at threonine 508 within the activa-tion looprdquo The Journal of Biological Chemistry vol 275 no 5pp 3577ndash3582 2000

[56] V E Meyers M Zayzafoon J T Douglas and J M McDonaldldquoRhoA and cytoskeletal disruption mediate reduced osteoblas-togenesis and enhanced adipogenesis of human mesenchymalstem cells in modeled microgravityrdquo Journal of Bone and Min-eral Research vol 20 no 10 pp 1858ndash1866 2005


[57] M Zayzafoon V E Meyers and J M McDonald ldquoMicrograv-ity the immune response and bonerdquo Immunological Reviewsvol 208 no 1 pp 267ndash280 2005

[58] C E Walczak ldquoMicrotubule dynamics and tubulin interactingproteinsrdquo Current Opinion in Cell Biology vol 12 no 1 pp 52ndash56 2000

[59] F Yang ZDai Y Tan andY Li ldquoEffects of altered gravity on thecytoskeleton of neonatal rat cardiocytesrdquo Microgravity Scienceand Technology vol 22 no 1 pp 45ndash52 2010


[61] J Tabony N Rigotti N Glade and S Cortes ldquoEffect of weight-lessness on colloidal particle transport and segregation in self-organising microtubule preparationsrdquo Biophysical Chemistryvol 127 no 3 pp 172ndash180 2007


[63] H Daub K Gevaert J Vandekerckhove A Sobel and A HallldquoRacCdc42 and p65PAK regulate the microtubule-destabiliz-ing protein stathmin through phosphorylation at serine 16rdquoTheJournal of Biological Chemistry vol 276 no 3 pp 1677ndash16802001

[64] L Sciola M Cogoli-Greuter A Cogoli A Spano and P PippialdquoInfluence of microgravity on mitogen binding and cytoskele-ton in Jurkat cellsrdquo Advances in Space Research vol 24 no 6pp 801ndash805 1999

[65] H Goto H Kosako and M Inagaki ldquoRegulation of interme-diate filament organization during cytokinesis possible roles ofRho-associated kinaserdquoMicroscopyResearch andTechnique vol49 no 2 pp 173ndash182 2000

[66] L Chang and R D Goldman ldquoIntermediate filaments mediatecytoskeletal crosstalkrdquo Nature Reviews Molecular Cell Biologyvol 5 no 8 pp 601ndash613 2004

[67] BMUvaM AMasiniM Sturla et al ldquoMicrogravity-inducedapoptosis in cultured glial cellsrdquoEuropean Journal ofHistochem-istry vol 46 no 3 pp 209ndash214 2002

[68] C M Hale A L Shrestha S B Khatau et al ldquoDysfunctionalconnections between the nucleus and the actin andmicrotubulenetworks in laminopathic modelsrdquo Biophysical Journal vol 95no 11 pp 5462ndash5475 2008

[69] D Michaelson W Abidi D Guardavaccaro et al ldquoRac1 accu-mulates in the nucleus during the G2 phase of the cell cycle andpromotes cell divisionrdquo Journal of Cell Biology vol 181 no 3 pp485ndash496 2008

[70] Z Q Dai R Wang S K Ling Y M Wan and Y H Li ldquoSim-ulated microgravity inhibits the proliferation and osteogenesisof rat bone marrow mesenchymal stem cellsrdquo Cell Proliferationvol 40 no 5 pp 671ndash684 2007

[71] T B Damm A Franco-Obregon and M Egli ldquoGravitationalforce modulates G

2M phase exit in mechanically unloaded

myoblastsrdquo Cell Cycle vol 12 no 18 pp 3001ndash3012 2013[72] L Yuge T Kajiume H Tahara et al ldquoMicrogravity potentiates

stem cell proliferation while sustaining the capability of differ-entiationrdquo Stem Cells and Development vol 15 no 6 pp 921ndash929 2006

[73] AGuignandonCGenty LVicoM-H Lafage-Proust S Palleand C Alexandre ldquoDemonstration of feasibility of automated

osteoblastic line culture in space flightrdquo Bone vol 20 no 2 pp109ndash116 1997

[74] F J Alenghat and D E Ingber ldquoMechanotransduction allsignals point to cytoskeleton matrix and integrinsrdquo SciencersquosSTKE Signal Transduction Knowledge Environment vol 2002no 119 article PE6 2002

[75] N Q Balaban U S Schwarz D Riveline et al ldquoForce and focaladhesion assembly a close relationship studied using elasticmicropatterned substratesrdquoNature Cell Biology vol 3 no 5 pp466ndash472 2001

[76] N Nabavi A Khandani A Camirand and R E HarrisonldquoEffects of microgravity on osteoclast bone resorption and oste-oblast cytoskeletal organization and adhesionrdquoBone vol 49 no5 pp 965ndash974 2011

[77] J Gebken B Luders H Notbohm et al ldquoHypergravity stimu-lates collagen synthesis in human osteoblast-like cells evidencefor the involvement of p4442 MAP-kinases (ERK 12)rdquo TheJournal of Biochemistry vol 126 no 4 pp 676ndash682 1999

[78] W A Loesberg X FWalboomers J JW A Van Loon and J AJansen ldquoSimulated microgravity activates MAPK pathways infibroblasts cultured on microgrooved surface topographyrdquo CellMotility and the Cytoskeleton vol 65 no 2 pp 116ndash129 2008

[79] L Liu C Zong B Li et al ldquoThe interaction between1205731 integrinsand ERK12 in osteogenic differentiation of human mesenchy-mal stem cells under fluid shear stress modelled by a perfu-sion systemrdquo Journal of Tissue Engineering and RegenerativeMedicine vol 8 no 2 pp 85ndash96 2014

[80] D FWard Jr W AWilliams N E Schapiro et al ldquoFocal adhe-sion kinase signaling controls cyclic tensile strain enhanced col-lagen I-induced osteogenic differentiation of humanmesenchy-mal stem cellsrdquoMolecular and Cellular Biomechanics vol 4 no4 pp 177ndash188 2007

[81] B M Spiegelman and C A Ginty ldquoFibronectin modulation ofcell shape and lipogenic gene expression in 3T3-adipocytesrdquoCell vol 35 no 3 part 2 pp 657ndash666 1983

[82] J Liu S M DeYoung M Zhang M Zhang A Cheng andA R Saltiel ldquoChanges in integrin expression during adipocytedifferentiationrdquo Cell Metabolism vol 2 no 3 pp 165ndash177 2005

[83] P Barbe J Galitzky I de Glisezinski et al ldquoSimulated micro-gravity increases 120573-adrenergic lipolysis in human adipose tis-suerdquo The Journal of Clinical Endocrinology amp Metabolism vol83 no 2 pp 619ndash625 1998

[84] H Maass J Transmontano and F Baisch ldquoResponse of adren-ergic receptors to 10 days head-down tilt bedrestrdquo Acta Physi-ologica Scandinavica Supplement vol 144 no 604 pp 61ndash681992

[85] V A Convertino J L Polet K A Engelke G W HofflerL D Lane and C G Blomqvist ldquoEvidence for increased 120573-adrenoreceptor responsiveness induced by 14 days of simulatedmicrogravity in humansrdquo American Journal of Physiology vol273 no 1 part 2 pp R93ndashR99 1997

[86] J-M Dong T Leung E Manser and L Lim ldquocAMP-inducedmorphological changes are counteracted by the activated RhoAsmall GTPase and the Rho kinase ROK120572rdquoThe Journal of Biolog-ical Chemistry vol 273 no 35 pp 22554ndash22562 1998

[87] Z Cao RM Umek and S L McKnight ldquoRegulated expressionof three CEBP isoforms during adipose conversion of 3T3-L1cellsrdquo Genes amp Development vol 5 no 9 pp 1538ndash1552 1991

[88] Q-Q Tang M-S Jiang and M D Lane ldquoRepressive effect ofSp1 on the CEBP120572 gene promoter role in adipocyte differen-tiationrdquoMolecular and Cellular Biology vol 19 no 7 pp 4855ndash4865 1999


[89] J E Reusch L A Colton and D J Klemm ldquoCREB activationinduces adipogenesis in 3T3-L1 cellsrdquo Molecular amp CellularBiology vol 20 no 3 pp 1008ndash1020 2000

[90] A Mettouchi S Klein W Guo et al ldquoIntegrin-specific activa-tion of Rac controls progression through theG

1phase of the cell

cyclerdquoMolecular Cell vol 8 no 1 pp 115ndash127 2001[91] QWan E ChoH Yokota and SNa ldquoRac1 andCdc42GTPases

regulate shear stress-driven 120573-catenin signaling in osteoblastsrdquoBiochemical and Biophysical Research Communications vol 433no 4 pp 502ndash507 2013

[92] M Hughes-Fulford and V Gilbertson ldquoOsteoblast fibronectinmRNA protein synthesis andmatrix are unchanged after expo-sure tomicrogravityrdquoFASEB Journal vol 13 no 8 pp S121ndashS1271999

[93] A Guignandon C Faure T Neutelings et al ldquoRac1 GTPasesilencing counteracts microgravity-induced effects on osteo-blastic cellsrdquo The FASEB Journal vol 28 no 9 pp 4077ndash40872014

[94] F H Brembeck M Rosario andW Birchmeier ldquoBalancing celladhesion andWnt signaling the key role of 120573-cateninrdquo CurrentOpinion in Genetics and Development vol 16 no 1 pp 51ndash592006

[95] F M van Roy and P D McCrea ldquoA role for kaiso-p120ctncomplexes in cancerrdquoNature Reviews Cancer vol 5 no 12 pp956ndash964 2005

[96] J Noritake M Fukata K Sato et al ldquoPositive role of IQGAP1an effector of Rac1 in actin-meshwork formation at sites of cell-cell contactrdquo Molecular Biology of the Cell vol 15 no 3 pp1065ndash1076 2004

[97] M Fukata and K Kaibuchi ldquoRho-family GTPases in cadherin-mediated cell-cell adhesionrdquo Nature Reviews Molecular CellBiology vol 2 no 12 pp 887ndash897 2001

[98] CHMCastro C S Shin J P Stains et al ldquoTargeted expressionof a dominant-negative N-cadherin in vivo delays peak bonemass and increases adipogenesisrdquo Journal of Cell Science vol117 no 13 pp 2853ndash2864 2004

[99] S L Ferrari K Traianedes M Thorne et al ldquoRole for N-cadherin in the development of the differentiated osteoblasticphenotyperdquo Journal of Bone andMineral Research vol 15 no 2pp 198ndash208 2000

[100] A J Engler S Sen H L Sweeney and D E Discher ldquoMatrixelasticity directs stem cell lineage specificationrdquo Cell vol 126no 4 pp 677ndash689 2006

[101] T P Stein and C E Wade ldquoMetabolic consequences of muscledisuse atrophyrdquo The Journal of Nutrition vol 135 no 7 pp1824Sndash1828S 2005

[102] B Nusgens G Chometon A Guignandon et al ldquoRole ofthe RhoGTPases in the cellular receptivity and reactivity tomechanical signals including microgravityrdquo Journal of Gravita-tional Physiology vol 12 no 1 pp 269ndash270 2005

[103] Z-G Zhang C A Lambert S Servotte et al ldquoEffects of con-stitutively active GTPases on fibroblast behaviorrdquo Cellular andMolecular Life Sciences vol 63 no 1 pp 82ndash91 2006

[104] R McBeath D M Pirone C M Nelson K Bhadriraju and CS Chen ldquoCell shape cytoskeletal tension and RhoA regulatestem cell lineage commitmentrdquo Developmental Cell vol 6 no4 pp 483ndash495 2004

[105] Y-R V Shih K-F Tseng H-Y Lai C-H Lin and O KLee ldquoMatrix stiffness regulation of integrin-mediated mechan-otransduction during osteogenic differentiation of humanmes-enchymal stem cellsrdquo Journal of Bone andMineral Research vol26 no 4 pp 730ndash738 2011

[106] D L Long J S Willey and R F Loeser ldquoRac1 is requiredfor matrix metalloproteinase 13 production by chondrocytes inresponse to fibronectin fragmentsrdquo Arthritis and Rheumatismvol 65 no 6 pp 1561ndash1568 2013

[107] D C Radisky D D Levy L E Littlepage et al ldquoRac1b andreactive oxygen species mediate MMP-3-induced EMT andgenomic instabilityrdquo Nature vol 436 no 7047 pp 123ndash1272005

[108] J Settleman ldquoTension precedes commitmentmdasheven for a stemcellrdquoMolecular Cell vol 14 no 2 pp 148ndash150 2004

[109] X Yao R Peng and J Ding ldquoEffects of aspect ratios of stem cellson lineage commitments with and without induction mediardquoBiomaterials vol 34 no 4 pp 930ndash939 2013

[110] L Gao R McBeath and C S Chen ldquoStem cell shape regulatesa chondrogenic versus myogenic fate through Rac1 and N-cadherinrdquo Stem Cells vol 28 no 3 pp 564ndash572 2010

[111] R Sordella W Jiang G-C Chen M Curto and J Settle-man ldquoModulation of Rho GTPase signaling regulates a switchbetween adipogenesis and myogenesisrdquo Cell vol 113 no 2 pp147ndash158 2003

[112] S Huang C S Chen and D E Ingber ldquoControl of cyclin D1p271198701198941199011 and cell cycle progression in human capillary endothe-lial cells by cell shape and cytoskeletal tensionrdquo MolecularBiology of the Cell vol 9 no 11 pp 3179ndash3193 1998

[113] SHuang andD E Ingber ldquoThe structural andmechanical com-plexity of cell-growth controlrdquo Nature Cell Biology vol 1 no 5pp E131ndashE138 1999

[114] M F Olson A Ashworth and A Hall ldquoAn essential role forRho Rac and Cdc42 GTPases in cell cycle progression throughG1rdquo Science vol 269 no 5228 pp 1270ndash1272 1995

[115] MYamamotoNMarui T Sakai et al ldquoADP-ribosylation of therhoA gene product by botulinum C3 exoenzyme causes Swiss3T3 cells to accumulate in the G1 phase of the cell cyclerdquo Onco-gene vol 8 no 6 pp 1449ndash1455 1993

[116] D Joyce B Bouzahzah M Fu et al ldquoIntegration of Rac-dependent regulation of cyclin D1 transcription through anuclear factor-120581B-dependent pathwayrdquoThe Journal of BiologicalChemistry vol 274 no 36 pp 25245ndash25249 1999

[117] J K Westwick Q T Lambert G J Clark et al ldquoRac regulationof transformation gene expression and actin organization bymultiple PAK-independent pathwaysrdquo Molecular amp CellularBiology vol 17 no 3 pp 1324ndash1335 1997

[118] M L Coleman and C J Marshall ldquoA family outing smallGTPases cyclinrsquo through G1rdquo Nature Cell Biology vol 3 no 11pp E250ndashE251 2001

[119] A J Ridley ldquoCyclinrsquo round the cell with Racrdquo DevelopmentalCell vol 1 no 2 pp 160ndash161 2001

[120] J Rosenblatt L P Cramer B Baum and KMMcGee ldquoMyosinII-dependent cortical movement is required for centrosomeseparation and positioning during mitotic spindle assemblyrdquoCell vol 117 no 3 pp 361ndash372 2004

[121] M Glotzer ldquoAnimal cell cytokinesisrdquoAnnual Review of Cell andDevelopmental Biology vol 17 pp 351ndash386 2001

[122] S B Ting E Deneault K Hope et al ldquoAsymmetric segregationand self-renewal of hematopoietic stem and progenitor cellswith endocytic Ap2a2rdquo Blood vol 119 no 11 pp 2510ndash25222012

[123] R Nusse and H Varmus ldquoThree decades of Wnts a personalperspective on how a scientific field developedrdquo The EMBOJournal vol 31 no 12 pp 2670ndash2684 2012


[124] L Mellor T Bake M Hiremath E G Loboa and J T OxfordldquoSimulated microgravity affects Wnt signaling in articularcartilage possible implications for crosstalk between cartilageand subchondral bonerdquo inProceedings of the 2014NASAHumanResearch Program InvestigatorsrsquoWorkshop Galveston Tex USAFebruary 2014

[125] M Capulli A Rufo A Teti and N Rucci ldquoGlobal transcrip-tome analysis in mouse calvarial osteoblasts highlights setsof genes regulated by modeled microgravity and identifies Aldquomechanoresponsive osteoblast gene signaturerdquordquo Journal of Cel-lular Biochemistry vol 107 no 2 pp 240ndash252 2009

[126] Q Wan E Cho H Yokota and S Na ldquoRhoA GTPase interactswith beta-catenin signaling in clinorotated osteoblastsrdquo Journalof Bone andMineralMetabolism vol 31 no 5 pp 520ndash532 2013

[127] S J Rodda and A P McMahon ldquoDistinct roles for Hedgehogand caronicalWnt signaling in specification differentiation andmaintenance of osteoblast progenitorsrdquo Development vol 133no 16 pp 3231ndash3244 2006

[128] J-B Kim P Leucht K Lam et al ldquoBone regeneration is regu-lated by Wnt signalingrdquo Journal of Bone and Mineral Researchvol 22 no 12 pp 1913ndash1923 2007

[129] D Baksh and R S Tuan ldquoCanonical and non-canonical Wntsdifferentially affect the development potential of primary isolateof human bone marrow mesenchymal stem cellsrdquo Journal ofCellular Physiology vol 212 no 3 pp 817ndash826 2007

[130] E J Arnsdorf P Tummala and C R Jacobs ldquoNon-canonicalWnt signalling andN-cadherin related120573-catenin signalling playa role in mechanically induced osteogenic cell faterdquo PLoS ONEvol 4 no 4 Article ID e5388 2009

[131] N Case M Ma B Sen Z Xie T S Gross and J Rubin ldquo120573-Catenin levels influence rapid mechanical responses in osteo-blastsrdquoThe Journal of Biological Chemistry vol 283 no 43 pp29196ndash29205 2008

[132] A Santos A D Bakker J M A De Blieck-Hogervorst andJ Klein-Nulend ldquoWNT5A induces osteogenic differentiationof human adipose stem cells via rho-associated kinase RockrdquoCytotherapy vol 12 no 7 pp 924ndash932 2010

[133] J Rossol-Allison L N Stemmle K I Swenson-Fields et alldquoRho GTPase activity modulates Wnt3a120573-catenin signalingrdquoCellular Signalling vol 21 no 11 pp 1559ndash1568 2009

[134] L Peng Y Li K Shusterman M Kuehl and C W GibsonldquoWnt-RhoA signaling is involved in dental enamel develop-mentrdquo European Journal of Oral Sciences vol 119 supplementS1 pp 41ndash49 2011

[135] M-H Kim M Kino-oka N Maruyama A Saito Y SawaandM Taya ldquoCardiomyogenic induction of human mesenchy-mal stem cells by altered Rho family GTPase expression ondendrimer-immobilized surface with d-glucose displayrdquo Bio-materials vol 31 no 30 pp 7666ndash7677 2010

[136] D Diekmann A Abo C Johnston A W Segal and A HallldquoInteraction of Rac with p67phox and regulation of phagocyticNADPHoxidase activityrdquo Science vol 265 no 5171 pp 531ndash5331994

[137] J D Lambeth ldquoNoxDuox family of nicotinamide adenine din-ucleotide (phosphate) oxidasesrdquo Current Opinion in Hematol-ogy vol 9 no 1 pp 11ndash17 2002

[138] R Takeya and H Sumimoto ldquoMolecular mechanism for activa-tion of superoxide-producingNADPHoxidasesrdquoMolecules andCells vol 16 no 3 pp 271ndash277 2003

[139] M Almeida E Ambrogini L Han S C Manolagas andR L Jilka ldquoIncreased lipid oxidation causes oxidative stress

increased peroxisome proliferator-activated receptor-120574 expres-sion and diminished pro-osteogenicWnt signaling in the skele-tonrdquo The Journal of Biological Chemistry vol 284 no 40 pp27438ndash27448 2009

[140] C-L Kao L-K Tai S-H Chiou et al ldquoResveratrol promotesosteogenic differentiation and protects against dexamethasonedamage in murine induced pluripotent stem cellsrdquo Stem Cellsand Development vol 19 no 2 pp 247ndash257 2010

[141] S W Lane S de Vita K A Alexander et al ldquoRac signaling inosteoblastic cells is required for normal bone development butis dispensable for hematopoietic developmentrdquo Blood vol 119no 3 pp 736ndash744 2012

[142] M-K Kwak K Itoh M Yamamoto T R Sutter and T WKensler ldquoRole of transcription factor Nrf2 in the induction ofhepatic phase 2 and antioxidative enzymes in vivo by the cancerchemoprotective agent 3H-1 2-dimethiole-3-thionerdquo Molecu-lar Medicine vol 7 no 2 pp 135ndash145 2001

[143] M K Cho W D Kim S H Ki et al ldquoRole of G12057212and G120572

13

as novel switches for the activity of Nrf2 a key antioxidativetranscription factorrdquo Molecular amp Cellular Biology vol 27 no17 pp 6195ndash6208 2007

[144] M Daugaard R Nitsch B Razaghi et al ldquoHace1 controlsROS generation of vertebrate Rac1-dependent NADPH oxidasecomplexesrdquo Nature Communications vol 4 article 2180 2013

[145] A Y Chi G B Waypa P T Mungai and P T SchumackerldquoProlonged hypoxia increases ros signaling and RhoA activa-tion in pulmonary artery smooth muscle and endothelial cellsrdquoAntioxidants and Redox Signaling vol 12 no 5 pp 603ndash6102010

[146] D Kondrikov R B Caldwell Z Dong and Y Su ldquoReactiveoxygen species-dependent RhoA activation mediates collagensynthesis in hyperoxic lung fibrosisrdquo Free Radical Biology andMedicine vol 50 no 11 pp 1689ndash1698 2011

[147] A S Nimnual L J Taylor and D Bar-Sagi ldquoRedox-dependentdownregulation of Rho by Racrdquo Nature Cell Biology vol 5 no3 pp 236ndash241 2003

[148] S Versari G Longinotti L Barenghi J A M Maier and SBradamante ldquoThe challenging environment on board the Inter-national Space Station affects endothelial cell function by trig-gering oxidative stress through thioredoxin interacting proteinoverexpression the ESA-SPHINX experimentrdquo The FASEBJournal vol 27 no 11 pp 4466ndash4475 2013

[149] T P Stein ldquoSpace flight and oxidative stressrdquo Nutrition vol 18no 10 pp 867ndash871 2002

[150] I Testard M Ricoul F Hoffschir et al ldquoRadiation-inducedchromosome damage in astronautsrsquo lymphocytesrdquo InternationalJournal of Radiation Biology vol 70 no 4 pp 403ndash411 1996

[151] M FukataMNakagawa andKKaibuchi ldquoRoles of Rho-familyGTPases in cell polarisation and directionalmigrationrdquoCurrentOpinion in Cell Biology vol 15 no 5 pp 590ndash597 2003

[152] C D Lawson and K Burridge ldquoThe on-off relationship of RhoandRac during integrin-mediated adhesion and cellmigrationrdquoSmall GTPases vol 5 no 1 Article ID e27958 2014

[153] TMitsuharaM Takeda S Yamaguchi et al ldquoSimulatedmicro-gravity facilitates cell migration and neuroprotection after bonemarrow stromal cell transplantation in spinal cord injuryrdquo StemCell Research andTherapy vol 4 no 2 article 35 2013

[154] P A Plett R Abonour S M Frankovitz and C M OrschellldquoImpact of modeled microgravity onmigration differentiationand cell cycle control of primitive human hematopoietic pro-genitor cellsrdquo Experimental Hematology vol 32 no 8 pp 773ndash781 2004


[155] A Espinosa-Jeffrey P M Paez V T Cheli V Spreuer IWanner and J De Vellis ldquoImpact of simulated microgravity onoligodendrocyte development implications for central nervoussystem repairrdquo PLoS ONE vol 8 no 12 Article ID e76963 2013

[156] K Burridge and K Wennerberg ldquoRho and Rac take centerstagerdquo Cell vol 116 no 2 pp 167ndash179 2004

[157] A Higashibata M Imamizo-Sato M Seki T Yamazaki andN Ishioka ldquoInfluence of simulated microgravity on the activa-tion of the small GTPase Rho involved in cytoskeletal forma-tionmdashmolecular cloning and sequencing of bovine leukemia-associated guanine nucleotide exchange factorrdquo BMC Biochem-istry vol 7 article 19 2006

[158] X Zhang YNanHWang et al ldquoModelmicrogravity enhancesendothelium differentiation of mesenchymal stem cellsrdquoNatur-wissenschaften vol 100 no 2 pp 125ndash133 2013

[159] K Hamamura G Swarnkar N Tanjung et al ldquoRhoA-mediatedsignaling in mechanotransduction of osteoblastsrdquo ConnectiveTissue Research vol 53 no 5 pp 398ndash406 2012

Research ArticleA Tissue Retrieval and Postharvest ProcessingRegimen for Rodent Reproductive Tissues Compatible withLong-Term Storage on the International Space Station andPostflight Biospecimen Sharing Program

Vijayalaxmi Gupta1 Lesya Holets-Bondar1 Katherine F Roby23

George Enders2 and Joseph S Tash1

1Department of Molecular amp Integrative Physiology University of Kansas Medical Center Mail Stop 3050 3901 Rainbow BoulevardHLSIC 3098 Kansas City KS 66160 USA2Department of Anatomy and Cell Biology University of Kansas Medical Center Kansas City KS 66160 USA3Institute for Reproductive Health and Regenerative Medicine University of Kansas Medical Center Kansas City KS 66160 USA

Correspondence should be addressed to Joseph S Tash jtashkumcedu

Received 6 June 2014 Revised 18 September 2014 Accepted 20 October 2014


Copyright copy 2015 Vijayalaxmi Gupta et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Collection and processing of tissues to preserve space flight effects from animals after return to Earth is challenging Specimensmust be harvestedwithminimal time after landing tominimize postflight readaptation alterations in protein expressiontranslationposttranslational modifications and expression as well as changes in gene expression and tissue histological degradation aftereuthanasia We report the development of a widely applicable strategy for determining the window of optimal species-specific andtissue-specific posteuthanasia harvest that can be utilized to integrate into multi-investigator Biospecimen Sharing Programs Wealso determined methods for ISS-compatible long-term tissue storage (10 months at minus80∘C) that yield recovery of high qualitymRNA and protein for western analysis after sample return Our focus was reproductive tissues The time following euthanasiawhere tissues could be collected and histological integrity was maintained varied with tissue and species ranging between 1 and3 hours RNA quality was preserved in key reproductive tissues fixed in RNAlater up to 40min after euthanasia Postfixationprocessing was also standardized for safe shipment back to our laboratory Our strategy can be adapted for other tissues underNASArsquos Biospecimen Sharing Program or similar multi-investigator tissue sharing opportunities

1 Introduction

With the current paucity of opportunities for studying wholeanimal mammalian physiology in space flight the Biospec-imen Sharing Program (BSP) for postflight tissue collectionoffers the opportunity to broaden access to biological samplesshortly after return and maximize the data generated fromflight animal payloads The logistics of space flight experi-ments involving live animals often requires harvesting tissuesat a remote site followed by shipping the specimens to thePrinciple Investigatorsrsquo laboratories for detailed analysis Fur-thermore as the capabilities to house rodent andother animalson the International Space station (ISS) and to conduct long-term space flight experiments using animals are enabled the

need to harvest and fix tissues for long-term storage on ISSthat will retain high quality RNA and protein for subsequentanalysis in laboratories on Earth is also required Theseapproaches to live animal experimentation in space flightthat include tissue harvest for multiple investigators requiredetermination of (1) tissue-specific windows of time betweeneuthanasia and tissue fixation that retain quality of histologyand (2) tissue-specific windows of time for tissue fixation thatretain high quality protein and RNA for subsequent analysisDetermination of these windows provides a quantitativelogical approach to generate appropriately prioritized andoptimized tissue harvest and fixation logistics in a multi-investigator Biospecimen Sharing Program scenario be it onEarth or during tissue harvest on the ISS In addition tissue



storage methods should retain high sample quality underlong-term storage as samplesmay be harvested and stored onthe ISS but may not be returned to Earth for many monthsdepending on ISS to Earth flight frequency and payloadcapacities This issue has been addressed for preserving plantmaterial for gene expression analysis [1] but there is nodata available for animal tissues Our participation in theBSP program involved tissue harvest from female mice atKennedy SpaceCenter Florida USA (KSC) for animals flownfor 12ndash15 days in orbit on three space shuttle flights STS-131STS-133 and STS-135 In addition we harvested tissues frommale mice at the Institute for Biomedical Problems (IMBP)laboratory in Moscow RU for animals flown for 30 daysin orbit on the BION M1 satellite Both the STS and BIONseries of flight experiments involved age- and time-matchedground control groups of animals During the early flightplanning phase of BION there were possibilities that maleor female mice would be flown and that male gerbil tissuesmay also be provided to the BSPThus to be prepared for anyof these possibilities we undertook to determine the optimaltissue harvest windows for all of the species and reproductivetissues that we might be able to obtain Our participationin these multi-investigator specimen sharing efforts coveringfour different primary flight PI experimental designs andharvest logistics necessitated a determination of the windowof time between euthanasia and harvest and preservation ofour tissues of interest that would allow us sufficient flexibilityto obtain the highest possible quality of tissue for histopathol-ogy RNA for gene-transcription analysis and protein forexpression and posttranslational modification analysis Adetermination of these time windows is essential since thetissues that are made available to the BSP investigators areprovided after the primary flight PIrsquos have obtained their tis-sues Knowledge of the optimal windows for all of the tissuesof interest aids in the preparation of targeted tissue harvestflow logistics that can provide each of the BSP teammembersthe highest possible quality tissue respectively Therefore aswe report here we developed a strategy to determine thewindowof time between euthanasia and fixation for retentionof high quality histology for male and female reproductiveorgans We also determined methods for long-term tissuestorage for 10 months that provide for recovery of both highquality protein and RNA These strategies are adaptable andcan be applied to harvest and storage of other time-sensitivelabile tissues from animals and plants Furthermore thesemethods can be used to optimize logistics and data collectionundermulti-investigator tissue harvest and sharing programsoperated by any space agency commercial entity or flightplatform


21 Animals Approximately 8 wk old male and femaleC57Bl6J (Jackson Lab Bar Harbor ME) and sim10-month oldmale and female Mongolian gerbils (Charles River Wilming-ton MA) were used throughout this study All animal useprotocols were approved by the University of Kansas Insti-tutional Animal Care andUse Committee (IACUC) Animals

were maintained in standard cages with 12 12 h dark lightcycle and standard food and water were provided ad libitumAll animals were euthanized using CO

2asphyxiation fol-

lowed by cervical dislocation prior to tissue harvest

22 Determination of the Limit of Time andTemperature between Euthanasia and Tissue Fixation

221 Male Mice Our protocol consisted of three groups ofsix mice each with one mouse for each time point In groupA all mice were euthanized at one time and the testes andepididymides were harvested and separated and then imme-diately placed in Hamrsquos F-10 medium (Sigma Aldrich StLouis MO) on ice At time intervals of 0 05 10 15 20 and25 hr each testis (one mouse per time point) and epididymis(2 mice per time point) were separated and then transferredfrom Hamrsquos F-10 medium to Bouinrsquos fixative (Sigma AldrichSt Louis MO) at room temperature (RT 21∘C) In group Bthe same procedure was followed except that the testes andepididymides separated were placed in Hamrsquos F-10 mediumat RT until transfer to Bouinrsquos at the same time interval asabove In group C all mice were euthanized at one time thecarcasses were maintained at (RT) and then at time 0 05 1015 20 and 25 hr the testis and epididymis were harvestedfrom the carcasses respectively At each of the time pointsabove groupC testes were placed in Bouinrsquos solution and pro-cessed as per standard histology protocols as detailed belowFor all groups the tissues fixed in Bouinrsquos were processed asdetailed in Section 23 below At the same time point groupC epididymides were processed to obtain cauda sperm toassess sperm motility by computer assisted sperm analysis(CASA) [2] For animals used for collection of cauda epi-didymal sperm formotility analysis two animals were used ateach time point

222 Female Mice Female mice were euthanized (carcassesmaintained at RT as per groupC above) and their ovaries anduteri were harvested at 0 05 10 15 20 25 and 30 hr aftereuthanasia (one animal per time point) At the times indi-cated the tissues were fixed at RT in Bouinrsquos overnight andthen processed for histology as detailed below (Section 23)

223 Mongolian Gerbils The six male gerbils were euth-anized at one time and the carcasses were kept at roomtemperature At 0 05 10 15 20 and 30 hr after euthanasiatestis and epididymis were harvested from the carcasses andseparated respectively (one animal per time point) Similarlyall six female gerbils were euthanized and the carcasses werekept at RT Ovaries and uterine horns were harvested fromeach carcass at the same time interval as the males At thetimes indicated above the tissues were placed in Bouinrsquos fixa-tive at RT and then processed for histology as detailed below(Section 23) Sperm were immediately harvested from caudaepididymis as mentioned above and sperm motility analysiswas carried out using CASA as described above [2]

23 Postharvest Processing of Mouse Testicular Tissues Testesand epididymides from the mature mice were harvested as


detailed above Unless indicated otherwise all procedureswere done at RT Tissues were fixed in Bouinrsquos solution for48 h washed in 70 ethanol (ETOH) until the yellow colorof Bouinrsquos disappeared (sim48 hr with frequent changes of 70ETOH and gentle agitation) and divided into four groupsControl group tissue was stored in 70 ETOH until paraffinembedding Rapid transition 70 ETOH was immediatelyreplaced with PBS (Sigma Aldrich St Louis MO) pH 74for one wk and then rapidly replaced with 70 ETOHSlow Re-ETOH only 70 ETOH was immediately replacedwith PBS for one wk and then sequentially replaced at 2 hrintervals with each of 10 30 50 and 70 ETOH Slowrehydration-dehydration 70 ETOH was sequentially (7050 30 10 then PBS at 2 hr each) substituted with PBSfor one wk and then replaced at 2 hr intervals with each of10 30 50 and 70 ETOH To prevent contaminationwe kept tissue in PBS at 4∘C and ETOH replacement wascompleted at RT with 2 hr intervals between changes asdetailed above After each of the respective final dehydrationsteps above the tissues were paraffin-embedded and pro-cessed for histology and hematoxylin and eosin staining (HE)using standard methods as performed previously [3]

24 Total RNA Extraction and Preservation Freshly harves-ted mouse testes were immediately stabilized in 10 volumesof TRIzol reagent (Invitrogen Carlsbad CA) or RNAlatersolution (Ambion Austin TX) The samples placed inRNAlater were stored for (1) two wks at 4∘C (2) one wk at RTand one wk at 4∘C or (3) two wks at RT Before RNAextraction tissues were retrieved from RNAlater solutionwith sterile forceps and then submerged in TRIzol reagentRNAwas isolatedwithTRIzol reagent according to themanu-facturerrsquos instructions Time of placement into RNAlater wasnoted to determine if RNA quality was related to the durationof window from euthanasia to placement in RNAlater RNAintegrity and quantity were determined using Agilent RNAkit andAgilent Bioanalyzer 2100 (Santa Clara CA) One 120583g ofRNA was subjected to RT PCR with the primers specific formouse GAPDH (Forward 51015840CCTTCATTGACCTCAAC-TAC Reverse 51015840ATGACAAGCTTCCCATTCTC) and inte-rleukin-1alpha (IL-1120572) (Forward 51015840ACTTGTTTGAAGAC-CTAAAG Reverse 51015840GTTTCAGAGGTTCTCAGAG) Pri-mers were designed using online Primer Design Tool Primer3 Uteri and ovaries from STS-135 were harvested in RNAlatersolution and stored at minus80∘C for 10 months Total RNA wasisolated from ovaries and uterine horns with Gene EluteMammalian RNA kit (Sigma Aldrich St Louis MO) PCRproducts were verified on 30 agarose gels using standardprocedures

25 Protein Extraction from Ovaries and Uterus Tissue Pre-served in RNAlater Ovaries stabilized in RNAlater for onewk at RT and two wks at 4∘C were used for protein extrac-tion The tissue was removed from RNAlater briefly rinsedwith ice-cold PBS pH 74 and then homogenized in RIPAbuffer containing protease inhibitor cocktail (all from SigmaAldrich St Louis MO) or ProteoJet Mammalian Cell LysisReagent (Fermentas Pittsburgh PA) Uteri from STS-135

mission mice were preserved in RNAlater for 10 months atminus80∘C prior to processing in RIPA as described above Pro-tein concentration was determined using the DC Assay (Bio-Rad Hercules CA) and 15 120583g protein was electrophoresedunder denaturing conditions on 4ndash15 polyacrylamide geland transferred to nitrocellulose membrane (Bio-Rad Her-cules CA) The membranes were blocked for 1 hr with 5nonfat milk in TBS-T (Tris buffered saline with Tween-20SigmaAldrich St LouisMO) and probedwith 1120583gmL rabbitanti-mouse estrogen receptor alpha (ER120572) antibody (SantaCruz Biotechnology CA) To verify equal loading of proteinsmembranes were stripped and reprobed with a goat anti-120573-actin antibody (Santa Cruz Biotechnology CA) Incubationswith primary antibodies were carried out overnight at 4∘CAfter washing in TBS-T and probing with the correspondinghorseradish peroxidase-labeled secondary antibody (PierceBiotechnology Rockford IL) bound antibodies were iden-tified using AmershamACL PlusWestern Blotting DetectionReagents (GEHealthcare Pittsburgh PA) and luminographyWestern blots were quantitated by densitometric analysis

3 Results

31 Effect of Delayed Processing on Quality of Mouse andGerbil Reproductive Tissue Histology

311 Male Mouse Testes harvested any time between time0 after euthanasia up to 25 hr after euthanasia showedhistological properties comparable to the time 0-harvestedmice For comparative purposes the 0min harvested testesrepresent the control for subsequent time points for eachstorage treatment respectively Histological quality of thetestes was excellent in all postharvest treatment proceduresnamely tissues were kept in the carcass until fixation or fixedin Bouinrsquos immediately or stored in Hams F10 on ice or RTbefore fixation Testis tubules appeared normal in all treat-ment groups compared to the 0min controls with noshrinkage of tissue and complete retention of histologic archi-tectural details (Figures 1(a) 1(b) and 1(c))

Total motility and progressive motility of cauda epididymalmouse sperm at times 0 05 10 15 20 and 25 hr were notsignificantly different between any of the tissue harvest timepoints and also not significantly different between the threetissue harvest scenarios Table 1 shows the percent progressivemotility for each time point under all three tissue storageconditions Table 2 shows the percent total motility for eachtime point under all three tissue storage conditions Based oncomparable results between the three tissue harvest regimenswe focused on the ldquotissue in carcassrdquo at RT regimen forsubsequent experimentswith femalemice aswell asmale andfemale gerbils

312 Male Gerbil Testis showed normal distinct histologicaldetails with all spermatogenic cells arranged in a normal pat-tern in the tubule when collected up to 25 hr after euthanasia(Figure 2) Total motility of the gerbil cauda epididymalsperm harvested at each at time point was analyzed and ispresented in Table 3 Since the data represent a single animal


(a) (b)

(c)

Figure 1 Effect of delayed processing on C57Bl6J testicular morphology Each panel represents light microscopy (40x objective) of sectionsof adult mouse testis stained with hematoxylin and eosin (HE) (magnification bar is 100 120583m) (a) Tissues kept on ice for 0 to 25 hr afterharvesting (b) tissues allowed to remain in the carcass for 0 to 25 hr (c) tissues removed and kept at RT for 0 to 25 hr in Hamrsquos F10 afterharvesting


Figure 2Gerbil testicularmorphology (40x objective) at various time points after tissue harvest from the carcass (magnification bar is 50 120583m)All gerbils were euthanized at once and tissues were harvested from the carcass at 05 hr interval from 0 to 25 hr HE staining demonstratedthe retention of histological features at every time point

Table 1 Percent progressivemotility (plusmnSD) ofmouse cauda epididy-mal sperm after exposure of epididymis to various conditions

Time aftereuthanasia In carcass On icelowast

lowastAt roomtemperature

0 h 23 plusmn 18 25 plusmn 26 26 plusmn 1905 h 28 plusmn 13 22 plusmn 13 27 plusmn 1210 h 31 plusmn 3 32 plusmn 19 31 plusmn 915 h 27 plusmn 6 47 plusmn 9 30 plusmn 1120 h 26 plusmn 7 18 plusmn 9 28 plusmn 1425 h 26 plusmn 9 23 plusmn 12 25 plusmn 16lowastTissue was submerged in Hamrsquos F-10 medium in a 15mL tissue culture tubewhich was placed on ice or at room temperature Values are mean plusmn standarddeviation (119899 = 2mice at each time point)There was no significant differencein motility between the three testing conditions at each time point

statistical analysis cannot be done Given the variation inmotilitywith time the data suggest thatmotilitywas relativelystable at all time points except with perhaps a drop at 25 hr

313 Female Mouse The ovaries (Figure 3(a)) and uteri(Figure 3(b)) harvested from female mice up to 3 hr aftereuthanasia showed excellent histological properties devoid ofapparent tissue degradation

314 FemaleGerbil Gerbil ovaries (Figure 3(c)) harvested upto 1 hr after euthanasia showed normal healthy follicles and

Table 2 Percent total motility (plusmnSD) of mouse cauda epididymalsperm after exposure of epididymis to various posteuthanasiaconditions



0 h 48 plusmn 18 49 plusmn 29 45 plusmn 2105 h 51 plusmn 12 47 plusmn 11 50 plusmn 1510 h 53 plusmn 6 60 plusmn 21 40 plusmn 915 h 57 plusmn 12 67 plusmn 16 54 plusmn 1120 h 51 plusmn 14 46 plusmn 14 43 plusmn 1425 h 50 plusmn 17 49 plusmn 18 52 plusmn 12lowastTissue was submerged in Hamrsquos F-10 medium in a 15mL culture tube whichwas placed on ice or at room temperature as indicated Values are mean plusmnstandard deviation (119899 = 2mice at each time point) There was no significantdifference inmotility between the three testing conditions at each time point

Table 3 Percent total motility of gerbil cauda epididymal spermrecovered from the epididymis after storage in the carcass at RT forthe times indicated (119899 = 1 at each time point)

Time aftereuthanasia

Total motility()

Progressivemotility ()

0 h 983 85005 h 824 58010 h 826 61015 h 952 86220 h 879 73225 h 756 475Since we used one animal per time point standard deviation could not bedetermined


(a)

(b)

(c)

(d)

(e)Figure 3 Mouse and gerbil ovarian and uterine horn histology HE staining was used to evaluate quality of oocyte and follicles Rowsrepresent HE staining of (a) mouse ovary up to 3 hr after euthanasia (4x objective magnification bar is 500120583m) (b) mouse uteri up to3 hr after euthanasia (10x objective magnification bar is 500120583m) (c) gerbil ovary up to 15 hr after euthanasia (4x objective magnificationbar is 500 120583m) (d) significantly high number of vacuoles are indicated in the yellow circles in gerbil ovaries from 15 h after euthanasia (40xobjective magnification bar is 100 120583m) (e) gerbil uteri up to 25 hr after euthanasia (10x objective magnification bar is 500 120583m)

healthy oocytes devoid of signs of tissue degradation how-ever at 15 hr after euthanasia high numbers of unhealthy fol-licles and shrunken oocytes were seen suggesting tissue dete-rioration due to the delay in fixation process after euthanasiaSignificantly high numbers of ldquovacuoles-likerdquo structures werealso seen at 15 hr after euthanasia (Figure 3(d)) which isindicative of tissue degradation (indicated by the yellow cir-cles) Control (0min) gerbil ovaries had negligible ldquovacuolesrdquo(indicated by two black arrows) The ovary harvested after

15 hr did not sustain the histological processing as they weretoo fragile and degraded implying that gerbil ovaries weremore sensitive and should be harvested within 1 hr of euth-anasia Uterine histology indicated tissues were intact andcomparable to control at all tested time points (Figure 3(e))

32 Effect of Different Postfixation Procedures on Quality ofTesticular Histology Testicular and epididymis morphology


60x

(a)

60x

(b) (c)

(d)

Figure 4 Morphological analysis of mouse testis after different postfixative manipulations Sections of adult mouse testis were stained withHE (all at 60x magnification bar is 40 120583m) (a) Control (b) slow rehydration-dehydration stepwise replacement of ETOH-PBS-ETOH (c)slow Re-ETOH only stepwise replacement (d) rapid ETOH-PBS-ETOH transition Sertoli cell (Se) spermatogonia (Sp) spermatocytes (Sc)spermatids (Sd) and Leydig cell (L) Black arrowsmdashabnormal open spaces in seminiferous epithelium yellow arrowsmdashabnormal wavy andthinner basement membrane

was evaluated for histological changes after different post-fixation processing (Figures 4 and 5) In control testiculartissue (Figure 4(a)) all types of spermatogenic cells sper-matogonia (Sp) Sertoli cells (Se) spermatocytes (Sc) andspermatids (Sd) were evident Lymphatic spaces betweenseminiferous tubules and adjacent to Leydig cells (L) clustersare clearly defined After the slow rehydration-dehydrationstepwise replacement with ETOH testis tubules appearednormal with histological architecture similar to the controlgroup (Figure 4(b)) Although all types of spermatogenic cellwere identified after slow rehydration-dehydration stepwisereplacement spermatogonia and Sertoli cell nuclei weremoredifficult to distinguish compared to control The quality oftesticular histology observed after slow Re-ETOH only (Fig-ure 4(c)) was similar to that observed in the slow rehydration-dehydration stepwise replacement group However openspaces (black arrows) were observed within portions ofseminiferous tubules in the slow Re-ETOH only group andthe seminiferous tubule basement membrane in this groupoften appeared wavy and thinner (yellow arrows) comparedto the control (Figure 4(a))The histologic quality was poor inthe group treated by single step (rapid transition) change ofsolution (Figure 4(d)) as evidenced by a diffuse appearance ofthe tissue (not due to focus) and limited clarity of nucleardetails in spermatogonia and Sertoli cells In addition theseminiferous tubule basement membrane was occasionallyindistinct and some spermatid artifactual loss of residualbodies is also observed

We found epididymal tissue to be sensitive to dehy-drationrehydration shock (Figure 5) Slow rehydration-dehydration ETOHreplacement had no visible negative effect

on the quality of epididymal morphology (Figure 5(b)) how-ever histological examination revealed differences in epi-didymal morphology after slow Re-ETOH only (Figure 5(c))and rapid transition procedures (Figure 5(d)) compared tocontrol (Figure 5(a)) The differences included alterations inthe thickness of columnar epithelium and basal and principalcells are not very sharp The slow rehydration-dehydrationcaused less destruction of testicular and epididymal tissuethan the rapid single-step changes of solution and has anoverall better morphological detail preservation compared tothe rapid ETOH-PBS solution change

33 Total RNA and Protein Evaluation after TissuePreservation in RNAlater under Long-Term Storage

331 RNA Stability We compared RNA integrity afterpreservation testicular tissue in TRIzol reagent and RNAlater(Figure 6(a)) Total RNA integrity analysis demonstrated highquality RNA after storing tissue in RNAlater for 1-2 wk at RTor 4∘C compared to TRIzol preservation Distinct 28S and 18Sribosomal RNA and absence of degraded RNAwere observedon the gel RT-PCR analysis of RNAwith primers forGAPDHand IL-1120572 indicated high quality of expected PCR products inall analyzed samples (Figure 6(b)) Integrity analysis of totalRNA isolated from STS-131 ground controls (G10 G11 andG12) and flight (F10 F11 and F12) mouse uteri after 30 to40min after euthanasia demonstrate high RNA quality inthese samples (Figure 6(c)) We found no differences in RNAquality between all analyzed uteri and ovaries samples placedinto RNAlater from 15min to 40min after euthanasia


40x

(a)

40x

(b) (c)

(d)

Figure 5 Morphological analysis of mouse epididymis after different postfixative modifications Sections of adult mouse epididymis werestained with HE (all at 40x magnification bar is 50 120583m) (a) Control (b) slow rehydration-dehydration stepwise replacement of ETOH-PBS-ETOH (c) slow Re-ETOH only replacement (d) rapid ETOH-PBS-ETOH transition Epithelium (E) sperm (S) basal cell (B) and principalcell (P)

332 Long-Term Storage for RNA Analysis Using ovary anduteri harvested from STS 135 mice we also determined theeffect of long-term (10 months) preservation in RNAlater onRNA and protein quality (Figure 7) This time frame waschosen to mimic a possible storage scenario that could occuron the ISS Ovaries and uterus from STS-135 ground controlsstabilized in RNAlater for 10 months showed excellent RNAquality (Figure 7(a)) and yield in range 6ndash8 120583g

333 Protein Stability Although not commonly done tissuestabilized in RNAlater can be used for subsequent proteinextraction Protein obtained from samples stored inRNAlateris suitable for western blotting or 2D gel electrophoresis butnot for applications that require native protein (AmbionGuideline for RNAlater) We optimized the protocol for pro-tein extraction from ovaries preserved in RNAlater for onewk at RT or two wks at 4∘C by using RIPA buffer or ProteoJetLysis reagent (LR) Western analysis of 120573-actin integritydemonstrated that one wk ambient storage of testicular tissuein RNAlater did not affect actin integrity as evidenced by theabsence of proteolytic fragments and consistent signal inten-sity in replicate samples (Figure 6(d))

334 Long-Term Storage for Protein Analysis Protein fromuteri stabilized in RNAlater for 10 months after STS-135mission was extracted with RIPA buffer Immunoblot forER120572 and 120573-actin verified excellent expression levels of bothproteins with no evidence of proteolytic degradation of bothproteins (Figure 7(b)) Quantitative densitometry analysis ofwestern blots indicated reduced levels ER120572 in mostly flight

animals These results demonstrate that RNAlater is aneffective sample collection and stabilization reagent for pro-tecting both RNA and protein under long-term conditionscompatible for the ISS

4 Discussion

We report here a logical method to determine the optimaltime window of tissue harvest and fixation after euthanasiafor use in multi-investigator tissue harvest programs that iscompatible with processing of tissue sample obtained fromspace flight animals We also demonstrate here a long-termstorage regimen for animal tissues compatible with recoveryof high quality RNA and protein under conditions similar tothat on the ISS when there may be many months betweensample collection and return to Earth The procedures setforth also include methods for tissue harvest at the site ofreturn and for safe shipment to external laboratories forfurther processing for histopathology and recovery of proteinand RNA Several fixationpreservation studies have beencarried out for plant samples [1 4 5] One European SpaceAgency report included a brief discussion on fixation ofmammalian cells in tissue culture for microscopy [6] Freidinet al [7] have demonstrated significant alterations on geneexpression in lung carcinoma tissues collected about 30minutes after harvest Durrenberger et al [8] reported thatin human brain samples collected from several brain banksantemortem events appeared to negatively affect the RNAquality but postmortem delays caused no significant dete-rioration This observation supported earlier report that


Ladd

er

1 2 3

(s)

70

65

60

55

50

45

40

35

30

25

20

4

120573-Actin

RIPA-R RIPA-4∘C LR-RT LR-4∘C

IL-1120572 GAPDH1 2 3 4 1 2 3 4

Ladd

er

G10

ut

G11

ut

G12

ut

F10

ut

F11

ut

F12

ut

4000

2000

1000

500

200

25

(nt)

(a)

(b) (d)

(c)

Figure 6 Effect of different extraction and storage methods on RNA and protein quality in ovaries and testes extracted andor stabilized inTRIzol or RNAlater (a) Total RNA integrity analysis Total RNA was isolated from mouse testis and stabilized in TRIzol reagent (RNA yieldand quality control (lane 1) RNAlater for 2wks at 4∘C (lane 2) RNAlater for 1 wk at room temperature and 1 wk at 4∘C (lane 3) RNAlater for2 wks at RT (lane 4) (b) Agarose gel electrophoresis of PCR products with primers for IL-1120572 and GAPDH For RT PCR total RNA was usedafter stabilization in TRIzol reagent (lane 1) RNAlater for 2wks at 4∘C (lane 2) RNAlater for 1 wk at RT and 1wk at 4∘C (lane 3) RNAlater for2 wks at RT (lane 4) (c) Total RNA integrity analysis STS-131 ground (G10 G11 and G12) and flight (F10 F11 and F12) uteri fixed in RNAlaterafter 30ndash40min after euthanasia (d) Comparison of buffers to remove RNAlater for subsequent western analysis of 120573-actin integrity inovaries Mouse ovaries were stored at RNAlater for 1 wk at RT or 2wk at 4∘C tissues were homogenized in RIPA lysis buffer (lanes 1 2) orProteoJET Lysis reagent (LR) (lanes 3 4) Total cell lysates were prepared and subjected to SDS-PAGE Western for 120573-actin is presented

postmortem delay had negligible effect on RNA quality [9]Human stomach has been described as the tissue showingthe earliest sign of postmortem [10] Presnell and Cinadescribed stomach and pancreas as the earliest human tissuesto deteriorate following death [11] The significance of quickprocessing of histopathological specimen in a clinical settinghas been identified by Rohr et al [12] However we havenot come across similar studies for animal tissues used inbiomedical research Prior to our study reported here therewas a significant knowledge gap in the literature for methodsto process animal tissues compatible for multi-investigatorBiospecimen Sharing Programs for space flight logistical sce-narios For our flight studies using male and female mice onthree different space shuttle flights and the BIONM1 flight itwas critical to determine the optimum conditions of tissueharvest and processing for tissues of our interest namelytestis epididymis ovary and uteri

Space flight studies usually comprise remotely dispersedmulti-investigator collaborationsThus there is a need to ship

tissue samples from the site of collection at the return-to-Earth laboratory facility for initial tissue harvest to the site offinal processing and detailed data collection With respect tocollection of RNA samples TRIzol reagent gives excellentRNA integrity however it is a phenol-based solution andtissue preserved in TRIzol cannot be shipped internationallydue to airline safety restrictions Lyophilization of fresh tissuespecimen has been shown to preserve RNA and protein qual-ity and levels by Wu et al [13] Though lyophilization makesshipping easier especially across international borders itis not a viable option for preservation of highest qualityhistological analysis To obtain optimum RNA and proteinquality our results demonstrate that mouse testicular tissuecan be submerged in RNAlater and stored successfully foranalysis at a later time point at least 10 months at minus80∘C Itshould be noted that 10 months represents a minimum limitand longer storage intervals would still need to be directlyassessed Given the current estimate for ISS SpaceX Dragonflights at approximately 3-month intervals this would span


(nt)La

dder

G16

OV

G18

OV

G20

OV

G22

OV

G24

OV

G26

OV

G16

utG18

ut

G20

ut

G22

ut

G24

ut

G26

ut

L 1 2 3 4 5 6 7 8 9 10 11 12

4000

20001000

500200

25

(a)

16

14

12

10

08

06

04

02

00

G16

G18

G20

G22

G24

G26

G28

G32

G34

G40

G42

G44

F46

F50

F52

F54

F56

F58

F60

F62

F64

F66

F68

F70

Ground control group Flight group

Rela

tive E

R120572120573

-act

in p

rote

in ex

pres

sion

G1618 20 22 24 26 G28 32 38 40 42 44 F46 F6050 52 54 56 58 62 64 66 68 70

ER120572

120573-Actin

(b)

Figure 7 RNA and protein quality of STS-135 uteri and ovaries stabilized in RNAlater for 10 months at minus80∘C (a) STS-135 ground control(G16ndashG26) ovarian and uterine RNA integrity analysis Total RNA was extracted and examined for RNA quality (b) Western blot analysis ofER120572 and actin in STS-135 mouse uterus Total cell lysates were prepared and subjected to SDS-PAGE (15120583glane) Western blot analysis wasperformed using the corresponding antibodies to check expression levels of the proteins Representative Immunoblot (top) and its graphicalpresentation (bottom) Densitometric intensities of specific protein bands were digitally obtained and normalized to 120573-actin

three opportunities for sample return after an experimentis terminated and ensure maintenance of RNA and proteinsample quality It is known that in order to isolate high qualityRNA and protein from mammalian tissue the tissue mustbe processed directly after harvest We determined for ourtissues of interest that excellent RNA stability was achieved ifthe tissue samples were placed into RNAlater up to 40minafter euthanasia RNAlater is a popular reagent that inac-tivates all cellular enzymes including RNAses thus RNAexpression profiles can be preserved in situations whenimmediate RNA isolation is not feasible The tissue can bestored in RNAlater for a long time without nucleic acid deg-radation RNAlater has been used by investigators for col-lection of human tissue [14] and used in RNA expressionmicroarrays [15] Our results indicated that RNAlater enableslong-term tissue preservation for RNA and protein extractioncompatible with delayed sample return from flight Thisprocess should be evaluated for use on other tissues to max-imize optimal histology and gene transcription data collec-tion in the primary flight experiments as well as the Biospec-imen Sharing Programs (BSP) investigators Freidin et al [7]have confirmed the significance of RNAlater as a medium topreserve gene expression of lung tissues Tissue processingmethods should be standardized for best storage and analysisof harvested tissues Our methods for tissue fixation long-term storage and recovery of protein and RNA are compati-ble for planned in-flight tissue harvest on ISS

With regard to obtaining the best possible tissue fixationfor histology collected under a multi-investigator Biospec-imen Sharing Program obtaining many tissues of interest

to the participating investigators has to be considered in anintegrated way to accommodate the scientific requirementsof the overarching flight project as well as the natural degra-dation process that tissues undergo as soon as euthanasia hasoccurred Depending on the tissue and species differentwindows of time from euthanasia to fixationmay exist withinwhich histological architecture (as well as RNA and proteinintegrity) is stable In the study design reported here we havedetermined the window of time and temperature for optimalpostharvest maintenance of male and female tissue quality(testis epididymis ovary and uterine horn) and sperm via-bility in mice and gerbils These harvest protocols provide alogical method for integrating the tissue flow logistics forpostflight animals for any project involving multiple inves-tigators Protocols compatible with investigators who requiremore rapid tissue retrieval can be identified and prioritized toensure data preservation We determined that mouse testeswere able to retain excellent histological details when pro-cessed up to 3 hr after euthanasia Sperm motility showedgradual decline with time The authors would reiterate asmentioned in the Results section that sperm motility is ahighly sensitive and variable parameter It is normal to seemajor differences inmotility of sperm obtained fromnot onlyone mouse to another but also within samples obtained fromthe same mouse or gerbil Nonetheless even within the spanof variability seen our results indicate that in case of mousethere was no major difference in sperm motility between anyof the time points whereas in case of gerbil we saw sug-gestions of a slight drop at 25 hr after harvest Future studiesfor flight will require analysis of larger 119899rsquos during definition


phase if gerbils will be used Of the tissues studied hereMongolian gerbil ovaries appear most sensitive to delay inprocessing and require more rapid posteuthanasia process-ing than mouse ovaries Determining optimum conditionsfor tissue handling after harvest is very crucial and can help inmaximizing tissue retrieval form animal models therebymaximizing data output Based on the time-sensitivity inves-tigators may be able to plan the sequence in which the tissuesare harvested starting with the most sensitive tissue to leastsensitive

Finally with regard to the requirement to ship tis-sues fixed for histopathology airline and ground transportproviders (especially international carriers) have specificsafety regulations that prohibit shipment of samples contain-ing many widely used fixatives and preservatives of tissuehistologic integrity Science requirements may present chal-lenges in using alternative fixatives In this regard Bouinrsquosfixative (which contains picric acid) is the best fixative for thetestis if the tissue is to be embedded in paraffin [16] Howeverin space flight experiments safety regulations prevent its useon flight platforms and its presence in tissue samples beingshipped Thus we optimized protocols for fixative removaland storage of tissues in PBS for safe shipping as well asreconstitution protocols for storing tissues in 70 ethanol toretain excellent histology

5 Conclusion

Optimal time frames for harvesting testis epididymis ovaryand uteri without compromising the histological qualitysperm motility and RNA quality have been determinedDifferences in tissue-specific optimal fixation time windowswere noted between mice and gerbils We provide here newmethods for (1) fixative removal and transfer of tissues intoaqueousmedia for safe shipping and (2) reconstitution proto-cols into 70 ethanol that retains excellent histologyWe con-clude that stepwise replacement of ETOH-PBS-ETOHcausedless degradation of histological quality of tissue than a single-step change of solution Our results demonstrate that maleand female mouse reproductive tissues stored in RNAlatersolution were stable and gave high quality RNA and proteinafter 10months of storage atminus80∘CThuswe have determinedmethods for postharvest tissue processing to replace Bouinrsquoswith 70 ethanol for safe shipping across USA and alsoreplace 70 ethanol with PBS to enable shipping of tissuesacross international borders These protocols will facilitateintegration of tissue harvest logistics in multi-investigatorBiospecimen Sharing Programs for optimal tissue histologyand retention of high quality RNA and protein recovery fromanimal tissues on long-term space flight experiments on ISSas well as other flight platforms

Abbreviations

TBS-T Tris buffered saline with 01 Tween 20BSP Biospecimen Sharing ProgramETOH Ethyl alcoholSTS Space Transport System

PBS Phosphate buffered salineISS International Space StationCASA Computer assisted sperm analysisWk WeekRT Room temperatureIACUC Institutional Animal Care and Use CommitteeKSC Kennedy Space CenterRNA Ribonucleic acidCO2 Carbon dioxide


The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

This study is supported by NASA Grant NNX09AP04G toJoseph S TashThe authors wish to acknowledge the excellentsupport for the project from Richard Boyle Paula DumarsVera Vizir Gwo-Shing and Kenny Vassigh from AmesResearch Center NASA and from Stacy Engel AshleighRuggles and Ramona Bobber at Kennedy Space CenterFlorida The authors thank Stanton Fernald of the KUMCImaging Core for assistance in preparation of final figures

References

[1] A-L Paul H G Levine W McLamb et al ldquoPlant molecularbiology in the space station era utilization of KSC fixation tubeswith RNAlaterrdquo Acta Astronautica vol 56 no 6 pp 623ndash6282005

[2] J Luo V Gupta B Kern et al ldquoRole of FYN kinase inspermatogenesis defects characteristic of FYN-null sperm inmicerdquo Biology of Reproduction vol 86 no 1 article 22 2012

[3] J S Tash B Attardi S A Hild R Chakrasali S R Jakkaraj andG I Georg ldquoA novel potent indazole carboxylic acid derivativeblocks spermatogenesis and is contraceptive in rats after a singleoral doserdquo Biology of Reproduction vol 78 no 6 pp 1127ndash11382008

[4] M Braun B Buchen and A Sievers ldquoFixation procedure fortransmission electronmicroscopy ofChara rhizoids under mic-rogravity in a Spacelab (IML-2)rdquo Journal of Biotechnology vol47 no 2-3 pp 245ndash251 1996

[5] V D Kern F D Sack N J White K Anderson W Wells andCMartin ldquoSpaceflight hardware allowing unilateral irradiationand chemical fixation in petri dishesrdquo Advances in SpaceResearch vol 24 no 6 pp 775ndash778 1999

[6] F J Medina A Cogoli C Dournon et al ldquoPreservation ofsamples during space experimentsrdquo in Topical Teams in theLife amp Physical Sciences Towards New Research Applications inSpace pp 200ndash208 European Space Agency 2005

[7] M B Freidin N Bhudia E Lim A G Nicholson W OCookson andM FMoffatt ldquoImpact of collection and storage oflung tumor tissue on whole genome expression profilingrdquoJournal ofMolecular Diagnostics vol 14 no 2 pp 140ndash148 2012

[8] P F Durrenberger S Fernando S N Kashefi et al ldquoEffects ofantemortem and postmortem variables on human brainmRNAquality a brainNet Europe studyrdquo Journal of Neuropathology ampExperimental Neurology vol 69 no 1 pp 70ndash81 2010


[9] J F Ervin E L Heinzen K D Cronin et al ldquoPostmortemdelay has minimal effect on brain RNA integrityrdquo Journal ofNeuropathology and Experimental Neurology vol 66 no 12 pp1093ndash1099 2007

[10] A R Thomas A Practical Guide for Making Post-MortemExaminations BiblioBazaar 2009

[11] S E Presnell and S J Cina ldquoPostmortem changesrdquo MedscapeDrugs amp Diseases 2013 httpemedicinemedscapecomarticle1680032-overview

[12] L R Rohr L J Layfield DWallin andDHardy ldquoA comparisonof routine and rapid microwave tissue processing in a surgicalpathology laboratory quality of histologic sections and advan-tages of microwave processingrdquoThe American Journal of Clini-cal Pathology vol 115 no 5 pp 703ndash708 2001

[13] YWuMWuY Zhang et al ldquoLyophilization is suitable for stor-age and shipment of fresh tissue samples without altering RNAand protein levels stored at room temperaturerdquo Amino Acidsvol 43 no 3 pp 1383ndash1388 2012

[14] S R Florell C M Coffin J A Holden et al ldquoPreservation ofRNA for functional genomic studies a multidisciplinary tumorbank protocolrdquo Modern Pathology vol 14 no 2 pp 116ndash1282001

[15] G L Mutter D Zahrieh C Liu et al ldquoComparison of frozenand RNALater solid tissue storage methods for use in RNAexpressionmicroarraysrdquoBMCGenomics vol 5 article 88 2004

[16] L D Russell R A Ettlin H A Sinha and E D Clegg EdsHistological andHistopathological Evaluation of the Testis CacheRiver Press Clearwater Fla USA 1990

Research ArticleLarge Artery Remodeling and Dynamics following SimulatedMicrogravity by Prolonged Head-Down Tilt Bed Rest in Humans

Carlo Palombo1 Carmela Morizzo1 Martino Baluci1 Daniela Lucini2 Stefano Ricci3

Gianni Biolo4 Piero Tortoli3 and Michaela Kozakova1

1Department of Surgical Medical Molecular and Critical Area Pathology University of Pisa 56124 Pisa Italy2Department of Medical Biotechnologies and Translational Medicine University of Milan 20129 Milan Italy3Department of Information Engineering University of Florence 50139 Florence Italy4Department of Medicine Surgery and Health Sciences University of Trieste 34127 Trieste Italy

Correspondence should be addressed to Carlo Palombo carlopalombomedunipiit

Received 16 May 2014 Revised 26 October 2014 Accepted 27 October 2014

Academic Editor Mariano Bizzarri

Copyright copy 2015 Carlo Palombo et alThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

The effects of simulated microgravity on the static and dynamic properties of large arteries are still mostly unknown The presentstudy evaluated using an integrated vascular approach changes in structure and function of the common carotid and femoralarteries (CCA and CFA) after prolonged head-down tilt bed rest (HDTBR) Ten healthy men were enrolled in a 5-week HDTBRstudy endorsed by the Italian Space Agency (ASI) Arterial geometry flow stiffness and shear rate were evaluated by ultrasoundLocal carotid pulse pressure and wave reflection were studied by applanation tonometry After five weeks of HDTBR CFA showeda decrease in lumen diameter without significant changes in wall thickness (IMT) resulting in an inward remodeling Local carotidpulse pressure decreased and carotid-to-brachial pressure amplification increasedThe ratio of systolic-to-diastolic volumetric flowin CFA decreased whereas in CCA it tended to increase Indices of arterial stiffness and shear rate did not change during HDTBReither in CCA or CFA In summary prolonged HDTBR has a different impact on CCA and CFA structure and flow probablydepending on the characteristics of the vascular bed perfused

1 Introduction

Prolonged head-down tilt bed rest (HDTBR) representsan established experimental model allowing investigatingthe physiologic adaptations to microgravity conditions onthe ground [1] Studies evaluating the effect of simulatedmicrogravity on cardiovascular system have demonstratedthat the prolonged HDTBR is followed by a significantdecrease in left ventricular (LV) mass accompanied by areduction in LV performance [2] Our group has previouslydemonstrated that a reduction in echocardiographic indicesof LV systolic and diastolic performance after a 5-weekperiod of HDTBR does not reflect an impairment in intrinsicmyocardial function but simply an adaptive response tocirculatory unloading [3] Data regarding response of thearterial system to bed rest are less clear Prolonged unloading

has been shown to induce an inward remodeling of thefemoral artery with time-dependent decrease in arterial sizereaching 17 after 52 days of bed rest [4] Eight weeks ofphysical inactivity have been also shown to increase carotidand femoral arterywall thickness andwall-to-lumen ratio [5]Pathophysiologic mechanisms underlying these structuralchanges are supposed to include inactivity-related muscleatrophy associated with a reduced metabolic demand ofthe downstream muscle tissue [6] as well as an impact ofaltered hemodynamic stimuli on the arterial wall It hasbeen demonstrated that arteries are capable to respond tochanges in hemodynamic stimuli (flow and shear rate) andmechanical forces (circumferential and pulsatile stress) bymodification of their geometry [7]However previous studiesdid not provide definite evidence on bed rest induced changesin flow shear rate or wall stress and data regarding impact



of deconditioning on arterial stiffness and wave reflection aresporadic [8] In the present study the common carotid andfemoral arteries were investigated at baseline and after a 5-week HDTBR by an integrated vascular approach allowingevaluating impact of deconditioning on different structuraland functional properties of the arterial system

2 Methods

21 Subjects Ten healthy young volunteers all men meanage 23 plusmn 2 years were enrolled in amultidisciplinary HDTBRstudy endorsed by the Italian Space Agency (ASI) and takingplace at the Orthopedic Hospital Valdoltra Ankaran Slove-nia None of the volunteers was a smoker Medical historyphysical examination laboratory examinations resting andstress ECG and echocardiography have excluded any acuteor chronic medical problem The National Committee forMedical Ethics of the Slovene Ministry of Health (LjubljanaSlovenia) approved the study All participants were informedabout the aim of the investigation the procedures andthe methods and signed a written informed consent formaccording to the Declaration of Helsinki

22 Study Protocol All participants underwent a 5-weekperiod of bed rest in a 6∘ head-down tilt position (HDTBR)During the bed rest period participants were kept strictlyin bed for 24 hours a day and none of them took anymedication or underwent any physical or pharmacologicalcountermeasure Dietary intake was 2300 kcalday and waterintakewas 10ndash15 LdayDiuresiswasmonitored daily andBPand heart rate were measured every 4 hours during daytimeBody composition and hematocrit were measured before andat the end of the bed rest Carotid and femoral ultrasoundcarotid applanation tonometry carotid-femoral pulse wavevelocity (PWV) and cardiac ultrasound were performed theday before entering bed rest and within 24 hours after itstermination Vascular and cardiac examinations were per-formed in a quiet room three hours after a light breakfast andafter an acclimatization period of 30min in supine positionAll vascular acquisitions and readings were performed by asingle operator (CM)

23 Measurements

231 BodyCompositionAssessment Bodyweight and fat-freemass were measured by electrical bioimpedance (BioScan916S Maltron International Ltd Essex UK)

232 Carotid and Femoral Ultrasound On the right com-mon carotid and femoral artery (CCA CFA) two sequentialacquisitions were performed using a modified commerciallyavailable equipment (MyLab30 Esaote Firenze Italy with a75ndash12MHz broadband linear transducer LA435) in orderto obtain the following measures (a) intima-media thick-ness (IMT) systolic diastolic and mean arterial luminaldiameters (b) centerline blood flow velocity (by conventionalDuplex ultrasound) (c) shear rate values directlymeasured atthe near and far arterial wall (by multigate Doppler system)

For all ultrasound acquisitions the angle of inclination forDoppler velocity measurements was consistently adjustedto 60∘ whereas the vessel lumen was set parallel to thetransducer

(a) Longitudinal B-mode images of the right CCA andCFA with well-defined intima-media complex of the nearand far wall were obtained and a loop over 5 cardiac cycleswas stored Brachial pressure and heart rate were mea-sured during loop acquisition (Omron 705 Tokyo Japan)Vascular ultrasound scans were analyzed by the computer-driven image analysis system MIP (Medical Image Pro-cessing Institute of Clinical Physiology CNR Pisa Italy)end-diastolic and end-systolic frames of the CCA or CFAwere selected end-diastolic far-wall IMT and minimumand maximum luminal diameters were measured within aregion of interest Arterial remodeling was assessed as aratio of end-diastolic IMT and luminal radius (IMTradius)where radius was calculated as minimum diameter2 End-diastolic wall stress (kPa) was calculated as follows diastolicBP (in kPa) lowast end-diastolic radiusIMT Delta diameter (Δdiameter) was calculated as the difference betweenmaximumand minimum diameter and the stiffness index beta wascalculated as minimum diameter lowast ln(Systolic BPDiastolicBP)Δ diameter The values reported represent the averageof three cardiac cycles Intraindividual variability of IMTand arterial diameter measurement byMIP in our laboratoryis 48 plusmn 28 and 31 plusmn 19 respectively To estimatemean volumetric flow per beat CCA and CFA diameteraveraged over the entire cardiac cycle was measured from theradiofrequency signal processed by a dedicated software tool(QIMT Esaote EuropeMaastricht Netherlands) in a 1 regionof interest placed at the same area as flow-velocity integral wasmeasured

(b) In spectral Doppler recordings peak systolic anddiastolic velocities as well as systolic diastolic and systo-diastolic flow-velocity integrals weremeasured both for CCAand CFA Resistive index was calculated as follows (peak sys-tolic velocity minus peak diastolic velocity)peak systolic velocitySystolic and diastolic volumetric flows per beat were calcu-lated as systolic and diastolic arterial area (Πlowastdiameter24)multiplied by the corresponding flow-velocity integral Meanvolumetric flow over cardiac cycle was calculated as systo-diastolic flow-velocity integral multiplied by area of luminaldiameter averaged over cardiac cycle as obtained from radio-frequency signal (see above) All values are reported as theaverage of 3 cardiac cycles

(c) Shear rate was assessed by a validated multigateDoppler system determining a flow velocity profile from amatrix of 128-point power spectral densities correspondingto 128 different depths along the Doppler beam [9] A customPC board based on a high-speed digital signal processor wasused to process the quadrature demodulated echo signalsderived from theMyLab30 and to display results in real timeA polynomial least-square fit is applied off-line on the 128experimental velocity points and the resulting profile is usedto evaluate the gradient with respect to radius The localpeak shear rate at the near and far blood-wall interfaces wascalculated


233 Carotid Applanation Tonometry Carotid applanationtonometry was performed on the right CCAusing a validatedsystem (PulsePen Diatecne Milan Italy) [10] Carotid pres-sure waveforms were calibrated according to brachial meanand diastolic pressure as previously described [11] In thecarotid pressure waveform the following parameters weremeasured local systolic BP local pulse pressure and aug-mentation index (AIx) Pulse pressure index was calculatedas local pulse pressure divided bymean BP and pulse pressureamplification as the ratio of brachial to carotid pulse pressure[12] The mean of 3 measurements was used for statisticalanalysis

234 Carotid-Femoral Pulse Wave Velocity Carotid-femoralPWV was measured according to current guidelines [13]using the Complior device (Alam Medical VincennesFrance) Briefly arterial waveforms were obtained transcu-taneously over the right CCA and femoral artery and thetime delay (t) was measured between the feet of the twowaveforms The distance (D) covered by the waves wasestablished as the distance between the two recording sitesPWV was then calculated as D (meters)t (seconds) Themeasurement was performed three times and the mean valuewas used for statistical analysis Simultaneous BP measure-ment was performed at the left brachial artery (OmronKyoto Japan) In our laboratory intraindividual variability ofPWVmeasurement is 45 plusmn 28

235 Cardiac Ultrasound Cardiac ultrasound was per-formed as previously described [3] Stroke volume was mea-sured as a product of aortic area and flow-velocity integralin aortic orifice [14] Flow-velocity integral was obtained alsoin ascending aorta from the suprasternal notch Results onchanges in LV mass performance and loading conditionsobserved in the same study group were previously publishedin detail [3]

24 Statistical Analysis Quantitative data are expressed asmean plusmn sd Paired 119905-test was used to compare the measure-ments obtained before and after HDTBR Linear univariateregression analysis was used to test the relationships betweenbed rest-induced changes in arterial diameter or flow and inFFM or stroke volume Statistical significance was set at avalue of119875 less than 005 Statistical analysis was performed byJMP software version 802 (SAS Institute Inc Cary NorthCarolina USA)

3 Results

During the bed rest period body weight BMI fat FFMand Doppler-derived stroke volume and flow-velocity inte-gral in ascending aorta diminished peripheral BP did notchange significantly and heart rate and hematocrit increased(Table 1)

After 5 weeks of HDTBR no significant changes wereobserved in CCA geometry and stiffness (Table 2) CFAdiameter significantly decreased (minimum diameter by 10plusmn 4) CFA intima-media thickness did not change and

Table 1 Main anthropometric and hemodynamic characteristicsand hematocrit in 10 healthy volunteers before and after HDTBR

Before After 119875

Weight (kg) 75 plusmn 10 73 plusmn 9 lt001BMI (kgm2) 233 plusmn 20 228 plusmn 16 lt005Fat-free mass (kg) 64 plusmn 5 61 plusmn 5 lt00001Hematocrit () 444 plusmn 29 479 plusmn 21 0001Systolic BP (mmHg) 115 plusmn 17 113 plusmn 10 051Diastolic BP (mmHg) 62 plusmn 7 65 plusmn 4 033Pulse pressure (mmHg) 53 plusmn 11 48 plusmn 10 019Heart rate (bpm) 60 plusmn 10 71 plusmn 7 lt0005Stroke volume (mL) 76 plusmn 11 63 plusmn 10 lt001FVI ascending aorta (cm) 217 plusmn 21 192 plusmn 28 001BMI body mass index FFM fat-free mass BP blood pressure FVI flow-velocity integral

therefore the ratio end-diastolic CFA IMTradius increasedand circumferential wall stress decreased (Table 2) Thechanges in CFA minimum diameter showed a trend tocorrelate with changes in fat-free mass (119903 = 049 119875 = 015)CFA beta stiffness index remained unchanged after HDTBR(Table 2)

Responses in flow velocities and volumes differedbetween CCA and CFA In CCA peak systolic and diastolicvelocity did not change significantly during the bed restperiod In CFA both peak systolic and diastolic velocitiesincreased but the increase was higher for diastolic velocityand consequently the resistive index decreased Systolicvolumetric flow per beat remained stable both in CCA andin CFA In contrast diastolic volumetric flow showed a trendto decrease in CCA whereas it increased in CFA (Table 2)Consequently the ratio of systolic-to-diastolic flow in CCAtended to increase while in CFA it significantly decreasedThe relationships between volumetric flow per beat in CCAand stroke volume or ascending aorta flow-velocity integral(estimated by Doppler echocardiography) as well as therelationship between volumetric flow per beat in CFA andstroke volume were tested In CCA the mean and diastolicflow per beat at baseline were strongly related to baselinestroke volume (119903 = 075 119875 = 001 and 119903 = 082 119875 lt 0001)and the changes in mean and diastolic flow per beat duringHDTBR were related to changes in stroke volume (119903 = 070119875 lt 005 and 119903 = 054 119875 = 010) as well as to changes inascending aorta flow-velocity integral (119903 = 078 119875 lt 001and 119903 = 048 119875 = 015) None of these relationships wereobserved for CFA

Wall shear rate at near and far arterial wall did not changeduring HDTBR either in CCA or in CFA (Table 2) In CCAthe mean luminal diameter was positively related to wallshear rate at anterior (119903 = 062 119875 = 005) and posterior wall(119903 = 063 119875 = 005) however this correlation was lost afterthe period of bed rest No relationship between shear rate andluminal diameter was observed for CFA

Carotid femoral PWV and AIx did not change after 5weeks ofHDTBRwhile local carotid pulse pressure and pulsepressure index decreased and pressure amplification index


Table 2 Common carotid artery and common femoral artery structure stiffness and flow before and after HDTBR in 10 healthy volunteers

CCA119875

CFA119875

Before After Before AfterIMT (120583m) 503 plusmn 48 520 plusmn 36 027 515 plusmn 79 523 plusmn 57 058Diameter minimum (mm) 51 plusmn 03 50 plusmn 03 012 74 plusmn 09 67 plusmn 10 lt001Diameter maximum (mm) 59 plusmn 03 58 plusmn 03 009 81 plusmn 10 74 plusmn 10 lt001Δ diameter (mm) 080 plusmn 013 078 plusmn 011 071 074 plusmn 022 072 plusmn 023 072End-diastolic IMTradius 018 plusmn 002 019 plusmn 002 014 014 plusmn 003 016 plusmn 002 lt001End-diastolic wall stress (kPa) 424 plusmn 59 415 plusmn 56 070 607 plusmn 109 549 plusmn 79 005Beta index 32 plusmn 07 29 plusmn 07 029 65 plusmn 19 56 plusmn 21 020Peak velocity systolic (cms) 124 plusmn 25 125 plusmn 22 083 89 plusmn 15 116 plusmn 35 009Peak velocity diastolic (cms) 25 plusmn 5 26 plusmn 6 066 6 plusmn 3 10 plusmn 5 lt005Resistive index 079 plusmn 004 079 plusmn 004 088 094 plusmn 002 091 plusmn 002 005Mean flow per beat (mL) 93 plusmn 16 86 plusmn 13 042 94 plusmn 21 93 plusmn 24 090Systolic flow per beat (mL) 49 plusmn 09 50 plusmn 06 074 78 plusmn 21 71 plusmn 23 037Diastolic flow per beat (mL) 34 plusmn 07 29 plusmn 06 008 15 plusmn 05 19 plusmn 07 005Ratio systdiast flow per beat 15 plusmn 05 17 plusmn 04 007 58 plusmn 20 39 plusmn 06 001WSR peak near wall (sminus1) 524 plusmn 80 575 plusmn 120 029 569 plusmn 177 557 plusmn 197 089WSR peak far wall (sminus1) 460 plusmn 107 494 plusmn 95 049 357 plusmn 52 326 plusmn 69 034IMT intima-media thickness WSR wall shear rate

Table 3 Carotid-femoral pulse wave velocity and carotid pressurewaveform analysis before and after HDTBR in 10 healthy volunteers

Before After 119875

C-F PWV (ms) 69 plusmn 10 69 plusmn 07 053Local SBP (mmHg) 106 plusmn 11 101 plusmn 7 023Local PP (mmHg) 44 plusmn 11 36 plusmn 7 lt005PPI 055 plusmn 011 046 plusmn 009 005AIx 66 plusmn 59 54 plusmn 44 050Pressure amplification 124 plusmn 011 131 plusmn 010 lt005C-F PWV carotid-femoral pulse wave velocity SBP systolic blood pressurePP pulse pressure PPI pulse pressure index AIx augmentation index

increased (Table 3) Changes in hematocrit were not relatedto changes in vascular measures

4 Discussion

The present study compares the response of large elasticand muscular artery to prolonged HDTBR and providessome novel information about arterial mechanics and flowdynamics during deconditioning that are summarized inFigure 1 A complex vascular approach integrating establishedinvestigative modalities with new advanced techniques wasexploited to this purpose

41 Bed Rest Deconditioning and Vascular Geometry In ouryoung healthy volunteers an inward remodeling of femoralartery due to luminal diameter reduction and a diminutionof circumferential wall stress was observed after a 35-day bedrest Carotid geometry on the other hand was not signif-icantly influenced by deconditioning a finding confirming

the differences in response of carotid and femoral artery tobed rest Observed reduction in femoral artery diameter isin agreement with results of the Berlin Bed Rest (BBR) study[4] and may reflect structural andor functional changesextensively discussed in a review paper of Thijssen et al[7] In our study the changes in CFA diameter showed atrend to correlate directly with changes in fat-free massSuch a correlation might suggest that the reduction in CFAlumen reflects a reduced metabolic demand in a downstreammuscle tissue as the gravitational unloading involves bothartery and muscle Yet similar to the BBR study femoralartery volumetric flow did not decrease after bed rest Thisapparent discrepancy could be explained by the fact thatconduit arteries adapt primarily to peak blood flow andoxygen demand during exercise [15]The association betweenconduit artery diameter and muscle work has been suggestedalso in a recent study in which a reduction in femoralartery diameter was demonstrated in subjects wearing amechanical device (HEPHAISTOS) allowing an ldquounloadedorthosisrdquo [16] that is a reduction of muscle work withunchanged gravitational acceleration

In contrast with results of the second BBR study [5]reporting an increment in CCA and femoral artery IMTafter a 60-days bed rest period we did not observe asignificant change in carotid or femoral wall thickness ashorter duration of bed rest in our study could explain thediscrepancy

42 Bed Rest Deconditioning and Blood Flow Previous stud-ies evaluating the effect of unloading on the blood flow inthe lower extremity have produced inconclusive evidenceIn the BBR study [4] the mean blood flow did not changein CFA and superficial femoral artery after bed rest in theHEPHAISTOS study [16] blood flow volume in superficial


5-week HDTBR

mass and work

inward remodeling

of flow in CFA

CCA CFA

Systemic hemodynamics

vasodilation

dArr Lower limb muscle

dArr Metabolic demand

dArr CFA diameter

dArr Wall stress

uArr Diastolic component

dArr Resistive index

dArr Stroke volume

dArr Aortic flow

dArr CCA diastolic flow

dArr Vascular resistance

dArr Sympathetic tone

dArr Wave reflection from peripherydArr Pulsatile component of flow

dArr Local pulse pressuredArr Pulse pressure indexuArr Carotid-brachial pressure amplification

Figure 1 Schematic representation of changes observed at common carotid level at femoral artery level and in central hemodynamics after35-days head-down tilt bed rest in 10 young healthy volunteers

femoral artery remained unaffected by a reduction in mus-cle work while flow velocity increased by 17 in studiesusing plethysmography the blood flow at the arteriolarlevel decreased [17] In a HDTBR study a large portionof blood flow reduction as measured by plethysmographywas observed already after the first day of unloading [18]However plethysmographic and Doppler measurements arehardly comparable

Our study is the first to look separately at the systolicand diastolic flow velocities and volumes both at CCA andCFA levels In CCA the diastolic component of volumetricflow after HDTBR showed trend to decrease and the changesin both diastolic and mean volumetric flow were directlyrelated to changes in stroke volume and in flow-velocityintegral in ascending aorta This observation suggests thatcarotid artery flow simply mirrors the changes occurring inaortic flow In contrast in CFA the diastolic component oflocal blood flow significantly increased and consequentlythe resistive index and the ratio of systolic to diastolic flowvolume decreased This behavior may reflect a decrease inlocal vascular resistance at arteriolar level of the leg Basedupon evidence from previous HDTBR studies a reductionof sympathetic firing to lower limb vessels could explain ourfinding Stout et al reported that during simulated micro-gravity a cutaneous microcirculatory vasodilation is moremarked in the lower than in the upper part of the body beingrelated to a baroreflex-mediated withdrawal of a sympathetictone [19] More recently in healthy volunteers maintained for90 days in HDTBR Ferretti et al demonstrated a significantreduction of the efferent muscle sympathetic nerve activity inthe leg [20]

Wall shear rate describing the tangential force exerted bythe flow stream on the arterial wall did not change duringthe study either in elastic or muscular artery A role of shearrate in arterial diameter control was suggested by a directrelationship between near- and far-wall shear rate and CCAluminal diameter at baseline conditions [21] However such arelationship was not observed at femoral artery level

43 Bed Rest Deconditioning Large Artery Stiffness and Cen-tral Pressure The lack of bed rest induced changes in indicesof either local carotid and femoral stiffness or segmentalaortic stiffness (Table 3) further supports the premise thatthe changes in large artery geometry and flow depend uponfunctional instead of structural vascular changes The signif-icant reduction in the local carotid pulse pressure a goodsurrogate of aortic pressure [22] together with the reductionin the pulse pressure index and the increase in the carotid-to-brachial pressure amplification estimated by means ofcarotid waveform analysis reflect a significant decrease inthe pulsatile component of central pressure compared to thesteady one and possibly a reduction of wave reflection froma vasodilated periphery

5 Study Limitations

This study has several limitations First the population stud-ied is small and consists only of men Second all participantswere young and thus the results do not provide informationon the effect of bed rest on arterial structure and function inolder subjects Third vascular examinations were performed


only one day before and one day after HDTBR consequentlywe could not evaluate a sequence of changes over the bedrest period or after termination of bed rest Fourth plasmaviscosity was not measured so that only wall shear rate butnot wall shear stress could be assessed Furthermore theexperimental model used in this study although establishedfor simulating unloading conditions related to microgravitydoes not allow separating the effects of a reduced muscleactivity from those of a reduced gravitational accelerationFinally AIx values were not adjusted for heart rate Dueto the significant increase in heart rate observed after bedrest we could have overestimated AIx and underestimated areduction in wave reflection

6 Conclusion

An integrated vascular approach combining established andexperimental ultrasound arterial contour wave analysis andpulse wave velocity assessment was exploited to investigatethe adaptation of large arteries to microgravity conditionssimulated by HDTBR Prolonged HDTBR showed a differentimpact on CCA and CFA structure and flow probablydepending on the characteristics of the vascular bed perfusedChanges in CCA blood flow seem to reflect bed rest induceddecrease in stroke volume and aortic flow and did notalter CCA geometry Reduction in CFA luminal diameterand inward remodeling may result from reduced metabolicdemand in a downstream unloaded muscle tissue andchanges in CFA flow may reflect decrease in local vascularresistance secondary to withdrawal of a sympathetic toneObserved changes in systemic hemodynamics that includeddecrease in local pulse pressure and pulse pressure index andincrease in carotid-brachial pressure amplification suggest areduction of wave reflection from a vasodilated periphery(Figure 1) Therefore 5-week HDTBR results in a relativereduction of the pulsatile versus the steady component ofblood flow and arterial pressure possibly reflecting changesin systemic hemodynamics and in sympathetic control of thearteriolar tone

In the prospect of improving the management of subjectsundergoing real microgravity conditions data obtained inthis study confirm the indication to active counter-measure-ments aimed to prevent unloading-related sarcopenia as wellas the possible usefulness of common carotid artery as aldquowindowrdquo to monitor central hemodynamic changes



Acknowledgments

The authors are grateful to Dr Bostjan Simunic Institute ofKinesiology Research University of Primorska Koper andto the personnel at the Valdoltra Orthopaedic Hospital inAnkaran (Slovenia) for their valuable medical assistance andtechnical support This study was partly supported by grants

of the Italian SpaceAgency (ASI) projectsDisorders ofMotorand Cardio-Respiratory Control (DMCR) and Osteoporosisand Muscle Atrophy (OSMA) and a Grant (PRIN 2010-2011)of the Italian Ministry of University and Research (MIUR)

References

[1] J-O Fortrat D Sigaudo R L Hughson A Maillet YYamamoto and C Gharib ldquoEffect of prolonged head-downbed rest on complex cardiovascular dynamicsrdquo AutonomicNeuroscience Basic and Clinical vol 86 no 3 pp 192ndash201 2001

[2] M A Perhonen J H Zuckerman and B D Levine ldquoDeteri-oration of left ventricular chamber performance after bed restldquocardiovascular deconditioningrdquo or hypovolemiardquo Circulationvol 103 no 14 pp 1851ndash1857 2001

[3] M Kozakova EMalshi CMorizzo et al ldquoImpact of prolongedcardiac unloading on left ventricular mass and longitudinalmyocardial performance an experimental bed rest study inhumansrdquo Journal ofHypertension vol 29 no 1 pp 137ndash143 2011

[4] M W P Bleeker P C E de Groot G A Rongen et alldquoVascular adaptation to deconditioning and the effect of anexercise countermeasure results of the Berlin Bed Rest studyrdquoJournal of Applied Physiology vol 99 no 4 pp 1293ndash1300 2005

[5] N T L vanDuijnhovenD J GreenD FelsenbergD L BelavyM T E Hopman and D H J Thijssen ldquoImpact of bed rest onconduit artery remodeling effect of exercise countermeasuresrdquoHypertension vol 56 no 2 pp 240ndash246 2010

[6] P C E de Groot M W P Bleeker and M T E HopmanldquoMagnitude and time course of arterial vascular adaptations toinactivity in humansrdquo Exercise and Sport Sciences Reviews vol34 no 2 pp 65ndash71 2006

[7] D H J Thijssen D J Green and M T E Hopman ldquoBloodvessel remodeling and physical inactivity in humansrdquo Journalof Applied Physiology vol 111 no 6 pp 1836ndash1845 2011

[8] E V Nosova P Yen K C Chong et al ldquoShort-term phys-ical inactivity impairs vascular functionrdquo Journal of SurgicalResearch vol 190 pp 672ndash682 2014

[9] P Tortoli T Morganti G Bambi C Palombo and K V Ram-narine ldquoNoninvasive simultaneous assessment of wall shearrate and wall distension in carotid arteriesrdquo Ultrasound inMedicine amp Biology vol 32 no 11 pp 1661ndash1670 2006

[10] P Salvi G Lio C Labat E Ricci B Pannier and A BenetosldquoValidation of a new non-invasive portable tonometer fordetermining arterial pressure wave and pulse wave velocity thePulsePen devicerdquo Journal of Hypertension vol 22 no 12 pp2285ndash2293 2004

[11] L M Van Bortel E J Balkestein J J van der Heijden-Spek et al ldquoNon-invasive assessment of local arterial pulsepressure comparison of applanation tonometry and echo-trackingrdquo Journal of Hypertension vol 19 no 6 pp 1037ndash10442001

[12] A P Avolio LM Van Bortel P Boutouyrie et al ldquoRole of pulsepressure amplification in arterial hypertension expertsrsquo opinionand review of the datardquoHypertension vol 54 pp 375ndash383 2009

[13] P Boutouyrie and S J Vermeersch ldquoDeterminants of pulsewavevelocity in healthy people and in the presence of cardiovascularrisk factors establishing normal and reference valuesrdquo Euro-pean Heart Journal vol 31 no 19 pp 2338ndash2350 2010

[14] J F Lewis L C Kuo J G Nelson M C Limacher and M AQuinones ldquoPulsed Doppler echocardiographic determinationof stroke volume and cardiac output clinical validation of two


new methods using the apical windowrdquo Circulation vol 70 no3 pp 425ndash431 1984

[15] F A Dinenno H Tanaka K D Monahan et al ldquoRegularendurance exercise induces expansive arterial remodelling inthe trained limbs of healthymenrdquoThe Journal of Physiology vol534 no 1 pp 287ndash295 2001

[16] T Weber M Ducos E Mulder et al ldquoThe specific role ofgravitational accelerations for arterial adaptationsrdquo Journal ofApplied Physiology vol 114 no 3 pp 387ndash393 2013

[17] V A Convertino D F Doerr K L Mathes S L Stein andP Buchanan ldquoChanges in volume muscle compartment andcompliance of the lower extremities in man following 30 daysof exposure to simulated microgravityrdquo Aviation Space andEnvironmental Medicine vol 60 no 7 pp 653ndash658 1989

[18] F Louisy P Schroiff and A Guell ldquoChanges in leg vein fillingand emptying characteristics and leg volumes during long-termhead-down bed restrdquo Journal of Applied Physiology vol 82 no6 pp 1726ndash1733 1997

[19] M S Stout D E Watenpaugh G A Breit and A R HargensldquoSimulated microgravity increases cutaneous blood flow in thehead and leg of humansrdquo Aviation Space and EnvironmentalMedicine vol 66 no 9 pp 872ndash875 1995

[20] G Ferretti F Iellamo P Pizzinelli et al ldquoProlonged head downbed rest-induced inactivity impairs tonic autonomic regulationwhile sparing oscillatory cardiovascular rhythms in healthyhumansrdquo Journal of Hypertension vol 27 no 3 pp 551ndash5612009

[21] S K Samijo J M Willigers R Barkhuysen et al ldquoWall shearstress in the human common carotid artery as function of ageand genderrdquoCardiovascular Research vol 39 no 2 pp 515ndash5221998

[22] T Weber S Wassertheurer B Hametner et al ldquoReferencevalues for central blood pressurerdquo Journal of the AmericanCollege of Cardiology vol 63 no 21 article 2299 2014

Research ArticleSpace Flight Effects on Antioxidant Molecules inDry Tardigrades The TARDIKISS Experiment

Angela Maria Rizzo1 Tiziana Altiero2 Paola Antonia Corsetto1 Gigliola Montorfano1

Roberto Guidetti3 and Lorena Rebecchi3

1Department of Pharmacological and Biomolecular Sciences Universita degli Studi di MilanoVia D Trentacoste 2 20134 Milano Italy2Department of Education and Human Sciences University of Modena and Reggio Emilia Via A Allegri 9 42121 Reggio Emilia Italy3Department of Life Sciences University of Modena and Reggio Emilia Via G Campi 213D 41125 Modena Italy

Correspondence should be addressed to Angela Maria Rizzo angelamariarizzounimiitand Lorena Rebecchi lorenarebecchiunimoreit

Received 10 July 2014 Accepted 22 September 2014


Copyright copy 2015 Angela Maria Rizzo et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

The TARDIKISS (Tardigrades in Space) experiment was part of the Biokon in Space (BIOKIS) payload a set of multidisciplinaryexperiments performed during the DAMA (Dark Matter) mission organized by Italian Space Agency and Italian Air Force in2011 This mission supported the execution of experiments in short duration (16 days) taking the advantage of the microgravityenvironment on board of the Space Shuttle Endeavour (its last mission STS-134) docked to the International Space StationTARDIKISS was composed of three sample sets one flight sample and two ground control samples These samples provided thebiological material used to test as space stressors includingmicrogravity affected animal survivability life cycle DNA integrity andpathways of molecules working as antioxidants In this paper we compared the molecular pathways of some antioxidant moleculesthiobarbituric acid reactive substances and fatty acid composition between flight and control samples in two tardigrade speciesnamely Paramacrobiotus richtersi and Ramazzottius oberhaeuseri In both species the activities of ROS scavenging enzymes thetotal content of glutathione and the fatty acids composition between flight and control samples showed few significant differencesTARDIKISS experiment together with a previous space experiment (TARSE) further confirms that both desiccated and hydratedtardigrades represent useful animal tool for space research

1 Introduction

As the interest in space exploration grows it becomes ofgreat importance to predict and know the response of uni-and multicellular organisms to unfavourable space condi-tions including microgravity This allows us to elaboratethe opportune countermeasures to avoid the risks imposedby space environmental stressors To date many studiesfor understanding physiological biochemical and molecu-lar mechanisms against space stressors are performed onunicellular organisms or cultivated cells of multicellularorganisms [1] Although the experiments on cell culturesare useful it is equally clear that cell cultures representonly the first level of life organization and they cannot be

compared to the response of an entire multicellular livingorganism The use of animals in space research allows us toconduct experiments with organisms characterized by a highlevel of hierarchical biological complexity and physiologicalprocesses comparable to those of humans [2]

Even though animals could be useful models in spaceresearch their use is often limited by the fact that many ofthem need specific rearing bioreactors of large volume [1 3]Tardigrades or water bears are little known and neglectedanimals that allow overcoming this problem Their use inspace research is supported by several reasons (i) they areminiaturized animals (from 200 to 1000120583m in length) thatcan be kept and reared in small facilitiesbioreactors (ii)while having tissues and organs they are simpler than several



other animals having a limited cell number (about 1000)(iii) they can be easily reared under lab conditions (iv)many of them are parthenogenetic often apomictic so clonallineages can be obtained [1 2] Although all tardigrades areaquatic animals they thrive in terrestrial habitats subjectedto periodic desiccation thanks to their ability to enter ahighly stable state of suspended metabolic activity calledanhydrobiosis [4] Entering in this physiological state tardi-grades lose up to 97 of their body water and shrivel into adesiccated structure about one-third of its original sizeWhenrehydrated tardigrades can return to their active metabolicstate in a few minutes to a few hours [4 5] Desiccatedtardigrades can persist in anhydrobiosis for several years anda remarkable resilience to physical and chemical extremeshas been documented [4ndash6] By possessing the abilities towithstand complete desiccation severe cold microgravityvacuum and high levels of ionizing and UV radiationsanhydrobiotic tardigrades fulfill the most important criteriafor tolerating exposure to natural space conditions includingopen space [2]

Tardigrades have already been exposed to space stressorson Low Earth Orbit during the FOTON-M3 mission in2007 with different projects (TARDIS [7] TARSE [1 8]RoTaRad [9]) With the TARSE (Tardigrade Resistance toSpace Effect) project we analyzed the responses of bothdesiccated and hydrated physiological state of the tardigradeParamacrobiotus richtersi to spaceflight conditions withinthe spacecraft [1 8] Microgravity and radiation had noeffect on animal survival and life history traits even thougha higher number of laid eggs a shorter egg developmenttime and a higher number of flight-born juveniles wererecordedwith respect to tardigrades reared onEarth [1 10] Inaddition spaceflight induced in active tardigrades an increaseof glutathione content an increase of glutathione peroxidaseactivity and a decrease of catalase superoxide dismutase andglutathione reductase activities [1] Lastly no change in thio-barbituric acid reactive substances was detected On the basisof these results we developed the new project TARDIKISS(Tardigrades in Space) with the aim to deepen the study ofsurvivorship life history traits and regulation of antioxidantdefences on alive desiccated tardigrades under space stressorsincluding microgravity exposure The flight tardigrades ofthe project TARDIKISS have had a very high survival (morethan 91) and females laid eggs which were able to hatchproducing normal newborns able to reproduce in adulthood[11] In this paper we compared the molecular pathwaysof molecules with antioxidant activity thiobarbituric acidreactive substances and fatty acid composition between flighttardigrades and ground control ones with the final aim toprovide news about the biochemical mechanisms underlyingresistance to space stress conditions

2 Material and Methods

21 TARDIKISS Project The TARDIKISS project was partof the BIOKIS (Biokon in Space) payload a set of multidis-ciplinary experiments in the field of biology and dosimetryperformed in microgravity condition during the DAMA(Dark Matter) mission organized by Italian Space Agency

(ASI) and Italian Air Force in 2011 This mission sup-ported the execution of experiments in short duration (16days) taking the advantage of the microgravity environmenton board of the last mission (STS-134) of Space ShuttleEndeavour docked to the International Space Station (ISS)[11]

TARDIKISSwas composed of three sample sets one flightsample (F) and two ground control samples The formercontrol (temperature control TC) was a postflight controlin which samples were exposed to the temperature profileexperienced by tardigrades the days immediately beforeduring and just after the flight mission the latter (laboratorycontrol LC) was maintained in Modena laboratory for theduration of the flight at constant temperature These samplesprovided the biological material used to test as space stres-sors including microgravity affected animal survivabilitylife cycle DNA integrity and changes of the pathways ofmolecules working as antioxidants

Two anhydrobiotic eutardigrade species were consid-ered namely Paramacrobiotus richtersi (Murray 1911) (Mac-robiotidae) and Ramazzottius oberhaeuseri (Doyere 1840)(Ramazzottiidae) Paramacrobiotus richtersi is the modelspecies already used in the FOTON mission [1] P richtersiwas extracted fromahazel leaf litter (sample codeC3499) it iscarnivorous white in colour and the population here consid-ered is bisexual and amphimictic R oberhaeuseri (Figure 1)was extracted from the lichen Xanthoria parietina (L) ThFr (1860) (sample code C3282) it is herbivorous brownredin colour and the population considered in this study isunisexual and parthenogenetic To extract tardigrades fromtheir substrates leaf litter and lichen were sprinkled with tapwater and after 15min submerged in water for 15min at roomtemperature Later each substrate was sieved (mesh size ofsieves 250 120583mand 37 120583m)under runningwater then animalswere picked up from the sieved sediments with a glass pipetteunder a stereomicroscope

For both tardigrade species animals in desiccated (anhy-drobiotic) physiological state were used To obtain desiccatedspecimens tardigrades were dehydrated in lab under con-trolled air relative humidity (RH) and temperature Afterextraction from their substrates tardigrades were kept inwater for 24 h at 15∘C without any food source Then theywere forced into anhydrobiosis by placing groups of animalson a square (1 cm2) blotting paper with natural mineral water(30 120583L) The paper with tardigrades was initially exposedto 80 RH and 18∘C for 4 h then to 50 RH at 18∘C for4 h in a climatic chamber and finally to 0ndash3 RH at roomtemperature for 12 h [1]

Papers with desiccated tardigrades were stored in twelvesmall plastic Petri dishes (18 cm times 10 cm) enveloped withparafilm and integrated within the Biokon facility (KayserItalia) where a radiation dosimeter for neutrons and i-buttondata logger recorded temperature were also present [11]During the entire flight mission the temperature profile wasrelatively constant ranging from 21∘C to 25∘C [11] while thedose equivalent rates due to space radiation exposure were320 120583Sv (measured by TLD 100 and TLD 700) and 360 120583Sv(measured by TLD 600) [11]


(a) (b)

Figure 1 Micrographs by scanning electronmicroscopy of the tardigrade Ramazzottius oberhaeuseri showing its two physiological states (a)Hydrated and metabolically active specimen (b) Desiccated and metabolically inactive specimen Bars a = 10 120583m b = 5 120583m

22 Biochemical Assays Biochemical assays were performedon desiccated tardigrades comparing F samples with TCsamples

The activities of the enzymes superoxide dismutase(SOD total activity) catalase (CAT) glutathione peroxidase(GPx) and glutathione reductase (GR) were evaluated Thetotal glutathione (GSH) content thiobarbituric acid reactivesubstances (TBARS) and fatty acid composition were alsodetermined as previously described [12]

Substrates and reagents for enzyme determinations wereNAD(P)H DTNB GSH GSSG glutathione reductase andtert-butyl hydroperoxide all of them were purchased fromSigma-Aldrich (St Louis Missouri USA) For each sampleset and each species 6 or 8 (with the exception of SOD)replicates each made up by 10 in toto tardigrades werehomogenized in water on ice with potter using 3 cycles of30 sec each The homogenate was assayed for protein content(according to [13]) and used for enzyme determination Foreach enzyme homogenates were analyzed in duplicate

Briefly the activity of the enzyme superoxide dismutasewas assayed using the method based on NAD(P)H oxidationinhibition (according to [14]) the inhibition of NADPHoxidation by superoxide which was chemically generatedwas measured at 340 nm for 20min in the presence of tissueextracts The incubation mixture included 213 120583L of TDB(triethanolaminediethanolamine 100mM pH 74) 10 120583L ofNADPH 75 120583M 7 120583L of EDTA-MnCl

2(100mMndash50mM)

and 20 120583L of sample or blank One unit of SOD activity wasdefined as the amount of enzyme required to inhibit the rateof NADPH oxidation by 50

To evaluate the activity of catalase samples were assayedby measuring the consumption of H

2O2(according to [15])

Consumption of hydrogen peroxide by the tissue extracts wasdetermined at 240 nm for 1min at 30∘CThe incubation mix-ture included 10 120583L of H

2O2200mM 20 120583L of homogenate

and 170 120583L of Na-phosphate buffer (50mM pH 70) One unitofCATactivitywas defined as the amount of enzyme requiredto catalyze the decomposition of 1mmol of H

2O2minminus1

The activity of the glutathione reductase was assayedfollowing the oxidation of NADPH (according to [16])Briefly GSSG reduction and NADPH consumption werefollowed at 340 nmThe incubation mixture included 5 120583L ofGSSG 125mM 3120583L of NADPH 11mM animal homogenatefrom 20 to 50 120583L and K-phosphate buffer (100mM pH 70)to reach a final volume of 025mL One unit of GR activity

was defined as the amount of enzyme required to catalyze theoxidation of 1 120583mol NADPHminminus1

To evaluate the activity of selenium-dependent glu-tathione peroxidase the enzyme activitywas assayed (accord-ing to [17]) following the decrease in the absorbance at340 nm for 3min which corresponds to the rate of GSHoxidation to GSSG in the presence of NADPH and glu-tathione reductase The incubation mixture included 5 120583L ofGSH 100mM 3 120583L of NADPH 22mM GR 1 unit 5 120583L oftert-butyl hydroperoxide 20mM from 20 to 50 120583L of animalhomogenate and EDTA-K phosphate buffer (3mMndash100mMpH 70) to reach a final volume of 025mL One unit of GPxactivity was defined as the amount of enzyme required tocatalyze the oxidation of 1 120583mol of NADPHminminus1

To measure the total glutathione tardigrades werehomogenized on ice in 5 metaphosphoric acid thehomogenatewas centrifuged at 5000timesg for 10min at 4∘C andthe supernatant was assayed (according to [18]) with someslight modifications Briefly the sulfhydryl group of GSHalso generated from GSSG by adding GR reacts with DTNB(551015840-dithiobis-2-nitrobenzoic acid) and produces a yellow-coloured 5-thio-2-nitrobenzoic acid (TNB) The rate of TNBproduction is directly proportional to this reaction whichin turn is directly proportional to the concentration of GSHin the sample The measurement of the absorbance of TNBat 412 nm provides an accurate estimation of the GSH levelpresent in the sample

To evaluate the thiobarbituric acid reactive substances(TBARS) tardigrade samples standards (from 25 to100 pmol TEP 11-33 tetraethoxypropane) and blanks wereassayed (according to [19]) both before and after inductionof lipid peroxidation by FeSO

4and ascorbic acid TBARS

were determined using a fluorescence spectrophotometer(Carly Eclipse Varian CA USA) at an excitation wavelengthof 517 nm and an emission wavelength of 550 nm For eachsample set (F and TC) and species (R oberhaeuseri and Prichtersi) 2 or 4 replicates were analyzed

To evaluate the fatty acid composition lipids wereextracted from groups of 10 desiccated tardigrades withchloroformmethanol (according to [20]) The total extractwas used for derivatization with sodium methoxide inmethanol 333wv to obtain the fatty acid methylesters(FAME) FAME were injected into a gas chromatograph(Agilent Technologies 6850 Series II) equipped with a flameionization detector (FID) under the following experimental


Table 1 Percentage of fatty acid composition in the tardigrades Paramacrobiotus richtersi and Ramazzottius oberhaeuseri

Fatty acid Paramacrobiotus richtersi Ramazzottius oberhaeuseriTC F TC F

C160 2886 (156) 2941 (353) 2965 (184) 3264 (115)C161 844 (198) 891 (079) 656 (168) 977 (023)C180 1453 (268) 1786 (487) 1622 (176) 1856 (455)C181 1985 (343) 1713 (455) 2145 (171) 2004 (621)C182 n-6 975 (324) 1318 (299) 1211 (659) 1297 (169)C183 n-3 261 (205) 225 (169) 403 (348) 150 (030)C203 n-6 111 (089) 103 (067) 022 (015) 024 (032)C204 n-6 978 (756) 505 (520) 587 (174) 291 (202)C205 n-3 117 (046) 199 (078) 130 (155) 057 (077)C225 n-3 051 (051) 014 (019) 014 (023) 023 (033)C226 n-3 400 (127) 303 (036) 245 (036) 056 (060)lowast

PUFA 2892 (749) 2668 (445) 2612 (325) 1899 (073)lowast

TBARS basal (pmoles120583g proteins) 281 (104) 251 (055) 277 (056) 260 (108)TBARS induced (pmoles120583g proteins) 2606 (365) 2825 (127) 4365 (161) 3291 (258)TC = ground temperature control samples F = flight samples PUFA = polyunsaturated fatty acids TBARS = thiobarbituric reactive substances lowast119875 lt 005 inbrackets SD

conditions capillary column AT Silar length 30m filmthickness 025120583m gas carrier helium temperatures injector250∘C detector 275∘C oven 50∘C for 2min and rate of10∘Cminminus1 until 200∘C for 20min A standard mixture con-taining methyl ester fatty acids was injected for calibrationFor each sample set and species 2 or 4 replicates wereanalyzed

23 Statistical Analysis Data were analyzed with Mann-Whitney test and expressed as mean plusmn SD using the pro-gramme SPSS

3 Results

The results of the enzyme activities in the tardigradesParamacrobiotus richtersi and Ramazzottius oberhaeuseri arealways indicated in relation to 120583g of proteins It is worthnoting that R oberhaeuseri contains a lower amount ofproteins compared to P richtersi (Figure 2)

In both species the comparative analysis of the enzymeactivities and other antioxidant molecules between flight (F)and temperature control samples (TC) showed few significantdifferences (Figures 3 and 4) In particular a significantdecrease (119875 lt 005) of the glutathione reductase activitywas detected in R oberhaeuseri F samples with respect toTC samples (Figure 4(b)) Although not statistically sup-ported in this species a tendency to decrease catalasesuperoxide dismutase and glutathione peroxidase activityand in glutathione content was detected In P richtersi atendency to decrease catalase superoxide dismutase and glu-tathione reductase activities and to increase the glutathioneperoxidase activity was detected Noteworthy differenceswere recorded in the activities of ROS scavenging enzymesbetween the two species

The total percentage fatty acid composition of F and TCsamples is reported in Table 1 In R oberhaeuseri a significant

000

020

040

060

080

100

120

P richtersi R oberhaeuseri

TCFlight

(120583g

prot

eins

aa)

Figure 2 Total protein content in flight and ground temperaturecontrol (TC) samples in the tardigrades Paramacrobiotus richtersiand Ramazzottius oberhaeuseri The bars show the mean with SD

decrease (119875 lt 005) was recorded for the fatty acidC22-6 n-3 and polyunsaturated fatty acids (PUFA) in theF samples with respect to the TC samples Moreover Roberhaeuseri has significantly lower amount of C22-6 n-3compared to P richtersi The amount of thiobarbituric acidreactive substances (TBARS) in tardigrades both before andafter induction of peroxidation in vitro is also reported inTable 1 No differences were detected between F and TCsamples in both species for basal levels and after inductionof peroxidation


0

005

01

015

02

025

03

035

04


Superoxide dismutase (SOD)

TCFlight

(Un

g pr

otei

ns)

(a)

0

10

20

30

40

50

60

70

80


Catalase

TCFlight

(mU

120583 g

prot

eins

)

(b)

Figure 3 Superoxide dismutase (a) and catalase (b) activities in flight and ground temperature control (TC) samples in the tardigradesParamacrobiotus richtersi and Ramazzottius oberhaeuseri The bars show the mean with SD

0

05

1

15

2

25

3


Glutathione peroxidase

TCFlight

(mU

120583g

prot

eins

)

(a)

0010203040506070809

1


Glutathione reductase

lowast

TCFlight

(mU

120583g

prot

eins

)

(b)

0

02

04

06

08

1


Glutathione (GSH)

TCFlight

(mm

oles

120583g

prot

eins

)

(c)

Figure 4 Glutathione peroxidase (a) glutathione reductase (b) and total glutathione content (c) in flight and ground temperature control(TC) samples in the tardigrades Paramacrobiotus richtersi and Ramazzottius oberhaeuseri The bars show the mean with SD lowast119875 lt 005


4 Discussion

Exposure to space stress conditions induces oxidative stressOxidative stress resulting from an imbalance between theexcessive production of reactive oxygen species (ROS) andlimited action of antioxidant defences is implicated inthe development of many important human pathologiesincluding atherosclerosis hypertension inflammation can-cer Parkinson and Alzheimer diseases [21] Oxidative stressmay be highly destructive also in anhydrobiotic organismseven if the lower cellular water content decreases the produc-tion of ROS [21 22] Under normal conditions antioxidantsystems minimize the adverse effects caused by ROS butdesiccation stress could cause the loss or reduction of thesedefence control mechanisms since the metabolic activity isabsent or reduced [21ndash24]

The ability of some animals tardigrades among themto survive extreme desiccation involves a complex arrayof factors working at structural physiological and molec-ular level From a molecularbiochemical point of viewanhydrobiotic organisms synthesize molecules working asbioprotectants during entering permanence and leaving ina desiccated state [25] For example trehalose and sucrosestabilise the biological membrane avoiding protein unfoldingand membrane disturbances late embryogenesis abundantproteins and heat shock proteins work as chaperone systemsrepairing or eliminating damaged molecules while antiox-idant molecules counteract the negative effects of oxidativestress [25]

Since it is known that both hydrated and desiccatedtardigrades have a good natural capability to overcomeoxidative stress [26] they have been used in TARDIKISSexperiments to evaluate the role of antioxidant defence inovercoming oxidative stress induced by exposure to spacestress conditions such as ionizing and UV radiations

The first space experiment (TARSE) conducted withhydrated starved specimens of the tardigrade P richtersidemonstrated that some of the enzymes involved in antiox-idant defences were significantly influenced by the flightstresses [1] In particular there was a significant decrease incatalase and SOD activities the more active enzymes in Prichtersi In addition the glutathione system the less activesystem in not stressed specimens of this species [26] wassignificantly induced during space flight [1] These resultscould be related to the stresses experienced by the hydratedand metabolically active animals (microgravity starvationand radiations) during the flight On the contrary the analysisof antioxidant defences in desiccated tardigrades of theTARDIKISS experiment showed fewer differences relatedto space flight even if the tendency was similar to thatrecorded in hydrated metabolically active animals of theTARSE experiment A similar trend between TARSE andTARDIKISS experiments was also detected in regard to tardi-grade survival since flight animals did not show significantdifferences in survival from temperature laboratory controlones [1 11] Only inR oberhaeuseri (TARDIKISS experiment)a significant decrease in survival rate was recorded between Fand TC samples the species in which a significant decrease ofthe C226 n-3 fatty acid and of glutathione reductase activity

and even though not significant of the activity of the otherROS scavenging enzymes were detected

In conclusion TARDIKISS experiment together withprevious space experiments using tardigrades [1 7ndash9] furtherconfirms that both desiccated and hydrated physiologicalstates of tardigrades represent useful animal tool for spaceresearch To further develop the space research using tardi-grades the setup of experimentswith the possibility to changethe exposition condition of metabolically hydrated animalsas well as the possibility to expose desiccated tardigrades toopen space is necessary Experiments under true space con-dition provide a realistic evaluation of the mechanisms thatcould allowmulticellular organisms including tardigrades tosurvive the combined and synergic effects of space stressorsNevertheless experiments on ground using simulators ofmicrogravity radiation temperature and other space stressesare an essential part of space research complementing exper-iments under true space conditions The comparisons of twodifferent sets of data (ground and space data) will allow betterunderstanding of the physiological and molecular pathwaysof living organisms under space environment



Acknowledgments

The authors are very grateful to the Italian Space Agency(ASI) and the Italian Air Force (AM) which funded theDAMA mission The authors are also very grateful to KayserItalia (KI) which developed and manufactured the hardwareinvolved in the BIOKIS payload They are grateful to anony-mous reviewers for their constructive suggestions

References

[1] L Rebecchi T Altiero R Guidetti et al ldquoTardigrade resistanceto space effects First results of experiments on the LIFE-TARSEMission onFOTON-M3 (September 2007)rdquoAstrobiology vol 9no 6 pp 581ndash591 2009

[2] R Guidetti A M Rizzo T Altiero and L Rebecchi ldquoWhatcan we learn from the toughest animals of the Earth Waterbears (tardigrades) asmulticellularmodel organisms in order toperform scientific preparations for lunar explorationrdquoPlanetaryand Space Science vol 74 no 1 pp 97ndash102 2012

[3] H Marthy ldquoDevelopmental biology of animal models undervaried gravity conditions a reviewrdquo Vie et Milieu vol 52 no4 pp 149ndash189 2002

[4] N Moslashbjerg K A Halberg A Joslashrgensen et al ldquoSurvival inextreme environmentsmdashon the current knowledge of adapta-tions in tardigradesrdquo Acta Physiologica vol 202 no 3 pp 409ndash420 2011

[5] R Guidetti T Altiero and L Rebecchi ldquoOn dormancy strate-gies in tardigradesrdquo Journal of Insect Physiology vol 57 no 5pp 567ndash576 2011

[6] T Altiero R Guidetti V Caselli M Cesari and L RebecchildquoUltraviolet radiation tolerance in hydrated and desiccated


eutardigradesrdquo Journal of Zoological Systematics and Evolution-ary Research vol 49 supplement 1 pp 104ndash110 2011

[7] K I Jonsson E Rabbow R O Schill M Harms-Ringdahl andP Rettberg ldquoTardigrades survive exposure to space in low Earthorbitrdquo Current Biology vol 18 no 17 pp R729ndashR731 2008

[8] L Rebecchi T Altiero M Cesari et al ldquoResistance of the anhy-drobiotic eutardigrade Paramacrobiotus richtersi to space flight(LIFE-TARSE mission on FOTON-M3)rdquo Journal of ZoologicalSystematics and Evolutionary Research vol 49 supplement 1 pp98ndash103 2011

[9] D Persson K A Halberg A Joslashrgensen C Ricci N Moslashbjergand R M Kristensen ldquoExtreme stress tolerance in tardigradessurviving space conditions in low earth orbitrdquo Journal of Zoolog-ical Systematics and Evolutionary Research vol 49 supplement1 pp 90ndash97 2011

[10] T Altiero L Rebecchi and R Bertolani ldquoPhenotypic varia-tions in the life history of two clones of Macrobiotus richtersi(Eutardigrada Macrobiotidae)rdquo Hydrobiologia vol 558 no 1pp 33ndash40 2006

[11] M Vukich P L Ganga D Cavalieri et al ldquoBIOKIS amodel payload for multisciplinary experiments in micrograv-ityrdquo Microgravity Science and Technology vol 24 pp 397ndash4092012

[12] A M Rizzo L Adorni G Montorfano F Rossi and B BerraldquoAntioxidant metabolism of Xenopus laevis embryos duringthe first days of developmentrdquo Comparative Biochemistry andPhysiologymdashB Biochemistry and Molecular Biology vol 146 no1 pp 94ndash100 2007

[13] O H Lowry N J Rosebrough A L Farr and R J RandallldquoProtein measurement with the Folin phenol reagentrdquo TheJournal of Biological Chemistry vol 193 no 1 pp 265ndash275 1951

[14] F Paoletti and A Mocali ldquoDetermination of superoxide dis-mutase activity by purely chemical system based on NAD(P)HoxidationrdquoMethods in Enzymology vol 186 pp 209ndash220 1990

[15] H Aebi ldquoCatalase in vitrordquoMethods in Enzymology vol 105 pp121ndash126 1984

[16] M C Pinto A M Mata and J Lopez-barea ldquoReversible inacti-vation of Saccharomyces cerevisiae glutathione reductase underreducing conditionsrdquo Archives of Biochemistry and Biophysicsvol 228 no 1 pp 1ndash12 1984

[17] J R Prohaska and H E Ganther ldquoSelenium and glutathioneperoxidase in developing rat brainrdquo Journal of Neurochemistryvol 27 no 6 pp 1379ndash1387 1976

[18] O W Griffith ldquoGlutathione and glutathione disulphiderdquo inMethods of Enzymatic Analysis H U Bergmeyer Ed vol 3 pp521ndash529 Academic Press New York NY USA 1984

[19] H E Wey L Pyron and M Woolery ldquoEssential fatty acid defi-ciency in cultured human keratinocytes attenuates toxicity dueto lipid peroxidationrdquo Toxicology and Applied Pharmacologyvol 120 no 1 pp 72ndash79 1993

[20] J Folch M Lees and G H S Stanley ldquoA simple method for theisolation and purification of total lipides from animal tissuesrdquoThe Journal of Biological Chemistry vol 226 no 1 pp 497ndash5091957

[21] M B Franca A D Panek and E C A Eleutherio ldquoOxidativestress and its effects during dehydrationrdquoComparative Biochem-istry and PhysiologymdashA Molecular and Integrative Physiologyvol 146 no 4 pp 621ndash631 2007

[22] R Cruz de Carvalho M Catala J Marques da Silva CBranquinho and E Barreno ldquoThe impact of dehydration rateon the production and cellular location of reactive oxygen

species in an aquatic mossrdquo Annals of Botany vol 110 no 5 pp1007ndash1016 2012

[23] I Kranner and S Birtic ldquoA modulating role for antioxidants indesiccation tolerancerdquo Integrative and Comparative Biology vol45 no 5 pp 734ndash740 2005

[24] R Cornette and T Kikawada ldquoThe induction of anhydrobiosisin the sleeping chironomid current status of our knowledgerdquoIUBMB Life vol 63 no 6 pp 419ndash429 2011

[25] L Rebecchi ldquoDry up and survive the role of antioxidantdefences in anhydrobiotic organismsrdquo Journal of Limnology vol72 no 1 pp 62ndash72 2013

[26] A M Rizzo M Negroni T Altiero et al ldquoAntioxidant defencesin hydrated and desiccated states of the tardigrade Paramac-robiotus richtersirdquo Comparative Biochemistry and Physiology BBiochemistry and Molecular Biology vol 156 no 2 pp 115ndash1212010

Research ArticleIdentification of Reference Genes in Human MyelomonocyticCells for Gene Expression Studies in Altered Gravity

Cora S Thiel123 Swantje Hauschild123 Svantje Tauber123 Katrin Paulsen12

Christiane Raig1 Arnold Raem4 Josefine Biskup12 Annett Gutewort12 Eva Huumlrlimann1

Felix Unverdorben1 Isabell Buttron1 Beatrice Lauber1 Claudia Philpot5 Hartwin Lier6

Frank Engelmann67 Liliana E Layer1 and Oliver Ullrich1238

1 Institute of Anatomy Faculty of Medicine University of Zurich Winterthurerstraszlige 190 8057 Zurich Switzerland2Department of Machine Design Engineering Design and Product Development Institute of Mechanical EngineeringOtto-von-Guericke-University Magdeburg Universitatsplatz 2 39106 Magdeburg Germany3Study Group ldquoMagdeburger Arbeitsgemeinschaft fur Forschung unter Raumfahrt- und Schwerelosigkeitsbedingungenrdquo (MARS)Otto-von-Guericke-University Magdeburg Universitatsplatz 2 39106 Magdeburg Germany4Arrows Biomedical Deutschland GmbH Center for Nanotechnology at the Westfalische Wilhelms-Universitat MunsterHeisenbergstraszlige 11 48149 Munster Germany5German Aerospace Center Space Agency Konigswinterer Straszlige 522-524 53227 Bonn Germany6KEK GmbH Kemberger Straszlige 5 06905 Bad Schmiedeberg Germany7University of Applied Science Jena Carl-Zeiss-Promenade 2 07745 Jena Germany8Zurich Center for Integrative Human Physiology (ZIHP) University of Zurich Winterthurerstraszlige 190 8057 Zurich Switzerland


Received 14 May 2014 Accepted 4 September 2014


Copyright copy 2015 Cora S Thiel et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Gene expression studies are indispensable for investigation and elucidation of molecular mechanisms For the process ofnormalization reference genes (ldquohousekeeping genesrdquo) are essential to verify gene expression analysis Thus it is assumed thatthese reference genes demonstrate similar expression levels over all experimental conditions However common recommendationsabout reference genes were established during 1 g conditions and therefore their applicability in studies with altered gravity has notbeen demonstrated yetThemicroarray technology is frequently used to generate expression profiles under defined conditions andto determine the relative difference in expression levels between two ormore different states In our study we searched for potentialreference genes with stable expression during different gravitational conditions (microgravity normogravity and hypergravity)which are additionally not altered in different hardware systems We were able to identify eight genes (ALB B4GALT6 GAPDHHMBS YWHAZ ABCA5 ABCA9 and ABCC1) which demonstrated no altered gene expression levels in all tested conditions andtherefore represent good candidates for the standardization of gene expression studies in altered gravity

1 Introduction

Since several limiting factors for human health and perfor-mance in microgravity have been clearly identified [1] ithas been concluded that substantial research and develop-ment activities are required in order to provide the basicinformation for appropriate integrated risk managementincluding efficient countermeasures and tailored life supportsystems [2] In particular bone loss during long stays in

weightlessness still remains an unacceptable risk for long-term and interplanetary flights [3] and serious concernsarose whether spaceflight-associated immune system weak-ening ultimately precludes the expansion of human presencebeyond Earthrsquos orbit [4] The immune and skeletal systemsare tightly linked by cytokine and chemokine networks anddirect cell-cell interactions [5 6] and the immune systeminfluences metabolic structural and functional changes inbones directly [6] Both systems share common cellular



players such as the osteoclasts which are bone-residentmacrophages and derivatives of monocytic cells Thereforeknowing the cellular and molecular mechanisms of howgravity influences cell function is a valuable requirementto provide therapeutic or preventive targets for keepingimportant physiological systems fully functional during long-term space missions

Since the first pioneering in vitro studies that revealed thatcells of the immune system are sensitive to changes of grav-itational force [7ndash10] several studies in real and simulatedmicrogravity have confirmed microgravity-induced alter-ations in the molecular mechanisms and signal transductionprocesses in leukocytes including themonocytemacrophagesystem (MMS) [11 12] The MMS belongs to the innateimmune system and is characterized by a fast but nonspecificimmune reaction the first line of defense against invadingpathogens Cells of the MMS in microgravity demonstrateddisturbed cytokine release [13ndash15] reduced oxidative burst[16 17] alteration of the cytoskeleton [18] and reductionin their locomotion ability [19] Importantly analysis ofgene expression of monocytes during an ISS experimentrevealed significant changes in gene expression associatedwith macrophageal differentiation [20]

Differential gene expression analyses are a widely usedmethod to investigate the influence of different treatmentsor conditions on a cell system The resulting changes on themolecular level can be investigated either by reverse tran-scription quantitative real-time PCR (RT-qPCR) as majortechnique for the sensitive and robust analysis of expressionlevels of specific genes [21ndash27] and microarrays for wholegenome or transcriptome analyses [28]

After the genome sequencing era when numerousgenomes were completely decoded the focus of interest shi-fted towards genome wide expression level analyses so that asnapshot of the whole genome expression profile is obtainedin a single experiment [28ndash30] offering also a possibility toobtain an insight into networks and pathways of biomolecularinteractions on a large scale [29ndash33] The technology behindmicroarray analysis developed fast and different suppliersused different protocols for for example hybridization anddata normalization Therefore it was and still is difficultto establish standards for the experimental procedure andprocessing of the raw data obtained [30] Consequently aconcept for the development of standards for microarrayexperiments and data has been presented by the microarraygene expression database group (MGED) describing themin-imum information about a microarray experiment (MIAME[34]) This compilation covers (1) the experimental design(2) the array design (3) samples (4) hybridizations (5)measurements and (6) normalization controls [34] Also forRT-qPCR technique [35ndash38] standard guidelines (MIQE =minimum information for the publication of quantitativereal-time PCR) were developed [27 39ndash41]

One of the most crucial requirements of standardizationare suitable internal controls so called reference genes that areused for data normalization which are important to accountfor differences in the amount and quality of starting materialas well as reaction efficiency [42] GAPDH HPRT 120573-actintubulin and ribosomal RNA genes are typical examples for

frequently used reference genes [43ndash45] However referencegenes have to be tested for their suitability as an endogenouscontrol in each case prior to the experiment This is ofhigh importance substantiated by many studies reportingexpression effects of classical ldquohousekeeping genesrdquo uponexperimental treatments [46ndash49] A selection of severalreference genes used simultaneously can also be a good wayto further increase reliability of the resulting data [50 51] Infact recommendations state to identify three stable referencegenes for each planned assay to assure a reliable outcome[50 52]The identification of stably expressed reference genescan be performed in a pilot study using dedicated algorithmslike geNorm or BestKeeper or a combination hereof wherea minimum of eight potential reference genes are tested andranked according to their stability being an indication fortheir suitability as control genes for normalization [50 51]Candidate reference genes for such a study may be forexample chosen from the literature or from experimentaldata obtained from microarray analysis [27 51]

However common recommendations about referencegenes were established during 1 g conditions and there-fore their applicability in studies with altered gravity condi-tions has not been intensively demonstrated so far Altho-ugh there are numerous publications describing differentialgene expression analyses under simulated and real microgra-vity conditions in various cells types and tissues (suppleme-ntary Table 1 available online at httpdxdoiorg1011552014363575) a systematic research on reference genes stableunder altered gravity conditions has not been published yet

In our study we used microarray analyses to investigatethe differential gene expression in U937 cells a myelomono-cytic human cell line exposed to short-term (20 seconds) andmiddle-term (6 to 7 minutes) microgravity and hypergravityduring parabolic flights and sounding rocket flights twoplatforms commonly used by researchers to investigate theeffects of real microgravity Our experimental goal was toidentify potential reference genes that can be recommendedto the community of gravitational biology for differentialexpression analysis performed with cells of the immunesystem on those two frequently used platforms Thereforewe chose 22 reference genes widely used throughout theliterature and screened our microarray data for these partic-ular genes evaluating their stability for possible applicationas control genes Besides the highly conserved ribosomalRNA genes and others ABC transporter and tRNA genesbelong to evolutionary well-conserved genes as well Sinceribosomal RNA and tRNA genes are not represented onthe array we decided to adhere to tRNA related genes liketRNA synthetases as these play a central role in basal cellularfunctions and should be robustly expressed to ensure cellsurvival Therefore our study comprised published referencegenes ABC transporters and tRNA related genes


21 Cell Culture U937 cells (ATCC CRL15932) originatingfrom a diffuse histiocytic lymphoma displayingmanymono-cytic characteristics were used as a model cell line to inves-tigate the differential gene expression under altered gravity


conditions in monocyticmacrophageal cells U937 cells werecultured in RPMI 1640medium (BiochromMerckMilliporeGermany) supplemented with 10 fetal bovine serum (FBSSuperior BiochromMerck Millipore Germany) 2mM glu-tamine (GibcoLife Technologies Germany) and 100UmLpenicillin as well as 100 120583gmL streptomycin (GibcoLifeTechnologies Germany) Cells were seeded with a densityof 02 times 106 cellsmL and the medium was exchanged every48 hours Cells were harvested by centrifugation at 300 g for5min at room temperature resuspended in fresh mediumand an aliquot was used for an adequate dilution with trypanblue to count the vital cell number Cells were reseeded infresh medium at a concentration of 02 times 106 cellsmL

22 Parabolic Flight Experiments We designed and constru-cted an experiment module suitable to perform cell cultureexperiments with living mammalian cells during parabolicflights on board the Airbus A300 ZERO-G During the 19thDLR parabolic flight campaign (PFC) we focused on theanalysis of differential gene expression in U937 cells consi-dering the different gravity conditions in-flight 1 g 18 gand 0 g Experiments were only performed during the firstparabola to assure that the investigated differential geneexpressions are generated by a direct effect of gravitationalchange and not an accumulated long-term effect Duringthe 19th DLR PFC experiments were reproduced on twoindependent flight days

In search of rapidly responsive molecular alterationsin mammalian cells short-term microgravity provided byparabolic flight maneuvers is an ideal instrument to eluci-date initial and primary effects without the influence andinterference of secondary signal cascades Parabolic flightsprovide 1 g 18 g and microgravity (120583g) with a quality ofapproximately 10minus2 to 10minus3 g For the 19th DLR PFC 1 times 107U937 cells in 10mL medium (RPMI 1640 supplemented with100UmL penicillin 100 120583gmL streptomycin 250 ngmLamphotericin B (GibcoLife Technologies Germany) 2mMglutamine and 2 FBS (ie serum starved)) were filledinto 200mL Nutrimix bags (B Braun Melsungen Germany)and transported from the home laboratory to the preflightpreparation laboratories at the NOVESPACE premises inBordeaux France After arrival cells were destarved byaddition of 08mL FBS per Nutrimix bag and used for theflight experiment on the following day For the flight daythe Nutrimix bags were placed in a solid plastic containerto create a double containment to prevent spillage of fluidsin the aircraft in case of leakage which is strictly prohibitedby the NOVESPACE regulations The rapid preservation ofthe effects of altered gravity on the gene expression in theU937 cells was achieved by injection of 50mL of RLT buffer(Qiagen Germany) a lysis buffer immediately lysing cellsand tissues prior to RNA isolation The 1 g in-flight controlswere performed 5min before the first parabola and the 18 gsample directly before the microgravity phase of the firstparabola The 120583g samples were fixed directly at the end ofthe microgravity phase of the first parabola Samples weretransported to the laboratory immediately after landing 1 gground controls were performed immediately after landing

using the experimental module in the aircraft In total 30samples were obtained during two parabolic flight days 6x 1 gground controls 9x 1 g in-flight controls 6x 18 g and 9x 120583g

23 RNA Isolation after the Parabolic Flight ExperimentsAfter landing of the aircraft and transport of the samples tothe laboratory on site facilities the containers were disassem-bled the Nutrimix bags were gently agitated and the lysedcell solution was filled into a T75 straight neck cell cultureflaskThe cell solutionwas vortexed for 10 sec and passed fourtimes through aOslash 08times 120mmneedle (B BraunMelsungenGermany) fitted to a 50mL syringe 50mL of absoluteethanol was added and precipitates were resuspended byvigorous shaking A valve and a sterile connective piecewere placed on a QIAvac 24 plus vacuum system (QiagenGermany) and an RNAmaxi column (Qiagen Germany) wasattached to the connective piece A vacuum of minus200mbarwas adjusted and the column was loaded with the lysed cellsuspension Then the valve was closed and the column wascentrifuged at 4000 g for 3min 15mL of buffer RW1 (QiagenGermany) was applied for washing membrane bound RNAAfter centrifugation at 4000 g for 7min the flow throughwas discarded and two washing steps with 10mL RPE buffer(Qiagen Germany) followedwith centrifugation at 4000 g for3min and 10min respectively The column bound RNA waseluted by application of 600 120583L of RNase-free water (QiagenGermany) incubation for 1min at room temperature andcentrifugation for 4min at 4000 g The elution step wasrepeated with the first eluate The RNA was transported atapproximately minus150∘C in a Cryo Express dry shipper (CX-100 Taylor-Wharton USA) prepared with liquid nitrogenand stored at minus80∘C until the processing of the RNA for themicroarray analysis

24 Experiments during the TEXUS-49 Sounding Rocket Cam-paign For theTEXUS-49 campaign at ESRANGE (EuropeanSpace and Sounding Rocket Range Kiruna Sweden) U937cells were cultured in the fully installed laboratories on siteCells were seeded with a density of 02 times 106 cellsmL andthe medium was exchanged every 48 hours as describedabove On the launch day cells were visually inspectedharvested counted and pooled to a concentration of 5 times 107cellsmL 05mL of this cell suspension was filled in a sterile3mL plastic syringe shortly before the launch Additionallyone syringe was filled with 03mL of cell culture mediumand another one with 1mL Trizol LS (Life TechnologiesGermany) The three syringes were mounted on a plasticblock with a tubing system connecting them This unit wasfinally integrated into the automatically operated experimentsystem In total 35 of these experiment units were preparedand were kept at 37∘C until the integration into the payloadof the rocket

During the experimental run firstly the 03mL ofmedium as a potential placeholder for an activation solutionand secondly the 1mL of Trizol LS were injected to thecell suspension at defined time points to lyse the cells andpreserve the current status of differential gene expressionThis sequential injection of fluids was performed at 75 secafter launch to monitor the so-called baseline (BL) directly


before the 120583g phase and at 375 sec after launch shortly beforethe end of the 120583g phase A group of 1 g ground controls werekept on ground in the incubator simultaneously to the 120583gsample group

TEXUS-49 consisted of a VSB-30 engine (S-30 solidrocket stage with an S-31 second stage) and of the payloadThe rocket was launched on March 29 2011 at 0601 amfrom the ESRANGE Space Center near Kiruna SwedenDuring the ballistic suborbital flight an altitude of 268 kmand 378 sec of microgravity with a quality of 10minus5 g wereachieved Further parameters include first stage peak thrustacceleration 63 g mean thrust acceleration 503 g burnout at123 sec and engine separation at 136 sec second stage peakthrust acceleration 135 g mean thrust acceleration 730 gburnout at 430 sec yo-yo despin at 560 sec and engineseparation at 590 sec

25 RNA Isolation after the TEXUS-49 Sounding RocketCampaign Directly after landing localization and recoveryof the payload the experiment modules were dismantledand handed over to the scientists The cell suspension wassheared three times with a 20G needle (B BraunMelsungenGermany) and distributed in two 20mL tubes 01mL ofchloroform (Sigma-Aldrich Germany) was added and thehomogenate was vortexed for 15 sec and incubated for 5minat room temperature before a 15min centrifugation step at11000 g and 4∘C The upper phase of both 20mL tubeswas transferred into a 15mL tube and 4mL of RLT bufferas well as 3mL of absolute ethanol was added and mixed4mL of this solution was pipetted on an RNA Midi column(Qiagen Germany) and centrifuged for 30 sec at 3000 g androom temperature The flow through was discarded and theresidual 4mL of RNA solution was loaded on the column andcentrifuged for 5min at 3000 g at room temperature Thenthe columns were washed twice with 25mL of RPE bufferand centrifuged for 2min and 5min respectively at 3000 gat room temperature The RNA was eluted by addition of250120583L RNase-free water (Qiagen Germany) to the columnincubation for 1min at room temperature and centrifugationfor 3min at 3000 g and room temperature The eluate wasloaded again onto the column followed by a 1min incubationand centrifugation for 5min at 3000 g and room temperatureThe isolated RNA was transferred into sterile cryotubes andstored until the return transport at approximately minus150∘Cin a Cryo Express dry shipper (CX-100 Taylor-WhartonUSA) preparedwith liquid nitrogen After arrival in the homelaboratory samples were stored at minus80∘C until the processingof the RNA for the microarray analysis

26 RNA Processing and Microarray Analysis RNA quantityand purity were analyzed spectrophotometrically using aNanoDrop 1000 (Thermo Scientific USA) Isolated RNAsamples were all of high quality with 260280 nm ratiosbetween 19 and 21 The RNA integrity was measured usingan Agilent 2100 Bioanalyzer (Agilent Technologies USA)Only RNA with an RNA integrity number (RIN) gt 87 wasused for the following microarray analysis 400 ng total RNAwas applied to Cy3-labeling with the ldquoLow RNA Input LinearAmplification Kit PLUS One-Colorrdquo (Agilent Technologies)

and hybridized for 175 h to a NimbleGen expressionmicroar-ray (12 times 135000 features) employing the ldquoGene Expres-sion Hybridization Kitrdquo (Agilent Technologies) Afterwardsarrays were washed and scanned by the Microarray ScannerG2505B (Agilent Technologies)

The image files of the scanner were analyzed with theNimbleScan Software 26 using the robustmultiarray analysis(RMA) with the default parameters RMA a probe-levelsummarization method identifies probes that are outliers inthe overall behavior of the expression measured for a givengene The contribution of outlier probes is reduced in thereported gene expression level which has been demonstratedto improve the sensitivity and reproducibility of microarrayresults In addition to screening outlier probes NimbleScansoftwarersquos implementation of RMA [53] used quantile nor-malization and background correction

The normalized microarray data were analyzed usingPartek Genomics Suite 66 Statistical analysis was performedusing the one-wayANOVAand the false discovery rate (FDR)[54] for multiple-testing correction Further the coefficientof variation (CV) expressed in percent was calculated alsoknown as ldquorelative variabilityrdquo It equals the standard devi-ation divided by the mean An integration tool (availableat httpwwwleonxiecomreferencegenephp) [50 51 55]of four algorithms (geNorm NormFinder BestKeeper andthe comparative delta-CT method) was used to evaluate theexpression stability of the reference genes On the basis ofthe resulted rankings from the four algorithms an overallranking of the candidate genes was achieved

27 Statistical Analysis of Selected Genes Genes of interestwere identified and the log 2 values of the measured fluo-rescent intensities returned by the Partek software were backcalculated to linear values Then means of all values of thesame gene generated by different probes were calculated ifat least three values existed excluding outliers Subsequentlystandard deviations were calculated for the means and anunpaired t-test with Welch correction was performed usingExcel 2011 (119905-test tails 2 type 3) to obtain statistical signifi-cance

3 Results

The aim of our study was to identify a group of genes thatshow a stable nonchanging expression profile in immunecells under altered gravity conditions over a time rangeof seconds until several minutes Therefore we performedexperiments on the 19th DLR PFC and the sounding rocketmission TEXUS-49 two platforms that offer microgravitytimes of 20 seconds and 6 minutes respectively During bothmissions U937 cells a model for monocyticmacrophagealcells of the human immune system were exposed to differentgravity conditions for various time periods (see Table 1)During the 19th DLR PFC cells were exposed only to the firstparabola with the following sequence 1 g in-flight control18 g and microgravity (120583g) Cells were subjected to alteredgravity conditions of 18 g and 120583g for 20 seconds in each caseand were immediately fixed and stored cooled until RNAisolation In case of theTEXUS-49 campaign cells underwent


134

12

105

9

75

6

461 3 5 7 9 11 13 15 17 19 21 23 25 27

Sample IDGravity condition

Log

inte

nsiti

es19th DLR PFC

120583gHW

1g18 g

(a)

1 2 3 4 5 6 7 8 9 10 11 12 16 17Sample ID

13 14 15 18

134

12

105

9

75

6

46

Log

inte

nsiti

es

TEXUS-49

Gravity condition120583gHW

BL

(b)Figure 1 Boxplots showing the log expression values of individual microarrays The central line represents the 50th percentile or medianwhereas the upper and lower boundaries of the box display the 75th and 25th percentile respectively The upper and lower bars represent the9th and the 91st percentile Two experimental data sets are displayed (a) 28 microarrays hybridized with samples from the 19th DLR PFC (8x120583g 6x HW 8x 1 g 6x 18 g) and (b) 18 microarrays hybridized with samples originating from the TEXUS-49 campaign (7x 120583g 6x HW 5x1 g) The expression data show an even distribution for the displayed log intensities

Table 1 Gravity conditions 19th DLR PFC and TEXUS-49

Gravity condition 19th DLR PFC TEXUS-491 g ground controls(hardware HW) HW HW

Microgravity 120583g (20 sec) 120583g (378 sec)1 g in-flight control 1 g mdashIn-flight baseline (hyper-gphase directly before 120583gBL)

18 g (20 sec) BL (1 gmdashmax135 g 75 sec)

the following sequence of altered gravity hypergravity up to135 g during the first 75 seconds after liftoff and 120583g for 378seconds Hypergravity is defined as the baseline (BL) becausesamplesmirror the vibration and hypergravity effects directlybefore the microgravity phase In both experimental setupson ground 1 g hardware controls (HW) were performed tobe able to differentiate between the effects caused by theconditions experienced before hypergravity and 120583g and thealtered gravity conditions themselves After the campaignsthe RNA samples were analyzed for quantity and quality byNanoDrop spectrophotometry and a bioanalyzer analysisand only samples with an RNA integrity number (RIN)higher than 87 were chosen for subsequent microarrayanalysis 12 times 135 K Roche NimbleGen arrays were hybridizedand data were collected after the normalization procedure Intotal we obtained data from 46 single microarrays (19th DLRPFC 8x 120583g 6x HW 8x 1 g and 6x 18 g TEXUS-49 7x 120583g6x HW and 5x BL)

Data tables were compiled individually for the 19th DLRPFC and TEXUS-49 including all gravity conditions listedin Table 1 and a first overview of the datasets was providedby a boxplot diagram (Figures 1(a) and 1(b)) Boxplots are auseful tool to visualize the variation within a microarray and

between microarrays The central line shows the position ofthe median while the upper and the lower boundaries repre-sent the upper (75th percentile) and lower (25th percentile)quartile The ends of the tails display the 9th and the 91stpercentile The boxplots of the microarray data show thatthere is only little variation within a single array and betweenthe arrays that belong to the same gravity condition Figure 1shows that the quality of both data sets (19th DLR PFC andTEXUS-49) is sufficient to proceed with further analyses

In search of potential reference genes for gravitationalstudies in this monocyticmacrophageal cell system we firstperformed PubMed database search to identify commonlyused reference genes in RNA expression analyses in humancells We found 22 genes that were used in several reversetranscription quantitative real-time PCR (RT-qPCR) studiesas control genes for normalization (Table 2 supplementaryTable 2) The microarray data tables were screened for these22 widely used reference genes and 20 of them could belocated on the Roche NimbleGen 12 times 135 K array that wasused in our experiments Two genes coding for 5s and 18srRNAs could not be identified since they are not spotted onthe array The PFC and TEXUS data sets were screened forthose 20 selected potential reference genes and fluorescenceintensities were compiled for each gene and each gravitycondition in heatmaps (Figures 2(a) and 2(b)) Overall flu-orescence intensities for all samples showed only minor dif-ferences in the heatmaps A more detailed visual inspectionrevealed completely equal fluorescence intensities for ACTBALB B4GALT6 HMBS HPRT1 PPIA RPLP0 and YWHAZfor the gravity conditions prevailing during the 19thDLRPFC(Figure 2(a)) The gravity conditions investigated during theTEXUS-49 campaign showed stable expression values for thegenes ACTB ALB B4GALT6 GUSB PLA2G4A POLR2APPIA TBP UBC and YWHAZ (Figure 2(b)) For furthercharacterization and identification of stable reference genes


Table 2 List of potential reference genes

Potential reference gene Genesymbol Citation

5s rRNA [85]18s rRNA [86]120573-Actin ACTB [49 51]Albumin ALB [49]120573-2 microglobin B2M [49 51 56]UDP-GalbGlcNAcb14-galactosyl-transferasepolypeptide 6

B4GALT6 [86]

Glucose 6-phosphatedehydrogenase G6PD [49]

Glyceraldehyde-3-phosphatedehydrogenase GAPDH [49 51 56 85

87 88]Glucuronidase beta GUSB [86]Hydroxymethylbilane synthase(porphobilinogen deaminase) HMBS [49 51 56]

Hypoxanthinephosphoribosyltransferase 1 HPRT1 [49 51 56 86]

Heat shock protein 90 kDa HSP90AA1 [86]Phospholipase A2 PLA2G4A [49]RNA polymerase II POLR2A [49 86]Peptidylprolyl isomerase A(Cyclophilin A) PPIA [49 86]

Ribosomal protein L13 RPL13A [49 51 56 86]Acidic ribosomalphosphoprotein P0 RPLP0 [89]

Succinate dehydrogenasecomplex subunit A SDHA [51 56 86]

TATA box binding protein TBP [49 51 56 86]120572-Tubulin TUBA1 [49]Ubiquitin C UBC [51 56]Tyrosine 3-monooxygenasetryptophan 5-monooxygenaseactivation protein

YWHAZ [51 56 86]

we performed a geNorm pilot study [51] and calculated thecoefficient of variation (CV) for all 20 potential referencegenes (Figure 3) For homogeneous groups CV values below25 and for heterogeneous groups CV values below 50 areacceptable [56] Rapid and extreme changes in gravity inducestrong changes in cellular functions Therefore we classifiedour samples as heterogeneous groups According to the setcriteria all analyzed potential reference genes showed CVvalues below 50 for the PFC and TEXUS data sets (Figures3(a) and 3(b)) In the sample set of the 19th DLR PFC allgenes but HMBS fulfill even the more stringent criterion of aCVbelow 25 (Figure 3(a)) For the samples collected duringthe TEXUS-49 campaign all genes butHPRT1 andPLA2G4Adisplay CV values below 25 (Figure 3(b))

To increase the number of potential reference genes thatcan be used as standards for differential expression anal-yses in gravitational studies we extended our analysis toevolutionary highly conserved genes We hypothesized that

ACTB

ALB

B2M

B4GALT

6GAPD

HG

USB

HMBS

HPR

T1H

SP90

AA

1H

SP90

AB1

PLA

2G4A

POLR

2APP

IARP

L13A

RPLP

0SD

HA

TBP

TUBA

1U

BCYW

HAZ

10000

19th DLR PFC

120583g

1g

HW

108 g

1755000 3500000

(a)

ACTB

ALB

B2M

B4GALT

6GAPD

HG

USB

HMBS

HPR

T1H

SP90

AA

1H

SP90

AB1

PLA

2G4A

POLR

2APP

IARP

L13A

RPLP

0SD

HA

TBP

TUBA

1U

BCYW

HAZ

TEXUS-49

120583g

HW

BL

10000 1755000 3500000

(b)

Figure 2 Heatmaps for selected reference genes The graph illus-trates fluorescent intensity levels of the 20 potential reference genesfromTable 2 between the three and four different gravity conditionsrespectively Each gene is represented in one column and eachgravity condition is represented in one row (a) 120583g HW 1 g and18 g (19th DLR PFC) and (b) HW BL and 120583g (TEXUS-49) Theheatmap shows large variation in fluorescence intensities for thedifferent genes However within the same gene expression levels aresimilar for all tested conditions The lower bar with the graduatedred colors is the measure for the different fluorescence intensities

genes stable over time and taxonomic kingdoms should havevery fundamental functions within a cell and thus be largelyindependent from external influences to ensure basic cellularfunctions Besides ribosomal RNA genes which are notrepresented on the microarray applied in this study ABCtransporters and tRNA genes are also evolutionary highlyconserved over a wide variety of organisms Unfortunatelythe 12 times 135 K Roche NimbleGen array does also not containprobes for tRNA Therefore we had a look at expressionprofiles of ABC transporters and tRNA related genes (sup-plementary Tables 2 and 3) Since almost all fluorescencevalues of tRNA related genes showed a high variance making


43387344301258215172129

9643

Coe

ffici

ent o

f var

iatio

n (C

V) (

) 19th DLR PFC

ACTB

B2M

GU

SB

HPR

T1H

SP90

AA1

HSP

90

AB1

PLA2

G4

APO

LR2

APP

IARP

L13

A

SDH

A

TUBA

1

UBCALB

B4GALT

6GAPD

H

HMBS

RPLP

0

TBP

YWHAZ

0

(a)

43387344301258215172129

9643

0Coe

ffici

ent o

f var

iatio

n (C

V) (

)

ACTB

ALB

B2M

B4GALT

6GAPD

HG

USB

HMBS

HPR

T1H

SP90

AA1

HSP

90

AB1

PLA2

G4

APO

LR2

APP

IARP

L13

ARP

LP0

SDH

ATB

PTU

BA1

UBC

YWHAZ

TEXUS-49

(b)Figure 3 Coefficient of variation calculation for the potential reference genes This bar chart displays the coefficient of variation (CV) in of the 20 potential reference genes across the gravity conditions for the 19th DLR PFC (HW 1 g 18 g 120583g) and TEXUS-49 (HW BL 120583g) Alower value corresponds to higher stability in gene expression (a) 19th DLR PFC All calculated CV values are below the threshold of 50(b) TEXUS-49 all CV values are below 50 but in total more genes show higher coefficients of variation

19th DLR PFC

ABC

C1

ABC

C4

ABC

D4

40ABC

F2

TAP2

1 3 5 7 9 25 27 33 35 4511 13 15 17 19 21 31 39 41 4323 29 37 47

10000

120583g

HW

130000 250000

1g

18 g

(a)

HW

BL

120583g

TEXUS-49

ABC

C1

ABC

C4

ABC

D440

ABC

F2

TAP2

1 3 5 7 9 25 27 33 35 4511 13 15 17 19 21 31 39 41 4323 29 37 47

10000 130000 250000

(b)

Figure 4 Heatmaps for highly conserved ABC transporters The fluorescent intensity levels of the 47 ABC transporter genes shown insupplementary Table 2 were quantified for the different gravity conditions Each gene is represented in one column and each gravity conditionis represented in one row (a) 120583g 1 g HW and 18 g (19thDLRPFC) and (b) 120583g BL andHW (TEXUS-49)The heatmaps show large variationin fluorescence intensities for the different genes However within the same gene expression levels are mostly similar for all tested conditionsThe lower bar with the graduated red colors is the measure for the different fluorescence intensities

reasonable analysis impossible we concentrated on the ABCtransporters Heatmap analyses were carried out to obtain afirst impression on the gene stability (Figure 4) The samplesfrom the 19th DLR PFC and TEXUS-49 mission also showa rather high variation in fluorescence intensities (Figures4(a) and 4(b)) The calculation of the CV for these samples(Figure 5) displays higher values compared to the potentialreference genes however taken together all analyzed samplesof the 19th DLR PFC fulfill the criterion of CV values lessthan 50 in case of ABC transporter signals (Figure 5(a))Out of 47 samples 36 are also below 25CV Althoughthree samples from TEXUS-49 showed values above 50CV

(Figure 5(b)) 37 samples stayed below the 25 threshold(Figure 5(b)) Selected reference genes and ABC transporters(marked in bold Figures 4 and 5) were chosen for furtherdetailed analysis of differential gene expression under alteredgravity conditions

For nine of the potential reference genes from the litera-ture there were at least three values returned by themicroarray generated by independent probes targeting thesame gene Two of these genes were excluded from furtheranalysis due to high variance between their single values(HSP90AA1 and PPIA) and the remaining seven genes (ALBB4GALT6 GAPDH HMBS RPLP0 TBP and YWHAZ (see


401 3 5 7 9 25 27 33 35 4511 13 15 17 19 21 31 39 41 4323 29 37 47

ABC

C1

ABC

C4

ABC

D4

ABC

F2

TAP2

69

621

552

483

414

345

276

207

138

69

0

Coe

ffici

ent o

f var

iatio

n (C

V) (

)

19 th DLR PFC

(a)

401 3 5 7 9 25 27 33 35 4511 13 15 17 19 21 31 39 41 4323 29 37 47

ABC

C1

ABC

C4

ABC

D4

ABC

F2

TAP2

69621552483414345276207138

690

Coe

ffici

ent o

f var

iatio

n (C

V) (

)

TEXUS-49

(b)Figure 5 Coefficient of variation calculation for the ABC transporter genes This bar chart displays the coefficient of variation (CV) in ofthe 47 ABC transporter genes across the gravity conditions for the 19th DLR PFC (HW 1 g 18 g 120583g) and TEXUS-49 (HW BL 120583g) A lowervalue corresponds to higher stability in gene expression (a) 19th DLR PFC all calculated CV values are below the threshold of 50 and fulfillthe criterion (b) TEXUS-49 three genes showCV values higher than 50 andwere excluded from further analysesThe numbers correspondto the ABC transporters listed in supplementary Table 2 Genes that were further analyzed are labeled and marked in bold (ABCC1 ABCC4ABCD4 ABCF2 and TAP2)

GAPDH HMBS RPLP0 TBP YWHAZ0

5000

10000

15000

20000

25000

30000 19th DLR PFC potential reference genes

Fluo

resc

ence

inte

nsity

lowast

lowast lowast

lowastlowastlowast

HW1g

18 g120583g

(a)

Fluo

resc

ence

inte

nsity

ALB B4GALT60

20

40

60

HW1g

18 g120583g

19th DLR PFC potential reference genes

(b)

Figure 6 Influence of altered gravity during parabolic flight on potential reference genes RNA expression levels after 1 g (light gray) 18 g(dark gray) and 120583g (black) conditions during the 19th DLR parabolic flight campaign Hardware ground controls (HW striped) are shownfor each experimental group RNA expression levels are shown as fluorescence intensities (a) The expression values for GAPDH HMBSRPLP0 TBP and YWHAZ are displayed (b) ALB and B4GALT6 show low but stable fluorescent intensities GAPDH HMBS YWHAZALB and B4GALT6 show no significant change in RNA levels upon altered gravity for 20 sec while RPLP0 displays 120583g sensitivity comparedto 1 g and TBP reacts sensitively to all g conditionsMean values of at least threemeasurements with standard deviations are shown lowast119875 lt 005lowastlowast119875 lt 0005

Table 3)) were subjected to further statistics The calculationof the mean fluorescence intensity levels revealed that dif-ferent ranges of transcript abundance are present in bothexperimental setups While ALB and B4GALT6 seem to beexpressed rather low GAPDH HMBS RPLP0 TBP andYWHAZ are represented inmuch higher abundance (Figures6 and 7) The comparison of mean fluorescence intensities ofone gene under different g conditions revealed that GAPDHHMBS YWHAZ ALB and B4GALT6 are stably expressedwith respect to all investigated gravity conditions duringparabolic flight of the 19th DLR PFC (Figures 6(a) and 6(b))

RPLP0 is significantly upregulated by 120583g compared to 1 gwhile TBP is initially downregulated by 18 g and recoversduring 120583g (Figure 6(a)) Furthermore comparison of in-flight 1 g controls to 1 g ground controls (HW) shows asignificantly reduced mRNA level of TBP portending thatduring the preexperimental phase a certain kind of stress wasaccumulated in the cells influencing its expression level

The data analysis of the TEXUS-49 sounding rocketexperiment reveals stable RNA expression levels throughoutthe different g levels for GAPDH HMBS RPLP0 YWHAZALB and B4GALT6 (Figures 7(a) and 7(b) Table 4 and


Table3Selected

potentialreference

genes(19th

DLR

PFC)

Gene

symbo

lAc

cession

number

19th

DLR

PFC

potentialreference

genes

HW

1g18

g120583g

119875values

MeanFI

SDMeanFI

SDMeanFI

SDMeanFI

SDHW

versus

1g1g

versus

18g

18gversus120583g

1gversus120583g

ALB

NM

000477

213

204

4028

122

4062

088

4010

071

02627

07154

04724

08383

BC035969

BC034023

B4GALT

6NM

004775

4055

765

3930

762

4341

1138

3992

762

08506

06342

06847

09253

AF0

69054

BC074835

GAPD

HNM

002046

2179123

91423

2116

232

144799

1912

458

131950

2082321

200500

05652

01466

02971

08250

BC001601

BC00

9081

HMBS

NM

000190

218420

97377

161283

6774

999873

44339

122698

5315

904563

02695

05994

04833

NM

001024382

BC008149

RPLP

0

NM

001002

1135

234

172826

1163377

42619

1138089

206

050

1271495

83710

07397

08004

02346

004

25BC

001127

BC008594

BC00

0087

BC070194

TBP

NM

003194

415856

31025

32714

32172

622872

01610

9277682

17076

00192

000

4200226

00390

X54993

BC109053

YWHAZ

NM

003406

913631

207039

944522

199027

984330

212525

985306

218444

07808

07239

09934

07214

BC06

8456

BC108281

BC101483

BC083508

BC072426

BC003623



5000

10000

15000

20000

25000

30000 TEXUS-49 potential reference genes

Fluo

resc

ence

inte

nsity

lowast lowast

HWBL120583g

(a)

Fluo

resc

ence

inte

nsity

ALB B4GALT60

20

40

60

HWBL120583g

TEXUS-49 potential reference genes

(b)Figure 7 Influence of altered gravity during sounding rocket flight on potential reference genes GAPDHHMBS RPLP0 TBP and YWHAZ(a) ALB and B4GALT6 (b) RNA expression levels after launch and acceleration (BL dark gray) and 120583g (black) conditions of TEXUS-49Hardware ground controls (HW striped) are shown for each experimental group RNA levels are displayed as fluorescence intensitiesGAPDH HMBS RPLP0 YWHAZ ALB and B4GALT6 show no significant change in RNA levels upon altered gravity while TBP reactssensitively to all g conditions Mean values of at least three measurements with standard deviations are shown lowast119875 lt 005

supplementary Table 1) TBP RNA levels were reduced in 120583gsamples compared to the in-flight BL Interestingly compar-isons between the HW ground controls and BL revealed apostlaunch increase in RNA expression most likely inducedby the launch vibrations or hypergravity (Figure 7(a))

Only a very low number of tRNA related genes fulfilledour criterion of being represented by at least three probes(four out of 32) For three out of those four genes fluorescentintensity showed a great variance between the single valuesas mentioned above Only one tRNA synthetase (SARS)yielded reasonable resultsThe exposure of the cells to alteredgravity conditions during the parabolic flight resulted ina decreased SARS expression in 1 g in-flight control and18 g samples compared to the HW ground control and 1 gcontrol respectively (supplementary Table 3) Although notsignificant there is a visible increase of SARS mRNA upon120583g compared to 18 g arguing for an immediate expressionrecovery after termination of 18 g This is in line with theresults from TEXUS-49 flight campaign where SARS showsno significant expression change in in-flight baseline controlor in 120583g compared to HW ground controlThis could be dueto fast expression recovery of the gene during g alterationshyper-g phase and 120583g

The highly conserved ABC transporters were representedas a large group of genes on the applied microarray We ana-lyzed a total of 47 ABC transporters belonging to nine differ-ent sub-families (supplementary Table 2) 19 ABC transportergenes were represented by three or more individual probeson the microarray and 11 of them had similar fluorescentintensities meeting the requirements for a statistical analysisExemplarily five of those 11 ABC transporters are shown inFigures 8 and 9

During the short-term gravity alterations achieved byparabolic flights ABCC1 and ABCF2 displayed no significantdifferential expression between all g conditions analyzedTAP2 showed a significant reduction of RNA expression

ABCC1 ABCC4 ABCD4 ABCF2 TAP20

2000

4000

6000

8000

10000

Fluo

resc

ence

inte

nsity

19th DLR PFC ABC transporter genes

lowast lowast lowast

lowast

lowastlowast lowastlowast

HW1g

18 g120583g

Figure 8 Influence of altered gravity during parabolic flight onABC transporter genes ABCC1 ABCC4 ABCD4 ABCF2 andTAP2 RNA levels after 1 g (light gray) 18 g (dark gray) and 120583g(black) conditions during the 19th DLR parabolic flight campaignHardware ground controls (HW striped) are shown for eachexperimental group RNA expression levels are displayed as fluo-rescence intensities ABCC1 and ABCF2 show no significant changein RNA expression levels upon altered gravity while ABCC4 andTAP2 display 120583g sensitivity compared to 18 g and to 18 g and 1 grespectively ABCD4 reacts sensitively to 18 g compared to 1 g andABCD4 and TAP2 show vibration sensitivity comparing 1 g to HWMean values of at least threemeasurements with standard deviationsare shown lowast119875 lt 005 lowastlowast119875 lt 0005

comparing 120583g samples to 18 g samples while ABCC4 showedan increase ABCD4 revealed hyper-g sensitivity by reducingits RNA level during 18 g compared to 1 g And ABCD4 andTAP2 displayed reduced expression during preflight phasecompared to HW control (Figure 8 Table 5)

A prolonged exposure of the cells to 120583g (378 sec ver-sus 20 sec) during TEXUS-49 experiment led to significantreduction of mRNA levels of ABCC4 ABCD4 ABCF2 and


Table4Selected

potentialreference

genes(TE

XUS-49)

Gene

symbo

lAc

cessionnu

mber

TEXU

S-49potentia

lreference

genes

HW

BL120583g

119875values

MeanFI

SDMeanFI

SDMeanFI

SDHW

versus

BLBL

versus120583g

HW

versus120583g

ALB

NM

000477

1802

129

1747

118

1788

188

06184

07708

09215

BC035969

BC034023

B4GALT

6NM

004775

1715

426

1767

605

1595

326

09093

06933

07197

AF0

69054

BC074835

GAPD

HNM

002046

2338582

172178

2486207

22710

22545117

169290

04238

07383

02126

BC001601

BC00

9081

HMBS

NM

000190

29214

010

8402

30874

910

9422

256823

8996

508609

05612

06872

NM

001024382

BC008149

RPLP

0

NM

001002

802710

123345

903611

153820

740641

262246

02871

02728

06499

BC001127

BC008594

BC00

0087

BC070194

TBP

NM

003194

373825

14390

446

601

25759

35176

95387

00215

002

0101037

X54993

BC109053

YWHAZ

NM

003406

397064

149810

573874

211820

377601

159671

00993

00758

08180

BC06

8456

BC108281

BC101483

BC083508

BC072426

BC003623


Table5Selected

ABC

transporters(19

thDLR

PFC)

No

Gene

symbo

lAc

cession

Num

ber

19th

DLR

PFC

ABC

transporterg

enes

HW

1g18

g120583g

119875values

MeanFI

SDMeanFI

SDMeanFI

SDMeanFI

SDHW

versus

1g1g

versus

18g

18gversus120583g

1gversus120583g

23ABC

C1AB2

09120

314046

7517

133092

974399

225845

32589

293479

49693

07959

0119

301312

05141

NM

019900

NM

004996

29ABC

C4BC

041560

149737

1997

514

3784

2794

58570

53065

106701

9126

07805

00676

004

5901380

AY133678

NM

005845

37ABC

D4

NM

005050

171937

16462

125413

24026

76594

23575

115589

25753

00222

00273

00673

05972

BC012815

NM

020326

NM

020325

40ABC

F2NM

005692

751096

66499

677422

4078

5602209

52591

652551

51517

01916

01263

03019

05496

BC001661

AF0

91073

47TA

P2NM

018833

336561

13624

261483

9454

248522

19087

20278

47461

000

2303711

00394

000

14BC

002751

AF105151


Table6Selected

ABC

transporters(TEX

US-49)

No

Gene

symbo

lAc

cession

number

TEXU

S-49A

BCtransporterg

enes

HW

BL120583g

119875values

MeanFI

SDMeanFI

SDMeanFI

SDHW

versus

BLBL

versus120583g

HW

versus120583g

23ABC

C1AB2

09120

240

502

68683

309607

63460

202076

39849

02701

00797

04598

NM

019900

NM

004996

29ABC

C4BC

041560

76534

29367

138237

29464

5378

714092

00620

002

2803164

AY133678

NM

005845

37ABC

D4

NM

005050

7599

219032

84389

9439

56511

13898

04697

00194

01539

BC012815

NM

020326

NM

020325

40ABC

F2NM

005692

838241

75110

796343

7214

6637322

57328

05244

004

3000238

BC001661

AF0

91073

47TA

P2NM

018833

490284

1208

400

089

3526

343097

4927

000

01000

02000

02BC

002751

AF105151



2000

4000

6000

8000

10000TEXUS-49 ABC transporter genes

Fluo

resc

ence

inte

nsity

lowast

lowast lowast

lowast

lowastlowastlowast

lowastlowastlowast

lowastlowastlowast

HWBL120583g

Figure 9 Influence of altered gravity during sounding rocket flighton ABC transporter genes ABCC1 ABCC4 ABCD4 ABCF2 andTAP2 RNA expression levels after launch and acceleration (BL darkgray) and 120583g (black) conditions of TEXUS-49 Hardware groundcontrols (HW striped) are shown for each experimental groupRNA levels are depicted as fluorescence intensities Only ABCC1expression is stable over all g conditions ABCC4 ABCD4 ABCF2and TAP2 display 120583g sensitivity compared to BL and to HW inthe case of ABCF2 and TAP2 TAP2 also shows vibration sensitivitycomparing BL to HW Mean values of at least three measurementswith standard deviations are shown lowast119875 lt 005 lowastlowast119875 lt 0005 andlowastlowastlowast119875 lt 00005

TAP2 in 120583g compared to in-flight BL Furthermore TAP2expression already decreases in the first phase after launch(BL versus HW) while the other ABC transportersrsquo mRNAlevels appeared stable (Figure 9 Table 6)

Taken together in this study we identified eight genesas nonchanging reference genes suitable for studies underaltered gravity conditions and nine genes as candidates forg-sensitivity and 83 genes could not be assigned to eithergroup due to low probe number on the microarray or to greatvariance between the probe values (Table 7)

4 Discussion

Microarray expression data are intensely used to analyzedifferential gene expression in cells tissues and organismsthat are exposed to various conditions [29 30] Even inthe field of gravitational biology gene expression analysesare utilized with increasing frequency Recently an articlewas released giving an overview of all published microarraybased microgravity studies [57] describing the difficulties tocombine and overlay the data from different experimentsstudy objects microgravity platforms (simulated micro-gravity sounding rocket space shuttle and ISS missions)and different microarray experimental designs [58ndash64]The different analyses were done mostly in simulated micro-gravity and investigated various organisms and cell types likeArabidopsis thaliana Salmonella enterica and rat and mousetissues as well as human osteoblasts and T-cells [58ndash61 65ndash70] (for the complete list see [57]) The goal was to screen thevast amount of data to identify a list of major ldquospace genesrdquo

that are sensitive to microgravity throughout all involvedplatforms The data inspection revealed a huge number ofdifferentially expressed genes butwith only little or no overlapbetween closely related studies on the level of single genesIn contrast on the level of pathway analysis it was possibleto define major pathways like ECM-receptor interactionfocal adhesion TGF-beta signaling and glycolysis beingaffected in many species (human mouse rat and Xenopusin different combinations) by the exposure to microgravity[57] Moreover major ldquospace genesrdquo sensitive to microgravitywere defined if they were found to be differentially expressedin at least four of the examined studies The results showedin total eight potential space genes (CD44 MARCKS FN1TUBA1 CTGF CYR61 MT2 and MT1) which are involvedin T-cell development cell motility extracellularmatrix com-ponents cytoskeleton and oxidative stress protection [57]The study describes in detail the difficulties of combininggene expression data from different groups due to varyingexperimental setups and conditions It elucidates that it is ofhigh relevance to be able to standardize gene expression datathat arose from RT-qPCR or microarray studies A key com-ponent for standardization within a single experiment andbetween different experiments is normalization An impo-rtant factor for normalization is the use of stable referencegenes There are numerous studies describing that commo-nly used reference genes could represent a pitfall becausethey are often differentially expressed under specific experi-mental conditions and that they have to be considered carefu-lly before the experiment [46ndash49] Different guidelineshave been published to facilitate standardized experimentaldesign and increase comparability between analyses (MIQEMIAME) [34 39 40] It is for example highly recommendedto perform a pilot study with programs like geNorm or Best-Keeper prior to the experiment to identify several stable refe-rence genes that can be used simultaneously as controls fornormalization in the differential gene expression analysis[50 51]

Alternatively tomicroarrays a novel technique RNA-Seqis under development for whole genome expression analysesIt is reported that thismethod has advantages in detecting lowabundance transcripts genetic variants and splice isoformsof genes as well as distinguishing biologically critical isoforms[71] Despite the described technical advantages of RNA-Seq microarrays remain popular for some reasons Themicroarray platforms have a proven track record spanningnearly two decades in the lab The arrays are generallyconsidered easier to use with less complicated and less labor-intensive sample preparation than RNA-Seq The same holdstrue for the data storage and data analysis Moreover despitethe rapid drop in the cost associated with next-generationsequencing (NGS) arrays are still more economical andyield higher throughput providing significant advantageswhen working with a large number of samples Thereforemicroarray analyses are still more commonly used for tran-scriptional profiling experiments [71]

Taking into account that many other studies throughoutthe last few years have reported a considerable portion of thetraditionally used reference genes not being stably expressedunder various experimental conditions it becomes rather


Table 7 Overview of g-stable (+) and g-sensitive genes (minus)

Gene symbol Accession number 19th DLR PFC TEXUS-49(HW

versus 1 g)(1 g versus

18 g)(18 g

versus 120583g)(1 g versus120583g)

(HWversus BL)

(BL versus120583g)

(HWversus 120583g)

Potential reference genes

ALBNM 000477

+ + + + + + +BC035969BC034023

B4GALT6NM 004775

+ + + + + + +AF069054BC074835

GAPDHNM 002046

+ + + + + + +BC001601BC009081

HMBSNM 000190

+ + + + + + +NM 001024382BC008149

RPLP0

NM 001002

+ + + minus + + +BC001127BC008594BC000087BC070194

TBPNM 003194

minus minus minus minus minus minus +X54993BC109053

YWHAZ

NM 003406

+ + + + + + +

BC068456BC108281BC101483BC083508BC072426BC003623

ABC transporter genes

ABCA5NM 018672

+ + + + + + +AJ275973AY028897

ABCA9NM 080283

+ + + + + + +BC062472NM 172386

ABCC1AB209120

+ + + + + + +NM 019900NM 004996

ABCC4BC041560

+ + minus + + minus +AY133678NM 005845

ABCC12

AK127951

minus + + + minus minus +NM 145187NM 033226BC036378

ABCD4

NM 005050

minus minus + + + minus +BC012815NM 020326NM 020325

ABCF2NM 005692

+ + + + + minus minusBC001661AF091073

TAP2NM 018833

minus + minus minus minus minus minusBC002751AF105151


apparent that a natural constant as 1 g might have evenmore an effect on the expression of genes than other testcircumstances Therefore we focused in this study on theinvestigation of the expression qualities of several potentialreference genes under 1 g compared to altered gravity con-ditions generated by two widely used platforms parabolicflights and sounding rockets These two platforms are ofspecial interest because of the rather easy access comparedto the extremely limited accessibilities of long-term micro-gravity experiments on satellites and the ISS We present amicroarray based analysis identifying stable reference genesin cells of the immune system exposed to short-term (severalseconds) and middle-term (several minutes) altered gravityconditions on the twowidely used platforms parabolic flightsand sounding rockets

Our analyses of commonly used reference genes ABCtransporters and tRNA related genes revealed that nineof the 17 genes suspected to be ubiquitously expressed areg-sensitive and therefore inappropriate for our purposesamongst them being TATA box binding protein (TBP) afundamental transcription factor for many genes and seryl-tRNA synthetase (SARS) an essential enzyme for mRNAtranslation also regulating vascular development (Table 7)

Two of the g-sensitive genes that we identified in thisstudy are involved in multidrug resistance processes likethe ABC transporters ABCC4 and transporter associatedwith antigen presenting 2 (TAP2) ABCC4 is of particularinterest because it has the ability to provide resistance toantiviral and anticancer nucleotide analogs andmethotrexate[72 73] acts as an independent regulator of intracellularcAMP mediates cAMP dependent signal transduction tothe nucleus and controls human and rat smooth musclecell (SMC) proliferation [74] It is known that cAMP haslargely inhibitory effects on components of macrophageactivation and elevation of cAMP levels which suppressesFcgammaR-mediated phagocytosis [75] Therefore it wouldbe interesting to look at this multidrug resistance-associatedprotein (ABCC4) in more detail in microgravity exposedcells to elucidate its role in the signaling cascades importantfor immune cell action and reaction under space conditionsas ABCC4 proved to be 120583g-sensitive during parabolic andsounding rocket flight (Figures 8 and 9) Furthermore TAP2seems to be evenmore g-sensitive because it shows significantdifferential gene expression under 120583g and hypergravity con-ditions during parabolic and sounding rocket flight (Figures8 and 9) It will be interesting to further analyze the potentialeffects of differential gene expression of TAP2 because it isa key player in endogenous pathways for antigen presenta-tion and involved in the cellular transport of antigens forsubsequent association with MHC class I molecules [76] Animbalance in its gene expression could lead to an impairedreactivity of cells of the immune system under altered gravityconditions

Further standard genes as well as ABC transporterslike RPLP0 ABCD4 and ABCF2 also turned out in ouranalysis to be g-sensitive RPLP0 encodes for a ribosomalprotein that is a component of the 60S subunit and interactswith P1 and P2 to form pentameric complexes [77] It isinvolved for example in Chagas disease [78] as well as mixed

connective tissue disease [79]TheABC transporters ABCD4and ABCF2 are involved in transport of molecules acrossextra- and intracellular membranes like in peroxisomalimport of fatty acids andor fatty acyl-CoAs in the organelle[80] and play a role in suppression of volume-sensitiveoutwardly rectifying Cl channel (VSOR) respectively [81]Altered expression levels of those genes by microgravity orhypergravity could have an impact on the translational levelor the supply of the cell with essential resources important forproper cellular function Recently it was shown that duringparabolic flights the activity of the MRP2-ABC-transporterwas significantly reduced [82] Furthermore under shortduration spaceflight missions certain ABC transporter genesin the medically relevant species Salmonella sp and Candidasp were upregulated [83 84]

Interestingly we identified many of the g-sensitive genesnot only reacting on 120583g but also on hypergravity indicatingthat not only the experimental g-conditions should be takeninto account when selecting an appropriate reference genebut also the accompanying g-conditions prevailing usuallybefore 120583g is achieved A detailed differential gene expressionanalysis of the parabolic flight and sounding rocket flight datasets for g-sensitive genes is currently ongoing

Genes that proved to be stable over all g-conditions testedwere

(i) albumin (ALB) a protein comprising about one halfof blood serum protein

(ii) UDP-GalbetaGlcNAc beta 14-galactosyltransferasepolypeptide 6 (B4GALT6) a type II membrane-bound glycoprotein important for glycolipid biosyn-thesis

(iii) glyceraldehyde-3-phosphate dehydrogenase(GAPDH) a protein with several distinct functionsfor example the reversible oxidative phosphorylationof glyceraldehyde-3-phosphate

(iv) hydroxymethylbilane synthase (HMBS) a proteincatalyzing the head to tail condensation of four por-phobilinogen molecules into the linear hydroxyme-thylbilane

(v) tyrosine 3-monooxygenasetryptophan 5-monooxy-genase activation protein zeta (YWHAZ) a gene pro-duct belonging to the 14-3-3 family of proteins thatinteracts with IRS1 suggesting a role in regulatinginsulin sensitivity

(vi) ATP-binding cassette subfamily A member 5(ABCA5) a membrane-associated protein belongingto the only major ABC subfamily found exclusivelyin multicellular eukaryotes with unknown function

(vii) ATP-binding cassette subfamily A member 9(ABCA9) another ABC1 family member inducedduring monocyte differentiation into macrophagesand

(viii) ATP-binding cassette subfamily C member 1(ABCC1) a member of the MRP subfamily of ABCtransporters involved in multidrug resistance and


functioning as a multispecific organic anion transpo-rter

Taken together the compilation of genes that we presentin Table 7 gives an overview about which genes are stablyexpressed during all investigated gravitational conditionslasting from seconds to minutes and can therefore be con-sidered as suitable reference genes Furthermore Table 7 canbe regarded as a tool for the community that can be easilyadapted to select potential control genes in the design phaseof a new immune cell based experiment on parabolic flightsand sounding rocket flights because it provides valuableinformation about gene expression levels in 120583g as well as in18 g in-flight 1 g and hardware ground control Our resultsalso allow for the identification of adaptation mechanisms bycomparing short (parabolic flight) and intermediate (sound-ing rocket) microgravity periods and spot those genes thatconvert from sensitive into stable and vice versa Our workshould considerably facilitate identification of appropriatereference genes for individual experiments performed duringparabolic flight and sounding rocket campaignswith immunecells especially of the monocytemacrophage system inaltered gravity

Abbreviations

BL BaselineCEV Centre drsquoessai en volCV Coefficient of variationESA European Space AgencyFBS Fetal bovine serumFDR False discovery rateDLR German Aerospace CenterHW Hardware ground controlsHKG Housekeeping geneshyper-g HypergravityIL InterleukinISS International Space StationLPS Lipopolysaccharide120583g MicrogravityMMS Monocyte-macrophage systemPFC Parabolic flight campaignqPCR Quantitative real-time PCRRIN RNA integrity numberRPM Random positioning machineROS Reactive oxygen speciesRWV Rotating wall vesselSD Standard deviationTNF-120572 Tumor necrosis factor-alpha



Authorsrsquo Contributions

Oliver Ullrich and Cora S Thiel developed the study ideaconcept and the overall study design in addition to planningcoordinating and supervising the study Cora SThiel Liliana

E Layer and Swantje Hauschild wrote the paper OliverUllrich edited the paper Beatrice Lauber contributed tothe paper Liliana E Layer Cora S Thiel Oliver UllrichSvantje Tauber Swantje Hauschild Claudia Philpot AnnettGutewort Eva Hurlimann Josefine Biskup andHartwin Lierperformed the experiments during the 19th DLR parabolicflight campaign Svantje Tauber Cora S Thiel Felix Unver-dorben and Oliver Ullrich performed the experiments dur-ing the TEXUS-49 mission Cora S Thiel was responsible forsample analysis from the 19th DLR parabolic flight campaignand TEXUS-49 mission Liliana E Layer contributed tothe sample analysis Frank Engelmann contributed to andsupervised the technical procedures during the 19th DLRparabolic flight campaign

Acknowledgments

The authors gratefully acknowledge financial support byGerman Aerospace Center DLR (grants nos 50WB0912 and50WB1219) They also gratefully acknowledge the support of(in alphabetic order) Gesine Bradacs Markus Braun MiriamChristen Giovanni Colacicco Ulrike Friedrich NadineGolzAndre Hilliger AndreasHuge Schirin IbrahimOtfried JoopSonja Krammer Andre Melik Shirin Milani Brice MouttetMarianne Ott Irina Rau Frank Ruhli Chen Sang BurkhardSchmitz Brita Scholte Andreas Schutte Johanna StahnMarcStuder Susanne Wolf and Fengyuan Zhuang

References

[1] B Comet ldquoLimiting factors for human health and performancemicrogravity and reduced gravity In study on the survivabilityand adaptation of humans to long-duration interplanetary andplanetary environmentsrdquo Technical Note 2 Critical Assess-ments of the Limiting Factors for Human Health and Perfor-mance and Recommendation of Countermeasures HUMEX-TN-002 2001

[2] G Horneck R Facius M Reichert et al ldquoHUMEX a studyon the survivability and adaptation of humans to long-durationexploratory missions part II missions to Marsrdquo Advances inSpace Research vol 38 no 4 pp 752ndash759 2006

[3] G Horneck and B Comet ldquoGeneral human health issues forMoon and Mars missions results from the HUMEX studyrdquoAdvances in Space Research vol 37 no 1 pp 100ndash108 2006


[5] H Takayanagi ldquoOsteoimmunology shared mechanisms andcrosstalk between the immune and bone systemsrdquo NatureReviews Immunology vol 7 no 4 pp 292ndash304 2007

[6] J Caetano-Lopes J E Canhao and H Fonseca ldquoOsteoimmu-nologymdashthe hidden immune regulation of bonerdquo Autoimmu-nity Reviews vol 8 no 3 pp 250ndash255 2009

[7] A Cogoli B Bechler O Muller and E Hunzinger ldquoEffect ofmicrogravity on lymphocyte activationrdquo in Biorack on SpacelabD1 pp 89ndash100 European Space Agency Paris France 1988

[8] A Cogoli and A Tschopp ldquoLymphocyte reactivity during spa-ceflightrdquo Immunology Today vol 6 no 1 pp 1ndash4 1985


[9] A Cogoli A Tschopp and P Fuchs-Bislin ldquoCell sensitivity togravityrdquo Science vol 225 no 4658 pp 228ndash230 1984

[10] A Cogoli ldquoGravitational physiology of human immune cellsa review of in vivo ex vivo and in vitro studiesrdquo Journal ofGravitational Physiology vol 3 no 1 pp 1ndash9 1996


[12] J W Armstrong R A Gerren and S K Chapes ldquoThe effectof space and parabolic flight onmacrophage hematopoiesis andfunctionrdquo Experimental Cell Research vol 216 no 1 pp 160ndash168 1995

[13] M Limouse S Manie I Konstantinova B Ferrua and LSchaffar ldquoInhibition of phorbol ester-induced cell activation inmicrogravityrdquo Experimental Cell Research vol 197 no 1 pp 82ndash86 1991

[14] D A Schmitt J P Hatton C Emond et al ldquoThe distribution ofprotein kinase C in human leukocytes is altered in micrograv-ityrdquo FASEB Journal vol 10 no 14 pp 1627ndash1634 1996

[15] C-L Hsieh P-D L Chao and S-H Fang ldquoMorin sulphatesglucuronides enhance macrophage function in microgravityculture systemrdquo European Journal of Clinical Investigation vol35 no 9 pp 591ndash596 2005

[16] I Kaur E R Simons V A Castro C M Ott and D L PiersonldquoChanges inmonocyte functions of astronautsrdquoBrain Behaviorand Immunity vol 19 no 6 pp 547ndash554 2005

[17] A Adrian K Schoppmann J Sromicki et al ldquoThe oxidativeburst reaction in mammalian cells depends on gravityrdquo CellCommunication and Signaling vol 11 no 1 article 98 2013

[18] M A Meloni G Galleri P Pippia and M Cogoli-GreuterldquoCytoskeleton changes and impaired motility of monocytes atmodelled low gravityrdquo Protoplasma vol 229 no 2ndash4 pp 243ndash249 2006

[19] MAMeloniGGalleri G Pani A Saba P Pippia andMCog-oli-Greuter ldquoSpace flight affects motility and cytoskeletal struc-tures in human monocyte cell line J-111rdquo Cytoskeleton vol 68no 2 pp 125ndash137 2011

[20] M Hughes-Fulford T Chang and C-F Li ldquoEffect of gravity onmonocyte differentiationrdquo in Proceedings of the 10th ESA LifeSciences Symposium29th Annual ISGP Meeting24th AnnualASGSB MeetingELGRA Symposium ldquoLife in Space for Life onEarthrdquo pp 22ndash27 2008

[21] C Porcher M-C Malinge C Picat and B Grandchamp ldquoAsimplified method for determination of specific DNA or RNAcopy number using quantitative PCR and an automatic DNAsequencerrdquo BioTechniques vol 13 no 1 pp 106ndash114 1992

[22] R Higuchi C Fockler G Dollinger and R Watson ldquoKineticPCR analysis real-time monitoring of DNA amplificationreactionsrdquo Nature Biotechnology vol 11 no 9 pp 1026ndash10301993

[23] P-W Chiang W-J Song K-Y Wu et al ldquoUse of a fluorescent-PCR reaction to detect genomic sequence copy number andtranscriptional abundancerdquo Genome Research vol 6 no 10 pp1013ndash1026 1996

[24] U E Gibson C A Heid and P M Williams ldquoA novel methodfor real time quantitative RT-PCRrdquoGenome Research vol 6 no10 pp 995ndash1001 1996

[25] C A Heid J Stevens K J Livak and PMWilliams ldquoReal timequantitative PCRrdquoGenome Research vol 6 no 10 pp 986ndash9941996

[26] H D VanGuilder K E Vrana and W M Freeman ldquoTwenty-five years of quantitative PCR for gene expression analysisrdquo Bio-Techniques vol 44 no 5 pp 619ndash626 2008

[27] R Biassoni and A Raso Eds Quantitative Real-Time PCRMethods and Protocols Humana Press New York NY USA2014

[28] W M Freeman D J Robertson and K E Vrana ldquoFundamen-tals of DNA hybridization arrays for gene expression analysisrdquoBioTechniques vol 29 no 5 pp 1042ndash1055 2000

[29] S Draghici Data Analysis for DNA Microarrays Chapman ampHallCRC Boca Raton Fla USA 2003

[30] H C Causton J Quackenbush and A Brazma MicroarrayGene Expression Data Analysis A Beginnerrsquos Guide BlackwellPublishing Malden Mass USA 2003

[31] ldquoThe chipping forecastrdquo Nature Genetics vol 21 pp 1ndash60 1999[32] P O Brown and D Botstein ldquoExploring the new world of the

genome with DNA microarraysrdquo Nature Genetics vol 21 no 1pp 33ndash37 1999

[33] D J Lockhart and E A Winzeler ldquoGenomics gene expressionand DNA arraysrdquoNature vol 405 no 6788 pp 827ndash836 2000

[34] A Brazma PHingamp J Quackenbush et al ldquoMinimum infor-mation about a microarray experiment (MIAME)mdashtowardstandards for microarray datardquo Nature Genetics vol 29 no 4pp 365ndash371 2001

[35] R D Canales Y Luo J C Willey et al ldquoEvaluation of DNAmicroarray results with quantitative gene expression platformsrdquoNature Biotechnology vol 24 no 9 pp 1115ndash1122 2006

[36] S Lefever J Vandesompele F Speleman and F Pattyn ldquoRTPri-merDB the portal for real-time PCR primers and probesrdquoNucleic Acids Research vol 37 no 1 pp D942ndashD945 2009

[37] F Pattyn P Robbrecht A de Paepe F Speleman and J Vande-sompele ldquoRTPrimerDB the real-time PCR primer and probedatabase major update 2006rdquo Nucleic Acids Research vol 34supplement 1 pp D684ndashD688 2006

[38] X Wang and B Seed ldquoA PCR primer bank for quantitativegene expression analysisrdquoNucleic Acids Research vol 31 no 24article e154 2003

[39] S A Bustin V Benes J A Garson et al ldquoTheMIQE guidelinesminimum information for publication of quantitative real-timePCRexperimentsrdquoClinical Chemistry vol 55 no 4 pp 611ndash6222009

[40] S Taylor MWakem G Dijkman M Alsarraj andM NguyenldquoA practical approach to RT-qPCR-Publishing data that con-form to theMIQE guidelinesrdquoMethods vol 50 no 4 pp S1ndashS52010

[41] V Marx ldquoPCR living life amplified and standardizedrdquo NatureMethods vol 10 no 5 pp 391ndash395 2013

[42] S A Bustin ldquoAbsolute quantification of mrna using real-timereverse transcription polymerase chain reaction assaysrdquo Journalof Molecular Endocrinology vol 25 no 2 pp 169ndash193 2000

[43] T Suzuki P J Higgins and D R Crawford ldquoControl selectionfor RNAquantitationrdquoBioTechniques vol 29 no 2 pp 332ndash3372000

[44] O Thellin W Zorzi B Lakaye et al ldquoHousekeeping genes asinternal standards use and limitsrdquo Journal of Biotechnology vol75 no 2-3 pp 291ndash295 1999

[45] N Tanic M Perovic A Mladenovic S Ruzdijic and SKanazir ldquoEffects of aging dietary restriction and glucocorticoidtreatment on housekeeping gene expression in rat cortex andhippocampusmdashevaluation by real time RT-PCRrdquo Journal ofMolecular Neuroscience vol 32 no 1 pp 38ndash46 2007


[46] T D Schmittgen and B A Zakrajsek ldquoEffect of experimentaltreatment on housekeeping gene expression validation byreal-time quantitative RT-PCRrdquo Journal of Biochemical andBiophysical Methods vol 46 no 1-2 pp 69ndash81 2000

[47] J A Warrington A Nair M Mahadevappa and M Tsygan-skaya ldquoComparison of human adult and fetal expression andidentification of 535 housekeepingmaintenance genesrdquo PhysiolGenomics vol 2 no 3 pp 143ndash147 2000

[48] K Dheda J F Huggett S A Bustin M A Johnson G Rookand A Zumla ldquoValidation of housekeeping genes for norma-lizing RNA expression in real-time PCRrdquo BioTechniques vol 37no 1 pp 112ndash119 2004

[49] A Radonic SThulke IMMackayO LandtW Siegert andANitsche ldquoGuideline to reference gene selection for quantitativereal-time PCRrdquo Biochemical and Biophysical Research Commu-nications vol 313 no 4 pp 856ndash862 2004

[50] M W Pfaffl A Tichopad C Prgomet and T P NeuviansldquoDetermination of stable housekeeping genes differentiallyregulated target genes and sample integrity bestKeepermdashexcel-based tool using pair-wise correlationsrdquo Biotechnology Lettersvol 26 no 6 pp 509ndash515 2004

[51] J Vandesompele K De Preter F Pattyn et al ldquoAccurate nor-malization of real-time quantitative RT-PCR data by geometricaveraging of multiple internal control genesrdquo Genome Biologyvol 3 no 7 Article ID RESEARCH0034 2002

[52] J H Cai S Deng SW Kumpf et al ldquoValidation of rat referencegenes for improved quantitative gene expression analysis usinglow density arraysrdquo BioTechniques vol 42 no 4 pp 503ndash5122007

[53] RA Irizarry BHobbs F Collin et al ldquoExploration normaliza-tion and summaries of high density oligonucleotide array probelevel datardquo Biostatistics vol 4 no 2 pp 249ndash264 2003

[54] Y Benjamini and Y Hochberg ldquoControlling the false discoveryrate a practical and powerful approach to multiple testingrdquoJournal of the Royal Statistical Society Series B Methodologicalvol 57 no 1 pp 289ndash300 1995

[55] N Silver S Best J Jiang and S L Thein ldquoSelection of house-keeping genes for gene expression studies in human reticu-locytes using real-time PCRrdquo BMC Molecular Biology vol 7article 33 2006

[56] J Hellemans G Mortier A de Paepe F Speleman and JVandesompele ldquoqBase relative quantification framework andsoftware for management and automated analysis of real-timequantitative PCR datardquo Genome biology vol 8 no 2 p R192007

[57] J Q Clement ldquoGene expression microarrays in microgravityresearch toward the identification of major space genesrdquo inInnovations in Biotechnology E C Agbo Ed pp 319ndash348InTech 2012

[58] A-I Kittang J J van Loon O Vorst R D Hall K Fossumand T-H Iversen ldquoGround based studies of gene expressionin Arabidopsis exposed to gravity stressesrdquo Journal of Gravita-tional Physiology vol 11 no 2 pp P223ndashP224 2004

[59] M Martzivanou M Babbick M Cogoli-Greuter and RHampp ldquoMicrogravity-related changes in gene expression aftershort-term exposure of Arabidopsis thaliana cell culturesrdquoProtoplasma vol 229 no 2ndash4 pp 155ndash162 2006

[60] V Chopra A A Fadl J Sha S Chopra C L Galindo andA K Chopra ldquoAlterations in the virulence potential of ente-ric pathogens and bacterial-host cell interactions under sim-ulated microgravity conditionsrdquo Journal of Toxicology and

Environmental Health Part A Current Issues vol 69 no 14 pp1345ndash1370 2006

[61] S Yamada T Ganno N Ohara and Y Hayashi ldquoChitosanmonomer accelerates alkaline phosphatase activity on humanosteoblastic cells under hypofunctional conditionsrdquo Journal ofBiomedical Materials Research Part A vol 83 no 2 pp 290ndash295 2007

[62] M L Lewis L A Cubano B Zhao et al ldquocDNA microarrayreveals altered cytoskeletal gene expression in space-flownleukemic T lymphocytes (Jurkat)rdquo The FASEB Journal vol 15no 10 pp 1783ndash1785 2001

[63] M A Meloni G Galleri S Carta et al ldquoPreliminary studyof gene expression levels in human T-cells exposed to cosmicradiationsrdquo Journal of Gravitational Physiology vol 9 no 1 ppP291ndashP292 2002

[64] S J PardoM J PatelMC Sykes et al ldquoSimulatedmicrogravityusing the Random Positioning Machine inhibits differentiationand alters gene expression profiles of 2T3 preosteoblastsrdquo TheAmerican Journal of PhysiologymdashCell Physiology vol 288 no 6pp C1211ndashC1221 2005

[65] J W Wilson R Ramamurthy S Porwollik et al ldquoMicroarrayanalysis identifies Salmonella genes belonging to the low-shearmodeled microgravity regulonrdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 99 no21 pp 13807ndash13812 2002

[66] M Wittwer M Fluck H Hoppeler S Muller D Desplanchesand R Billeter ldquoProlonged unloading of rat soleus musclecauses distinct adaptations of the gene profilerdquoThe FASEB Jou-rnal vol 16 no 8 pp 884ndash886 2002


[68] K M Fridley I Fernandez M-T A Li R B Kettlewell and KRoy ldquoUnique differentiation profile of mouse embryonic stemcells in rotary and stirred tank bioreactorsrdquo Tissue EngineeringPart A vol 16 no 11 pp 3285ndash3298 2010

[69] A Qian S Di X Gao et al ldquocDNA microarray reveals thealterations of cytoskeleton-related genes in osteoblast underhigh magneto-gravitational environmentrdquo Acta Biochimica etBiophysica Sinica vol 41 no 7 pp 561ndash577 2009

[70] N E Ward N R Pellis S A Risin and D Risin ldquoGene expre-ssion alterations in activated humanT-cells induced bymodeledmicrogravityrdquo Journal of Cellular Biochemistry vol 99 no 4 pp1187ndash1202 2006

[71] S Zhao W-P Fung-Leung A Bittner K Ngo and X LiuldquoComparison of RNA-Seq and microarray in transcriptomeprofiling of activated T cellsrdquo PLoS ONE vol 9 no 1 ArticleID e78644 2014

[72] Z-S Chen K Lee and G D Kruh ldquoTransport of cyclicnucleotides and estradiol 17-120573-D-glucuronide by multidrugresistance protein 4 Resistance to 6-mercaptopurine and 6-thioguaninerdquo The Journal of Biological Chemistry vol 276 no36 pp 33747ndash33754 2001

[73] Z-S Chen K Lee S Walther et al ldquoAnalysis of methotrex-ate and folate transport by multidrug resistance protein 4(ABCC4) MRP4 is a component of the methotrexate effluxsystemrdquo Cancer Research vol 62 no 11 pp 3144ndash3150 2002

[74] Y Sassi L Lipskaia G Vandecasteele et al ldquoMultidrug resi-stance-associated protein 4 regulates cAMP-dependent signal-ing pathways and controls human and rat SMC proliferationrdquo


The Journal of Clinical Investigation vol 118 no 8 pp 2747ndash2757 2008

[75] D M Aronoff C Canetti C H Serezani M Luo and MPeters-Golden ldquoCutting edge macrophage inhibition by cyclicAMP (cAMP) differential roles of protein kinase A andexchange protein directly activated by cAMP-1rdquoTheThe Journalof Immunology vol 174 no 2 pp 595ndash599 2005

[76] E Procko and R Gaudet ldquoAntigen processing and presentationTAPping into ABC transportersrdquo Current Opinion in Immunol-ogy vol 21 no 1 pp 84ndash91 2009

[77] K-M Lee C W Yu D S Chan et al ldquoSolution structure of thedimerization domain of ribosomal protein P2 provides insightsfor the structural organization of eukaryotic stalkrdquoNucleic AcidsResearch vol 38 no 15 pp 5206ndash5216 2010

[78] I Ferrari M J Levin G Wallukat et al ldquoMolecular mimicrybetween the immunodominant ribosomal protein P0 of Try-panosoma cruzi and a functional epitope on the human 1205731-adrenergic receptorrdquo Journal of Experimental Medicine vol 182no 1 pp 59ndash65 1995

[79] R W Hoffman and M E Maldonado ldquoImmune pathogenesisof Mixed Connective Tissue Disease a short analytical reviewrdquoClinical Immunology vol 128 no 1 pp 8ndash17 2008

[80] S Kemp and R J A Wanders ldquoX-linked adrenoleukodys-trophy very long-chain fatty acid metabolism ABC half-transporters and the complicated route to treatmentrdquoMolecularGenetics and Metabolism vol 90 no 3 pp 268ndash276 2007

[81] Y Ando-Akatsuka T Shimizu T Numata and Y OkadaldquoInvolvements of the ABC protein ABCF2 and 120572-actinin-4 inregulation of cell volume and anion channels in human epithe-lial cellsrdquo Journal of Cellular Physiology vol 227 no 10 pp3498ndash3510 2012

[82] S Vaquer E Cuyas A Rabadan A Gonzalez F Fenollosa andR de la Torre ldquoActive transmembrane drug transport in micro-gravity a validation study using an ABC transporter modelrdquoF1000Research vol 3 article 201 2014

[83] JWWilson CMOtt KHoner Zu Bentrup et al ldquoSpace flightalters bacterial gene expression and virulence and reveals a rolefor global regulator Hfqrdquo Proceedings of the National Academyof Sciences of the United States of America vol 104 no 41 pp16299ndash16304 2007

[84] A Crabbe S M Nielsen-Preiss C M Woolley et al ldquoSpace-flight enhances cell aggregation and random budding in Can-dida albicansrdquo PLoS ONE vol 8 no 12 Article ID e80677 2013

[85] C K Mantri J P Dash J V Mantri and C C V DashldquoCocaine Enhances HIV-1 Replication in CD4+ T Cells byDown-Regulating MiR-125brdquo PLoS ONE vol 7 no 12 ArticleID e51387 2012

[86] F Jacob R Guertler S Naim et al ldquoCareful selection ofreference genes is required for reliable performance of RT-qPCR in human normal and cancer cell linesrdquo PLoS ONE vol8 no 3 Article ID e59180 2013

[87] A Marcant A Denys A Melchior et al ldquoCyclophilin B attenu-ates the expression of TNF-120572 in lipopolysaccharide-stimulatedmacrophages through the induction of B cell lymphoma-3rdquoTheJournal of Immunology vol 189 no 4 pp 2023ndash2032 2012

[88] B P Barna I Huizar A Malur et al ldquoCarbon nanotube-induced pulmonary granulomatous disease twist1 and alveolarmacrophage M1 activationrdquo International Journal of MolecularSciences vol 14 no 12 pp 23858ndash23871 2013

[89] J P Chou CM Ramirez J EWu andR B Effros ldquoAcceleratedaging in HIVAIDS novel Biomarkers of Senescent HumanCD8+ T Cellsrdquo PLoS ONE vol 8 no 5 Article ID e64702 2013

Research ArticleA Whole-Genome Microarray Study of Arabidopsis thalianaSemisolid Callus Cultures Exposed to Microgravity andNonmicrogravity Related Spaceflight Conditions for 5 Dayson Board of Shenzhou 8

Svenja Fengler1 Ina Spirer1 Maren Neef1 Margret Ecke1

Kay Nieselt2 and Ruumldiger Hampp1

1Physiological Ecology of Plants University of Tubingen Auf der Morgenstelle 1 72076 Tubingen Germany2Center for Bioinformatics University of Tubingen Sand 14 72076 Tubingen Germany

Correspondence should be addressed to Svenja Fengler svenjafengleruni-tuebingende

Received 8 May 2014 Revised 26 August 2014 Accepted 9 September 2014


Copyright copy 2015 Svenja Fengler et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

The Simbox mission was the first joint space project between Germany and China in November 2011 Eleven-day-old Arabidopsisthaliana wild type semisolid callus cultures were integrated into fully automated plant cultivation containers and exposed tospaceflight conditions within the Simbox hardware on board of the spacecraft Shenzhou 8 The related ground experiment wasconducted under similar conditions The use of an in-flight centrifuge provided a 1 g gravitational field in space The cells weremetabolically quenched after 5 days via RNAlater injectionThe impact on theArabidopsis transcriptomewas investigated bymeansof whole-genome gene expression analysis The results show a major impact of nonmicrogravity related spaceflight conditionsGenes that were significantly altered in transcript abundance are mainly involved in protein phosphorylation and MAPK cascade-related signaling processes as well as in the cellular defense and stress responses In contrast to short-term effects of microgravity(seconds minutes) this mission identified only minor changes after 5 days of microgravity These concerned genes coding forproteins involved in the plastid-associated translation machinery mitochondrial electron transport and energy production

1 Introduction

Gravitation biology is a field of research which hasmade con-siderable progress within the last years involving prokary-otes fungi plants and animals Plants are especially interest-ing because as sessile organisms they possess high versatilityin responding to environmental challenges and abiotic aswell as biotic ones In order to investigate responses toaltered gravitation a large range of methods is availablethat allows for modification of the Earthrsquos gravitational fieldThese involve centrifugation (hypergravity) clinorotationmagnetic levitation and random positioning (simulatedmicrogravity) or parabolic flights of planes and sound-ing rockets as well as satellites and spacecrafts (delivermicrogravity) Experiments with plants show that not onlytissues and organelles [1 2] but also single-cell systems

like characean rhizoids [3ndash7] as well as spores (Ceratopterisrichardii [8 9]) and protoplasts [10ndash12] or homogeneous cellcultures (Arabidopsis thaliana) exhibit gravisensitivity [2 13ndash16] Experimental approaches that analyze the response toaltered gravitation such as transcriptomics proteomics andmetabolomics dominate recently First molecular approacheswere using transcriptomics that is the search for genes whichchange their expression under altered gravitation In plantslike in other organisms the improvement of gene expressionquantification technologies together with growing databasessupports this development considerably To date databasesare available that exhibit plant datasets representing theirresponse to diverse experimental stimuli [17ndash20] They showthat external signals are translated into biochemical onesresulting in molecular signaling cascades which eventuallyresult in a life-sustaining adaptation process



For Arabidopsis (Arabidopsis thaliana) cell suspensioncultures the response to short-term microgravity was inves-tigated intensely in our group by means of parabolic flights[21] A combination of transcriptomics with phosphopro-teomics showed that changes in gene expression and proteinmodification occur within seconds The investigation ofeffects caused by longer-lasting microgravity depends onmuch scarcer availability of respective flight opportunitiesHowever data on cellular andmolecular long-term responsesof plants such as Brassicaceae (Arabidopsis) Fabaceae andPoaceae has recently been published [2 15 22ndash31] Withregard to long-term experiments on gene expression thereare conflicting reports Stutte et al [30] for example couldnot observe differentially expressed genes (DEGs) above a 2-fold cut-off in 24-day-old wheat leaves after a 21-day-spacemission In contrast Paul et al [15 24] detected many DEGsin nearly 20-day-old Arabidopsis callus cultures and 18-day-old seedlings after a nearly 13-day-space mission Further-more the set of altered genes detected in whole seedlingswas different from that in callus cultures [15] Thereby thespaceflight-mediated upregulated expression of heat shockproteins appeared to be an age-independent cell culturespecific response [15 16] Within the so-called TROPI-2experiment only 24 genes were altered in their abundance inArabidopsis seedlings [2] due to possible microgravity effectsafter 4 days In addition these authors reported differencesbetween the 1 g ground sample and the 1 g in-flight controlswith over 200 DEGs [2] Also Zhang et al [32] observed agreater difference between flight and ground samples withrespect to 1 g in-flight conditionsThese observations indicatethat the differing results could be related to the organismsinvestigated the time of exposure hardware experimentalparameters and set-up

In this study we report on results of a spaceflightexperimentThis experiment was part of the Simbox (Sciencein Microgravity Box) mission a joint project between thespace agencies from Germany (Deutsches Zentrum fur Luft-und Raumfahrt e V) and China (China Manned SpaceEngineering) in November 2011 As one out of 17 biologicalexperiments semisolid callus cultures of Arabidopsis wereexposed to a 17-day spaceflight on board of the Chinesespacecraft Shenzhou 8 Due to reduced viability after longerperiods of exposure within the flight hardware the calluscultures were metabolically quenched after 5 days in spaceResults of a whole-genomemicroarray screening (120583g exposedsamples 1 g in-flight samples kept in a reference centrifugeand 1 g ground samples) revealed major differences betweenboth 1 g controls but a minor impact of microgravity


21 Experiment-Specific Hardware (HW) The Simbox wasa modification of the Biobox-6 [33 34] which was devel-oped for unmanned recoverable capsules and space shuttlemissions Development and production were carried out byAstriumEADS Friedrichshafen Germany [35] This incu-bator (size of 461 times 551 times 273mm internal volume of34 L max power consumption of 130W and empty mass

Staticflight platform

Reference centrifuge


Position SP04 [FSGS]

Position C05 [FC]Inside of incubator housing

Figure 1 Photograph of the inside of the Simbox incubator usedwithin the flightground experiment (housing removed) The rotorof the reference centrifuge (position C05 for sample group FC) isindicated by a circle The static experimental platform is in themiddle and outside of the centrifuge rotor (position SP04 for samplegroup FS within the flight experiment and GS within the groundexperiment resp) (photograph DLRAstrium)

Slide with spikes

Window of CC biofoil and frame removed

Front CC

Substrate holder

Callus cultures

12a media with agar

Figure 2 Photograph of the inside of one culture chamber (CC)(experiment container (EC) window biofoil and frame removed)The semisolid callus cultures were positioned on substrate holders(slides) with plastic spikes on 12a agar containing culture media

of 17 kg) served as carrier for an experimentstatic platformwith an integrated centrifuge rotor (provides 1 g in-flightcontrol) The Simbox incubator (Figure 1) enabled samplecultivation at 22ndash24∘C (nominal temperature range) and 30ndash40 humidity throughout the mission A duplicate model ofthe Simbox was constructed for the ground experiment Ourbiological approach (experiment number 16) was realizedby means of three fully automated type V Experiment UnitEnvelopes (EUE plant cultivation unit without illumina-tion) EUEs consisted of support housing made of polyether-ketone with two culture chambers each (front and rear CC317 times 24 times 143mm plusmn 015mm) Our biological material waspositioned on substrate holders (slides) with plastic spikes(Figure 2) The latter were needed to keep the cultures inplace In order to allow gas exchange the CCs were sealedwith a biofoil made of polysulfone (Tecason S PolysulfoneEnsinger Inc Washington-Pennsylvania USA) In additiona peristaltic pump (flow rate of ge243mLmin) was used to


Window frame of CC with glued biofoil

Substrate holder slide with spikes

Rear CC Front CCType 5 EUE

housingCavity cover Type 1 EC

Fixativewaste tank Tubing system Liquid pump

Fittings for tube connection

Figure 3 Photograph of the fully automated plant cultivation unit type V EUE (left side) and EC removed (right side) (photographDLRAstrium)

connect the CC to a fixativewaste unit (volume 203mL plusmn05mL) EUEs were accommodated inside type I ExperimentContainers (ECs) (Figure 3) Via sensors parameters such astemperature humidity CO

2 and O

2content as well as acti-

vation of the pump system were recorded and transmitted

22 Cell Cultures Sterile cuttings (about 50mm long) ofstems of wild type Arabidopsis thaliana (cv Columbia Col-0) plants were used for callus formation on 12a media [36]containing 1 agar (Sigma-Aldrich Germany) Calli weretransferred to 500mL Erlenmeyer flasks with 200mL liquid12a medium and cultivated under sterile conditions at 23∘Cin the dark on a rotary shaker (130 rpm Infors BottmingenSwitzerland) as described previously [14] New medium wasadded every week to the resulting cell suspension Eightmonths before the Simbox mission an aliquot of this culture(3 g) was spread on 6 cm Petri dishes (Greiner Bio-OneFrickenhausen Germany) containing agar and 12a mediumCell cultures were mailed to the Institute of Physiologyand Ecology Shanghai (Laboratory of Prof Zheng) and thecultivation continued (as liquid suspension) as describedabove These suspension cultures were transferred to thePITC (Payload Integration and Test Center Beijing China)The cultivation was then continued on agar plates (see above)and finally these semisolid calli were brought to the launchsite (Jiuquan Satellite Launch Center Jiuquan China) byplane

23 Preparation of Final Experiment Configuration Oneday before the launch 11-day-old semisolid callus cultureswere transferred into the CCs with 2mL agar containingmedium (Figures 2 and 3) Two ECs were used for thespaceflight (flight models FM 16001 and FM 16002) and one

for the ground experiment (FM 16003) respectively One ofthe two ECs was contained in the centrifuge rotor and theother one was fixed at the experimentstatic platform (flightplatform) respectively (Figure 1) Metabolic quenching ofthe samples was by the injection of RNAlater (Ambion LifeTechnologiesDarmstadtGermany)This reagent is also usedto stabilize nucleic acids Twenty mL of this fixative wasfilled into the fixativewaste unit attached to the bottom ofthe EC Between handover and integration into the Simboxflightground incubator the ECs were stored at nominallaboratory temperature conditions (22ndash24∘C) The Simboxincubator was unpowered for about 3 hours during transportto the spacecraft During this time the lowest temperaturewas 21∘C (Figure 4)

24 The Experiment in Orbit The Simbox was launched onboard of the unmanned spacecraft Shenzhou 8 on Octo-ber 31 2011 at 2158 UTC (universal time coordinated)with a Long March 2F rocket from the cosmodrome inJSLC The precise mission timings including sample fixationtime points are illustrated in Figure 5 (for a gravity-levelprofile see Supplementary Material S1 available online athttpdxdoiorg1011552014547495) Experiment zero time(EZT) was set when the spacecraft reached the orbit At EZTthe centrifuge was activated to run with 7440 rpm Withinthe spacecraft the oxygen partial pressure ranged from1804 to 2732 kPa and the carbon dioxide partial pressurewas between minus003 and 046 kPa Radiation measurementsyielded a total dose of 593 to 81mSv and an equivalent doseof 037 to 051mSvd near the Simbox incubator (telemetrydata Chinese authorities personal communication) Thepump system was activated after 5 days in space and injectedthe fixative solution from the fixativewaste unit into the CCrsquos


Unpowered transport toShenzhou

Lowest temp

Launch215800(UTC) N

omin

al te

mp

rang

e

Time (hhmm)

Power supply on

Power supply off

2500

2450

2400

2350

2300

2250

2200

2150

2100

2050

22000000 0130 0300 0430 0600 0730 0900 1030 1200

rarr21∘C

TP1TP2TP3

Tem

pera

ture

(∘C)

3h 06min

Figure 4 Temperature profile as recorded by 3 temperature sensors (TP1-3) attached to the Simbox incubator during integration of ECs intothe incubator transport to Shenzhou and launch (data Astrium)

Figure 5 Precise mission timeline of the experiment in orbit (grey) and related ground experiment (white) Universal time coordinated(UTC) time units are given in hoursminutes seconds experimental zero time (EZT) Arrowheads (nabla) indicate sample fixation time pointsof sample groups FS FC and GS respectively

of FMs This yielded a final RNAlater concentration of about90 (vv) after mixing Temperature in CCs was kept at anominal range of 22 to 24∘C before during and after fixation(Figure 6) After 17 days in space the spacecraftwas separatedfrom Tjangong-1 and touched ground on November 17 2011

After landing and recovery of the capsule samples wereretrieved within 6 hours The ECs were disassembled andstored around 4∘C until they arrived in Tubingen on Novem-ber 25 2011 In the home laboratory calli were harvested andstored at minus80∘C until processing


Nom

inal

tem

p ra

nge

Transport to spacecraft (lowest

FC

FS2500

2450

2400

2350

2300

2250

2200

2150

21000000 4800 9600 14400 19200 24000 28800 33600 38400 43200

TX1

Time (hhmm)

TX2TX3TX4

EZT + 1202800

EZT + 1283830Te

mpe

ratu

re (∘

C)

temp 21∘C)

Figure 6 Temperature profile as recorded by 4 temperature sensors (TX1-4) attached to the Simbox incubator during the whole Simboxmission (data Astrium) Sample fixation time points for the spaceflight samples (FS and FC) are indicated by arrowheads (grey triangle)

25 Ground Control Immediately after the launch the labo-ratory equipment and cell cultures were brought back to thePITC by Chinese scientists The ground experiment startedwith a one-day delay on November 2 2011 (Figure 5) TheEUE was integrated into the Simbox duplicate accordingto the position in the flight incubator (experimentstaticplatform) and kept at 23∘C As in the experiment in orbitsamples weremetabolically quenched after 5 days (November7) The ground experiment ended on November 19 Thesamples were handled as described for the experiment inorbit

26 Experiment Conditions and Specification of GeneratedSamples During the Simbox mission the samples wereexposed to different experimental conditions In the experi-ment in orbit FM 16002 was attached to the static platform ofthe Simbox incubator and experienced 5 days ofmicrogravity(group FS Flight Static) FM 16001 was centrifuged resultingin a 1 g control (group FC in-flight centrifugation) In theground experiment the same experimental design was usedFM 16003 was fixed to the static platform (group GS groundstatic) In summary we obtained one biological sample perCC resulting in two replicates for each FM (front and rearCC) and for each experimental condition respectively

27 Isolation of Total RNA and High-Density OligonucleotideArrays Total RNA was extracted using the RNeasy Plus kit(Qiagen Hilden Germany) according to the manufacturerrsquos

instructions Quantity and quality controls were performedand samples were processed using theMessageAmp II-BiotinEnhanced Single Round aRNA Amplification Kit (AmbionLife Technologies Darmstadt Germany) as described earlier[21 37] Fragmented biotin-labeled aRNA was then submit-ted to a high throughputmicroarray analysis (GeneChipAra-bidopsis ATH1 Genome Array Ref 510690 LOT 4155830Affymetrix Santa Clara California USA) Hybridizationwasperformed according to the manufacturerrsquos instructions (fordetails see httpwwwaffymetrixcomsupporttechnicalmanualsaffx) The Affymetrix protocol EukGE-WS2 V4 wasused for washing and staining procedures

28 Gene Expression Analysis Expression data were calcu-lated from raw values of the detected signal intensity ofhybridization events of all spotted probe sets and savedas CEL data files Microarray data are available in theArrayExpress database (httpwwwebiacukarrayexpress[38]) under accession number E-MTAB-2518 For integrativedata analysis we used the open-source software Mayday[39] Normalization was performed using the robust multi-array average method of background-adjustment quantile-normalization andmedian-polish to ensure comparability ofarrays and estimate log

2expression values [40ndash42] Hierar-

chical clustering was performed by means of the neighbourjoining method [43] in order to reconstruct and visualizerelationships within expression values due to experimentconditions The Pearson Correlation coefficient was used to


Experiment in orbit [FS]

(a)

Experiment in orbit [FC]

(b)

Ground experiment [GS]

(c)

Figure 7 Photograph of Arabidopsis thaliana semisolidcallus cultures after a 5-day 120583g cultivation in orbit ((a) FS) 1 g in-flight cultivation((b) FC) or on ground ((c) GS) The photographs were taken after fixation by RNAlater and recovery of the spacecraft

calculate the distance between each experimental condition(FS FC andGS) and biological replicates (front and rearCC)The matrix of variant genes was filtered and subjected to aStudentrsquos 119905-test (119875 le 01) with combined false discovery rate(FDR) correction to identify significantly altered transcripts(119875 lt 01) between the sample groups FS and FC FS andGS and FC and GS respectively Differentially expressedgenes were determined by fold change (fc) calculation of log

2

transformed expression dataThereby the thresholdwas set atminus1ge log

2(fold change)ge 1 for at least 2-fold altered transcripts

[40 41 44] Additionally the Affymetrix probe identifierswere tested by Gene Set Enrichment Analysis (GSEA [45])for enrichment of functional ontologies usingGeneOntologyterms [46] within Mayday Thereby we focused on genesthat share their function in identical biological processes forinterpreting the genome-wide expression profiles

3 Results

The aim of this experiment was to characterize the transcrip-tome of Arabidopsis semisolid callus cultures after 5 daysin space Due to the availability of an in-flight centrifugeit was possible to compare expression data with (a) realmicrogravity samples (thought to yield the microgravityrelated alterations) and with (b) those from the groundcontrols (which should deliver effects of nonmicrogravityrelated spaceflight conditions) This was achieved with high-density oligonucleotide arrays

31 Performance of Hardware and Biological Material Thehardware was thoroughly tested in order to retain viabilityof the callus cultures for as long as possible These tests werefocused on the biocompatibility of the used materials gas-exchange properties of membranes and viability of the cellcultures under the cultivation conditions within the EC Wealso recorded the oxygen content within the CC [37] As this

declined from 8 to about 2mgL after 5 days automated sam-ple fixationwas set at day 5 after take-offMission parameterssuch as temperature were within nominal range during themission Radiation measurements recorded increased valuesAfter landing and return of the biological material to theUniversity of Tubingen (Germany) the samples were visuallychecked The fixed calli showed good morphology and hadwell grown during the initial culture of 5 days in space Thecalli from the 1 g controls (flight and ground experiment)were smaller compared to those exposed to microgravity(Figure 7)

32 Biology of Samples and Gene Expression Analysis Thequality of the extracted total ribonucleic acid was satisfyingfor GeneChip hybridization (for RNA quality see Supple-mentary Material S2) with clear bands representing the 28Sand 18S rRNA Whole-genome microarray screening wasperformed for each sample Due to the limited amount oftotal RNA the confirmation of expression data by quan-titative real-time PCR was not possible The data analysisrevealed experiment-specific properties of biological repli-cates which were visualized by hierarchical clustering on thebasis of the calculation of the Pearson Correlation coefficient(Figure 8) In this graph a relatively short distance impliesa high correlation between the samples As obvious fromFigure 8 the flight and ground experiment showed group-based clustering The short distance between FS and FC (FSand FC boxes) in contrast to GS (GS boxes) indicates thatnonmicrogravity related spaceflight conditions have majorimpact The transcriptome of the biological replicates withinthe experiment groups (front and rear chamber of FS FC GS119899 = 2) showed a high degree of similarity (Figure 8)This factwas confirmed by heat map generation based on calculatedcorrelations (Figure 9) The Pearson Correlation was about099 between front and rear CC for all three modules


FC re

ar

FC front

GS

front

GS rear

Flight experimentGround experiment

FS rear

FS front

Figure 8 Hierarchical clustering bymeans of the neighbour joiningmethod of generated sample groups (white ground experiment GSground static grey flight experiment FS flight space FC in-flightcentrifugation) Each EUE consisted of two culture chambers (frontand rear chambers illustrated by boxes)

(FS FC and GS 119899 = 2 Figure 9) Statistical (Studentrsquos 119905-test 119875 lt 01 and FDR correction) and comparative analysisshowed a relatively low response of semisolid callus cultures(Figure 10) Interestingly microgravity conditions did notinduce statistically significant changes (119875 lt 01) at thegene expression level although 298 genes were at least 2-fold differentially expressed (275 up- and 23 downregulated)within flight space (FS) samples In contrast nonmicro-gravity related spaceflight conditions interfered with geneexpression considerably Eight hundred ninety-seven geneswere significantly and differentially expressed (at least 2-fold119875 lt 01) when 1 g ground and 120583g exposed flight sampleswere compared Among them 463 were upregulated and 434geneswere downregulatedwithin FS (Figure 10) Comparisonbetween both 1 g controls (in-flight ground) resulted in 826significantly (119875 lt 01) differentially altered genes (543 up and283 downregulated Figure 10)Thereby 573 significant DEGs(119875 lt 01) were identical in both comparisons (Figure 10)

33 Identification of Altered Genes after Long-Term Micro-gravity For detection of gene expression changes due to 120583gexposure we compared data generated out of the samplegroups flight space (FS) and in-flight centrifugation (FC)Two hundred seventy-five genes were at least 2-fold differ-entially upregulated and 23 downregulated (Figure 10) Theapplication of statistics showed that there were no significant(119875 lt 01) alterations at the expression level after 5 daysin space By means of a Gene Ontology [46] based GeneSet Enrichment Analysis (GSEA) the DEGs were related tocommon biological processes In order to identify processeswhich are specifically influenced by microgravity conditionswe compared overrepresented processes that were identicalbetween sample group FS versus FC and FS compared toGS (Table 1) Most prominent were effects on the translationmachinery (Table 1 gene set number 24) Interestingly allgenes that were differentially upregulated and involved in

Ground static [GS front]

Ground static [GS rear]

Flight centrifugation [FC rear]

Flight centrifugation [FC front]

Flight space [FS rear]

Flight space [FS front]

Gro

und

stat

ic [G

S fro

nt]

Gro

und

stat

ic [G

S re

ar]

Flig

ht ce

ntrif

ugat

ion

[FC

rear

]

Flig

ht ce

ntrif

ugat

ion

[FC

front

]

Flig

ht sp

ace [

FS re

ar]

Flig

ht sp

ace [

FS fr

ont]

1 0994 0972 0973 0973 0974

0994 1 0981 098 0979 0983

0972 0981 1 0995 0987 099

0973 098 0995 1 0992 0991

0973 0979 0987 0992 1 0996

0974 0983 099 0991 0996 1

Figure 9 Pearson correlation heat map shows high degree ofsimilarity between front and rear culture chamber of each samplewithin each sample group Flight space (FS) in-flight centrifugation(FC) and ground static (GS)

translation processes were chloroplast-encoded This geneset comprises genes coding for several protein subunits andcomponents of ribosomes (eg ATCG00065 ATCG00660ATCG00770 andATCG00790) but also the nucleus-encodedtranslation initiation factor EIF-5A (AT1G13950) that is wellknown to regulate translation initiation and terminationwithin the cytoplasma of eukaryotes (Table 2) The otherpart of identified differentially upregulated genes is involvedin electron transport chains located within mitochondria(Table 1 gene sets number 4 8 and 11) such as subunits ofthe NADH dehydrogenase multi-enzyme complex of the res-piratory chain (ATMG00650 ATMG00070 ATMG00580)(Table 2) Mitochondrial electron transport is connectedto the production of adenosine triphosphate (ATP) Thusthe gene set representative for ATP biosynthesis was alsopart of the DEGs (ATCG00120 ATMG00410 ATCG00480and ATCG00150) (Table 2) Within the 23 downregulatedgenes (at least 2-fold) no special gene sets could befound but the largest group codes for heat shock pro-teins (AT4G27670 AT2G29500 AT5G12020 AT5G59720AT4G25200 AT1G53540 and AT5G12030)

34 Attempt to Distinguish between Effects of Microgravityand Nonmicrogravity Related Spaceflight Conditions on GeneExpression One aim of this investigation was to separateresponses to microgravity from those of nonmicrogravityrelated spaceflight conditions Until today onlymarginal dataexist about these effects on plants in spaceThus we screenedfor genes that were significantly (119875 lt 01) altered withinspaceflight samples (FS and FC) compared to the 1 g groundcontrol and were identical between FS and FC comparedto GS This overlap yielded 573 significantly altered (119875 lt01) DEGs (Figure 10) The GSEA of these genes representeddiverse biological processes (Table 1 bold font) The majority


Flig

ht

Gro

und

897 in total 20 in total 826 in total 155 in total

DEGs

298 in totalFlight space

FS front FS rear

Flight centrifugation

FC front FC rear

Ground static

GS front GS rear

573 in total 9 in total

Sign DEGs (P lt 01)

Uncorrected P value

Sign DEGs (P lt 01)

Uncorrected P value FDR-corrected P value

Sign DEGs (P lt 01)

Uncorrected P value FDR-corrected P valueFDR-corrected P value

463uarr434darr in FS 13uarr7darr in FS 543uarr283darr in FC 96uarr59darr in FC

275uarr23darr in FS

Figure 10 Overview of the number of differentially (fold change (fc) at least 2) and significantly expressed genes (DEGs 119875 lt 01) withinthe flight (grey) and ground (white) experiment The different sample groups are illustrated by boxes Up- and downregulated transcripts aresymbolized by arrows behind the number of altered genes Genes that are significantly (119875 lt 01) differentially expressed are shown in boxesframed in black (bold lines)

of these genes could be related to intracellular signalingpathways such as mitogen-activated protein kinase (MAPK)cascades and protein phosphorylation (Table 1 gene setnumber 6 and 12) Included were different MAP kinases (egAT1G01560 AT1G73500) serinethreoninetyrosine kinases(eg AT1G20650 AT5G16900 and AT4G38470) and manyother kinases (Table 3) Furthermore we identified genescoding for members of the calcium-binding EF-hand proteinfamily (AT3G01830 AT3G47480) and the WRKY transcrip-tion factors 54 70 and 38 (AT2G40750 AT3G56400 andAT5G22570) that have also transcription regulation activity(Table 3) Additionally the spaceflight environment otherthan microgravity had a significant (119875 lt 01) impact ongeneral stress-responsive (gene set number 20) and defense-related genes (3) especially those involved in the response tooxidative stress and respiratory burst responses (21) Theseare peroxidases 21 4 52 and 25 (AT2G37130 AT1G14540AT5G05340 and AT2G41480) catalase 3 (AT1G20620) andreceptor-like kinases (AT5G46330 AT2G19190) The lattercan be induced upon contact with the bacterial proteinflagellin which is an important elicitor of the plant defenseresponse These kinases are also important members ofthe MAP kinase signaling cascade Furthermore generalmetabolic processes (gene set number 7) protein targeting(13) and rRNA processing (21) were overrep-resented due tononmicrogravity related conditions in space

4 Discussion

The expression data of Arabidopsis semisolid callus culturesshow alterations in differential gene expression in responseto microgravity However the influence of the spaceflightenvironment in addition to microgravity is significant

41 Identification of Altered Genes after 5 Days of Micro-gravity Comparison between microgravity and 1 g spacecontrols revealed about 298 differentially (but not signifi-cantly) expressed genes This number is low in comparisonto short-term exposures to microgravity within a rangeof minutes (TEXUS 47 sounding rocket experiment [47])or seconds (14 DLR parabolic flight campaign [21]) Thisfinding could be due to the small number of biologicalreplicates (2 biological replicates only due to limited mate-rial and hardware) However similar observations are alsoreported by others After 4 days in space Arabidopsis plantsexhibited only 27 transcripts which were at least 2-foldaltered at their expression level [2] This might indicate thatplants respond immediately to a microgravity environmentbut then adapt to the new situation on the longer runAlso Zhang et al [32] could also identify only 45 proteinschanged in expression after 14 days in space (same mission)Genes with prolonged changes in expression could howeverprovide important information about the physiological needs


Table 1 Visualization of enriched Gene Ontology categorization terms (GSEA Gene Set Enrichment Analysis of biological processes) Genesets identical in FSFC and FSGS are not colored the ones identical in FSGS FCGS and the overlap of both are in bold font (FS = flightspace FC = flight centrifugation and GS = ground static)

Number Enriched gene set (biological process) FSFC FSGS FCGS OverlapGene set size

1 ATP catabolic process 0 8 8 72 ATP biosynthetic process 10 9 0 03 Defense response 0 20 26 144 Mitochondrial electron transport chain 7 7 0 05 Lipid metabolic process 0 8 7 66 MAPK cascade 0 29 36 277 Metabolic process 0 25 20 168 Mitochondrial electron transport 11 11 0 09 Oxidation-reduction process 0 13 12 810 Photosynthesis light harvesting 0 5 5 511 Photosynthetic electron transport chain 5 5 0 012 Protein phosphorylation 0 23 31 1813 Protein targeting to membrane 0 12 13 10

14 Regulation of transcriptionDNA-dependent 0 12 11 8

15 Respiratory burst involved in defenseresponse 0 22 26 21

16 Response to chitin 0 7 6 617 Response to ethylene stimulus 0 5 6 518 Response to hypoxia 0 6 9 619 Response to oxidative stress 0 15 13 920 Response to stress 0 9 9 621 rRNA processing 0 16 15 1422 Toxin catabolic process 0 7 7 623 Transition metal ion transport 0 10 12 824 Translation 27 28 0 0

25 Two-component signal transductionsystem 0 6 5 5

after a few days in space These include an upregulatedgroup of genes which code for proteins that constitute theribosomal complex within plastids These are necessary fortranslation of mRNAThe upregulation of the mitochondrialelectron transport chain could indicate an increased need forATP The upregulated expression of NADH dehydrogenasecould have the same reason Interestingly gene productsinvolved in processes like the response to stress proteindegradation or programmed cell death appeared not to bealtered in expression The involvement of a series of geneswith still unknown functions (not shown) suggests that thespace environment induces also unknown cellular processesTogether with the fact that there were no significant changesin gene expression detectable after 5 days of microgravity letsus suggest that at this stage the impact of a lack of gravitationon cell physiology was not too heavyThe space environmentper se however causes possibly an increased energy demandas shown by the upregulation of respiratory componentsThisaspect should be taken into consideration when plants will

be used to provide nutrients oxygen and energy on longduration space missions

Heat shock proteins (HSPs) dominate the group oftranscripts which are reduced in amount (not shown) Theseproteins are involved in many forms of stress response Theyenable the folding and membrane translocation of proteinsand are thought to reconstitute the tertiary structure ofproteins affected by stress events This way they can increasethe stress tolerance A decreased expression (our study)should thus indicate a lower number of proteins affectedin their structure and was also reported for Arabidopsis invitro callus cultures under simulatedmicrogravity conditions(magnetic levitation magnetic field strength 101 Tesla) [48]as well as for the single-cell system of the fern Ceratopterisrichardii [9] There are however also reports on increasedexpression of HSPs [15 16 21 24]

A group of plant genes which are always affected byaltered gravity are those involved in cell wall modification[2 49ndash51] This reflects the need for increased stability


Table 2 Differentially expressed genes (fold change (fc) at least 2) within the sample group flight space (FS frontrear CC) compared toin-flight centrifugation (FC) Samples taken after 5-day cultivation at microgravity and sorted according to the overrepresented biologicalprocesses identified by GSEA to be the most prominent

Number ATG number Gene namedescription log (fc) Enriched Gene set (biological process)1 ATCG00065 Ribosomal protein S12 236 Translation2 ATCG00660 Ribosomal protein L20 214 Translation3 ATCG00770 30S ribosomal protein S8 196 Translation4 ATCG00160 Ribosomal protein S2 184 Translation5 ATCG00790 Ribosomal protein L16 18 Translation6 ATCG00780 Ribosomal protein L14 163 Translation7 AT1G13950 Eukaryotic translation initiation factor 5A-1 114 Translation8 ATCG01120 Ribosomal protein S15 111 Translation9 ATCG00750 Ribosomal protein S11 105 Translation10 ATCG00800 Ribosomal protein S3 104 Translation11 ATMG00650 NADH dehydrogenase subunit 4L 23 Mitochondrial electron transport12 ATMG00060 NADH dehydrogenase subunit 5 184 Mitochondrial electron transport13 AT2G07751 NADH-ubiquinoneplastochinone oxidoreductase 175 Mitochondrial electron transport14 ATCG01050 Subunit of NAD(P)H dehydrogenase complex 174 Mitochondrial electron transport15 ATMG00160 Cytochrome c oxidase subunit 2 166 Mitochondrial electron transport16 ATMG00070 NADH dehydrogenase subunit 9 15 Mitochondrial electron transport17 ATCG00420 NADH dehydrogenase subunit J 143 Mitochondrial electron transport18 ATCG01250 NADH dehydrogenase ND2 125 Mitochondrial electron transport19 ATMG00510 NADH dehydrogenase subunit 7 124 Mitochondrial electron transport20 ATMG00270 NADH dehydrogenase subunit 6 124 Mitochondrial electron transport21 ATMG00580 NADH dehydrogenase subunit 4 119 Mitochondrial electron transport22 ATCG01070 NADH dehydrogenase ND4L 113 Mitochondrial electron transport23 ATCG00120 ATPase 120572-subunit 215 ATP biosynthesis24 ATCG00140 ATPase III subunit 159 ATP biosynthesis25 ATMG00410 ATPase subunit 6 156 ATP biosynthesis26 ATCG00130 ATPase F subunit 147 ATP biosynthesis27 ATCG00480 120573-Subunit of ATP synthase 133 ATP biosynthesis28 ATCG00150 Subunit of ATPase complex CF0 112 ATP biosynthesis

(hypergravity) or more flexibility (microgravity) In thepresent study expression of expansins (cell wall loosening)is increased (not shown) This might be the reason for theenhanced size of the microgravity cultures when comparedto the 1 g controls (Figure 7)

42 Impact of the Nonmicrogravity Related Spaceflight Condi-tions on Gene Expression The availability of a 1 g referencecentrifuge enabled us to screen for genes affected by nonmi-crogravity related spaceflight conditions in that we comparedexpression data between 120583g exposed and 1 g space with 1 gground samples This resulted in a considerable numberof identical genes altered in mRNA abundance (573 genes)(Figure 10) We thus assume that this could be due to effectsof spaceflight-related environmental conditions includingspace radiation Radiation measurements inside the capsulein a position close to our samples yielded a total dose of 59to 81mSV (milliSieverts) and an equivalent dose of 037 to051mSVd (data Chinese authorities) This is considerablymore compared to terrestrial conditions (1 to 2mSVa) and

could be one of the reasons for the alterations at transcriptlevels obviously not related to 120583g Also Zhang et al [32]reported a greater difference on protein expression of non-120583gconditions Analysis showed that both experimental condi-tions (120583g and non-120583g spaceflight conditions) affect differentbiological processes (Table 1) Overrepresented processesshould not be regarded separately as they are closely linkedtogether within a plant cell For example the formation ofreactive oxygen species (ROS) is one of the initial responsesupon most kinds of stresses They are also produced asby-products of redox reactions They are important secondmessengers as well as toxic species and their cellular levelsare closely controlled by detoxification systems [52ndash54] Therole of ROS in response to environmental changes canhowever also be deduced from alterations in gene productsinvolved in ROS production and turnover In this studywe observed that many ROS-related genes are significantlyregulated (Table 3) These comprise peroxidases catalaseand a glutathione S-transferase (Table 3) These proteins aresuggested to be part of the stress-induced antioxidant sys-tem [55] Glutathione S-transferases also possess peroxidase


Table 3 Differentially (at least 2-fold) and significantly expressed genes (119875 lt 01 573 in total) that are identical between flight space (FS) aswell as in-flight centrifugation (FC) compared to ground static (GS) Changes are due to nonmicrogravity related spaceflight conditions Thegenes are sorted according to the overrepresented biological processes identified by GSEA to be most prominent

No ATG number Gene namedescription log (fc) (119875 value)FS versus GS

log (fc) (119875 value)FC versus GS Biological process

1 AT1G01560 MAP kinase 11 183 (0034) 213 (0027) MAPK cascade2 AT1G73500 MAP kinase 9 134 (0006) 137 (0038) MAPK cascade

3 AT3G01830 Calcium-binding EF-hand familyprotein 13 (0032) 185 (0027) MAPK cascade


5 AT2G40750 WRKY DNA-bindingtranscription factor 54 129 (0008) 174 (0006) MAPK cascade



8 AT3G15500 NAC-domain containingtranscription factor 3 298 (572119864 minus 4) 263 (0003) MAPK cascade

9 AT1G35670Calcium-dependentcalmodulin-independent proteinkinase 2

12 (0002) 123 (0003) Proteinphosphorylation

10 AT1G20650 Serinethreonine protein kinasesuperfamily protein minus14 (0024) minus15 (0018) Protein

phosphorylation

11 AT3G61160 Serinethreonine protein kinasefamily protein minus122 (0007) minus14 (0008) Protein

phosphorylation

12 AT1G78290 Serinethreonine protein kinasefamily protein 2C 171 (0019) 20 (0034) Protein

phosphorylation

13 AT4G18640 Serinethreonine protein kinasefamily protein 108 (0019) 107 (0014) Protein

phosphorylation

14 AT4G18950 Serinethreoninetyrosineprotein kinase family protein 253 (0031) 317 (002) Protein

phosphorylation

15 AT5G16900 Leucine-rich repeat proteinkinase family protein 142 (0023) 20 (0012) Protein

phosphorylation


phosphorylation

17 AT4G11480 Cysteine-rich receptor-likeprotein kinase family protein 156 (005) 189 (0033) Protein

phosphorylation


phosphorylation

19 AT4G38470 Tyrosine kinase family protein 46 114 (0008) 134 (0015) Proteinphosphorylation

20 AT1G69790 Protein kinase superfamilyprotein 119 (0038) 112 (0009) Protein

phosphorylation

21 AT5G53450 Protein kinase 188 (0088) 189 (0075) Proteinphosphorylation

22 AT1G51620 Protein kinase family protein 18 (0052) 231 (0048) Proteinphosphorylation

23 AT3G04530 Phosphoenolpyruvatecarboxylase kinase 2 minus16 (006) minus119 (043) Protein

phosphorylation

24 AT5G63650 Protein kinase 25 minus126 (0028) minus101 (0032) Proteinphosphorylation

25 AT1G16260 Cell-wall associated proteinkinase family protein 173 (0006) 213 (0003) Protein

phosphorylation

26 AT1G68690 Proline-rich extension-likereceptor kinase family protein 104 (0002) 103 (004) Protein

phosphorylation


Table 3 Continued



27 AT5G46330 Flagellin 2-induced receptor-likekinase minus185 (0043) minus238 (0016) Defense response

28 AT2G19190 Flagellin 22-inducedreceptor-like kinase 248 (0065) 227 (0075) Defense response

29 AT2G15120 Disease-resistance family protein 268 (0035) 253 (004) Defense response30 AT1G59780 Disease resistance protein 137 (0092) 198 (0052) Defense response31 AT1G63880 Disease resistance protein minus181 (0003) minus179 (0016) Defense response

32 AT2G39200

Transmembranedomain-containing proteinsimilar to mildew resistanceprotein 12

26 (0059) 255 (0063) Defense response

33 AT1G19610 Pathogenesis-related protein 14 minus217 (0002) minus214 (0029) Defense response

34 AT3G20600 Nonrace specific diseaseresistance protein 105 (0037) 212 (0011) Defense response

35 AT1G02360 Chitinase family protein 26 (0026) 287 (0019) Defense response36 AT3G54420 Chitinase family protein class IV 173 (0055) 253 (0026) Defense response

37 AT4G21390 Serinethreonine protein kinasefamily protein 15 (0068) 186 (0031) Defense response

38 AT3G46280 Protein kinase family protein 183 (0074) 23 (0048) Defense response39 AT5G35750 Histidine kinase 2 minus121 (0042) minus135 (0026) Defense response

40 AT2G37130 Peroxidase 21 minus306 (0014) minus34 (0008) Response tooxidative stress

41 AT1G14540 Peroxidase 4 335 (0018) 328 (002) Response tooxidative stress


43 AT4G37530 Peroxidase family protein 217 (0035) 215 (0026) Response tooxidative stress


45 AT1G20620 Catalase 3 minus112 (007) minus139 (0052) Response tooxidative stress

46 AT2G29490 Glutathione S-transferase 19 classtau 1 175 (007) 17 (0073) Response to

oxidative stress

47 AT3G22370 Oxidase family protein 13 (0013) 10 (0089) Response tooxidative stress

48 AT4G37220 Stress-responsive protein 287 (0004) 187 (0049) Response to stress49 AT4G21870 Heat shock protein 265 minus131 (0002) minus143 (0012) Response to stress

50 AT2G38750 Calcium-dependentphospholipid binding protein 148 (002) 115 (0032) Response to stress

activity and can thus prevent cell damage by peroxides suchas hydrogen peroxide [56 57]The increase in detoxification-related transcripts appears reasonable as radiation in orbitconsists of highly energetic (HZE) particles from interplan-etary galactic sources or results from solar particle eventswhich could have an impact on cells [58ndash62] Wan et al [6364] showed that X-rays 120574-rays protons and heavy chargedparticles increased oxidative stress in different cell types andcountermeasures for space radiation effects are the use ofantioxidants [62] Similar responses are probable for plantcells Therefore the impact of long-term space radiation on

the transcriptome of Arabidopsis should be investigated inground-based studies in simulation testbeds for the spaceenvironments [59]

In addition a range of WRKY transcription factors andcomponents of signaling chains (Ca2+-dependent proteinsMAP kinases) were identified (Table 3) These responsivekinases (Table 3) are potentially also modulated by cytosolicfluctuations of H

2O2and can thus be part of signal trans-

duction chains starting from hydrogen peroxide (for defense-related genes in tomato see Orozco-Cardenas et al [65]) Incontrast to other observations to altered gravitation [2 15]


in this study genes which are defense- resistance- andpathogen-related are significantly altered due to non-120583grelated spaceflight conditions

5 Conclusions

In this study gene expression changes within Arabidopsiswild type semisolid callus cultures were investigated aftera 5-day spaceflight and compared to on-board and groundcontrols Faced with limited HW capacities (only 3 EUEs)and small amounts of biological material (119899 = 2 foreach sample group) high-density oligonucleotide arrays wereused to screen for changes at the gene expression level Forfuture investigations it would thus be desirable to have flightrepetitions and an adequate amount of samples for addi-tional analysis (eg qPCR) Unexpectedly the response ofcallus cultures to long-term microgravity was less prominentcompared to nonmicrogravity related spaceflight conditionsThe latter including space radiation induced differential andsignificant expression changes of transcripts that are involvedin the stress-induced antioxidant system signalling chainsand defense-resistance-related genes These findings clearlyhighlight that the use of an in-flight reference centrifuge (1 gin-flight control) should be mandatory during space flightmissions



Acknowledgments

Thisworkwas supported by a grant of theDeutsches Zentrumfur Luft- und Raumfahrt (DLR) (Grant no 50WB0723) toRudiger Hampp The authors are indebted to Dr MarkusBraun (DLR) for perfect campaign organization and toAchim Schwarzwalder Dr Astrid Horn and the EADSAstrium team for hardware construction and technical sup-port They thank the China Manned Space Engineering andthe Chinese scientists especially Professor Zheng for goodcooperation at launch site They are grateful to Margret Eckefor skilful production and maintenance of the cell culturesand Fabian Bergwitz for assistance in China as well as AnneHennig for ground-based experiments before the mission

References

[1] G Perbal and D Driss-Ecole ldquoMechanotransduction ingravisensing cellsrdquo Trends in Plant Science vol 8 no 10 pp498ndash504 2003

[2] M J Correll T P Pyle K D L Millar et al ldquoTranscriptomeanalyses of Arabidopsis thaliana seedlings grown in spaceimplications for gravity-responsive genesrdquo Planta vol 238 no3 pp 519ndash533 2013

[3] M Braun ldquoGravitropism in tip-growing cellsrdquo Planta vol 203no 1 pp S11ndashS19 1997

[4] M Braun B Buchen and A Sievers ldquoElectron microscopicanalysis of gravisensing Chara rhizoids developed undermicro-gravity conditionsrdquo The FASEB Journal vol 13 no 8 pp S113ndashS120 1999

[5] M Braun ldquoGravity perception requires statoliths settled onspecific plasma membrane areas in characean rhizoids andprotonematardquo Protoplasma vol 219 no 3-4 pp 150ndash159 2002

[6] M Braun J Hauslage A Czogalla and C Limbach ldquoTip-localized actin polymerization and remodeling reflected by thelocalization of ADF profilin and villin are fundamental forgravity-sensing and polar growth in characean rhizoidsrdquoPlantavol 219 no 3 pp 379ndash388 2004

[7] M Braun and C Limbach ldquoRhizoids and protonemata ofcharacean algae model cells for research on polarized growthand plant gravity sensingrdquo Protoplasma vol 229 no 2ndash4 pp133ndash142 2006

[8] M L Salmi T J Bushart and S J Roux ldquoAutonomous gravityperception and responses of single plant cellsrdquo Gravitationaland Space Biology vol 25 pp 6ndash13 2011

[9] M L Salmi and S J Roux ldquoGene expression changes inducedby space flight in single-cells of the fern Ceratopteris richardiirdquoPlanta vol 229 no 1 pp 151ndash159 2008

[10] O Rasmussen D A Klimchuk E L Kordyum et al ldquoTheeffect of exposure to microgravity on the development andstructural organisation of plant protoplasts flown onBiokosmos9rdquo Physiologia Plantarum vol 84 no 1 pp 162ndash170 1992

[11] E Hoffmann K Schonherr and R Hampp ldquoRegenerationof plant cell protoplasts under microgravity investigation ofprotein patterns by SDS-PAGE and immunoblottingrdquo Plant CellReports vol 15 no 12 pp 914ndash919 1996

[12] R Hampp E Hoffmann K Schonherr P Johann and L DeFilippis ldquoFusion and metabolism of plant cells as affected bymicrogravityrdquo Planta vol 203 pp S42ndashS53 1997


[14] M Martzivanou and R Hampp ldquoHyper-gravity effects on theArabidopsis transcriptomerdquo Physiologia Plantarum vol 118 no2 pp 221ndash231 2003

[15] A L Paul A K Zupanska D T Ostrow et al ldquoSpaceflighttranscriptomes Unique responses to a novel environmentrdquoAstrobiology vol 12 no 1 pp 40ndash56 2012

[16] A K Zupanska F C Denison R J Ferl and A-L PaulldquoSpaceflight engages heat shock protein and other molecularchaperone genes in tissue culture cells of Arabidopsis ThalianardquoThe American Journal of Botany vol 100 no 1 pp 235ndash2482013

[17] R Edgar M Domrachev and A E Lash ldquoGene ExpressionOmnibus NCBI gene expression and hybridization array datarepositoryrdquo Nucleic Acids Research vol 30 no 1 pp 207ndash2102002

[18] A Brazma H Parkinson U Sarkans et al ldquoArrayExpressmdashapublic repository for microarray gene expression data at theEBIrdquo Nucleic Acids Research vol 31 no 1 pp 68ndash71 2003

[19] J Kilian D Whitehead J Horak et al ldquoThe AtGenExpressglobal stress expression data set protocols evaluation andmodel data analysis of UV-B light drought and cold stressresponsesrdquo Plant Journal vol 50 no 2 pp 347ndash363 2007

[20] D Swarbreck C Wilks P Lamesch et al ldquoThe ArabidopsisInformation Resource (TAIR) Gene structure and function


annotationrdquo Nucleic Acids Research vol 36 no 1 pp D1009ndashD1014 2008

[21] N Hausmann S Fengler A Hennig M Franz-Wachtel RHampp and M Neef ldquoCytosolic calcium hydrogen peroxideand related gene expression and protein modulation in Ara-bidopsis thaliana cell cultures respond immediately to alteredgravitation parabolic flight datardquo Plant Biology vol 16 no 1pp 120ndash128 2014

[22] J Z Kiss W J Katembe and R E Edelmann ldquoGravitropismand development of wild-type and starch-deficient mutants ofArabidopsis during spaceflightrdquo Physiologia Plantarum vol 102no 4 pp 493ndash502 1998

[23] A L Paul C J Daugherty E A Bihn D K Chapman KL L Norwood and R J Ferl ldquoTransgene expression patternsindicate that spaceflight affects stress signal perception andtransduction in Arabidopsisrdquo Plant Physiology vol 126 no 2pp 613ndash621 2001

[24] A-L Paul M P Popp W B Gurley C Guy K L Norwoodand R J Ferl ldquoArabidopsis gene expression patterns are alteredduring spaceflightrdquo Advances in Space Research vol 36 no 7pp 1175ndash1181 2005

[25] K D L Millar P Kumar M J Correll et al ldquoA novelphototropic response to red light is revealed in microgravityrdquoNew Phytologist vol 186 no 3 pp 648ndash656 2010

[26] J Allen P A Bisbee R L Darnell et al ldquoGravity control ofgrowth form in Brassica Rapa and Arabidopsis Thaliana (Bras-sicaceae) consequences for secondary metabolismrdquo AmericanJournal of Botany vol 96 no 3 pp 652ndash660 2009

[27] M EMusgrave A Kuang Y Xiao et al ldquoGravity independenceof seed-to-seed cycling in Brassica rapardquo Planta vol 210 no 3pp 400ndash406 2000

[28] G Perbal and D Driss-Ecole ldquoSensitivity to gravistimulus oflentil seedling roots grown in space during the IML 1 Missionof Spacelabrdquo Physiologia Plantarum vol 90 no 2 pp 313ndash3181994

[29] B C Tripathy C S Brown H G Levine and A D KrikorianldquoGrowth and photosynthetic responses of wheat plants grownin spacerdquo Plant Physiology vol 110 no 3 pp 801ndash806 1996

[30] G W Stutte O Monje R D Hatfield A L Paul R J Ferl andC G Simone ldquoMicrogravity effects on leaf morphology cellstructure carbon metabolism and mRNA expression of dwarfwheatrdquo Planta vol 224 no 5 pp 1038ndash1049 2006

[31] J Ueda KMiyamoto T Yuda et al ldquoGrowth and developmentand auxin polar transport in higher plants under microgravityconditions in space BRIC-AUX on STS-95 space experimentrdquoJournal of Plant Research vol 112 no 1108 pp 487ndash492 1999

[32] Y Zhang L Wang J Xie and H Zheng ldquoDifferential proteinexpression profiling ofArabidopsis thaliana callus undermicro-gravity on board the Chinese SZ-8 spacecraftrdquo Planta In press

[33] M Vukich A Donati and V Zolesi ldquoKayser Italia hardware forradiation and microgravity experiments in spacerdquo RendicontiLincei vol 25 no 1 pp 7ndash11 2014

[34] E Brinckmann ldquoCentrifuges and their application for biologi-cal experiments in spacerdquoMicrogravity Science and Technologyvol 24 no 6 pp 365ndash372 2012

[35] Astrium GmbH Astrium Space Transportation Astrium SpaceBiology Product Catalog Friedrichshafen Germany 2012

[36] H Kleinig ldquoPflanzlicheGewebekultur Ein PraktikumrdquoBiologiein unserer Zeit vol 16 no 4 p 128 1986

[37] S Fengler M Neef M Ecke and R Hampp ldquoThe Simboxexperiment with Arabidopsis thaliana cell cultures hardware

tests and first results from theGerman-Chinese satellitemissionShenzhou 8rdquo in Proceedings of the Life in Space for Life on EarthSymposium vol ESA SP-706 2013

[38] G Rustici N Kolesnikov M Brandizi et al ldquoArrayExpressupdate-trends in database growth and links to data analysistoolsrdquo Nucleic Acids Research vol 41 no 1 pp D987ndashD9902013

[39] F Battke S Symons and K Nieselt ldquoMaydaymdashintegrativeanalytics for expression datardquo BMC Bioinformatics vol 11article 121 2010

[40] R A Irizarry BM Bolstad F Collin LMCope BHobbs andT P Speed ldquoSummaries of Affymetrix GeneChip probe leveldatardquo Nucleic Acids Research vol 31 no 4 2003


[42] B M Bolstad R A Irizarry M Astrand and T P Speed ldquoAcomparison of normalizationmethods for high density oligonu-cleotide array data based on variance and biasrdquo Bioinformaticsvol 19 no 2 pp 185ndash193 2003

[43] M Simonsen T Mailund and C N Pedersen ldquoRapidneighbour-joiningrdquo inAlgorithms in Bioinformatics vol 5251 ofLecture Notes in Computer Science pp 113ndash122 Springer BerlinGermany 2008

[44] D M Mutch A Berger R Mansourian A Rytz and M-ARoberts ldquoThe limit fold change model a practical approach forselecting differentially expressed genes from microarray datardquoBMC Bioinformatics vol 3 article 17 2002

[45] A Subramanian P Tamayo V K Mootha et al ldquoGene setenrichment analysis a knowledge-based approach for inter-preting genome-wide expression profilesrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 102 no 43 pp 15545ndash15550 2005

[46] M Ashburner C A Ball J A Blake et al ldquoGene ontology toolfor the unification of biologyrdquoNature Genetics vol 25 no 1 pp25ndash29 2000

[47] M Babbick Z Barjaktarovic and R Hampp ldquoAlterations in theexpression of transcription factors in Arabidopsis thaliana cellcultures during sounding rocket 120583Grdquo in Proceedings of the 18thESA Symposium on European Rocket and Balloon Programmespp 473ndash477 June 2007

[48] A I Manzano J J W A van Loon P C M Christianen J MGonzalez-Rubio F J Medina and R Herranz ldquoGravitationaland magnetic field variations synergize to cause subtle varia-tions in the global transcriptional state of Arabidopsis in vitrocallus culturesrdquo BMC Genomics vol 13 no 1 article 105 2012

[49] HWang Q Z HuiW Sha R Zeng and C X Qi ldquoA proteomicapproach to analysing responses of Arabidopsis thaliana calluscells to clinostat rotationrdquo Journal of Experimental Botany vol57 no 4 pp 827ndash835 2006

[50] T Hoson K Soga R Mori et al ldquoStimulation of elongationgrowth and cell wall loosening in rice coleoptiles under micro-gravity conditions in spacerdquo Plant and Cell Physiology vol 43no 9 pp 1067ndash1071 2002

[51] A Nasir S M Strauch I Becker et al ldquoThe influence ofmicrogravity on Euglena gracilis as studied on Shenzhou 8rdquoPlant Biology vol 16 no supplement 1 pp 113ndash119 2014

[52] K Apel and H Hirt ldquoReactive oxygen species metabolismoxidative stress and signal transductionrdquo Annual Review ofPlant Biology vol 55 pp 373ndash399 2004


[53] S Davletova K Schlauch J Coutu and R Mittler ldquoThe zinc-finger protein Zat12 plays a central role in reactive oxygen andabiotic stress signaling in Arabidopsisrdquo Plant Physiology vol139 no 2 pp 847ndash856 2005

[54] R Mittler ldquoAbiotic stress the field environment and stresscombinationrdquo Trends in Plant Science vol 11 no 1 pp 15ndash192006

[55] P G Sappl A J Carroll R Clifton et al ldquoThe Arabidopsisglutathione transferase gene family displays complex stressregulation and co-silencing multiple genes results in alteredmetabolic sensitivity to oxidative stressrdquoThe Plant Journal vol58 no 1 pp 53ndash68 2009

[56] V P Roxas R K Smith Jr E R Allen and R D Allen ldquoOver-expression of glutathione S-transferaseglutathione peroxidaseenhances the growth of transgenic tobacco seedlings duringstressrdquo Nature Biotechnology vol 15 no 10 pp 988ndash991 1997

[57] V P Roxas S A Lodhi D K Garrett J R Mahan and RD Allen ldquoStress tolerance in transgenic tobacco seedlings thatoverexpress glutathione S-transferaseglutathione peroxidaserdquoPlant and Cell Physiology vol 41 no 11 pp 1229ndash1234 2000

[58] G Horneck ldquoRadiobiological experiments in space a reviewrdquoNuclear Tracks and Radiation Measurements vol 20 no 1 pp185ndash205 1992

[59] G Horneck ldquoAstrobiology studies of microbes in simulatedinterplanetary spacerdquo in Laboratory Astrophysics and SpaceResearch pp 667ndash685 1999

[60] C Baumstark-Khan C E Hellweg A Arenz and M MMeier ldquoCellular monitoring of the nuclear factor 120581B pathwayfor assessment of space environmental radiationrdquo RadiationResearch vol 164 no 4 pp 527ndash530 2005

[61] C E Hellweg and C Baumstark-Khan ldquoGetting ready forthe manned mission to Mars the astronautsrsquo risk from spaceradiationrdquoNaturwissenschaften vol 94 no 7 pp 517ndash526 2007

[62] A R Kennedy ldquoBiological effects of space radiation anddevelopment of effective countermeasuresrdquo Life Sciences inSpace Research vol 1 no 1 pp 10ndash43 2014

[63] X S Wan Z Zhou and A R Kennedy ldquoAdaptation of thedichlorofluorescein assay for detection of radiation-inducedoxidative stress in cultured cellsrdquo Radiation Research vol 160no 6 pp 622ndash630 2003

[64] X S Wan Z Zhou J H Ware and A R Kennedy ldquoStandard-ization of a fluorometric assay for measuring oxidative stress inirradiated cellsrdquo Radiation Research vol 163 no 2 pp 232ndash2402005

[65] M L Orozco-Cardenas J Narvaez-Vasquez and C A RyanldquoHydrogen peroxide acts as a second messenger for the induc-tion of defense genes in tomato plants in response to woundingsystemin and methyl jasmonaterdquo The Plant Cell vol 13 no 1pp 179ndash191 2001

Research ArticleRCCS Bioreactor-Based Modelled Microgravity InducesSignificant Changes on In Vitro 3D Neuroglial Cell Cultures

Caterina Morabito12 Nathalie Steimberg23 Giovanna Mazzoleni23 Simone Guarnieri12

Giorgio Fanograve-Illic12 and Maria A Mariggiograve124

1 Department of Neuroscience Imaging and Clinical Sciences Unit of Functional Biotechnology Aging Research Center (CeSI)ldquoG drsquoAnnunziordquo University of Chieti-Pescara Via dei Vestini 29 66100 Chieti Italy

2 Interuniversity Institute of Myology Italy3 Laboratory of Tissue Engineering Department of Clinical and Experimental Sciences School of Medicine University of BresciaViale Europa 11 25123 Brescia Italy

4 Section of Physiology and Physiopathology Department of Neuroscience Imaging and Clinical SciencesldquoG drsquoAnnunziordquo University of Chieti-Pescara Via dei Vestini 31 66013 Chieti Italy

Correspondence should be addressed to Maria A Mariggio mariggiounichit

Received 24 April 2014 Revised 10 September 2014 Accepted 10 September 2014


Copyright copy 2015 Caterina Morabito et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

We propose a human-derived neuro-glial cell three-dimensional in vitro model to investigate the effects of microgravity on cell-cell interactions A rotary cell-culture system (RCCS) bioreactor was used to generate a modelled microgravity environment andmorphofunctional features of glial-like GL15 and neuronal-like SH-SY5Y cells in three-dimensional individual cultures (monotypicaggregates) and cocultures (heterotypic aggregates) were analysed Cell survival was maintained within all cell aggregates over 2weeks of cultureMoreover compared to cells as traditional staticmonolayers cell aggregates cultured undermodelledmicrogravityshowed increased expression of specific differentiation markers (eg GL15 cells GFAP S100B SH-SY5Y cells GAP43) andmodulation of functional cell-cell interactions (eg N-CAM and Cx43 expression and localisation) In conclusion this culturemodel opens a wide range of specific investigations at the molecular biochemical and morphological levels and it represents animportant tool for in vitro studies into dynamic interactions and responses of nervous system cell components to microgravityenvironmental conditions

1 Introduction

Microgravity modulates numerous features and functionsof biological organisms through its effects on physical phe-nomena such as hydrostatic pressure in fluid-filled com-partments sedimentation of organelles and convection pro-cesses of flow and heat These physical parameters can inturn directly and indirectly influence cellular and tissuemorphology metabolism and signalling and consequentlya wide range of cell functions [1] Several years ago it wasproposed that gravity is involved in embryonic developmentthrough effects on morphogenesis and organogenesis of thecentral nervous system and on sensory organs in inverte-brates and vertebrates In particular when amphibian eggs

were fertilised in vivo or in vitro under microgravity con-ditions some abnormalities during embryonic developmentwere observed even if compensatory mechanisms producednearly normal larvae [2] Also during space flight signsof neurophysiological impairment have been observed forastronauts although few studies have been carried out toinvestigate such effects on the nervous system in particularat the cellular level [3]

Recently Pani and colleagues reported that neuronalmonolayers showed alterations in morphology and viabilitywhen exposed to short- and middle-term simulated micro-gravity in the random positioning machine while long-termexposures revealed high adaptation of single neurons to thenew gravity conditions [4] Also other neuronal cell models



showed morphological andor cytoskeletal alterations whenexposed to simulated weightlessness or during changinggravity [5 6] These effects appeared conditioned by thepresence of microgravity conditions and after short-termexposures under ground-conditions the cells were able tofully recover their features and the ability to form adherentmonolayer cultures [4 7]

Traditional monolayer cell cultures that are kept understatic conditions (two-dimensional (2D) cell culture) haveprovided great advances in our understanding of the phys-iological regulatory processes of single cells On the otherhand the intrinsic complexity of cell-cell extracellular sig-nalling and the remarkable plasticity in the compositionand structure of the extracellular matrix have made it verydifficult to study these interactions using conventional cell-culture techniques For these reasons advanced methods areneeded to grow cells while maintaining their native three-dimensional (3D) cytoarchitecture and the specific tissue-like microenvironment Interestingly 3D cultures have beenshown to favour the maintenance of tissue-specific pheno-types and tissue-like cytoarchitectureHowever an importantlimitation for long-term culture in three dimensions is thelow diffusion of oxygen and nutrients and the absence of ablood supply to the deeper parts of the tissue construct Thisis particularly the case for neural cells and it can result in theappearance of a central core of dead cells [8 9]

In the 1990s after the beginning of themany internationalspace programmes attempts were made to grow 3D cellcultures or tissue explants in particular microenvironmentsto test the effects of reduced gravity Major efforts have beenaddressed to the building of a system that can reproducea tissue-like microenvironment in vitro and to study thecytoskeletal and nuclear matrix protein interactions duringcell exposure to simulated microgravity as is present in space[10] Engineers at the US National Aeronautics and SpaceAdministration (NASA) devised a rotating bioreactor whichis a useful device for culturing cells on Earth as well as inspace Briefly thismonoaxial clinostat (the rotary cell-culturesystem (RCCS) bioreactor) is a horizontally rotating andfluid-filled culture vessel that is equippedwith a gas-exchangemembrane that optimises the oxygen supply to the biologicalsamples Without air bubbles or air-liquid interface the fluiddynamic conditions inside the culture chamber generate alaminar flow state that greatly reduces shear stress and turbu-lence which are hazardous for cell survival These dynamicconditions provided by the RCCS bioreactor favour spatialcolocalisation and three-dimensional assembly of single cellsinto aggregates [11] The rotational speed of the culturechamber can be modified to set conditions in which the 3Dcell constructsaggregates also rotate around their own axesfurther providing an efficient high mass transfer of nutrientand wastes When cultured cell aggregates grow in sizethe rotational speed of the culture vessel can be increasedto compensate for the increased sedimentation rates Theoperational conditions of the RCCS bioreactor can also beadjusted so that the gravitational vectors are randomised upto reach a modelled microgravity state [12 13] In this way3D biological samples can remain in a constant orientationwith respect to the chamber wall and move in near-solid

body rotation with the fluid thus fulfilling the requirementsneeded to successfully model microgravity conditions [14]

In the present study we aimed to develop a 3D dynamicin vitro neuroglial coculture system to evaluate the capacityof the cells to reproduce at least in part neuronal featuresTo this end we used two well-characterised cell lines GL15and SH-SY5Y cells which are astrocyte-like and neuronal-like cells respectively The human glioblastoma GL15 cellline is an established in vitro astrocyte model that hasbeen functionally characterised by our group and others[15 16] and these express a typical astroglial phenotypeand functions The human neuroblastoma-derived SH-SY5Ycells are a widely used and well-characterised neuronalcell model that has been extensively used for in vitroneurotoxicity testing and has been shown to differentiatetowards either adrenergic or cholinergic phenotypes [17ndash20] In addition the human origin of these cell lines makesthem an appealing model for basic in vitro research studiesThus to develop astrocyte-like or neuronal-like in vitromodels 3D monotypic cultures (GL15 cells only or SH-SY5Ycells only) were established in a RCCS bioreactor Of noteit has been demonstrated that cell-cell interactions as forexample those between glial cells and neurons are crucialfor both glial and neuronal differentiation and developmentalprocesses as well as for response to neural injury [21 22]For these reasons we also established 3D neuronalglialheterotypic cultures (cocultures) to more closely reproducethe in vivo microenvironment of the nervous tissue andto bridge the gap between in vitro systems and animalmodels These analyses were also performed under modelledmicrogravity when the 3D cell aggregates were sufficientlygrown in size to adjust to the operational conditions of theRCCS bioreactor so as to reach a state of vector-averagedmicrogravity Under such conditions their cell morphologyviability and functional features were analysed and com-pared


All of the reagents for cell culturewere fromLife Technologies(Milan Italy) The plasticware was from BD Falcon (SaccoMilan Italy)

21 Cell Culture The SH-SY5Y cell line (from the EuropeanCollection of Cell Cultures supplied through Sigma-AldrichUK) and the GL15 cell line were both cultured in Dulbeccorsquosmodified Eaglersquos medium (DMEM) with 10 foetal bovineserum 100 IUmL penicillin 100 120583gmL streptomycin and1mM glutamine The cells were amplified in monolayersand detached for subculturing using 005 trypsin and002 EDTA SH-SY5Y and GL15 cell cultures used forexperimental assays were prepared by seeding cells in T75Falcon flasks to form 2D static monolayer cultures or in theRCCS bioreactor to establish 3D cultures subjected to micro-gravity Both culture models were cultured in DMEM with10 foetal bovine serum 100 IUmL penicillin 100 120583gmLstreptomycin and 1mM glutamine and maintained in thesame incubator (5 CO

2 at 95 humidity) for the same

times and the medium was refreshed twice a week


22 3D Culture in the RCCS Bioreactor The RCCS biore-actor (Synthecon Houston USA) can generate a specialmicroenvironment where highmass transfer is achieved withlow shear stress It is equipped with a cylindrical growthchamber that contains an inner corotating cylinder with agas-exchange membrane (a 55mL autoclavable slow-turninglateral vessel) where specific hydrodynamic and physicalconditions are attained The culture of cell spheroids wasperformed in this device in a 5 CO

2incubator at 95

humidity The horizontally rotating culture vessel was filledwith the complete medium (without air-liquid interface toreduce the shear stress) After a defined rotational speed wasreached the cells were cultured under Earth gravity in a nearlaminar fluid flow environment (ie a free-fall state) Undersuch conditions the cells grew in the form of 3Dmulticellularaggregates [23 24]

The cell-density seeding for both GL15 and SH-SY5Ycells was approximately 15 times 106 cellsmL The medium wasrefreshed twice a week For cocultures the SH-SY5Y and theGL15 cells were each seeded at a density of 075times 106 cellsmLThe rotational speed of the culture chamberwas initially set atbetween 6 rpmand 8 rpm and then it was gradually increasedas the multicellular aggregates increased in size to maintainthe aggregates in constant equilibrium (ie under free-fallconditions)

At the indicated times the cells were harvested andaccording to the experimental conditions required themulticellular aggregates were either included in Tissue-TekOCT compound (VWR International Srl USA) (for in situanalysis) or centrifuged at 2300 rpm for 5min at 4∘C and theresulting cell pellets were kept atminus80∘CuntilWestern blottingwas carried out

For the embedded aggregates slices (6120583m to 10 120583m)wereprepared with a CM1900 cryostat (Leica Milan Italy) andprocessed for cell viability assays or frozen at minus20∘C forfurther investigations

23 Morphological Analysis The frozen sections were left towarm up to room temperature and were subsequently incu-bated for 12min in Harrisrsquo haematoxylin solution washedtwice in water and incubated for 15 s in Eosin solution Afterwashing the sections were dehydrated and mounted withEukitt mounting medium (Electron Microscopy Sciences)The sections were examined under a Vanox optical micro-scope (Olympus Opera Zerbo Italy)

24 Cell Viability Assay The sections were incubated for15min at room temperature in a solution containing recombi-nant Annexin V conjugated to the Alexa 488 fluorophore andpropidium iodide (Vybrant kit 2 Life Technologies Italy)as described by the manufacturer Moreover to quantify thetotal number of cells in the aggregates 41015840-6-diamidino-2-phenylindole (DAPI) was added to this solution at a finalconcentration of 010 120583gmL The sections were mounted inProlong antifade medium (Life Technologies) and examinedunder an inverted fluorescence microscope (Axiovert ZeissArese Italy) equipped with an image analyser Photomicro-graphs were analysed with the ProImage+ and Scion Imagesoftware (httpproimagesoftwareinformercom andhttp

scion-imagesoftwareinformercom) to determine the cellviability

25 Immunostaining Assay The frozen GL15 and SH-SY5Ycells in OCT sections were fixed with 37 paraformaldehydeat room temperature for 30min Slices were then perme-abilised with 01 Triton X-100 at room temperature for15min and incubated for 1 h in 10 bovine serum albumin atroom temperature and then for 1 h at 37∘C with the primaryantibody followed by 1 h at 37∘C with either an Alexa 488-or an Alexa 633-conjugated secondary antibody (MolecularProbesMilan Italy) For double staining the second primaryantibody was incubated with the constructs after removal ofthe first Alexa 488-conjugated secondary antibody After theantibody incubation the cells were washed three times with01 Tween 20 each for 5min at room temperature Finallythe nuclei were stained with 1 120583gmL propidium iodide for30min After three washes with phosphate-buffered salinethey were mounted on coverslips and examined

Primary monoclonal mouse antibodies neuronal celladhesion molecule (N-CAM) tyrosine hydroxylase growthassociated protein 43 (GAP43) glial fibrillary acidic protein(GFAP) and S100B were from Sigma-Aldrich (Milan Italy)and connexin 43 (Cx43) was from Chemicon InternationalInc (Temecula CA USA)

The fluorescence images were obtained using a ZeissLSM510 META confocal system (Jena Germany) connectedto an inverted Zeiss Axiovert 200 microscope equipped witha Plan Neofluar oil-immersion objective (40x13 NA)

26 Western Blotting Frozen pellets of the GL15 andSH-SY5Y cell aggregates were lysed in cell lysis buffer(50mM Tris-HCl 100mM NaCl 50mM NaF 40mM 120573-glycerophosphate 5mM EDTA 1 Triton X-100 200120583Msodium orthovanadate 100 120583gmL phenylmethylsulfonyl flu-oride 10 120583gmL leupeptin 5 120583gmL pepstatin A 10 120583gmLbenzamidine and pH 74) After vortexing for 5 min thesamples were centrifuged at 1000 rpm for 10min at 4∘C ina microcentrifuge The protein content of each supernatantwas quantified colorimetrically (Bio-Rad Laboratories SrlMilan Italy) and aliquots containing 40 120583g protein wereadded to Laemmli buffer (8 SDS 10 glycerol 5 120573-mercaptoethanol 25mM Tris-HCl 0003 bromophenolblue and pH 65) and applied to and separated by SDS-PAGEon 7 to 10 SDS polyacrylamide slab gels Proteins wereelectroblotted onto hydrophobic polyvinylidene difluoridemembranes (Immobilon Millipore Milan Italy) using atank transfer system (Bio-Rad Laboratories Srl) Transferefficiency was verified by Ponceau red staining of the blotsand Coomassie blue staining of the gels The SH-SY5Y cellblots were incubated with the following mouse monoclonalantibodies anti-N-CAM (1 100 dilution Sigma-Aldrich)anti-tyrosine hydroxylase (1 1000 dilution Sigma-Aldrich)anti-GAP43 (1 1000 dilution Sigma-Aldrich) andor anti-Cx43 (1 1000 dilution Chemicon) The GL15 cell blots wereincubated with the following mouse monoclonal antibodiesanti-glial fibrillary acidic protein (GFAP 1 500 dilutionSigma-Aldrich) anti-S100B (1 500 dilution Sigma-Aldrich)andor also Cx43 (1 1000 dilution Chemicon) These were


then detected by chemiluminescence (ECL plus GE Health-care) Moreover after a membrane-stripping procedure theGL15 and SH-SY5Y cell membranes were immunostainedwith a mouse monoclonal anti-actin antibody (1 1000 dilu-tion Sigma-Aldrich)

3 Results

31 Cell Aggregates in the RCCS BioreactorMorphology and Viability

311 GL15 Cells Initial experiments were performed toestablish the most suitable protocol to prepare the cell aggre-gates GL15 cells were incubated in the RCCS bioreactor aspreinduced cell clusters or as homogeneous cell suspensionsthat were left to spontaneously aggregate The preinducedaggregates were obtained using the hanging drop method(see [24]) Both types of aggregates were maintained underconditions of microgravity in the RCCS bioreactor for upto 2 weeks The single cells spontaneously aggregated within48 h of culture although some features of the spontaneouscell aggregates were different compared to the preinducedaggregates

The preinduced aggregates provided relatively uniformclusters while the spontaneously formed aggregates appearedmore irregular in shape In addition after 2 weeks inthe RCCS bioreactor the spontaneously formed aggregatesshowed a trend (not significant) towards a greater meanarea (298 plusmn 026mm2) compared to that of the preinducedaggregates (189 plusmn 128mm2) (Figure 1)

The cell viability in the preinduced aggregates and thespontaneously formed aggregates was also assessed after 2weeks in the RCCS bioreactor to determine the apoptotic orthe necrotic cells (Figure 2) To this aim the cells were testedto measure early apoptosis by detecting phosphatidylserineexpression revealed by Annexin V binding or necrosis bymembrane permeability to the propidium iodide (PI) vitaldye Cells positive to Annexin V green fluorescence signalare known to be apoptotic cells while those positive to PIred fluorescence signal are necrotic cells the absence of greenor red signal and the nuclear staining with DAPI revealedviable cells The image analyses of stained cells revealedthat some apoptotic cells (Figure 2(a) green fluorescence)were evident at a similar extent in preinduced and spon-taneous GL15 aggregates (Figure 2(b)) A relevant amountof necrotic cells (Figure 2(a) red fluorescence) was presentin preinduced aggregates compared to spontaneous ones inwhich necrotic cells were nearly absent (Figure 2(b)) Thesedata revealed the presence of possible stress conditions inthe preinduced aggregates while the spontaneously formedaggregates showed cells that were in a more healthy stateThis cell stress might be the result of hypoxic processes inthe central core of the preinduced aggregates in particularpotentially due to the static conditions encountered in thehanging drops For this reason the rest of the investigationsused only the spontaneously formed cell aggregates andthose formed by the GL15 cells are henceforth referred to asthe G-aggregates

312 SH-SY5Y Cells Following the same proceduredescribed above for the formation of the G-aggregatesthe SH-SY5Y cells were cultured to form spontaneousaggregates in the RCCS bioreactor for up to 2 weeks andtheir cell morphology and viability were then assessedTheseaggregates formed by the SH-SY5Y cells are henceforthreferred to as the S-aggregates

At the end of the incubation period the S-aggregatesshowed variable and irregular shapes with a mean area of268 plusmn 013mm2 In addition there were very low levels ofapoptotic and necrotic cells (Figure 2(b)) which indicatedthat these 3D dynamic culture conditions are a suitablemethod to sustain cell viability also for neuronal-like cells

32 Qualitative Analysis of Phenotype-Specific Markers

321 GL15 Cells To analyse the expression of the GL15cell specific phenotype in the G-aggregates cultured in theRCCS bioreactor for up to 2 weeks immunostaining for glialmarkers was carried out The G-aggregates showed glial-cell-specific protein expression similar to that observed inthe GL15 cells cultured as monolayers under 2D conditions(Figure 3) The G-aggregates and the GL15 cells cultured asmonolayers both showed cytoplasmic localisation of GFAPand S100B as two markers of the glial cytoskeleton (Figures3(a)ndash3(d))

Cell interactions due to gap-junction-mediated intercel-lular communication have been shown to have crucial rolesin the regulation of the glial-cell network and nervous systemfunctions [25] For this reason the expression of Cx43 wasalso investigated as Cx43 is the main gap-junction proteinexpressed by astrocytes As shown in Figure 3(e) the GL15cells grown in two dimensions expressed Cx43 near theplasmalemma at cell-cell contact areas and in the cytoplasmA similar distribution was also seen for Cx43 in the G-aggregates (Figure 3(f))

322 SH-SY5Y Cells We characterised the phenotypeexpressed by the SH-SY5Y cells in the S-aggregatesmaintained in the dynamic 3D culture in the RCCSbioreactor for 2 weeks by determining the expression of theneuronal specific markers N-CAM GAP43 and tyrosinehydroxylase Immunofluorescence analysis revealed thatN-CAM in the S-aggregates was localised towards theplasma membrane and near cell-cell contact areas thusresembling its distribution in the SH-SY5Y cells culturedin 2D monolayers (Figures 4(a) and 4(b)) which showedcell-cell adhesion interactions GAP43 is involved in neuriteoutgrowth and neuronal plasticity [26] and in SH-SY5Ycell monolayers it was localised into neurite-like processes(Figure 4(c)) In the S-aggregates GAP43 was localised in thecytoplasmic compartment (Figure 4(d)) The distributionof tyrosine hydroxylase (TH) which is a rate-limitingenzyme in dopaminenorepinephrine synthesis [27] was inthe cytoplasm under both of these cell-culture conditions(Figures 4(e) and 4(f))


Preinduced Spontaneous

Spontaneous

0

1

2

3

4

Agg

rega

te ar

ea (m

m2 )

GL15-composed aggregates

Preinduced

Figure 1 GL15 cell aggregate morphology Representative images and quantification of sections from preinduced and spontaneously formedGL15 aggregates (as indicated) Data are means plusmn SEM 119899 = 15 for the averaged areas of the aggregate sections calculated using the ImageJsoftware (httpimagejnihgovij)

33 Quantitative Analysis of Phenotype-Specific Markers

331 GL15 Cells The differentiation status of cells is charac-terised not only by marker localisation but also by markerexpression levels To evaluate potential quantitative dif-ferences between the G-aggregate modelled microgravity-exposed cultures and the GL15 cells as 2D static monolayercultures the expression levels of the GFAP S100B and Cx43proteins were determined byWestern blotting (Figure 5)TheG-aggregates showed increased levels of GFAP S100B andCx43 after the first 48 h of culture These levels graduallydecreased over the following 2 weeks when those of S100Band Cx43 were similar to those observed in the GL15 cells as2D static cultures while those of GFAP remained increasedin the G-aggregates (Figure 5)

332 SH-SY5Y Cells Western blotting carried out for the S-aggregates showed that N-CAM-140 and GAP43 expressionlevels were increased during the incubation compared to theSH-SY5Y cells as 2D static monolayer cultures (Figure 6) Inparticular for the S-aggregates N-CAM-140 reached a peakafter 2 weeks while GAP43 peaked after 48 h ConverselyN-CAM-180 and tyrosine hydroxylase did not significantly

change in the S-aggregates compared to SH-SY5Y cells as 2Dstatic cultures (Figure 6)

34 Coculture of GL15 and SH-SY5Y Cells in the RCCSBioreactor The SH-SY5Y cells were also cocultured with theGL15 cells in the RCCS bioreactor with the aim to reestablisha more neural-like microenvironment and thus to be closerto in vivo conditions Initial experiments were carried outto determine if it was possible to establish viable GL15 plusSH-SY5Y cocultures in the RCCS bioreactor henceforthreferred to as GS-aggregates GL15 and SH-SY5Y cells werethus cocultured in the RCCS bioreactor at a 1 1 ratio forup to 2 weeks At the end of this period the sizes of theGS-aggregates were similar to those of the monotypic G-aggregates and S-aggregates (Figure 7) and although the S-aggregates appeared smaller than the others these differencesdid not reach significance Cell viability assays also showedthat the GS-aggregates had low levels of apoptotic andnecrotic cells (data not shown)

To characterise the cell phenotype in theseGS-aggregatesimmunostaining was carried out for N-CAM GFAP andCx43 These coculture conditions induced the establishment


SH-SY5Y

spontaneously

formed

aggregates in

GL15

spontaneously

formed

aggregates in

GL15

preinduced

aggregates in

hanging drop

DAPI PI

RCCS

RCCS

Annexin V

(a)

Total cell number Apoptotic cells ()

Necrotic cells ()

GL15 preinduced aggregates in hanging drop

512 16 19

914 14 007

491 7 1

SH-SY5Y spontaneously

formedaggregates in

GL15 spontaneously

formedaggregates in

RCCS

RCCS

(b)

Figure 2 Cell viability assay (a) Representative images of preinduced and spontaneously formed GL15 aggregates and spontaneously formedSH-SY5Y aggregates (as indicated) Aggregates were stained with DAPI (blue) Annexin V-Alexa 488 (green) and propidium iodide (PIred) DAPI-positiveAnnexin V-Alexa 488-negativePI-negative cells are healthy DAPI-positiveAnnexin V-Alexa 488-positivePI-negativeand PI-positive cells are considered apoptotic (Annexin V arrowheads) DAPI-positiveAnnexin V-Alexa 488-negativePI-positive cellsare necrotic (PI arrowheads) (b) Quantification of apoptotic and necrotic cells in aggregate sections Data derived from 3 independentexperiments


GFAP

2D

(a)

GFAPPI

RCCS

(b)

S100B

(c)

S100BPI

(d)

Cx43

(e)

Cx43PI

(f)

Figure 3 Glial marker localisation in GL15 cells Representative confocal images of GL15 cells cultured as a monolayer (2D (a) (c) and (e))and under the modelled microgravity (RCCS bioreactor (b) (d) and (f)) and immunostained with anti-GFAP ((a) and (b)) anti-S100B ((c)and (d)) and anti-Cx43 ((e) and (f)) antibodies (as indicated)The RCCS G-aggregate sections were also stained with propidium iodide (PI)Insets in (b) (d) and (f) show image magnification Scale bars 25120583m

of GS-aggregates that contained both glial-like and neuronal-like cell phenotypes After the 2 weeks of culture in the RCCSbioreactor these GS-aggregates showed specific fluorescencesignals for astrocyte (GFAP-positive) andneuronal (N-CAM-positive) phenotypes (Figure 8)

The GS-aggregates were double-stained for N-CAM andCx43 These N-CAM-specific and Cx43-specific fluorescentsignals revealed a particular distribution of these proteinswhereby even if colocalisation of the N-CAM and Cx43patterns was not evident possible heterotypic cell-cell inter-actions could not be excluded In particular within the GS-aggregates N-CAM localised to the peripheral areas of thecells while Cx43-specific fluorescent spots appeared to be

sparsely distributed which indicated a low level of cell-cellfunctional interactions (Figure 9) In addition in the sameGS-aggregates there were also evident N-CAM-negativeandor Cx43-negative cells which indicated potential differ-ent cell activities due to different protein expression levels

Western blotting of N-CAM and Cx43 expression levelsrevealed that in the GS-aggregates the monomeric Cx43protein (43 kDa) was downregulated during the RCCS biore-actor incubation Interestingly homogenates from the GS-aggregates showed a Cx43-positive band at 86 kDa whichdemonstrates the presence of a dimeric form of Cx43 whichwas highly expressed in the initial phases of the coculture(over the first 24 h) and which significantly decreased over


2D

N-CAMPI

(a)

N-CAMPI

RCCS

(b)

GAP43PI

(c)

GAP43PI

(d)

THPI

(e)

THPI

(f)

Figure 4 Neuronal marker localisation in SH-SY5Y cells Representative confocal images of SH-SY5Y cells cultured as a monolayer (2D (a)(c) and (e)) and under the modelled microgravity (RCCS bioreactor (b) (d) and (f)) and immunostained with anti-N-CAM ((a) and (b))anti-GAP43 ((c) and (d)) and anti-tyrosine hydroxylase (TH) ((e) and (f)) antibodies (as indicated) All of the cells were also stained withpropidium iodide (PI) Insets show image magnification Scale bars 20120583m


0

1

2

0

1

2

3

4

0

1

2

3

GFAP 50 kDa

S100B 20kDa

Cx43 43kDa

2D 24h 48h 96h 2w

2D 24h 48h 96h 2w2D 24h 48h 96h 2w 2D 24h 48h 96h 2w

GFAP S100B Cx43

Ratio

Ratio

Ratio

RCCS

lowast

lowast


lowast

Figure 5 Expression of glial cell markers RepresentativeWestern blotting and quantification of the levels of GFAP S100B and Cx43 in GL15cells cultured in 2Dmonolayers and in the RCCS bioreactor for 24 h 48 h 96 h and 2 weeks (2w) Data are from densitometric ratio analysesas means plusmn SEM from 3 independent experiments lowast119875 lt 005 versus 2D monolayers

0

1

2

3

4

0

1

2

3

0

1

2

3

N-CAM180 kDa140kDa

TH 70kDa

2D 24h 48h 96h 2w

2D 24h 48h 96h 2w 2D 24h 48h 96h 2w 24h 48h 96h 2w

THN-CAM 140

N-CAM 180

Ratio

Ratio

Ratio

lowastlowastlowast

RCCS

GAP43

C0

GAP43 46kDa

Figure 6 Expression of neuronal cell markers Representative Western blotting and quantification of the levels of N-CAM GAP43 andtyrosine hydroxylase (TH) in SH-SY5Y cells cultured in 2D monolayers and in the RCCS bioreactor for 24 h 48 h 96 h and 2 weeks (2 w)Data are densitometric ratio analyses as means plusmn SEM from 3 independent experiments lowast119875 lt 005 and lowastlowast119875 lt 001 versus 2D monolayers


GL15 SH-SY5Y Coculture0

1

2

3

4

Agg

rega

te ar

ea (m

m2 )

Figure 7 Cell aggregate sizes Quantification of section area ofGL15 SH-SY5Y and cocultured (GL15 plus SH-SY5Y) cell aggre-gates (as indicated) Data aremeans plusmn SEM (119899 = 15) for the averagedareas of the aggregate sections calculated using the ImageJ software(httpimagejnihgovij)

the 2 weeks of the GS-aggregates in the RCCS bioreactor(Figure 10) The N-CAM isoform expression pattern showeda slight but not significant decrease in N-CAM-180 levelsand a significant increase in N-CAM-140 levels (Figure 10)which resembled the N-CAM-140 increase observed in theS-aggregate homogenates

4 Discussion

There are evidences available showing that microgravity canaffect the functioning of the nervous system although thepossible physiological mechanisms of these effects remaindifficult to determine [7 28] Such difficulties in investiga-tions into microgravity effects are mainly due to the poormodels that are available either because of their high costand low availability (eg spaceflight) or because they are littlerepresentative of truemicrogravity conditions (eg hindlimbsuspensiondisusemodel) Among the ground-basedmodelsin vitro culture of cellstissues within clinorotation-basedsystems (eg random positioning machine RCCS bioreac-tor) represents a reasonable alternative to spaceflight TheRCCS bioreactor in particular was initially developed byNASA engineers to maintain cells in culture during spacemissions and to counteract the forces faced during shuttlelaunch and landing The RCCS bioreactor was further usedto maintain cells in dynamic 3D culture on the ground andbecause of its particular properties the RCCS bioreactoralso allows the modelling of microgravity on the groundSetting standardised parameters it is possible to also promotethe colocalisation of cells the establishment of cell-cellcontacts and consequently the spontaneous formation ofmulticellular aggregates [11 13] Moreover the rotation speedcan also be regulated in such away that it is possible to reach avector-averaged gravity that simulates low-gravity conditions[14]

In the present study we designed and investigated a pow-erful human-derived 3D organotypic-like model of nervoussystem tissue The experimental strategy was to study this

3D cell aggregation in terms of the cell phenotypes followingshort-term culture (up to 48 h as a time that allows theformation of multicellular aggregates) and long-term (upto 2 weeks) culture to analyse the effects of this modelledmicrogravity system on cell behaviour However apart fromthe effects related to microgravity the development of areliable neuroglia cell in vitro model is of great interest forbasic and clinical research in the field of the nervous systemThus we developed astrocyte-like and neuron-like in vitromodels here as 3D monotypic (GL15 cells only SH-SY5Ycells only) and heterotypic (cocultures of both GL15 and SH-SY5Y cells) cell cultures in the RCCS bioreactor

The particular dynamic conditions in the RCCS bioreac-tor have been shown to favour the differentiated phenotypeexpression for numerous cell and tissue types [13 24 29ndash31]In our hands over 48 h of culture these optimal dynamicconditions favoured spontaneous formation of healthy mul-ticellular aggregates according to the cell type consideredas demonstrated by the low cell death in these spontaneouscell aggregates The survival of these G-aggregates and S-aggregates and also of the GS-aggregates was assessed for upto 2 weeks in the RCCS bioreactor cultures and the dataconfirm the absence of significant necrosis in their centralcores in contrast to what has been reported in the literaturefor similar static culture conditions [32] This evidence sup-ported our choice to use the spontaneously formed aggregatemethod as this allowed the random distribution of the cellsinside the aggregates which is a feature that is particularlyimportant for the establishment of the heterotypic coculturemodel

Under our 3D cell culture conditions in growth mediumthe GL15 cells showed an astrocyte-like phenotype withthe expression of the glial-specific markers GFAP [33] andS100B Interestingly under these conditions Cx43 expressionwas also evident in these G-aggregates These data confirmthe importance of cell-cell interactions in the regulation ofphenotypic expression The modulation of Cx43 expressionmight be related to the formation of these G-aggregates in the3D culture During the first phase of G-aggregate formationthere was upregulation of Cx43 expression In a previousstudy we showed that these GL15 cells express Cx43 andform junctional channels where the permeability is directlyrelated to the cell proliferation rate as it decreased whentheir differentiated status was reached [16] In the presentstudy this transient upregulation of Cx43 duringG-aggregateformation might support the hypothesis that Cx43 has acrucial role and function in cellular aggregation in additionto its well-known involvement in differentiation processes[34] This hypothesis was also supported by Cotrina andcolleagues who demonstrated a role for Cx43 hemichannelsin cellular adhesion of C6 glioma cells [35]

The optimal dynamic culture conditions provided by theRCCS bioreactor were also demonstrated by the favouredexpression of neuronal-specificmarkers by the SH-SY5Y cellsin the S-aggregates such as tyrosine hydroxylase GAP43and N-CAM The expression levels of tyrosine hydroxylaseappeared similar in both the 2D and the 3D cultures atall of the times tested which demonstrated the adrener-gic phenotype that was expressed by these S-aggregates


N-CAM N-CAMGFAPGFAP

Figure 8 Localisation of glial and neuronal cell markers in the GS-aggregates Representative confocal images of GS-aggregates culturedunder the modelled microgravity for 2 weeks and immunostained with anti-N-CAM and anti-GFAP antibodies (as indicated) Insets showimage magnification Scale bars 20120583m

N-CAM N-CAMCx43PICx43

Figure 9 Localisation of cell-cell interaction markers in the GS-aggregates Representative confocal images of GS-aggregates cultured undermodelled microgravity for 2 weeks and immunostained with anti-N-CAM and anti-Cx43 antibodies (as indicated) The GS-aggregatesections were also stained with propidium iodide (PI right) Insets show image magnification Scale bars 10 120583m

GAP43 expression increased during the cell aggregation(48 h) which confirms active cell-cell interactions with thecytoskeletal modifications shown by GAP43 regulation Thestabilisation of these cell aggregates is supported by theincrease in N-CAM expression

During the long-termexposure tomodelledmicrogravityspecific protein expression was differently regulated in thecell aggregates Even after 2 weeks under culture in themodelled microgravity in the G-aggregates the glial-specificand functional markers (ie GFAP S100B and Cx43) showedlocalisation patterns thatwere similar to those observed in themonolayers under normal gravity conditions Interestinglyunder microgravity S100B and Cx43 expression levels in theG-aggregates were downregulated over two weeks as com-pared to those in the G-aggregate cultures at 48 h exposurewhereas there was a slight transient although not significanteffect on GFAP expression

The exact physiological roles of GFAP in astrocytesremain incompletely understood although they appear to beinvolved in cell-shape maintenance nervous system cytoar-chitecture mechanical stability and synaptic function [36]On the other hand it is well-known that Cx43 modulationis involved in neuronal development and plasticity [37] and

that S100B is expressed and also secreted by astrocytes andcan thus be an extracellular mediator of cell signalling [38]This evidence supports the hypotheses that the microgravitycan affect not only cell shape but also cell function

In the S-aggregates the modelled microgravity condi-tions did not have any significant effects on the localisationof N-CAM and tyrosine hydroxylase but they were shownto induce a switch of the GAP43 protein from the neurite-like processes to the cytoplasmic compartment In additionthe microgravity induced a slight although not significantdecrease in the expression levels of GAP43 and tyrosinehydroxylase while it had no effect on the expression of theN-CAM-180 isoform but significantly increased the expressionof the N-CAM-140 isoform which has been shown to havea key role in neuronal survival and signal transduction [3940] This suggests the involvement of N-CAM-140 duringthis modelled microgravity exposure that could promote asignificant degree of neuronal remodelling and survival

It has been previously reported that in neuroglialcocultures the neurons induce a reduction in astrocyteproliferation [41] In particular this effect was mediated bymembrane-membrane interactions between the neurons andthe astroglia in vitro and raised the possibility that membrane


00

05

10

15

20

N-CAM 180 kDa

N-C

AM

180

N-CAM 140kDa

N-C

AM

140

Cx43 86kDa

(dimer)

Cx43 43kDa

(monomer)

24h 2w

24h2w

GS-aggregates

Ratio

Cx43

(dim

er)

Cx43

(mon

omer

)

lowast

lowast

lowast

Figure 10 Expression levels of functional markers in the GS-aggregates Representative Western blotting and quantification of the levels ofN-CAM and Cx43 in GS-aggregates cultured in the RCCS bioreactor for 24 h and 2 weeks (2 w) Data are densitometric ratio analyses asmeans plusmn SEM from 3 independent experiments lowast119875 lt 005 for 2-week versus respective 24 h cultures

elements involved in glial cell growth regulation includeneuron-glial interaction molecules [41] In our neuron-likeand glial-cell-like coculture we focused our attention on N-CAM and Cx43 expression as these participate in importantintercellular signal interactions The GL15 cells used heredid not express N-CAM isoforms (data not shown) and theimmunofluorescence signals might reveal SH-SY5Y homo-typic interactions even if N-CAM heterotypic interactionscould not be excluded (such as N-CAM-integrins) Howeverthe N-CAM expression pattern in homogenates from GS-aggregates resembled that for the S-aggregates with anincrease in the expression of N-CAM-140 one of the threemain isoforms of N-CAM that is implicated in regenerationand remodelling of the nervous system [42]

Gap-junction-mediated intercellular communicationamong astrocytes has long been thought to contributeto tissue homeostasis in the brain [43] Cx43 has beenused as a marker to investigate neuron-glia interactions[22] Astrocytes express Cx30 and Cx43 which can formhomotypic (Cx43Cx43 andCx30Cx30) but not heterotypicjunctions [44] Interestingly in the homogenates from theGS-aggregates in addition to the classical 43 kDa formof Cx43 a dimeric form of Cx43 (ie 86 kDa) was alsoexpressed This Cx43 dimeric form has been related to

stress conditions In other models oxidative stress status hasbeen related to the appearance of such a higher molecularweight band for Cx43 which suggests that this representsan aggregated form of Cx43 [45] Under our conditions thepresence of this dimeric form of Cx43 might reveal a firstphase of impact between the neuronal-like and the glial-likephenotypes subsequently these Cx43 forms significantlydecreased when the GS-aggregates reached dynamic adaptiveconditions

5 Conclusions

In conclusion the evidence presented here suggests that the3D laminar flow the high mass transfer and the low shear-stress microenvironment generated by the RCCS bioreactorrepresent optimal conditions for the well-being of monotypicneural-like and glial-like cells as well as for heterotypicaggregates and for long-term culture Moreover such modelsystem can reproduce 3D cell-cell interactions that are similarto those under in vivo conditions [46] and can mimic themicrogravity conditions of exposure Our data highlight howsome phenotypic markers of monotypic and heterotypicneuroglia culture models can be influenced by microgravity


The data presented here open a wide range of specificinvestigations in terms of cell transduction pathways cell-cellinteractions and signalling and heterotypic culture biologyand the cell models we have described and analysed hererepresent important tools in the study of in vitro biologicaland pathological processes of the nervous system




Caterina Morabito and Nathalie Steimberg equally con-tributed to this study

Acknowledgments

The authors wish to thank Jennifer Boniotti (UNIBS) fortechnical support This study was supported by ASI 2013 Dec1342013 to MAM and by G drsquoAnnunzio University researchfunds to MAM

References

[1] G Vunjak-Novakovic N Searby J de Luis and L E FreedldquoMicrogravity studies of cells and tissuesrdquo Annals of the NewYork Academy of Sciences vol 974 pp 504ndash517 2002

[2] A-M Duprat D Husson and L Gualandris-Parisot ldquoDoesgravity influence the early stages of the development of thenervous system in an amphibianrdquo Brain Research Reviews vol28 no 1-2 pp 19ndash24 1998

[3] B M Uva M A Masini M Sturla et al ldquoMicrogravity-induced apoptosis in cultured glial cellsrdquo European Journal ofHistochemistry vol 46 no 3 pp 209ndash214 2002

[4] G Pani N Samari R Quintens et al ldquo Morphological andphysiological changes in mature in vitro neuronal networkstowards exposure to short- middle- or long-term simulatedmicrogravityrdquo PLOS ONE vol 8 no 9 Article ID e73857 2013

[5] R Gruener and G Hoeger ldquoVector-averaged gravity altersmyocyte and neuron properties in cell culturerdquo Aviation Spaceand Environmental Medicine vol 62 no 12 pp 1159ndash1165 1991

[6] H Rosner T Wassermann W Moller and W Hanke ldquoEffectsof altered gravity on the actin and microtubule cytoskeleton ofhuman SH-SY5Y neuroblastoma cellsrdquo Protoplasma vol 229no 2ndash4 pp 225ndash234 2006

[7] A Crestini C Zona P Sebastiani et al ldquoEffects of simulatedmicrogravity on the development and maturation of disso-ciated cortical neuronsrdquo In Vitro Cellular amp DevelopmentalBiologyminusAnimal vol 40 no 5-6 pp 159ndash165 2004

[8] L Lossi S Alasia C Salio and A Merighi ldquoCell deathand proliferation in acute slices and organotypic cultures ofmammalian CNSrdquo Progress in Neurobiology vol 88 no 4 pp221ndash245 2009

[9] K Rambani J Vukasinovic A Glezer and S M PotterldquoCulturing thick brain slices an interstitial 3D microperfusionsystem for enhanced viabilityrdquo Journal of NeuroscienceMethodsvol 180 no 2 pp 243ndash254 2009

[10] L E Freed and G Vunjak-Novakovic ldquoSpaceflight bioreactorstudies of cells and tissuesrdquo Advances in Space Biology andMedicine vol 8 pp 177ndash195 2002

[11] D A Wolf and R P Schwarz ldquoRP analysis of gravity-inducedparticle motion and fluid perfusion flow in the NASA-designedrotating zero-heaed-space tissue culture vesselrdquo NASA Techni-cal Paper 3143 1991

[12] C M Begley and S J Kleis ldquoThe fluid dynamic and shearenvironment in the NASAJSC rotating-wall perfused-vesselbioreactorrdquo Biotechnology and Bioengineering vol 70 no 1 pp32ndash40 2000

[13] T G Hammond and J M Hammond ldquoOptimized suspensionculture the rotating-wall vesselrdquo The American Journal ofPhysiologymdashRenal Physiology vol 281 no 1 pp F12ndashF25 2001

[14] P S Ayyaswamy and K Mukundakrishnan ldquoOptimal condi-tions for simulating microgravity employing NASA designedrotating wall vesselsrdquo Acta Astronautica vol 60 no 4ndash7 pp397ndash405 2007

[15] V Bocchini T Beccari C Arcuri L Bruyere C Fages and MTardy ldquoGlial fibrillary acidic protein and its encoding mRNAexhibit mosaic expression in a glioblastoma multiform cellline of clonal originrdquo International Journal of DevelopmentalNeuroscience vol 11 no 4 pp 485ndash492 1993

[16] M A Mariggio G Mazzoleni T Pietrangelo et al ldquoCalcium-mediated transductive systems and functionally active gapjunctions in astrocyte-like GL15 cellsrdquo BMC Physiology vol 1no 1 article 4 2001

[17] Y-T CheungW K-W Lau M-S Yu et al ldquoEffects of all-trans-retinoic acid on human SH-SY5Y neuroblastoma as in vitromodel in neurotoxicity researchrdquo NeuroToxicology vol 30 no1 pp 127ndash135 2009

[18] S Guarnieri R Pilla C Morabito et al ldquoExtracellular guano-sine and GTP promote expression of differentiation mark-ers and induce S-phase cell-cycle arrest in human SH-SY5Yneuroblastoma cellsrdquo International Journal of DevelopmentalNeuroscience vol 27 no 2 pp 135ndash147 2009

[19] M Miloso D Villa M Crimi et al ldquoRetinoic acid-inducedneuritogenesis of human neuroblastoma SH-SY5Y cells is ERKindependent and PKC dependentrdquo Journal of NeuroscienceResearch vol 75 no 2 pp 241ndash252 2004

[20] S Pahlman J C Hoehner E Nanberg et al ldquoDifferentiationand survival influences of growth factors in human neuroblas-tomardquo European Journal of Cancer Part A General Topics vol31 no 4 pp 453ndash458 1995

[21] T Fellin ldquoCommunication between neurons and astrocytesrelevance to the modulation of synaptic and network activityrdquoJournal of Neurochemistry vol 108 no 3 pp 533ndash544 2009

[22] A Vernadakis ldquoGlia-neuron intercommunications and synap-tic plasticityrdquo Progress in Neurobiology vol 49 no 3 pp 185ndash214 1996

[23] G Mazzoleni D Di Lorenzo and N Steimberg ldquoModellingtissues in 3D the next future of pharmaco-toxicology and foodresearchrdquo Genes and Nutrition vol 4 no 1 pp 13ndash22 2009

[24] N Steimberg J Boniotti and G Mazzoleni ldquo3D culture ofprimary chondrocytes and bonecartilage tissue explants insimulated microgravityrdquo inMethods in Bioengineering Alterna-tive Technologies to Animal Testing M Yarmush and R LangerEds Artech House 2010

[25] D A Goodenough and D L Paul ldquoGap junctionsrdquo Cold SpringHarbor Perspectives in Biology vol 1 no 1 Article ID a0025762009

[26] M I Mosevitsky ldquoNerve ending ldquosignalrdquo proteins GAP-43MARCKS and BASP1rdquo International Review of Cytology vol245 pp 245ndash325 2005


[27] D A Lewis D S Melchitzky and J W Haycock ldquoFourisoforms of tyrosine hydroxylase are expressed in human brainrdquoNeuroscience vol 54 no 2 pp 477ndash492 1993

[28] I B Krasnov ldquoGravitational neuromorphologyrdquo Advances inSpace Biology and Medicine vol 4 pp 85ndash110 1994

[29] X Chen H Xu C Wan M McCaigue and G Li ldquoBioreactorexpansion of human adult bone marrow-derived mesenchymalstem cellsrdquo Stem Cells vol 24 no 9 pp 2052ndash2059 2006

[30] G Mazzoleni F Boukhechba N Steimberg J Boniotti J MBouler and N Rochet ldquoImpact of dynamic culture in theRCCSTM bioreactor on a three-dimensional model of bonematrix formationrdquo Procedia Engineering vol 10 pp 3670ndash36752011

[31] B R Unsworth and P I Lelkes ldquoGrowing tissues in micrograv-ityrdquo Nature Medicine vol 4 no 8 pp 901ndash907 1998

[32] P Humphreys S Jones and W Hendelman ldquoThree-dimensional cultures of fetal mouse cerebral cortex in acollagen matrixrdquo Journal of Neuroscience Methods vol 66 no1 pp 23ndash33 1996

[33] G Moretto N Brutti V de Angelis C Arcuri and V BocchinildquoA time-dependent increase in glial fibrillary acidic proteinexpression and glutamine synthetase activity in long-termsubculture of the GL15 glioma cell linerdquo Cellular and MolecularNeurobiology vol 17 no 5 pp 509ndash519 1997

[34] T Nakase and C C G Naus ldquoGap junctions and neurologicaldisorders of the central nervous systemrdquo Biochimica et Biophys-ica ActamdashBiomembranes vol 1662 no 1-2 pp 149ndash158 2004

[35] M L Cotrina J H-C LIN and M Nedergaard ldquoAdhesiveproperties of connexin hemichannelsrdquo Glia vol 56 no 16 pp1791ndash1798 2008

[36] VMenetMGimenezYRibottaNChauvet et al ldquoInactivationof the glial fibrillary acidic protein gene but not that ofvimentin improves neuronal survival and neurite growth bymodifying adhesion molecule expressionrdquo Journal of Neuro-science vol 21 no 16 pp 6147ndash6158 2001

[37] R RozentalM Srinivas S Gokhan et al ldquoTemporal expressionof neuronal connexins during hippocampal ontogenyrdquo BrainResearch Reviews vol 32 no 1 pp 57ndash71 2000

[38] R Donato G Sorci F Riuzzi et al ldquoS100Brsquos double lifeintracellular regulator and extracellular signalrdquo Biochimica etBiophysica Acta vol 1793 no 6 pp 1008ndash1022 2009

[39] M A Mariggio C Morabito S Guarnieri A Gentile KKolkova and G Fano ldquoIgIII (270-280)-fragment-like H

2N-

DDSDEEN-COOH peptide modulates N-CAM expression viaCa2+-dependent ERK signaling during ldquoin vitro neurogenesisrdquordquoPeptides vol 29 no 9 pp 1486ndash1497 2008

[40] P S Walmod K Kolkova V Berezin and E Bock ldquoZippersmake signals NCAM-mediated molecular interactions andsignal transductionrdquoNeurochemical Research vol 29 no 11 pp2015ndash2035 2004

[41] M E Hatten ldquoNeuronal inhibition of astroglial cell prolifera-tion is membrane mediatedrdquo Journal of Cell Biology vol 104no 5 pp 1353ndash1360 1987

[42] L C B Roslashnn V Berezin and E Bock ldquoTheneural cell adhesionmolecule in synaptic plasticity and ageingrdquo International Jour-nal of Developmental Neuroscience vol 18 no 2-3 pp 193ndash1992000

[43] M Theis G Sohl J Eiberger and K Willecke ldquoEmergingcomplexities in identity and function of glial connexinsrdquoTrendsin Neurosciences vol 28 no 4 pp 188ndash195 2005

[44] J L Orthmann-Murphy M Freidin E Fischer S S Schererand C K Abrams ldquoTwo distinct heterotypic channels mediategap junction coupling between astrocyte and oligodendrocyteconnexinsrdquo Journal of Neuroscience vol 27 no 51 pp 13949ndash13957 2007

[45] C M L Hutnik C E Pocrnich H Liu D W Laird and QShao ldquoThe protective effect of functional connexin43 channelson a human epithelial cell line exposed to oxidative stressrdquoInvestigative Ophthalmology and Visual Science vol 49 no 2pp 800ndash806 2008

[46] E Fennema N Rivron J Rouwkema C van Blitterswijk andJ de Boer ldquoSpheroid culture as a tool for creating 3D complextissuesrdquo Trends in Biotechnology vol 31 no 2 pp 108ndash115 2013

Review ArticleThe Impact of Microgravity and Hypergravityon Endothelial Cells

Jeanette A M Maier1 Francesca Cialdai2 Monica Monici2 and Lucia Morbidelli3

1 Department of Biomedical and Clinical Sciences ldquoL Saccordquo Universita di Milano Via Gian Battista Grassi 74 20157 Milan Italy2 ASAcampus Joint Laboratory ASA Research Division Department of Experimental and Clinical Biomedical Sciences ldquoM SeriordquoUniversity of Florence Viale Pieraccini 6 50139 Florence Italy

3 Department of Life Sciences University of Siena Via A Moro 2 53100 Siena Italy

Correspondence should be addressed to Lucia Morbidelli morbidelliunisiit

Received 4 July 2014 Revised 20 October 2014 Accepted 4 November 2014


Copyright copy 2015 Jeanette A M Maier et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

The endothelial cells (ECs) which line the inner surface of vessels play a fundamental role in maintaining vascular integrity andtissue homeostasis since they regulate local blood flow and other physiological processes ECs are highly sensitive to mechanicalstress including hypergravity and microgravity Indeed they undergo morphological and functional changes in response toalterations of gravity In particular microgravity leads to changes in the production and expression of vasoactive and inflammatorymediators and adhesion molecules which mainly result from changes in the remodelling of the cytoskeleton and the distributionof caveolae These molecular modifications finely control cell survival proliferation apoptosis migration and angiogenesis Thisreview summarizes the state of the art on how microgravity and hypergravity affect cultured ECs functions and discusses somecontroversial issues reported in the literature

1 The Endothelium

The concept of endothelium as an inert barrier lining theinner side of blood vessels has been overcome by thefinding that the endothelium is a dynamic heterogeneousand disseminated organ which orchestrates blood vessel andcirculatory functions thus exerting a critical role for tissuehomeostasis Indeed the endothelial cells (ECs) possessessential secretory synthetic metabolic and immunologicactivities [1 2]

The endothelium is semipermeable and regulates thetransport of variousmolecules between the blood and under-lying interstitial space by expressing specific carriers ECs alsocontrol vascular permeability especially in microvasculardistricts Moreover ECs importantly contribute to main-taining a nonthrombogenic blood-tissue interface since theyrelease various antithrombotic and fibrinolytic factors as wellas molecules that impact on platelets [1 2]

The endothelium is an immunocompetent organ becauseit exposes histocompatibility and blood group antigens canbe induced to express adhesion molecules for leukocytes

and produce cytokines Finally a functional relation existsbetween endothelial and smooth muscle cells as a conse-quence of the presence of junctions allowing the passageof electric charges and metabolites and the productionand release of vasoactive mediators [1 2] Indeed ECsfinely control vasomotor responses through the productionand metabolism of vasoactive molecules acting on smoothmuscle cells as endothelin-1 (ET-1) nitric oxide (NO) andangiotensin II (AngII) They also tightly control smoothmuscle cells proliferation [1 2] ECs are protagonists inangiogenesis that is the formation of new blood vessels frompreexisting ones Angiogenesis involves the most dynamicfunctions of the endothelium since it requires the migrationof ECs their ability to degrade the extracellular matrixtheir proliferation and differentiation ultimately leading tofunctional capillaries [3] This highly organised process ismodulated by the balance between stimulators and inhibitorsof angiogenesis

Vascular endothelium is structurally and function-ally heterogeneous [4] This heterogeneity is detectable at



different levels that is markers of cell activation gene expres-sion responsiveness to growth factors and antigen com-position and differentiates the behaviour between micro-and macrovascular ECs as well as between cells isolatedfrom different organs and from different vascular districtsof the same organ In fact the arteriolar endothelium isdifferent from the venous one as well as from the micro- andmacrovessel derived ECs The endothelium of the cerebralcirculationmdashwhich is the main component of the blood-brain barrier to protect the brain from toxic substancesmdashdeserves special consideration It is continuous has tightjunctions and differs both from fenestrated endotheliumwhere cells have pores and from discontinuous endotheliumwhere cells have intracellular and transcellular discontinuities[2]

ECs are normally quiescent in vivo with a turnover rateof approximately once every three years [5] Most of ECs inthe adult have a cell cycle variable from months to yearsunless injury to the vessel wall or angiogenesis occurs Onlyendothelium from endometrium and corpus luteum has adoubling time of weeks

ECs act as mechanotransducers whereby the transmis-sion of external forces induces various cytoskeletal changesand activates second messenger cascades which in turnmay act on specific response elements of promoter genesTherefore it is not surprising that ECs are sensitive tovariations of gravity

2 Methods to Simulate Microgravityand Hypergravity on Earth

Gravity is exerted permanently on organisms which are inconstant orientation in the gravity field (static stimulation)and if their orientation is changed with respect to the gravityvector (dynamic stimulation) [6]

The only way to achieve real microgravity is to useparabolic flights rockets space crafts or space labs asavailable on the International Space Station (ISS) Howeverthe possibility to perform experiments in real microgravityis limited because of high costs and the limited numberof missions On the other hand the short duration ofmicrogravity conditions achieved by using parabolic flightsor rockets limits the studies of many complex and prolongedbiological processesTherefore many efforts have been madeto establish methods to simulate microgravity on Earth Allthe devices available however mimic only some aspects ofreal microgravity

21 Clinostat Clinostats are considered reasonably effectiveground-based tools for simulating microgravity [7ndash10] andhave been used to study the effects of microgravity [11ndash19]

The clinostat randomizes motion and theoreticallyreduces the uniform gravity influence In the more widelydiffuse design that is the random positioning machine(RPM) the clinostat consists of an inner chamber containingthe samples which rotate clockwise anticlockwise verticallyand horizontally The horizontal and vertical motions areprovided by an outer chamber All the chambers are operated

by small motors under computer controlThe cells are grownin cell chambers or in flasks filled completely with mediathus diminishing the likelihood of turbulence and shearforces during culture rotation When using the clinostatto simulate microgravity the shear stress and vibrationsgenerated by the clinostat must be taken into account Shearstress can be limited by completely filling the chamberwith the culture medium Parallel controls are necessary toeliminate the effects of vibrations It is also important toconsider the distance of the samples from the centre of theplatform where the maximal reduction of gravity occursAnother important parameter to monitor is the speed ofrotation It has been verified that the effects of clinostat-determined microgravity are similar to those obtained inspace labs [8ndash19]

22 Rotating Wall Vessel Bioreactor This device wasdeveloped at NASArsquos Johnson Space Center to simulatethe effects of microgravity on cells in a ground-basedculture system The bioreactor the rotating wall vessel(RWV)rotating cell culture system (RCCS) from Synthecon(httpwwwsyntheconcom) is a cylindrical vessel thatmaintains cells in suspension by slow rotation around itshorizontal axis with a coaxial tubular silicon membranefor oxygenation Adherent cells need to be cultured onbeads This system represents a new cell culture technologydeveloped for 3D cultures of different cell types andbiotechnological applications The vessel wall and themedium containing cells bound to microcarrier beads or3D cultures rotate at the same speed producing a vector-averaged gravity comparable with that of near-Earth free-fallorbit [20] Most results obtained using the RWV wereconfirmed by experiments in real microgravity [12 21ndash23]

23 Magnetic Levitation This is a relatively novel Earth-based simulation technique used to investigate the biologicalresponse to weightlessness Magnetic levitation takes placewhen the magnetic force counterbalances the gravitationalforce Under this condition a diamagnetic sample is in asimulatedmicrogravity environment However the magneticfield which is generated affects cell behaviour thereforeconfounding the effects of simulated microgravity Mouseosteoblastic MC3T3-E1 cultured in a superconducting mag-net for 2 days showed marked alterations of gene expression[24] Random rotation and magnetic levitation inducedsimilar changes in the actin of A431 cells that were alsodescribed in real microgravity [25] At the moment howeverno studies are available on ECs under magnetic levitationbut they should be fostered as levitation as an alternative tosimulate microgravity might yield novel information or con-firm previous data thereby helping in designing successfulexperiments in real microgravity

24 Models to Generate Hypergravity Variation in gravityexposure is also related to hypergravity as the one to whichthe astronauts are transiently exposed during launch andreturn to Earth Also military pilots and subjects engaged


in certain sports such as motor racing motorcycling bob-sledding and the luge experience hypergravity The com-parison among the conditions of microgravity normogravity(1timesg) and hypergravity may be helpful to understand themechanisms underlying the effects of gravitational alterationson endothelial function and to understand what happenswhen humans quickly pass fromhypergravity tomicrogravityconditions and vice versa

Centrifuges constructed for research under hypergravityconditions are characterized by high precision control of rpmTheir speed and the angle of inclination of the sample can beregulated to obtain the desired hypergravity in a range from 1to many g Centrifuges are also used to perform 1timesg controlexperiments on board of the ISS and spacecraft Studies onendothelial cells in hypergravity are available [12 26 27]

3 The Effects of Microgravity on ECs

Exposure to microgravity during space missions impacts onvarious systems In humansmicrogravity-induced alterationsinclude bone loss muscle atrophy cardiovascular decondi-tioning impairment of pulmonary function and immuneresponse [50 51] The cardiovascular system is affected byspaceflight with changes manifesting as cardiac dysrhyth-mias cardiac atrophy orthostatic intolerance and reducedaerobic capacity [52] These changes can cause adaptationproblems when astronauts return back to Earth especiallyafter long-duration spaceflights [53]

Because ECs are key players in the maintenance ofvascular integrity inflammation and angiogenesis severalstudies have been devoted to the mechanisms by whichmicrogravity affects EC functions

Various reports have indicated that ECs are highly sensi-tive to microgravity and undergo morphological functionaland biochemical changes under these conditions [11 12 2328 29 37 38 46 49 54] These studies have used a varietyof in vitro cell models with divergent results One of thereasons for these discrepancies can be EC heterogeneity orthe isolation from different species Indeed human bovinemurine and porcine endothelial cells have been investigatedunder gravitational unloading With concern to human cellsstudies are available on human ECs from the umbilicalvein (HUVEC) widely considered a model of macrovas-cular endothelial cells as well as on human microvascularECs (HMEC) Moreover studies have been performed onEAhy926 cells a fusion of HUVEC with the lung carcinomacell line A549 [55] Although immortalized cell lines offersignificant logistical advantages over primary cells in in vitrostudies they exhibit important differences when compared totheir primary cell counterparts Indeed microarrays used fora genome-wide comparison between HUVEC and EAhy926in their baseline properties have shown that EAhy926 cellsare useful in studies on genes encodingmolecules involved inregulating thrombohemorrhagic features while they appearto be less suited for studies on the regulation of cell prolifera-tion and apoptosis [56] Moreover immortalized endothelialcell lines show different expression pattern of biomarkerswhen compared to primary cells [57] The controversial

results reported about the response of ECs to microgravitycould be due also to the diverse experimental approachesutilized such as the device simulating microgravity theduration of exposure to simulated microgravity and thedegree of reduction of the gravity that can be reachedoperating these devices differently (see above) Neverthelessaltered EC morphology cell membrane permeability andsenescence are documented by spaceflight experiments oncultured endothelium [21 30 58]

Several aspects of endothelial behaviour have been stud-ied in simulated and real microgravity Table 1 summarizesthe published findings

31 Migration Controversy exists on this topic No signif-icant modulation of cell migration under basal conditionand in response to the angiogenic factor hepatocyte growthfactor (HGF) was observed in HUVEC as well as in HMECcultured in the RPM [12 46] Shi et al [39] demonstratedthat after 24 h of exposure to simulated microgravity ina clinostat HUVEC migration was significantly promotedthrough the eNOS pathway upregulation by means of PI3K-Akt signalling On the contrary the endothelial cell lineEAhy926 in simulated microgravity migrated more thancontrols [31] while in a study on porcine aortic endothelialcells (PAEC) microgravity modelled by a RPM caused amarked impairment of cell migration induced by serum orthe angiogenic factors vascular endothelial growth factor(VEGF) and fibroblast growth factor-2 (FGF-2) [11]

32 Proliferation and Formation of 3D Structures Carlssonand Versari using the RWV and the RPM respectivelyfound that the proliferation rate of HUVECs was reversiblyincreased under simulated microgravity [12 28] Also bovineaortic ECs (BAEC) grew faster in the RWV than controls[35] On the contrary simulated microgravity inhibited thegrowth of HMEC and murine microvascular ECs [23 46]The results obtained using microvascular EC are reinforcedby the in vivo finding showing an impairment of angiogenesisin space Wound healing in which neovascularization is anearly and fundamental step is retarded in space-flown animalmodels [59] and the development of vascular channels in arat fibular osteotomy model is inhibited after flight as shownby an experiment carried out during a shuttle mission [60]

Also in PAECs a marked impairment of EC responsive-ness to angiogenic factors and a reduced ability to proliferatewere reported [11] Using the endothelial cell line EAhy926Grimm et al [43] showed the formation of 3D tubularstructures in clinorotation After two weeks a subtype of 3Daggregates was observed with a central lumen surroundedby one layer of ECs These single-layered tubular structuresresembled the intimas of blood vessels Characterization ofthese tubular structures revealed that they might originatefromdouble-row cell assemblies formedbetween the fifth andseventh day of culture under simulated microgravity [43]

33 Apoptosis Increased apoptosis after culture in the RPMhas been observed in PAEC and the endothelial cell line


Table1Th

eeffectso

frealorsim

ulated

microgravity

ondifferent

endo

thelialcelltypes

Experim

entalm

odel

Experim

entalcon

ditio

nsEff

ects

Authors

Prim

aryhu

man

umbilicalvein

ECs

(HUVEC

)

Rotatin

gwallvessel(RW

V)

Rand

ompo

sitioning

machine

(RPM

)48or

96h

Growth

stim

ulation

uarrNOprod

uctio

nAc

tinremod

ellin

gdarrAc

tin

Versarietal2007

[12]

Spaceflight

(Progress4

0Pmission)

10d

uarrTh

ioredo

xin-interactingprotein

darrhsp-70

and90

uarrsecretionof

IL-1120572andIL-1120573

Ionchannels(TPC

N1KC

NG2KC

NJ14

KCN

G1KC

NT1T

RPM1CL

CN4

CLCA

2)m

itochon

drialoxidativ

epho

spho

rylatio

nandfocaladh

esionwerew

idely

affected

Versarietal2013

[21]

RWV72

hor

96h

uarrPG

I2andNO

Carlssonetal2002

[22]

RPM

24ndash4

8huarrNOuarrCa

v-1p

hospho

rylatio

n(Tyr

14)

Spisn

ietal2006

[26]

RWV4244896144

h

uarruarrhsp70

darrIL-1120572

Remod

ellin

gof

cytoskele

ton

darractin

Carlssonetal2003

[28]

RPM

24h

uarreN

OSCa

v-1and

-2darrof

thelengthandwidth

ofthec

ells

darrICAM-1V

CAM-1E

-sele

ctinand

IL-6

andTN

F-120572

Greno

netal2013

[29]

Spaceflight

12d

Cytoskele

tald

amage

uarrcellmem

branep

ermeability

Inreadaptedcells

persistingcytoskele

talchanges

darrmetabolism

andcellgrow

th

Kapitono

vaetal2012

[30]

2D-C

linostat(developedby

ChinaA

stron

autR

esearchand

Training

Center)30

rpm24h

uarrHUVEC

tube

form

ationandmigratio

ndarrnu

mbero

fcaveolaeinthem

embrane

uarreN

OSactiv

ityby

phosph

orylationof

Akt

andeN

OS

Siam

walae

tal2010

[31]

RWV5m

in30m

in1hand24

h

uarrICAM-1expressio

nDepolym

erizationof

F-actin

andclu

sterin

gof

ICAM-1on

cellmem

brane(short

term

)Ac

tinfib

errearrangem

entand

stablec

luste

ringof

ICAM-1(afte

r24h

)uarrICAM-1andVC

AM-1RN

Aaft

er30

min

Zhangetal2010

[32]

RPM

96h

Alteratio

nof

proteins

regu

latin

gcytoskele

tonassembly

darrIL-1120572IL-8andbF

GF

uarrchem

okines

Rantes

andEo

taxin

involved

inleuk

ocytes

recruitm

ent

Griff

onietal2011

[33]

RPM

24h

uarriN

OSby

amechanism

depend

ento

nsupp

ressionof

AP-1

Wangetal2009

[34]


Table1Con

tinued

Experim

entalm

odel

Experim

entalcon

ditio

nsEff

ects

Authors

Bovine

aorticEC

s(BA

EC)

RWVforu

pto

30d

Growth

stim

ulation

uarrNO

Prod

uctio

nof

NOdepend

ento

ntheR

WVrotatio

nrate73

increase

at8r

pm

262

increase

at15rpmand

500

increase

at20

rpm

Sanfordetal2002

[35]

Porcinea

ortic

ECs(PA

EC)

RPM

72h

uarrproapo

ptoticgenes(p53FA

S-LBA

X)darrantia

poptoticgenes(Bc

l-2)

Diss

olutionof

mito

chon

drialm

embraneintegrity

Impairm

ento

fcellrespo

nsivenesstoexogenou

sstim

uli

Morbidelli

etal2005

[11]

Bovine

coronary

venu

larE

Cs(C

VEC

)RP

M72

h

uarrFibron

ectin

(form

ationof

intricaten

etworkof

FNfib

ers)

uarrLaminin

uarr120573-Actin

(form

ationof

stressfi

bers)

uarr120572120573-Integrin

(form

ationof

cluste

rs)

Mon

icietal2011

[36]

Hum

anEC

lineE

Ahy926

RPM

10days

uarrCa

spase-3Ba

xandBc

l-2uarrcollagentypesI

andIII

Alteratio

nsof

thec

ytoskeletal120572

-and120573-tu

bulin

sand

F-actin

darrbrain-deriv

edneurotroph

icfactorplatele

ttissue

factorV

EGFandET

-1

Infanger

etal2007

[37]

RPM

7days

Mod

ulationof

genese

ncod

ingforsignaltransdu

ctionandangiogenicfactorscell

adhesio

nmem

branetranspo

rtproteinsore

nzym

esinvolved

inserin

ebiosynthesis

Mae

tal2013

[38]

RPM

2h

uarrcellu

larm

igratio

nuarrfilop

odiaandlamellipod

iaAc

tinrearrangem

ents

uarrNO

Shietal2012

[39]

RPM

4122448and

72h

uarrextracellularm

atrix

(ECM

)proteins

Alteratio

nin

cytoskeletalcompo

nents

uarrexpressio

nof

ECM

proteins

(collagentype

Ifib

ronectinoste

opon

tinlam

inin)

andflk

-1protein

Morph

ologicalandbiochemicalsig

nsof

apop

tosis

after

4hfurther

increasin

gaft

er72

h

Infanger

etal2006

[40]

Parabo

licflight(22

smicrogravity18xg

2perio

dsof

20s)

Parabolic

flight

darrCO

L4A5CO

L8A1ITGA6ITGA10and

ITGB3

mRN

Asaft

erP1

(firstp

arabolas)

uarrED

N1and

TNFR

SF12AmRN

Asaft

erP1

darrADAM19C

ARD

8CD

40G

SNP

RKCA

mRN

AsuarrPR

KAA1(AMPK1205721)mRN

Ascytoplasmicrearrangem

ent

uarruarrABL

2aft

erP1

andP3

1

Grossee

tal2012

[41]

Parabo

licflight(22

smicrogravity18xg

2perio

dsof

20s)

Parabolic

flight

Actin

networkrearrangem

ent

uarrCC

NA2CC

ND1CD

C6C

DKN

1AE

ZRM

SNO

PNV

EGFA

CASP

3CA

SP8

ANXA1ANXA2andBIRC

5darrFL

K1uarrEZ

RMSN

OPN

ANXA2andBIRC

5aft

er31P

Wehland

etal2013

[42]


Table1Con

tinued

Experim

entalm

odel

Experim

entalcon

ditio

nsEff

ects

Authors

RPM

71421and28

d

Differentrespo

nsivenesstoVEG

FandbF

GFaddedexogenou

slyAlteredgene

andproteinexpressio

nof

phosph

okinaseA

catalytic

subu

nit

phosph

okinaseC

alph

aandER

K-1and

2darrVEG

FbF

GFsolubleT

NFR

SF5TN

FSF5ICA

M-1T

NFR

2IL-18complem

ent

C3and

vonWillebrand

factor

Grim

metal2010

[43]

RPM

7and28

dDelayed

3Dcellgrow

th

uarrbeta(1)-integrinlam

ininfibron

ectin

120572-tu

bulin

intube-like

structuresa

fter4

weeks

ofcultu

ring

Grim

metal2009

[44]

Hum

anEC

lineE

Ahy926

Bovine

lung

microvascular

ECS

Bovine

pulm

onaryaorticEC

sPo

rcinev

entricular

endo

cardialE

Cs

RPM

2h

Results

indicatethatiN

OSisam

olecular

switchforthe

effectsof

microgravity

ondifferent

kind

sofend

othelialcells

uarrangiogenesisviathe

cyclicg

uano

sinem

onop

hosphate(cGMP-)P

KGdepend

ent

pathway

Siam

walae

tal2010

[45]

Hum

anderm

almicrovascular

cells

(HMEC

)RW

VRP

M48

or96

or168h

uarrTIMP-2

uarrNOdarrproteasomea

ctivity

Mariotti

andMaier2008

[46]

Murinelun

gcapillary

ECs(1G

11cells)

RWV72

h

darrendo

thelialgrowth

uarrp21

darrIL-6

uarreN

OSandNO

Cotrupi

etal2005

[23]

Hum

anpu

lmon

arymicrovascular

ECs

(HPM

ECs)

MG-3

clino

stat(develo

pedby

the

Institu

teof

Biop

hysic

sChinese

Academ

yof

Sciences)

uarrapop

tosis

darrPI3K

Akt

pathway

uarrNF-120581Banddepo

lymerizationof

F-actin

Kang

etal2011

[47]

Hum

anandbo

vine

microvascular

ECs

RWV96

huarrhsp70in

cells

which

maintainedthec

apabilityto

proliferatein

microgravity

Cotrupi

andMaier

2004

[48]

Cocultureso

fend

othelialm

onolayers

human

lymph

ocytesimmun

ecells

and

myeloleucem

ic(K

-560)cells

Spaceflight

(ISS)

uarradhesio

nof

PMA-

activ

ated

lymph

ocytes

Retained

abilityof

immun

ecellsto

contactrecogn

izeanddestroyon

cogenicc

ells

invitro

Buravkovae

tal2005

[49]

Legend

uarrincreaseddarrdecreased


EAhy926 [11 40] In particular following exposure to sim-ulated hypogravity PAEC change their morphology andgene expression pattern triggering proapoptotic signals Thegene expression profile demonstrated the upregulation ofp53 FAS-L and BAX genes and the concomitant down-regulation of the antiapoptotic protein Bcl-2 and prolif-eration marker PCNA The induction of apoptosis wasaccompanied by mitochondrial disassembly thus suggest-ing the activation of the mitochondrial intrinsic pathways[11]

In pulmonary HMEC simulated microgravity-inducedapoptosis by downregulating the PI3KAkt pathways andincreasing the expression of NF120581B [47] On the con-trary no apoptosis was observed in HUVEC and dermalHMEC cultured for various times in the RWV or in theRPM and this has been linked to the rapid inductionof heat shock protein (hsp)-70 [28 46 48] Indeed hsp-70 protects endothelial cells from apoptotic stimuli act-ing downstream of cytochrome c release and upstream ofcaspase 3

34 Alterations of Cytoskeleton and Extracellular Matrix Thecytoskeleton plays a key role in the adaptation to mechanicalstress including alterations of gravity [61 62] Therefore thechanges that cytoskeletal components such as microtubulesundergo inmicrogravity can be a key to explaining the effectsof weightlessness on cells [63 64]

Carlsson et al [28] studied actin microfilaments inHUVEC exposed to microgravity simulated by the RWVIn comparison with controls the cells showed elongatedand extended podia disorganization of actin microfilamentsthat clustered in the perinuclear area and decrease in stressfibers Moreover after 96 h exposure actin RNA levels weredownregulated and total actin amounts were reduced Thecytoskeletal modifications were reversible upon return tonormal growth conditions (1timesg) The authors speculatedthat the reduction in actin amount could be an adaptivemechanism to avoid the accumulation of redundant actinfibers The same results were obtained when the experimentwas replicated by using a RPM to model the micrograv-ity conditions [12] More recently in HUVEC exposed tomechanical unloading by RPM Grenon et al [29] found dis-organization of the actin network with clustering of the fibersaround the nucleus Moreover they observed that caveolin-1was less associated with the plasma membrane and adopteda perinuclear localization Thus the authors advanced thehypothesis that disruption of the actin cytoskeleton orga-nization could impair the translocation of caveolin-1 to thecaveolae

After spaceflight (Soyuz TMA-11) readapted HUVECcells with subsequent passages exhibited persisting changesin the organization of microtubules with prominent bundlesthat occupied the peripheral cytoplasm [30]

In a study carried out by Zhang et al [32] HUVEC acti-vated with TNF-120572 and exposed to microgravity modelled byRWV demonstrated that after 30min depolymerization ofF-actin and clustering of ICAM-1 on cellmembrane occurredMoreover ICAM-1 and VCAM-1 RNA were upregulated

After 24 h actin fiber rearrangement was initiated clusteringof ICAM-1 became stable and the mRNAs of ICAM-1 andVCAM-1 returned to levels comparable with the controlsThe authors speculated that actin cytoskeleton rearrangementand changes in levels and distribution of surface adhesionmolecules could significantly affect transendothelial migra-tion processes

Grosse et al [41] studied the effect of parabolic flighton the cytoskeleton of the endothelial cell line EAhy926Every parabola (P) included two hypergravity (18 g) periodsof 20 s separated by a 22 s microgravity period After P1they observed a rearrangement of120573-tubulin that accumulatedaround the nucleus After P31 120573-tubulin and vimentin weredownregulated Using the EAhy926 cell line exposed toparabolic flight Wehland et al [42] reported that the actinnetwork underwent a drastic rearrangement mostly affectedby vibration

Grimm et al [44] studied the walls of tube-like structuresspontaneously formed by the endothelial cell line EAhy926cultured in a RPM They found that the walls consistedof single-layered endothelial-like cells which had producedsignificantly more 120573

1-integrin laminin (LM) fibronectin

(FN) and 120572-tubulin than controls Microgravity-inducedupregulation of proteins involved in the extracellular matrixbuilding was confirmed in studies carried out by Moniciet al [36] on cultured bovine coronary venular endothelialcells (CVECs) exposed for 72 h to microgravity modelledby a RPM The authors observed an increase in actincontent and impressive production of actin stress fibersaccompanied by the overexpression and clustering of 120573

1-

integrin 40 increase in LM 111 increase in FN contentand formation of a tight and intricate network of FNfibrils

Since FN and LM are strongly involved in the regulationof cell adhesionmigration their upregulation and alterednetworking together with the changes in actin and integrinpatterns induced the authors to hypothesize that the exposureto microgravity causes a dysregulation in cell motility andadhesion to the substrate

In summary all of the studies carried out so far demon-strated that microgravity strongly affects cytoskeleton orga-nization and induces a rearrangement of the actin networkwith clustering of the fibers in the perinuclear area A similarbehaviour has been observed also analysing the microtubulenetwork Moreover clustering of adhesion molecules on theplasma membrane and overexpression of proteins of theextracellular matrix have been reported by some authors

The results are less consistent when considering theexpression of cytoskeleton proteins or their RNA Probablythe discrepancies are due to differences in experimentalmodels (different cell populations) protocols and analyticalprocedures

However it is widely accepted that the microgravity-induced changes in the cytoskeleton can strongly affect thebehaviour of endothelial cells in terms of adhesion migra-tion and production of extracellular matrix and can interferewith other processes such as translocation ofmolecules insidethe cells transendothelial migration and even inflammationand angiogenesis


35 Synthesis of VasoactiveMolecules The levels of vasoactivemolecules such as NO and ET-1 are modified under micro-gravity conditions which also indicates that microgravitymay influence both hemodynamic changes and angiogenesis[33 37] In particular HUVEC and HMEC exposed to simu-lated microgravity using RWV and RPM produce more NOthan controls as the result of increased levels of endothelial-nitric oxide synthase (e-NOS) [12] which correlates withthe increase of caveolins [26 29] In particular Grenonet al suggested that the alterations in NO production aremediated by changes in the cytoskeleton detected in all theendothelial types studied [29] Wang et al [34] explained theincreased amounts of NO in HUVEC after 24 h in simulatedmicrogravity as the results of the upregulation of inducibleNOS through a mechanism dependent on the suppression ofthe activity of the transcription factorAP-1 Also inBAECNOproduction was increased [35]

In the endothelial cell line EAhy926 a reduced releaseof ET-1 and VEGF was reported [37] while the productionof NO was increased via the iNOS-cGMP-PKG pathway [3945] If confirmed in vivo in space these results might in partexplain the hemodynamic changes and the redistribution ofblood flows induced by microgravity

36 Genomic and Proteomic Analysis Microgravity affectsseveral molecular features of ECs markedly modulatinggene expression In HUVEC cultured in the RPM thesecretome was evaluated by a 2D proteomic approach [33]The proangiogenic factor FGF-2 and the proinflammatorycytokines interleukin-1 (IL-1) and IL-8 were decreased insimulatedmicrogravity whereas two chemokines involved inleukocyte recruitment Rantes and Eotaxin were increased[33] The unprecedented gene profile analysis on HUVECcultured on the ISS for 10 days was performed by Versariet al [21] 1023 genes were significantly modulated themajority of which are involved in cell adhesion oxidativephosphorylation stress responses cell cycle and apopto-sis thioredoxin-interacting protein being the most upregu-lated Briefly in cultured HUVEC real microgravity affectsthe same molecular machinery which senses alterations offlow and generates a prooxidative environment that altersendothelial function and promotes senescence [21] Similarconclusions were reached by Kapitonova et al [30 58] whodescribed premature senescence in space-flown HUVEC Byaccelerating some aspects of senescencemicrogravity offers abig challenge to study themechanisms implicated in the onsetof aging

Looking at the endothelial cell line EAhy926 a shortterm lack of gravity (22 s) generated by parabolic flightssignificantly influences the signalling pathways [41] Whenthese cells are cultured for various times from 4 to 72 h onthe RPM a number of proteins of the extracellular matriximplicated in apoptosis are modulated when compared tocontrol cells [40] In theRPMsomeEAhy926 cells form tube-like 3D aggregates while others continue to grow adherently3D aggregates and adherent cells were analyzed by genearray and PCR techniques and compared to controls [38]1625 differentially expressed genes were identified and in

particular the levels of expression of 27 genes changedat least 4-fold in RPM-cultured cells when compared tocontrolsThese genes code for angiogenic factors and proteinsimplicated in signal transduction cell adhesion membranetransport or enzymes Fifteen of them with IL-8 and vonWillebrand factor being the most affected showed linkagesto genes of 20 proteins that are important in the maintenanceof cell structure and in angiogenesis

EAhy926 cell line and human dermal microvascular ECs(HMVECs) were then compared after culture on the RPM for5 and 7 days [54] A total of 1175 types of proteins were foundin EAhy926 cells and 846 in HMVECs 584 of which werecommon and included metabolic enzymes structure-relatedand stress proteins This proteomic study also highlightsthat HMVECs develop tube-like 3D structures faster thanEAhy926 possibly through a transient augmentation ofribosomal proteins during the 3D assembling of ECs

4 The Effects of Hypergravity on ECs

A summary of published data on endothelial cell behaviouris reported in Table 2 HUVECs exposed to hypergravity(3timesg) for 24ndash48 h showed inhibition of cell growth butunaltered apoptosis increased COX-2 eNOS and Cav-1suggesting a possible role of caveolae in mechanotransduc-tion Also an increased synthesis of PGI2 and NO whichare also proangiogenic was observed However surprisinglythe formation of capillary-like structure was inhibited [65]Versari et al [12] studying the same cells exposed to 35timesgfound increased NO production enhanced cell migrationbut no effects on proliferation Moreover altered distributionof actin fibers without modifications of the total amountsof actin was detected [12] In the same conditions HUVECshowed a time-dependent decrease in occludin correlatingwith an increase in paracellular permeability and a decreasein transendothelial electrical resistance indicating a decreasein EC barrier function [66 67] with exactly opposingresults in BAEC cultured under hypogravity in RWV whereincreased barrier properties were detected [35]

Koyama et al [69] reported that after a few minutesof exposure to 3timesg in a centrifuge BAECs showed actinreorganization via Rho activation and FAK phosphorylationincreased cell proliferation andATP release Adaily exposureof 1-2 h repeated for 5 consecutive days promoted cell migra-tion Wehland et al [42] investigated short term (s) effectsof hypergravity (18timesg) on EAhy926 cells and found that thecells were weakly affected by loading in the conditions usedfor the experiment On the contrary short term effects ofmicrogravity were much more evident

In order to evaluate these results two considerations haveto be made

(1) Very different protocols and parameters have beenused for EC exposure to hypergravity continuousversus discontinuous exposure different g force valueand exposure times ranging from minutes to days

(2) ECs both derived from the microvasculature andmacrocirculation are very sensitive to mechanicalstress It should be underscored that in physiological


Table2Th

eeffectso

fhypergravity

cond

ition

sondifferent

endo

thelialcelltypes

Experim

entalm

odel

Experim

entalcon

ditio

nsEff

ects

Authors

Prim

aryhu

man

umbilicalvein

ECs(HUVEC

)

Hypergravity

cond

ition

s(generatedby

aMidiCAR

centrifugea

t35xg)for

24ndash4

8h

uarrmigratio

nuarrNO

Altereddistrib

utionof

actin

fibers

Versarietal2007

[12]

Hypergravity

cond

ition

s(generatedby

acentrifu

geat

3xg)

for2

4ndash48

h

uarrcav-1

uarrdistrib

utionof

caveolae

inthec

ellinterior

uarrCO

X-2NOand

PGI2

prod

uctio

ndarrangiogenesis(th

roug

hap

athw

ayno

tinvolving

apop

tosis)

Spisn

ietal2003

[65]

Lifto

ffsim

ulationby

centrifuge(75

min

simulationof

thep

attern

ofgforces

experie

nced

durin

glift

offof

the

NASA

spaces

huttle)

darrMAPK

phosph

orylation

uarrocclu

dinexpressio

nSumanasekerae

tal2006

[66]

Lifto

ffsim

ulationby

centrifuge(75

min

simulationof

thep

attern

ofgforces

experie

nced

durin

glift

offof

the

NASA

spaces

huttle)

uarrParacellu

larp

ermeability

darrOccludin

darrTransend

othelialelectric

alresistance

darrMAPK

activ

ation

darrEC

barrierfun

ction

Sumanasekerae

tal2007

[67]

Bovine

aorticEC

s(BA

EC)

Hypergravity

(thermostated3-18KSigm

aZentrifu

gen

5perio

dsof

10min

expo

sure

to10xg

spaced

with

10min

at1xg)

Mod

ified

integrin

distrib

ution

Reorganizatio

nof

cytoskele

taln

etwork

darrgenesc

ontro

lling

vasoconstrictio

nandinflammation

darrProapo

ptoticsig

nals

Morbidelli

etal2009

[68]

Hypergravity

(3xg)app

liedby

lowspeedcentrifuge

uarrAT

Prelease

uarractin

reorganizatio

nviaR

hoA

activ

ationandFA

Kph

osph

orylation

uarrcellproliferatio

nandmigratio

n

Koyamae

tal2009

[69]

Bovine

coronary

venu

larE

Cs(C

VEC

)

Hypergravity


aZentrifu

gen

5perio

dsof

10min

expo

sure

to10xg

spaced

with

10min

at1xg)

darrproapo

ptoticgenes(FA

DDFasFas-L)

uarrantia

poptoticgene

NF120581

BChangeincytoskele

tonorganizatio

nAlteratio

nof

cellenergy

metabolism

Mon

icietal2006

[27]

Hum

anEC

lineE

Ahy926

Hypergravity

Experim

ents(M

uSICD

LRC

ologneG

ermany

centrifuge18xg)

Vibrationexperim

ents(V

ibraplex

vibrationplatform

frequ

ency

range0

2ndash14k

Hz)

darrCA

RD8NOS3V

ASH

1SE

RPIN

H1(allP

1)C

AV2ADAM19

TNFR

SF12AC

D40

and

ITGA6(P31)m

RNAs

Nosig

nificantchanges

ongene

expressio

nandmorph

olog

yof

the

cells

Grossee

tal2012

[41]

Hypergravity

Experim

ents(M

uSICD

LRC

ologneG

ermany

centrifuge18xg)

Vibrationexperim

ents(V

ibraplex

vibrationplatform

frequ

ency

range0

2ndash14k

Hz)

darrPan-actin

tub

ulinand

Moesin

darrgene

expressio

nof

CCND1MSN

RDX

OPN

BIRC5

and

ACTB

uarrPan-actin

tub

ulinand

ezrin

darr120573-Actin

andMoesin

darrAC

TBC

CND1CD

C6C

DKN

1AV

EGFA

FLK

-1E

ZRITB

G1

OPN

CASP

3CA

SP8ANXA2andBIRC

5

Wehland

etal2013

[42]

Legend



Table 3 Summary of the principal parameters influenced by simulated microgravity and hypergravity in different types of ECs

Microgravity HypergravityEndothelial cellline EAhy926 Dermal HMEC HUVEC PAEC BAEC Endothelial cell

line EAhy926 HUVEC CVEC BAEC

Migration darr = =uarr darr ND ND uarr ND uarr

Proliferation ND darr uarr darr uarr ND (=uarr) = ND uarr

Apoptosis uarr = = uarr = =darr = = =NO synthesis uarr uarr uarr ND uarr ND uarr ND NDCytoskeletal rearrangements +++ +++ +++ +++ +++ +++ ++ +++ +++Legend uarr increased darr decreased = no change ND not determined ++ and +++ highly and strongly upregulated

conditions the quality and intensity of mechanicalstimulation to which the endothelium is exposeddepend on the vascular district

Following the latter consideration we hypothesized thatEC response to gravitational alteration could depend on thedistrict from which the cell population derives and could bedifferent in cells derived from macro- or microcirculationTo verify this hypothesis we studied and compared thebehaviour of coronary venular endothelial cells (CVEC) [27]and BAEC [68] exposed to 10timesg for 5 periods of 10 minuteseach spaced with four recovery periods of the same durationFollowing exposure both the cell types showed similarchanges in cytoskeleton organization and 120572v1205733 integrindistribution The peripheral ring of actin microfilaments wassubstituted by trans-cytoplasmic stress fibers microtubulesand intermediate filaments gathered in the perinuclear areafocal contacts in the protruding lamellipodia disappearedand 120572v1205733 integrin molecules clustered in the central bodyof the cells Both in CVEC and in BAEC the expressionof the cytoskeletal proteins 120573-actin and vimentin increasedIn BAEC the transcripts for the matrix proteins LM andFN decreased In both the cell types exposure to hypergrav-ity decreased the transcription of genes encoding for theproapoptotic factors Fas and FasL Bcl-XL [27 68]

Cell energy metabolism assessed by autofluorescencespectroscopy and imaging did not change significantly inBAECs On the contrary CVECs exposed to hypergravityshowed an increase of the anaerobic metabolism in compar-ison with 1timesg controls [27]

The phenotypic expression of molecules involved ininflammation and angiogenesis such as eNOS FGF-2 andCOX-2 which is not expressed in basal conditions didnot significantly change as assessed by immunofluorescencemicroscopy in CVECs Nevertheless in BAECs the expres-sion of COX-2 and other genes controlling the calibre of thevessels that is renin ET processing enzyme and inflamma-tion such as TNF120572 and its receptor CD40 P and E selectinsCD54 was downregulated Briefly hypergravity does notseem to affect significantly the survival of both macro- andmicrovascular ECs However significant changes have beenobserved in cytoskeleton and integrin distribution in all theECs studied and changes in cell energy metabolism havebeen observed only in CVECs while the downregulation ofsome genes involved in inflammation and vasoconstrictionhas been found only in BAECs Considering the expression of

growthmodulators hypergravity increasedVEGF expressionwhile it decreased a series of interleukins acting as inhibitorsof EC proliferation [27 68] These results are consistentwith the hypothesis that the EC response to gravitationalalterations depends at least in part on the vascular districtfrom which the cells are derived

5 Concluding Remarks

The effects of simulated gravity changes on endothelialcells described in various papers are rather discordant butall converge in the indication that endothelial behaviouris significantly altered (Table 3) Briefly from studies ondifferent types of ECs exposed to simulated microgravity wecan summarize the following

(i) Impact on cell proliferation and survival all thestudies indicate alterations of cell proliferation OnlyHUVEC and BAEC have been reproducibly foundto proliferate faster in microgravity than controlsMicrovascular EC and other endothelial cells aregrowth inhibited or induced to apoptosis

(ii) Impact on NO synthesis most studies agree on theincreased production of NO through the modulationof NOS isoforms

(iii) Impact on cytoskeleton all the studies describedimportant cytoskeletal remodelling in all the differentEC analyzed

(iv) Impact on gene expression no doubt exists aboutthe profound modifications of gene expression byexposure to simulated or real microgravity

The impact of hypergravity on ECs is less defined Dueto the different experimental approaches adopted on variouscell types the findings are not consistent and deserve furtherconsideration

The effects of gravitational forces on mechanotransduc-tion in ECs responses have been the matter of only a fewinvestigations and remain largely unknown The plausiblemechanosensing targets for gravity changes appear to be thecytoskeletal structure and particularly caveolae [26 29 65]

In conclusion because (i) endothelial cells are crucial forthe integrity of the vessel wall and (ii) vessels are responsiblefor the homeostasis of all the tissues it is pivotal to continuestudies on this topic since the modulation of endothelial


functions can contribute to cardiovascular deconditioningand other disorders observed in space from bone loss tomuscle atrophy However it would be recommended toclearly define the experimental models to use A clear cutdefinition of endothelial cell models to be used and theconditions to model gravity need to be standardized



Acknowledgments

Part of this work has been funded byAgenzia Spaziale Italiana(ASI) and European Spatial Agency (ESA)

References

[1] D B Cines E S Pollak C A Buck et al ldquoEndothelial cells inphysiology and in the pathophysiology of vascular disordersrdquoBlood vol 91 no 10 pp 3527ndash3561 1998

[2] H F Galley andN RWebster ldquoPhysiology of the endotheliumrdquoBritish Journal of Anaesthesia vol 93 no 1 pp 105ndash113 2004

[3] S P Herbert and D Y R Stainier ldquoMolecular control ofendothelial cell behaviour during blood vessel morphogenesisrdquoNature ReviewsMolecular Cell Biology vol 12 no 9 pp 551ndash5642011

[4] E R Regan and W C Aird ldquoDynamical systems approach toendothelial heterogeneityrdquo Circulation Research vol 111 no 1pp 110ndash130 2012

[5] K E Foreman and J Tang ldquoMolecular mechanisms of replica-tive senescence in endothelial cellsrdquo Experimental Gerontologyvol 38 no 11-12 pp 1251ndash1257 2003

[6] B Buchen M Braun Z Hejnowicz and A Sievers ldquoStatolithspull onmicrofilamentsmdashexperiments undermicrogravityrdquo Pro-toplasma vol 172 no 1 pp 38ndash42 1993

[7] W Briegleb ldquoSome qualitative and quantitative aspects ofthe fast-rotating clinostat as a research toolrdquo ASGSB BulletinPublication of the American Society for Gravitational and SpaceBiology vol 5 no 2 pp 23ndash30 1992

[8] T F B Kraft J JW A van Loon and J Z Kiss ldquoPlastid positionin Arabidopsis columella cells is similar in microgravity and ona random-positioning machinerdquo Planta vol 211 no 3 pp 415ndash422 2000

[9] M A Kacena P Todd L C Gerstenfeld and W J LandisldquoExperiments with osteoblasts cultured under varying orienta-tions with respect to the gravity vectorrdquo Cytotechnology vol 39no 3 pp 147ndash154 2002

[10] Z Barjaktarovic A Nordheim T Lamkemeyer C Fladerer JMadlung and R Hampp ldquoTime-course of changes in amountsof specific proteins upon exposure to hyper-g 2-D clinorota-tion and 3-D random positioning of Arabidopsis cell culturesrdquoJournal of Experimental Botany vol 58 no 15-16 pp 4357ndash43632007

[11] L Morbidelli M Monici N Marziliano et al ldquoSimulatedhypogravity impairs the angiogenic response of endotheliumby up-regulating apoptotic signalsrdquoBiochemical and BiophysicalResearch Communications vol 334 no 2 pp 491ndash499 2005

[12] S Versari A Villa S Bradamante and J A M Maier ldquoAlter-ations of the actin cytoskeleton and increased nitric oxidesynthesis are common features in human primary endothelialcell response to changes in gravityrdquo Biochimica et BiophysicaActa vol 1773 no 11 pp 1645ndash1652 2007

[13] R Gruener R Roberts and R Reitstetter ldquoReduced receptoraggregation and altered cytoskeleton in cultured myocytes afterspace-flightrdquo Biological Sciences in Space vol 8 no 2 pp 79ndash931994

[14] D Grimm P Kossmehl M Shakibaei et al ldquoEffects of sim-ulated microgravity on thyroid carcinoma cellsrdquo Journal ofGravitational Physiology vol 9 no 1 pp P253ndashP256 2002

[15] K Hirasaka T Nikawa L Yuge et al ldquoClinorotation preventsdifferentiation of rat myoblastic L6 cells in association withreduced NF-120581B signalingrdquo Biochimica et Biophysica Acta vol1743 no 1-2 pp 130ndash140 2005

[16] Z Li Y Song M Yuzhong et al ldquoInfluence of simulatedmicrogravity on avian primordial germ cell migration andreproductive capacityrdquo Journal of Experimental Zoology vol292 no 7 pp 672ndash676 2002

[17] D Sarkar T Nagaya K Koga F Kambe Y Nomura andH SeoldquoRotation in clinostat results in apoptosis of osteoblastic ROS1728 cellsrdquo Journal of Gravitational Physiology vol 7 no 2 ppP71ndashP72 2000


[19] C CWoods K E Banks R Gruener andDDeLuca ldquoLoss of Tcell precursors after spaceflight and exposure to vector-averagedgravityrdquoThe FASEB Journal vol 17 no 11 pp 1526ndash1528 2003


[21] S Versari G Longinotti L Barenghi J A M Maier andS Bradamante ldquoThe challenging environment on board theInternational Space Station affects endothelial cell functionby triggering oxidative stress through thioredoxin interactingprotein overexpression the ESA-SPHINX experimentrdquo FASEBJournal vol 27 no 11 pp 4466ndash4475 2013

[22] S I Carlsson M T Bertilaccio I Ascari S Bradamante andJ A Maier ldquoModulation of human endothelial cell behaviourin simulated microgravityrdquo Journal of Gravitational Physiologyvol 9 no 1 pp P273ndashP274 2002

[23] S Cotrupi D Ranzani and J A M Maier ldquoImpact of modeledmicrogravity on microvascular endothelial cellsrdquo Biochimica etBiophysica ActamdashMolecular Cell Research vol 1746 no 2 pp163ndash168 2005

[24] B E Hammer L S Kidder P C Williams and W W XuldquoMagnetic levitation of MC3T3 osteoblast cells as a ground-based simulation of microgravityrdquo Microgravity Science andTechnology vol 21 no 4 pp 311ndash318 2009

[25] M J A Moes J C Gielen R-J Bleichrodt J J W A VanLoon P C M Christianen and J Boonstra ldquoSimulation ofmicrogravity by magnetic levitation and random positioningeffect on human A431 Cell morphologyrdquo Microgravity Scienceand Technology vol 23 no 2 pp 249ndash261 2011

[26] E Spisni M Toni A Strillacci et al ldquoCaveolae and caveolaeconstituents in mechanosensing effect of modeled micrograv-ity on cultured human endothelial cellsrdquo Cell Biochemistry andBiophysics vol 46 no 2 pp 155ndash164 2006

[27] M Monici N Marziliano V Basile et al ldquoHypergravity affectsmorphology and function in microvascular endothelial cellsrdquo


Microgravity Science and Technology vol 18 no 3-4 pp 234ndash238 2006

[28] S I M Carlsson M T S Bertilaccio E Ballabio and J AMMaier ldquoEndothelial stress by gravitational unloading effectson cell growth and cytoskeletal organizationrdquo Biochimica etBiophysica Acta Molecular Cell Research vol 1642 no 3 pp173ndash179 2003

[29] S M Grenon M Jeanne J Aguado-Zuniga M S Conteand M Hughes-Fulford ldquoEffects of gravitational mechanicalunloading in endothelial cells association between caveolinsinflammation and adhesion moleculesrdquo Scientific reports vol 3p 1494 2013

[30] M Y Kapitonova S Muid G R A Froemming et al ldquoRealspace flight travel is associated with ultrastructural changescytoskeletal disruption and premature senescence of HUVECrdquoMalaysian Journal of Pathology vol 34 no 2 pp 103ndash113 2012

[31] J H Siamwala S H Reddy S Majumder et al ldquoSimulatedmicrogravity perturbs actin polymerization to promote nitricoxide-associated migration in human immortalized Eahy926cellsrdquo Protoplasma vol 242 no 1 pp 3ndash12 2010


[33] C Griffoni S di Molfetta L Fantozzi et al ldquoModificationof proteins secreted by endothelial cells during modeled lowgravity exposurerdquo Journal of Cellular Biochemistry vol 112 no1 pp 265ndash272 2011

[34] Y-C Wang S Zhang T-Y Du B Wang and X-Q SunldquoClinorotation upregulates inducible nitric oxide synthase byinhibiting AP-1 activation in human umbilical vein endothelialcellsrdquo Journal of Cellular Biochemistry vol 107 no 2 pp 357ndash363 2009

[35] G L Sanford D Ellerson C Melhado-Gardner A E Sroufeand S Harris-Hooker ldquoThree-dimensional growth of endothe-lial cells in the microgravity-based rotating wall vessel bioreac-torrdquo In Vitro Cellular and Developmental Biology-Animal vol38 no 9 pp 493ndash504 2002

[36] M Monici F Cialdai G Romano et al ldquoAn in vitro studyon tissue repair impact of unloading on cells involved in theremodelling phaserdquo Microgravity Science and Technology vol23 no 4 pp 391ndash401 2011

[37] M Infanger C Ulbrich S Baatout et al ldquoModeled gravitationalunloading induced downregulation of endothelin-1 in humanendothelial cellsrdquo Journal of Cellular Biochemistry vol 101 no6 pp 1439ndash1455 2007

[38] X Ma M Wehland H Schulz et al ldquoGenomic approach toidentify factors that drive the formation of three-dimensionalstructures by EAhy926 endothelial cellsrdquo PLoS ONE vol 8 no5 Article ID e64402 2013

[39] F Shi Y-CWang T-Z Zhao et al ldquoEffects of simulatedmicro-gravity on human umbilical vein endothelial cell angiogenesisand role of the PI3K-Akt-eNOS signal pathwayrdquo PLoS ONE vol7 no 7 Article ID e40365 2012

[40] M Infanger P Kossmehl M Shakibaei et al ldquoInductionof three-dimensional assembly and increase in apoptosis ofhuman endothelial cells by simulated microgravity impact ofvascular endothelial growth factorrdquo Apoptosis vol 11 no 5 pp749ndash764 2006

[41] J Grosse M Wehland J Pietsch et al ldquoShort-term weight-lessness produced by parabolic flight maneuvers altered geneexpression patterns in human endothelial cellsrdquo FASEB Journalvol 26 no 2 pp 639ndash655 2012

[42] M Wehland X Ma M Braun et al ldquoThe impact of alteredgravity and vibration on endothelial cells during a parabolicflightrdquo Cellular Physiology and Biochemistry vol 31 no 2-3 pp432ndash451 2013

[43] D Grimm J Bauer C Ulbrich et al ldquoDifferent responsivenessof endothelial cells to vascular endothelial growth factor andbasic fibroblast growth factor added to culture media undergravity and simulated microgravityrdquo Tissue Engineering A vol16 no 5 pp 1559ndash1573 2010

[44] D Grimm M Infanger K Westphal et al ldquoA delayed typeof three-dimensional growth of human endothelial cells undersimulated weightlessnessrdquo Tissue Engineering A vol 15 no 8pp 2267ndash2275 2009

[45] J H Siamwala SMajumder K P Tamilarasan et al ldquoSimulatedmicrogravity promotes nitric oxide-supported angiogenesis viathe iNOS-cGMP-PKG pathway in macrovascular endothelialcellsrdquo FEBS Letters vol 584 no 15 pp 3415ndash3423 2010

[46] M Mariotti and J A M Maier ldquoGravitational unloadinginduces an anti-angiogenic phenotype in humanmicrovascularendothelial cellsrdquo Journal of Cellular Biochemistry vol 104 no1 pp 129ndash135 2008

[47] C-Y Kang L Zou M Yuan et al ldquoImpact of simulated micro-gravity on microvascular endothelial cell apoptosisrdquo EuropeanJournal of Applied Physiology vol 111 no 9 pp 2131ndash2138 2011

[48] S Cotrupi and J A M Maier ldquoIs HSP70 upregulation crucialfor cellular proliferative response in simulated microgravityrdquoJournal of Gravitational Physiology vol 11 no 2 pp P173ndash1762004

[49] L Buravkova Y Romanov M Rykova O Grigorieva andN Merzlikina ldquoCell-to-cell interactions in changed gravityground-based and flight experimentsrdquo Acta Astronautica vol57 no 2ndash8 pp 67ndash74 2005


[51] J Pietsch J Bauer M Egli et al ldquoThe effects of weightlessnesson the human organism and mammalian cellsrdquo Current Molec-ular Medicine vol 11 no 5 pp 350ndash364 2011

[52] V A Convertino ldquoStatus of cardiovascular issues related tospace flight implications for future research directionsrdquo Respi-ratory Physiology and Neurobiology vol 169 supplement 1 ppS34ndashS37 2009

[53] B J Yates and I A Kerman ldquoPost-spaceflight orthostatic intol-erance possible relationship to microgravity-induced plasticityin the vestibular systemrdquoBrain Research Reviews vol 28 no 1-2pp 73ndash82 1998

[54] X Ma A Sickmann J Pietsch et al ldquoProteomic differencesbetween microvascular endothelial cells and the EAhy926 cellline forming three-dimensional structuresrdquo Proteomics vol 14no 6 pp 689ndash698 2014

[55] C J S Edgell C C McDonald and J B Graham ldquoPermanentcell line expressing human factor VIII-related antigen estab-lished by hybridizationrdquo Proceedings of the National Academyof Sciences of the United States of America vol 80 no 12 pp3734ndash3737 1983

[56] M Boerma G R Burton J Wang L M Fink R E McGeheeJr and M Hauer-Jensen ldquoComparative expression profiling inprimary and immortalized endothelial cells changes in geneexpression in response to hydroxy methylglutaryl-coenzyme Areductase inhibitionrdquo Blood Coagulation and Fibrinolysis vol17 no 3 pp 173ndash180 2006


[57] H F Galley M G Blaylock A M Dubbels and N R WebsterldquoVariability in E-selectin expression mRNA levels and sE-selectin release between endothelial cell lines and primaryendothelial cellsrdquo Cell Biology International vol 24 no 2 pp91ndash99 2000

[58] M Y Kapitonova S L Kuznetsov G R A Froemming etal ldquoEffects of space mission factors on the morphology andfunction of endothelial cellsrdquo Bulletin of Experimental Biologyand Medicine vol 154 no 6 pp 796ndash801 2013

[59] J M Davidson A M Aquino S C Woodward and W WWilfinger ldquoSustained microgravity reduces intrinsic woundhealing and growth factor responses in the ratrdquo FASEB Journalvol 13 no 2 pp 325ndash329 1999

[60] M E Kirchen K M OrsquoConnor H E Gruber et al ldquoEffects ofmicrogravity on bone healing in a rat fibular osteotomymodelrdquoClinicalOrthopaedics andRelatedResearch vol 318 pp 231ndash2421995

[61] D Ingber ldquoHow cells (might) sense microgravityrdquo The FASEBJournal vol 13 no 8 pp S3ndashS15 1999

[62] M Hughes-Fulford and J Boonstra ldquoCell mechanotransduc-tion cytoskeleton and related signalling pathwaysrdquo in CellMechanochemistry Biological Systems and Factors InducingMechanical Stress such as Light Pressure andGravityMMoniciand J W A van Loon Eds pp 75ndash95 Transworld ResearchNetwork Trivandrum India 2010


[64] R G Bacabac D Mizuno and G H Koenderink ldquoMechanicalproperties of living cells on mechanosensing and micrograv-ityrdquo in Cell Mechanochemistry Biological Systems and FactorsInducing Mechanical Stress Such as Light Pressure and GravityM Monici and J W A van Loon Eds pp 23ndash54 TransworldResearch Network Trivandrum India 2010

[65] E Spisni M C Blanco C Griffoni et al ldquoMechanosensingrole of caveolae and caveolar constituents in human endothelialcellsrdquo Journal of Cellular Physiology vol 197 no 2 pp 198ndash2042003

[66] W K Sumanasekera L Zhao M Ivanova et al ldquoEffectof estradiol and dihydrotestosterone on hypergravity-inducedMAPK signaling and occludin expression in human umbilicalvein endothelial cellsrdquo Cell and Tissue Research vol 324 no 2pp 243ndash253 2006

[67] W K Sumanasekera G U Sumanasekera K A Mattingly SM Dougherty R S Keynton and C M Klinge ldquoEstradioland dihydrotestosterone regulate endothelial cell barrier func-tion after hypergravity-induced alterations in MAPK activityrdquoAmerican Journal of Physiology Cell Physiology vol 293 no 2pp C566ndashC573 2007

[68] L Morbidelli N Marziliano V Basile et al ldquoEffect of hyper-gravity on endothelial cell function and gene expressionrdquoMicrogravity Science andTechnology vol 21 no 1-2 pp 135ndash1402009

[69] TKoyamaCKimuraMHayashiMWatanabe Y KarashimaandMOike ldquoHypergravity induces ATP release and actin reor-ganization via tyrosine phosphorylation and RhoA activationin bovine endothelial cellsrdquo Pflugers Archiv European Journal ofPhysiology vol 457 no 4 pp 711ndash719 2009

Research ArticleA Functional Interplay between 5-Lipoxygenase and120583-Calpain Affects Survival and Cytokine Profile of HumanJurkat T Lymphocyte Exposed to Simulated Microgravity

Valeria Gasperi1 Cinzia Rapino23 Natalia Battista45 Monica Bari1

Nicolina Mastrangelo1 Silvia Angeletti6 Enrico Dainese45 and Mauro Maccarrone56

1 Department of Experimental Medicine amp Surgery Tor Vergata University of Rome Via Montpellier 1 00133 Rome Italy2 Faculty of Veterinary Medicine University of Teramo Piazza A Moro 45 64100 Teramo Italy3 StemTeCh Group Via Colle dellrsquoAra 66100 Chieti Italy4 Faculty of Bioscience Technology for Food Agriculture and Environment University of Teramo Piazza A Moro 4564100 Teramo Italy

5 European Center for Brain Research (CERC) IRCCS Santa Lucia Foundation Via del Fosso di Fiorano 64-65 00143 Rome Italy6Center of Integrated Research Campus Bio-Medico University of Rome Via Alvaro del Portillo 21 00128 Rome Italy

Correspondence should be addressed to Valeria Gasperi valeriagasperiuniroma2it Enrico Dainese edaineseuniteitand Mauro Maccarrone mmaccarroneunicampusit

Received 15 May 2014 Accepted 18 August 2014 Published 16 September 2014


Copyright copy 2014 Valeria Gasperi et alThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

A growing body of evidence strongly indicates that both simulated and authentic weightlessness exert a broad range of effects onmammalian tissues and cells including impairment of immune cell function and increased apoptotic deathWe previously reportedthat microgravity-dependent activation of 5-lipoxygenase (5-LOX) might play a central role in the initiation of apoptosis in humanT lymphocytes suggesting that the upregulation of this enzyme might be (at least in part) responsible for immunodepressionobserved in astronauts during space flights Herein we supplement novel information about themolecular mechanisms underlyingmicrogravity-triggered apoptotic cell death and immune system deregulation demonstrating that under simulated microgravityhuman Jurkat T cells increase the content of cytosolic DNA fragments and cytochrome c (typical hallmarks of apoptosis) andhave an upregulated expression and activity of 120583-calpain These events were paralleled by the unbalance of interleukin- (IL-) 2and interferon- (INF-) 120574 anti- and proapoptotic cytokines respectively that seemed to be dependent on the functional interplaybetween 5-LOX and 120583-calpain Indeed we report unprecedented evidence that 5-LOX inhibition reduced apoptotic death restoredthe initial IL-2INF-120574 ratio and more importantly reverted 120583-calpain activation induced by simulated microgravity

1 Introduction

Several studies have shown that authentic space condi-tions markedly alter physiological processes thus leadingto cardiovascular changes [1] loss of bone density [2 3]muscle atrophy [2 4] and immunodepression [5 6] Todate it is well established that cells of the immune systemare severely affected by microgravity conditions [5ndash8] Inparticular alterations observed in astronauts and rodentsflown in space included altered distribution and function ofcirculating leukocytes [9ndash11] lymphocytopenia [12ndash14] and

impaired T cell activation [9 14ndash16] In addition several invivo and in vitro studies reported a weightlessness-dependentalteration of cytokine secretion from T-helper 1 (Th1) and T-helper 2 (Th2) cells that in turn results in a deregulation ofcell-to-cell crosstalk as well as of inflammatory responses [9ndash11 17]

It has been reported that several proinflammatory Th1cytokines including interferon- (INF-) 120574 tumor necro-sis factor- (TNF-) 120573 and interleukin- (IL-) 2 and anti-inflammatory Th2 cytokines like IL-4 and IL-10 as well asleukaemia inhibitory factor (LIF) are related to programmed



cell death (PCD) These glycoproteins indeed are able toinduce or protect cells from apoptosis [18ndash23] so that analternative classification distinguishes them as anti-(LIF IL-2IL-4 IL-10) or proapoptotic (INF-120574) substances A hot topic isthe study of the effect of microgravity (be it real or simulated)on apoptosis of different mammalian cell types includingcerebral vascular smooth muscle [24] thyroid cancer [25]endothelial cells [26] cultured glial cells [27] spermatozoa[28] B lymphocytes [29] and T cells [6 30] In particular5-lipoxygenase (5-LOX) has been proposed as a ldquogravityresponderrdquo which executes the apoptotic events induced bymicrogravity in human lymphocytes [6 30]

Evidence is accumulating that the execution of PCDis finely regulated by a distinct set of signal transductionpathways and catabolic mechanisms (eg mitochondriallysosomal and nuclear alterations lipid modifications andcytosolic calcium accumulation) and recent data providedfirst hints that lipid hydroperoxides impact on PCD [31]Indeed LOX-catalyzed lipid peroxidation has been reportedto be a specific downstream event that triggers apoptosis-inducing factor- (AIF) mediated PCD in primary neuronsin culture and in mice [31] In the same context calpainscleave multiple substrates potentially involved in PCD andincluding cyclin-dependent kinase-5 [32] plasmamembraneCa2+ ATPase isoform-1 [33] and calcineurin [34] Also AIFis a calpain substrate implicated in neuronal death becauseits proteolysis activates PCD through a translocation of AIFitself from the mitochondria to the nucleus [35 36]

Against this background the present study aimed atbetter defining the influence of the space environment onsurvival and cytokine profile of human lymphocytes in orderto identify a possible link between these events In thiscontext we report an unprecedented functional interplaybetween 5-LOX and 120583-calpain in modulating PCD inducedby simulated microgravity


21 Reagents Chemicals were of the purest analytical gradeHuman recombinants IL-2 IL-4 IL-6 IL-10 INF-120574 and LIFcalpain substrate [N-Suc-Leu-Tyr-AMC (7-amido-4-methyl-coumarin)] AA861 (specific inhibitor of 5-LOX) and E64D(specific inhibitor of calpain) were purchased from SigmaChemical Co (St Louis MO USA) Mouse anti-cytochromec antibody was from Cell Signalling Technology Inc (Dan-vers MA USA) mouse anti-calpain-1 was from Calbiochem(Merck Darmstadt Germany) Rabbit anti-LIF anti-IL-2anti-IL-4 anti-IL-6 anti-IL-10 anti-INF-120574 secondary anti-bodies conjugated to horseradish peroxidase (HRP) andenhanced chemiluminescence (ECL) kit were from SantaCruz Biotechnology Inc (Santa Cruz CA USA) Goat anti-rabbit conjugated to alkaline phosphatase (GAR-AP) wasfrom Bio-Rad (Hercules CA USA)

22 Simulated Microgravity Cell Cultures To simulate spaceconditions the rotary cell culture system (RCCS) devel-oped by the National Aeronautics and Space Administration(Washington DC USA) and manufactured by Synthecon

(Houston TX USA) was used Human Jurkat T cells (CloneE6-1) (ATCC Manassas VA USA) were grown in RPMI1640 medium supplemented with 2mM glutamine 25mMsodium pyruvate 100UmL penicillin 100 120583gmL strepto-mycin and 10 heat-inactivated foetal bovine serum Cellswere placed in completely filled 50mL vessels to avoid thepresence of air bubbles that could lead to shear force damageof cells on the RCCS Vessels were rotated at a speed of72 rpm (simulated microgravity and referred to as sim-120583g)as reported [30 37] or cultured at ground gravity (1 g) ascontrols Incubation of 1 g and sim-120583g cells with differentcompounds was performed at 37∘C in an atmosphere of 5CO2 at the indicated concentrations and for the indicated

periods of time

23 Evaluation of PCD PCD was estimated by the cell-deathdetection enzyme-linked immunosorbent assay (ELISA) kit(Boehringer Mannheim Germany) based on evaluationof histone-associated DNA fragments in the cytoplasm aspreviously reported [30]

Cytochrome c release from mitochondria was analyzedas reported [38] Briefly cells were lysed in HB buffer (5mMTris-HCl pH 74 10mM KCl 1mMMgCl

2 and 1mM DTT)

containing protease inhibitor cocktail and centrifuged at1000 timesg for 10min to completely remove nuclei and wholecells The resulting supernatant was centrifuged at 3000 timesgfor 10min then the pellet was saved as membrane-boundorganellar fraction enriched with mitochondria while thesupernatant after centrifugation at 100000 timesg for 40minwas collected as cytosolic fraction These two fractions wereanalyzed for cytochrome c localization by means of ELISAmitochondrial and cytosolic proteins (20 120583gwell) were incu-bated with anti-cytochrome c antibody (diluted 1 500) andafter incubation with a GAR-AP (diluted 1 2000) colourdevelopment of the alkaline phosphatase reaction was mea-sured at 405 nm (A

405 nm) by using 119901-nitrophenyl phosphateas substrate

24 Analysis of 120583-Calpain Activity and Expression Detectionof 120583-calpain mRNA was performed by quantitative reversetranscriptase-polymerase chain reaction (q-RT-PCR) as pre-viously reported [6] Briefly total RNA was extracted fromJurkat cells using the RNeasy extraction kit (Qiagen CrawleyUK) following the manufacturerrsquos instructions RT-PCRreactions were performed using the RT-PCR SuperScript IIIPlatinum Two-Step qRT-PCR Kit (Invitrogen Carlsbad CAUSA) One 120583g total RNA was used to synthesize cDNA with10U120583L SuperScript III reverse transcriptase in the presenceof 2U120583L RNaseOUT 125 120583Moligo (dT) 125 ng120583L randomhexamers 5mMMgCl

2 05mM dNTP mix and DEPC-

treatedwaterThe reactionwas performedusing the followingRT-PCR program 25∘C for 10min 42∘C for 50min 85∘C for5min and then after addition of 01 U120583L of E coli RNase Hthe product was incubated at 37∘C for 20min For expressionstudies target transcripts were amplified in ABI PRISM 7700sequence detector system (Applied Biosystems Foster CityCA USA) Thermal cycling involved 40 cycles of 95∘C for15 sec and 60∘C for 30 sec after initial denaturation for 10min


at 95∘C TaqMan MGB probe was synthesized by AppliedBiosystems (Foster City CA USA) The probe was labelledwith the fluorescent dye 6-carboxyfluorescein at the 51015840 endand a dark quencher at the 31015840 end (Applied Biosystems)Fluorescence was measured after each cycle of PCR andto confirm the quality of isolated RNA and to standardizethe amount of RNA applied glyceraldehyde-3-phosphatedehydrogenase (GAPDH) was used as endogenous controlwith FAMTM dye label and MGB Real-time PCR mixturescontained template cDNA 20x PrimerProbe Mix TaqManMGB Probe with FAMTM dye label no primer limitationMinorGroove Binder andNonfluorescentQuencher Univer-sal PCRMasterMix no AmpErase UNGApplied Biosystems(Foster City CA USA) in a total volume of 25120583L in a 96-wellplate Relative 120583-calpain expression levels were measured byΔΔCT method (PE-Applied Biosystems Sequence DetectorUser Bulletin)

Calpain protein expression was evaluated by Westernblot analysis Briefly cell lysates (20120583gwell) were sub-jected to SDS-PAGE electroblotted onto PVDF membranesincubated with mouse anti-120583-calpain antibody (1 4000)which detects both the full-length (large subunit) and theautoproteolytically cleaved forms of 120583-calpain and detectedwith ECL Calpain quantification was also evaluated throughELISA method by incubating protein lysates (20 120583gwell)with mouse anti-120583-calpain-1 (1 2000) as primary anti-body and HRP-conjugated antibody (1 5000) as secondaryantibody The HRP enzymatic activity was determinedby adding 100 120583Lwell of tetramethylbenzidine containing0002 H

2O2 and the absorbance was read on a microplate

reader (ELISA Ascent Software per Multiskan) at 450 nmAbsorbance values of the samples were within the linearityrange of the ELISA test assessed by calibration curveswith known amounts of 120583-calpain (in the range of 75ndash600 ngwell)

The enzymatic activity of 120583-calpain was measured asreported [39] Briefly cell lysates (40 120583gtest) were incubatedwith 150120583M calpain substrate (N-Suc-Leu-Tyr-AMC) in10mMHepes pH 74 1TritonX-100 and 100 120583MCaCl

2 for

2 hours at 37∘C After incubation hydrolyzed AMC groupswere measured on a fluorimeter LS50B (Perkin-Elmer LifeSciences Inc Boston MA USA) with an excitation filter of380 nm and emission filter of 460 nm

25 5-LOX Activity The activity of 5-LOX (arachido-nateoxygen 5-oxidoreductase EC 1131134) was deter-mined as previously reported [6] Briefly the end prod-uct leukotriene (LT) B

4was extracted from Jurkat cells

(5 times 106 celltest) and quantified at 405 nm by using theLeukotriene B

4EIA Kit (Cayman Chemical Company Ann

Arbor MA USA) and calibration curves drawn according tothe customerrsquos instructions

26 Cytokine Profile Analysis Jurkat cells harvested after 48hours were centrifuged at 200 timesg for 10min to collect cellsand culture medium Cells were lysed in 50mM Tris-HCl(pH 74) containing protease inhibitors and cytokine contentwas quantified by coating proteins (20120583gwell) from whole

lysates overnight in a 96-well ELISA microplate as reported[40] Rabbit anti-LIF anti-IL-2 anti-IL-4 anti-IL-6 anti-IL-10 and anti-INF120574 (diluted 1 500) were used as primaryantibodies GAR-AP (diluted 1 2000) was used as secondaryantibody and absorbance values were read at 405 nm Releaseof LIF and other cytokines from Jurkat cells into the mediumwas quantified through Quantikine Immunoassay kit (RampDSystem Minneapolis MN USA) and a specific MultiproteinProfiling ELISA Kit (SuperArray Bioscience Co Germany)respectively according to the manufacturerrsquos instructions Tothis aim 50120583L of culture medium was used and the contentof each protein was evaluated by comparing A

405 nm values tothose of antigen standard curves (positive controls)

27 Statistical Analysis All values were expressed as meansplusmn SEM of at least three independent experiments Studentrsquosunpaired 119905-test or one-way ANOVA (followed by Bonferronipost hoc analysis) was used to compare quantitative data withnormal distributions and equal varianceThe statistical InStat3 program (GraphPAD Software for Science San DiegoCalifornia) was used and a value of 119875 lt 005 was consideredstatistically significant

3 Results

31 Prolonged Exposure to Simulated Microgravity InducesApoptosis in Human Jurkat T Cells Jurkat T cells wereexposed to simulatedmicrogravity for different times (from 0to 48 hours) and the hallmarks of apoptosis DNA fragmenta-tion and cytochrome c release were analyzed In agreementwith previously reported data [30] RCCS treatment ledto a time-dependent increase of cytosolic DNA fragmentsthat were undetectable after a brief exposure (4 hours) tosimulated microgravity increased after 24 hours (sim2-foldover 1 g cells) and reached a maximum level of sim3-foldover controls 24 hours later (Table 1) Then the subcellularlocalization of cytochrome c upon simulated microgravitywas checked Jurkat cells exposed to weightlessness showed aloss of mitochondrial cytochrome c and a parallel increase inthe cytosolic content with a time-dependence comparable tothat observed for DNA fragmentation (Table 1) ConverselyJurkat cells incubated at 1 g under the same experimental con-ditions did not show significant signs of PCD (Table 1) SinceRCCS treatment for 48 hours yielded a significant increase inPCD we chose to perform all subsequent experiments usingthis time point

32 Prolonged Exposure to Simulated Microgravity Upreg-ulates 120583-Calpain Expression and Activity in Human JurkatT Cells We have previously reported that after 48 hoursof exposure to authentic microgravity human lymphocytesshow increased mRNA levels of 120583-calpain [6] a Ca2+-dependent intracellular cysteine protease that is implicated indifferent physiological functions including cell growth andapoptosis [41] Therefore once established that under ourexperimental conditions Jurkat cells underwent apoptosis wecheckedwhether RCCS treatmentmight engage 120583-calpain Inagreement with our previous data [6] RT-qPCR experiments


Table 1 Time-dependent effect of simulated microgravity on apoptotic markers in Jurkat T cells exposed to simulated microgravity (sim-120583g)or kept at normal gravity (1g)

Parameter 1g 4-hour sim-120583g 24-hour sim-120583g 48-hour sim-120583gDNA fragmentation 100 plusmn 8 107 plusmn 8 270 plusmn18lowast 342 plusmn21lowastlowast

Cytochrome c release (cytosolmitochondria ratio) 100 plusmn 9 111 plusmn 7 347 plusmn41lowast 450 plusmn47lowastlowast

Results are expressed as percentage of 1g cells set to 100 For DNA fragmentation 100 = 0300 plusmn 0030A405 nm for cytochrome c release 100 = 0074 plusmn 0005lowastdenotes 119875 lt 001 versus 1g cells lowastlowastdenotes 119875 lt 0001 versus 1g cells

demonstrated a significant increase of 120583-calpain mRNA insim-120583g Jurkat cells (sim2-fold over 1 g cells) (Figure 1(a))Interestingly upregulation of capn 1 gene which encodes 120583-calpain was paralleled by increased protein content (Figures1(b) and 1(c)) Western blot analysis indeed showed that 48hours of RCCS treatment dramatically increased 120583-calpainprotein levels in particular larger amounts of autocleavedand active fragment of 120583-calpain (sim75 kDa) [42] were foundin sim-120583g whereas no active enzyme was observed in1 g cells (Figure 1(b)) Such a result was further corrobo-rated by ELISA revealing that RCCS almost doubled 120583-calpain protein content after 48 hours (Figure 1(c)) We nextdetermined whether increased mRNA and protein contentmight result in increased enzyme activity By analysing thecleavage of a fluorogenic 120583-calpain substrate we observedan enhanced protease activity in sim-120583g T cells (sim2-foldover 1 g cells) (Figure 1(d)) Specific proteolytic activity ofcalpain was confirmed by the addition of 5 120583M calpastatin(Figure 1(d)) the natural calpain inhibitor [43] Since calpainactivation seemed to be implicated in DNA fragmentation[44 45] we analyzed the effect of E64D a cell permeableand selective inhibitor of the same protease [46] on sim-ulated microgravity-induced PCD As shown in Figure 1(e)inhibition of calpain activity significantly lowered internucle-osomal DNA fragmentation thus preventing weightlessness-induced cell death of T cells

33 Prolonged Exposure to Simulated Microgravity Affects theBalance between Proapoptotic and Antiapoptotic Cytokinesin Jurkat Cells Then we characterized the cytokine profilein Jurkat T cells exposed to simulated microgravity Asdemonstrated by ELISA assay 48 hours of RCCS treatmentsignificantly reduced the synthesis and release of antiapop-totic cytokines like LIF IL-4 and IL-2 while increasingprotein levels of the proapoptotic cytokine INF-120574 (Figure 2)Instead no change in IL-6 and IL-10 content was observedupon simulated microgravity treatment (Figure 2)

Next we went further by investigating whetherRCCS-induced PCD might be related to the unbalancebetween proapoptotic and antiapoptotic cytokines Tothis aim we analyzed apoptosis in Jurkat cells culturedunder simulated microgravity for 48 hours in the presenceof the cytokines that changed upon RCCS exposureNeither LIF nor IL-4 (both at 10 ngmL) reduced cytosolicDNA fragments (Figure 3(a)) and cytochrome c content(Figure 3(b)) on the other hand 10 ngmL IL-2 wasable to protect Jurkat cells from simulated microgravity-triggered cell death since it significantly reduced both DNAfragmentation and cytochrome c release (Figures 3(a) and

Table 2 Time-dependent effect of simulated microgravity on 5-LOX activity in Jurkat cells

Sample LOX activity ( of 1g cells)a

1g cells 100 plusmn 114-hour sim-120583g cells 213 plusmn18lowast

24-hour sim-120583g cells 249 plusmn23lowast

48-hour sim-120583g cells 235 plusmn21lowasta100 of 5-LOX activity = 89 plusmn 7 pg of LTB41 times 10

6 cells lowastdenotes 119875 lt 001versus 1g cells

3(b)) To validate our hypothesis we also analyzed theeffect of INF-120574 (10 ngmL) In agreement with the previousdata (Figure 2) the latter cytokine drastically enhancedRCCS-induced PCD of Jurkat cells (sim 45- and 25-fold over1 g cells and sim-120583g cells resp) (Figures 3(a) and 3(b))

To gain further insights on the evaluation of a possiblerelationship between altered IL-2INF-120574 content and calpainactivation we measured the activity of the latter enzymein the presence of these two cytokines Interestingly IL-2reduced calpain activation due to simulated microgravitywhile INF-120574 did not significantly affect enzyme activity(Figure 3(c))

34 Effect of Inhibition of 120583-Calpain and 5-LOX on Apoptosisand Cytokine Release Since we observed that simulatedmicrogravity triggers apoptosis by altering the content of IL-2and INF-120574 we asked whether such an event might engage 5-LOX which has been proposed as a ldquogravity responderrdquo [30]First we analyzed 5-LOX activity by quantifying the contentof its LTB

4product upon RCCS exposure In agreement

with previous data we found an early increase of 5-LOXactivity (sim2 fold over 1 g cells) with values that remainedunchanged over the whole time period tested (Table 2)Hence we subjected Jurkat cells to simulated microgravityin the presence of 10 120583M AA861 a specific 5-LOX inhibitor[47] As shown inTable 3 we observed that 5-LOX inhibitionreduced DNA fragmentation and cytochrome c release andreverted calpain activation More interestingly it was able torestore the balance between IL-2and INF-120574 that was alteredbyRCCS treatmentThese data seem to suggest that increased5-LOX activity might be (at least in part) responsible foraltered cytokine levels

4 Discussion

The effects of LTs on the secretion of cytokines havebeen reported both in vitro and in vivo [46] Here wedemonstrated that increased LTB

4synthesis upon simulated


0

50

100

150

200

250lowast

1g

120583-c

alpa

in m

RNA

leve

l (

of1

gJu

rkat

cells

)

sim-120583g

(a)

Full length

Cleaved form

1g

80kDa

75kDa

sim-120583g

(b)

0

50

100

150

200

250

lowast

1gng o

f120583-c

alpa

in120583

g pr

otei

n (

of1

gJu

rkat

cells

)

sim-120583g

(c)

Calpastatin0

20

40

60

80

100

120

140

160

180

200

sect

lowast

1g

120583-c

alpa

in ac

tivity

( o

f1g

Jurk

at ce

lls)

mdashsim-120583g

(d)

0

50

100

150

200

250

300

350

400

E64D

sect

lowast

1g mdashDN

A fr

agm

enta

tion

( o

f1g

Jurk

at ce

lls)

sim-120583g

(e)

Figure 1 Effect of simulated microgravity on 120583-calpain expression and activity in Jurkat T cells (a) RT-qPCR analysis of 120583-calpain geneexpression in Jurkat cells exposed to simulated microgravity (sim-120583g) at 37∘C for 48 hours Gene levels were normalized to the housekeepingGAPDH and expressed as percentage of 1 g cells considered as control set to 100 (b)Western blot analysis of protein expression in Jurkat cellstreated as in (a) (c) ELISA analysis of 120583-calpain protein content in Jurkat cells treated as in (a) Results are expressed as percentage of 1 g cellsconsidered as control set to 100 (= 948 plusmn 050 ngper 120583g protein) (d) 120583-calpain activity analysis in Jurkat cells treated as in (a) in absence (minus)or in presence of 5120583Mof calpastatin Results are expressed as percentage of 1 g cells considered as control set to 100 (= 6626 plusmn 365 pmolminper mg protein) (e) DNA fragmentation in Jurkat cells exposed to simulatedmicrogravity for 48 hours in absence (minus) or in presence of 10120583ME64D Values are expressed as percentage of 1 g cells considered as control lowastdenotes 119875 lt 0001 versus 1 g cells sectdenotes 119875 lt 005 versussim-120583g cells


0

50

100

150

200

250

SynthesisRelease

( o

f1g

Jurk

at T

cells

)

LIF IL-2 IL-4 IL-6 IL-10 INF-120574

lowast

lowastlowast lowast

lowast

1g

sim-120583g

Figure 2 Effect of simulated microgravity on cytokine profile ofJurkat T cells Cells were exposed (sim-120583g) or not exposed (1 g) tosimulated microgravity at 37∘C for 48 hours and cytokine synthesis(gray bars) and release (black bars) were analyzed as reported inSection 26 Results are reported as percentage of 1 g cells set to100 For synthesis 100 of IL-2 = 027 plusmn 001 A

405 nm 100 of IL-4 = 034 plusmn 002 A

405 nm 100 of LIF = 022 plusmn 002 A405 nm 100

of IL-6 = 018 plusmn 001 A405 nm 100 of IL-10 = 034 plusmn 002 A

405 nm100 of INF-120574 025 plusmn 002 A

405 nm For release 100 of LIF = 042plusmn 003 Abs

405 nm 100 of IL-2 = 66 plusmn 05 pgmL 100 of IL-4 = 12plusmn 01 pgmL 100 of IL-6 = 20 plusmn 2 pgmL 100 of IL-10 = 25 plusmn03 pgmL 100 INF-120574 = 122 plusmn 01 pgmL lowastdenotes 119875 lt 005 versus1 g cells denotes 119875 lt 001 versus 1 g cells

Table 3 Effect of 5-LOX inhibition on Jurkat T cells exposed for 48hours to simulated microgravity (sim-120583g) or kept at normal gravity(1g)

Parameter 1g sim-120583g sim-120583g +10 120583MAA861

DNA fragmentation 100 plusmn 10 342 plusmn21lowastlowastlowast 250 plusmn11

Cytochrome c release(cytosolmitochondriaratio)

100 plusmn 9 450 plusmn47lowastlowastlowast 230 plusmn24

Calpain activity 100 plusmn 11 177 plusmn9lowastlowastlowast 31 plusmn2

IL-2 protein content 100 plusmn 9 67 plusmn5lowast 93 plusmn8

INF-120574 protein content 100 plusmn 9 179 plusmn15lowastlowast 120 plusmn4

Values are reported as percentage of relative control set to 100 For DNAfragmentation 100 = 030 plusmn 003A405 nm for cytochrome c release 100= 0074 plusmn 0005 for calpain activity 100 = 6626 plusmn 365 pmolmin per mgprotein for IL-2 synthesis 100 = 027 plusmn 001 A405 nm for INF-120574 synthesis100 = 025 plusmn 002 A405 nm

lowastdenotes = 119875 lt 005 versus 1g cells lowastlowastdenotes119875 lt 001 versus 1g cells lowastlowastlowastdenotes 119875 lt 0001 versus 1g cells denotes 119875 lt001 versus sim-120583g cells

microgravity exposure is paralleled by a reduced release ofantiapoptotic cytokines such as LIF IL-4 and IL-2 [19ndash23] as well as by a significant increase of the productionof the proapoptotic cytokine INF-120574 [18 23] These dataare in line with the immunomodulatory role postulatedfor 5-LOX metabolites and especially for LTB

4 Indeed

the latter substance is a powerful chemoattractant forinflammatory cells and induces degranulation superoxide

anion production and adherence of neutrophils to vascularendothelial cells [48] LTB

4has been already demonstrated

to affect the production of several cytokines including IL-1120573 [49 50] IL-2 [51 52] IL-6 [53] INF-120574 [54] IL-4 [55]and IL-10 [56] Moreover LTB

4has been also demonstrated

to modulate the expression of the IL-2 receptor 120573-chain innatural killer cells and in CD8+ lymphocytes [57]

In this context our data add further information onthe mechanism of PCD activation suggesting a crosstalkbetween 5-LOX and 120583-calpain signalling In particular wedemonstrate that exposure of Jurkat T cells to simulatedmicrogravity induced activation of 120583-calpain and 5-LOXOur results suggest that the functional interplay betweenthese two enzymes could be related to the synthesis of aspecific pattern of cytokines In line with this our resultsshow that 5-LOX inhibition (i) reduced DNA fragmentationand cytochrome c release (typical apoptotic markers) (ii)reestablished the initial IL-2INF-120574 ratio and (iii) moreimportantly reverted 120583-calpain activation induced by simu-lated microgravity (Table 3) Furthermore we showed thattreatment of Jurkat T cell with IL-2 whose levels are down-regulated upon simulated microgravity exposure (Figure 2)significantly reduced 120583-calpain activation upon RCCS treat-ment Remarkably the latter result is in agreement with thewell-known antiapoptotic effect of IL-2 [21 23] It shouldbe noted that the lack of any increase in 120583-calpain activityin the presence of 5-LOX inhibitors might be suggestivethat additional and as-yet-unknown 5-LOX products areable to directly activate 120583-calpain Thus in addition to aspecific role of distinct cytokines in modulating the crosstalkbetween 5-LOX and 120583-calpain we can speculate that 5-LOX activation could also induce the formation of specificlipid hydroperoxides that could trigger PCD via 120583-calpainactivation In line with the latter hypothesis hydroperoxidesof cardiolipin and phosphatidylserine have been detectedas byproducts upon PCD [58] Consistently it has beendemonstrated that LOX-induced lipid peroxidation triggersAIF-mediated PCD [31] Indeed although a finely regulatedlipid peroxidationmay have beneficial effects for the cells andthe whole organism leading to different physiological roles ofLOXs (such as eicosanoid synthesis cell maturation and lipidmobilization) when the lipid bilayer of biologicalmembranesis oxidized in an uncontrolled manner (as in the case ofexternal stimuli like microgravity) it may lose its barrierfunction and thus harm the integrity of subcellular organellesor of thewhole cell [59] Consistently an overactivated 5-LOXcan open pore-like structures in mitochondrial membranes[60 61] thus forming the basis for a converging role of thisenzyme in the induction of PCD by unrelated stimuli [59]

Overall our results demonstrate that simulatedmicrogravity-dependent increase in 5-LOX activity regulatessurvival and cytokine release of human T lymphocytes byengaging 120583-calpain

5 Conclusions

Our findings seem to add biochemical support to the immun-odepression observed in astronauts exposed to authentic


0

100

200

300

400

500D

NA

frag

men

t con

tent

( o

f1g

Jurk

at T

cells

)

LIF IL-2 IL-4 INF-120574

lowast

1g mdash

sim-120583g

(a)

0100200300400500600700800900

Cyto

chro

me c

(cyt

osol

mito

chon

dria

ratio

)

lowast

LIF IL-2 IL-4 INF-120574mdash

1gsim-120583g

(b)

0

50

100

150

200

120583-c

alpa

in ac

tivity

( o

f1g

Jurk

at ce

lls) lowast

IL-2 INF-120574mdash

1g

sim-120583g

(c)

Figure 3 Cytokine effects on Jurkat cell apoptosis under simulated microgravity conditions Jurkat T cells were exposed (sim-120583g) or notexposed (1 g) to simulated microgravity in absence (minus) or in presence of the indicated cytokines (10 ngmL) and DNA fragmentation (a)cytochrome c release (b) and 120583-calpain activity (c) were evaluated as reported in Section 2 Results are reported as percentage of 1 g cells setto 100 For DNA fragmentation 100 = 0300 plusmn 0030 A

405 nm for cytochrome c release 100 = 0074 plusmn 0005 for 120583-calpain activity 100 =6626 plusmn 365 pmolmin per mg protein lowastdenotes119875 lt 0001 versus1 g cells denotes 119875 lt 005 versus sim-120583g cells denotes 119875 lt 001 versussim-120583g cells

microgravity for long periods of time (eg InternationalSpace Station crewmembers or astronauts travelling toMars)Taking into account that Jurkat E61 cells are somewhatdifferent from normal human T cells [61] nonetheless theyare considered a valid experimental model especially in thelight of their exaggerated signaling making changes mucheasier to detect Therefore only authentic space conditionswill give a conclusive answer onwhether or not the unbalancebetween proapoptotic and antiapoptotic cytokines due toimpaired 5-LOX and 120583-calpain activities can affect immuneresponse helping to design countermeasures against apopto-sis observed in space

Abbreviations

LOX LipoxygenaseAMC 7-Amido-4-methyl-coumarinAIF Apoptosis-inducing factorRCCS Rotary cell culture system

ECL Enhanced chemiluminescenceELISA Enzyme-linked immunosorbent assayGAPDH Glyceraldehyde-3-phosphate

dehydrogenaseGAR-AP Goat anti-rabbit conjugated to alkaline

phosphatase1 g Ground gravityHRP Horseradish peroxidaseINF-120574 Interferon-120574IL-2 Interleukin-2LIF Leukaemia inhibitory factor(LT)B

4 Leukotriene B

4

sim-120583g Simulated microgravityPCD Programmed cell deathq-RT-PCR Quantitative reverse transcriptase-

polymerase chain reactionTh1 T-helper 1Th2 T-helper 2TNF-120573 Tumor necrosis factor-120573



The authors declare that there is no conflict of interests

Acknowledgment

This investigation was supported under contracts from ldquoDis-turbi del Controllo Motorio e Cardiorespiratoriordquo and ldquoFromMolecules to Manrdquo 2006ndash2009 to Mauro Maccarrone

References

[1] R M Baevsky V M Baranov I I Funtova et al ldquoAuto-nomic cardiovascular and respiratory control during prolongedspaceflights aboard the International Space Stationrdquo Journal ofApplied Physiology vol 103 no 1 pp 156ndash161 2007

[2] L C Shackelford ldquoMusculoskeletal response to space flightrdquoin Principles of Clinical Medicine for Space Flight M R Barrattand S L Pool Eds pp 293ndash306 Springer Science and BusinessMedia New York NY USA 2008

[3] J H Keyak A K Koyama A LeBlanc Y Lu and T F LangldquoReduction in proximal femoral strength due to long-durationspaceflightrdquo Bone vol 44 no 3 pp 449ndash453 2009

[4] EWang ldquoAge-dependent atrophy andmicrogravity travel whatdo they have in commonrdquoThe FASEB Journal vol 13 no 8 ppS167ndashS174 1999

[5] D Williams A Kuipers C Mukai and R Thirsk ldquoAcclimationduring space flight effects on human physiologyrdquo CanadianMedical Association Journal vol 180 no 13 pp 1317ndash1323 2009

[6] N Battista M A Meloni M Bari et al ldquo5-Lipoxygenase-dependent apoptosis of human lymphocytes in the Interna-tional Space Station data from the ROALD experimentrdquo TheFASEB Journal vol 26 no 5 pp 1791ndash1798 2012



[9] B E Crucian R P Stowe D L Pierson and C FSams ldquoImmune system dysregulation following short- vslong-duration spaceflightrdquo Aviation Space and EnvironmentalMedicine vol 79 no 9 pp 835ndash843 2008

[10] B Crucian R Stowe S Mehta et al ldquoImmune system dysreg-ulation occurs during short duration spaceflight on board thespace shuttlerdquo Journal of Clinical Immunology vol 33 no 2 pp456ndash465 2013

[11] B S Crucian S R Zwart S Mehta et al ldquoPlasma cytokineconcentrations indicate that in vivo hormonal regulation ofimmunity is altered during long-duration spaceflightrdquo Journalof Interferon amp Cytokine Research 2014

[12] S K Chapes S J Simske A D Forsman T A Bateman andR J Zimmerman ldquoEffects of space flight and IGF-1 on immunefunctionrdquo Advances in Space Research vol 23 no 12 pp 1955ndash1964 1999

[13] S K Chapes S J Simske G Sonnenfeld E S Miller andR J Zimmerman ldquoEffects of spaceflight and PEG-IL-2 on ratphysiological and immunological responsesrdquo Journal of AppliedPhysiology vol 86 no 6 pp 2065ndash2076 1999

[14] A T Ichiki L A Gibson T L Jago et al ldquoEffects of spaceflighton rat peripheral blood leukocytes and bonemarrowprogenitorcellsrdquo Journal of Leukocyte Biology vol 60 no 1 pp 37ndash43 1996

[15] Z Allebban A T Ichiki L A Gibson J B Jones C CCongdon and R D Lange ldquoEffects of spaceflight on thenumber of rat peripheral blood leukocytes and lymphocytesubsetsrdquo Journal of Leukocyte Biology vol 55 no 2 pp 209ndash2131994

[16] D S Gridley JM Slater X Luo-Owen et al ldquoSpaceflight effectson T lymphocyte distribution function and gene expressionrdquoJournal of Applied Physiology vol 106 no 1 pp 194ndash202 2009

[17] K Felix KWise SManna et al ldquoAltered cytokine expression intissues of mice subjected to simulated microgravityrdquoMolecularand Cellular Biochemistry vol 266 no 1-2 pp 79ndash85 2004

[18] S P Tu M Quante G Bhagat et al ldquoIFN-120574 inhibits gastriccarcinogenesis by inducing epithelial cell autophagy and T-cellapoptosisrdquo Cancer Research vol 71 no 12 pp 4247ndash4259 2011

[19] D Duval B Reinhardt C Kedinger and H Boeuf ldquoRole ofsuppressors of cytokine signaling (Socs) in leukemia inhibitoryfactor (LIF) -dependent embryonic stem cell survivalrdquo TheFASEB Journal vol 14 no 11 pp 1577ndash1584 2000

[20] H Slaets D Dumont J Vanderlocht et al ldquoLeukemiainhibitory factor induces an antiapoptotic response in oligoden-drocytes throughAkt-phosphorylation andup-regulation of 14-3-3rdquo Proteomics vol 8 no 6 pp 1237ndash1247 2008

[21] L R Devireddy and M R Green ldquoTranscriptional programof apoptosis induction following interleukin 2 deprivationidentification of RC3 a calciumcalmodulin binding protein asa novel proapoptotic factorrdquoMolecular and Cellular Biology vol23 no 13 pp 4532ndash4541 2003

[22] L M Minter and B A Osborne ldquoNotch and the survival ofregulatory T cells location is everythingrdquo Science Signaling vol5 no 234 article pe31 2012

[23] F C H Pinto G B Menezes S A L Moura G D Cassali MM Teixeira and D C Cara ldquoInduction of apoptosis in tumorcells as a mechanism of tumor growth reduction in allergicmicerdquo Pathology Research and Practice vol 205 no 8 pp 559ndash567 2009

[24] M-J Xie Y-G Ma F Gao et al ldquoActivation of BK119862119886

channel isassociated with increased apoptosis of cerebrovascular smoothmuscle cells in simulated microgravity ratsrdquo American Journalof Physiology Cell Physiology vol 298 no 6 pp C1489ndashC15002010

[25] D Grimm J Bauer P Kossmehl et al ldquoSimulated microgravityalters differentiation and increases apoptosis in human follicu-lar thyroid carcinoma cellsrdquo The FASEB Journal vol 16 no 6pp 604ndash606 2002

[26] C Y Kang L Zou M Yuan et al ldquoImpact of simulated micro-gravity on microvascular endothelial cell apoptosisrdquo EuropeanJournal of Applied Physiology vol 111 no 9 pp 2131ndash2138 2011


[28] L H Yan Z Hong M G Ying et al ldquoSimulated microgravityconditions and carbon ion irradiation induce spermatogeniccell apoptosis and sperm DNA damagerdquo Biomedical and Envi-ronmental Sciences vol 26 no 9 pp 726ndash734 2013

[29] B Dang Y Yang E Zhang et al ldquoSimulated microgravityincreases heavy ion radiation-induced apoptosis in human Blymphoblastsrdquo Life Sciences vol 97 no 2 pp 123ndash128 2014


[30] M Maccarrone N Battista M Meloni et al ldquoCreating con-ditions similar to those that occur during exposure of cellsto microgravity induces apoptosis in human lymphocytesby 5-lipoxygenase-mediated mitochondrial uncoupling andcytochrome c releaserdquo Journal of Leukocyte Biology vol 73 no4 pp 472ndash481 2003

[31] A Seiler M Schneider H Forster et al ldquoGlutathione per-oxidase 4 senses and translates oxidative stress into 1215-lipoxygenase dependent- and AIF-mediated cell deathrdquo CellMetabolism vol 8 no 3 pp 237ndash248 2008

[32] M-S Lee Y T Kwon M Li J Peng R M Friedlander andL-H Tsai ldquoNeurotoxicity induces cleavage of p35 to p25 bycalpainrdquo Nature vol 405 no 6784 pp 360ndash364 2000

[33] D Guerini B Pan and E Carafoli ldquoExpression purificationand characterization of isoform 1 of the plasmamembrane Ca2+pump Focus on calpain sensitivityrdquo The Journal of BiologicalChemistry vol 278 no 40 pp 38141ndash38148 2003

[34] H-Y Wu K Tomizawa Y Oda et al ldquoCritical role of calpain-mediated cleavage of calcineurin in excitotoxic neurodegener-ationrdquo The Journal of Biological Chemistry vol 279 no 6 pp4929ndash4940 2004

[35] B M Polster G Basanez A Etxebarria J M Hardwick andD G Nicholls ldquoCalpain I induces cleavage and release ofapoptosis-inducing factor from isolated mitochondriardquo Journalof Biological Chemistry vol 280 no 8 pp 6447ndash6454 2005

[36] S A Susin H K Lorenzo N Zamzami et al ldquoMolecularcharacterization of mitochodrial apoptosis-inducing factorrdquoNature vol 397 no 6718 pp 441ndash446 1999

[37] R Mitteregger G Vogt E Rossmanith and D FalkenhagenldquoRotary cell culture system (RCCS) a new method for cultivat-ing hepatocytes on microcarriersrdquo The International Journal ofArtificial Organs vol 22 no 12 pp 816ndash822 1999

[38] MV Catani V Gasperi D Evangelista A F Agro L Aviglianoand M MacCarrone ldquoAnandamide extends platelets survivalthrough CB1-dependent Akt signalingrdquo Cellular and MolecularLife Sciences vol 67 no 4 pp 601ndash610 2010

[39] K G Daniel J S Anderson Q Zhong A Kazi P Gupta andQ P Dou ldquoAssociation ofmitochondrial calpain activationwithincreased expression and autolysis of calpain small subunit inan early stage of apoptosisrdquo International Journal of MolecularMedicine vol 12 no 2 pp 247ndash252 2003

[40] V Gasperi F Fezza N Pasquariello et al ldquoEndocannabinoidsin adipocytes during differentiation and their role in glucoseuptakerdquo Cellular and Molecular Life Sciences vol 64 no 2 pp219ndash229 2007

[41] P Łopatniuk and J M Witkowski ldquoConventional calpains andprogrammed cell deathrdquo Acta Biochimica Polonica vol 58 no3 pp 287ndash296 2011

[42] H Sorimachi S Ishiura and K Suzuki ldquoStructure and phys-iological function of calpainsrdquo Biochemical Journal vol 328article 3 pp 721ndash732 1997

[43] T Uemori T Shimojo K Asada et al ldquoCharacterization ofa functional domain of human calpastatinrdquo Biochemical andBiophysical Research Communications vol 166 no 3 pp 1485ndash1493 1990

[44] J Takano M Tomioka S Tsubuki et al ldquoCalpain mediatesexcitotoxic DNA fragmentation via mitochondrial pathwaysin adult brains evidence from calpastatin mutant micerdquo TheJournal of Biological Chemistry vol 280 no 16 pp 16175ndash161842005

[45] A Rami R Agarwal G Botez and JWinckler ldquo120583-Calpain acti-vation DNA fragmentation and synergistic effects of caspase

and calpain inhibitors in protecting hippocampal neurons fromischemic damagerdquo Brain Research vol 866 no 1-2 pp 299ndash3122000

[46] Y Yang Z H Liu C F Ware and J D Ashwell ldquoA cysteineprotease inhibitor prevents activation-induced T-cell apoptosisand death of peripheral blood cells from human immunodefi-ciency virus-infected individuals by inhibiting upregulation ofFas ligandrdquo Blood vol 89 no 2 pp 550ndash557 1997

[47] Y Tanihiro Y Chieko O Kenkichi et al ldquo235-Trimethyl-6-(12-hydroxy-510-dodecadiynyl)-l4-benzoquinone (AA861)a selective inhibitor of the 5-lipoxygenase reaction andthe biosynthesis of slow-reacting substance of anaphylaxisrdquoBiochimica et Biophysica Acta Lipids and Lipid Metabolism vol713 no 2 pp 470ndash473 1982

[48] J C Eun E E Moore A Banerjee et al ldquoLeukotriene B4and its metabolites prime the neutrophil oxidase and induceproinflammatory activation of human pulmonary microvascu-lar endothelial cellsrdquo Shock vol 35 no 3 pp 240ndash244 2011

[49] G Bonizzi J Piette M P Merville and V Bours ldquoDistinctsignal transduction pathwaysmediate nuclear factor- 120581B induc-tion by IL-1beta in epithelial and lymphoid cellsrdquo Journal ofImmunology vol 159 no 11 pp 5264ndash5272 1997

[50] J Marcinkiewicz A Grabowska K Bryniarski and B MChain ldquoEnhancement of CD4+ T-cell-dependent interleukin-2 production in vitro by murine alveolar macrophages the roleof leukotriene B4rdquo Immunology vol 91 no 3 pp 369ndash374 1997

[51] M Los H Schenk K Hexel P A Baeuerle W Droge and KSchulze-Osthoff ldquoIL-2 gene expression and NF-120581B activationthrough CD28 requires reactive oxygen production by 5-lipoxygenaserdquoThe EMBO Journal vol 14 no 15 pp 3731ndash37401995

[52] J Dornand C Sekkat J-C Mani and M Gerber ldquoLipoxy-genase inhibitors suppress IL-2 synthesis relationship withrise of [Ca++]i and the events dependent on protein kinase Cactivationrdquo Immunology Letters vol 16 no 2 pp 101ndash106 1987

[53] M A Brach S de Vos C Arnold H-J Gruszlig R Mertelsmannand F Herrmann ldquoLeukotriene B4 transcriptionally activatesinterleukin-6 expression involving NK-NB and NF-IL6rdquo Euro-pean Journal of Immunology vol 22 no 10 pp 2705ndash2711 1992

[54] H M Johnson and B A Torres ldquoLeukotrienes Positive signalsfor regulation of 120574-interferon productionrdquo Journal of Immunol-ogy vol 132 no 1 pp 413ndash416 1984

[55] NDugas B Dugas J-P Kolb K Yamaoka J F Delfraiss andCDamais ldquoRole of leukotriene B4 in the interleukin-4-inducedhuman mononuclear phagocyte activationrdquo Immunology vol88 no 3 pp 384ndash388 1996

[56] S Jozefowski R Biedron M Bobek and J MarcinkiewiczldquoLeukotrienes modulate cytokine release from dendritic cellsrdquoImmunology vol 116 no 4 pp 418ndash428 2005

[57] J Stankova N Gagnon and M Rola-PleszczynskildquoLeukotriene B4 augments interleukin-2 receptor-beta(IL-2R120573) expression and IL-2R120573-mediated cytotoxic responsein human peripheral blood lymphocytesrdquo Immunology vol 76no 2 pp 258ndash263 1992

[58] V E Kagan V A Tyurin J Jiang et al ldquoCytochrome c actsas a cardiolipin oxygenase required for release of proapoptoticfactorsrdquo Nature Chemical Biology vol 1 no 4 pp 223ndash2322005

[59] M Maccarrone G Melino and A Finazzi-Agro ldquoLipoxyge-nases and their involvement in programmed cell deathrdquo CellDeath and Differentiation vol 8 no 8 pp 776ndash784 2001


[60] K van Leyen R M Duvoisin H Engelhardt and M Wied-mann ldquoA function for lipoxygenase in programmed organelledegradationrdquo Nature vol 395 no 6700 pp 392ndash395 1998

[61] R R Bartelt N Cruz-Orcutt M Collins and J C D HoutmanldquoComparison of T cell receptor-induced proximal signaling anddownstream functions in immortalized and primary T cellsrdquoPLoS ONE vol 4 no 5 Article ID e5430 2009

Research ArticleHow Microgravity Changes Galectin-3 in Thyroid Follicles

Elisabetta Albi1 Francesco Curcio2 Andrea Lazzarini12

Alessandro Floridi1 Samuela Cataldi1 Remo Lazzarini1 Elisabetta Loreti3

Ivana Ferri3 and Francesco Saverio Ambesi-Impiombato2

1 Laboratory of Nuclear Lipid BioPathology CRABiON 06100 Perugia Italy2 Department of Medical and Biological Sciences University of Udine 33100 Udine Italy3 Institute of Pathologic Anatomy and Histology University of Perugia 06100 Perugia Italy

Correspondence should be addressed to Elisabetta Albi elisabettaalbiyahoocom

Received 22 April 2014 Revised 7 August 2014 Accepted 28 August 2014 Published 11 September 2014


Copyright copy 2014 Elisabetta Albi et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

After long-term exposure to real microgravity thyroid gland in vivo undergoes specific changes follicles are made up oflarger thyrocytes that produce more cAMP and express more thyrotropin-receptor caveolin-1 and sphingomyelinase andsphingomyelin-synthase parafollicular spaces loseC cells with consequent reduction of calcitonin productionHerewe studied fourimmunohistochemical tumor markers (HBME-1 MIB-1 CK19 and Galectin-3) in thyroid of mice housed in the Mouse DrawerSystem and maintained for 90 days in the International Space Station Results showed that MIB-1 proliferative index and CK19are negative whereas HBME-1 and Galectin-3 are overexpressed The positivity of Galectin-3 deserves attention not only for itsexpression but also and especially for its localization Our results highlighted that in microgravity conditions Galectin-3 leavesthyrocytes and diffuses in colloid It is possible that the gravity force contributes to the maintenance of the distribution of themolecules in both basal membrane side and apical membrane side and that the microgravity facilitates slippage of Galectin-3 incolloid probably due to membrane remodelling-microgravity induced

1 Introduction

Galectins are endogenous lectins which constitute a galacto-side-binding protein family of 15 members [1] All membersshare close sequence homology in their carbohydrate recog-nition domain but exhibit different affinities for differentsaccharide ligands and can be bi- or multivalent in terms oftheir ligand-binding activity in cell surface [2] Eukaryoticcell surfaces are dominated by the glycocalyx asim100 nmwidemacromolecular structure consisting of glycans attached toproteins and lipids and N-glycans appear to be the majorligand for galectins [3] Each member of the galectin familycontains at least one domain of about 130 amino acids thisdomain binds to saccharides and is designated the carbohy-drate recognition domain (CRD) Based on the number andorganization of domains in the polypeptides the galectinshave been classified into subfamilies (a) the prototype groupcontains one domain the CRD (b) the chimera group

contains a proline- (P-) and glycine- (G-) rich domain (alsoabout 130 amino acids) which fused amino terminal to theCRD and (c) the tandem repeat group contains two CRDs[4]

Galectin-3 (Gal-3) the only representative of the chimeragroup was first discovered as an IgE-binding protein andcharacterized as a 32 kDa antigen on the surface of murinemacrophages [5] It is mainly a cytosolic protein but caneasily traverse the intracellular and plasma membranes totranslocate into the nucleus or mitochondria or get exter-nalized [6] The protein shuttles between the cytoplasm andnucleus on the basis of targeting signals that are recognized byimportins for nuclear localization and exportin-1 for nuclearexport Depending on the cell type specific experimentalconditions in vitro or tissue location Gal-3 has been reportedto be exclusively cytoplasmic predominantly nuclear ordistributed between the two compartments [7] The pres-ence of Gal-3 in the nucleus is dependent on the integrity



of ribonucleoprotein complexes [8] and a Gal-3-U1 smallnuclear ribonucleoprotein (snRNP) complex has been iden-tified which provides a mechanism of incorporation of theGal-3 into the pre-mRNA splicing substrate [9] In additionGal-3 is secreted via nonclassical pathway outside of the cellindependent on the classical secretory pathway through theendoplasmic reticulumGolgi network thus being found onthe cell surface or in the extracellular space [10] Thus Gal-3is a multifunctional protein which regulates pleiotropic bio-logical functions such as cell growth cell adhesion cell-cellinteractions apoptosis angiogenesis and mRNA processingIts unique structure enables interacting with a plethora ofligands in a carbohydrate dependent or independent manner[6]

In thyroid gland Gal-3 plays a role in the pathogenesis ofwell-differentiated carcinoma particularly in papillary carci-noma [11]Therefore it is one of themarkers most commonlyused to assist in distinguishing thyroid lesions together tohuman bone marrow endothelial cell-1 (HBME-1) as a tumormarker of follicular origin and cytokeratin-19 (CK-19) withgeneral intense and diffuse expression in papillary carcinomaand heterogeneous labeling in carcinoma and in follicularadenoma [12] In addition MIB-1 is useful in evaluatingproliferative activity and in predicting the aggressiveness ofthyroid carcinoma [13]

We have previously demonstrated that microgravityinduces changes in the physiology of the thyroid gland Infact in comparison with control animals thyroids of space-flight animals have a more homogenous structure producemore cAMP and overexpress thyrotropin-receptor (TSHR)caveolin-1 [14] and sphingomyelinase and sphingomyelin-synthase [15] and are characterized by a loss of parafollicularcells with reduction of calcitonin production [16]

Data are not available at the time regarding the evaluationof thyroid tumor markers in microgravity We report for thefirst time the effect of long-term exposure to realmicrogravityenvironment on thyroid HBME-1 MIB-1 CK19 and Gal-3


21 Experimental Design and Animal Care All experimentalprocedures were authorized by the Public Veterinary HealthDepartment of the ItalianMinistry ofHealthThe experimentwas also conducted in accordance with the regulations for thecare and use of laboratory animals and with the guidelines ofthe Japanese Physiological Society Furthermore this studywas also approved by the Committee on Animal Care andUse at Graduate School of Medicine Osaka University (no22-071) Finally the protocol utilized in the study has beenauthorized by the Public Veterinary Health Department ofthe Italian Ministry of Health All experiments were carriedout using male C57BL10J mice (8 weeks old)

22 Microgravity Experiment 3 mice were individuallyhoused in the Mouse Drawer System (MDS) a 116 times 98 times84 cmpayload developed byThales-Alenia Space Italy and alltreatments were performed as previously reported [14] Foodand water were supplied ad libitum The MDS was launched

in the Space Shuttle Discovery within the Space TransportSystem (STS)-128 mission on August 28 2009 It was thenhoused in Japanese Experimental Module (Kibou) on the ISSuntil its return to the Earth by Space Shuttle Atlantis (STS-129 mission) on November 27 2009 Only 1 mouse returnedto the Earth alive after 91 days of space flight

Thyroids were sampled bilaterally from eachmouse killedby inhalation of carbon dioxide at the Life Sciences SupportFacility of Kennedy Space Center within 3-4 hours after land-ing and either processed or frozen immediately accordingto the various experimental protocols The procedure wasapproved by the IACUC protocol n∘ FLT-09-070(KSC)

After the spaceflight experiment the on-ground experi-ment was also carried out at the Vivarium of the AdvancedBiotechnology Center in Genoa Italy One group of 3 micewith the same species sex and age was housed in normalvivarium cage as the laboratory control Amount of food andwater supplementation and environmental conditions weresimulated as the flight group After 3 months thyroids weresampled bilaterally and treated for spaceflight mice

23 Thyroid Tissue Treatment The thyroid lobes were fixedin 4 neutral phosphate-buffered formaldehyde solution for24 h as previously reported [14] Thyroids were dropped withessentially randomorientation in paraffinTheparaffinblockswere sectioned into 4-120583m-thick sections All sections weremounted on silane-coated glass slides Each slide contained apair of sections at a distance equal to 140 120583m Between 5 and14 pairs of sections were sampled excluding the first and thelast sections 2 6 and 10 were used for HBME-1 detectionsections 3 7 and 11 for MIB-1 detection sections 4 8 and12 for CK19 detection and sections 5 9 and 13 for Gal-3detection Tissue sectionswere deparaffinized and rehydratedthrough a series of xylene and ethanol washes

24 Immunohistochemical Analysis For immunohistochem-ical analysis Bond Dewax solution was used for removal ofparaffin from tissue sections before rehydration and immun-ostaining on the Bond automated system (Leica BiosystemsNewcastle Ltd UK) as previously reported [17] Immunos-taining detection was performed according to Bancroft andStevens [18] by using HBME-1 and Ki-67 (MIB-1 clone)from Dako (Milano Italy) and CK19 and Gal-3 antibodiesand Bond Polymer Refine Detection from Leica Biosystems(Newcastle Ltd UK) The observations were performed byusing invertedmicroscopy EUROMEXFE 2935 (EDAmhemThe Netherlands) equipped with a CMEX 5000 camerasystem (40x magnification)The analysis of the tissue sectionsize was performed by ImageFocus software

25 Statistical Analysis The experiments have been con-ducted on the thyroid of 1 animal for themicrogravity experi-ment (the only ones that returned alive from themission) and3 control animals for the microgravity experiment (vivarium1) Median and range of sections 2 6 and 10 (HBME-1) ofsections 3 7 and 11 (MIB-1) of sections 4 8 and 12 (CK19)and of sections 5 9 and 13 (Gal-3) were given



Prolonged space flights are known to elicit changes in humancardiovascular musculoskeletal immune and nervous sys-tems whose functions are regulated by the thyroid gland[14] The structure of thyroid shows the presence of folliclescontaining colloid and surrounded by a single layer of thyroidepithelial cells or thyrocytes that produce the metabolicallyactive iodothyronines and parafollicular spaces with thyroidC cells that produce calcitonin [19] We have previouslyreported that thyrocyte cells in culture delay cell growth andenter into a proapoptotic state after long stay on the Inter-national Space Station (ISS) [20] In vivo experiments on theboard of ISS showed that thyroid of spaceflightmice hasmoreordered follicles with thicker thyrocytes containing increasednuclear volume [14] and reduction of interfollicular spacewith loss of C cells [16] in comparison with thyroid glandof ground mice In order to verify whether the structuralchanges of the thyroid gland inmicrogravity conditions couldlead to pathological conditions in this study we investigatedthe immunoexpression of markers known to be related toclinical outcome The limitation of the present paper is thatonly 1 mouse survived to the 91-day spaceflight Howeverthe MDS experiment was a unique opportunity to study themicrogravity long-term exposure effects on several tissuesof an animal model and to collect interesting observationsthat could prepare the field to future experimentsThe resultsshowed that microgravity gives a nonspecific staining inthe colloid during MIB-1 CK19 and Gal-3 immunohisto-chemistry analysis absent in control samples It is reallyhard to pinpoint the reason but it is possible to hypoth-esize an increase of membrane permeability microgravity-dependent on the basis of the observation that at the endof the spaceflight endothelial cells display profound changesindicating cytoskeletal lesions and increased cell membranepermeability [21] MIB-1 and CK19 immunopositivity do notshow changes in thyroid of spaceflight mice in comparisonwith control animals (Figure 1(a)) Differently the immunos-taining is present for HBME-1 and it is very strong for Gal-3(Figure 1(a)) Alshenawy demonstrated that no single markeris completely sensitive and specific for diagnosis of thyroidlesions but only their combination [22] with Gal-3 + HBME-1 was considered the best combination for distinguishingbenign from malignant lesions [23] In thyroid of spaceflightmice the structure of thyroid follicles is more organized thanthat of the control animals [14] and thyrocytes delay theirgrowth [20] and MIB-1 is negative So it is very difficult atthe moment to consider that the expression of HBME-1 andGal-3 markers is linked to tumor transformation Howeverthe possibility that HBME-1 and Gal-3 overexpression mightindicate a premaligne state of thyroid tissue cannot beexcluded by considering that in microgravity follicles aremade up of cells 2 times larger and colloid darker [14] similarto those of papillary carcinoma [24] Our result showedthat HBME-1 is present only in trace in thyroid of controlmice maintained in the vivarium whereas it appears evidentafter space flight with well-defined localization in thyrocytes(Figure 1(a)) Median and range value of immunopositivesurface area is 462 (551ndash457)mm2 and its ratio in relation

Control Microgravity

HBME-1

MIB-1

CK19

Galectin-3

(a)

ControlMicrogravity

HBME-1 Galectin-30

02

04

06

08

1

Posit

ive s

urfa

ceto

tal s

urfa

ce

(b)

Figure 1 Effect of microgravity on HBME-1 MIB-1 CK19 andGalectin-3 (a) Marker detection in thyroid tissue by immunohis-tochemical staining ldquoControlrdquo mice maintained in vivarium cagesldquomicrogravityrdquo experimental mice in space environment (b) Ratiobetween the immunopositive surface and total surface of thyroidlobe The values are expressed as median and range of two sectionsas reported inMaterial andMethods 40xmagnificationThe arrowsindicate positive areas



Galectin-3

Figure 2 Localization of Galectin-3 in colloid Gal-3 immunohistochemical staining ldquoControlrdquo mice maintained in vivarium cagesldquomicrogravityrdquo experimental mice in space environment 40x magnification The arrows indicate positive areas

to total surface is reported in Figure 1(b) Gal-3 labelling ispresent in some of follicular thyrocytes of control animals andit increases strongly in spaceflight mice (Figure 1(a)) Medianand range value of immunopositive surface area is 172 (199ndash125) mm2 in the control and 794 (859ndash700) in microgravityby increasing 467 times the positive surfacetotal surfaceratio (Figure 1(b)) The presence of Gal-3 in normal thyroidtissue has already been demonstrated [25] Our data show anoverexpression in microgravity We do not have support inthe literature since this is the first study on observation ofthe behavior of thyroid pathological markers in micrograv-ity Nevertheless Grosse et al demonstrated that NF-120581B isoverexpressed and different factors that interact with it aredifferentially regulated under altered gravity conditions [26]In addition spaceflight conditions change gene expressionprofile in thyroid cancer cells [27] Therefore microgravityinfluences gene expression and consequently protein contentHowever the positivity of Gal-3 deserves attention not onlyfor its expression but also and especially for its localizationOur results highlighted that in microgravity conditions Gal-3 leaves thyrocytes and diffuses in colloid (Figure 2) It ispossible that microgravity induces changes of cell membranethat in turn facilitates the escape of Gal-3 accumulated inthyrocytes We have previously demonstrated that thyrocytesin culture (FTRL-5 cell line) release thyrotropin receptorlinked to cholesterol and sphingomyelin in culture mediumduring space missions by indicating a depletion of lipid raftsand consequently cell membrane remodelling [20] Clarke etal told about microgravity-induced decrease in membraneorder [28] and Hsu et al localized Gal-3 in membrane lipidrafts [29] It is possible to suppose that Gal-3 overexpressed inthyrocytes moves into colloid due to the modification of thecell membrane following the variation of gravity force It hasbeen demonstrated that Gal-3 is mainly a cytosolic proteinbut it shuttles to the nucleus or extracellular space the basisof targeting signals [6] Here we do not have specific stainingin these locations but the molecules move in the oppositedirection they do not protrude from the basal membrane of

thyrocytes towards the extracellular space but from the apicalmembrane to the colloid On the other hand Delacour et alsuggested a direct role of Gal-3 in apical sorting as a sortingreceptor [30] It is possible that the gravity force contributes tothe maintenance of the distribution of the molecules in bothbasal membrane side and apical membrane side and that themicrogravity facilitates slippage of Gal-3 in colloid

4 Conclusion

To our knowledge this is the first study correlating thyroidtumor markers with long stay mice in microgravity con-ditions Here we found higher expression of HBME-1 andGal-3 in comparison with ground gravity However MIB-1 proliferative index and CK19 are negative Gal-3 usuallypresent in cytoplasm nuclei and extracellular space leavesthyrocytes and diffuses in colloid probably due to membraneremodelling-microgravity induced



Acknowledgment

This work has been partially supported by grants fromAgenzia Spaziale Italiana (ASI)

References

[1] S H Barondes V Castronovo D N W Cooper et al ldquoGalec-tins a family of animal 120573-galactoside-binding lectinsrdquo Cell vol76 no 4 pp 597ndash598 1994

[2] S Di Lella V Sundblad J P Cerliani et al ldquoWhen galectinsrecognize glycans from biochemistry to physiology and backagainrdquo Biochemistry vol 50 no 37 pp 7842ndash7857 2011


[3] A Grigorian S Torossian and M Demetriou ldquoT-cell growthcell surface organization and the galectin-glycoprotein latticerdquoImmunological Reviews vol 230 no 1 pp 232ndash246 2009

[4] J Hirabayashi and K-I Kasai ldquoThe family of metazoan metal-independent 120573-galactoside-binding lectins structure functionand molecular evolutionrdquo Glycobiology vol 3 no 4 pp 297ndash304 1993

[5] M K Ho and T A Springer ldquoMac-2 a novel 32000 Mr mousemacrophage subpopulation-specific antigen defined by mono-clonal antibodiesrdquo Journal of Immunology vol 14 pp 1221ndash12281982

[6] T Funasaka A Raz and P Nangia-Makker ldquoNuclear transportof galectin-3 and its therapeutic implicationsrdquo Seminars inCancer Biology vol 27C pp 30ndash38 2014

[7] K C Haudek P G Voss L E Locascio J L Wang and R JPatterson ldquoA mechanism for incorporation of galectin-3 intothe spliceosome through its association with U1 snRNPrdquo Bio-chemistry vol 48 no 32 pp 7705ndash7712 2009

[8] J G Laing and J L Wang ldquoIdentification of carbohydratebinding protein 35 in heterogeneous nuclear ribonucleoproteincomplexrdquo Biochemistry vol 27 no 14 pp 5329ndash5334 1988

[9] K C Haudek K J Spronk P G Voss R J Patterson J LWangand E J Arnoys ldquoDynamics of galectin-3 in the nucleus andcytoplasmrdquo Biochimica et Biophysica Acta vol 1800 no 2 pp181ndash189 2010

[10] J Dumic S Dabelic andM Flogel ldquoGalectin-3 an open-endedstoryrdquo Biochimica et Biophysica Acta vol 1760 no 4 pp 616ndash635 2006

[11] T Yoshii H Inohara Y Takenaka et al ldquoGalectin-3 maintainsthe transformed phenotype of thyroid papillary carcinomacellsrdquo International Journal of Oncology vol 18 no 4 pp 787ndash792 2001

[12] L L deMatos A B del Giglio C O Matsubayashi M de LimaFarah andM A da Silva Pinhal ldquoExpression of ck-19 galectin-3 and hbme-1 in the differentiation of thyroid lesions systematicreview and diagnostic meta-analysisrdquoDiagnostic Pathology vol7 no 1 article 97 2012

[13] P Kjellman G Wallin A Hoog G Auer C Larsson and JZedenius ldquoMIB-1 index in thyroid tumors a predictor of theclinical course in papillary thyroid carcinomardquoThyroid vol 13no 4 pp 371ndash380 2003

[14] M A Masini E Albi C Barmo et al ldquoThe impact of long-term exposure to space environment on adult mammalianorganisms a study onmouse thyroid and testisrdquo PLoSONE vol7 no 4 Article ID e35418 2012

[15] E Albi F Curcio R Spelat et al ldquoObserving the mouse thyroidsphingomyelin under space conditions a case study from theMDS mission in comparison with hypergravity conditionsrdquoAstrobiology vol 12 no 11 pp 1035ndash1041 2012

[16] E Albi F Curcio R Spelat et al ldquoLoss of parafollicular cellsduring gravitational changes (microgravity hypergravity ) andthe secret effect of pleiotrophinrdquoPLoSONE vol 7 no 12 ArticleID e48518 2012

[17] E Albi F Curcio R Spelat et al ldquoThe thyroid lobes the differ-ent twinsrdquo Archives of Biochemistry and Biophysics vol 518 no1 pp 16ndash22 2012

[18] J D Bancroft and A Stevens EdsTheory and Practice of Histo-logical Techniques Churchill Livingstone New York NY USA1996

[19] C C Capen and S L Martin ldquoThe effects of xenobiotics onthe structure and function of thyroid follicular and C-cellsrdquoToxicologic Pathology vol 17 no 2 pp 266ndash293 1989

[20] E Albi F S Ambesi-Impiombato M Peverini et al ldquoThyrot-ropin receptor and membrane interactions in FRTL-5 thyroidcell strain in microgravityrdquo Astrobiology vol 11 no 1 pp 57ndash64 2011


[22] H A Alshenawy ldquoUtility of immunohistochemical markers indiagnosis of follicular cell derived thyroid lesionsrdquo Pathology ampOncology Research 2014

[23] H A Saleh J Feng F Tabassum O Al-Zohaili M Husainand T Giorgadze ldquoDifferential expression of galectin-3 CK19HBME1 and Ret oncoprotein in the diagnosis of thyroidneoplasms by fine needle aspiration biopsyrdquo CytoJournal vol6 article 18 2009

[24] R V Lloyd D Buehler and E Khanafshar ldquoPapillary thyroidcarcinoma variantsrdquo Head and Neck Pathology vol 5 no 1 pp51ndash56 2011

[25] J Feilchenfeldt M Totsch S-Y Sheu et al ldquoExpression of ga-lectin-3 in normal and malignant thyroid tissue by quantitativePCR and immunohistochemistryrdquo Modern Pathology vol 16no 11 pp 1117ndash1123 2003

[26] J Grosse M Wehland J Pietsch et al ldquoGravity-sensitive sig-naling drives 3-dimensional formation of multicellular thyroidcancer spheroidsrdquoThe FASEB Journal vol 26 no 12 pp 5124ndash5140 2012

[27] X Ma J Pietsch M Wehland et al ldquoDifferential gene expres-sion profile and altered cytokine secretion of thyroid cancer cellsin spacerdquo FASEB Journal vol 28 no 2 pp 813ndash835 2014

[28] M S Clarke C R Vanderburg and D L Feeback ldquoThe effectof acute microgravity on mechanically-induced membranedamage and membrane-membrane fusion eventsrdquo The Journalof Gravitational Physiology vol 8 no 2 pp 37ndash47 2001

[29] D K Hsu A I Chernyavsky H-Y Chen L Yu S A Grandoand F-T Liu ldquoEndogenous galectin-3 is localized in membranelipid rafts and regulates migration of dendritic cellsrdquo Journal ofInvestigative Dermatology vol 129 no 3 pp 573ndash583 2009

[30] D Delacour C I Cramm-Behrens H Drobecq A Le BivicH Y Naim and R Jacob ldquoRequirement for galectin-3 in apicalprotein sortingrdquo Current Biology vol 16 no 4 pp 408ndash4142006

Research ArticleThe Influence of Simulated Microgravity on PurinergicSignaling Is Different between IndividualCulture and Endothelial and Smooth Muscle Cell Coculture

Yu Zhang12 Patrick Lau3 Andreas Pansky1 Matthias Kassack2

Ruth Hemmersbach3 and Edda Tobiasch1

1 Department of Natural Sciences Bonn-Rhine-Sieg University of Applied Sciences 53359 Rheinbach Germany2 Institute of Pharmacology and Medical Chemistry University of Dusseldorf 40225 Dusseldorf Germany3 Institute of Aerospace Medicine German Aerospace Center 51147 Cologne Germany

Correspondence should be addressed to Edda Tobiasch eddatobiaschh-brsde

Received 25 April 2014 Revised 30 June 2014 Accepted 23 July 2014 Published 28 August 2014


Copyright copy 2014 Yu Zhang et alThis is an open access article distributed under theCreativeCommonsAttribution License whichpermits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Exposure to microgravity conditions causes cardiovascular deconditioning in astronauts during spaceflight Until now no specificdrugs are available for countermeasure since the underlying mechanism is largely unknown Endothelial cells (ECs) and smoothmuscle cells (SMCs) play key roles in various vascular functions many of which are regulated by purinergic 2 (P2) receptorsHowever their function in ECs and SMCs under microgravity conditions is still unclear In this study primary ECs and SMCswere isolated from bovine aorta and verified with specific markers We show for the first time that the P2 receptor expressionpattern is altered in ECs and SMCs after 24 h exposure to simulated microgravity using a clinostat However conditioned mediumcompensates this change in specific P2 receptors for example P2X7 Notably P2 receptors such as P2X7 might be the importantplayers during the paracrine interaction Additionally ECs and SMCs secreted different cytokines under simulated microgravityleading into a pathogenic proliferation and migration In conclusion our data indicate P2 receptors might be important playersresponding to gravity changes in ECs and SMCs Since some artificial P2 receptor ligands are applied as drugs it is reasonable toassume that they might be promising candidates against cardiovascular deconditioning in the future

1 Introduction

Exposure to microgravity conditions during space missionsinduces a variety of health issues in astronauts includ-ing bone loss muscle atrophy decreased immune activityand cardiovascular deconditioning [1ndash3] The cardiovasculardeconditioning is very likely caused by the dysfunction ofthe major vascular cells endothelial cells (ECs) and smoothmuscle cells (SMCs) ECs build up the monolayer coatinginner surface of blood vessels Layers of SMCs arranged infibers support the EC monolayer by providing contractionand relaxation of vessels [4] Importantly the interactionbetween ECs and SMCs has been shown to be a key playerin human cardiovascular physiology [5] ECs are sensitive tomechanical stress and they secret cytokines inhibiting SMCproliferation [6] Purinergic receptors can bind extracellular

nucleotides such as ATP [7 8] and they are crucial playersin regulating a series of physiological and pathological car-diovascular processes such as atherosclerosis hypertensionand vascular pain [9 10] Purinergic receptors are dividedinto P1 receptors and P2 receptors [11] P2 receptors can besubdivided into P2X receptors that are ion channels and P2Yreceptors that are G protein-coupled receptors [12] Untilnow seven P2X (P2X1-7) and eight P2Y (P2Y1 2 4 6 1112 13 and 14) have been characterized However the roleof extracellular nucleotides on vascular cell function undermicrogravity condition is still unknown

Recent publications have shown that cytoskeletonarrangement gene expression of extracellular matrix andcell surface adhesion molecules in ECs were altered after22 seconds and 24 h exposure to microgravity [13ndash16]ECs formed tubes after culturing for longer term (7 days)



(a)

Rotation axis18mm

6mm

(b) (c)

Figure 1 Cell Culture in aClinostat to SimulateMicrogravityConditionsThe clinostat (a) was used to simulate themicrogravity environmentby rotating cells Only cells grown in the 6mm area in the middle of culture slide (b) received the optimal simulated microgravity and werethus harvested using a special cell scraper (c)

under simulated microgravity conditions using a RandomPositioning Machine (RPM) [17] On the other hand SMCsshowed suppressed proliferation and an enhanced rate ofapoptosis after 72 h exposure to simulatedmicrogravity usinga rotating wall vessel (RWV) [18] However these findingswere encountered when single cell type such as endothelialor smooth muscle cells was cultured under real or simulatedmicrogravity Considering ECs already showed to secretedifferent cytokines under simulated microgravity using RPM[19] the important interactions between ECs and SMCsunder microgravity condition should be evaluated and thusrequire investigations

In this study an indirect cell coculture model wasestablished by culturing SMCs with EC-conditionedmediumand vice versa The P2 receptor expression pattern was ana-lyzed and compared under three conditions normal gravitycontrol (1 g) simulated microgravity (MG) and simulatedmicrogravity with conditioned medium For the simulationof microgravity a fast rotating clinostat was used in whichthe cells were quickly rotated around one axis perpendicularto the direction of gravity [20] The influence of condi-tioned medium collected from normal gravity and simulatedmicrogravity on cell proliferation and migration was alsoinvestigated

2 Methods

21 Isolation and Characterization of Bovine Aortic Endothe-lial and Smooth Muscle Cells Bovine aorta was cut longi-tudinal into 5 cm sections and divided again into rectanglesafter removing residual and connective tissues The cut aortainto type I collagenase (10mgmL in PBS) coated cell culturedishes with the inner layer (endothelium) attached to thecollagenase and incubated for 60min at 37∘C The aorticendothelial cells were slightly scraped with a cell scraperand put onto gelatin (in PBS (1 vv)) coated culture plates[21] Medium was added to the freshly scraped cells and theplates were then incubated at 37∘C 5CO

2under humidified

conditions Aortic smooth muscle cells were isolated byobtaining the media layer through removal of the outer layer

and scraping off the endothelial cellsThemedia layer was cutinto 2mm times 2mm sections and put into cell culture dish for2 h without medium to allow these sections adhering tightlyto the surface [22] The medium was added and the pieceswere incubated at 37∘C 5CO

2under humidified conditions

for up to a week to let SMCs migrate and proliferate from thetissue pieces to the surface of culture dish

22 Cell Culture The cells were cultured in DMEMmedium(Merck Millipore Berlin Germany) supplied with 10 FCSand 1 penicillinstreptomycin ECs and SMCs were splitand seeded at a density of 5000 cellscm2 after they reacheda level of 80ndash90 confluence ECs and SMCs with passagenumber 2ndash4 were used The cell line human microvascularendothelial cell-1 (HMEC-1) C2C12 (ATCC number CRL-1772) andMG-63 (ATCC number CRL-1427) and U-87 MG(ATCC number HTB-14) were cultured in DMEM mediumand subsequently used as positive control

23 Clinostat Experiments The fast-rotating 2D clinostatused in this study was originally developed by the Institute ofAerospace Medicine German Aerospace Center (DLR) (seeFigure 1(a)) It has 6 parallel horizontal axes each for fixationfor up to 4 slide flasks ECs and SMCswere seeded at a densityof 10000 cellscm2 onto 9 cm2 cell culture slide flasks (NuncThermo Fisher Scientific Langenselbold Germany) Whenthey reached a confluence level of 60ndash70 the culture flaskswere filled up completely with DMEM medium To avoidshear stress and thus the induction of respective metabolicchanges in signal transduction pathways for example apop-tosis air bubbles were removed carefully The flasks wereinserted on the clinostat and rotated at 60 rpm for 24 h in theCO2incubator at 37∘CControls were also filledwithmedium

and placed simultaneously under normal gravityCells from the whole flask were first used to analyze the

P2 receptor expression pattern that altered subtypes couldbe distinguished from unaffected Later according to theclinostat principle only cells exposed to minimal g-forceswere taken for further analysis This means that only cellsfrom the middle of the flask were taken (see Figure 1(b))


Under a defined constant speed of 60 rpm the maximalresidual acceleration at an area of 6mm provided an optimalquality of simulated microgravity (le00121 g) [23] Thus onlycells within this 6mm area were isolated to evaluate alteredP2 subtypes for both gene and protein expression in detail Tomaintain the cells accurately and consistently in the center aspecial chamber consisting of two cover plates attached to abottom plate [23] was used to allow slide insertion withoutwiping off the cell layer A corresponding cell scraper wasused to scratch the cells from the specific 6mm width areain the center (Figure 1(c))

24 ConditionedMedium To investigate a possible paracrineinfluence on P2 receptor expression ECs and SMCs wereseeded in a density of 2500 cellscm2 Cell growth mediumwas collected when they were 80ndash90 confluent Theconditioned medium (CM) was composed out of cell growthmedium and normal DMEM medium in a ratio of 1 2 onrespective cell type The SMC-conditioned medium was sub-sequently fully added into the culture slide with ECs and setof a 24 h clinorotation as a group of ECMG+CMThe ECs innormal gravity group (EC 1 g) and in clinorotation but filledwith normal DMEM medium (EC MG) were set simultane-ouslyThe similar experimentswere set for SMCswith normalgravity (SMC 1 g) clinorotation (SMC MG) and clinorota-tion filled with EC-conditioned medium (SMCMG + CM)

To evaluate a possible paracrine effect on cell prolif-eration and migration cell growth medium was collectedfrom cells cultured 24 h in normal gravity and cultured24 h under clinorotation respectively The used conditionedmedium (CM) was composed out of cell growth mediumand normal DMEM medium in a ratio of 1 2 on respectivecell type The ECs were subsequently treated with normalDMEM medium SMC-conditioned medium from normalgravity (CM SMC + 1 g) and clinorotation (CM SMC +MG) separately for proliferation or migration assays Similarexperiments were set and performed with SMCs as wellAll experiments were performed with samples from threecows

25 RNA Isolation and Semiquantitative PCR RNAwas extracted after clinorotation using a Ribozol RNAreagent (Amresco OH USA) cDNAs were synthesizedfrom 20 120583g total RNA by using Revert Aid ReverseTranscriptase and oligo-dT primer (Thermo FisherScientific MA USA) Primers for P2 receptors EC andSMC specific markers in the human and bovine system weredesigned and shown in the supplementary data availableat httpdxdoiorg1011552014413708 The RT-PCRconditions such as annealing temperature and magnesiumconcentration are given in the supplementary data as well 1of agarose gels were set up to evaluate the RT-PCR productsAs positive control RNA extracts from the cell lines HMEC-1 MG-63 C2C12 and U-87 MG were used for respective P2receptor subtypes given in the supplementary data

26 Western Blot Analysis The proteins were extracted fromthe cells in a protein lysis buffer (Cell Signaling Technology

MA USA) and subsequently centrifuged at 22000 g for5min at 4∘C to remove cellular debris After boiling for5min the lysate samples were separated by a 12 SDS-PAGEelectrophoresis and electrotransferred to a PVDFmembraneThe membrane was blocked in TBST containing 5 BSAand incubated with anti-P2X7 P2Y1 P2Y2 P2Y11 VEGFR2VE-cadherin PECAM-1 calponin SMA-120572 MYH-11 (1 500)or GAPDH antibodies (1 5000) (Santa Cruz BiotechnologyCA USA) overnight at 4∘C The membranes were washedthree times with TBST and incubated with the secondaryantibodies (1 5000) (CALBIOCHEM CA USA) for 60minat RT After washing with TBST immune-detection wasaccomplished by using the Luminata Forte Western HRPsubstrate (MerckMilliporeMAUSA) and imageswere takenusing Bio-Rad Chemidoc system

27 Immunofluorescence The cells were fixed in 4paraformaldehyde for 15min Cells were incubated withprimary anti-VEGFR2 VE-cadherin PECAM-1 calponinSMA-120572 MHY-11 P2X7 P2Y1 P2Y2 and P2Y11 (SantaCruz Biotechnology CA USA) diluted in a ratio of 1 100in antibody dilution buffer containing 1 BSA and 02Triton-X-100 in PBS at 4∘C overnight After rinsing withPBS 3 times cells were stained with FITC-labeled anti-goator rabbit antibody respectively (1 100) (Southern BiotechAL USA) at RT for 60min Cell nuclei were stained withDAPI (Sigma MO USA) and the cell cytoskeleton waslabeled using rhodamine (1 2000) (Life Technologies CAUSA) After washing with PBS fluorescent signals wereanalyzed with an Axio Observer D1 fluorescence microscope(Carl Zeiss Germany) or a FW300 confocal fluorescentmicroscope (Olympus Japan) respectively

28 DiI-ac-LDL Uptake ECs were incubated with 10 120583gmLDiI-labeled acetylated-low density lipoprotein (DiI-ac-LDL)(Biomedical Technologies Inc MA USA) for 4 hours at 37∘Cand investigated with a fluorescent microscope (Carl ZeissGermany) at a wavelength of 565 nm After staining withLDL cells were fixed with 4 formaldehyde for 15min andsubsequently incubated with DAPI (1 10000 in PBS) andrinsed with PBS Images were taken with an Axio ObserverD1 fluorescent microscopy (Carl Zeiss Germany)

29 Proliferation and Wound Assay For the proliferationassay 20000 ECs were seeded separately in each well of 12-well plates ECs were grown in DMEM medium in a SMC-conditioned medium in normal gravity (SMC CM + 1 g) andin a SMC-conditioned medium in simulated microgravity(SMC CM + MG) (see Section 24 for conditioned mediumdetails) ECs incubated with the respective medium weresubsequently obtained after 24 h and 48 h under normalgravity incubation and numbers were calculated Similarlyexperiments were set for SMCs SMCs were cultured inDMEM medium in an EC-conditioned medium in normalgravity (EC CM + 1 g) and EC-conditioned medium insimulatedmicrogravity (ECCM+MG) for 24 h and 48 h Cellnumber in each well was counted


For the wound assay ECs and SMCs (10000cm2) wereseeded and grown to 80ndash90 confluence A straight scratchinjury was made using a sterile 1mL pipette tip on 6-wellplates The ECs were incubated for 24 h at 37∘C in a CO

2

incubator with normal DMEM medium SMC-conditionedmedium in normal gravity (SMC CM + 1 g) and SMC-conditioned medium in simulated microgravity (SMC CM+ MG) On the other hand SMCs were cultured in DMEMmedium EC-conditionedmedium innormal gravity (ECCM+ 1 g) and EC-conditioned medium in simulated micrograv-ity (ECCM+MG) Hydroxyurea (5mM)was added to inhibitcell proliferation Images were taken using a phase contrastmicroscope (Carl Zeiss Germany)The numbers of migratedcells in three individual areas were calculated and quantifiedusing Image J software (NIH)

210 Statistical Analysis Statistical analysis was applied forthe experiments using the Microsoft Office program Excel2010 and SPSS 120 Data are shown as means plusmn standarddeviation Experiments were repeated at least three times forthree donors which are given as 119899 = number of experimentsThe probability (119875) value was calculated using LSH test toassess differences between two groups Levels of significancewere labeled as follows lowast119875 le 005 lowastlowast119875 le 001 and lowastlowastlowast119875 le0001 Significance was given with the appropriate number ofasterisks or in numbers

3 Results

31 Characterization of Primary ECs and SMCs from BovineAorta The isolated ECs showed positive gene expressionof endothelial cell markers VEGFR2 VE-cadherin andPECAM-1 whereas SMCs positively expressed smooth mus-cle cell markers SMA-120572 calponin andMYH-11 (Figure 2(a))Western blot experiments confirmed the gene expressiondata on the protein level ECs positively expressed VEGFR2VE-cadherin and PECAM-1 while SMCs were positive forcalponin SMA-120572 and MYH-11 (Figure 2(b)) Importantlyboth gene and protein data showed that ECs were negativefor SMC markers except a weak band found in calponinSMCs were negative for the three tested endothelial cellmarkers These results indicate that isolated ECs and SMCswere without major cross contaminations The fluorescentstaining data further confirmed the results from the RT-PCRandWestern blot analysis (Figures 2(c) and 2(d)) In additionisolated ECs also showed the typical endothelial activity byuptaking LDL (Figure 2(e))

32 P2 Receptor Expression in ECs after 24 h Simulated Micro-gravity with and without SMC-Conditioned Medium Allfifteen P2 receptors were analyzed for their gene expressionby RT-PCR In the first experiment RNAwas collected in oneset of clinorotation experiments from the whole culture flaskAll P2 receptors were expressed in ECs with the exceptionof P2X3 and P2Y6 Next to this P2X5 P2Y4 P2Y11 andP2Y14 were upregulated while P2X7 P2Y1 and P2Y4 weredownregulated on the gene expression level in ECs under24 h simulated microgravity condition (MG) induced by

clinorotation if compared to ECs under normal gravity (1 g)(Figure 3(a)) In a further set of clinorotation experimentsthe conditioned medium collected from SMCs grown undernormal gravity condition (see Section 24) was added toECs For this experiment only cells grown in the 6mmdiameter area of the center were taken to isolate RNA andprotein RT-PCR data showed that although the expressionof P2X7 and P2Y1 was decreased after clinorotation P2X7in ECs showed an increase on the gene level when culturedin SMC-conditioned medium P2Y11 protein expression inECs was upregulated and further increased also on theSMC conditioned medium compared to P2X7 (Figure 3(b))Western blot and fluorescence confirmed the change of P2X7on protein level (Figures 3(c) and 3(d))

33 P2 Receptor Expression in SMCs after 24 h SimulatedMicrogravity with and without EC-Conditioned MediumIdentical operational steps were undertaken to investigateSMCs under simulated microgravity In SMCs all P2 recep-tors were expressed except P2X3 P2X7 P2Y6 and P2Y11After 24 h clinorotation RT-PCR showed an increased geneexpression of P2X4 P2X7 and P2Y2 whereas P2X2 P2Y1and P2Y14 were downregulated in SMCs under simulatedmicrogravity condition (MG) if compared to the SMCsunder normal gravity (1 g) (Figure 4(a)) After adding EC-conditionedmedium (see Section 24) within clinostat exper-iment clinorotation induced an upregulation of P2X7 geneexpression in SMCs as revealed by RT-PCR (Figure 4(b))Interestingly P2X7 showed a decreased gene expression afteradding EC-conditioned medium compared to its increasewithout EC-conditioned medium under 24 h clinorotationP2Y1 was upregulated in SMCs under simulated micro-gravity however conditioned medium showed no effect onits expression Gene level alterations of P2X7 and P2Y2were confirmed on the protein level by Western blot orfluorescent staining however P2Y1 showed an increasingprotein expression in simulated microgravity and with EC-conditioned medium (Figures 4(c) and 4(d))

34 Proliferation and Migration of ECs Cultured with SMC-Conditioned Medium Collected under Normal Gravity andSimulated Microgravity Conditioned medium from SMCscollected after 24 h normal gravity and after 24 h simu-lated microgravity was used to culture ECs evaluating theparacrine influence of SMCs on EC proliferation and migra-tion SMC-conditioned medium from normal gravity (SMCCM+ 1 g) did not have a significant influence on ECprolifera-tion after 24 h but caused a decrease of EC numbers after 48 hSMC-conditioned medium collected after simulated micro-gravity (SMC CM + MG) inhibited EC proliferation signif-icantly after both 24 h and 48 h respectively (Figure 5(a))To mimic a wound in the endothelium a straight scratchthrough the cells was set The SMC-conditioned mediumcultured under normal gravity (SMCCM+ 1 g) and simulatedmicrogravity (SMCCM+MG) condition was added to studyEC migration capacity The conditioned medium from MGenhanced EC migration after 24 h and even more signifi-cantly after 48 h in the presence of hydroxyurea (Figures 5(b)


pc EC SMC

PECAM-1

VEGFR2

MYH-11

Calponin

SMA-120572

VE-cadherin

120573-Actin

(a)

pc EC SMC

PECAM-1

VEGFR2

Calponin

SMA-120572

VE-cadherin

120573-Actin

(b)

NegDAPI VEGFR2DAPI

VE-cadherinDAPI PECAM-1DAPI

(c)

NegDAPI CalponinDAPI

MYOCDDAPI MYH11DAPI

(d)

Dil-ac-LDLDAPI

(e)

Figure 2 Characterization of Bovine Endothelial and Smooth Muscle Cells The isolated cells were verified with the EC specific markersVEGFR2 VE-cadherin PECAM-1 and the SMC specific markers calponin SMA-120572 MYH-11 by RT-PCR (a) and Western blot (b) GAPDHserved as internal control The endothelial (c) and smooth muscle cells (d) were also identified with the above-mentioned markers viafluorescent staining The isolated endothelial cells were further examined for the typical endothelial activity of LDL up-take (e) All picturesare representative of one cow sample out of three

and 5(c)) Figure 5 is a representative of example one of threecows The numbers of proliferating and migrating ECs fromthe three individual cows are given in the supplementary data

35 Proliferation andMigration of SMCs Grown in ECDerivedConditioned Medium Collected after 24 h under NormalGravity and SimulatedMicrogravity Experimentswith SMCswere performed in a comparable manner as described forECs The conditioned medium collected from EC grownunder normal gravity (ECCM+ 1 g) reduced the proliferationof SMCs However the conditioned medium collected fromEC grown under simulated microgravity condition (EC CM+ MG) compensated this effect (Figure 6(a)) Conditionedmedium under simulated microgravity induced SMCmigra-tion after 48 h but inhibited it after 24 h (Figures 6(b) and6(c)) Figure 6 is a representative of example one of threeindividual cows The numbers of proliferating and migrating

SMCs from the three individual cows are given in thesupplementary data

4 Discussion

In this study we showed for the first time that several specificP2 receptor expressionswere altered on gene and protein levelafter 24 h under simulated microgravity condition as shownin Figure 7 Culturing ECs with SMC-conditioned mediumunder normal gravity and vice versa can compensate the P2receptor expression change such as P2X7

Similar to the findings of Wang and colleagues [24]our data showed that ECs and SMCs expressed differentP2 receptors on cell membrane P2X4 P2Y1 P2Y2 andP2Y11 were predominantly expressed in ECs while P2X1 andP2Y2 were strongly expressed in SMCs Macrovascular and


P2Y1

P2Y2

P2Y6

P2Y14

P2Y13

P2Y12

P2Y11

P2Y4

GAPDH

pc

P2X5

P2X6

P2X4

P2X1

P2X3

P2X2

P2X7EC

MG

EC1

g

(a)

GAPDH

pc

P2Y13

P2Y11

P2Y1

P2Y2

P2X7

P2X5

EC M

G

ECM

G+

CM

EC1

g

(b)

P2Y1

pc

P2Y2

P2Y11

GAPDH

P2X7

P2X5

EC M

G

ECM

G+

CM

EC1

g

(c)

P2Y1 F-actin Merge P2Y11 F-actin MergeP2X7 F-actin Merge

MG

MG + CM

1g

(d)

Figure 3 P2 Receptor Expression in Endothelial Cells after 24 h under Normal Gravity and SimulatedMicrogravity All cells on the surface offlasks were isolated for RT-PCR P2X5 P2Y4 P2Y11 P2Y13 were up-regulated and P2X7 P2Y1 and P2Y2were down-regulated in the ECs after24 h in the clinostat (a) Cells grown within 6mm of the center had the optimal simulated microgravity condition and were therefore isolatedto confirm the above P2 receptor alteration on the RNA (b) and protein (c) level after 24 h simulated microgravity with and without SMC-conditioned medium collected under normal gravity P2X5 and P2Y11 were up-regulated in ECs but P2X5 up-regulation was not significanton protein level P2X7 P2Y1 P2Y2 were down-regulated on both gene and protein level The SMC-conditioned medium can compensate thedecrease of P2X7 expression but cause no significant effect on the alteration of P2Y1 P2Y2 and P2Y11The fluorescent staining confirmed theprotein change of P2X7 P2Y1 and P2Y11 (d)

microvascular ECs have shown the several functional differ-ences such as matrix metalloproteinase expression [25] andbeta-adrenergic regulation of transendothelial permeability[26]The expression of P2X3 and P2Y4 is low in bovine aorticECs if compared to the control HMEC-1 which suggests thatmacrovascular ECsmight differ in the P2 receptor expressionpattern compared to microvascular ECs

The P2 receptor expression patterns of ECs and SMCshave already been shown to play an important role invarious cardiovascular functions For example in controlling

vascular tone ATP and UTP released from ECs act on P2Y1P2Y2 and P2Y4 leading to the production of NO and sub-sequent vasodilatation Simultaneously ATP released by thesympathetic nerve acts on P2X1 P2X2 and P2X4 resultingin vasoconstriction [9] We found that the expression ofP2X2 and P2X4 in SMCs was significantly increased afterclinorotation indicating to maybe more vasoconstrictionNext to this P2Y1 and P2Y2 expressions were decreasedwhich suggests NO productionmight decrease and cause lessvasodilatation Kang and colleagues found that 72 h exposure


P2X5

P2X6

P2X4

P2X1

P2Y1

P2Y2

P2Y6

P2Y14

P2Y13

P2Y12

GAPDH

P2Y11

P2Y4

P2X7

P2X3

P2X2pc SM

C M

G

SMC1

g

(a)

P2Y2

P2Y1

GAPDH

P2X4pc

P2X7

P2Y14

P2X2

SMC

MG

SMC

MG+

CM

SMC1

g

(b)

pc

P2Y1

GAPDH

P2Y2

P2Y14

P2X7

SMC

MG

SMC

MG+

CM

SMC1

g

(c)

P2X7 F-actin Merge P2Y1 F-actin Merge P2Y2 F-actin Merge

MG

MG + CM

1g

(d)

Figure 4 P2 Receptor Expression in SmoothMuscle Cells after 24 h under Normal Gravity and SimulatedMicrogravityThe experiments forSMCwere performed similarly to those for the endothelial cells above All cells on the surface of flasks were isolated for RT-PCR P2X4 P2X7and P2Y2were up-regulated whereas P2X2 P2Y1 and P2Y14 were down-regulated in the SMCs after 24 h clinorotation (a) Cells grownwithin6mm of the center had the optimal simulated microgravity condition and were thus isolated to confirm the above P2 receptor alteration onboth RNA (b) and protein (c) level after 24 h clinorotation with and without EC-conditioned medium from normal gravity P2X2 and P2X4showed no significantly changed P2X7 and P2Y2 was up-regulated P2Y1 and P2Y14 were down-regulatedThe EC-conditioned medium cancompensate for the increase of P2X7 and P2Y2 expression but no significant effect was observed on P2Y1 and P2Y14 (d)

to clinorotation led to a decreased proliferation but increasedthe rate of apoptotic SMCs Additionally the SMC phenotypewas induced and transferred from the contractive to thesynthetic type [18] Our data showed that the expression ofP2X7 and P2Y2 was altered differentially between ECs andSMCs under simulated microgravity which indicates thatthey could be the key P2 receptor subtypes responding tothe change of gravity To point out P2X7 has an importantrole in cell apoptosis and can activate a series of downstreamsignals due to several protein kinase binding sites on itslong intracellular tail [10] A mechanical force such as shear

stress can induce endothelial cell apoptosis that might beregulated through P2X7 It is of interest that P2X7 which wasdownregulated in ECs was upregulated in SMCsThis changecould be compensated by adding conditioned medium fromthe other cell type Such compensation was not found in theother P2 receptor subtypes which might point to P2X7 as amajor player in the interaction between ECs and SMCs insimulated microgravity with respect to apoptosis regulation

Various proteins of endothelial cells are altered under realor simulated microgravity such as F-actin [27] tubulin [28]cell adhesion molecules [16] integrins [27] eNOS [29] and


0

10000

20000

30000

40000

50000

60000

70000

80000

0

(h)24 48

DMEM SMC CM + MG

lowastlowast

lowastlowast

lowast

lowastlowastlowast

SMC CM + 1g

DMEM SMC CM + MG

200

150

100

50

0

lowastlowast

lowastlowastlowast

SMC CM + 1g

DMEM SMC CM + MGSMC CM + 1g

(a)

(b)

(c)

Figure 5 Effect of SMC-ConditionedMedium on EC Proliferation andMigration To evaluate the EC proliferation normal DMEMmediumand conditioned medium of SMCs after normal gravity (SMC CM + 1 g) and simulated microgravity (SMC CM +MG) was added for a 24 hand 48 h culture period The conditioned medium from SMCs after normal gravity showed a decrease of EC numbers after 48 h SMC-conditioned medium under simulated microgravity inhibited EC proliferation significantly after 24 h and 48 h incubation (a) To evaluateEC migration ECs were scratched and normal DMEM medium SMC-conditioned medium under normal gravity and under simulatedmicrogravity was added for a 24 h culture period (b) Conditioned medium from SMCs under simulated microgravity (SMC CM + MG)enhanced EC migration significantly if compared to the normal DMEMmedium and SMC-conditioned medium under normal gravity (c)

iNOS [30] In line with that ECs also showed a decreasedproliferation rate increased apoptosis [28 31] and migration[29] in simulated microgravity However these data wereobserved based on cultured ECs as single cell type eitherunder real or simulated microgravity An EC and SMCcoculture model was successfully created via EC-conditionedmedium culturing SMCs and vice versa In addition to theparacrine effect on P2 receptor expression such as P2X7our results showed that this effect found under simulatedmicrogravity could influence EC or SMC behavior for cellproliferation and migration SMC proliferation has beendemonstrated to be a crucial process in atherosclerosis sincemigrated and proliferating SMCs form amajor cell type in theplaque [32]

In healthy vessels ECs secrete cytokines that inhibit SMCproliferation and form a monolayer to block small moleculesfrom blood that might cause SMC proliferation We foundthat conditioned medium collected from ECs in normalgravity inhibited SMCproliferation but conditionedmediumcollected from ECs after clinorotation was not able to doso On the other hand EC damage or dysfunction is oneof the first steps during the development of pathologicalchange in atherosclerosis The conditioned medium of SMCgrown in simulated microgravity reduced EC proliferationEnhanced apoptosis was observed when only ECs werecultured in simulatedmicrogravity by Infanger and coauthors[28] They found several caspases such as caspase-3 andcaspase-9 activated after simulated microgravity treatment


0

10000

20000

30000

40000

50000

60000

70000

80000

0 24 48

(h)

DMEM EC CM + MG

lowast

lowastlowast

lowastlowastlowast

lowastlowastlowast

lowastlowastlowast

EC CM + 1g

DMEM

200

150

100

50

0

250

lowastlowast

lowastlowast

lowastlowastlowast

EC CM + MG

EC CM + 1g

DMEM EC CM + MGEC CM + 1g

(a)

(b)

(c)

Figure 6 Effect of EC-Conditioned Medium on SMC Proliferation and Migration To investigate the SMC proliferation normal DMEMmedium and EC-conditioned medium under normal gravity (EC CM + 1 g) and simulated microgravity (EC CM + MG) was added andSMCs were subsequently incubated for 24 h and 48 h Conditionedmedium from EC grown under normal gravity (ECCM+ 1 g) significantlyinhibited SMC proliferation after 24 h and 48 h incubation and conditioned medium from EC grown under simulated microgravity (EC CM+ MG) led to a significant decrease of SMC numbers after 24 h which is not obvious after 48 h (a) To evaluate SMC migration SMCs werescratched and cultured under normal DMEMmedium EC-conditionedmedium under normal gravity and under simulatedmicrogravity for24 h (b) EC-conditioned medium under normal gravity (EC CM + 1 g) inhibited the SMCmigration significantly Whereas EC-conditionedmedium under simulated microgravity (EC CM +MG) enhanced migrated SMC numbers (c)

Apoptosis might be induced by activation of NF-120581B via thePI3KAkt pathway [28 31]These data suggest that astronautsmay be more prone to suffer from cardiovascular diseasessuch as atherosclerosis during space missions and paracrineeffects between ECs and SMCs might be the key factors inthis process On the other hand migration of ECs is thefirst step in angiogenesis and a major factor in metastasisIn addition it also plays an important role in restenosisin the vascular system after application of a stent Thereare evidences that microgravity can promote angiogenesisin both macrovascular and microvascular ECs when onlyECs were cultured under simulated microgravity [29 30]Our data showed an enhanced number of migrated ECswhen cultured with SMC-conditioned medium derived after

clinostat application compared to the DMEM control andSMC-conditioned medium collected after 24 h exposure tonormal gravityThis indicates the effect ofmicrogravitymightenhance the angiogenesis via both autocrine and paracrinesignals

Contradictory observations have been demonstrated inseveral publications for example EC migration increasedin simulated microgravity both in this study and in thestudy of Siamwala and colleagues [30] while Versari andcolleagues found a decreased EC migration under simulatedmicrogravity [33] One explanation for these findings mightbe that different endothelial cells were used such as primaryendothelial cells from an artery or umbilical vein or anendothelial cell line (EAhy926) which might give a different


SMC

EC EC

SMC

Clinorotation(MG)

Paracrine effect

(protective)

Paracrine effect

(pathogenic)

P2X7 P2Y1

P2Y2

P2Y14

P2X5

P2X7

P2Y11P2Y4

P2Y1

P2Y2P2Y13

P2X4

1g

Figure 7 Scheme of P2 Receptor Alteration and the Postulated Paracrine Effect in ECs and SMCs under Simulated Microgravity SeveralP2 receptor expressions were altered in ECs and SMCs under 24 h simulated microgravity condition using a clinostat To point out that theexpression P2X7 and P2Y2 was altered differentially between ECs and SMCs under simulated microgravity Especially the change of P2X7in ECs was compensated under SMC-conditioned medium and vice versa The conditioned medium collected under simulated microgravityshowed the pathogenic influence of EC and SMC proliferation and migration if compared to condition medium from normal gravity

response due to its immortalization and thus prolongedtime in culture Another explanation could be that differentdevices were applied to simulate microgravity conditionssuch as the clinostat the RPM and the RWV Differentequipment might produce different qualities of microgravityas well as a different amount and quality of shear stressduring rotation [20] Clinorotationwas shown to produce thelowest shear forces and the central area used in our studyhas an optimized simulated microgravity environment [23]Furthermore different ECs from different body parts wereused Macrovascular andmicrovascular ECs already revealeda difference in promoting angiogenesis under realsimulatedmicrogravity conditions which is regulated via the iNOS-cGMP-PKG pathway in macrovascular ECs but via eNOS-PI3K-Akt in microvascular ECs [29 30] Taken togetherthe simulated microgravity data independently of ground-based facility we use have to be approved and verified in realmicrogravity for a final statement on the outcome

5 Conclusion

Our data show for the first time that P2 receptor gene andprotein expression in both ECs and SMCs were altered undersimulated microgravity SMC-conditionedmedium collectedunder simulated microgravity influenced some P2 receptorexpressions as well as proliferation and migration of ECs andvice versa Additionally proliferation and migration of ECsand SMCs differed between conditioned medium collectedunder normal gravity and under simulated microgravityThese data suggest that the extracellular environment suchas paracrine signals is an important factor and cannot beignored considering the impact of microgravity on vascularcells Since some P2 receptor artificial ligands are already

applied as drugs for cardiovascular patients specific P2receptor ligandsmight be reasonable candidates to investigat-ing their function for cardiovascular deconditioning undermicrogravity in the future


All authors declare that there are no conflicts of interest andagree with the contents of the paper

Acknowledgments

This work was supported by the Bundesministerium furBildung und Forschung- (BMBF-) FHprofUnt [FKZ03FH012PB2 to ET] NRW FH-Extra [FKZ z1112fh012 toET] DAAD PPP Vigoni [FKZ 314-vigoni-dr and FKZ54669218 to ET] and BMBF-AIF [FKZ 1720X06 to ET]the fellowship of YZ was funded by China ScholarshipCouncil [no 20100602024] and the Helmholtz Space LifeSciences Research School (SpaceLife) SpaceLife is funded inequal parts by the Helmholtz Association and the GermanAerospace Center (DLR)

References


[2] A R Hargens and S Richardson ldquoCardiovascular adaptationsfluid shifts and countermeasures related to space flightrdquo Respi-ratory Physiology and Neurobiology vol 169 pp S30ndashS33 2009

[3] M Coupe J O Fortrat I Larina G Gauquelin-Koch CGharib and M A Custaud ldquoCardiovascular deconditioning


from autonomic nervous system to microvascular dysfunc-tionsrdquo Respiratory Physiology and Neurobiology vol 169 ppS10ndashS12 2009

[4] E L Schiffrin ldquoThe endothelium and control of blood ves-sel function in health and diseaserdquo Clinical and InvestigativeMedicine vol 17 no 6 pp 602ndash620 1994


[6] C A Limbach M Lange M Schulze and E TobiaschldquoRecent patents on biomedical applications for the treatment ofatherosclerosisrdquo Recent Patents on RegenerativeMedicine vol 2no 2 pp 75ndash102 2012

[7] N Zippel C A Limbach N Ratajski et al ldquoPurinergicreceptors influence the differentiation of human mesenchymalstem cellsrdquo Stem Cells and Development vol 21 no 6 pp 884ndash900 2012

[8] Y Zhang D Khan J Delling and E Tobiasch ldquoMechanismsunderlying the osteo- and adipo-differentiation of humanmesenchymal stem cellsrdquoThe ScientificWorld Journal vol 2012Article ID 793823 14 pages 2012

[9] G Burnstock ldquoControl of vascular tone by purines and pyrim-idinesrdquo The British Journal of Pharmacology vol 161 no 3 pp527ndash529 2010

[10] G Burnstock ldquoPurinergic signallingrdquo British Journal of Phar-macology vol 147 no 1 pp S172ndashS181 2006

[11] G Burnstock ldquoPurine and pyrimidine receptorsrdquo Cellular andMolecular Life Sciences vol 64 no 12 pp 1471ndash1483 2007

[12] Y Zhang and E Y Tobiasch ldquoThe role of purinergic receptorsin stem cells in their derived consecutive tissuesrdquo in AdultStem Cell Standardization P di Nardo Ed pp 73ndash98 RiverPublishers 2011






[18] H Y Kang Y B Fan A Q Sun X L Jia and X Y DengldquoSimulated microgravity exposure modulates the phenotype ofcultured vascular smooth muscle cellsrdquo Cell Biochemistry andBiophysics vol 66 no 1 pp 121ndash130 2013

[19] C Griffoni S Di Molfetta L Fantozzi et al ldquoModificationof proteins secreted by endothelial cells during modeled lowgravity exposurerdquo Journal of Cellular Biochemistry vol 112 no1 pp 265ndash272 2011


[21] S M Schwartz ldquoSelection and characterization of bovine aorticendothelial cellsrdquo In Vitro vol 14 no 12 pp 966ndash980 1978

[22] M A Stepp M S Kindy C Franzblau and G E SonensheinldquoComplex regulation of collagen gene expression in culturedbovine aortic smooth muscle cellsrdquo The Journal of BiologicalChemistry vol 261 no 14 pp 6542ndash6547 1986

[23] P Eiermann S Kopp J Hauslage R Hemmersbach R Gerzerand K Ivanova ldquoAdaptation of a 2-D clinostat for simulatedmicrogravity experiments with adherent cellsrdquo MicrogravityScience and Technology vol 25 pp 153ndash159 2013

[24] L Wang L Karlsson S Moses et al ldquoP2 receptor expressionprofiles in human vascular smooth muscle and endothelialcellsrdquo Journal of Cardiovascular Pharmacology vol 40 no 6 pp841ndash853 2002

[25] C J Jackson and M Nguyen ldquoHuman microvascular endothe-lial cells differ from macrovascular endothelial cells in theirexpression of matrix metalloproteinasesrdquo International Journalof Biochemistry and Cell Biology vol 29 no 10 pp 1167ndash11771997

[26] S Zink P Rosen and H Lemoine ldquoMicro- and macrovascularendothelial cells in 120573-adrenergic regulation of transendothelialpermeabilityrdquo The American Journal of PhysiologymdashCell Physi-ology vol 269 no 5 pp C1209ndashC1218 1995



[29] F Shi Y C Wang T Z Zhao et al ldquoEffects of simulated micro-gravity on human umbilical vein endothelial cell angiogenesisand role of the PI3K-Akt-eNOS signal pathwayrdquo PLoS ONE vol7 no 7 Article ID e40365 2012


[31] C Kang L Zou M Yuan et al ldquoImpact of simulated micro-gravity on microvascular endothelial cell apoptosisrdquo EuropeanJournal of Applied Physiology vol 111 no 9 pp 2131ndash2138 2011

[32] V J Dzau R C Braun-Dullaeus and D G Sedding ldquoVascularproliferation and atherosclerosis new perspectives and thera-peutic strategiesrdquo Nature Medicine vol 8 no 11 pp 1249ndash12562002

[33] S Versari A Villa S Bradamante and J A M MaierldquoAlterations of the actin cytoskeleton and increased nitric oxidesynthesis are common features in human primary endothelialcell response to changes in gravityrdquo Biochimica et BiophysicaActa vol 1773 no 11 pp 1645ndash1652 2007

Review ArticleHuman Locomotion under Reduced Gravity ConditionsBiomechanical and Neurophysiological Considerations

Francesca Sylos-Labini12 Francesco Lacquaniti123 and Yuri P Ivanenko2

1 Centre of Space Bio-Medicine University of Rome Tor Vergata Via Montpellier 1 00133 Rome Italy2 Laboratory of Neuromotor Physiology IRCCS Santa Lucia Foundation Via Ardeatina 306 00179 Rome Italy3 Department of Systems Medicine University of Rome Tor Vergata Via Montpellier 1 00133 Rome Italy

Correspondence should be addressed to Yuri P Ivanenko yivanenkohsantaluciait

Received 24 April 2014 Accepted 12 June 2014 Published 28 August 2014


Copyright copy 2014 Francesca Sylos-Labini et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Reduced gravity offers unique opportunities to study motor behavior This paper aims at providing a review on current issues ofthe known tools and techniques used for hypogravity simulation and their effects on human locomotion Walking and runningrely on the limb oscillatory mechanics and one way to change its dynamic properties is to modify the level of gravity Gravity has astrong effect on the optimal rate of limb oscillations optimal walking speed andmuscle activity patterns and gait transitions occursmoothly and at slower speeds at lower gravity levels Altered center of mass movements and interplay between stance and swingleg dynamics may challenge new forms of locomotion in a heterogravity environment Furthermore observations in the lack ofgravity effects help to reveal the intrinsic properties of locomotor pattern generators and make evident facilitation of nonvoluntarylimb stepping In view of that space neurosciences research has participated in the development of new technologies that can beused as an effective tool for gait rehabilitation

1 Introduction

Life evolved in the presence of gravity which has two majorimpacts on motor functions specific body orientation inspace and antigravity muscle tone and specific rules ofmotion in the gravity field Gravity plays an essential role interrestrial locomotion The dominant hypothesis regardingtemplates for bipedal walking in the gravity field is thependular mechanism of walking up to intermediate speedsand the bouncing mechanism of running up to the highestspeeds attainable [1]The inverted pendulum-likemechanismof energy exchange taking place during walking would beoptimized at slower speeds in reduced gravity [2 3] Despiteour intuitive appreciation for the influence of gravity wedo not fully understand how gravity interacts with otherforces such as inertia to affect many biological and physicalprocesses and what type of gait andor limb synchronization(trot gallop lateral sequencewalk pace skipping etc) wouldevolve at other gravity levels

Understanding locomotion characteristics is critical forthose working in the area of gait biomechanics and neu-rophysiology f pattern generation networks and of exer-cise countermeasures for astronauts Many researchers haveinvestigated the effects of reducing and eliminating gravityon locomotive kinematics and kinetics [4ndash8] Others havestudied locomotion in actual weightlessness or hypogravity[9 10] The techniques have included supine and erect cablesuspension parabolic aircraft flights water immersion andcentrifugal methods [6] Increased knowledge of locomotionkinematics kinetics muscular activity patterns and sensoryfeedback modulation may help to facilitate more effectiveexercise countermeasures develop innovative technologiesfor gait rehabilitation and provide new insights into ourunderstanding of the physiological effects of gravity In thisreview we will consider the known tools and techniquesused for hypogravity simulation and their effects on humanlocomotion



2 Methods and Apparatuses for ReducedGravity Simulation

Spaceflights are the more direct way to assess the effect ofgravity on locomotion but studying locomotion in actualhypogravity is demanding and expensive [6] The drawbacksto spaceflight experiments include difficulty in using neces-sary data collection hardware and performing an experimentwith adequate sample size Parabolic flight offers a viablealternative but periods of weightlessness are limited to sim20 swhich only allows for acute locomotion investigations [11]

There are several apparatuses that have been used inthe past to simulate reduced gravity locomotion One ofthe more used systems is the vertical body weight support(BWS) (Figures 1(a) and 1(b)) These kinds of simulators areusually obtained supporting the subjects in a harness thatapplies a controlled upward force For example the WARD[12] mechanism consists of a mechanical gear driven by apneumatic cylinder (Figure 1(b)) It is held in a cart that slidesforward and backward over a track Low-friction sliding ofthe mechanism ensures that only vertical forces are appliedto the subject Vertical BWS systems may also make use ofa small increase in air pressure around the userrsquos lower bodyto create a lifting force approximately at the personrsquos centerof mass [13] Other vertical systems [8 14] use a series ofcompliant rubber spring elements that are stretched to createthe upward (to simulate gravity less than 1 g) or downward(to simulate gravity greater than 1 g) force (Figure 1(a))The main limitation of these reduced gravity simulators (inaddition to high local skin pressure via a harness) is that eachsupporting limb experiences a simulated reduction of gravityproportional to the applied force while the swinging limbexperiences 1 g

The tilted BWS systems (Figures 1(c) and 1(d)) are con-structed to simulate more realistic effects of gravity changeson both the stance and swing legs in the sagittal planeThese simulators that have been used in the past by bothRoscosmos (Russian Federal Space Agency) and NASA totrain astronauts before space flights [15ndash17] are based onthe idea of neutralizing the component of the gravity forcenormal to the lying surface [mg sdot cos(120572) where 120572 is theangle of inclination] while the component of the gravityforce acting on the body and swinging limbs in the sagittalplane is reduced in relation to the tilt angle [mg sdot sin(120572)]A similar concept has been used in the reduced gravitysimulator (Figure 1(d)) designed by Ivanenko et al (Italianpatent number Rm2007A000489) the subject lies on the sideon a tilted couch (up to 40∘ from the horizontal position)with both legs suspended in the exoskeleton and steps onthe treadmill which is tilted to the same angle [7 18 19]This simulator included additional mass of the tilted chassis(sim15 kg) and exoskeleton (15 kg for each leg) Thus theentire assembly had a mass of sim18 kg that increased bothgravitational and inertial forces during walking

Another class of gravity-related manipulations is ldquosubjectload devicerdquo (SLD) that applies a gravity replacement forcein the direction down to the surface This type of SLD can beused in the vertical systems to increase the gravity [8] or in thelying position (Figure 1(e)) When an astronaut walks or runs

on a treadmill in weightlessness a subject load device is usedto return him or her back to the treadmill belt and to loadthe limbs The gravity replacement load is transferred via aharness to the pelvis andor the shoulders Gravity simulatorscan simulate active treadmill running in weightlessness andprovide a method of testing proposed improvements in SLDdesign and exercise protocols [20 21] In supine suspensionsystems (Figure 1(e)) subjects are suspended horizontallyattached to latex rubber cords A cloth sleeve and rubber cordare attached each to the upper and lower arms and legs (eighttotal) [20] The limitation of this system is a local pressureon some parts of the body (eg shoulders) andmodificationsin the swing phase dynamics due to nonconstant forcesof rubber cords and gravity acting in the anterioposteriordirection of leg movements (Figure 1(e))

Based on the passive gravity balancing technology Ma etal [22 23] proposed a design concept of a passive reducedgravity simulator to simulate human walking or other activ-ities in a reduced-gravity environment for potential applica-tions of training astronauts and space travelers (Figure 1(f))The system consists of a 3-DOF dual parallelogram mech-anism a 2-DOF torso support assembly and a pair of 3-DOF leg exoskeletons The weight of the body and the legsis compensated by the spring-balanced dual-parallelogrammechanism and torso-support assembly and the weight ofeach leg is compensated by a leg exoskeleton The systemis capable of simulating human walking and jumping in ahypogravity environment [24] Hardware prototyping andexperimental study of the new system are currently under-way

In the following section we discuss the basic principles ofadaptation of locomotion to different gravity values using thetechnologies described here

3 Biomechanical Aspects of Locomotion inReduced Gravity

Despite some differences all reduced gravity simulationapproaches show a reasonable approximation of the reduc-tion in the gravitational force acting on the center of bodymass (COM) and similar results concerning the speed ofgait transitions An important consequence of the pendulum-like behavior of the limbs in the gravity field is the principleof dynamic similarity [29] which states that geometricallysimilar bodies that rely on pendulum-like mechanics ofmovement have similar gait dynamics at the same Froudenumber

Fr sim 1198812

119892119871 (1)

where 119881 is the speed of locomotion 119892 is the accelerationof gravity and 119871 is a characteristic leg length That is alllengths times and forces scale by the same factors Inorder to optimize the recovery of mechanical energy thekinetic energy and the potential energy curves must beequal in amplitude and opposite in phase as in a pendulumAssuming that the change in kinetic energy within each step


PL

R

R

M

B 20 m

(a) Vertical system for altered grav-ity simulation

F

(b) Vertical BWS

(c) Tilted BWS

120572

(d) Tilted BWS

D

E

F

G

CB

A

(e) Supine suspension system

q2

q1

q3

q4

q6

q5

q11 q12

q78

q910

(f) Passive gravity balancing system

Figure 1 Reduced gravity simulators for locomotion (a) Schema of the vertical system used to simulate different gravity values (redrawnfrom [8]) R rubber bands B light metal bars M electric motor to stretch the elastic band system PL pulleys to invert the direction ofthe pull on the subject (dashed lines) (b) Vertical body weight support (BWS) system subject walks on a treadmill with different levels ofBWS while being supported in a harness pulled upwards by a preset unloading force 119865 (c) Tilted BWS system used by Roscosmos (RussianFederal Space Agency) to train astronauts before space flights [15] the subject walks on a truncated cone (60m height 92∘ inclination relativeto the vertical) supported by five ropes sustaining the head trunk and legs (picture portraying Professor Gurfinkel reproduced with his kindpermission) (d) Tilted unloading system for stepping on a treadmill the subject lies on the side on a tilted couch (up to 40∘ from the horizontalposition) with both legs suspended in the exoskeleton and steps on the treadmill which is tilted to the same angle The component of thegravity force acting on the stance and swing limb segments is proportional to the tilting angle 120572 [18] (e) Supine suspension system (adaptedfrom [20] courtesy of Professor Peter Cavanagh) the subject is suspended horizontally attached to latex rubber cords A cloth sleeve andrubber cord are attached each to the upper and lower arms and legs (eight total) The subject is actively pulled toward the treadmill by agravity replacement load through cables attached to a load splitter (f) Passive reduced gravity walking simulator (courtesy of Dr Ou Ma)The system consists of a 3-DOF dual parallelogram mechanism a 2-DOF torso support assembly and a pair of 3-DOF leg exoskeletons Theweight of the body and the legs is compensated by the spring-balanced dual-parallelogram mechanism and torso-support assembly and theweight of each leg is compensated by a leg exoskeleton [22ndash24]

is an increasing function of the walking speed (while thechange in the potential energy is proportional to gravity)the hypothesis was proposed that the inverted pendulum-like mechanism of energy exchange during walking wouldbe optimized at slower speeds in reduced gravity [3 10] An

optimal exchange between potential and kinetic energies ofthe COM occurs at Fr sim 025 [2] (Figure 2(a)) Even thoughspecific limb segment proportions may play an essential rolein the kinematics and energetics of walking [30] animalanatomy and individualized limb segment dimensions are


Margaria and Cavagna [3] Cavagna et al [68]Cavagna et al [10]Griffin et al [67]

Kram et al [4]Ivanenko et al [18]

0 05 10

1

2

15Gravity (g)

Wal

king

spee

d (m

s)

Walk-runtransitionFr = 05

Optimal speed Fr = 025

(a)

0

20

40

60

F

BW

andashpmndashl

0

4

2

0

20

40

60

1 g065 g 05 g

025 g 005 g

005 g

05 g

1 g

(b)

Simulated locomotion on Moon (016 g)

Gravity replacement load mg

Tilted BWS Vertical BWS

Horiz foot excursion Max horiz foot vel

Tilted BWS systemVertical BWS system

Gravity replacement load Vertical BWS system(matched foot excursion)

0

20

40

60

80

(cm

)

0

2

4(km

h)

100 10

6

8

x

1 g 1 g

Fp = 016middotmgFg

120572

Fvert = 084middotmg

(c)Figure 2 Biomechanical features of locomotion in reduced gravity conditions (a) Optimal (blue) and walk-to-run transition (green) speedsas a function of gravity Dynamically similar speeds predicted by Fr = 025 and Fr = 05 are indicated by blue and green dashed curvesrespectively [25] Green circles and stars refer to measurements of optimal walk-to-run transition speeds in simulated low-gravity conditions[5 18]The grey triangle indicates an earlier estimate of optimal walking speed predicted for theMoon gravitational environment byMargariaand Cavagna [3] Blue triangles refer to the optimal speeds (at which most of the mechanical exchange between potential and kinetic energyof the body center of mass occurs) obtained in a simulation study of Griffin et al [26] Blue circles represent measurements of optimal speedobtained during parabolic flight [10 27] (b) Time course of the net vertical component of in-shoe reaction forces plotted as a function ofthe spatial coordinates of the foot at different reduced gravity levels Note change in vertical scale in the 005 g condition The lower rightpanel shows the trajectories of the center of pressure superimposed on a foot outline (adapted from [28]) (c) Maximum longitudinal footvelocity and foot excursion (119909) during walking at 2 kmh at 016 g using three different reduced gravity simulators (represented schematicallyin the upper panels) Horizontal dashed lines indicate values for walking at 1 g The hatched bar (right panel) corresponds to the maximalfoot velocity for the vertical BWS system approximated by matching the foot excursion to that of the tilted BWS system [7] Note significantlylower foot velocities during swing using tilted BWS systems


optimized in such a way that the Froude number can explainoptimal walking velocity

On Earth walking and running gaits are usually adoptedfor different speeds of locomotion with a preferred transitionoccurring at sim2ms for human adults and at slow speeds forchildren (Frsim 05) in accordance with the dynamic similaritytheory [29] Different studies [4 18] demonstrated that atlower levels of gravity the walk-run transition occurred atprogressively slower absolute speeds but at approximately thesame Froude number (Figure 2(a))

Despite similarities in approximating reduced gravitythere are nevertheless essential differences between differentsimulation approaches The variables that showed the great-est differences between vertical and tilted reduced gravitysystems (Figure 1) were maximal longitudinal foot velocityand longitudinal foot excursion (Figure 2(c)) in agreementwith significant influences of gravity on swing leg dynamics[7] Even though the maximal longitudinal foot velocity forthe tilted BWS condition decreased only slightly relative tothe vertical BWS however the actual decrement was muchmore obvious if one takes into account that it was significantlycompensated for ormasked by increments in the stride length[7] A previous modeling study also predicted differentialeffects of gravity during stance and swing phases [31] In factthe changes in the longitudinal foot excursion were basicallyopposite for the vertical and tilted BWS systems (Figure 2(c))For the former system the amplitude of longitudinal footmotion decreased while for the latter system it increasedrelative to the 1 g condition Considering a monotonic (pre-sumably proportional [32]) relationship between the stridelength and the maximal foot velocity at a given gravity level(1 g) the peak foot velocity would be expected to be sim15times higher for the vertical than for tilted BWS conditionif the stride lengths were similar (Figure 2(c)) The previousstudies on parabolic flights investigating the effect of gravityon walkingmechanics demonstrated increments in the swingphase duration (by 29 at 025 g [33] see also [11]) in linewith the substantial contribution of gravity to the swing legOverall the findings demonstrate that gravity acting on bothstance and swing legs plays an important role in shapinglocomotor patterns

4 Nonlinear Reorganization of EMG Patterns

It is known that load plays a crucial role in shaping patternedmotor output during stepping [34ndash36] and humans producea specific heel-to-toe rolling pattern during stance in normalgravity conditions Ground contact forces reflect the netvertical and shear forces acting on the contact surface andresult from the sum of the mass-acceleration products ofall body segments while the foot is in contact with ground[37] Simulating reduced gravity between 005 and 1 g revealsdrastic changes of kinetic parameters but limited changesof the kinematic coordination [28] The reported accuratecontrol of limbfoot kinematics [28]maydependon load- anddisplacement-compensationmechanisms working effectivelythroughout a wide range of ground contact forces from fullbody weight up to lt5 of its value The peak vertical contactforces decrease proportionally to gravity but at 005 g they are

applied at the forefoot only (Figure 2(b)) During lower limbloading a variety of receptors can be activated such as Golgitendon organs cutaneous receptors of the foot and spindlesfrom stretched muscles [36] These sensory signals interactwith central rhythm-generating centers and help in shapingthe motor patterns controlling phase-transitions and rein-forcing ongoing activity [38 39] For instance loading ofthe limb enhances the activity in antigravity muscles duringstance and delays the onset of the next flexion [40] It isimportant to understand the mechanisms of sensorimotoradaptation to the biomechanics of locomotion and footplacementloading in heterogravity especially to longer-termchanges of load

A key feature of adaptation to hypogravity is a remarkablenonlinear scaling of muscle activity patterns contrary tomonotonic changes in foot loading The simplest kind ofchange with simulated reduced gravity [28] was seen inankle extensors the mean amplitude of activity decreasedsystematically with decreasing simulated gravity consistentwith their antigravity function [35 41] By contrast thebehavior of other muscles could not be predicted simply onthe basis of the static load during stance The amplitude andpattern of muscle activity generally depended on speed andcould vary nonmonotonically with body unloading Therewas also a complex reorganization of the pattern of activityof thigh muscles with decreasing simulated gravity as well asnoteworthy individual differences [28] Figure 3(a) illustratesan example of nonlinear reorganization of EMG patterns inone subject walking at 3 kmh With body weight unloadinggluteus maximus and distal leg extensors decreased theiractivity while other muscles demonstrated a ldquoparadoxicalrdquoincrement of activation (eg quadriceps) or considerablechanges in the activation waveforms (hamstring muscles)Note also the absence of the typical burst of RF at thebeginning of the swing phase at low simulated gravity levels(Figure 3(a)) consistent with other studies on the effect ofbody weight unloading [42] and walking speed [43] It isunlikely that these changes are due to the order of trials orthe consequence of learning the hypogravity condition sincepresentation order of speeds andBWSwas randomized acrosssessions and experiments [28] Also the duration of each trialwas sim1min with at least 2 min rest between trials and ashort (sim30 s) training period of walking at different speedswas allowed for each simulated reduced gravity level beforethe actual data collection was begun (the walking patternstypically adapt rapidly to simulated reduced gravity [4 5])This reorganization is presumably related to the multifunc-tional (biarticular) action of these muscles and to the needto repartition the joint torque contributions across differentmuscles as a function of the changes induced by gravity At1 g the main peak of m biceps femoris activity occurringbefore heel-contact serves to decelerate the swinging limb[37] However as gravity is decreased its main activity occursin mid-stance and late stance presumably in relation to theneed to assist vaulting over an inverted pendulum of thestance limb and swing initiation

There might be various factors accounting for the non-linear reorganization of muscle activity patterns with gravityTo start with nonlinear scaling also occurs during walking


005 g

GM VL RF

0 20 40 60 80 100Cycle () Cycle () Cycle ()

0 20 40 60 80 100 0 20 40 60 80 100

025 g

20 120583

V30

120583V

50 120583

V

50 120583

V

30 120583

V

30 120583

V

05 g065 g

1 g

005 g 025 g

05 g065 g

1 g

BF TA LG

(a)

Speed5 kmh 3 kmh

2 kmh 11 kmh

012345

GM

(120583V

)

0 02 04 06 08 111235

0 5

10 15 20

BF

g

Speed (kmh)

(120583V

)

010203040

TA

(120583V

)

05

101520

LG(120583

V)

02468

10

VL

(120583V

)

02468

10

RF

(120583V

)

(b)

Figure 3 Nonlinear reorganization of muscle activity patterns (a) An example of ensemble-averaged electromyographic (EMG) activity oflower limb muscles versus the normalized gait cycle is shown for a subject walking at 3 kmh at different simulated reduced gravity levels[28] (b) Mean EMG activity computed over the gait cycle and averaged across all cycles and subjects (119899 = 8) For each muscle values fortrials performed at each speed are plotted as a function of simulated reduced gravity (adapted from [28]) GM gluteus maximus VL vastuslateralis RF rectus femoris BF biceps femoris TA tibialis anterior LG and lateral gastrocnemius

at different speeds at 1 g For instance VL and RF activity isquite small at low speeds (less than sim3 kmh) but becomesprominent at higher speeds (gt4 kmh) (Figure 3(b)) a speedeffect consistent with that reported in the literature [28 4345 46] Given that it should be stressed that walking at lowergravity levels at the same speed (Figure 3(a)) corresponds towalking at higher speeds if one uses the Froude number as adimensionless parameter (eg walk-run transition at 025 g

occurs at sim4 kmh Figure 2(a)) so that ldquoparadoxicalrdquo incre-ments of VL and RF EMG activity in Figure 3(a) may reflecthigher biomechanical demands on proximal leg muscles athigher dimensionless speeds Nonlinear reorganization ofEMG patterns was also observed when using exoskeletonrobotic devices that provide body weight support [42 47]Changes in the body reference configuration during stance(slightly flexed posture [48 49]) may contribute to a greater


SOL

Walking speed

Time (s)

0

1 g

0 50 100 150 200 250 300

W-R R-W

0 50 100 150

Time (s)

350

0

048

(km

h)

0

100Cycle

()

Cycle ()0 1001 s

Right

Left100 120583V

600 120583VCycle ()

0 1001 s RightLeft

25 120583V

016 g

Δ

(a)

Mars

Spee

d (k

mh

)

0

2

4

6

8

0 02 04 06 08 1 12

10

Fr ~ 05

PlutoMoon

Walking

Abrupt switch

Smooth transitions

Running

Earth

Simulated gravity gEarth

Fr ~ 05

(b)

Figure 4 Smoothnessabruptness of gait transitions at different gravity levels (a) Soleus (SOL) EMG patterns during slow changes intreadmill belt speed (lower panels) in one representative subject at 016 g (left) and 1 g (right)Upper panels examples of SOL EMGwaveforms(left plotted versus time right plotted versus normalized cycle) during 5 consecutive strides of both legs around the transition from walking(black lines) to running (gray lines) Dotted curves denote the (transition) stride of the leg in which the swing phase first exceeded 50gait cycle Bottom horizontal bars denote stance (black) and swing (white) phases Lower panels the color maps represent a sequence ofdiscrete activation waveforms (vertical slices) 119909-axis indicates the number of the gait cycles (corresponding to the appropriate timing of thetrial) 119910-axis indicates normalized gait cycle (from touchdown to another touchdown) and color indicates EMG amplitude The white lineindicates when toe off occurred Vertical dashed lines indicate walk-to-run (W-R) and run-to-walk (R-W) transitions Note abrupt changesin the relative stance duration and muscle activation patterns at gait transitions at 1 g and no obvious distinction in these parameters at thetransition from walking to running at 016 g (b) Schematic representation of the smoothness of gait transitions as a function of gravity Theorange curve symbolizes the dimensionless walk-run transition speed consistent with the theory of dynamic similarity (Fr sim 05) [19 29 44]The blue color range of gravitational levels represents a discontinuous switch from walk to run whereas the white region indicates smoothtransitions

activity of proximal extensors as well Finally there is adifferential effect of speed on quadriceps muscle activity atreduced gravity levels VL and RF activity increases at lowspeeds (lt3 kmh) while it decreases at a high speed (5 kmh)(Figure 3(b)) Potential nonlinear scaling of muscle activityfor most whole body movements in microgravity shouldalso be taken into account for exercise countermeasures forastronauts

5 Different Gaits

Considering complex high-dimensional dynamically cou-pled interactions between an organism and gravitationalenvironment in principle one challenging solution is toadopt different coordination patterns and not only an optimalspeed of locomotion Are different gaits possible on otherplanets

One approach to study locomotor adaptations is to lookat the effect of gravity on gait transitions A gait has beendefined as ldquoa pattern of locomotion characteristic of a limitedrange of speeds described by quantities of which one or morechange discontinuously at transitions to other gaitsrdquo [29]An important aspect of gait transitions is a discontinuousswitch that occurs at some point while varying the speedof progression (although some exceptions may exist [50ndash52]) As already discussed (Figure 2(a)) gravity has a strong

effect on the speed at which gait transitions occur (Fr sim05) Surprisingly however we found [18 19] that at lowerlevels of simulated gravity the transition betweenwalking andrunningwas generally gradual without any noticeable abruptchange in gait parameters or EMG bursts (Figure 4(a)) Thiswas associated with a significant prolongation of the swingphase whose duration became virtually equal to that of stancein the vicinity of the walk-run transition speed and witha gradual shift from inverted-pendulum gait (walking) tobouncing gait (running) A lack of discontinuous changes inthe pattern of speed-dependent locomotor characteristics in ahypogravity environment (Figure 4(b)) is consistent with theidea of a continuous shift in the state of a given set of centralpattern generators rather than the activation of a separate setof central pattern generators for each distinct gait [19]

Interestingly the smoothness of gait transitions is accom-panied by a gradual shift from inverted-pendulum gaitto bouncing gait resulting in a ldquoparadoxicalrdquo inverted-pendulum running in the vicinity of run-walk and walk-runtransitions [18]The swing phasemay havemore influence ongait than it was previously thought For instance relativelyslower swing and longer foot excursions (tilted BWS con-dition Figure 2(c)) may raise questions about optimality orcomfort ofwalking and could account for potentially differentpreferred gaits such as loping on the Moon observed inApollo astronauts (though the Lunar suit limits the range


(a)

25

20

15

10

5

0

10 13 16 19 22 25

10

5

0

Predicted available impulse (N s)

Pred

icte

d av

aila

ble

impu

lse (N

s)Successful subjects

Succ

essfu

l sub

ject

s

MoonEnceladusEuropa Io

Simulated gravity gEarth ()

(b)

Figure 5 Running on water at simulated reduced gravity The blue curve represents the net vertical impulse available to run on water aspredicted by the model used by Minetti et al [53] Bars represent the number of subjects out of 6 capable of avoiding sinking at differentsimulated gravity values Both variables show that 22 of Earth gravity (119892EARTH) is the maximum gravity at which humans can run on waterwhen assisted by a small rigid fin (as illustrated in the left panel)

of motion in the leg joints and may also contribute to theloping gait on the Moon [9]) The resulting changes in theintersegmental and interlimb coordinationmay in turn affectthe COM motion Overall the results support the idea oflooking for new forms of locomotion (both bipedal andquadrupedal) in a heterogravity environment [54] based onthe interplay between stance and swing leg dynamics alteredinterlimb coupling and altered center of mass movements

Other significant influences of gravity on short-term andlong-term gait adaptations may be related to its effects onthe body reference configuration [48 49] and anticipatorymechanisms of limb and body movements [55 56] Forinstance the basis of habitual human posture is postural toneof the skeletal muscles and microgravity elicits substantialchanges in muscle tone and posture [48 49] Based onclinical observations it has been recently argued that anyreflection on the nature and choice of preferred gait (egbipedal versus quadrupedal) should include a considerationof the mechanisms determining the choice of unconscioushabitual posture [57] Also in analogy with the resultsbased on upper-limb movements related to time-to-contact[55] or movement planning [58] anticipatory postural andlocomotor adjustments for lower limb movements (eg forthe control of heel strike or accurate foot placement) shouldtake gravity into consideration Therefore altered gravityconditions may also affect locomotor-related tasks such asthe negotiation of stationary andmoving obstructions duringwalking or gait initiationtermination [56 59 60]

Finally the repertoire of known gaits can be expandedto a variety of animals For instance on Earth only afew legged species such as water strider insects and someaquatic birds and lizards can run on water For most other

species including humans this is precluded by body sizeand proportions lack of appropriate appendages and limitedmuscle power However if gravity is reduced to less thanEarthrsquos gravity running on water should require less musclepower Recently Minetti et al [53] used this hydrodynamicmodel of Glasheen and McMahon [61] to predict the gravitylevels at which humans should be able to run on water andtested the hypothesis in the laboratory using a reduced gravitysimulator (Figure 5)The results showed that a hydrodynamicmodel of Basilisk lizards running on water [61] can also beapplied to humans despite the enormous difference in bodysize andmorphology Particularly 22of Earthrsquos gravity is themaximum at which humans can run on water when assistedby a small rigid fin (Figure 5) [53] It is also worth notingthe limitations for our musculoskeletal system for producingforcepower (endurance) for instance the stride frequencyin humans is limited to about 2Hz whatever the planet isOn Earth the biggest animal that can run on water is likelyWestern Grebes and even these birds can run only for severalseconds since the force production is basically anaerobic(participants in [53] could run at simulated ldquoMoonrdquo gravityonly for sim10 s) In contrast at reduced gravity (Moon) thesebirds could run on water in a charming manner for muchlonger time

6 Clinical Implications

Reduced gravity also offers unique opportunities for adjust-ing the basic patterns to altered locomotor conditions forgait rehabilitation Bodyweight support systems coupledwithrobotic devices or pharmacologic treatments are now oftenused in the rehabilitation practice to assist physical therapy of


ES of peroneal nerve

RF

BF

TA

LG

Hip

Knee

Ankle

Delay

Flex

Ext

5 s

300 120583V

30∘

30∘

15∘

(a)

Hand-walking

Hand

Foot

DELTa

FCU

BIC

ST

BF

5 s

50 cm

30 cm

100 120583V

800 120583V

200 120583V

(b)

Figure 6 Eliciting nonvoluntary limb stepping movements in simulated weightlessness (gravity neutral) conditions (a) An example ofnonvoluntary rhythmic movements of the suspended legs induced by electrical stimulation (ES) of peroneal nerve from the study of Selionovet al [62] Note the absence of ankle joint rotations during evoked air-stepping (b) An example of evoked rhythmic leg movements duringhand walking in one subject from the study of Sylos-Labini et al [63] RF rectus femoris BF biceps femoris TA tibialis anterior LG lateralgastrocnemius FCU flexor carpi ulnaris BIC biceps brachii DELTa anterior deltoid ST and semitendinosusHand and foot denote anterior-posterior displacements of the left hand and foot

individuals with neurological disorders We will not reviewany detailed analysis of clinical outcomes for ambulationwhen using locomotor training with body weight supportsystems and refer to other reviews [64] Nevertheless it isworth emphasizing a facilitatory effect of the lack of gravityon rhythmogenesis and its potential for gait recovery

Novel pharmacological strategies [65] and electromag-netic stimulation techniques [62 66ndash68] are being developedaimed at modulating spinal activity and restoring the loco-motor function The spinal central pattern generator (CPG)circuitry can be easily activated in healthy humans in a gravityneutral position by applying tonic central and peripheralsensory inputs To minimize interference with the ongoingtask of bodyweight and balance control steppingmovementsare elicited during air-stepping in the absence of gravityinfluences and external resistance Figure 6 illustrates exam-ples of nonvoluntary rhythmic movements of the suspendedlegs induced by electrical stimulation of peroneal nerve [62]and during hand walking [63] It has been suggested thatfunctional multisensory stimulations and a functional neuralcoupling between arm and legs can inspect CPG access bysensory and central activations and entrain locomotor neuralnetworks and promote gait recovery Such investigations may

contribute to the clinical development of central patterngenerator-modulating therapies and neuroprosthetic tech-nologies [65 69]


This perspective outlines an interdisciplinary approach toextend our knowledge on adaptation of human locomotionto a hypogravity environment including biomechanical neu-rophysiological and comparative aspects effective exercisecountermeasures for astronauts and even exobiology ofnew forms of locomotion on different planets The toolsand techniques used for hypogravity simulation and theireffects on human locomotion provide new insights into ourunderstanding of the physiological effects of gravity Thebeneficial effect of weightlessness on rhythmogenesis wouldfurther enhance the utility of this approach and developmentsof innovative technologies for gait rehabilitation




Acknowledgments

This work was supported by the Italian Health MinistryItalian Ministry of University and Research (PRIN project)and Italian Space Agency (DCMC CRUSOE and COREAgrants)

References

[1] R J Full and D E Koditschek ldquoTemplates and anchorsneuromechanical hypotheses of legged locomotion on landrdquoThe Journal of Experimental Biology vol 202 no 23 pp 3325ndash3332 1999

[2] A E Minetti ldquoInvariant aspects of human locomotion indifferent gravitational environmentsrdquo Acta Astronautica vol49 no 3ndash10 pp 191ndash198 2001

[3] R Margaria and G A Cavagna ldquoHuman locomotion insubgravityrdquo Aerospace Medicine vol 35 pp 1140ndash1146 1964

[4] R Kram A Domingo and D P Ferris ldquoEffect of reducedgravity on the preferred walk-run transition speedrdquoThe Journalof Experimental Biology vol 200 no 4 pp 821ndash826 1997

[5] J M Donelan and R Kram ldquoThe effect of reduced gravity onthe kinematics of human walking a test of the dynamic simi-larity hypothesis for locomotionrdquo The Journal of ExperimentalBiology vol 200 no 24 pp 3193ndash3201 1997

[6] B L Davis and P R Cavanagh ldquoSimulating reduced gravity areview of biomechanical issues pertaining to human locomo-tionrdquo Aviation Space and Environmental Medicine vol 64 no6 pp 557ndash566 1993

[7] F Sylos-Labini Y P Ivanenko G Cappellini A Portone MJ Maclellan and F Lacquaniti ldquoChanges of gait kinematicsin different simulators of reduced gravityrdquo Journal of MotorBehavior vol 45 no 6 pp 495ndash505 2013

[8] G A Cavagna A Zamboni T Faraggiana and R MargarialdquoJumping on the moon power output at different gravityvaluesrdquo Aerospace Medicine vol 43 no 4 pp 408ndash414 1972

[9] C E Carr and J McGee ldquoThe apollo number space suits self-support and the walk-run transitionrdquo PLoS ONE vol 4 no 8Article ID e6614 2009

[10] G A Cavagna P A Willems and N C Heglund ldquoThe role ofgravity in human walking pendular energy exchange externalwork and optimal speedrdquo Journal of Physiology vol 528 part 3pp 657ndash668 2000

[11] J K De Witt G P Perusek B E Lewandowski et al ldquoLoco-motion in simulated and real microgravity horizontal Suspen-sion vs parabolic flightrdquo Aviation Space and EnvironmentalMedicine vol 81 no 12 pp 1092ndash1099 2010

[12] F Gazzani A Fadda M Torre and V Macellari ldquoWARD apneumatic system for body weight relief in gait rehabilitationrdquoIEEE Transactions on Rehabilitation Engineering vol 8 no 4pp 506ndash513 2000

[13] A M Grabowski and R Kram ldquoEffects of velocity and weightsupport on ground reaction forces and metabolic power duringrunningrdquo Journal of Applied Biomechanics vol 24 no 3 pp288ndash297 2008

[14] J P He R Kram and T A McMahon ldquoMechanics of runningunder simulated low gravityrdquo Journal of Applied Physiology vol71 no 3 pp 863ndash870 1991

[15] V A Bogdanov V S Gurfinkel and V E Panfilov ldquoHumanmotion under lunar gravity conditions (Human performancein various locomotive tasks under simulated lunar reduced

gravity conditions classifying test stands and equipment)rdquoKosmicheskaya Biologiya i Meditsina vol 5 pp 3ndash13 1971

[16] J R Hansen Spaceflight Revolution NASA Langley ResearchCenter from Sputnik to Apollo National Aeronautics and SpaceAdministration Washington DC USA 1995

[17] D E Hewes ldquoReduced-gravity simulators for studies of manrsquosmobility in space and on the moonrdquoHuman Factors vol 11 no5 pp 419ndash431 1969

[18] Y P Ivanenko F Sylos-Labini G Cappellini V MacellariJ McIntyre and F Lacquaniti ldquoGait transitions in simulatedreduced gravityrdquo Journal of Applied Physiology vol 110 no 3pp 781ndash788 2011

[19] F Sylos-Labini Y P Ivanenko G Cappellini S Gravanoand F Lacquaniti ldquoSmooth changes in the EMG patternsduring gait transitions under body weight unloadingrdquo Journalof Neurophysiology vol 106 no 3 pp 1525ndash1536 2011

[20] K O Genc V E Mandes and P R Cavanagh ldquoGravity replace-ment during running in simulatedmicrogravityrdquoAviation Spaceand Environmental Medicine vol 77 no 11 pp 1117ndash1124 2006

[21] J L McCrory H A Baron S Balkin and P R CavanaghldquoLocomotion in simulated microgravity gravity replacementloadsrdquo Aviation Space and Environmental Medicine vol 73 no7 pp 625ndash631 2002

[22] Q Lu C Ortega and O Ma ldquoPassive gravity compensationmechanisms technologies and applicationsrdquo Recent Patents onEngineering vol 5 no 1 pp 32ndash44 2011

[23] OMa and JWang ldquoApparatus andmethod for reduced-gravitysimulationrdquo 2012

[24] Q Lu J McAvoy and O Ma ldquoA simulation study of a reduced-gravity simulator for simulating human jumping and walkingin a reduced-gravity environmentrdquo in Proceedings of the ASMEDynamic Systems and Control Conference 2009

[25] A E Minetti ldquoWalking on other planetsrdquo Nature vol 409 no6819 pp 467ndash469 2001

[26] T M Griffin N A Tolani and R Kram ldquoWalking in simulatedreduced gravity mechanical energy fluctuations and exchangerdquoJournal of Applied Physiology vol 86 no 1 pp 383ndash390 1999

[27] G A Cavagna P A Willams and N C Heglund ldquoWalking onmarsrdquo Nature vol 393 no 6686 article 636 1998

[28] Y P Ivanenko R Grasso V Macellari and F LacquanitildquoControl of foot trajectory in human locomotion role ofground contact forces in simulated reduced gravityrdquo Journal ofNeurophysiology vol 87 no 6 pp 3070ndash3089 2002

[29] R Alexander McN ldquoOptimization and gaits in the locomotionof vertebratesrdquo Physiological Reviews vol 69 no 4 pp 1199ndash1227 1989

[30] F Leurs Y P Ivanenko A Bengoetxea et al ldquoOptimal walkingspeed following changes in limb geometryrdquo The Journal ofExperimental Biology vol 214 part 13 pp 2276ndash2282 2011

[31] D A Raichlen ldquoThe effects of gravity on human walking anew test of the dynamic similarity hypothesis using a predictivemodelrdquoThe Journal of Experimental Biology vol 211 no 17 pp2767ndash2772 2008

[32] Y Osaki M Kunin B Cohen and T Raphan ldquoThree-dimensional kinematics and dynamics of the foot duringwalking a model of central control mechanismsrdquo ExperimentalBrain Research vol 176 no 3 pp 476ndash496 2007

[33] J F Roberts ldquoWalking responses under lunar and low gravityconditionsrdquo AMRL-TR 6570th Aerospace Medical ResearchLaboratory 1963


[34] S H Scott and D A Winter ldquoBiomechanical model of thehuman foot kinematics and kinetics during the stance phase ofwalkingrdquo Journal of Biomechanics vol 26 no 9 pp 1091ndash11041993

[35] S J Harkema S L Hurley U K Patel P S Requejo B HDobkin and V R Edgerton ldquoHuman lumbosacral spinal cordinterprets loading during steppingrdquo Journal of Neurophysiologyvol 77 no 2 pp 797ndash811 1997

[36] J Duysens F Clarac and H Cruse ldquoLoad-regulating mecha-nisms in gait and posture comparative aspectsrdquo PhysiologicalReviews vol 80 no 1 pp 83ndash133 2000

[37] D A Winter The Biomechanics and Motor Control of HumanGait Normal Elderly and Pathological University of WaterlooPress Waterloo Canada 1991

[38] K G Pearson ldquoCommon principles of motor control invertebrates and invertebratesrdquo Annual Review of Neurosciencevol 16 pp 265ndash297 1993

[39] K G Pearson ldquoProprioceptive regulation of locomotionrdquoCurrent Opinion inNeurobiology vol 5 no 6 pp 786ndash791 1995

[40] J Duysens B M H van Wezel H W A A van de CrommertM Faist and J G M Kooloos ldquoThe role of afferent feedback inthe control of hamstrings activity during human gaitrdquo EuropeanJournal of Morphology vol 36 no 4-5 pp 293ndash299 1998

[41] L Finch H Barbeau and B Arsenault ldquoInfluence of bodyweight support on normal human gait development of a gaitretraining strategyrdquoPhysicalTherapy vol 71 no 11 pp 842ndash8551991

[42] J C Moreno F Barroso D Farina et al ldquoEffects of roboticguidance on the coordination of locomotionrdquo Journal of Neu-roEngineering and Rehabilitation vol 10 no 1 article 79 2013

[43] A R den Otter A C H Geurts T Mulder and J DuysensldquoSpeed related changes in muscle activity from normal to veryslow walking speedsrdquo Gait and Posture vol 19 no 3 pp 270ndash278 2004

[44] F Saibene and A E Minetti ldquoBiomechanical and physiologicalaspects of legged locomotion in humansrdquo European Journal ofApplied Physiology vol 88 no 4-5 pp 297ndash316 2003

[45] Y P Ivanenko R E Poppele and F Lacquaniti ldquoSpinalcord maps of spatiotemporal alpha-motoneuron activation inhumanswalking at different speedsrdquo Journal ofNeurophysiologyvol 95 no 2 pp 602ndash618 2006

[46] A Pepin K E Norman and H Barbeau ldquoTreadmill walkingin incomplete spinal-cord-injured subjects 1 Adaptation tochanges in speedrdquo Spinal Cord vol 41 no 5 pp 257ndash270 2003

[47] F Sylos-Labini V La Scaleia I Pisotta et al ldquoEMG patternsduring assisted walking in the exoskeletonrdquo Frontiers in HumanNeuroscience vol 8 article 423 2014

[48] J Massion K Popov J-C Fabre P Rage and V Gurfinkel ldquoIsthe erect posture in microgravity based on the control of trunkorientation or center of mass positionrdquo Experimental BrainResearch vol 114 no 2 pp 384ndash389 1997

[49] G Andreoni C Rigotti G Baroni G Ferrigno N A Colfordand A Pedotti ldquoQuantitative analysis of neutral body posturein prolonged microgravityrdquo Gait amp Posture vol 12 no 3 pp235ndash242 2000

[50] S M Gatesy and A A Biewener ldquoBipedal locomotion effectsof speed size and limb posture in birds and humansrdquo Journal ofZoology vol 224 no 1 pp 127ndash147 1991

[51] J Rubenson D B Heliams D G Lloyd and P A FournierldquoGait selection in the ostrich mechanical and metabolic char-acteristics of walking and running with and without an aerial

phaserdquo Proceedings of the Royal Society B Biological Sciencesvol 271 no 1543 pp 1091ndash1099 2004

[52] L Ren and J R Hutchinson ldquoThe three-dimensional locomotordynamics of African (Loxodonta africana) and Asian (Elephasmaximus) elephants reveal a smooth gait transition at moderatespeedrdquo Journal of the Royal Society Interface vol 5 no 19 pp195ndash211 2008

[53] A E Minetti Y P Ivanenko G Cappellini N Dominici andF Lacquaniti ldquoHumans running in place on water at simulatedreduced gravityrdquoPLoSONE vol 7 no 7 Article ID e37300 2012

[54] M Srinivasan and A Ruina ldquoComputer optimization of aminimal biped model discovers walking and runningrdquo Naturevol 439 no 7072 pp 72ndash75 2006

[55] J McIntyre M Zago A Berthoz and F Lacquaniti ldquoDoes thebrain model Newtonrsquos lawsrdquo Nature Neuroscience vol 4 no 7pp 693ndash694 2001

[56] G Clement V S Gurfinkel F Lestienne M I Lipshits andK E Popov ldquoAdaptation of postural control to weightlessnessrdquoExperimental Brain Research vol 57 no 1 pp 61ndash72 1984

[57] Y P IvanenkoW GWright R J St George andV S GurfinkelldquoTrunk orientation stability and quadrupedalismrdquo Frontiers inNeurology vol 4 article 20 2013

[58] C Papaxanthis T Pozzo K E Popov and J McIntyre ldquoHandtrajectories of vertical arm movements in one-G and zero-G environments Evidence for a central representation ofgravitational forcerdquo Experimental Brain Research vol 120 no4 pp 496ndash502 1998

[59] P Crenna D M Cuong and Y Breniere ldquoMotor programmesfor the termination of gait in humans organisation and velocity-dependent adaptationrdquo Journal of Physiology vol 537 no 3 pp1059ndash1072 2001

[60] B J McFadyen and H Carnahan ldquoAnticipatory locomotoradjustments for accommodating versus avoiding level changesin humansrdquo Experimental Brain Research vol 114 no 3 pp500ndash506 1997

[61] J W Glasheen and T A McMahon ldquoSize-dependence of water-running ability in basilisk lizards (Basiliscus basiliscus)rdquo TheJournal of Experimental Biology vol 199 no 12 pp 2611ndash26181996

[62] VA Selionov Y P Ivanenko I A Solopova andV S GurfinkelldquoTonic central and sensory stimuli facilitate involuntary air-stepping in humansrdquo Journal of Neurophysiology vol 101 no 6pp 2847ndash2858 2009

[63] F Sylos-Labini Y P Ivanenko M J Maclellan G Cappellini RE Poppele and F Lacquaniti ldquoLocomotor-like leg movementsevoked by rhythmic armmovements in humansrdquoPloSONE vol9 no 3 Article ID e90775 2014

[64] P Sale M Franceschini A Waldner and S Hesse ldquoUse ofthe robot assisted gait therapy in rehabilitation of patients withstroke and spinal cord injuryrdquo European Journal of Physical andRehabilitation Medicine vol 48 no 1 pp 111ndash121 2012

[65] P A Guertin ldquoPreclinical evidence supporting the clinicaldevelopment of central pattern generator-modulating therapiesfor chronic spinal cord-injured patientsrdquo 2014

[66] YGerasimenko PMusienko I Bogacheva et al ldquoPropriospinalbypass of the serotonergic system that can facilitate steppingrdquoJournal of Neuroscience vol 29 no 17 pp 5681ndash5689 2009

[67] V A Selionov I A Solopova D S Zhvansky et al ldquoLackof non-voluntary stepping responses in Parkinsonrsquos diseaserdquoNeuroscience vol 235 pp 96ndash108 2013


[68] C A Angeli V R Edgerton Y P Gerasimenko and S JHarkema ldquoAltering spinal cord excitability enables voluntarymovements after chronic complete paralysis in humansrdquo Brain2014

[69] D Borton M Bonizzato J Beauparlant et al ldquoCorticospinalneuroprostheses to restore locomotion after spinal cord injuryrdquoNeuroscience Research vol 78 pp 21ndash29 2014

Research ArticleConditioned Media from Microvascular Endothelial CellsCultured in Simulated Microgravity Inhibit Osteoblast Activity

Alessandra Cazzaniga Sara Castiglioni and Jeanette A M Maier

Dipartimento di Scienze Biomediche e Cliniche Luigi Sacco Universita di Milano Via GB Grassi 74 Milano Italy

Correspondence should be addressed to Jeanette A M Maier jeanettemaierunimiit

Received 23 April 2014 Revised 9 July 2014 Accepted 9 July 2014 Published 19 August 2014


Copyright copy 2014 Alessandra Cazzaniga et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Background and Aims Gravity contributes to the maintenance of bone integrity Accordingly weightlessness conditions duringspace flight accelerate bone loss and experimental models in real and simulated microgravity show decreased osteoblastic andincreased osteoclastic activities It is well known that the endotheliumandbone cells cross-talk and this intercellular communicationis vital to regulate bone homeostasis Because microgravity promotes microvascular endothelial dysfunction we anticipated thatthe molecular cross-talk between endothelial cells exposed to simulated microgravity and osteoblasts might be altered ResultsWe cultured human microvascular endothelial cells in simulated microgravity using the rotating wall vessel device developed byNASA Endothelial cells in microgravity show growth inhibition and release higher amounts of matrix metalloproteases type 2 andinterleukin-6 than controls Conditionedmedia collected frommicrovascular endothelial cells in simulatedmicrogravity were usedto culture human osteoblasts and were shown to retard osteoblast proliferation and inhibit their activityDiscussion Microvascularendothelial cells inmicrogravity are growth retarded and release high amounts of matrix metalloproteases type 2 and interleukin-6which might play a role in retarding the growth of osteoblasts and impairing their osteogenic activityConclusions We demonstratethat since simulated microgravity modulates microvascular endothelial cell function it indirectly impairs osteoblastic function

1 Introduction

Bone development and remodeling depend mainly uponcomplex interactions between osteoblasts and osteoclastsIndeed an intimate communication exists between osteo-blasts and osteoclasts since osteoclasts control osteoblasticgrowth and function while osteoblasts regulate the dif-ferentiation and the activity of osteoclasts [1] Recentlyother cells of the bone microenvironment are emerging asimplicated in bone health Among others endothelial cellsare players of the communication network in the bone [2]In embryonic skeletal tissue osteogenesis and angiogenesisare temporally related [3] and in the adults osteoblasts arealways located adjacent to endothelial cells in blood ves-sels at sites of new bone formation [4] The fact thatolder subjects with osteoporosis have decreased bloodvessels in their skeletal tissue accompanied by a paral-lel decrease in osteoblasts further highlights this closerelation [5] Several lines of evidence indicate that a

mutual communication system exists between the endothe-lium and the osteoblasts At the cellular and molec-ular levels vascular endothelial cells have been shown toregulate bone remodelling via cell signalling networks ofligand-receptor complexes and osteoblasts release growthfactors that influence endothelial cells [3]

In long duration space missions astronauts experienceconsiderable bone loss about 1-2 of bone mass per monthin the weight-bearing regions of the leg and the spinemainly because of an uncoupling between osteoblasts andosteoclasts [6ndash8] We anticipate that endothelial-osteoblastcommunication might be impaired in space and contributesto bone loss Indeed dysfunctions in human endothelial cellscultured in simulated microgravity have been described [9ndash15] and alterations in the capillaries of the epiphyses andmetaphyses of femoral bones of rats flown aboard the USlaboratory SLS-2 were detected [16]

Cross-talk between endothelial cells and osteoblasts insimulated microgravity has not been deciphered yet As



a first approach to investigate this issue we exposedosteoblasts to conditioned media (CM) from microvascularendothelial cells (HMEC) cultured in the rotating wall ves-sel (RWV) which simulates some aspects of microgravityStudies utilizing CM are considered a successful strategy forthe identification of soluble factors interconnecting differentcell types and candidate biomarkers for further validation inclinical samples [17] Indeed CM reveal the cell secretomethat is the collection of proteins that are released throughthe classical and nonclassical secretion pathways and alsoproteins shed from the cell surface These secreted proteinsinclude enzymes growth factors cytokines and other solublemediators and are important contributors to cell survivalgrowth and differentiation [17] We here show that CMfrom HMEC grown in simulated microgravity impair theproliferation and activity of cultured primary osteoblasts andosteoblast-like Saos-2 cells


21 Cell Culture HMEC were obtained from CDC (AtlantaUSA) and grown in MCDB131 containing epidermal growthfactor (10 ngmL) and 10 fetal bovine serum (FBS) on 2gelatin-coated dishes Normal human osteoblasts (NHOst)were maintained in osteoblast growth media (OGM) asindicated by the manufacturer (Lonza Basel Switzerland)at 37∘C in a humidified atmosphere containing 5 CO

2

[18] Saos-2 cells (American Type Culture Collection) werecultured inDMEMcontaining 10FBS Before beginning theexperiments with CM from HMEC NHOst and Saos-2 cellswere gradually adapted to be cultured in 1 1 HMEC growthmedium and OGM or DMEM respectively To simulatemicrogravity we utilized the RWV (Synthecom Inc HoustonTX USA) HMEC were seeded on beads (Cytodex 3 SigmaAldrich St Louis MO USA) as controls (CTR) HMECgrown on beads were cultured in the vessels not undergoingrotation [11] In the RWV the vessel rotates around a horizon-tal axis (28 rpm) and allows diffusion of oxygen and carbondioxide across a semipermeable membrane The vessel walland the medium containing cells bound to microcarrierbeads rotate at the same speed producing a vector-averagedgravity comparable with that of near-earth free-fall orbit [19]The beads do not form aggregates in the RWV and tend tobe evenly distributed throughout the vessel Such a rotationreduces gravity to approximately 3 times 10minus2 g [10] After 72 h inthe RWV or in the vessels without rotation the media fromHMEC were collected centrifuged filtered through 02 120583mfilter diluted 1 1 with fresh culture medium to replenishnutrients and used to culture NHOst and Saos-2 cells Inthese experiments the medium was changed every 48 h

22 DNA Fragmentation HMEC cell death was evalu-ated using the cell death detection ELISA (Roche) whichdetermines cytoplasmic histone-associated DNA fragmentsBriefly after 48 and 72 h in the RWV or under controlconditions the cells were lyzed and centrifuged and thesupernatant was analyzed according to the manufacturerrsquosinstruction As a positive control we used HMEC exposed

for 30min to H2O2(10 120583M) and cultured for additional 48 h

in their growth medium

23 Cell Proliferation For MTT assay NHOst and Saos-2at 50 confluence were cultured in 96-well plates for 24 hbefore being exposed for different times to the media col-lected from HMEC MTT measures the reduction of yellowtetrazolium salt MTT to dark purple formazan by succinatedehydrogenase mainly in mitochondria and it is now widelyaccepted as a reliable way to examine cell viability and prolif-eration [20] Briefly at the end of the experiment the mediawere replaced with medium containing 3-(45-Dimethyl-2-thiazolyl)-25-diphenyl-2H-tetrazolium bromide (MTT05mgmL) (Sigma Aldrich St Louis MO USA) Formazancrystals generated by the cellular reduction activity weredissolved in DMSO Absorbance was measured at 550 nm

Neutral red uptake assay was also used to estimateNHOstviability Briefly 24 h after seeding in 96-well dishes the cellswere exposed to CM from HMEC After 3 days neutralred was added to the medium to a final concentration of50 120583gmL 2 h later the wells were washedwith PBS and fixedAbsorbance was measured at 550 nm [21]

HMECand Saos-2 cells were trypsinized and stainedwithtrypan blue solution (04) and the viable cells were countedusing a Burker chamber

24 Osteoblast Activity NHOst and Saos-2 cells at 80confluence were cultured in 24-well plates with conditionedmedia from HMEC added with 100 nM dexamethasone50 120583M L-ascorbate-2-phosphate and 10mM glycerophos-phate at 37∘C in a 5CO

2for 7 and 14 days Osteoblast activ-

ity was evaluated quantifying alkaline phosphatase (ALP)enzymatic activity in the medium by a colorimetric assaybased on the hydrolysis of P-nitrophenyl phosphate Theabsorbance was measured at 405 nm [18] To analyze calciumdeposition the cellswere rinsedwith PBS fixed (70ethanol1 h) and stained for 10min with 2 Alizarin Red S (pH 42)Cultures were photographed with a digital camera AlizarinRed was then released from the cell matrix by incubationfor 15min in 10 cetylpyridinium chloride in 10mM sodiumphosphate (pH 70) The absorbance was measured at 562 nm[18]

25 Measurements of TIMP-2 and IL-6 by ELISA Condi-tioned media were centrifuged and filtered The amountsof tissue inhibitor of matrix metalloprotease (TIMP)-2 andinterleukin (IL)-6 were measured using a double-antibodysandwich ELISA (GE Healthcare) according to the manufac-turerrsquos instructions The concentrations of TIMP-2 and IL-6were determined by interpolation from a standard curve

26 Western Blot HMEC cells were lysed separated onSDS-PAGE and transferred to nitrocellulose sheets Westernanalysis was performed using antibodies against p21 p53 andGAPDH (Tebu Bio-Santa Cruz) Secondary antibodies werelabelled with horseradish peroxidase (Amersham PharmaciaBiotech) The SuperSignal chemiluminescence kit (Pierce)was used to detect immunoreactive proteins


0

1

2

3

4

5

6

7

8

24 48 72 96

HMEC-CHMEC-RWV

(h)

Tota

l cel

l num

ber (times105

)

(a)

C RWV

p21

GAPDH

p53

(b)

0100

200

300

400

500

600

700

48 72

HMEC-CHMEC-RWV

Abso

rban

ce 405

nm

(h)

lowastlowastlowast

HMEC + H2O2

(c)

Figure 1 Simulated microgravity inhibits HMEC growth (a) HMEC were cultured for different times in the RWV (HMEC-RWV) andtrypsinized and viable cells were counted HMEC-C control (b) Cell extracts (50120583glane) were loaded on a 15 SDS-PAGE blotted intonitrocellulose filter incubated with anti-p21 and anti-p53 antibodies and visualized by chemiluminescence as described After stripping theblot was incubated with an anti-GAPDH antibody to show that comparable amounts of proteins were loaded per lane (c) Apoptosis wasevaluated by ELISA on HMEC lysates after 48 and 72 h in the RWV or under control conditions Our positive control is represented byHMEC exposed to H

2O2for 30min and then cultured for additional 48 h

27 Statistical Analysis All experiments were repeated atleast three times in triplicate Data are presented as means plusmnstandard deviation Statistical differences were determinedusing the unpaired two-tailed Studentrsquos 119905 test Consider lowast119875 lt005 lowastlowast119875 lt 001

3 Results

31 Simulated Microgravity Alters HMEC BehaviourFigure 1(a) shows that culture in the RWV retarded HMECproliferation Accordingly growth inhibition correlatedwith the upregulation of p21 (WAF1) an inhibitor of cyclin-dependent kinases as detected by western blot and thisevent seems to be p53-independent since no modulationof p53 was observed in HMEC (Figure 1(b)) We also showthat no cell death is detectable after 48 and 72 h culture inthe RWV (Figure 1(c)) It is noteworthy that similar resultswere obtained when microgravity was simulated using therandom positioning machine (RPM) (data not shown) Onthe basis of results obtained by protein array on 40 proteinsinvolved in inflammation we validated the increase of IL-6and TIMP-2 in the CM from HMEC cultured for 48 and72 h in the RWV and relative controls by ELISA Figure 2(a)shows that TIMP-2 is significantly increased in the mediacollected from HMEC after 48 and 72 h in the RWV whilesecreted IL-6 was increased after 72 h culture in simulatedmicrogravity (Figure 2(b)) On these bases we decided touse 72 h conditioned media fromHMEC for the experimentson bone cells

32 HMEC Secreted Factors Impact on NHOst Cell Prolif-eration and Osteogenic Activity We evaluated the effects ofCM from HMEC on NHOst cell proliferation MTT assayrevealed a significant reduction of NHOst cell proliferationcultured in the presence of CM from HMEC in simulatedmicrogravity (Figure 3(a)) These results were confirmedby neutral red assay which estimates the number of viablecells in a culture on the basis of their ability to incorporateand bind the supravital dye neutral red in the lysosomes(Figure 3(b)) We did not detect any significant difference incell death in NHOst exposed to the conditioned media fromHMEC cultured for 72 h in the RWV and relative controls(not shown)

To evaluate osteoblastic activity NHOst cells were cul-tured for 7 and 14 days in a 24-well plate with CM fromHMEC added with an osteogenic cocktail containing 100 nMdexamethasone 50 120583ML-ascorbate-2-phosphate and 10mMglycerophosphate Two parameters were evaluated that isALP activity which has long been recognised as a reliableindicator of osteoblastic activity and calcium deposition byAlizarin Red Staining

ALP enzymatic activity was measured after 7 and 14 daysby a colorimetric assay Figure 4(a) shows that media fromHMEC in simulated microgravity inhibited ALP activityTo analyze calcium deposition we used the Alizarin Red SStaining Figure 4(b) shows that CM fromHMEC exposed tosimulated microgravity markedly inhibited the deposition ofmineral matrix


(h)

CM-CCM-RWV

0

10

20

30

40

50

60

48 72

lowast

lowastlowast

TIM

P-2

(ng106

cell)

(a)

(h)

CM-CCM-RWV

48 720

05

1

15

2

25

3

35lowast

IL-6

(ng106

cell)

(b)

Figure 2 Simulated microgravity induces TIMP-2 and IL-6 secretion by HMEC TIMP-2 (a) and IL-6 (b) were measured by ELISA in mediacollected after different times of culture in the RWV (CM-RWV) or from relative controls (CM-C)

0

01

02

03

04

05

06

07

08

24 48 72(h)

CM-CCM-RWV

Abso

rban

ce550

nm

(a)

0

02

04

06

08

1

12

14

0 24 48 72(h)

CM-CCM-RWV

Abso

rban

ce570

nm

(b)

Figure 3 CM from HMEC in simulated microgravity inhibit NHOst proliferation NHOst were cultured for different times with CM fromHMEC in simulated microgravity (CM-RWV) or by HMEC controls (CM-C) Viable cells were evaluated by MTT assay (a) and neutralred (b) and the absolute absorbance values are shown Data are expressed as the mean plusmn standard deviation of three different experimentsperformed in triplicate

33 HMEC Secreted Factors Impact on Saos-2 Cell Prolif-eration and Osteogenic Activity Many factors such as agegender and site of isolation influence the behavior of primaryosteoblasts [22]We therefore performed experiments also onan immortalized cell line to reproduce the results obtainedin NHOst and we chose Saos-2 cells because they closelyresemble primary osteoblasts [22] Indeed Saos-2 cells are

used as representative of primary osteoblasts when standardtests are evaluated [23]

Saos-2 cells were exposed toCM fromHMEC in theRWVand relative controls for different times MTT assay showsthat media fromHMEC in the RWV impair cell proliferation(Figure 5(a)) These results were confirmed when the cellswere counted (Figure 5(b))


0

01

02

03

04

05

06

07lowast

Abso

rban

ce405

nm

7 14

Days

(a)

0

02

04

06

08

1

12

14

16

18 lowast

lowast

Abso

rban

ce562

nm

7 14

Days

CM-CCM-RWV

7 days 14 days

CM-C

CM-R

WV

(b)

Figure 4 CM from HMEC in simulated microgravity inhibit NHOst activity NHOst were cultured for 7 and 14 days with mediumconditioned by HMEC in simulated microgravity (CM-RWV) or by HMEC controls (CM-C) both added with osteogenic stimuli (a) ALPenzymatic activitywas quantified by spectrophotometric analysis as described Absorbancewasmeasured at 405 nm (b)Alizarin Red Stainingwas performed Photographs were taken before acid extraction Absorbance was measured at 562 nm

Confluent Saos-2 cells were then cultured in CM fromHMEC in simulated microgravity or HMEC controls bothadded with the osteogenic cocktail and were stained withAlizarin Red to evaluate the formation of calcium phos-phate in culture [18] We found that 14-day culture in theconditioned media from HMEC in the RWV inhibited ALPactivity (Figure 6(a))The inhibition of Saos2 cell activity wasconfirmed by demonstrating lower amounts of deposition ofmineral matrix in cell cultured with the CM from HMEC inthe RWV (Figure 6(b))

4 Discussion

Bone loss in space has been reported in humans and in severalexperimental models [8] All the in vivo results obtained inspace point to major alterations of bone cells Bone cells

have been extensively studied in vitro both in space and onground using different devices to simulate microgravity toconclude that microgravity alters the morphology of thesecells [24] impairs the differentiation of osteoblasts [25] andincreases the activity of osteoclasts [8] All these results arenot surprising since gravitational forces contribute to themaintenance of bone integrity and affect bone remodeling toadjust to mechanical demands

Bone vasculature is important for skeletal developmentduring the embryonic stage postnatal growth and boneremodeling It supplies oxygen nutrients hormones cytok-ines and bone precursor cells Moreover the communicationbetween bone endothelium and bone cells is vital to regulateand modulate bone homeostasis The endothelium con-tributes to bone health by releasing osteogenic factors [26]and bone cells produce angiogenic factors that are crucial


0

05

1

15

2

25

3

35

0 h 2 days 4 days 6 days 8 days

CM-CCM-RWV

Abso

rban

ce570

nm

(a)

0

05

1

15

2

25

T0 2 days 5 days

CM-CCM-RWV

Tota

l cel

l num

ber (times105)

(b)

Figure 5 CM from HMEC in simulated microgravity inhibit Saos-2 proliferation Saos-2 were cultured for different times with CM fromHMEC in simulated microgravity (CM-RWV) or by HMEC controls (CM-C) Viable cells were evaluated by MTT assay (a) and the absoluteabsorbance values are shown After trypsinization viable cells were stained with trypan blue and counted (b)

for endothelial viability and survival under physiologicalconditions and that drive angiogenesis when needed [3]

We have shown that human endothelial cells from theumbilical vein widely used as a model of macrovascularendothelial cells are deeply influenced by simulated micro-gravity [10 11 27]These results were confirmed by our recentstudy performed on the International Space Station (ISS)[28] Other experiments have been performed on differenttypes of macrovascular endothelial cells with discordantresults which can be ascribed to poor definition of theendothelial cells used [14 15] the different culture conditionsthe use of different microgravity simulators and also theinadequate descriptions of how they were operated Less isknown about microvascular endothelial cells which coveran area 50 times greater than that of all large vesselscombined [29] In an animal model of wound healing andin a rat fibular osteotomy model microgravity retards neo-vascularization [30 31] thus indicating the occurrence ofmicrovascular endothelial dysfunction Moreover bed restwhichmimics some aspects of spaceflight causes impairmentof endothelium-dependent functions in the microcirculation[32] We have previously demonstrated that RWV-simulatedmicrogravity induces an antiangiogenic phenotype inHMEC[11] In the present study we confirm and broaden theseresults by showing that culture in the RWV retards HMECcell growth without inducing apoptosis This correlates withthe upregulation of p21 an inhibitor of the cyclinCDK2complexes necessary for the transition from the G1 to the Sphases through a p53-independent mechanism Our resultsare in disagreement with a recent report showing that culturein a clinostat induces apoptosis in pulmonary microvascularendothelial cells [12] As mentioned above these contrastingresults might be due to differences in the cells used in the cellculture conditions and in themicrogravity simulator utilized

The aim of this work was to understand whether simu-lated microgravity impairs endothelial-osteoblast communi-cation To this purpose we evaluated the effects producedon osteoblasts by CM from HMEC cultured in simulatedmicrogravity

We show that HMEC release factors that retard thegrowth of osteoblasts and severely impair their osteogenicactivity It is noteworthy that we found increased amounts ofsecreted TIMP-2 and IL-6 known to affect both endothelialcells and osteoblasts Interestingly TIMP-2 inhibits endothe-lial cell proliferation by a matrix metalloproteases (MMP)independent mechanism [33] and might therefore play arole in HMEC growth retardation in simulated microgravityTIMP-2 also impairs osteoblast activity Indeed TIMP-2nearly abolishes ALP expression [34] by inhibiting MT1-MMP (membrane type 1-metalloprotease) [34] a proteasewhich is implicated in multiple steps of osteogenic differ-entiation and is mainly involved in ALP upregulation [35]Interestingly TIMP-2 inhibits cell survival of osteoblastsforced to transdifferentiate into osteocytes [36] This resultmight offer a molecular explanation at least in part to thelysis of osteocytes in spaceflight described byBlaber et al [37]In media from HMEC cultured in the RWV we also foundincreased amounts of IL-6 a pleiotropic cytokine implicatedin acute phase response and inflammation IL-6 not onlypromotes endothelial dysfunction [38] but also affects humanosteoblast differentiation [39] thus contributing to osteope-nia

We therefore propose that microgravity impacts bothdirectly and indirectly on osteoblasts Microgravity has beenshown to directly inhibit osteoblasts In addition by modu-lating microvascular endothelial cell function microgravityindirectly exerts inhibitory effects on osteoblasts


0

01

02

03

04

05

06

07

08

7 14

Days

lowast

Abso

rban

ce405

nm

(a)

0

02

04

06

08

1

12

14

7 14

Days

CM-CCM-RWV

lowast

lowast

Abso

rban

ce562

nm

7 days 14 days

CM-C

CM-R

WV

(b)

Figure 6 CM fromHMEC in simulated microgravity inhibit Saos-2 activity Saos-2 were cultured for 7 and 14 days with CM fromHMEC insimulated microgravity (CM-RWV) or by HMEC controls (CM-C) both added with osteogenic stimuli (a) ALP enzymatic activity and (b)Alizarin Red Staining were performed as above

The current space programs onboard the ISS and thefuture human exploration of Mars require long durationmissions However several biomedical issues still need to beclarified before these missions can take place without causinghealth problems to the astronauts Our results suggest thatendothelial dysfunction might represent a common denomi-nator for cardiovascular deconditioning and for bone loss andoffer a new light to interpret the behaviour of mammalianskeleton in microgravity Eventually these results mightfoster studies to develop countermeasures that target theendothelium to improve both bone homeostasis and vascularfunction

Conflict of InterestsThe authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This work was supported by a grant from the European SpaceAgency to Jeanette A M Maier

References

[1] T C Phan J Xu andM H Zheng ldquoInteraction between osteo-blast and osteoclast impact in bone diseaserdquoHistology and His-topathology vol 19 no 4 pp 1325ndash1344 2004

[2] B Guillotin C Bourget M Remy-Zolgadri et al ldquoHuman pri-mary endothelial cells stimulate human osteoprogenitor celldifferentiationrdquo Cellular Physiology and Biochemistry vol 14no 4ndash6 pp 325ndash332 2004

[3] J Kular J Tickner S M Chim and J Xu ldquoAn overview of theregulation of bone remodelling at the cellular levelrdquo ClinicalBiochemistry vol 45 no 12 pp 863ndash873 2012


[4] B Decker H Bartels and S Decker ldquoRelationships betweenendothelial cells pericytes and osteoblasts during bone for-mation in the sheep femur following implantation of trical-ciumphosphate-ceramicrdquoAnatomical Record vol 242 no 3 pp310ndash320 1995

[5] R D Prisby M W Ramsey B J Behnke et al ldquoAging reducesskeletal blood flow endothelium-dependent vasodilation andno bioavailability in ratsrdquo Journal of Bone andMineral Researchvol 22 no 8 pp 1280ndash1288 2007

[6] L Vico P Collet A Guignandon et al ldquoEffects of long-termmicrogravity exposure on cancellous and cortical weight-bear-ing bones of cosmonautsrdquo The Lancet vol 355 no 9215 pp1607ndash1611 2000

[7] AD LeBlanc E R SpectorH J Evans and J D Sibonga ldquoSkel-etal responses to space flight and the bed rest analog a reviewrdquoJournal of Musculoskeletal and Neuronal Interactions vol 7 no1 pp 33ndash47 2007

[8] M P Nagaraja and D Risin ldquoThe current state of bone lossresearch data from spaceflight and microgravity simulatorsrdquoJournal of Cellular Biochemistry vol 114 no 5 pp 1001ndash10082013

[9] S I M Carlsson M T S Bertilaccio E Ballabio and J A MMaier ldquoEndothelial stress by gravitational unloading effectson cell growth and cytoskeletal organizationrdquo Biochimica etBiophysica Acta vol 1642 no 3 pp 173ndash179 2003

[10] S Versari A Villa S Bradamante and J A M Maier ldquoAlter-ations of the actin cytoskeleton and increased nitric oxide syn-thesis are common features in human primary endothelial cellresponse to changes in gravityrdquo Biochimica et Biophysica ActamdashMolecular Cell Research vol 1773 no 11 pp 1645ndash1652 2007

[11] M Mariotti and J A M Maier ldquoGravitational unloading indu-ces an anti-angiogenic phenotype in human microvascularendothelial cellsrdquo Journal of Cellular Biochemistry vol 104 no1 pp 129ndash135 2008


[13] S M Grenon M Jeanne J Aguado-Zuniga M S Conteand M Hughes-Fulford ldquoEffects of gravitational mechanicalunloading in endothelial cells association between caveolinsinflammation and adhesion moleculesrdquo Scientific Reports vol3 article 1494 2013

[14] LMorbidelli MMonici NMarziliano et al ldquoSimulated hypo-gravity impairs the angiogenic response of endothelium byup-regulating apoptotic signalsrdquo Biochemical and BiophysicalResearch Communications vol 334 no 2 pp 491ndash499 2005

[15] M Infanger P Kossmehl M Shakibaei et al ldquoInduction ofthree-dimensional assembly and increase in apoptosis ofhuman endothelial cells by simulated microgravity impact ofvascular endothelial growth factorrdquo Apoptosis vol 11 no 5 pp749ndash764 2006

[16] N V Rodionova and V S Oganov ldquoChanges of cell-vascularcomplex in zones of adaptive remodeling of the bone tissueunder microgravity conditionsrdquo Advances in Space Researchvol 32 no 8 pp 1477ndash1481 2003

[17] PDowling andMClynes ldquoConditionedmedia fromcell lines acomplementarymodel to clinical specimens for the discovery ofdisease-specific biomarkersrdquo Proteomics vol 11 no 4 pp 794ndash804 2011

[18] M Leidi F DelleraMMariotti and J AMMaier ldquoHighmag-nesium inhibits human osteoblast differentiation in vitrordquoMag-nesium Research vol 24 no 1 pp 1ndash6 2011


[20] S Castiglioni S Casati R Ottria P Ciuffreda and J A MMaier ldquoN6-isopentenyladenosine and its analogue N6-ben-zyladenosine induce cell cycle arrest and apoptosis in bladdercarcinoma T24 cellsrdquo Anti-Cancer Agents in Medicinal Chem-istry vol 13 no 4 pp 672ndash678 2013

[21] S Casati R Ottria E Baldoli E Lopez J A Maier and PCiuffreda ldquoEffects of cytokinins cytokinin ribosides and theiranalogs on the viability of normal and neoplastic human cellsrdquoAnticancer Research vol 31 no 3 pp 3401ndash3406 2011

[22] E M Czekanska M J Stoddart R G Richards and J S HayesldquoIn search of an osteoblast cell model for in vitro researchrdquoEuropean Cells and Materials vol 24 pp 1ndash17 2012

[23] L Saldana F Bensiamar A Bore and N Vilaboa ldquoIn searchof representative models of human bone-forming cells forcytocompatibility studiesrdquo Acta Biomaterialia vol 7 no 12 pp4210ndash4221 2011

[24] N Nabavi A Khandani A Camirand and R E HarrisonldquoEffects of microgravity on osteoclast bone resorption andosteoblast cytoskeletal organization and adhesionrdquo Bone vol49 no 5 pp 965ndash974 2011

[25] G Carmeliet G Nys and R Bouillon ldquoMicrogravity reducesthe differentiation of human osteoblastic MG-63 cellsrdquo Journalof Bone and Mineral Research vol 12 no 5 pp 786ndash794 1997

[26] S M Chim J Tickner S T Chow et al ldquoAngiogenic factorsin bone local environmentrdquo Cytokine amp Growth Factor Reviewsvol 24 no 3 pp 297ndash310 2013

[27] M Mariotti and J A M Maier ldquoHumanMicro- and macrovas-cular endothelial cells exposed to simulated microgravityupregulate hsp70rdquoMicrogravity Science and Technology vol 21no 1-2 pp 141ndash144 2009

[28] S Versari G Longinotti L Barenghi J A Maier and S Brad-amante ldquoThe challenging environment on board the Interna-tional Space Station affects endothelial cell function by trig-gering oxidative stress through thioredoxin interacting proteinoverexpression the ESA-SPHINX experimentrdquo FASEB Journalvol 27 pp 4466ndash4475 2013

[29] S Danese E Dejana and C Fiocchi ldquoImmune regulation bymicrovascular endothelial cells directing innate and adaptiveimmunity coagulation and inflammationrdquo The Journal ofImmunology vol 178 no 10 pp 6017ndash6022 2007

[30] J M Davidson A M Aquino S C Woodward and W WWilfinger ldquoSustained microgravity reduces intrinsic woundhealing and growth factor responses in the ratrdquo The FASEBJournal vol 13 no 2 pp 325ndash329 1999

[31] M E Kirchen K M OrsquoConnor H E Gruber et al ldquoEffects ofmicrogravity on bone healing in a rat fibular osteotomymodelrdquoClinical Orthopaedics and Related Research no 318 pp 231ndash2421995

[32] M Coupe J O Fortrat I Larina G Gauquelin-Koch CGharib and M A Custaud ldquoCardiovascular deconditioningfrom autonomic nervous system to microvascular dysfunc-tionsrdquo Respiratory Physiology amp Neurobiology vol 169 supple-ment 1 pp S10ndashS12 2009

[33] W G Stetler-Stevenson and D Seo ldquoTIMP-2 an endogenousinhibitor of angiogenesisrdquo Trends in Molecular Medicine vol 11no 3 pp 97ndash103 2005

[34] S Barthelemi J Robinet R Garnotel et al ldquoMechanical forces-induced human osteoblasts differentiation involves MMP-2MMP-13MT1-MMP proteolytic cascaderdquo Journal of CellularBiochemistry vol 113 no 3 pp 760ndash772 2012


[35] P Manduca A Castagnino D Lombardini et al ldquoRole of MT1-MMP in the osteogenic differentiationrdquo Bone vol 44 no 2 pp251ndash265 2009

[36] M A Karsdal T A Andersen L Bonewald and C Chris-tiansen ldquoMatrix metalloproteinases (MMPs) safeguard osteo-blasts from apoptosis during transdifferentiation into osteo-cytes MT1-MMP maintains osteocyte viabilityrdquo DNA and CellBiology vol 23 no 3 pp 155ndash165 2004

[37] E A Blaber N Dvorochkin C Lee et al ldquoMicrogravity inducespelvic bone loss through osteoclastic activity osteocytic oste-olysis and osteoblastic cell cycle inhibi tion by CDKN1ap21rdquoPLoS ONE vol 8 no 4 Article ID e61372 2013

[38] S Wassmann M Stumpf K Strehlow et al ldquoInterleukin-6induces oxidative stress and endothelial dysfunction by over-expression of t he angio tensin II type 1 receptorrdquo CirculationResearch vol 94 no 4 pp 534ndash541 2004

[39] B Peruzzi A Cappariello A del Fattore N Rucci F deBenedetti and A Teti ldquoC-Src and IL-6 inhibit osteoblastdifferentiation and integrate IGFBP5 signallingrdquo Nature Com-munications vol 3 article 630 2012

Research ArticlePhenotypic Switch Induced by Simulated Microgravity onMDA-MB-231 Breast Cancer Cells

Maria Grazia Masiello12 Alessandra Cucina2 Sara Proietti2 Alessandro Palombo3

Pierpaolo Coluccia2 Fabrizio DrsquoAnselmi2 Simona Dinicola2 Alessia Pasqualato2

Veronica Morini2 and Mariano Bizzarri3

1 Department of Clinical and Molecular Medicine ldquoSapienzardquo University of Rome Piazza Sassari 3 00161 Rome Italy2 Department of Surgery ldquoPietroValdonirdquo ldquoSapienzardquo University of Rome Via A Scarpa 14 00161 Rome Italy3 Department of Experimental Medicine ldquoSapienzardquo University of Rome Systems Biology Group Viale Regina Elena 324Via A Scarpa 14 00161 Rome Italy


Received 14 May 2014 Accepted 23 July 2014 Published 18 August 2014


Copyright copy 2014 Maria Grazia Masiello et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Microgravity exerts dramatic effects on cell morphology and functions by disrupting cytoskeleton and adhesion structures as wellas by interfering with biochemical pathways and gene expression Impairment of cells behavior has both practical and theoreticalsignificance given that investigations of mechanisms involved in microgravity-mediated effects may shed light on how biophysicalconstraints cooperate in shaping complex living systems By exposing breast cancer MDA-MB-231 cells to simulated microgravity(sim0001 g) we observed the emergence of twomorphological phenotypes characterized by distinctmembrane fractal values surfacearea and roundness Moreover the two phenotypes display different aggregation profiles and adherent behavior on the substrateThese morphological differences are mirrored by the concomitant dramatic functional changes in cell processes (proliferation andapoptosis) and signaling pathways (ERK AKT and Survivin) Furthermore cytoskeleton undergoes a dramatic reorganizationeventually leading to a very different configuration between the two populations These findings could be considered adaptive andreversible features given that by culturing microgravity-exposed cells into a normal gravity field cells are enabled to recover theiroriginal phenotype Overall these data outline the fundamental role gravity plays in shaping form and function in living systems

1 Introduction

Space flights induce relevant changes in human physiologysuch as bone loss muscle atrophy deregulation of immunefunction hematological anomalies and cardiovascular func-tion impairment Microgravity effects may be ascribed tosystemic interferences with body fluids distribution disap-pearance of fluid shear perturbation of the circadian clockaltered endothelial function and reduced loading on skeletalstructures [1] Yet a direct effect on cell and signalingpathways inside the cell has been documented despite thefact that microgravity has been previously thought to be tooweak for contrasting the intermolecular forces [2]Thereby itis likely that spaceflight could exert its detrimental effects onastronauts via changes in cellular structure andor functions

Several studies performed both in simulated and actualmicrogravity have shown that normal as well as neoplas-tic cells undergo dramatic changes after exposition to amicrogravity field Cell morphology as well as features ofsubcellular organelles and cytoskeleton structure has beenreported to be dramatically influenced by gravity [3 4] Sim-ilarly relevant modifications in tissue organization have beenrecorded in microgravity-exposed organs andor animals [56] Shape changes are likely to be mediated by concomitantstructural rearrangement of cytoskeleton (CSK) which isseverely disorganized undermicrogravity [7 8] CSK conveysmechanical signals into the cells and by that way it influencesboth biochemical pathways [9 10] and gene expression [1112] As a consequence many metabolic proliferative anddifferentiating processes end up to be deeply perturbed [13]



Microgravity effects may be ascribed to both indirect anddirect effects [14] Meanwhile specialized cells and structuresin the plant realm have been found to be sensitive to evensubtle change in gravity vector [15] no components in themammalian cells have been so far identified as having asufficiently large mass density difference in respect to the sur-roundingmedium thus the force exerted by the gravitationalfield is nowhere higher than the energy of random thermalmotion and cannot significantly modify the behaviour of anysingle subcellular structure Instead mammalian cells maybe able to sense some environmental changes due to gravityaffecting a wide range of biophysical parameters buoyancyshear forces viscosity diffusion process and many othersYet a lot of gravity-related phenomena at the cellular levelinvolving shape rearrangement cytoskeleton disruption andeven modified gene expression would hardly be explainedby only considering changes in ldquoexternalrdquo environmentalbiophysical parameters Indeed gravity may likely affectsome general properties of the systems acting ldquodirectlyrdquo asan organizing field parameter We have previously reportedthat by ldquoremovingrdquo the gravitational constraint according tothe nonequilibrium theory [16] murine osteoblasts under-went a transition after a bifurcation point thus recoveringdegrees of freedom enabling the system in accessing newattractor states that is new phenotypic configurations [17]Indeed microgravity induces the emergence of two distinctphenotypes characterized by different morphologies Hereinwe investigate if a similar pattern could be retrieved inbreast cancer cells and how such features are associated withdifferences in their biochemical pathways Indeed conflictingdata have been reported by investigations carried out oncancer cells exposed to microgravity some authors haverecorded an overall inhibitory effect on cancer cell prolif-eration motility and survival [18 19] whereas others haveobserved the opposite [20ndash22] We hypothesize that suchresults may be likely explained by the emergence of distinctcell phenotypes characterized by different functional andreproductive features


21 RPM (Random Positioning Machine) Microgravity con-ditions were simulated by a Desktop RPM a particular kindof 3D clinostat [23] manufactured by Dutch Space (LeidenThe Netherlands) The degree of microgravity simulationdepends on angular speed and on the inclination of the diskThese tools do not actually eliminate the gravity but allowyou to apply a stimulus rather than a unidirectional omni-directional 1 g Effects generated by the RPM are comparableto those of the real microgravity provided that the directionchanges are faster than the response time of the system togravity field The desktop RPM was located in a standardincubator (to maintain temperature CO

2 and humidity

levels) and connected to the control console through standardelectric cables

22 Cell Culture MDA-MB-231 human breast cancer cellline was purchased from European Collection of Cell Cul-tures (ECACC Sigma-Aldrich St Louis MO USA) Cells

were seeded into Nunc OptiCell Cell Culture Systems gas-permeable cell culture disks (Thermo Scientific RochesterUSA) and cultured in Dulbeccorsquos modified Eaglersquos medium(DMEM Euroclone Ltd Cramlington UK) supplementedwith 10 Fetal Bovine Serum (FBS HyClone LaboratoriesLogan UT USA) 200mM L-glutamine 100 IUmL Peni-cillin and 100 120583gmL Streptomycin (all from Euroclone LtdCramlington UK) Then OptiCells containing subconfluentmonolayers were fixed onto the RPM as close as possibleto the center of the platform which was rotated at aspeed of 60∘s using the random mode of the machineOn ground control (1 g static cultures) and RPM cultureswere kept in the same humidified incubator at 37∘C in anatmosphere of 5 CO

2in air Experiments were performed

for 24 and 72 hours After 24 and 72 hours of microgravityexposure cell clumps swimming in culture supernatants werefound in addition to adherent cells and separately collectedThe three cell populations (on ground control cells RPMadherent cells and RPM cell clumps) were characterizedseparately

23 Optical Microscopy Cell clumps were collected washedin PBS and deposited onto a clearly defined area of a glassslide using a Shandon CytoSpin 4 Cytocentrifuge ThermoScientific while maintaining cellular integrity Cell clumpsand adherent and on ground control cells were fixed in 4paraformaldehyde for 10 minutes at 4∘C and photographedwith Nikon Coolpix 995 digital camera coupled with ZeissAxiovert optical microscope The images were obtained witha 320x magnification saved as TIFF files and used for imageanalysis

24 Image Analysis Image analysis was performed on 10images for each group of MDA-MB-231 cells As the analysiswas performed blindly the image groups were classified asfollows A (on ground cells 24 h) B (RPM adherent cells24 h) C (RPM cell clumps 24 h) D (on ground cells 72 h)E (RPM adherent cells 72 h) and F (RPM cell clumps 72 h)In each image single randomly chosen cells (50 for eachgroup) were contoured with a fine black marker by differentresearchers simply scanned and cataloged according to thetime of study 24 and 72 hours This method was chosenbecause pathologists are used to correlate the shape the cellsacquire with their malignancy by means of morphologicalqualitative and subjective observations Thus we decided toperform a semiautomatic analysis coupling the expertise ofresearchers with a computerized parameterization methodAll the images were processed by Adobe Photoshop CS4All the pictures (ie all the sheets of the groups for eachtime point) were resized at 2560 times 1920 pixels according tooriginal scale of image acquisition For each black contouredcell edges were refined Then cells were black filled andthreshold was adjusted in order to exclude from the imageother cells and background For each time point a single sheetof all the cells considered was created To obtain single cellshape parameters (area 119860 roundness solidity and fractaldimension FD) ImageJ v147h software was used Then


the software analyzed single cells by the function ldquoshapedescriptorrdquo In addition to area 119860 were calculated

Roundness = 4119860120587radicma

Solidity = 119860CA

(1)

where 119860 is the area of the cell ma is the major axis and CAis the convex area namely the area of the convex hull of theregion The convex hull of a region is the smallest region thatsatisfies two conditions (a) it is convex and (b) it contains theoriginal region

As for FD it was obtained by means of box countingmethod using FracLac plugin

FD = lim120576rarr0

[1 minuslog [119871

120576(119862)]

log 120576] (2)

where 119862 is the considered curve 119871 is the length of the curve119862 and 120576 is the length of the segment used as unit to calculate119871

Single graphs about roundness solidity and FD wereobtained for each set of images

25 Fluorescence Microscopy MDA-MB-231 cells were fixedin 4 paraformaldehyde for 10 minutes at 4∘C and incu-bated over night at 4∘C with PBS (CMF Calcium andMagnesium Free) 15 goat serum plus the following specificantibodies anti-120572-tubulin (T5168 Sigma-Aldrich) and anti-vimentin (sc-6260 Santa Cruz biotechnology) For F-actinvisualization rhodamine-phalloidin (Invitrogen MolecularProbes Eugene) was used Cells were washed three timeswith PBS (1 BSA 02 Triton X 100) and incubated withrhodamine-phalloidin the anti-mouse IgG-FITC PN IM1619secondary antibody (Beckman-Coulter Inc Fullerton CAUSA) and HOECHST 33342 (Sigma-Aldrich St Louis MOUSA) to stain the DNA Finally cells were washed mountedin buffered glycerol (01M pH 95) and analyzed usinga Zeiss Fluorescent Microscope The images were scannedunder 40x objective

26 Cell Cycle Analysis Cell clumps were collected andcentrifuged and pellets were trypsinized and washed twicewith PBS (Phosphate Buffered Saline Sigma-Aldrich StLouis MO USA) Adherent and ground control cells weretrypsinized and washed twice with PBS Cells were fixed with70 ethanol at 4∘C for 24 h and stainedwithDNAPREP Stain(Beckman Coulter Fullerton USA) at 4∘C overnight Stainedcells weremeasured by flow cytometry Cell cycle analysis wasperformed three times

27 Annexin V7-AAD Staining Cell clumps were collectedand centrifuged and pellets were trypsinized and washedtwice with PBS Adherent cells and ground control cells weretrypsinized andwashed twicewith PBSThe cells were stainedwith FITC labeled annexin V7-AAD (7-aminoactinomycin-D) according to the manufacturerrsquos instructions (annexin

V7-AAD kit Beckman Coulter Marseille France) Brieflya washed cell pellet (5 times 104 cellsmL) was resuspended in500120583L binding buffer 10 120583L of annexinV together with 20120583L7-AAD was added to 470 120583L cell suspension The cells wereincubated for 15min on ice in the dark The samples wereanalyzed by flow cytometry Apoptosis assay was performedthree times

28 Flow Cytometry Flow cytometry was performed usingan EPICS Coulter XL (Beckman-Coulter Inc) The flu-orescence of 20000 events was measured An excitationwavelength of 488 nmwas used in combinationwith standardfilters to discriminate between the FL1 (forward scatter) andFL3 (side scatter) channels Data were analyzed byModFit LTSoftware (Verity Software Inc USA)

29 Western Blot Cell clumps were washed twice withice-cold PBS and resuspended in RIPA lysis buffer (SigmaChemical Co) Adherent and ground control cells werewashed twice with ice-cold PBS and scraped in RIPAlysis buffer (Sigma Chemical Co) A mix of proteaseinhibitors (Complete-Mini Protease Inhibitor CocktailTablets Roche Mannheim Germany) and phosphataseinhibitors (PhosStop Roche Mannheim Germany) wasadded just before use Cellular extracts were then centrifugedat 8000timesg for 10min The protein content of supernatantswas determined using the Bradford assay For westernblot analysis cellular extracts were separated on SDS-polyacrylamide gels with a concentration of acrylamidespecific for the proteins studied Proteins were blottedonto nitrocellulose membranes (BIO-RAD Bio-RadLaboratories Hercules CA USA) and probed with thefollowing antibodies anti-Cyclin D1 (AB-90009) fromImmunological Sciences anti-survivin (number 2808)anti-phospho-AKT (ser473) (number 9271S) anti-AKT(number 9272S) anti-phospho-ERK12 (number 9106)anti-cleaved PARP (number 9541) anti-GAPDH (number2118) all from Cell Signaling Technology anti-Bax (sc-493)anti-Bcl-2 (sc-492) anti-ERK1 (sc-94) all from Santa CruzBiotechnology Antigens were detected with enhancedchemiluminescence kit (Amersham Biosciences LittleChalfont Buckinghamshire England) according to themanufacturerrsquos instructions All Western blot images wereacquired and analyzed through Imaging Fluor S densitometer(Biorad-Hercules)

210 Statistical Analysis Data were expressed as mean plusmnstandard deviation (SD) or mean plusmn standard error (SE)Data were statistically analyzed with the analysis of variance(ANOVA) followed by the Bonferroni post-test Differenceswere considered significant at the level of 119875 lt 0 05 Statisticalanalysis was performed by using GraphPad Instat software(GraphPad Software Inc San Diego CA USA)

3 Results

31 Effect of Microgravity on MDA-MB-231 MorphologyMDA-MB-231 cell line grew as a monolayer when culturedunder static 1 g condition (on ground control Figures 1(a)


100120583m

(a)

100120583m

(b)

100120583m

(c)

100120583m

(d)

100120583m

(e)

100120583m

(f)

Figure 1 Microphotographs of MDA-MB-231 by optical microscopy MDA-MB-231 cell line in on ground control condition at 24 (a) and 72hours (c)MDA-MB-231 cells exposed tomicrogravity for 24 (b) and 72 hours (d) RPMcell clumps after reseeding into a normal gravitationalfield after 6 (e) and 24 hours (f) Magnification times320 (a) (b) (c) (d) times100 (e) (f)

and 1(c)) After 24 and 72 hours of simulated microgravityexposure cells were distributed into two populations thefirst adhering to the substrate represented by flat spindlecells the second represented by rounded cells aggregatedin cell clumps floating in the culture medium (Figures 1(b)and 1(d)) This distribution does not represent a transitorystate given that the percentage of cells at both 24 and 72hours still remains constant However beside the fact thatsuch changes are likely to involve several modifications onshape and biological function the observed nonadherentphenotype is still wholly reversible after 72 hours Indeedafter reseeding into a normal gravitational field cell clumpswere de novo able to adhere to the culture plate alreadyafter 6 hours (Figure 1(e)) and to fully recover their native

morphological traits and topological distribution after 24hours (Figure 1(f))

32 Effect of Microgravity on Quantitative MorphologicalParameters Quantitative image analysis was performed byquantifying roundness solidity and fractal dimension (FD)Significant differences were recorded among the three cellpopulations (Table 1) Roundness no statistically significantdifferences between on ground cells and RPM adherent cellshave been observed at both 24 and 72 hours Instead RPMcell clumps showed a significant strong increase in roundnesscompared to control and RPM adherent cells at 24 and72 hours Solidity at 24 hours no statistically significantdifferences between on ground cells and RPM adherent cells


Table 1

Roundness plusmnSE Solidity plusmnSE FD plusmnSE24 hours

On ground cells 04563 00301 06690 00225 17482 00091RPM adherent cells 03991 00275 06499 00200 17406 00063RPM cell clumps 07894 00219 lowastlowast 08966 00263 lowastlowast 14625 00015 lowastlowast

72 hoursOn ground cells 03227 00263 04687 00136 16677 00036RPM adherent cells 04081 00311 06115 00226 lowast 17245 00067RPM cell clumps 07961 00208 lowastlowast 08573 00469 lowastlowast 14990 00015 lowastlowast

Roundness solidity and fractal dimension (FD) mean values plusmn SE in on ground control cells RPM adherent cells and RPM cell clumps lowastP lt 001 versus onground control cells lowastlowastP lt 0001 versus on ground control and RPM adherent cells by ANOVA followed by Bonferroni post-test

were recorded meanwhile in RPM cell clumps the solidityindex was significantly higher with respect to on groundcells and RPM adherent cells At 72 hours the solidity indexsignificantly increased in both RPM cell populations reach-ing its highest level in RPM cell clumps Fractal dimensionno statistically significant differences between on groundcells and RPM adherent cells were recorded both at 24 and72 hours Instead RPM cell clumps showed a statisticallysignificant decrease in FD compared to control and RPMadherent cells at 24 and 72 hours These results are coherentwith the qualitativemorphological assessment and confirmedthat microgravity exposure leads to the emergence of twomorphologically distinct cell populations

33 Effect of Microgravity on MDA-MB-231 CytoskeletonArchitecture After 24 hours of microgravity exposure bothMDA-MB-231 RPM adherent cells and RPM cell clumpsshowed a large rearrangement of F-actin 120572-tubulin andvimentin compared to on ground control cells In onground control cells the network of cytosolic F-actin bundlesappeared well organized and the microtubules appearedorderly radiating from the perinuclear area throughout thecytoplasm toward the cell periphery (Figure 2(a)) In RPMadherent cells the actin filaments showed a disappearanceof the complex cytosolic network which appeared mostlylocalized on the cell border microtubules were disorganizedwith a more evident thickening in perinuclear position(Figure 2(b)) In floating cell clumps the actin meshworkappeared completely disrupted and the filaments weremainly localized behind the cell border Tubulin mesh-work was also completely disrupted and a slight diffusefluorescence was observed spreading throughout the entirecytoplasm (Figure 2(c)) In the on ground cells vimentinfilaments were well organized all over the cytoplasm (Fig-ure 3(a)) In both RPM adherent cells and cell clumps thevimentin network was disrupted appearing in the form ofdense aggregates closely associated with the nucleus (Figures3(b) and 3(c)) Cytoskeleton rearrangements were almoststable given that no significant changes have been observedeven after 72 hours in microgravity-exposed cells (data notshown)

34 Microgravity Modifies MDA-MB-231 Cell Cycle Distribu-tion MDA-MB-231 cells subjected tomicrogravity displayed

relevant modification in their cell cycle (Figures 4(a) and4(b)) Nonadherent RPM-treated MDA-MB-231 cells weredistributed in a significantly different manner when com-pared to both control and RPM-adherent cells indeed after24 hours floating cell clumps in G0G1 and in S phasewere significantly decreased whereas cells in G2M phaseincrease up to 6-fold On the contrary adherent RPM-treated cells displayed only a slight increase in the S phasedistribution when compared to controls After 72 hours ofmicrogravity exposure MDA-MB-231 RPM cell clumps stillshowed a relevant decrease in the S phase thus demonstratinga persistent inhibition of cell growth cells number in G2Mphase was significantly higher meanwhile no significantchange in G0G1 was observed (Figure 4(b)) Again besidesminor differences control and adherent RPM-treated cellsdisplayed an overlapping distribution in theG0G1 andG2Mphase whereas the percentage of cells in the S phase was stillhigher than that recorded in floating RPM cell clumpsThesedata are exemplarily mirrored by Cyclin D1 data Cyclin D1is one of the main factors that regulate the activation of thecell cycle and its increase is required to foster cell growth Asexpected a statistically significant decrease of CyclinD1 levelsin RPMcell clumpswas recordedmeanwhile CyclinD1 levelsare higher in adherent proliferating RPM cells as well as incontrol samples These effects were observed at both 24 and72 hours (Figure 4(c))

35 Microgravity Induces Apoptosis in MDA-MB-231 CellClumps Data obtained by cytofluorimetric assays reporteda statistically significant increase in the apoptotic rate after 24and 72 hours of microgravity exposure in cell clumps withrespect to both adherent cells and on ground control cells(Figure 5(a)) Western blot analysis revealed a statisticallysignificant increase of BaxBcl-2 ratio at 72 hours in RPMcell clumps compared to RPM adherent cells and on groundcontrol cells (Figure 5(b)) Similarly cleaved-PARP levels adirect marker of caspase-3 activation [24] were significantlyincreased at 24 and 72 hours in RPM cell clumps compared toRPM adherent cells and on ground control cells (Figure 5(c))Overall these data suggest that nonadherent cells were signif-icantly constrained in their viability given that microgravityinhibits cell growth and at the same time enhances theapoptotic process Adherent cells in microgravity on the


F-Actin Merge120572-Tubulin

(a)


(b)


(c)

Figure 2 Immunofluorescence images of F-actin and 120572-tubulin inMDA-MB-231 Rhodamine-phalloidin staining of MDA-MB-231 showingF-actin distribution patterns (red color) and immunostaining of 120572-tubulin (green color) and HOECHST 33342 to stain nuclei (blue color)after 24 hours in on ground control cells (a) RPM adherent cells (b) and RPM cell clumps (c) Magnification times400

contrary display only minor changes in both apoptosis andproliferation rate

36 Microgravity Modifies MDA-MB-231 Survival PathwaysMicrogravity exposure is associated with a statistically sig-nificant decrease in the phosphorylation of AKT in adherentcells and cell clumps with respect to on ground control cellsafter 24 h Instead after 72 hours of microgravity expositionRPM adherent cells showed a statistical increase of p-AKTexpression with respect to on ground control cells andRPM cell clumps (Figure 6(a)) such biphasic effect on AKTactivation may help explain the biphasic trend observedin apoptosis rate in adherent RPM-exposed cells apoptosisincreases indeed at 24 hours when p-AKT values are lowthe opposite is observed when p-AKT levels increase at72 hours A similar behavior may be described for thetwo other prosurvival factors Survivin and phosphorylated-ERKMicrogravity exposure induced a statistically significantdecrease in Survivin levels in both adherent and nonadherent

RPM-treated cells at 24 hours However at 72 hours Survivinlevels significantly increased in RPM adherent cells anddecreased in nonadherent RPM-treated cells (Figure 6(b))Likewise ERK phosphorylation was severely inhibited inRPM cell clumps after 24 and 72 hours in respect tovalues observed in both RPM adherent and control cells(Figure 6(c)) Taken as a whole prosurvival factors increasedin adherent RPM-treated cells meanwhile they decreasedin nonadherent RPM-exposed cells mirroring so far theobserved mentioned trend in apoptosis

4 Discussion

Breast cancer cells exposed to microgravity acquire twodistinct phenotypes already after the first 24 hours Such out-standing result has been previously observed in osteoblastscultured in microgravity [17] and can be interpreted inthe light of the nonequilibrium theory Briefly a dissipa-tive nonlinear system sufficiently far from the equilibrium


Vimentin Merge

(a)

Vimentin Merge

(b)

Vimentin Merge

(c)

Figure 3 Immunofluorescence images of vimentin in MDA-MB-231 Immunostaining of vimentin (green color) and HOECHST 33342 tostain nuclei (blue color) after 24 hours in on ground control cells (a) RPM adherent cells (b) and RPM cell clumps (c) Magnification times400

can form spatial stationary patterns after experiencing aphase transition leading to new asymmetric configurations[25] These states are equally accessible that is to say thatthere exists a complete symmetry between the emergingconfigurations as it is reflected in the symmetry of thebifurcation diagram However the superimposition of anexternal field even if a weak one like gravity may breakthe systemrsquos symmetry bestowing a preferential directionalityaccording to which the system will preferentially evolve intoone of the states and not the others Indeed bifurcations far

from equilibrium endow a system with a very pronouncedsensitivity allowing it to capture the slightest environmentalasymmetry and select a preferred polarity or chirality Inother words the ldquoweakrdquo force dramatically influences thesystem in selecting one out ofmany other configurations [26]On the contrary by removing the gravitational constraints thesystem can freely access different attractor states recoveringhenceforth new configuration states (ldquophenotypesrdquo) Suchmodel has been experimentally confirmed showing that sev-eral cell components characterized by a nonlinear dynamics


0

10

20

30

40

50

60

70

G0G1 S G2M

Cell

s in

each

pha

se (

)

On groundRPM (adherent cells)RPM (cell clumps)

24h

lowastlowastlowast

lowastlowast

lowast

(a)

Cell

s in

each

pha

se (

)

G0G1 S G2M


0102030405060708090

72h

lowastlowastlowast

(b)

Cyclin D1

005

115

225

335

4

Relat

ive e

xpre

ssio

n

GAPDH

On RPM RPMground (adherent (cell

cells) clumps)


cells) clumps)

24h

24h

72h

72h

lowastlowast

lowast


(c)

Figure 4 Cell cycle analysis in MDA-MB-231 Cells distribution along the different phases of the cell cycle at 24 (a) and 72 hours (b) in onground control cells RPM adherent cells and RPM cell clumps (c) Immunoblot bar chart showing the expression of Cyclin D1 in MDA-MB-231 in on ground control cells RPM adherent cells and RPM cell clumps at 24 and 72 hours Columns and bars represent densitometricquantification of optical density (OD) of specific protein signal normalized with the OD values of the GAPDH served as loading control Eachcolumn represents the mean value plusmn SD of three independent experiments lowast119875 lt 0 05 lowastlowast119875 lt 0 01 lowastlowastlowast119875 lt 0 001 versus on ground control119875 lt 0 01 119875 lt 0 001 versus RPM adherent cells by ANOVA followed by Bonferroni post-test

when exposed to microgravity may experience bifurcationtransitions leading to the appearance of new self-organizedstates from an initially homogeneous conformation [27 28]It is tempting to speculate that such transitions may arise inthe cell when self-organization processes (cytoskeleton com-ponents assembly andmitosis) take place In our experimentthe annihilation of gravity enables the system to recovermoredegree of freedom through subsequent symmetry breakingswith the appearance of new morphological and functionalphenotypes

IndeedMDA-MB-231 cells exposed tomicrogravity werealmost equally split into two distinct populations char-acterized by very different morphologies The first cluster

is represented by cells adherent to the substrate roughlypreserving their native spindle profile The second one isrepresented by rounded smallest cells grouped and linkedto each other forming aggregates floating in the supernatant

Fractal analysis provides a quantitative assessment ofthose qualitative differences [17 29] Adherent cells inmicrogravity showed fractal values significantly higher thansuspended cells coherently roundness values were greaterin suspended than in adherent cells Additionally solidityestimation evidences how different these populations are interms of ldquopotentialrdquo deformability Solidity is a good descrip-tor of cell deformability indeed as it describes in geometricalterms the stiffness and deformability of an object Thus the


0020406080

100120140160

Apop

totic

cells

()


24h 72h

lowastlowastlowast

(a)

0

05

1

15

2

25

3

Relat

ive e

xpre

ssio

n

Bcl-2Bax


24h 72h


cells) clumps)


cells) clumps)

24h 72h

lowastlowast

(b)

cl-PARPGAPDH

05

0

1

15

2

25

3

Relat

ive e

xpre

ssio

n


24h 72h


cells) clumps)


cells) clumps)

24h 72h

lowastlowast

lowastlowast

(c)

Figure 5 Apoptosis analysis in MDA-MB-231 Apoptotic rate in RPM cultured MDA-MB-231 and on ground cells was determined by a dualparameter flow cytometric assay (a) Histograms show the percentage of apoptotic cells (Annexin V+7-AAD-) each column represents themean value plusmn SD of three independent experiments Immunoblot bar chart showing the expression of BaxBcl-2 ratio (b) and cleaved PARP(c) in on ground control cells RPM adherent cells and RPM cell clumps at 24 and 72 hours Columns and bars represent densitometricquantification of optical density (OD) of specific protein signal normalized with the OD values of the GAPDH served as loading controlEach column represents the mean value plusmn SD of three independent experiments lowast119875 lt 0 05 lowastlowast119875 lt 0 01 versus on ground control 119875 lt 0 05119875 lt 0 01 versus RPM adherent cells by ANOVA followed by Bonferroni post-test

higher the solidity is the lower the cell deformability isNonadherent cells growing in microgravity are grouped indiscrete clusters and they establish tight cell-to-cell contactsAs expected their solidity value is higher than that recordedin isolated adherent cells growing in RPM given thatthe multiple cell-to-cell adhesion is thought to ldquostabilizerdquothe cells shape by mutually reinforcing their stiffness Thecombined estimation of these parameters suggests that thetwo emerging populations significantly exhibit differences intheir respective morphological features

Aggregates of floating cells retain their viability potentialand after reseeding into a normal gravitational field they

are able to fully recover their native morphological traitsalready after 24 hours This is not really an unexpectedevent since it has been previously reported that microgravityexposed cellsmay recover their original profilewhen replacedin a normal gravity environment [30] Thereby gravity-related phenotypic variability may be considered an adaptivereversible phenomenon

Changes in cell shape are likely mediated by associatedmodification in cytoskeleton architecture which also conveysmechanical stress to the cell biochemicalgenetic machin-ery Therefore different cytoskeleton arrangements will endup in activating different gene sequences leading hence


00

05

10

15

20

25

30

35

Relat

ive e

xpre

ssio

n

AKTp-AKT


cells) clumps)

24hOn RPM RPM

ground (adherent (cellcells) clumps)

72h

24h 72h



(a)

0

05

1

15

2

25

Rela

tive e

xpre

ssio

n

SurvivinGAPDH


cells) clumps)

24hOn RPM RPM


72h

24h 72h


lowastlowast

lowastlowast

lowast

lowast

(b)

005

115

225

335

4

Relat

ive e

xpre

ssio

n

ERKp-ERK


cells) clumps)

24hOn RPM RPM


72h

24h 72h


lowast

lowast

(c)

Figure 6 Survival pathways analysis in MDA-MB-231 Immunoblot bar chart showing the expression of p-AKTAKT ratio (a) Survivin(b) and p-ERKERK ratio (c) in on ground control cells RPM adherent cells and RPM cell clumps at 24 and 72 hours Columns and barsrepresent densitometric quantification of optical density (OD) of specific protein signal normalized with the OD values of the GAPDH servedas loading control Each column represents the mean value plusmn SD of three independent experiments lowast119875 lt 0 05 lowastlowast119875 lt 0 01 versus on groundcontrol 119875 lt 0 05 119875 lt 0 001 versus RPM adherent cells by ANOVA followed by Bonferroni post-test

to triggering different biochemical pathways The balancebetween tensional forces and the cytoskeleton architecturemodulates thereupon several complex cell functions likeapoptosis differentiation proliferation ECM remodellingand so forth [31] Compelling data demonstrated that bothsimulated and real space-based microgravity can severelyaffect cytoskeleton structure and function [8 32] The mostimpressive modifications were observed in nonadherentRPM-exposed cells in which stress fibers disappear and actinarchitecture is severely compromised thus jeopardizing thechances of cell anchoring to the substrate In the same cells

tubulin microfilaments are almost completely disorganizedThis finding may help in explaining the cell cycle inhibitionobserved in floating cell clumps given that a correct arrange-ment of the tubulin meshwork is required to ensure a properfunctioning of the mitotic process microtubules performindeed a special task during mitosis and meiosis by formingthe spindle assembly to align and separate the chromosomes[33] Yet it is worth of noting that cytoskeleton changes greatlydiffer between the two RPM-cultured populations outliningtherefore that microgravity enacted the emergence of twovery different cytoskeleton phenotypes


That architectural diversity is associated with significantdifferences in cell cycle and apoptosis Adherent breast cancercells growing in RPM are trying to counteract microgravityeffects by increasing the number of cells in the S phase and bystabilizing the apoptotic rate On the contrary suspended cellaggregates display a very different behavior characterized byreduced proliferative capability and enhanced apoptosis

However most of the cells in the floating clumps resultedto be viable in fact these cells readhered and grew up whenonce they were reseeded in normal gravity environmentHence cell population blocked in G2M underwent apopto-sis meanwhile cell population blocked in G0G1 recoveredthe original profile when they were reseeded

It is worth noting that such results have been obtainedby treating highly malignant growing cancer cells In thisregard both cell phenotypes cultured in RPM greatly differfrom their counterpart growing in 1 g gravity field Suchprocesses are remarkably mirrored by the concomitantcoherent changes in several biochemical pathways mech-anistically linked to both proliferation and programmedcell death Cyclin D1 a key regulatory factor for cellcycle G1S transition is significantly increased in adherentMDA-MB-231 cells meanwhile in suspended cell aggregatesCyclin D1 release is almost completely abolished Likewiseproapoptotic effectors (BAX PARP) dramatically increasein suspended RPM-cultured cells meanwhile prosurvivalfactors (Bcl-2 Survivin) significantly decrease Survivin awell-known critical factor triggering a plethora of survivalsignaling cascades was indeed dramatically downregulatedand resulted to be undetectable after 72 hours of exposi-tion Opposite findings were observed in adherent breastcancer cells exposed to microgravity the Bcl-2 inhibitor ofcaspase activation increases whereas proapoptotic effectorsconcomitantly decline

Regulation of apoptotic processes relies on the mod-ulation of an intricate interplay between several upstreammolecular pathways involving mainly activation of p-ERKand p-AKT expression As expected p-AKT and p-ERK weresignificantly reduced in suspended cell aggregates mean-while they increase in adherent apoptosis-resistant cellsOverall these results become evident already at early timesthat is after 24 hours of exposition

5 Conclusions

Our results confirm previous findings demonstrating thatmicrogravity enacted the emergence of distinct pheno-types characterized by significant recognizable differences inshape configuration biochemical pathways architecture andbehavioral processes [17] Additionally it should be remarkedthat the coexistence of two different cell populations maycontribute to explain some contradictory results providedby earlier studies [34 35] indeed increase or reduction incell proliferation as well as enhanced or reduced apoptosiscould well be both found during microgravity experimentsgiven that such opposite behaviors must be ascribed to verydifferent cell clusters

Spontaneous emergence of different phenotypes inmicrogravity after the system has experienced a symmetry

breaking is a finding worth of interest and may have relevantconsequences for human space flights Phenotypic switchleading to divergent morphological and biochemical config-uration is triggered by nonlinear processes taking place nearthe transition point Such transition enables the system torecover new degree of freedom and as such it may be viewedas a spontaneous process allowed by the nonequilibriumthermodynamics That finding highlights the relevance ofbiophysical constraints in shaping the form biological dis-sipative systems acquire [36] and may help understand howcells and tissues behave during development pathologicalevents or in extreme environmental fields



Acknowledgment

Thisworkwas partially supported by a grant fromASI (ItalianSpace Agency) COREA Program

References

[1] J C Buckey Space Physiology Oxford University Press 2006[2] G Albrecht-Buehler ldquoPossible mechanisms of indirect gravity

sensing by cellsrdquo ASGSB Bulletin vol 4 no 2 pp 25ndash34 1991[3] MHughes-Fulford andM L Lewis ldquoEffects ofmicrogravity on

osteoblast growth activationrdquo Experimental Cell Research vol224 no 1 pp 103ndash109 1996

[4] G Carmeliet and R Bouillon ldquoThe effect of microgravity onmorphology and gene expression of osteoblasts in vitrordquo TheFASEB Journal vol 13 no 8 pp S129ndashS134 1999


[6] D A Riley J L W Bain J L Thompson et al ldquoThin filamentdiversity and physiological properties of fast and slow fibertypes in astronaut leg musclesrdquo Journal of Applied Physiologyvol 92 no 2 pp 817ndash825 2002

[7] M L Lewis L A Cubano B Zhao et al ldquocDNA microarrayreveals altered cytoskeletal gene expression in space-flownleukemic T lymphocytes (Jurkat)rdquo The FASEB journal vol 15no 10 pp 1783ndash1785 2001

[8] D Vorselen W H Roos F C MacKintosh G J L Wuite andJ J W A van Loon ldquoThe role of the cytoskeleton in sensingchanges in gravity by nonspecialized cellsrdquoThe FASEB Journalvol 28 no 2 pp 536ndash547 2014

[9] A Cogoli ldquoSignal transduction in T lymphocytes inmicrograv-ityrdquo Gravitational and Space Biology Bulletin vol 10 no 2 pp5ndash16 1997

[10] J P Hatton F Gaubert M L Lewis et al ldquoThe kinetics oftranslocation and cellular quantity of protein kinaseC in humanleukocytes aremodified during spaceflightrdquoTheFASEB Journalvol 13 no 8 pp S23ndashS33 1999

[11] T GHammond E Benes K C OrsquoReilly et al ldquoMechanical cul-ture conditions effect gene expression gravity-induced changes


on the space shuttlerdquo Physiological Genomics vol 2000 no 3pp 163ndash173 2000

[12] J Renn D Seibt R Goerlich M Schartl and C WinklerldquoSimulated microgravity upregulates gene expression of theskeletal regulator Core binding Factor 1205721Runx2 inMedaka fishlarvae in vivordquo Advances in Space Research vol 38 no 6 pp1025ndash1031 2006

[13] J Boonstra ldquoGrowth factor-induced signal transduction inadherent mammalian cells is sensitive to gravityrdquo The FASEBJournal vol 13 no 8 pp S35ndashS42 1999

[14] M Bizzarri A Cucina A Palombo and M G MasielloldquoGravity sensing by cells mechanisms and theoretical groundsrdquoRendiconti Lincei Scienze Fisiche e Naturali vol 25 no 1 ppS29ndashS38 2014

[15] M Braun ldquoPrimary responses of gravity sensing in plantsrdquoin Biology in Space and Life on Earth Effects of Spaceflighton Biological Systems E Brinckmann Ed chapter 2 WileyWeinheim Germany 2007

[16] D K Kondepudi and P B Storm ldquoGravity detection throughbifurcationrdquo Advances in Space Research vol 12 no 1 pp 7ndash141992

[17] F Testa A Palombo and S Dinicola ldquoFractal analysis ofshape changes in murine osteoblasts cultured under simulatedmicrogravityrdquo Rendiconti Lincei Scienze Fisiche e Naturali vol25 no 1 pp S39ndashS47 2014

[18] J Vassy S Portet M Beil et al ldquoWeightlessness acts on humanbreast cancer cell line MCF-7rdquo Advances in Space Research vol32 no 8 pp 1595ndash1603 2003

[19] AQianW Zhang L Xie et al ldquoSimulatedweightlessness altersbiological characteristics of human breast cancer cell lineMCF-7rdquo Acta Astronautica vol 63 no 7ndash10 pp 947ndash958 2008

[20] J M Jessup M Frantz E Sonmez-Alpan et al ldquoMicrogravityculture reduces apoptosis and increases the differentiation ofa human colorectal carcinoma cell linerdquo In Vitro Cellular ampDevelopmental BiologymdashAnimal vol 36 no 6 pp 367ndash3732000

[21] E N Grigoryan H J Anton and V I Mitashov ldquoReal andsimulated microgravity can activate signals stimulating cellsto enter the S phase during lens regeneration in urodeleanamphibiansrdquoAdvances in Space Research vol 38 no 6 pp 1071ndash1078 2006

[22] K Nakamura H Kuga T Morisaki et al ldquoSimulated micro-gravity culture system for a 3-D carcinoma tissue modelrdquoBioTechniques vol 33 no 5 pp 1068ndash1076 2002


[24] A H Boulares A G Yakovlev V Ivanova et al ldquoRole ofpoly(ADP-ribose) polymerase (PARP) cleavage in apoptosisCaspase 3-resistant PARPmutant increases rates of apoptosis intransfected cellsrdquo The Journal of Biological Chemistry vol 274no 33 pp 22932ndash22940 1999

[25] G Nicolis and I Prigogine Exploring Complexity FreemanNew York NY USA 1989

[26] G Nicolis and I Prigogine ldquoSymmetry breaking and patternselection in far-from-equilibrium systemsrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 78 no 2 part 1 pp 659ndash663 1981

[27] J Tabony N Glade J Demongeot and C Papaseit ldquoBiologicalself-organization by way of microtubule reaction-diffusionprocessesrdquo Langmuir vol 18 no 19 pp 7196ndash7207 2002

[28] S J Moorman and A Z Shorr ldquoThe primary cilium as agravitational force transducer and a regulator of transcriptionalnoiserdquo Developmental Dynamics vol 237 no 8 pp 1955ndash19592008

[29] A R Qian D Li J Han et al ldquoFractal dimension as a measureof altered actin cytoskeleton inMC3T3-E1 cells under simulatedmicrogravity using 3-D2-D clinostatsrdquo IEEE Transactions onBiomedical Engineering vol 59 no 5 pp 1374ndash1380 2012

[30] R Coinu A Chiaviello G Galleri F Franconi E Crescenziand G Palumbo ldquoExposure to modeled microgravity inducesmetabolic idleness in malignant human MCF-7 and normalmurine VSMC cellsrdquo FEBS Letters vol 580 no 10 pp 2465ndash2470 2006

[31] M E Chicurel C S Chen and D E Ingber ldquoCellular controllies in the balance of forcesrdquo Current Opinion in Cell Biologyvol 10 no 2 pp 232ndash239 1998


[33] J L Carminati and T Stearns ldquoMicrotubules orient the mitoticspindle in yeast through dyne independent interactionswith thecell cortexrdquoThe Journal of Cell Biology vol 138 no 3 pp 629ndash641 1997

[34] L Vico ldquoWhat do we know about alteration in the osteoblastphenotype with microgravityrdquo The Journal of MusculoskeletalNeuronal Interactions vol 6 no 4 pp 317ndash318 2006

[35] M Hughes-Fulford ldquoPhysiological effects of microgravity onosteoblast morphology and cell biologyrdquo Advances in SpaceBiology and Medicine vol 8 pp 129ndash157 2002


Research ArticleOxidative Stress and NO Signalling in the Root Apex asan Early Response to Changes in Gravity Conditions

Sergio Mugnai12 Camilla Pandolfi1 Elisa Masi1 Elisa Azzarello1 Emanuela Monetti1

Diego Comparini1 Boris Voigt3 Dieter Volkmann3 and Stefano Mancuso1

1 DISPAA University of Florence Viale delle Idee 30 50019 Sesto Fiorentino Italy2HSO-USB ESTEC European Space Agency Keplerlaan 1 2200 AG Noordwijk The Netherlands3 IZMB University of Bonn Kirschallee 1 53115 Bonn Germany

Correspondence should be addressed to Sergio Mugnai sergiomugnaiunifiit



Copyright copy 2014 Sergio Mugnai et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Oxygen influx showed an asymmetry in the transition zone of the root apex when roots were placed horizontally on ground Theinflux increased only in the upper side while no changes were detected in the division and in the elongation zone Nitric oxide(NO) was also monitored after gravistimulation revealing a sudden burst only in the transition zone In order to confirm theseresults in real microgravity conditions experiments have been set up by using parabolic flights and drop tower The production ofreactive oxygen species (ROS) was also monitored Oxygen NO and ROS were continuously monitored during normal and hyper-and microgravity conditions in roots of maize seedlings A distinct signal in oxygen and NO fluxes was clearly detected only inthe apex zone during microgravity with no significant changes in normal and in hypergravity conditions The same results wereobtained by ROS measurement The detrimental effect of Drsquoorenone disrupting the polarised auxin transport on the onset of theoxygen peaks during the microgravity period was also evaluated Results indicates an active role of NO and ROS as messengersduring the gravitropic response with probable implications in the auxin redistribution

1 Introduction

During evolution plants have developed elaborate sensoryand signaling systems to cope with and adjust to rapidenvironmental changes Among them gravity remains aconstant stimulus playing a central role in driving theevolution of plants on Earth [1] Gravitropism involves a fineand reliable coordination of the activity of different cells andtissues deputed to gravity sensing with a growth responseoccurring in spatially distinct regions In roots for examplethe centrally located columella cells in the root cap are theproposed site of gravity sensing but the growth response(root curvature) occurs in the elongation zone (EZ)Themostcommon and accepted explanation for gravity sensing inplants is the starch-statolith hypothesis (which is the physicalsedimentation of starch-filled organelles called amyloplasts(statoliths) in gravity-sensing cells (statocytes) located at theroot tip) which triggers biochemical and physiological signals

[2] After the first event (sensing the change in the gravityvectorlevel) a signal is transduced and then transmitted tothe EZ stimulating the differential cellular growthmentionedabove This response is mainly driven by auxin [3] whichaccumulates to higher levels along the lower side of the rootthus provoking the inhibition of growth (Cholodny-Wenttheory) Recently other investigators highlighted the role ofcytoskeleton in the gravitropic response [4 5] Its central roleinmodulating cell polarity organellemovement intracellulartransport and cell expansion leads cytoskeleton to be astrong candidate in mediating the gravity signal transductioncascade (tensegrity model [6])

In the last years experimental evidences depicted thestructure of the root apex as divided into three different zonesa transition zone (TZ) located between two other regionsthe apical division zone (DZ) and the elongation zone (EZ)[7] The cells belonging to TZ have a specific cytoarchi-tecture with centralized postmitotic nuclei surrounded by



perinuclear microtubules radiating toward the cell peripheryIn contrast to the mitotically active DZ cells which arecontinually assembling and disassembling mitotic spindlesthe TZ cells are not deputed to perform these activities but tohave more specific sensory activities [3] Experimental datasuggest that the TZ is more a sort of sensory and informationprocessing zone devoted to a continuous monitoring ofthe environmental parameters and triggering appropriateresponses rather than being implicated in the division andgrowth processTheTZ cells are very sensitive to a wide rangeof stress sources such as touch [8] water and salt stress [9]aluminium [10] and hypoxia [11 12] However little is knownabout the role of TZ cells in the graviresponse especiallyrelated to the transductiontransmission phases after gravitysensing by the cells of the root tip

Among key signalling molecules in plants nitric oxide(NO) has recently emerged as an essential compound [13]Among its tasks NO regulates the actin cytoskeleton [14]endocytosis vesicle trafficking and the polarity of growingtip cells [15] root formation [16] and stomatal regulation [17]In addition NO is widely implicated in the plant response toenvironmental stress [18] but its exact role in the response ofplants to change in gravity levels is still unclear and not wellinvestigated

Reactive oxygen species (ROS) such asH2O2are generally

considered to be toxic by-products of respiration Howeverrecent experiments suggest that the production of ROSshould have an important and active role as components ofintracellular and extracellular signalling [19] In particularH2O2is starting to be accepted as a second messenger for

signals generated by means of ROS because of its relativelylong life and high permeability across membranes [20] Therole of ROS in the gravitropic response is still under debateas this topic has been rarely investigated [21]

Elucidating the mechanisms behind the signal transmis-sion from the site of gravity sensing to the site of gravibendingis therefore the main objective of this paper with particularinterest towards the role of the root apex (and in particular theTZ) in the transduction process and the importance of sens-ing molecules such as NO and H

2O2 in the gravity response

process After preliminary experiments on ground using themethod of gravistimulation via a horizontal displacement ofthe root the response of maize seedlings to gravity changeshas been studied for the first time ever in a real situationof microgravity thanks to a set of experiments performedduring three ESAparabolic flight campaigns and a drop towercampaign


21 Gravistimulation on Ground Oxygen fluxmeasurementswere performed using the vibrating probe technique [22]Briefly healthy Z mays L root apices (5-6mm long) werecut carefully washed with deionized water and placed indi-vidually at the bottom of a measuring chamber containingan electrophysiological solution (10mM CaCl

2 pH 65)

The flux measurement was performed at 24 plusmn 025∘C bypositioning a custom-built oxygen-selective microelectrode

(tip diameter of 1120583m[23 24]) near the root surface To ensurethe flux detection on the bottompart of a gravistimulated root(gravistimulation was performed by rotating the seedlingand the measurement system by 90 degrees) an electrodewith a hooked tip was also built During the recording themicroelectrode oscillated in a square wave parallel to theelectrode axis over a distance of 10 120583m (01Hz frequency)moving along the entire root length The calculation of thedifference between the voltage of each electrode position andthat of the previous one at the other extreme position as wellas the evaluation of amoving average of these differences overany desired time period producing the potential differencewere computer generated The O

2influxes were calculated

using Fickrsquos first law of diffusion assuming a cylindricaldiffusion geometry The flux measurements were performedon at least 10 different root apices per treatment (119899 ge 10)

To localize the production of NO in the different regionsof the root apex with a spatial resolution of a few microm-eters a NO-selective microelectrode of carbon fibers withdiameters as small as 5 120583m was constructed [12] using thesame system described above for the measurement Thedimensions and the response time (lt05 s) allowed the useof this electrode in a self-referencing mode [23 25] with aresolution as small as 50 fmol cmminus2 sminus1

22 Parabolic Flight Experiments All parabolic flight exper-iments were conducted aboard the Airbus A300 ZERO-Gwhich is operated by Novespace and is based in BordeauxFrance Every parabolic flight which lasts sim3 h includ-ing takeoff and landing encompasses 31 parabolas Everyparabola started from a steady normal horizontal flightand typically included 2 hypergravity (18 g) periods of 20 sseparated on average by a 22-s microgravity period (lt005 g)The first test parabola was followed by 6 series of 5 parabolasseparated by breaks of 4 and 8min respectively The datapresented emerged from the 41st and 45th Parabolic FlightCampaigns (PFC) and the ldquoFly Your Thesis 2012rdquo (studentcampaign) of the European Space Agency (ESA) represent-ing a total of 8 parabolic flights or 248 parabolas

During the 41st ESA Parabolic Flight Campaign (PFC)the measurements of oxygen influxefflux from the seedlingroots were conducted For each parabolic flight of the PFCa set of three 3-day-old seedlings of Zea mays L with ahomogeneous length of 5 plusmn 05 cm have been installed intoan Eppendorf vial (1 seedling = 1 vial) filled with an electro-physiological solution (10mMCaCl

2 pH 65)The fourth vial

was left empty and used as a control without seedlings Acouple ofO

2needlemicrosensors (OX50UnisenseDK) have

been horizontally inserted at two different levels for each vialcorresponding to the root apex and to the mature zone of theseedling root The tip of the electrode was placed close to theroot tissues (distance lt 1mm) The vials with the seedlingsand the electrodes were placed inside a thermostated cham-ber (temperature = 24-25∘C) Each electrode was connectedto a picoammeter (PA2000 Unisense DK) a four-channellaboratory amplifier that enables the measurement of mul-tiple parameters The output of the picoammeter was thenconnected to a datalogger A dedicated LabView software on


a laptop recorded the oxygenmeasurement Concurrently anaccelerometer provided gravity measurement

During the 45th ESA PFC the respiration rate of detachedroot apices (119899 = 6 with known weight) from 3-day-oldZea mays L seedlings was measured The root apices wereplaced inside an oximeter chamber (Oxytherm HansatechInstruments) with controlled temperature (25∘C) A smallmagnetic stirrer provided a continuous stirring of the solu-tion Measurements have been performed with the apicesin distilled water or in a solution containing Drsquoorenone(C18

ketone (5E7E)-6-methyl-8-(266-trimethylcyclohex-1-enyl) octa-57-dien-2-one 10 120583gmL) The oximeter chamberwas connected to a laptop with a dedicated software fordatalogging

During the ldquoFly your thesis 2012rdquo PFC campaign theproduction of H

2O2was assessed The Amplex Red reagent

was used after a preliminary evaluation in our lab due to itshigh sensitivity and successful use for the measure of H

2O2

production in plant root as previously reported by [26] Weused the Amplex Red Hydrogen PeroxidePeroxidase AssayKit (Invitrogen A22188) in combination with horseradishperoxidase (HRP Invitrogen) to detect H

2O2released from

biological samples In the presence of HRP Amplex redreagent reacts with H

2O2in a 1 1 stoichiometry to produce

the red-fluorescent oxidation product resorufin As first stepthe tips of maize roots were cut and immediately washedtwice for 15 minutes in PBS to eliminate ROS derived fromthe cut Tips were divided in samples constituted by 10mgof fresh tissue Then 50 120583L of working solution (containingdye and HRP) was added to 10mg of plant tissue (root)and the samples were incubated in a 96-well microplate atroom temperature for 30 minutes in darkness During theflight fluorescence was measured using excitation at 530 plusmn125 nm and fluorescence detection at 590plusmn175 nm by usinga microplate reader (Tecan Infinite 200 PRO) A dataloggerconnected to the Tecan and a laptop provided data storagewhich were then normalised to plant biomass

23 Drop Tower Campaign The Drop Tower in Bremen(Germany) is one of only a few facilities worldwide providinggravitational forces as small as 10minus5 g even if only for ashort time of 47 seconds The cylindrical falling capsule ofa diameter of 80 cm a height of 28m and a mass of 500 kg isdropped from 110m height of the tower whose inner tube isevacuated within 2 h to an air pressure of less than 10 mPaOn the bottom of the tower the capsule (reaching 170 kmhminus1) is decelerated within 130ms by a huge basket of app 25times 8m filled with styropor grains There the motion energy(6 times 105Nm) is converted into heat Gravitational sensorswere provided by ZARM Deceleration of the capsule leads togravitational values of about 30 g An oximeter (see ParabolicFlights section) was used and adapted for the measurementof nitric oxide by using selective microelectrodes (amiNO-30 Innovative Instruments Inc USA) connected to a NOelectrochemical detector with automatic temperature com-pensation (in NO-T-II Innovative Instruments Inc USA)The output of the detector was connected to a laptop via

200

175

150

125

100

75

50

25

0

175

150

125

100

75

50

25

0

Oxy

gen

influ

x (p

mol

cmminus

2sminus

1 )

00 05 10 15 20 25 30 35 40 45 50

Distance from root apex (mm)

DZ TZ EZ

Upper side

Bottom side

GravistimulatedNongravistimulated

Figure 1 Oxygen influx measured on the two sides of a gravistim-ulated and a vertical root Upper graph refers to the upper side of agravistimulated root whereas the second graph refers to the bottomside of a gravistimulated root

200

175

150

125

100

75

Oxy

gen

influ

x (p

mol

cmminus

2sminus

1 )

minus60 0 60 120 180 240 300

Time (s)

Upper side

Bottom side

Gra

visti

mul

atio

n

Figure 2 Timeline of the oxygen influx change at TZ level aftergravistimulation (time = 0)


14

12

10

8

6

4

2

14

12

10

8

6

4

2

14

12

10

8

6

4

2

0

1 3 5 7 9 11 13 15 17 19

Time (s)

Division zone

Transition zone

Elongation zone

NO

(nM

)N

O (n

M)

NO

(nM

)

minus3 minus1

Figure 3 Timeline of the NO production in a maize root apex inthe three different constituent zones after gravistimulation (time =0)

USB port running the inNO-T-II specific software for dataacquisition

3 Results

31 Gravistimulation on Ground In normal (vertical) con-ditions strong differences between the constituent zones ofthe maize root (DZ TZ and EZ) were clearly evident TheTZ appeared to be the most active zone in the uptake of

Table 1 Average time of burst appearance calculated for differentgroups of parabolas Data were analyzed by ANOVA using Tukeyrsquostest (119875 lt 005)

Group ofparabolas

Average time(secs)

Standarddeviation Significance

1ndash10 251 108 ns11ndash20 265 122 ns21ndash30 207 099 ns1ndash30 24 11 mdash

oxygen from the surrounding solution (Figure 1) The spatialpatterns of the oxygen influxes in the entire root apex showedamarked peak in theTZ (110 pmol cmminus2 sminus1) at 1ndash15mm fromthemaize root tip Aminor oxygen influx peak (75 pmol cmminus2sminus1) was also evident in the DZ Importantly the TZ was theonly root apex region significantly affected with regards togravistimulation in fact the marked peak of oxygen influxwas greatly enhanced in the upper part of the horizontal(gravistimulated) root whereas the DZ maintained a similarpattern On the contrary the bottom side of the horizontalroot showed a normal behaviour The increase of oxygeninflux at TZ level appeared as a very quick response followinggravistimulation as it was clearly evident after less than 30seconds in the upper side of the root (Figure 2) while thebottom part remained unaffected

Gravistimulation also promoted a very fast NO produc-tion from the root apex (Figure 3) A burst of NO was sud-denly produced after only 2-3 seconds from gravistimulationreaching a peak of 10 nM and lasting approximately 8 s beforereturning to the steady-state values in the TZ Only a smalland negligible efflux of NO was detectable in the DZ andimportantly NO bursts were not detected in the EZ region

32 Parabolic Flight and Drop Tower Campaigns A typicalparabola is shown in Figure 4 with a period of microgravity(lt005 g 20ndash22 secs) inserted between two periods of hyper-gravity (18 g 20 secs) Oxygen concentrations in the solutionmeasured at two different root levels (root apex and maturezone) are also reported for three different parabolas Burstsand peaks of oxygen concentration are clearly evident at rootapex level during the microgravity periods with significantdetection during the entire flight Interestingly no peakswere detected in the mature zone and during hypergravityperiods in the root apex Control without seedlings showedno activity demonstrating that the results previously shownwere not related to a background noise of themicroelectrodesduring microgravity

For each parabola the average detection time (1198791) of the

first oxygen peaks from the start of the microgravity period(119892 lt 1 119879

0 see Figure 5) has been calculated The average

time for all the parabolas is 24 secs In order to evaluatea difference in the appearance of the first peak during theflight the parabolas have been separated in three differentgroups (Table 1) Each group was composed of 10 parabolasThe objective was to investigate if a sort of ldquomemory effectrdquoof the stimulus during the repeated parabolas could cause a


400

375

350

325

300

275

250

225

200

Oxy

gen

conc

entr

atio

n (p

A)

0 10 20 30 40 50 60 70 80 90 100

Gra

vity

0

1Apex

Mature zone

Time (s)

(a)

Apex

Mature zone

400

375

350

325

300

275

250

225

200

Oxy

gen

conc

entr

atio

n (p

A)

0 10 20 30 40 50 60 70 80 90 100

Gra

vity

0

1

2

Time (s)

(b)

Apex

Mature zone

400

375

350

325

300

275

250

225

200

Oxy

gen

conc

entr

atio

n (p

A)

0 10 20 30 40 50 60 70 80 90

Time (s)

Gra

vity

0

1

2

Gravity

(c)

Apex-no root

Mature zone-no root

400

350

300

250

200

150

100

Oxy

gen

conc

entr

atio

n (p

A)

0 10 20 30 40 50 60 70 80 90 100

Gra

vity

0

1

2

Gravity

Time (s)

(d)

Figure 4 Oxygen concentration in the solution measured at two different levels close to (distance lt 1mm) the root apex and to the rootmature zone during three different parabolas one parabola at the beginning (a) one in the middle (b) and one at the end of the experiment(c) Control experiment with no roots inside the Eppendorf vial is shown in (d)

180

1

005

011

010

009

008

Oxy

gen

conc

entr

atio

n (p

A)

50 100

Time (s)

T0 T1

Gra

vity

Figure 5 Timeline of burst appearance (1198791) after the onset of

microgravity (1198790) during a parabola

different response during the time of the flight (higherloweranticipatedretarded)The results showed a reduction but notstatistically significant in the onset of the first peak during thelast set of parabolas

Interesting results have been also obtained from themeasurement of the respiration rate by oximeters The res-piration rate inside a single parabola has been divided intofive segments each segment being related to a differentgravity level 1 g 2 g 0 g 2 g (after microgravity) and 1 g(after microgravity) Negative values indicate oxygen influxRespiration by root apices led to an unavoidable reductionof the oxygen content in the solution due to the plantmetabolism thus the parabolas have been divided intodifferent groups by taking into account the real oxygenconcentration because the respiration rate is directly relatedto the amount of oxygen present in the solution Fourgroups related to different [O

2] in the solution have been

therefore created gt1500 nM 1000ndash1500 nM 500ndash1000 nMand lt500 nM The values relative to the control (Figure 6)show no significant differences among the different gravitylevels in each parabola for every oxygen concentration groupexcept for the last group ([O

2] lt 500 nM) with an increased

respiration rate during the second period of hypergravity Onthe contrary the presence ofDrsquoorenone in the solution did notlead to any variation neither in the respiration rate among thedifferent gravity levels nor compared to the control (Figure 7)


Control

ns ns

ns

2

0

minus2

minus4

minus6

1G 2G 0G 2G after 1G after 1G 2G 0G 2G after 1G after

1G 2G 0G 2G after 1G after1G 2G 0G 2G after 1G after

0

minus1

minus2

minus3

minus4

aa a

ab

b

minus03

minus06

minus09

minus12

minus15

minus18

minus21

minus24

025

000

minus025

minus050

minus075

minus100

O2

(nM

sminus1 )

O2

(nM

sminus1 )

O2

(nM

sminus1 )

O2

(nM

sminus1 )

gt1500nM O2 1500ndash1000nM O2

500ndash0nM O21000ndash500nM O2

Figure 6 Respiration rate in control roots for the different groups related to the oxygen concentration in the solution

Drsquoorenone has been utilized in this experiment because itincreases PIN2 protein abundance without affecting PIN2transcripts with the consequence that the PIN2 expressiondomain enlarges and shifts basipetally resulting in moreactive auxin transport To deeply analyse the previous resultsthe behaviour of the respiration rate during a single parabolahas been evaluated It has been noted that when [O

2] was

lt 700 nM a sudden burst of oxygen was produced only inthe control a few seconds after the onset of microgravity(Figure 8)This large amount of oxygenwas quickly absorbedby the roots for respiration thus explaining the increasedrespiration rate during the second hypergravity period Thisphenomenon was clearly evident during each parabola with[O2] lt 700 nM The fact that the bursts were evident only

when [O2] lt 700 nM was probably due to the electrode

sensitivity which was not able to discriminate very lowdifferences in the respiration rate (around 25 nM)with higheroxygen concentrations in the solution These oxygen burstshave been characterised by calculating the area inside thecurve (Figure 9) The values of area response time after theonset of microgravity peak duration and peak amplitudeare reported in Figure 10 with a discrimination based onthe parabolasrsquo groups No significant differences among thegroups were noted in the peak area with an average value

of 27473 nM representing the moles of oxygen producedduring the microgravity period and then consumed in theresponse time after the onset of microgravity (average valueof 079 seconds) and the timing of the maximum peak (1103seconds) On the contrary significant differences amonggroups were registered in the peak duration The first 20parabolas had an average peak duration of 20-21 secondswhile the last 10 parabolas had a longer duration (averagevalue around 30 seconds) Finally peak amplitude showedno significant differences As expected when no roots werepresent in the oximeter a stable signal was registered (datanot shown)

Production ofH2O2measuredwithAmplex Red is shown

in Figure 11 Data were grouped according to the class ofgravity level Data recorded during microgravity (0 g) werestatistically compared with data recorded during normograv-ity (1 g) and hypergravity (2 g) conditions Transition from 1 gto 2 g and 2 g to 1 g had no significant effect on H

2O2 On the

contrary transition from 2 g to 0 g results stimulated a higherH2O2production from the root samples Interestingly we did

not observe any difference in H2O2production between 2 g

condition in comparison to 1 g Control experiment withoutany root tip inside the microplate showed no significantdifference between all the gravity levels (data not shown)


ns ns

nsns

minus10

minus15

minus20

minus25

minus30

minus35

1G 2G 0G 2G after 1G after

minus050

minus075

minus100

minus125

minus150

minus175

minus200

minus225

minus025

minus050

minus075

minus100

minus125

minus150

minus175

minus200

025

000

minus025

minus050

Drsquoorenone



O2

(nM

sminus1 )

O2

(nM

sminus1 )

O2

(nM

sminus1 )

O2

(nM

sminus1 )

gt1500nM O21500ndash1000nM O2


Figure 7 Respiration rate in roots incubated in Drsquoorenone for the different groups related to the oxygen concentration in the solution

ControlDrsquoorenone

1200

1000

800

600

400

200

0

350 450 550 650 750 850 950

Time (s)

2

1

0

Gra

vity

O2

(nM

)

Figure 8 Respiration rate during single parabolas when [O2] in the

solution was lt700 nM Both control roots and roots incubated inDrsquoorenone are reported

Finally the production of NOwas detected and evaluatedduring an ESA Drop Tower campaign (Figure 12) Interest-ingly a burst of NO was clearly evident after 2 secondsfrom the start of the microgravity period which then started

O2

(nM

)

350

325

300

275

250

Time (s)T0 T1 Tmax Tend

Peak

Baseline T1

18

1

005

Gra

vity

Figure 9 Characterisation of an oxygen burst measured with anoximeter 119879

0is the time of the onset of microgravity 119879

1is the time

when the burst of oxygen begins Its value in the 119884-axis is taken asthe baseline for the calculation of the area amplitude and durationbetween 119879

1and 119879end

to decline resembling the behaviour of the gravistimulatedroots on ground Oximeter chambers without roots showed


O2

(nM

)

O2

(nM

)

500

400

300

200

100

0

0ndash10 11ndash20 21ndash30Parabolas

Peak area Response time

Resp

onse

tim

e (s)

20

18

16

14

12

10

08

06

04

02

00

nsns

ns50

40

30

20

10

0

Peak amplitudePeak duration50

40

30

20

10

0

Peak

dur

atio

n (s

)

a

bb




Figure 10 Peak area response time peak duration and peak amplitude measured for different groups of parabolas by using the methoddescribed in Figure 9

no bursts or signal detected by the microelectrode (data notshown)

4 Discussion

Although both sensing (Cholodny-Went theory and tenseg-rity model) and signal transduction (role of auxin in the rootbending) in the gravitropic response are well and comprehen-sively described in the literature little is known about the greyarea of signal transmission the series of events comprisedbetween sensing and bending Hu et al [27] reported thatgravistimulation induced the asymmetric accumulation ofnitric oxide (NO) on the lower side of the apical region ofgravistimulated (horizontal) soybean seedling roots leadingto a subsequent auxin accumulation in the upper part Ourresults confirmed this hypothesis with a massive productionof NO in a very short time (2-3 seconds) Moreover we alsointegrated these results with the interesting information thatthe NO is mainly produced at TZ level thus confirming therole of TZ as a sensing zone of the root directly and activelyimplicated in the response to gravity changes

Gravistimulation also induced a sudden burst of oxygenin the upper part of TZ level 20ndash30 seconds after gravis-timulation Our results suggest the hypothesis that after the

displacement of statolythes under gravistimulation the chainof events is related to a sudden emission of NO which leadsto an improved plant metabolism which needs more oxygenfor respiration especially at TZ level to produce ATP tobe used as a source of energy Rapid changes in cytosolicCa2+ and pH have been proposed as components of thegravisignaling machinery [28] therefore it is plausible thatthe control of Ca2+ and H+ channels would require moreATP (ie more oxygen consumption) after gravistimulationand during microgravity The fact that oxygen burst at TZlevel following gravistimulation can be inhibited by BFA [3]and that TZ shows significant higher auxin secretion via theendocytic vesicle recycling [29] might correlate the oxygenbursts observed during gravistimulationunder microgravityand the auxin metabolism thus provoking a differentialgrowth response

In plants the simultaneous generation of O2and NO has

a synergistic function in defense responses [30] as well as inplants exposed to abiotic stress [31] NO is also generated atthe same time as ROS such as hydrogen peroxide for exam-ple during abiotic stress [32] Root gravitropism appears to beanother example of a physiological process in which bothNOand ROS play key roles in a simultaneous way [33] as ROSwere recently associated with auxin-mediated gravitropic


5

4

3

2

1

0

H2O

2(120583

mol

mgminus

1 )

1 I 2 I 0 2

(g)1 II

bbbb

a

Figure 11 H2O2production measured with Amplex Red during

different gravity conditionsThe letters I and II refer to the 1 g and 2 gphases prior (I) and after (II) the 0 g condition Data are presentedas average among each parabola of two parabolic flight days (62parabolas in total are considered) Data were analyzed by ANOVAData statistically different are indicated with different letters (119875 lt005) Error bars are also indicated

100

095

090

NO

(nM

)

minus5 0 5

Time (s)

Gravity

Gra

vity

(g)

NO

0

1

Figure 12 Typical nitric oxide curve measured in the oximeterchamber during a drop in the ESA Drop Tower campaign

responses in maize [21] and in the graviresponsive pulvinusof maize [34] In gravistimulated roots ROS accumulatedasymmetrically to the lower cortex within 30min of reori-entation becoming symmetrical upon longer stimulation[21] Interestingly Long et al [35] have shown that auxinasymmetries are detectable only after 2 h of gravistimulationof the pulvinus making the ROS changes reported muchfaster than the generation of gradients in auxin and so incontrast to the conclusions from the gravitropically respond-ing root possibly placing them upstream of the action ofthis hormone Our results support this hypothesis with thegeneration of an oxygen burst after a few seconds after theonset of microgravity which could be directly linked to theproduction of ROS as a stress messenger The fact that therespiration rate in root apices increased during microgravitycould also be related to the necessity of activating defensiveand scavenging mechanisms for ROS molecules In fact

the production of ROS during real microgravity has beenconfirmed during a parabolic flight campaign

Drsquoorenone rapidly and significantly activates theDR5 pro-moter [36] and also targets processes that are related to PIN2degradation [16] causing slower turnover and increasedprotein levels of this auxin efflux transporter thus suggestingthat this apocarotenoid interacts with auxin signaling at theroot apex Our results indicate that Drsquoorenone has also aninhibitor activity on the respiration rate and on the oxygenproduction thus giving indirect clue to a link between thesudden increase of oxygen during microgravity and auxinredistribution via PIN2 activity which is one of the majorresponses to changes in the gravity vectorlevels

5 Conclusions

For the first time ever a systemic and comprehensive seriesof experiments concerning the role of oxygen and stressmessengers (NO and ROS) during a real microgravity envi-ronment has been conductedThe timeline and the cascade ofevents detected during these experiments suggest an activerole of NO and ROS during the transmission step of thegravity response with probable implications in the auxinredistribution



Acknowledgments

The authors would like to thank ESA (European SpaceAgency) Novespace and ZARM for their kind supportduring the parabolic flight campaigns and the drop towercampaigns

References

[1] D Volkmann and F Baluska ldquoGravity one of the driving forcesfor evolutionrdquo Protoplasma vol 229 no 2ndash4 pp 143ndash148 2006

[2] E B Blancaflor ldquoRegulation of plant gravity sensing andsignaling by the actin cytoskeletonrdquoAmerican Journal of Botanyvol 100 no 1 pp 143ndash152 2013

[3] F Baluska and S Mancuso ldquoRoot apex transition zone asoscillatory zonerdquo Frontiers in Plant Science vol 4 article 3542013

[4] H Tatsumi T Furuichi M Nakano et al ldquoMechanosensitivechannels are activated by stress in the actin stress fibres andcould be involved in gravity sensing in plantsrdquo Plant Biologyvol 16 pp 18ndash22 2014

[5] F Baluka and D Volkmann ldquoMechanical aspects of gravity-controlled growth development and morphogenesisrdquo inMechanical Integration of Plant Cells and Plants pp 195ndash223Springer Berlin Germany 2011

[6] Y Chebli and A Geitmann ldquoGravity research on plants use ofsingle-cell experimental modelsrdquo Frontiers in Plant Science vol2 no 56 pp 1ndash10 2011


[7] F Baluska S Mancuso D Volkmann and P W Barlow ldquoRootapex transition zone a signalling-response nexus in the rootrdquoTrends in Plant Science vol 15 no 7 pp 402ndash408 2010

[8] H Ishikawa and M L Evans ldquoInduction of curvature inmaize roots by calcium or by thigmostimulation role of thepostmitotic isodiametric growth zonerdquo Plant Physiology vol100 no 2 pp 762ndash768 1992

[9] E S Ober and R E Sharp ldquoElectrophysiological responses ofmaize roots to low water potentials relationship to growth andABA accumulationrdquo Journal of Experimental Botany vol 54 no383 pp 813ndash824 2003

[10] M Amenos I Corrales C Poschenrieder P Illes F Baluskaand J Barcelo ldquoDifferent effects of aluminum on the actincytoskeleton and brefeldin A-sensitive vesicle recycling in rootapex cells of twomaize varieties differing in root elongation rateand aluminum tolerancerdquo Plant and Cell Physiology vol 50 no3 Article ID pcp013 pp 528ndash540 2009

[11] S Mugnai A M Marras and S Mancuso ldquoEffect of hypoxicacclimation on anoxia tolerance in vitis roots response ofmetabolic activity andK+ Fluxesrdquo Plant and Cell Physiology vol52 no 6 pp 1107ndash1116 2011

[12] S Mugnai E Azzarello F Baluska and S Mancuso ldquoLocal rootapex hypoxia induces no-mediated hypoxic acclimation of theentire rootrdquo Plant andCell Physiology vol 53 no 5 pp 912ndash9202012

[13] M Simontacchi C Garcıa-Mata C G Bartoli G E Santa-Marıa and L Lamattina ldquoNitric oxide as a key component inhormone-regulated processesrdquo Plant Cell Reports vol 32 no 6pp 853ndash866 2013

[14] K A Wilkins J Bancroft M Bosch J Ings N Smirnoff andV E Franklin-Tong ldquoReactive oxygen species and nitric oxidemediate actin reorganization and programmed cell death in theself-incompatibility response of papaverrdquo Plant Physiology vol156 no 1 pp 404ndash416 2011

[15] M C Lombardo and L Lamattina ldquoNitric oxide is essentialfor vesicle formation and trafficking in Arabidopsis root hairgrowthrdquo Journal of Experimental Botany vol 63 no 13 pp4875ndash4885 2012

[16] M Schlicht J Ludwig-Muller C Burbach D Volkmann and FBaluska ldquoIndole-3-butyric acid induces lateral root formationvia peroxisome-derived indole-3-acetic acid and nitric oxiderdquoNew Phytologist vol 200 no 2 pp 473ndash482 2013

[17] A M Distefano D Scuffi C Garcıa-Mata L Lamattina andA M Laxalt ldquoPhospholipase D120575 is involved in nitric oxide-induced stomatal closurerdquo Planta vol 236 no 6 pp 1899ndash19072012

[18] W Qiao C Li and L M Fan ldquoCross-talk between nitric oxideand hydrogen peroxide in plant responses to abiotic stressesrdquoEnvironmental and Experimental Botany vol 100 pp 84ndash932014

[19] S S Gill and N Tuteja ldquoReactive oxygen species and antioxi-dant machinery in abiotic stress tolerance in crop plantsrdquo PlantPhysiology and Biochemistry vol 48 no 12 pp 909ndash930 2010

[20] L J Quan B Zhang W W Shi and H Y Li ldquoHydrogenperoxide in plants a versatile molecule of the reactive oxygenspecies networkrdquo Journal of Integrative Plant Biology vol 50 no1 pp 2ndash18 2008

[21] J H Joo Y S Bae and J S Lee ldquoRole of auxin-induced reactiveoxygen species in root gravitropismrdquo Plant Physiology vol 126no 3 pp 1055ndash1060 2001

[22] C Pandolfi S Pollastri E Azzarello E Masi S Mugnai andS Mancuso ldquoThe vibrating probe technique in root studiesrdquo inMeasuring Roots pp 67ndash81 Springer Berlin Germany 2011

[23] S Mancuso G Papeschi and A M Marras ldquoA polarographicoxygen-selective vibrating-microelectrode system for the spa-tial and temporal characterisation of transmembrane oxygenfluxes in plantsrdquo Planta vol 211 no 3 pp 384ndash389 2000

[24] S Mancuso and A M Marras ldquoDifferent pathways of theoxygen supply in the sapwood of young Olea europaea treesrdquoPlanta vol 216 no 6 pp 1028ndash1033 2003

[25] S Mancuso and M Boselli ldquoCharacterisation of the oxygenfluxes in the division elongation and mature zones of Vitisroots influence of oxygen availabilityrdquo Planta vol 214 no 5pp 767ndash774 2002

[26] R Shin R H Berg and D P Schachtman ldquoReactive oxygenspecies and root hairs in arabidopsis root response to nitrogenphosphorus and potassium deficiencyrdquo Plant and Cell Physiol-ogy vol 46 no 8 pp 1350ndash1357 2005

[27] X Hu S J Neill Z Tang and W Cai ldquoNitric oxide mediatesgravitropic bending in soybean rootsrdquo Plant Physiology vol 137no 2 pp 663ndash670 2005

[28] G B Monshausen N D Miller A S Murphy and S GilroyldquoDynamics of auxin-dependent Ca2+ and pH signaling in rootgrowth revealed by integrating high-resolution imaging withautomated computer vision-based analysisrdquo The Plant Journalvol 65 no 2 pp 309ndash318 2011

[29] S Mancuso A M Marras V Magnus and F Baluska ldquoNonin-vasive and continuous recordings of auxin fluxes in intact rootapex with a carbon nanotube-modified and self-referencingmicroelectroderdquo Analytical Biochemistry vol 341 no 2 pp344ndash351 2005

[30] S Asai K Mase and H Yoshioka ldquoA key enzyme for flavinsynthesis is required for nitric oxide and reactive oxygen speciesproduction in disease resistancerdquo Plant Journal vol 62 no 6pp 911ndash924 2010

[31] G Tanou A Molassiotis and G Diamantidis ldquoHydrogenperoxide- andnitric oxide-induced systemic antioxidant prime-like activity under NaCl-stress and stress-free conditions incitrus plantsrdquo Journal of Plant Physiology vol 166 no 17 pp1904ndash1913 2009

[32] S J Neill R Desikan and J T Hancock ldquoNitric oxide signallingin plantsrdquo New Phytologist vol 159 no 1 pp 11ndash35 2003

[33] S Swanson and S Gilroy ldquoROS in plant developmentrdquo Physi-ologia Plantarum vol 138 no 4 pp 384ndash392 2010

[34] A M Clore S M Doore and S M N Tinnirello ldquoIncreasedlevels of reactive oxygen species and expression of a cytoplasmicaconitaseiron regulatory protein 1 homolog during the earlyresponse of maize pulvini to gravistimulationrdquo Plant Cell andEnvironment vol 31 no 1 pp 144ndash158 2008

[35] J C Long W Zhao A M Rashotte G K Muday and S CHuber ldquoGravity-stimulated changes in auxin and invertase geneexpression inmaize pulvinal cellsrdquo Plant Physiology vol 128 no2 pp 591ndash602 2002

[36] M Schlicht O Samajova D Schachtschabel et al ldquoDorenoneblocks polarized tip growth of root hairs by interfering with thePIN2-mediated auxin transport network in the root apexrdquo ThePlant Journal vol 55 no 4 pp 709ndash717 2008

Research ArticleCytoskeleton Modifications and Autophagy Induction inTCam-2 Seminoma Cells Exposed to Simulated Microgravity

Francesca Ferranti12 Maria Caruso2 Marcella Cammarota3

Maria Grazia Masiello45 Katia Corano Scheri2 Cinzia Fabrizi2 Lorenzo Fumagalli2

Chiara Schiraldi3 Alessandra Cucina56 Angela Catizone2 and Giulia Ricci3

1 Italian Space Agency (ASI) Via del Politecnico snc 00133 Rome Italy2 Department of Anatomy Histology Forensic Medicine and Orthopedics Sapienza University of RomeViale Regina Elena 336 00161 Rome Italy

3 Department of Experimental Medicine Second University of Naples Via Santa Maria di Costantinopoli 16 80138 Naples Italy4Department of Clinical and Molecular Medicine Sapienza University of Rome Viale Regina Elena 291 00161 Rome Italy5 Systems Biology Group Sapienza University of Rome Via A Scarpa 16 00161 Rome Italy6Department of Surgery ldquoPietro Valdonirdquo Sapienza University of Rome Viale del Policlinico 155 00161 Rome Italy

Correspondence should be addressed to Giulia Ricci giuliaricciunina2it

Received 12 May 2014 Revised 4 July 2014 Accepted 4 July 2014 Published 17 July 2014


Copyright copy 2014 Francesca Ferranti et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

The study of how mechanical forces may influence cell behavior via cytoskeleton remodeling is a relevant challenge of nowadaysthat may allow us to define the relationship between mechanics and biochemistry and to address the larger problem of biologicalcomplexity An increasing amount of literature data reported thatmicrogravity condition alters cell architecture as a consequence ofcytoskeleton structuremodifications Herein we are reporting themorphological cytoskeletal and behavioral modifications due tothe exposition of a seminoma cell line (TCam-2) to simulatedmicrogravity Even if no differences in cell proliferation and apoptosiswere observed after 24 hours of exposure to simulatedmicrogravity scanning electronmicroscopy (SEM) analysis revealed that thechange of gravity vector significantly affects TCam-2 cell surface morphological appearance Consistent with this observation wefound that microtubule orientation is altered by microgravity Moreover the confocal analysis of actin microfilaments revealed anincrease in the cell width induced by the low gravitational force Microtubules and microfilaments have been related to autophagymodulation and interestingly we found a significant autophagic induction in TCam-2 cells exposed to simulated microgravityThis observation is of relevant interest because it shows for the first time TCam-2 cell autophagy as a biological response inducedby a mechanical stimulus instead of a biochemical one

1 Introduction

An increasing number of experimental observations havedemonstrated that tissue homeostasis could be stronglyinfluenced and regulated by physical forces such as themodulation of gravity vector In the recent years many effortshave been made to elucidate the effect of microgravity oncell behavior and accumulating data show that micrograv-ity alters permanently or transiently important biological

processes such as mitosis differentiation survival cell mor-phology and gene expression profiles [1ndash7] However howcells sense these signals and convert them into a biochemicalresponse remains an important question that needs to beaddressed Recent studies have focused on the cytoskeletonas initial gravity sensor [1 8] Cytoskeleton plays importantroles in cell physiology being responsible for chromosomalsegregation during mitosis providing a mechanical supportto dividing cells contributing to maintain cell shape and



spatially organizing cell proteins and organelles in cell cyto-plasm Moreover cytoskeleton is involved in cell motilitymembrane trafficking signal transduction and cell adhesionIn addition cytoskeletal proteins can transduce and amplifymembrane receptor-captured signals transmitting the infor-mation to the nucleus and finally regulating gene expression[2 9 10] Considering all these observations it appearseasy to understand why cytoskeleton disorganization couldcompromise a lot of cell functions leading in some cases tocell death It is well known that microgravity exposure couldstrongly influence cytoskeleton organization [10ndash17] and itis commonly accepted that cellular tensegrity alteration inmicrogravity exposed cells could explain at least in part theconversion of a mechanical cue into a biological responseIn this regard recent studies have revealed the importanceof cytoskeletal integrity such as F-actin and microtubulesin the physiological specific aspects of autophagy and somepapers described the capability of microgravity to induceautophagy in living cells [18ndash22] Autophagy is an importanthousekeeping physiological process that is involved in cellu-lar remodeling during development elimination of aberrantorganelles or misfolded proteins and in the recycling ofunnecessary cellular components to compensate for thelimitation of nutrients during starvation It is of interestingnotice that this biological process is highly conserved fromyeast to mammals Despite several studies suggested a tumorsuppressive role for autophagy other reports support thehypothesis that this process is instead exploited by cancercells to prime their proliferation and promote their survival[23ndash27]

Microgravity condition is a stressful change in the physi-cal microenvironment for living cells however they seem tobe able to adapt to this change of gravitational force sincein the major part of studies available in the literature thebehavioral cellular modifications induced by microgravityare transient This observation has led to the intriguinghypothesis that cells in response to gravity changes reacttriggering adaptive biological processes and autophagy couldbe one of them

Testicular cells appear to be sensitive to microgravity ithas been demonstrated in fact that testicular function isimpaired by microgravity exposure [28ndash34] Moreover somein vitro observations revealed that microgravity influencescell proliferation apoptosis and testosterone secretion oftesticular organ cultures [35 36] In addition microgravitycondition has differentiating effect in cultured spermatocytesand influences germ cell survival [37 38] This effect onmale germ cell lineage has triggered the hypothesis that alsotesticular cancer germ cells could be altered by microgravitycondition For this reason we decided to study the effect ofmicrogravity on TCam-2 cells that are the only accreditedseminoma cell line [39ndash42] These cells have been recentlycharacterized at molecular and biochemical level [43ndash51]and thus represent a good tool to investigate male germcell behavior modification in response to a mechanical forcemodification In this paper we report for the first timecytoskeletal modifications and the activation of autophagicprocess induced by acute exposure to microgravity in TCam-2 cell line


21 Random Positioning Machine The random positioningmachine (RPM desktop RPM Dutch Space Leiden theNetherlands) we used in the investigation is a particularkind of 3D clinostat It consists of two independently rotatingframes One frame is positioned inside the other givinga very complex net change of orientation to a biologicalsample mounted in the middle The degree of microgravitysimulation depends on angular speed and on the inclinationof the disk These tools do not actually eliminate the gravitybut it is a microweight simulator based on the principleof ldquogravity-vector averagingrdquo it allows you to apply a 1 gstimulus omnidirectionally rather than unidirectionally andthe sum of the gravitational force vectors tends to zeroEffects generated by the RPM are comparable to those ofthe real microgravity provided that the direction changes arefaster than the response time of the system to gravity fieldThe desktop RPM we used has been positioned within anincubator (for maintaining temperature CO

2 and humidity


22 TCam-2 Cell Cultures The TCam-2 human cell line wasderived in 1993 from a primary testicular tumor sample ofpure classical seminoma [42] TCam-2 cells were culturedin RPMI 1640 (Lonza) supplemented with 10 fetal bovineserum (FBS Lonza) and penicillinstreptomycin (Invitrogen)at 37∘C in a humidified atmosphere with 5 carbon dioxide[41] The time 0 plating cell density is 3 times 104cm2 Asdescribed in the paragraph above microgravity conditionwas simulated using the randompositioningmachine (RPM)Experiments were performed on cells cultured for 24 and 48hours at 1 g or in RPM after additional 24 hours of preplatingon glass slides or IBIDImicroscopy chambers (IBIDI 80826)Glass slides were silicone fixed to the culture dishes at least 48hours before plating Cell culture dishes in both 1 g and RPMculture conditions were completely filled with fresh culturemedium in order to avoid air bubbles and to minimize liquidflow thus making negligible the effects of both buoyancy andshear stress during rotation

23 Proliferation Apoptosis and Autophagy QuantificationCells cultured at 1 g or under microgravity conditions (asdescribed above) were fixed in 4 paraformaldehyde (PFA)in phosphate buffered saline (PBS) 1X for 10 minutes at 4∘Cand permeabilized with 1 bovine serum albumin (BSA)01-Triton X-100 in PBS 1X for 1 hour at room temperature(RT) Nonspecific antibody binding was blocked with glycine1M pH 88 and with 1 BSA 01 Triton X-100 and 5donkey serum (Jackson ImmunoResearch Laboratories) inPBS 1X Cells were incubated overnight (ON) in PBS 1Xadded with 1 BSA01 Triton X-100 at 4∘C with thefollowing primary antibodies anticleaved Caspase-3 (CellSignaling rabbit polyclonal 9661 1 200 dilution) anti-p-histone H3 (Santa Cruz Biotechnology mouse monoclonalsc-374669 1 50 dilution) or anti-LC3 (Sigma-Aldrich L75431 120 dilution) After rinsing samples were incubated with


the opportune secondary antibody (FITC-conjugated donkeyanti-rabbit 711-095-152 or donkey anti-mouse 715-095-150IgG Jackson ImmunoResearch Laboratories 1 200 dilution)in PBS 1X for 90min at RT In negative controls primaryantibody was omitted After secondary antibody incubationsamples were washed and mounted in buffered glycerol(01M pH 95) All experiments were performed at least intriplicate

For proliferation and apoptosis analyses samples werephotographed with a Zeiss fluorescence microscope (Axio-scope) and positive cells were counted For LC3 immunolo-calization a Leica confocal microscope (Laser Scanning TCSSP2) equipped with ArArKr and HeNe lasers was usedImages were acquired utilizing the Leica confocal softwareThe laser line was at 488 nm for FITC excitation Theimages were scanned under a 20x objective or 40x oilimmersion objective In order to get a quantitative analysis offluorescence optical spatial series each composed of 2325optical sections with a step size of 2120583m were performed inareas in which cells reached confluence both in nonrotatedand in RPM cultured samples The fluorescence intensitywas determined by the Leica confocal software using thefollowing parameters the maximum amplitude of fluores-cence (MAX Amplitude) the sum of intensity (SUM (I))and the mean amplitude of fluorescence intensity (MEAN(A)) of LC3 positive areas The MAX Amplitude representsthe maximum intensity of fluorescence of each series TheSUM (I) represents the total amount of fluorescence intensityrecovered within the entire 119911-axis of each series The MEAN(A) represents the arithmeticalmean of fluorescence intensityrecovered within the entire 119911-axis of each series We analyzed12 equivalent sized regions (regions of interest (ROI)) for eachexperiment both in 1 g and in RPM culture conditions (36total ROI for each experimental condition)

24 Western Blotting of LC3 Autophagy Marker Cells cul-tured at 1 g and in RPM condition for 24 and 48 hourswere lysed in RIPA buffer (Sigma-Aldrich) Samples werethen clarified by centrifugation at 10000 rpm for 10minEquivalent amount of protein (10 120583g) from each samplewas electrophoretically resolved on 125 precast SDS-polyacrylamide gels (ExcelGel GE Healthcare Biosciences)using horizontal apparatus (Pharmacia Biotech UppsalaSweden) Then separated proteins were electrotransferredonto nitrocellulose membranes (Schleicher amp Schuell) by asemidry system (Novablot Pharmacia Biotech) Membraneswere blocked with 3 nonfat milk in PBS and then wereincubated (ON at 4∘C) with the LC3B monoclonal antibody(1 2000 Sigma) After extensive washing with PBS contain-ing 01 tween-20 (TBST) blots were incubated with 1 2000dilution of HRP-conjugated secondary antibody (AmershamBiosciences) for 1 hour at RT Immunopositive bands weredetected with a chemiluminescencersquos detection system (GEHealthcare Biosciences) To check for equal loading of thegel membranes were stripped and reprobedwithmouse anti-120573-actin antibody (1 20000 Sigma) and with anti-GAPDHantibody (1 1000 Cell Signalling Technology) Densitomet-ric analysis was performed with the Quantity One software(BioRad Laboratories)

25 F-Actin and Tubulin Distribution Pattern For F-actinvisualization Rhodamine Phalloidin (Invitrogen MolecularProbes Eugene 1 40 dilution) was used Cells were fixed in4 paraformaldehyde (PFA) in PBS for 10 minutes at 4∘Cand then permeabilized with cold ethanol Acetone 1 1 for10 minutes at 4∘C After rinsing cells were incubated withRhodamine Phalloidin for 25min in the dark Cells were thenwashed in PBS and mounted in buffered glycerol (01M pH95)

Cell height analysis (119911-axis) was performed using theconfocal microscope already described (Leica IRE SP2 LaserScanningTCS SP2) equippedwithArArKr andHeNe lasersImages of the optical sections were acquired using the Leicaconfocal software The Laser Line was at 543 nm for TRITCexcitation Images were scanned under a 40x oil objective Inorder to evaluate cell height three different experiments wereperformedusing cells cultured 1 g and inRPMconditions Foreach experiment 45 optical spatial series with a step size of2 120583mwere recovered and a total of at least 80 optical sectionswere analyzed for each experimental condition Cell heightof the examined samples was calculated using Leica confocalsoftware

For microtubules localization immunofluorescenceexperiments using anti-120572-tubulin (Biomeda mousemonoclonal V10178 1 75 dilution) as primary antibodywere performed The protocol used is the same alreadydescribed in the paragraph above Donkey anti-mouse(715-095-150 IgG Jackson ImmunoResearch Laboratories1 200 dilution) as secondary antibody was used Sampleswere then observed using both fluorescence microscope(Axioscope Zeiss) and confocal microscope (Leica)

26 Scanning Electron Microscopy Samples were fixed inGlutaraldehyde 25 in cacodylate buffer 01M pH 73 ONand then postfixed with 1 osmium tetroxide in cacodylatebuffer 1M dehydrated with increasing ethanol percentage(30ndash90 in water for 5min twice 100 for 15min) treatedin Critical Point Dryer (EMITECH K850) sputter coatedwith platinum-palladium (Denton Vacuum DESKV) andobserved with Supra 40 FESEM (Zeiss)

27 Statistical Analysis All experiments were performed atleast in triplicate All quantitative data are presented as themean value plusmn standard error (SEM) Studentrsquos 119905-test andANOVA test for multigroup comparison were carried outwhen appropriate to evaluate the significance of differencesThe significance level was fixed at a 119875 value lt 005


31 Microgravity Does Not Affect TCam-2 Cell Proliferationand Apoptosis Microgravity exposure is known to influencecell proliferation and apoptosis in normal and cancer cells[52] In order to asses proliferation rate of TCam-2 seminomacells maintained at 1 g or in RPM culture conditions for 24and 48 hours we performed immunofluorescence analysesof the M-phase marker p-histone H3 We observed thatactually this acute microgravity exposure does not affect


the number of mitotic cells at all the culture times considered(Figure 1) Literature data have demonstrated that TCam-2cells do not have a high proliferation rate (58 hours doublingtime) when compared with JKT1 (27 hours doubling time)that is another germ cell tumor cell line [40] Since thepercentage of proliferating cells we expect in the time frameof 24 and 48 hours is not high we can hypothesize that thisaltered gravitational stimulus is not long enough to determinea modification of cell proliferation in this particular cell lineInterestingly after 48 hours of culture the number of mitoticcells decreases significantly in a similar amount both in 1 gand in RPM cultured samples (Figure 1) indicating that cellproliferation in this particular cell line starts to be inhibitedby cell-to-cell contact even if these cells are cancer cells It hasto be noticed that we chose to plate cells at high density inorder to let them attach each other before the RPM exposureand react thanks to their tensional forces to the changes ofgravitational field Due to the high density of plating at theend of the longer culture time we analyzed cell culture dishesare crowded of cells so it appears not possible to prolongmore the culture without detaching and replate cells To thisregard it is fair to say that we cannot exclude that TCam-2 cellproliferationmight be altered by RPM exposure if they wouldhave been cultured at a different density

To test whether microgravity would be able to modifyTCam-2 cell apoptosis we performed immunofluorescencesfor the active fragment of the apoptosis marker Caspase-3We found that the change of gravity vector does not affect thenumber of apoptotic cells after 24 hours of culture (Figure 2)However it has to be noticed that after 48 hours of culturethe number of apoptotic cells increases significantly in theRPM cultured samples even if the large majority of cellsappear to tolerate this mechanical stress (Figure 2) and tosurvive The latter observation indicates that a small part ofTCam-2 cells appears more sensible to the change of gravityvector when the mechanical stimulus is prolonged a bitbut this sensibility does not seem related to mechanical cellstability because due to the high density of plating all cellsare stably attached to each other and to the substrate Inaddition apoptotic cells are observable uniformly dispersedin the culture dish On the basis of this observation wehypothesized that TCam-2 cells need to trigger rescue pro-cesses that let them survive after a prolonged change of gravityvector Possibly rescue processes are not correctly inducedor exploited by the whole population of TCam-2 cells andthis hypothesis may explain why a small percentage of themappears not able to survive to the change of gravity Thechange of physical forces is sensed by the cells through theircytoskeleton components and one of the first features thatreveal a cytoskeletal modification is the change in the plasmamembrane morphology We studied first membrane surfaceand cytoskeletal modifications due to RPM exposure to besure the TCam-2 cells are able to sense and modify theirshape in response to this mechanical stress Then we evalu-ated in the same culture conditions the autophagic processmodulation in response to RPM exposure since autophagyis the most known biological rescue mechanism that let cellto change rapidly and survive to suddenmicroenvironmentalchanges

32 Microgravity Strongly Influences TCam-2 Cell MembraneSurface To study if the alteration of the mechanical forcesacting on TCam-2 cells during microgravity simulation maymodify cell membrane surface morphology samples wereanalyzed by scanning electron microscopy We observed thepresence of two morphologically distinguishable cell popula-tions in the 1 g cultured samples one has smooth membranesurface and the other one is characterized by the presence ofmembrane expansions morphologically similar to microvilli(Figure 3) Noteworthy we found that microgravity stronglyaffects membrane surface appearance after 24 hours ofculture microvilli appeared collapsed and the differencesbetween the two cell populations are less evident (Figure 3)It is of interesting notice that cell microvilli are considered tobe an important site of mechanotransduction both in sensoryspecialized cells and not-sensory cells [53] After 48 hours ofculture the membrane surface differences appear recoveredand microvilli-like structures appear reconstituted in RPMcultured samples (Figure 3) On the basis of these obser-vations we hypothesized that cell mechanosensor-systemwas transiently altered by RPM exposure and this stronglysuggested the occurrence of cytoskeleton remodeling due toan acute exposure to gravitational vector change

33 Microgravity Induces TCam-2 Cytoskeleton RemodelingA huge amount of literature data demonstrated that micro-gravity is able to influence cell cytoskeletal architecture pro-moting cell morphofunctional alterations [54] In the light ofthese observations and on the basis of our scanning electronmicroscopy data we decided to evaluate the possible effectsof simulated microgravity on TCam-2 microfilament andmicrotubule organization Herein we report microfilamentdistribution pattern analyzed by F-actin staining of TCam-2 cells cultured at 1 g or in RPM culture conditions Even ifno apparent significant alterations in the actin cytoskeletonorganization were found both in 24 (Figure 4(a)) and 48hours of culture (not shown) a more detailed analysis byconfocal microscopy using Leica confocal software allowedus to evaluate cell height (cell 119911-axis) (Figures 4(b) 4(c)and 4(d)) in all the considered experimental conditions Weobserved that simulated microgravity significantly increasesTCam-2 cell height after 24 hours of RPM exposure withrespect to 1 g cultured cells (1562 plusmn 110 120583m versus 110 plusmn066 120583m 119875 lt 0001) indicating that RPM culture conditionwas able to modify TCam-2 cell shape Noteworthy after48 hours of culture the differences in cell height in 1 gand RPM cultured cells are no more statistically significant(Figure 4(d)) indicating that TCam-2 cells are able to recoverrapidly after the exposure to this mechanical stressThe latterobservation appears consistent with the reported recovery ofsurface membrane microvilli-like structures after 48 hours ofRPM exposure (Figure 3)

Microtubule distribution pattern was studied by anti-120572-tubulin immunofluorescence staining After 24 hours ofculture we observed that microtubule distribution is alteredinTCam-2 cells exposed toRPMculture condition centriolarpolarization is much less visible in these samples and micro-tubules appear to be distributed in an apparently random


p-H

iston

e H3

posit

ive c

ells

()

0

1

2

3

4

5

6

7

1GRPM

aa

bb

(h)24 48

(a)

(I) (II)

(b)

Figure 1 RPM exposure does not influence TCam-2 cell proliferation (a) Graphical representation of the percentage of proliferating cells(p-histone H3 positive cells) at 24 and 48 hours of culture No differences were observed between TCam-2 cells cultured at 1 g or in RPMculture conditions Data are expressed as the mean plusmn SEM Same letters indicate no statistical difference Different letters indicate 119875 lt 005(b) Representative images of TCam-2 cells cultured for 24 hours at 1 g (I) and in RPM condition (II) after p-histone H3 immunofluorescenceBar 50 120583m

manner within the cells (Figure 5) Microtubules are keyregulators of membrane trafficking organelle distributioninside the cells and together with actin microfilaments seemsto regulate autophagosome formation [55ndash57] In additionit is of interesting notice that LC3 the marker protein ofthe autophagic process is a microtubule associated protein(MAP) As well as actin filaments after 48 hours of culturethe microtubule distribution pattern appears recovered inRPM exposed samples since it is not possible to observesignificant differences between 1 g and RPM cultured cellsThese observations again clearly indicate the capability ofTCam-2 cell to sense the change of physical forces in theirmicroenvironment and also to recover rapidly from thisphysical stress These data strongly suggest the trigger ofrescue mechanisms due to TCam-2 RPM exposure

It is worth mentioning that the reported microtubulealteration does not appear to significantly alter the properformation of the mitotic spindle (Figure 5(g) white box)

This observation is consistent with the results reported inFigure 1 in which we observed that TCam-2 cell proliferationdoes not appear to be affected by RPM exposure

34 Microgravity Induces TCam-2 Cell Autophagy Somepapers in the literature reported that in other cellular sys-tems microgravity is involved in autophagy induction [18ndash20] and as previously stated cytoskeleton plays importantroles in autophagy regulation [22] In particular in mam-mals microtubules appear to be involved in the fusion ofautophagosome with late endosome and to bind and trans-port autophagosomes once terminally completedThe role ofactin filaments on mammalian autophagy process regulationis still a matter of debate but it is worth mentioning thatmicrofilaments depolymerization agents are able to blockautophagosome formation

TCam-2 cells cultured at 1 g and in RPM conditionswere immunostained to detect the autophagic marker LC3


123456789

10

Clea

ved

casp

ase-

3 po

sitiv

e cel

ls (

)

aa a

b

0

1GRPM

(h)24 48

(a)

(II)

(IV)

(I)

(III)

(b)

Figure 2 RPM exposure and TCam-2 cell apoptosis (a) Graphical representation of the percentage of apoptotic cell number (anticleavedCaspase-3 positive cells) No differences were observed between TCam-2 cells cultured for 24 hours at 1 g or in RPM culture conditions Onthe contrary a slight increase in apoptotic cell percentage is observed after 48 hours of culture Data are expressed as the mean plusmn SEM Sameletters indicate no statistical difference Different letters indicate 119875 lt 001 (b) Representative images of 1 g (I III) and RPM (II IV) exposedTCam-2 cells in 24 (I II) and 48 (III IV) hours of culture after cleaved Caspase-3 immunofluorescence Bar 50120583m (I and II) 35 120583m (III andIV)


RPM 24h1G 24h

RPM 48h

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

1G 48h

(i)

(l)

(m)

(n)

lowast

lowast

lowast

Figure 3 Microgravity effect on TCam-2 cell membrane surface Scanning electron microscopy pictures with increasing magnificationshowing cell membrane surfacemorphology of TCam-2 cells cultured for 24 (a b c and d) and 48 hours (i l) at 1 g or for 24 (e f g and h) and48 hours (m n) inRPMculture conditions In (a)white asterisks indicate TCam-2 cells with smoothmembrane surfacewhile the other TCam-2 cells of the same image are characterized by the presence of microvilli-like structures In (b) the boundary between one smooth membraneand one microvilli membrane presenting cells is reported (c) and (d) represent higher magnifications of the microvilli-like structures ofTCam-2 cells maintained at 1 g In (e) (f) (g) and (h) it is well evident that in RPM cultured cells membrane surface is more similar in allthe cells and it is difficult to clearly identify the two cell populations In particular in (h) it is possible to observe that microvilli-like structuresappeared collapsed in RPM exposed TCam-2 cells The morphological appearance of cell surface (i m) and microvilli-like structures (l n)appeared indistinguishable in 1 g (i l) and RPM exposed cells (m n) after 48 hours of culture


(I)

100200

300375

(I) (II)

RPM

0

5

10

15

20

1G RPM

48h culture

Cell

hei

ght (120583

m)

z-ax

is (c

ell w

idth

)z-

axis

(cel

l wid

th)

(I)(I)(I)(I) (II((II((IIII((( ))))))

1G

24h culture

24h culture

0

2

4

6

8

10

12

14

16

18 24h culture

1G RPM

Cell

hei

ght (120583

m)

(a)

(b)

(c)

(d)

lowast

100200

300375

x (120583m)x

(120583m)

100200

300375

100200

300375

(II)

x (120583m)x

(120583m)

Figure 4 Simulated microgravity influences TCam-2 cell height (a) Rhodamine-phalloidin staining of TCam-2 cells showing F-actindistribution pattern after 24 hours of culture at 1 g (I) or under RPM (II) conditions Bar 20120583m (b) Representative images of cell heightobtained using the Leica confocal software of samples cultured for 24 hours at 1 g (I) or in RPM (II) conditions (c) Graphical representationof cell height obtained by confocal microscopy analysis on 1 g and RPM exposed cells after 24 hours of culture (lowast1 562plusmn1 10120583m versus 110 plusmn066 120583m 119875 lt 0001) Data are expressed as the mean plusmn SEM (d) Graphical representation of cell height obtained by confocal microscopyanalysis on 1 g and RPM exposed cells after 48 hours of culture (1302 plusmn 132120583m versus 1602 plusmn 249 120583m resp) Data are expressed as themean plusmn SEMThe values are not statistically significant


1G 24h

(a)

RPM 48h1G 48h

(b)

(c)

(g)

(h)

RPM 24h

(d)

(e)

(f)

(i)

(l)

Figure 5 Microtubule distribution pattern in TCam-2 cells exposed to simulated microgravity Immunodetection of 120572-tubulin in TCam-2cells cultured for 24 hours (a b c d e and f) and 48 hours (g h i and l) at 1 g (a b c g and h) or under RPM conditions (d e f i and l) Inimages (g) and (i) in the white box representative images of mitotic spindles are also shown Bar 20120583m


As shown in Figures 6(a)(II) and 6(a)(IV) LC3 is detectableboth in 1 g and in RPM cultured samples and it ismainly localized in cytoplasmic vesicles Interestingly thenumber of these LC3 positive vesicles appears stronglyincreased in TCam-2 cells exposed to microgravity con-ditions (Figure 6(a)(IV)) with respect to 1 g cultured cells(Figure 6(a)(II)) after 24 hours of cultureMoreover a quanti-tative analysis carried out using the Leica confocal softwareallows us to quantify the fluorescence intensity increase ofLC3 stained cells exposed to simulated microgravity (Figures6(b) and 6(c)) In particular Figure 6(b) shows a stackprofile of 12 regions of interest (ROI) of a representativeexperiment both in 1 g (I) and in RPM cultured samples (II)The two groups of peaks reported in this figure representthe Max amplitude of fluorescence detected by the confocalmicroscope from the beginning to the end of the sample (total119911-axis) It is well evident thatMax amplitude of fluorescence isincreased in simulated RPM exposed samples We evaluatedalso the SUM (I) and the MEAN (A) of fluorescenceConsistent with the data reported in Figure 6(b) we observedalso an increase of both the SUM (I) and the MEAN (A) inRPM cultured cells after 24 hours of culture (Figure 6(c))According to the described confocal quantitative analyseswestern blots performed with the anti-LC3 antibody showedthat besides the increase of LC3-I protein amount LC3-II (the LC3 active isoform) protein content is increased inRPM with respect to 1 g cultured samples (Figure 7) Sameresults were obtained normalizing the LC3 bands versus 120573-actin (Figure 7) and versus GAPDH signal (not shown)Autophagy induction is a naturally transient process thisphenomenon is called autophagic flux [58] since when itworks autophagy protein machinery has to be degraded vialysosomes or proteasome together with the portion of the cellthat needs to be eliminated On the contrary when autophagyis blocked the autophagy protein machinery is not degradedand is maintained at high level in the cytoplasm In oursamples after 48 hours of culture autophagy active proteinLC3-II together with LC3-I appears quantitatively similar in1 g and RPM cultured cells demonstrating that autophagy isrestored at the same level with respect to 1 g culture conditionSame results were obtained normalizing the LC3 bands versus120573-actin (Figure 7) and versus GAPDH signal (not shown)Consistent with this observation the LC3 cytoplasmic fluo-rescence is lowered in the RPM exposed cells demonstratingthat autophagy was not blocked by this mechanical stress(Figure 6(a)(VI)) It has to be mentioned that LC3-II proteinis present at basal level at 24 and 48 hours of culture aswell as cytoplasmic LC3 dots even in cells cultured at 1 gindicating that autophagy is a housekeeping process thatworks inTCam-2 cells even in control samples and suggestingthat this cancer cell line may exploit autophagy as a survivalmechanism

There is a common agreement indicating that thereis a relationship between autophagy and apoptosis whenautophagy is not able to rescue cell frommicroenvironmentalchanges apoptotic process is triggered On the light of thistheory we might interpret the small increase in the apoptoticindex at 48 hours of culture in RPM cultured samples(Figure 2) as the autophagy efficiency threshold or the limit

of autophagy efficiency in the rescue of cell survival aftermechanical stress exposure

All together these qualitative and quantitative analysesallow us to conclude that microgravity is able to positivelymodulate the autophagic process in TCam-2 seminomacell line Autophagy induced in TCam-2 cells by Estrogenexposure through ER120573 activation was recently reported [59]Herein we reported for the first time autophagy induced inTCam-2 cells by a mechanical cue (or more precisely by aremoval of a mechanical stimulus) instead of a biochemicalone The analysis of the autophagy related pathways inducedby RPM exposure and the direct role of microtubules andmicrofilaments in this process as well as the other possiblebiological meanings of RPM induced TCam-2 autophagydeserves further investigations

4 Conclusions

Gravitational biology could be considered part ofmechanobi-ology the science that investigates the impact of forces onliving organisms At cellular level cytoskeleton elements arelikely candidates for force sensing and transduction pro-cesses These biomechanical properties of cell cytoskeletonexplain the capability to propagate a mechanical stimulusover long distances in living tissues and represent the basisof the intriguing hypothesis that many if not all reportedchanges in ion fluxes protein phosphorylation membranepotential changes and so forth are indeed provoked by amechanical modification somewhere within the cell or onits membrane [60 61] This paper is in line with this theoryand adds experimental data supporting the importance ofmechanotransduction and cell behavior In this paper in factwe reported the effects of the exposure to changes of gravityvector on TCam-2 seminoma cells In this experimentalmodel simulated microgravity is able to induce TCam-2 cellsurfacemodifications andmicrovilli-like structure alterationMoreover microtubules and microfilaments organizationresult to be influenced by microgravity (a) TCam-2 cellsshow actin cytoskeleton remodeling and cell height increase(b) centriolar polarization becomes much less visible in thesesamples and microtubules appear to be distributed in anapparent random manner within the cells All these modi-fications appear to be transient indicating that cells modifytheir cytoskeletal components in response to gravitationalforce change but that are also able to recover their shapewhen the gravitational change is prolonged InterestinglyRPM exposure is able to induce TCam-2 cell autophagyThe latter observation allows us to hypothesize that TCam-2 cells are able to rapidly respond to acute exposure tomicrogravity inducing adaptive biological processes suchas autophagy that probably allow them to survive in thechanging physical microenvironment Since autophagy isconsidered a biological survival mechanism the apoptosisinduction in a small percentage of TCam-2 cells after 48 hoursof culture might be speculated as the limit in the efficiencyof this survival process All together these data provideevidences of TCam-2 sensitivity to changes of gravitationalforce direction and lay the groundwork to further studies onTCam-2 cell autophagy and its biological meaning


(I) (III)

(IV) (VI)(II)

(V)

(a)

(120583m)

90

80

70

60

50

40

30

20

10

5 10 15 20 25 30 35 40

(I)Max amplitude

(120583m)

90

80

70

60

50

40

30

20

10

5 10 15 20 25 30 35 40

(II)Max amplitude

(b)

16

14

12

10

8

6

4

2

0

1G RPM

SUM

(I) (

au)

lowast

35

30

25

20

15

10

5

0

1G RPM

Mea

n(A

) (au

)

lowast

(c)

Figure 6 Autophagy induction in TCam-2 cells exposed to microgravity (a) Immunodetection of LC3 in TCam-2 cells cultured for 24 hoursat 1 g (II) or under RPM (IV) conditions In VI LC3 immunodetection of TCam-2 cells cultured in RPM condition for 48 hours is reported InI III and V the respective bright fields are shown (b) Stack profile of 24 hours of culture representative experiment showing the maximumamplitude (MAXAmplitude) of fluorescence in 12 regions of interest (ROI) randomly drawn in an area in which the cells reached confluencein nonrotated (I) and RPM cultured samples (II) It is evident an increase of maximum amplitude of fluorescence in microgravity exposedsamples (II) with respect to the 1 g-cultured cells (I) (c) MEAN (A) (lowast2 762plusmn104 versus 1434 plusmn 059 119875 lt 0001) and SUM (I) (lowast1 292plusmn085versus 695 plusmn 152 119875 lt 005) confirm an increase of LC3 positivity in RPM exposed sample with respect to 1 g cells after 24 hours of cultureData are expressed as the mean plusmn SEM


LC3-I

LC3-II

1G RPM

120573-Actin

24hours 48hours1G RPM

(a)

20

40

60

80

100

120

140

160

0

LC3-ILC3-II

1G RPM

Relat

ive i

nten

sity

(LC3

120573-a

ctin

)

24hours

lowastP le 005

lowast

(b)

20

40

60

80

100

120

140

0

LC3-ILC3-II

1G RPM

Relat

ive i

nten

sity

(LC3

120573-a

ctin

)

48hours

(c)

Figure 7 Western blot analysis of LC3 autophagy marker (a) Representative images of the bands revealed by anti-LC3 western blot analysison 24 and 48 hours cultured samples As expected anti-LC3 antibody detected both the LC3 isoforms (LC3-I cytosolic isoform LC3-IIautophagosomal membrane-conjugated isoform) (b) Graphical representation summarizing the densitometric analysis of the LC3-I andLC3-II bands normalized versus 120573-actin in 24 hours cultured samples Data are expressed as the mean plusmn DS lowastversus 1 g 119875 lt 005 (c)Graphical representation summarizing the densitometric analysis of the LC3-I and LC3-II bands normalized versus 120573-actin in 48 hourscultured samples Data are expressed as the mean plusmn DS The values are not statistically significant




Angela Catizone and Giulia Ricci are equal senior authors

Acknowledgments

The authors wish to thank Professor Mariano Bizzarri forthe valuable suggestions and the Italian Space Agency for thesupport received to carry out this research project

References

[1] D Vorselen W H Roos F C MacKintosh G J Wuite and JJ van Loon ldquoThe role of the cytoskeleton in sensing changes ingravity by nonspecialized cellsrdquoThe FASEB Journal vol 28 no2 pp 536ndash547 2014

[2] G Aleshcheva J Sahana X Ma et al ldquoChanges inmorphologygene expression and protein content in chondrocytes culturedon a random positioning machinerdquo PLoS ONE vol 8 no 11Article ID e79057 2013

[3] S Li Z Ma Z Niu et al ldquoNASA-approved rotary bioreactorenhances proliferation and osteogenesis of human periodontalligament stem cellsrdquo Stem Cells and Development vol 18 no 9pp 1273ndash1282 2009

[4] R Tamma G Colaianni C Camerino et al ldquoMicrogravityduring spaceflight directly affects in vitro osteoclastogenesisand bone resorptionrdquo The FASEB Journal vol 23 no 8 pp2549ndash2554 2009

[5] S J PardoM J PatelMC Sykes et al ldquoSimulatedmicrogravityusing the Random Positioning Machine inhibits differentiationand alters gene expression profiles of 2T3 preosteoblastsrdquoAmerican Journal of Physiology-Cell Physiology vol 288 no 6pp C1211ndashC1221 2005

[6] A Guignandon M H Lafage-Proust Y Usson et al ldquoCellcycling determines integrin-mediated adhesion in osteoblasticROS 1728 cells exposed to space-related conditionsrdquo TheFASEB Journal vol 15 no 11 pp 2036ndash2038 2001




[9] T D Ross B G Coon S Yun et al ldquoIntegrins inmechanotrans-ductionrdquo Current Opinion in Cell Biology vol 25 no 5 pp 613ndash618 2013

[10] B Geiger A Bershadsky R Pankov and K M YamadaldquoTransmembrane extracellular matrix-cytoskeleton crosstalkrdquoNature Reviews Molecular Cell Biology vol 2 no 11 pp 793ndash805 2001

[11] M Y Kapitonova N Salim and S Othman ldquoAlteration of cellcytoskeleton and functions of cell recovery of normal humanosteoblast cells caused by factors associated with real spaceflightrdquo Malaysian Journal of Pathology vol 35 no 2 pp 153ndash163 2013

[12] G Pani N Samari R Quintens et al ldquoMorphological andphysiological changes in mature in vitro neuronal networkstowards exposure to short- middle- or long-term simulatedmicrogravityrdquo PLoS ONE vol 8 no 9 Article ID e73857 2013

[13] J Nakashima F Liao J A Sparks Y Tang and E B BlancaflorldquoThe actin cytoskeleton is a suppressor of the endogenous skew-ing behaviour of Arabidopsis primary roots in microgravityrdquoPlant Biology vol 16 supplement 1 pp 142ndash150 2013

[14] C Nouri J A Tuszynski M W Wiebe and R Gordon ldquoSimu-lation of the effects of microtubules in the cortical rotation ofamphibian embryos in normal and zero gravityrdquo BioSystemsvol 109 no 3 pp 444ndash449 2012

[15] J Li S Zhang J Chen T Du Y Wang and ZWang ldquoModeledmicrogravity causes changes in the cytoskeleton and focaladhesions and decreases in migration in malignant humanMCF-7 cellsrdquo Protoplasma vol 238 no 1ndash4 pp 23ndash33 2009



[18] H W Ryu S H Choi S Namkoong et al ldquoSimulated micro-gravity contributes to autophagy induction by regulating AMP-activated protein kinaserdquo DNA and Cell Biology vol 33 no 3pp 128ndash135 2014

[19] Y SambandamM T Townsend J J Pierce et al ldquoMicrogravitycontrol of autophagy modulates osteoclastogenesisrdquo Bone vol61 pp 125ndash131 2014

[20] Y C Wang D Y Lu F Shi et al ldquoClinorotation enhancesautophagy in vascular endothelial cellsrdquo Biochemistry and CellBiology vol 91 no 5 pp 309ndash314 2013

[21] D Sandona J Desaphy G M Camerino et al ldquoAdaptation ofmouse skeletal muscle to long-term microgravity in the MDSmissionrdquo PLoS ONE vol 7 no 3 Article ID e33232 2012

[22] I Monastyrska E Rieter D J Klionsky and F ReggiorildquoMultiple roles of the cytoskeleton in autophagyrdquo BiologicalReviews vol 84 no 3 pp 431ndash448 2009

[23] J H Choi Y S Cho Y H Ko S U Hong J H Park andM A Lee ldquoAbsence of autophagy-related proteins expressionis associated with poor prognosis in patients with colorectaladenocarcinomardquo Gastroenterology Research and Practice vol2014 Article ID 179586 10 pages 2014

[24] N Orfali S L McKenna M R Cahill L J Gudas and N PMongan ldquoRetinoid receptor signaling and autophagy in acutepromyelocytic leukemiardquo Experimental Cell Research vol 324no 1 pp 1ndash12 2014

[25] C Cerella M H Teiten F Radogna M Dicato and MDiederich ldquoFrom nature to bedside Pro-survival and celldeath mechanisms as therapeutic targets in cancer treatmentrdquoBiotechnology Advances 2014

[26] C Fabrizi V S De F Somma et al ldquoLithium improves survivalof PC12 pheochromocytoma cells in high-density cultures andafter exposure to toxic compoundsrdquo International Journal of CellBiology vol 2014 Article ID 135908 7 pages 2014

[27] L Yu L Strandberg and M J Lenardo ldquoThe selectivity ofautophagy and its role in cell death and survivalrdquo Autophagyvol 4 no 5 pp 567ndash573 2008

[28] F Strollo G Riondino B Harris et al ldquoThe effect of micro-gravity on testicular androgen secretionrdquo Aviation Space andEnvironmental Medicine vol 69 no 2 pp 133ndash136 1998

[29] F Strollo M A Masini M Pastorino et al ldquoMicrogravity-induced alterations in cultured testicular cellsrdquo Journal ofGravitational Physiology vol 11 no 2 pp P187ndash188 2004

[30] Y Ding J Tang J Zou et al ldquoThe effect of microgravity ontissue structure and function of rat testisrdquo Brazilian Journal ofMedical and Biological Research vol 44 no 12 pp 1243ndash12502011

[31] T Kaneko S Sasaki Y Umemoto Y Kojima T Ikeuchiand K Kohri ldquoSimulated conditions of microgravity increasesprogesterone production in I-10 cells of Leydig tumor cell linerdquoInternational Journal ofUrology vol 15 no 3 pp 245ndash250 2008

[32] MAHMotabagani ldquoMorphological andmorphometric studyon the effect of simulated microgravity on rat testisrdquo ChineseJournal of Physiology vol 50 no 4 pp 199ndash209 2007

[33] F Strollo G Strollo M More et al ldquoChanges in humanadrenal and gonadal function onboard Spacelabrdquo Journal ofGravitational Physiology vol 4 no 2 pp 103ndash104 1997

[34] U Engelmann F Krassnigg and W- Schill ldquoSperm motilityunder conditions of weightlessnessrdquo Journal of Andrology vol13 no 5 pp 433ndash436 1992

[35] G Ricci A Catizone R Esposito and M Galdieri ldquoMicro-gravity effect on testicular functionsrdquo Journal of gravitationalphysiology vol 11 no 2 pp 61ndash62 2004

[36] G Ricci R Esposito A Catizone and M Galdieri ldquoDirecteffects of microgravity on testicular function analysis of hys-tological molecular and physiologic parametersrdquo Journal ofEndocrinological Investigation vol 31 no 3 pp 229ndash237 2008

[37] S di Agostino F Botti A di Carlo C Sette and R GeremialdquoMeiotic progression of isolated mouse spermatocytes undersimulatedmicrogravityrdquo Reproduction vol 128 no 1 pp 25ndash322004

[38] M Pellegrini S di SienaGClaps et al ldquoMicrogravity promotesdifferentiation andmeiotic entry of postnatal mouse male germcellsrdquo PLoS ONE vol 5 no 2 Article ID e9064 2010

[39] J de Jong H Stoop A J M Gillis et al ldquoFurther char-acterization of the first seminoma cell line TCam-2rdquo GenesChromosomes and Cancer vol 47 no 3 pp 185ndash196 2008

[40] D Eckert D Nettersheim L C Heukamp S Kitazawa K Bier-mann and H Schorle ldquoTCam-2 but not JKT-1 cells resembleseminoma in cell culturerdquo Cell and Tissue Research vol 331 no2 pp 529ndash538 2008

[41] N C Goddard A McIntyre B Summersgill D Gilbert SKitazawa and J Shipley ldquoKIT and RAS signalling pathways


in testicular germ cell tumours new data and a review of theliteraturerdquo International Journal of Andrology vol 30 no 4 pp337ndash348 2007

[42] YMizuno A Gotoh S Kamidono and S Kitazawa ldquoEstablish-ment and characterization of a new human testicular germ celltumor cell line (TCam-2)rdquoNihon Hinyokika Gakkai Zasshi vol84 no 7 pp 1211ndash1218 1993

[43] D Nettersheim L C Heukamp F Fronhoffs et al ldquoAnalysisof TET expressionactivity and 5mC oxidation during normaland malignant germ cell developmentrdquo PLoS ONE vol 8 no12 Article ID e82881 2013

[44] F Ferranti B Muciaccia G Ricci et al ldquoGlial cell line-derivedneurotrophic factor promotes invasive behaviour in testicularseminoma cellsrdquo International Journal of Andrology vol 35 no5 pp 758ndash768 2012

[45] F Ferranti F DAnselmi M Caruso et al ldquoCorrection TCam-2 seminoma cells exposed to egg-derived microenvironmentmodify their shape adhesive pattern and migratory behavioura molecular and morphometric analysisrdquo PLoS ONE vol 8 no10 2013

[46] S M Russell M G Lechner A Mokashi et al ldquoEstablishmentand characterization of a new human extragonadal germ cellline SEM-1 and its comparison with TCam-2 and JKT-1rdquoUrology vol 81 no 2 pp 464ndashe9 2013

[47] R Franco F Boscia V Gigantino et al ldquoGPR30 is overex-pressed in post-puberal testicular germ cell tumorsrdquo CancerBiology ampTherapy vol 11 no 6 pp 609ndash613 2011

[48] F Esposito F Boscia V Gigantino et al ldquoThe high-mobilitygroup A1-estrogen receptor 120573 nuclear interaction is impairedin human testicular seminomasrdquo Journal of Cellular Physiologyvol 227 no 12 pp 3749ndash3755 2012

[49] DNettersheimA JMGillis LH J Looijenga andH SchorleldquoTGF-1205731 EGF and FGF4 synergistically induce differentiationof the seminoma cell line TCam-2 into a cell type resemblingmixed non-seminomardquo International Journal of Andrology vol34 no 4 part 2 pp e189ndashe203 2011

[50] D Nettersheim A Gillis K Biermann L H J Looijengaand H Schorle ldquoThe seminoma cell line TCam-2 is sensitiveto HDAC inhibitor depsipeptide but tolerates various otherchemotherapeutic drugs and loss of NANOG expressionrdquoGenes Chromosomes and Cancer vol 50 no 12 pp 1033ndash10422011

[51] U Eppelmann F Gottardo J Wistuba et al ldquoRamanmicrospectroscopic discrimination of TCam-2 culturesreveals the presence of two sub-populations of cellsrdquo Cell andTissue Research vol 354 no 2 pp 623ndash632 2013


[53] O P Hamill and B Martinac ldquoMolecular basis of mechan-otransduction in living cellsrdquo Physiological Reviews vol 81 no2 pp 685ndash740 2001


[55] A Aplin T Jasionowski D L Tuttle S E Lenk and W ADunn Jr ldquoCytoskeletal elements are required for the formationand maturation of autophagic vacuolesrdquo Journal of CellularPhysiology vol 152 no 3 pp 458ndash466 1992

[56] E Fass E Shvets I Degani K Hirschberg and Z ElazarldquoMicrotubules support production of starvation-induced

autophagosomes but not their targeting and fusion withlysosomesrdquo Journal of Biological Chemistry vol 281 no 47 pp36303ndash36316 2006

[57] R Kochl X W Hu E Y W Chan and S A ToozeldquoMicrotubules facilitate autophagosome formation and fusionof autophagosomes with endosomesrdquo Traffic vol 7 no 2 pp129ndash145 2006

[58] N Mizushima T Yoshimori and B Levine ldquoMethods inMammalian Autophagy Researchrdquo Cell vol 140 no 3 pp 313ndash326 2010

[59] CGuido S PanzaM Santoro et al ldquoEstrogen receptor120573 (ER120573)produces autophagy and necroptosis in human seminoma cellline through the binding of the Sp1 on the phosphatase andtensin homolog deleted fromchromosome 10 (PTEN) promotergenerdquo Cell Cycle vol 11 no 15 pp 2911ndash2921 2012

[60] J J W A van Loon ldquoMechanomics and physicomics ingravisensingrdquo Microgravity Science and Technology vol 21 no1-2 pp 159ndash167 2009

[61] M Bizzarri A Cucina A Palombo and M G MasielloldquoGravity sensing by cells mechanisms and theoretical groundsrdquoRendiconti Lincei vol 25 pp 29ndash38 2014

Research ArticleGravity Affects the Closure of the Traps in Dionaea muscipula

Camilla Pandolfi1 Elisa Masi1 Boris Voigt2 Sergio Mugnai1

Dieter Volkmann2 and Stefano Mancuso1

1 DISPAA University of Florence Viale delle idee 30 50019 Sesto Fiorentino Italy2 IZMB University of Bonn Kirschallee 1 53115 Bonn Germany

Correspondence should be addressed to Camilla Pandolfi camillapandolfiunifiit

Received 12 May 2014 Accepted 27 June 2014 Published 15 July 2014


Copyright copy 2014 Camilla Pandolfi et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Venus flytrap (Dionaea muscipula Ellis) is a carnivorous plant known for its ability to capture insects thanks to the fast snapping ofits traps This fast movement has been long studied and it is triggered by the mechanical stimulation of hairs located in the middleof the leaves Here we present detailed experiments on the effect of microgravity on trap closure recorded for the first time during aparabolic flight campaign Our results suggest that gravity has an impact on trap responsiveness and on the kinetics of trap closureThe possible role of the alterations of membrane permeability induced by microgravity on trap movement is discussed Finally weshow how the Venus flytrap could be an easy and effective model plant to perform studies on ion channels and aquaporin activitiesas well as on electrical activity in vivo on board of parabolic flights and large diameter centrifuges

1 Introduction

The response of Venus flytrap (Dionaea muscipula) tomechanical stimulation has long been known and it is oneof the most rapid movements in the plant kingdom [1 2]

The plant produces a rosette of leaves each divided intotwo parts a lower part called the lamina and the upper partcalled the trap The trap catches prey thanks to a very rapidmovement of its bilobed halves that shut when the triggerhairs are stimulated At room temperature two touches acti-vate the trap which snaps shut in a fraction of second [3]At higher temperature only one stimulus is required for trapclosure [4] The stimulation of the trigger hairs activatesmechanosensitive ion channels and generates receptor poten-tials inducing the action potentials (APs) that initiate theclosure [5] electrical signals are the immediate cause of thetrapmovements irrespective of the way in which the signal istriggered (mechanical stimulation or electrostimulation) [5]Once the insect is caught the lobes seal tightly allowing diges-tion to take place [6 7] The APs in Dionaea muscipula havebeen extensively studied (eg [5 8 9]) Trigger hair-inducedgeneration of action potentials is not exclusively associatedwith the trap closure The struggling of the entrapped preyin the closed trap results in the generation of further action

potentials which cease to occur just when the prey stopsmovingTheseAPsmay induce inhibition of the dark reactionof photosynthesis [10] showing that chlorophyll-A fluores-cence is under electrochemical control [11] Although thisspectacular example of plant movement has long fascinatedscientists the mechanism by which the trap works remainspoorly understood [12] Some explanations proposed involvean irreversible cell wall loosening induced by the acidifica-tion of the cells [6] or a rapid loss of turgor pressure similarlyto what happens in stomata [13] However the validity ofboth mechanisms has been questioned because they cannotexplain the speed at which the movement happens Morerecently other two models have been proposed the elasticdeformation that results from a snap-buckling instability [14]and a hydroelastic curvature mechanism based on the fastopening ofwater channels [9] Bothmodelsmay convincinglyaccount for the speed of the movement

Thepossibility to study the effect ofmicrogravity on livingorganisms is a unique opportunity to observe the alterationof phenomena in the absence of the otherwise omnipresentgravity force Although the understanding of the effect ofgravity on animal and plant bodies is crucial in view of thepossible future space travels the research is moving slowly if



compared with other research fields due to the accessibilityto microgravity conditions and the challenging experimentalcondition

In the present study we report the effect of microgravityon trap closure conducted during a parabolic flight campaignThis gave us the opportunity to test Dionaea muscipula aspossible candidate to study the effect of gravity on the elec-trical activity of organisms bymonitoring the variation in theexcitability of the traps and the alteration in the kinetics oftheir closure Furthermore the changes in the kinetics of trapclosure gave us important hints on themechanism at the baseof the fast trap snapping


Parabolic flight experiments were performed in an A300 air-plane (Novespace France) during the 9th DLR parabolicflight campaign A typical parabolic flight manoeuvre pro-vides alternating acceleration levels of regular gravity (1 g)microgravity (0 g) for 22 s and two periods (20 s) of hyper-gravity (up to 18 g for 20 s) before and after each period ofmicrogravity Twenty Dionaea muscipula J Ellis plants weregrown in a growth chamber with 1410 h lightdark period inwell-drained peat moss in plastic pots irrigated with distilledwater All experiments were performed on healthy adult spec-imens The pots were sealed with parafilm to avoid the float-ing of the substrate during the zero gravity periods and plantswere secured inside a plexiglass growth chamber Digital HDvideo recorder Sony HDR-SR11E was used to film the Venusflytraps at 25 fps Each day experiments were done on 4different plants All the traps were mechanically stimulatedduring 0 1 and 2 g using a wooden stick by gently touchingthe trigger hairsThewooden stick was immediately removedafter stimulation The collected movies were analysed frameby frame and the trap closure was quantified by measuringthe change of distance between two laminas with ImageJ soft-ware [15] Space and time constraints limited the number ofplants that could be carried on board and consequently lim-ited the number of traps available each day More emphasiswas given to the zero gravity condition therefore 25 trapswere tested in microgravity 20 were tested in 1 g during theflight and 8 traps were devoted to 2 g The distance 119910(119905)between the edges of the trap leaf wasmeasured in the closingprocess In the open state the distance between the edges ofthe trap leaf is 119910max As individual plants have different open-ing distances data were normalized 119909 = 119910119910max The speedof trap closure was calculated as V = 119889119909119889119905 and it has thedimension of sminus1

3 Results

The trap closure was studied at different gravity conditions(Figure 1)

Results showed a low responsiveness of traps in micro-gravity 36 of the traps did not close at all and 48 mani-fested a slower closing motion The traps stimulated in nor-mal gravity demonstrated a normal closure in 80 of thecases and finally the traps stimulated in hypergravity reactedpromptly to the stimulation with 50of the traps being fasterthan controls (Table 1)

Table 1 Trap behaviour recorded under different gravity conditionsthe number of traps tested 119899 is reported in the table

119899No

responseNormalclosure

Slowclosure

Fastclosure

0G 25 36 16 48 01G 20 10 80 10 02G 8 0 50 0 50

In the graph (Figure 2) three representative exampleshave been reported

Trap closure is strongly affected by gravity in micrograv-ity the kinetics of snapping is slower (Figure 2(a)) and theacceleration is low if compared with 1 and 2 g where the speedof closure increases sharply after the trigger (Figure 2(b))

Because of the constraints involved in performing theexperiments on a plane we were unable to measure thereaction time between the trigger and the start of the closure(our time resolution was 40ms and the traps were stimulatedmanually) However visual observation revealed a delayedresponse in zero gravity and an anticipated response inhyper-g

4 Discussion

Volkov et al described the trap closure as consisting of threedifferent phases [9] (1) a mechanically silent period with noobservable movement immediately after stimulation (2) theperiod when the movement starts accelerating (3) the fastmovement of the trap when the leaves quickly relax to thenew equilibrium state In our results it appears that micro-gravity acts inDionaea at two different levels (i) by impairingthe signal transduction as suggested from the high per-centage of inactive traps and from the lower responsiveness(phase 1) and (ii) by altering the trap kinetics by signifi-cantly reducing the trap closing time (phases 2-3) eventuallysuggesting that the mechanisms leading to trap closure aregravity-related The electrical properties of excitable cells areextremely important in higher organisms Changes of theirparameters under microgravity can impair the functionalityof the neural systems and have significant consequences forhuman especially in view of long space travels The fewreports available on animal cells suggest that action potentialsare affected by gravity [16] in particular the propagationvelocity and their intensity seem to be gravity-dependentthat is they increase under hypergravity and decrease undermicrogravity compared to 1 g [17] Very little is known forhigher plants Masi et al monitored for the first time theelectrical activity of root cells during a parabolic flight andobserved alterations of the frequency of APs [18] Alteredparameters have been reported also under hypergravity (Masiet al unpublished) suggesting that the excitability of bothplant and animal cells is heavily affected by altered gravityconditions

InDionaeamuscipula it is well known that the stimulationof trigger hairs generates the twoAPs required for trap closure[9 19 20]Thehigh number of inactive traps (ie no responseto the trigger) and the apparent slower response time of


0g

0g

0g

0g

0g

0g

1g

1g

1g

1g

1g

1g

2g

2g

2g

2g

2g

2g

0ms

180ms

360ms

540ms

720ms

900ms

Figure 1 Closing of the trap in micro- (0 g) normal (1 g) and hypergravity (2 g)

the trap closure suggest an alteration in the generation orpropagation of the APs in microgravity In plants as well asin animals action potentials are induced by the fast openingand closing of ion channels whose functionality has beenlittle studied and understood in microgravity so far In fact

ion channels are integral membrane proteins and they couldbe affected either directly or indirectly by gravity Gravitycould directly affect the protein integrity whereas changesin the thermodynamical properties of the membrane couldhave an indirect effect on the ion channel functionality [21]


0 200 400 60000

05

10

Time (ms)

0g1g

2g

Trap

clos

ure (

yminusy

max

)

(a)

0 200 400 600 80000

01

02

03

04

05

Time (ms)

Spee

d (1

sminus1)

0g1g

2g

(b)

Figure 2 Effect of gravity on trap closure (a) Kinetics of trap closure under different gravity conditions 119910 is the distance between the edgesof the lobes (b) Dependency of the speed of trap closure on time after stimulation

In 2001 Goldermann and Hanke showed for the first timethat gravity influences the integral open state probabilityof ion channels providing a first explanation of the effectsof gravity on electrical signalling [22] Those findings werefurther confirmed by patch-clamp analysis [21]

Nothing similar has been done for plants The first silentstage of the trap closing involves transduction of electrical sig-nal and hence it is related to ion channel gating Interestinglyresults similar to the ones obtained here under microgravitywere observed when applying channel blockers to the traps[9]The use of BaCl2 ZnCl2 andTEACl significantly delayedtrap closure and altered its speed [9]

Of course the results presented here are just preliminaryFurther studies will be necessary to consolidate the resultsand to investigate in deeper detail the possible effect of gravityon the generation and propagation of action potentialsParticularly interesting would be to stimulate electrically thetraps allowing measuring and quantifying the delay in trapclosure under altered gravity conditions

To conclude our results demonstrate the role of micro-gravity on the events leading to trap closure The possiblealterations of ion and water channel permeability that couldbe at the base of the lower responsiveness and slow closureobserved in microgravity are a possibility worthy to be inves-tigated In fact if properly demonstrated it would strengthenthe validity of the hydroelastic curvature model suggested byVolkov et al [9] Finally we want to stress the fact that Venusflytrap could be an easy and effective model plant to performstudies on ion channels and aquaporin activities as well ason electrical activity in vivo on board of parabolic flights andlarge diameter centrifuges



References

[1] J Burdon-Sanderson ldquoOn the electrom otive properties of theleaf of dionaea in the excited and unexcited statesrdquoPhilosophicalTransactions of the Royal Society of London vol 173 pp 1ndash551882

[2] C Darwin and F Darwin Insectivorous Plants John MurrayLondon UK 1888

[3] B B E Juniper R J Robins and D M Joel The CarnivorousPlants Academic Press London UK 1989

[4] WH Brown andLW Sharp ldquoThe closing response in dionaeardquoBotanical Gazette vol 49 no 4 pp 290ndash302 1910

[5] A G Volkov T Adesina and E Jovanov ldquoClosing of venusflytrap by electrical stimulation of motor cellsrdquo Plant Signalingamp Behavior vol 2 no 3 pp 139ndash145 2007

[6] S E Williams and A B Bennett ldquoLeaf closure in the venusflytrap an acid growth responserdquo Science vol 218 no 4577 pp1120ndash1122 1982

[7] J Scala K Iott D Schwab and F Semersky ldquoDigestive secretionof Dionaea muscipula (venusrsquos flytrap)rdquo Plant Physiology vol44 no 3 pp 367ndash371 1969

[8] K Trebacz and A Sievers ldquoAction potentials evoked by light intraps ofDionaea muscipula ellisrdquo Plant and Cell Physiology vol39 no 4 pp 369ndash372 1998

[9] A G Volkov T Adesina V SMarkin and E Jovanov ldquoKineticsand mechanism of dionaea muscipula trap closingrdquo PlantPhysiology vol 146 no 2 pp 694ndash702 2008

[10] A Pavlovic L Slovakova C Pandolfi and S Mancuso ldquoOn themechanism underlying photosynthetic limitation upon triggerhair irritation in the carnivorous plant Venus flytrap (Dionaeamuscipula Ellis)rdquo Journal of Experimental Botany vol 62 no 6pp 1991ndash2000 2011

[11] A Pavlovic and SMancuso ldquoElectrical signaling and photosyn-thesis can they co-exist togetherrdquo Plant Signaling and Behaviorvol 6 no 6 pp 840ndash842 2011

[12] D Hodick and A Sievers ldquoOn themechanism of trap closure ofVenus flytrap (Dionaea muscipula Ellis)rdquo Planta vol 179 no 1pp 32ndash42 1989


[13] B S Hill and G P Findlay ldquoThe power of movement in plantsthe role of osmotic machinesrdquo Quarterly Reviews of Biophysicsvol 14 no 2 pp 173ndash222 1981

[14] Y Forterre JM Skotheim J Dumals and LMahadevan ldquoHowthe Venus flytrap snapsrdquoNature vol 433 no 7024 pp 421ndash4252005

[15] C A Schneider W S Rasband and K W Eliceiri ldquoNIH imageto imageJ 25 years of image analysisrdquo Nature Methods vol 9no 7 pp 671ndash675 2012

[16] M Wiedemann F P Kohn H Roesner and W R HankeldquoBehavior of action potentials under variable gravity condi-tionsrdquo in Self-Organization and Pattern-Formation in NeuronalSystems under Conditions of Variable Gravity pp 95ndash109 2011

[17] K Meissner and W Hanke ldquoAction potential properties aregravity dependentrdquoMicrogravity Science and Technology vol 17no 2 pp 38ndash43 2005

[18] E Masi M Ciszak S Mugnai et al ldquoElectrical network activityin plant roots under gravity-changing conditionsrdquo Journal ofGravitational Physiology pp 167ndash168 2008

[19] D Hodick and A Sievers ldquoThe action potential of Dionaeamuscipula Ellisrdquo Planta vol 174 no 1 pp 8ndash18 1988

[20] W H Brown ldquoThe mechanism of movement and the durationof the effect of stimulation in the leaves of dionaeardquo TheAmerican Journal of Botany vol 3 no 2 pp 68ndash90 1916

[21] M Wiedemann F P Kohn H Roesner and W R HankeldquoInteraction of gravity with molecules and membranesrdquo in Self-Organization and Pattern-Formation in Neuronal Systems underConditions of Variable Gravity pp 57ndash93 2011

[22] M Goldermann and W Hanke ldquoIon channel are sensitive togravity changesrdquo Microgravity Science and Technology vol 13no 1 pp 35ndash38 2001

Review ArticleThe Impact of Simulated and Real Microgravity onBone Cells and Mesenchymal Stem Cells

Claudia Ulbrich1 Markus Wehland2 Jessica Pietsch2 Ganna Aleshcheva2 Petra Wise3

Jack van Loon456 Nils Magnusson7 Manfred Infanger2 Jirka Grosse8 Christoph Eilles8

Alamelu Sundaresan9 and Daniela Grimm10

1 Department of Physiology Membrane Physiology University of Hohenheim 70593 Stuttgart Germany2 Clinic for Plastic Aesthetic and Hand Surgery Otto-von-Guericke University 39120 Magdeburg Germany3HematologyOncology Childrenrsquos Hospital Los Angeles University of Southern California Los Angeles CA 90027 USA4Department of Oral and Maxillofacial SurgeryOral Pathology VU University Medical Center Amsterdam1007 MB Amsterdam The Netherlands

5 Department of Oral Cell Biology Academic Centre for Dentistry Amsterdam (ACTA) University of Amsterdam andVU University Amsterdam 1081 LA Amsterdam The Netherlands

6 European Space Agency Technology Center Gravity Lab (ESA-ESTEC-TEC-MMG) 2201 AZ Noordwijk The Netherlands7Medical Research Laboratory Institute of Clinical Medicine Aarhus University 8000 Aarhus C Denmark8Department of Nuclear Medicine University of Regensburg 93052 Regensburg Germany9Department of Biology Texas Southern University 3100 Cleburne Houston TX 77004 USA10Institute of Biomedicine Pharmacology Aarhus University Wilhelm Meyers Alle 4 8000 Aarhus C Denmark

Correspondence should be addressed to Daniela Grimm danielagrimmfarmaudk

Received 4 April 2014 Revised 6 June 2014 Accepted 6 June 2014 Published 10 July 2014


Copyright copy 2014 Claudia Ulbrich et alThis is an open access article distributed under theCreative CommonsAttribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Howmicrogravity affects the biology of human cells and the formation of 3D cell cultures in real and simulatedmicrogravity (r- ands-120583119892) is currently a hot topic in biomedicine In r- and s-120583119892 various cell types were found to form 3D structures This review willfocus on the current knowledge of tissue engineering in space and on Earth using systems such as the random positioning machine(RPM) the 2D-clinostat or the NASA-developed rotating wall vessel bioreactor (RWV) to create tissue from bone tumor andmesenchymal stem cells To understand the development of 3D structures in vitro experiments using s-120583119892 devices can providevaluable information about modulations in signal-transduction cell adhesion or extracellular matrix induced by altered gravityconditions These systems also facilitate the analysis of the impact of growth factors hormones or drugs on these tissue-likeconstructs Progress has beenmade in bone tissue engineering using the RWV andmulticellular tumor spheroids (MCTS) formedin both r- and s-120583119892 have been reported and were analyzed in depth Currently these MCTS are available for drug testing andproteomic investigations This review provides an overview of the influence of 120583119892 on the aforementioned cells and an outlook forfuture perspectives in tissue engineering

1 Introduction

It is well known that microgravity influences different bio-logical systems like bone and muscle as well as the heart andbrain and it enhances cancer risk [1] During their stay atthe MIR astronauts and cosmonauts did show a distinct lossof bone mineral density in the lumbar spine the pelvis and

the proximal femur [2] and the extent of bone loss varied upto 20 [3]

As it is not feasible to gather enough material fromastronauts to do in-depth investigations another device hasbeen developed for the International Space Station (ISS) themice drawer system (MDS) as a facility to study long-timeinfluence of radiation on the biology and behavior of mice



Tavella et al for example report an altered bone turnoverin different strains of mice which were kept on the ISS for91 days This resulted in bone loss due to increased boneresorption and a decreased bone deposition [4]

While the past biological physiological and medicalresearch nearly exclusively focused on investigating the bio-chemical processes of living cells and organisms more andmore attention was paid to the biomechanical properties andmechanical environment of cells and tissues during the lastdecades When culturing cells on Earth they usually settleon the bottom of the culture flask forming two-dimensional(2D) monolayers A three-dimensional (3D) growth moreresembling the tissue environment found in living organismsis prevented by the presence of the gravitational field Fora scaffold-free 3D tissue growth it is therefore necessary tocircumvent this problem by effectively eliminating the influ-ence of the gravitational pull during cultivation One of thebyproducts of various space flight endeavors is the possibilityto perform long-term near-weightlessness or microgravity(120583119892) experiments [5 6] In a 120583119892 environment cells will notsettle like on Earth This provides an increased opportunityfor freely floating cells to interact with each other and develop3D structures [7]

2 Space Flights forCell-Biological Experiments

Long-term orbital space flight experiments are however nottrivial Flight opportunities are very scarce and the costs ofhardware development are high Furthermore science is notalways a priority in space flight activities Such preconditionsare delaying the advancement of research in areas such as cellbiology and tissue engineering disciplines which could profittremendously from more frequent research options in a realmicrogravity (r-120583119892) environment

Some researchers recently pointed out that osteoblastsundergo a disintegration of their cytoskeleton which mayexplain dramatic changes in size and shape of the cells andtheir surface specializations [47] Also other studies havebeen performed using the ISS or space shuttle flights to learnmore about the behavior of bone cells in space [48] but flightopportunities are sparse and therefore other platforms hadto be elucidated

It is due to the aforementioned limitations that over theyears various devices have been developed in an attemptto reduce the impact of gravity and simulate a near-weightlessness environment (s-120583119892) on Earth From a physicalpoint of view gravity is a force exhibiting bothmagnitude anddirection Therefore the influence of gravity can be reducedby either manipulating magnitude or direction An orbitalspace flight as on the ISS is physically identical to a free-fall Here the gravitation acts in a perpendicular manneron the spacecraftrsquos velocity vector effectively changing itsdirection constantly but not affecting its magnitude Free-fallis also found when using sounding rockets which provider-120583119892 during a time span of up to 15 minutes On Earthr-120583119892 can also be attained although only for periods inthe range of seconds in drop towers and during parabolic

flights missions [49 50] Although time periods of secondsor minutes limit their use for tissue engineering studiessuch periods can be useful to explore various intra- andintercellular processes responsible for gene expression andprotein content changes which can be observed after only afew hours of culturing cells in 120583119892 [49ndash51]

3 Devices Simulating Microgravity on Earth

In this respect we should mention an instrument that wasintroduced by the European Space Agency (ESA) in the earlynineties called the free fall machine (FFM) [52] This instru-ment was specifically developed for biological experimentsand could generate a free fall for a period of about 800mswith an intermediate ldquobouncerdquo ofsim20 g for around 50msTheparadigm of the FFM is that cells might not be sensitive tothe relatively short period of 50ms of hypergravity while theyexperience the relatively longer period of free-fall Long-termexperiments (hours days) which might be useful for tissueengineering studies could be performed on this platformHowever thus far only two studies were published usingthe FFM one investigatingChlamydomonas [53] and anotherone researching T-lymphocytes [54] The Chlamydomonasstudy showed similar results to what was found in realspace flight while the T-lymphocytes experiments did notConsidering the very limited number of studies performedonthis ground-based device the FFM still might deserve somemore exploration

Levitating magnets are also used to produce s-120583119892 onEarth Such systems compensate themagnitude of the gravityvector by preventing sedimentation of relatively heavy struc-tures like cells by the application of a high gradientmagneticfield This principle was first described for biological systemsby Berry and Geim in 1997 [55] who demonstrated that atoad could be levitated and survivewhile exposed to a 16 Teslamagnetic field Various experiments in cell biology havemadeuse of such systems [56ndash58] The magnetic field acts on indi-vidual molecules and atomswithin a cell based on theirmag-netic susceptibility preventing them from sedimentationHowever the magnetic field as such confounds possible s-120583119892effects The direction of the field might force (bio-)polymersinto a certain orientation Different polymers within a cellor on the cell membrane have different susceptibilitiespossibly producing artifacts by forcing polymers into specificarrangements which may not reflect the actual physiologicalsituation [59ndash61] Superconducting high gradient magnetsare especially capable of performing long-term experimentsand might be useful in the area of tissue engineering [62ndash64] In this context another promising technique should bementioned This method is the use of magnetic particles for3D cell cultures It is not based on a high-gradient magneticfield but on ferromagnetic particles attached to cells whichcan subsequently be levitated by a conventional magnetfacilitating the formation of 3D structures [65 66]

Another option is to manipulate the direction of thegravity vector with respect to the sample The reduction ofthe gravitational impact on biological systems by constantlychanging its orientation was shown first in experiments by


the German botanist von Sachs in 1879 growing Lepidiumsativum and Linumusit [67] He constructed a slowly rotatingsystem and named it a clinostat in which for example a plantcan be placed horizontally and rotated around its longitudinalaxis In doing so the gravity vector stimulus is constantlychanging its impact angle on the sample As a result aplant grows straight without the characteristic gravitropiccurvature seen when the plant is placed horizontally and notrotating Based on these initial studies other rotating systemslike the fast rotating clinostat have been developed

The initial clinostats were rotating relatively slowly in arange fromone rotation per couple of hours up to amaximumof about 10 rpmThis is adequate for relatively ldquosolid samplesrdquosuch as plants but too slow for cell culture systems thatinvolve a large liquid phase In a biphasic system that isa liquid with particles (cells) both of different density theheavy particles tend to settle Rotating such a system around ahorizontal axis keeps the heavy particles in suspension Thisphenomenon depends mainly on the relative density of theliquid and the particles the viscosity of the liquid the rotationspeed and the diameter of the rotated container When a cellis in a static vessel and the vessel is rotated by 90∘ the cell willsettle in the direction of the gravity vector One can repeatthis for a full 360∘ and upon an increase in the frequency ofrotation the traveling distance of the cell decreases If thisrotation is performed constantly with increased speed wefinally end up rotating a cell around its own axis Such acontrolled rotation not only applies to the cells but also itssurrounding boundary liquid phase [68]

Another well-known device to simulate 120583119892 is the so-called random positioning machine (RPM) a 3D clinostat[69] consisting of two frames each driven by a dedicatedmotor This allows a randomized movement of both framesindependent of each other [69ndash74] One of the advantages ofthe RPM is its size as cell culture flasks can easily bemountedon it so it is possible to work with quite large liquid volumesThis ranges from regular T25 flasks [75 76] to multi-wellplates [77] flasks on slides [78] or more dedicated devices[79] As cells move freely within the liquid they usuallyinteract with each other and form multicellular spheroids

The best simulation of 120583119892 is achieved in the rotationcenter of the two axes which limits the preferred volume sizeof the samples Depending on the speed of rotation and thedistance from the center an acceptable residual gravity canbe obtained in the order of 10minus4 g by a maximum angularvelocity of 60∘ sminus1 at a radial distance of 10 cm [70] EarlierRPM models had no possibility to add constituents duringthe experiment but newer models have been developed toenable fluid management during rotation [73 74] RPMsare commercially available by Mitsubishi Heavy Industries(Kobe Japan) and Dutch Space (Leiden The Netherlands)while various academic groups developed similar systems ontheir own [80ndash84] (Figure 1)

The rotating wall vessel (RWV) prevents cells from set-tling via a constant rotation It has been developed by NASA[85] and is now commercially available through SyntheconInc (Houston TX USA) Basically RWVs consist of aslow rotating relatively large liquid filled container (vessel)

The rotation speed has to be adapted to the specific weightof the cells the fluid density and viscosity The cells andtissues in the RWV are constantly falling within the fluidThesettling velocity and direction combined with the rotation ofthe fluid create spiral trajectories within the vessel [86] Thismotion of the sample relative to the fluid generates fluid shearforces on a particle surface ranging from 180 to 320mPa (18ndash32 dynecm2) for 50 120583m beads [87] sim500mPa (5 dynecm2)with 3D aggregates of BHK-21 cells [88] to 520ndash780mPa (52ndash78 dynescm2) for a 200 or 300120583mspherical object [89] Overthe years variousmodels based on the initial RWVhave beendeveloped differing in vessel geometry aspect ratio and gassupply such as the slow turning lateral vessel (STLV) [90]the high aspect ratio vessel (HARV) [91] or the rotating-wallperfused vessel (RWPV) [92]

Hence it can be concluded that annulling the gravityforces which pull the cells constantly towards the Earthdeliver the ultimate trigger to eukaryotic cells to leave a cellmonolayer and assemble in 3D aggregates [5]

It is still unknown which cellular and biochemical mech-anisms are involved in the altered signal transduction and inthe change of the cellular growth behavior

4 Transition from Two- toThree-Dimensional Cell Growth

A few publications appeared in the literature in recent yearsproviding some clues for understanding the weightlessness-induced transition from two- (2D) to three-dimensional (3D)cell growth

Several signaling pathways are affected by annullinggravity forces in the cell interior [93] However it is unknownwhich of these signaling pathways contribute to the formationof three-dimensional aggregatesWhen endothelial cells formtubes the nitric oxide signaling pathway appears to beaffected [94] Siamwala et al reported that iNOS (induciblenitric oxide synthase) acts as a molecular switch whichcontrolswhether the effects of120583119892on vascular endothelial cellsinduce angiogenesis via the cyclic guanosinemonophosphate(cGMP)-PKG-dependent pathway [94] iNOS is upregulatedinHUVECby amechanismdependent on suppression ofAP-1 after clinorotation of the cells [95] In addition the endothe-lial nitric oxide synthase is phosphorylated by phosphoinosi-tide 3-kinase under weightlessness simultaneously with Akt[96] The organoid formation by PC12 pheochromocytomacells in a RWV bioreactor is accompanied by prolongedactivation of the ERK p38 and jnk signaling pathways [97]

3D cell culture techniques have attracted much attentionnot only among biologists but also clinicians interested intissue engineering [98 99] of artificial vessels [100ndash104] orcartilage [105ndash108] Moreover osteoarthritis and cartilagetrauma occur in patients with a high incidence but currenttreatmentmethods are still limited [109] Even aminor injuryto articular cartilage may lead to progressive damage anddegeneration [110]


(a) (b) (c)

Figure 1 (a) Two 2D clinostat devices in an incubator constructed by the German Aerospace Center (DLR) Institute of AerospaceMedicineBiomedical Science Support Center Gravitational Biology Cologne Germany (b c) Random Positioning Machine simulating microgravityIt was developed by T Hoson in Japan and manufactured by Dutch Space (former Fokker Space)The basic principle consists of an inner andan outer frame rotating independently from each other in random directionThe samples in the center of the machine experience low gravityas the gravity vector is averaged to zero over time The redesign of the classical RPM with a CO

2-Incubator with temperature and CO

2-level

control was realized by Professor Jorg Sekler Fachhochschule Nordwestschweiz (FHNW) Institut fur Automation Switzerland and tested byPD Dr Marcel Egli Hochschule LuzernmdashTechnik amp Architektur CC Aerospace Biomedical Science amp Technology Hergiswil Switzerland

5 Tissue Engineering of Bone

Bone loss has been documented for many years in 120583119892 (1-2 a month) Increased bone loss and risk of fractures isan identified risk in the bioastronautics critical roadmap forlong-term cosmic missions to the moon and mars In vitrodrug screening both in 1 g 120583119892 and in artificial gravity isessential to adequately address countermeasures for boneloss Bone loss in 120583119892 is the second most important risk tospace missions [5 6]

Exposure to the 120583119892 environment of space causes astro-nauts to lose calcium from bones [5 6] This loss occursbecause the absence of Earthrsquos gravity disrupts the processof bone maintenance in its major function of supportingbody weight Exposure to the 120583119892 environment of space causesmen and women of all ages to lose up to 1 of their bonemass per month due to disuse atrophy a condition similar toosteoporosis It is not yet clear whether loss in bone mass willcontinue as long as a person remains in the 120583119892 environmentor level off in time

There are indeed four major bone cell types and eachof them seems to be influenced by 120583119892 Bone mesenchymalstem cells (MSC) are able to differentiate into adipocytesosteoblasts and osteoclasts Proliferation and differentiationare very sensitive to 120583119892 as the lack of gravity in spacecan reduce mechanical stress leading to a decreased rate ofosteogenesis and an increased adipogenesis rate [111] As thesignaling pathways involved in MSC differentiation form acomplicated network it has been found that the reduction inthe osteogenesis ofMSCs in the presence of 120583119892 is mediated bya decrease in the integrinmitogen-activated protein kinase(MAPK) signaling pathway [112] as well as RhoA andcytoskeletal disruption [113]

Osteoblasts are derived from MSCs but in 120583119892 the dif-ferentiation does not function properly and the resultingbone loss has been attributed to osteoblasts due to their (1)reduced proliferation and activity (2) reduced differentiationand (3) decreased responsiveness to bone-related factors inthe microenvironment [114] Observations have also beenmade regarding the cytoskeleton of osteoblasts there isgrowing evidence that the cytoskeleton is closely connected tonuclear morphology and function [115] The enlarged nucleiobserved in flight osteoblasts could be a result of cytoskeletaldisruption [116]

Osteocytes regulate bone resorption and formationand are considered the terminal differentiation stage ofosteoblasts The osteocytes in cortical bone and periosteumdegenerated after a 125-day flight in space on the CosmosBiosatellite [117] Osteocyte apoptosis has been observedafter a 2-week flight increasing the number of functionallyactive osteoclasts [118] Apoptotic osteocytes are essential forthe initiation of bone remodeling but it is the neighboringnonapoptotic osteocytes that produce proosteoclastogenicsignaling [119] Osteocytes seem to be the key effectors of 120583119892induced bone loss [120]

Osteoclasts are bone-resorbing cells and their differenti-ation seems to be enhanced in 120583119892 [121]This could be anotherexplanation of bone-loss in space

Themystery for the moment is what signals permit bonetissue to adapt to a weightless or an Earth (1 g) environmentResearchers do not yet know whether the biomechanicalstimuli that are changed by 120583119892 directly affect osteoblastand osteoclast function or if other physiological factorssuch as hormone levels or poor nutrition contribute tobone loss NASA investigators are studying gravity-sensingsystems in individual bone cells by flying cultures of these


cells on the space shuttle and observing how they functionDiscoveries made in the course of space biomedical researchon bone are already contributing to a better understandingof osteoporosis and the treatment of bone mass loss on Earthas well as in space The single most important contributionthat NASA research has made to the understanding of bonedeterioration in osteoporosis is heightened awareness of theimportance of gravity activity and biomechanicsmdashthat is themechanical basis of biological activitymdash in bone remodeling

Mechanical forcesmdashthe action of energy on mattermdashappear to coordinate bone shaping processes The standardtheory of bone remodeling states the body translatesmechan-ical force into biochemical signals that drive the basic pro-cesses of bone formation and resorption Aging especially inpostmenopausal women and exposure to 120583119892 uncouple boneresorption and formation When this uncoupling occursformation lags behind resorption and the result is bone loss

Researchers are not yet certain whether bone resorptionspeeds up or the bone formation slows down though recentexperimentation in space indicates that 120583119892 might somehowaffect both processes Progress in developing methods of pre-venting or treating disuse atrophy and osteoporosis dependson better understanding of the mechanisms that cause theproblem Determining how the body translates mechanicalloading (physical stress or force) into the signals that controlbone structure may reveal how aging inactivity and spaceflight uncouple bone formation and resorption Only in theabsence of gravity can we determine the influence of weightand stress on bone dynamics

By studying whatmechanisms translatemechanical stresson bones into biochemical signals that stimulate bone for-mation and resorption space life scientists may be able todetermine how tomaintain bonemass Researchers donot yetknow exactly what type and amount of exercise hormonesor drugs might prevent bone loss or promote bone forma-tion However some combination of sex hormones growthhormones and exercise seems to be the key to preventingbone mass loss associated with chronological aging andpostmenopausal hormone changes on Earth

Bone is made up of several different cell populationsOsteoclasts are responsible for the breakdown of mineralizedbone in preparation for bone remodeling In contrast theosteoblasts synthesize mineralized bone in the remodelingprocess The goal of this project is to develop an ldquoinvitrordquo three-dimensional cellular model of osteoclasts andosteoblasts (human and rodent) cultured together in 120583119892analog culture conditions to identify the underlying biomark-ers related to bone loss in 120583119892 and the cellular mechanismsinvolved in bone resorption The NASA rotating-wall vessel(RWV) permits the growth of mixed cell cultures for muchlonger periods than traditional culture methods This wouldset the stage for development of countermeasure strategies forbone loss in space as well as in osteoporosis and rheumatoidarthritis which are increased health risks on Earth ProfessorSundaresan and collaborators [122ndash124] have developed a 3Dcell culture bone tissue model using a specialized rotating-wall vessel culture system to address a more physiologicallyrelevant model to the human body The use of the cells by

themselves also eliminates confounding variables such asneuroendocrine stress found in vivo (Figure 2(a))

The human body needs a framework to withstand gravityThis framework is given by the skeletal system Duringlong-term space missions bone loss has been reported inastronauts at a rate that is both substantial and progressivewith time spent in 120583119892 [125ndash128] But what is the reason forthismassive bone loss Some studies suggested that this effectmight be attributed to increased resorption in load-bearingregions of the skeleton [129ndash131] and evidence of a decreasein bone formation had also been described For example theloss of bone in 120583119892 is about 10 times greater than the bonemineral density loss per month of postmenopausal womenon Earth who are not on estrogen therapy [132ndash135]The lossof bone mineral density in a six-month mission appeared tobe reversible in 1000 days after return to Earth [136 137] butchanges in the bone structure are irreversible and seem tomimic changes in the elderly [137]

Until now there are still knowledge gaps on the mecha-nism of bone loss especially on the molecular and cellularmechanisms also the question of fracture repair arisesMoreover more information is needed on the influence ofradiation hormones and fluid shifts

Investigations in humans and animals are quite difficultdue to the lack of long-term flight opportunities the absenceof animal housing facilities in space and the problem ofmaterial collection from returning astronauts Thus otherpossibilities have to be sought in order to investigate bone Sofar most commonly used are bone cell culture experimentswhich are a viable opportunity for investigating cells in 3Dacting as tissue like samples while they are cultivated underconditions of weightlessness However 3D embryonic bonetissue cultures have been used in the past and show a cleardecrease in matrix mineralization in mineralizing cartilageand by osteoblasts combined with an increased mineralresorption by osteoclasts [138]

Besides this tissue engineering is a very up-to-date topicThe ultimate goal is to generate functional 3D constructswhich can be used as replacement organs or structures withnormal function or serve for in vitro studies [5 139] Bonereplacement especially is quite difficult as large bone defectsusually require reconstructive surgery to restore function[140] Up to date the treatment includes autograft or allografttransplantation and the use of syntheticmaterials [141]Whileautograft transplantation is the preferred treatment it suffersfrom limited supply and donor site morbidity [142] Asthe autogenous origin of cells prevents potential immunerejection the amount of bone marrow suitable for trans-plantation is limited New techniques have been developedallowing selection of bone marrow osteoprogenitor cells andexpanding them in culture so that a large amount of trans-plantable cells can be generated after only one biopsy [143ndash145]

In principle culturing bone cells is not that easy A combi-nation of osteoconductive matrices bone-forming cells andosteogenic growth factors is needed for the engineering ofbone tissue [146] The first important factor is the cell typeOsteoblasts are in a close to mature stage showing a lowproliferative potential Mesenchymal stromal cells represent


(a)

TC 4d RPM MCTS

(b)

Figure 2 (a) Production of large numbers of small (200 120583m diameter) immature (7-day-old) osteospheres with labeled osteoclast cells (red)viewed by confocal imaging in living constructs-USPTO 80736136 and (b) follicular thyroid cancer cells (TC) cultured on the RPM Severalmulticellular tumor spheroids are visible after 4 days

a proliferating and undifferentiated cell source but theiravailability is limited [147 148] An option to increase theirlifespan in vitro is the overexpression of human telomerasereverse transcriptase (hTERT) The second factor is an idealscaffold which possesses mechanical properties comparableto bone It should support cell adhesion and should bebiodegradable to facilitate natural bone remodeling [146]As of now different studies have shown the advantagesand disadvantages of several types of scaffolds like chitingelatin poly(lactic acid) poly(glycolic acid) poly(lactic acid-co-glycolic acid) polycaprolactone hydroxyapatite coraland so forth Several in vitro studies revealed an ideal scaffoldpore size for osteoblasts from 200 to 400 120583m [149 150] It isimportant to recognize that the scaffold architecture influ-ences the distribution of shear stress the range of mechanicalstimuli as well as the proliferation and differentiation ofosteoprogenitor cells [151 152]

To simulate an ideal in vivo situation for in vitro cellsspecific cytokines and growth factors are necessary For bonemorphogenesis the bone morphogenetic proteins (BMP)which belong to the transforming growth factor beta (TGF-120573) superfamily are essential [153] Currently only BMP-2and -7 are commercially available so alternatives to stimulateosteoprogenitor cells by growth factors are required It hasbeen reported that autologous platelet-rich plasma is aneffective bioactive supplement as it contains osteogenic andangiogenic growth factors [154]

Several different bioreactor systems are already availablefor bone tissue engineering Awell-known and simple systemis the spinner flask bioreactor Convective forces are providedby a stirrer and the medium flows around the cells Theemerging shear stress is not applied homogenously as thereappears to form a gradient in the flask [146] This factorcertainly needs to be considered when conducting studieswith the spinner flask system

Other suitable instruments are rotating bioreactor sys-tems for example the RWV It has been used with different

kind of bone cells which are often grown with the help ofmicrocarriers [8 155] or scaffolds [8ndash11 15 155] The highaspect ratio vessel (HARV) [91] was used by Lv et al [12]to engineer tissue on poly(lactic acid glycolic acid)nano-hydroxyapatite composite microsphere-based scaffolds

Some researchers used bone marrow mesenchymal stemcells for their investigations Jin et al [16] were able to trans-plant RWV-grown bone constructs in cranial bone defectsof Sprague-Dawley rats and found them to be more effectivein repairing the defects than the 1 g controls after 24 weeksMoreover a 3D environment as in a rotary cell culture systemenhanced osteoblast cell aggregation andmineralization [13]Preosteoblasts cultured in a RWV could be engineered intoosseous-like tissue [14]

6 Mesenchymal Stem Cells and Microgravity

Mesenchymal stem cells (MSCs) are cells capable of long-term proliferation and differentiation into various stromaltissue cell types The state of MSCs rests on the cellularmicroenvironment and several soluble factors In additiongravity can influence MSC features Disuse as encounteredduring long-term bed-rest or space travel and the accom-panying absence of mechanical stimuli lead to an inhibitionof osteogenesis and simultaneously to an induction of adi-pogenesis in MSCs Hence it is crucial to provide a propermechanical stimulation for cellular viability and osteogenesisparticularly under unusual conditions

In 2004 Merzlikina et al [27] studied the effects of pro-longed clinorotation on cultured human MSC morphologyproliferation rate and expression of specific cellular markersAfter exposure of the cells to clinorotation for time framesfrom 1 h to 10 days it was shown that the proliferativerate decreased in the experimental cultures as compared tocells growing under normal conditions Clinorotated MSCsseemed more flattened and reached confluence at a lower


cell density which advocates that cultured hMSCs sense thechanges in the gravity vector and respond to s-120583119892 by alteredfunctional activity The group around Myoui [28] examinedwhether gravity-induced stress is linked to osteoblast dif-ferentiation and function Rat marrow mesenchymal cells(MMCs) were cultured in pores of interconnected porouscalcium hydroxyapatite (IP-CHA) for 2 weeks on a 3D clino-stat In MMCs subjected to s-120583119892 the marker of osteoblasticdifferentiation alkaline phosphatase activity was decreased by40 compared to the control group Also the clinostat groupexhibited less extensive extracellular matrix formation thanthe control group The implantation of the IP-CHAMMCcomposites in syngeneic rats showed that bone formationwas significantly lower for the clinostat group than for thecontrol group Yuge et al [29] also used a 3D clinostat fortheir experiments on the proliferation behavior of hMSCsThe proliferation rate of the cells of the clinostat group waselevated almost 3-fold in comparison to the control groupand the number of hMSCs double-positive for CD44CD29or CD90CD29 in the clinostat group after 7 days in cultureincreased 6-fold The hMSCs cultured in a 3D-clinostat werestill able to differentiate into hyaline cartilage after trans-plantation into cartilage defective mice and displayed thestrong proliferative characteristic of stem cells thus showingthat s-120583119892 may be used to expand stem cell populations invitro In contrast to these findings Dai et al [24] reportedin 2007 that 120583119892 simulated by a clinostat inhibited populationgrowth of bone marrow mesenchymal stem cells (rBMSCs)and their differentiation towards osteoblasts The cells grownon the clinostat were arrested in the G(0)G(1) phase ofcell cycle and growth factors such as insulin-like growthfactor-I epidermal growth factor and basic fibroblast growthfactor had only a slight stimulatory effect compared to thestatic control group Gershovich and Buravkovarsquos [17] worksupports this hypothesis After 20 days of clinostat-exposurethe proliferative activity of hBMCs was reduced whereas itincreased the number of large flat cells in the culture andstimulated migration activity of cells In 2009 Gershovichand Buravkova [30] demonstrated the effects of s-120583119892 byclinostat and RPMon the interleukin production by hBMSCsand MSC osteogenous derivatives 20-day exposure on aclinostat increased the interleukin-8 (IL-8) content 14 to 32times in the culturemedium while the average increase of IL-production on the RPM amounted to 15ndash6 times (10 days)and 16ndash21 times (20 days) respectively This suggests thatresults of s-120583119892 vary by the use of different modeling systemsrMSCs grown in a clinostat demonstrate that s-120583119892 can boostthe differentiation of MSCs into neurons as demonstratedby Chen et al [156] In s-120583119892 neuronal cells derived fromrMSCs were found to express higher microtubule-associatedprotein-2 (MAP-2) tyrosine hydroxylase (TH) and cholineacetyltransferase (CHAT) Furthermore the excretion ofneurotrophins such as nerve growth factor (NGF) brainderived neurotrophic factor (BDNF) or ciliary neurotrophicfactor (CNTF) was increased In comparison to 1 g controlsneuronal cells from the s-120583119892 group generated more matureaction potentials and displayed repetitive action potentialsThis might benefit the search for new strategies for thetreatment of central nervous system diseases

Zayzafoon et al [18] demonstrated that s-120583119892 inhibitsthe osteoblastic differentiation of hMSC and induces thedevelopment of an adipocytic phenotype In the effort ofunderstanding space flight-induced bone loss the group usedthe rotary cell culture system (RCCS) to model 120583119892 and deter-mine its effects on osteoblastogenesis Human MSCs werecultured and osteogenic differentiation was induced beforethe initiation of s-120583119892 As a result the important mediator ofadipocyte differentiation peroxisome proliferator-activatedreceptor gamma (PPARgamma2) and adipsin leptin andglucose transporter-4 was highly expressed These changeswere not adjusted after 35 days of readaptation to normalgravity Moreover 120583119892 decreased ERK- and increased p38-phosphorylation pathways known to regulate the activityof runt-related transcription factor 2 and PPARgamma2These results were supported by Saxena et al [19] in 2007who demonstrated that s-120583119892 inhibited osteoblastogenesisand increased adipocyte differentiation in hMSCs incubatedunder osteogenic conditions using the RCCS They couldshow that a reduced RhoA activity and cofilin phosphoryla-tion disruption of F-actin stress fibers and decreased inte-grin signaling through focal adhesion kinase were involvedin this process Meyers et al [20] also investigated the effectsof s-120583119892 on integrin expression and function in hMSCssince a reduced osteoblastic differentiation might be causedby impaired type I collagen (Col I)-integrin interactionsor a reduction of integrin signaling Culturing of hMSCsfor 7 days in s-120583119892 lead to reduced expression of Col Iwhile Col I-specific alpha2 and beta1 integrin protein expres-sion increased However autophosphorylation of adhesion-dependent kinases focal adhesion kinase (FAK) and proline-rich tyrosine kinase 2 (PYK2) was significantly reducedThese findings indicate that a reduction in osteoblastogenesisin s-120583119892 is at least in part caused by a reduced integrinMAPKsignaling The group around Duan [16] studied the relation-ships between the composition and mechanical properties ofengineered bone constructs BMSCs were grown for 15 dayson ceramic bovine bone scaffolds in different environmentsnamely static flasks and the RWV DNA content and alkalinephosphatase (ALP) were higher for cells grown on the RWVAfter transplantation into Sprague-Dawley rats with cranialbone defects the bone constructs engineered on the RWVrepaired the defects better and showed histologically betterbone connection

Sheyn et al [21] evaluated the effect of s-120583119892 on all genesexpressed in hMSCs with the hypothesis that many impor-tant pathways are affected during culture on a rotating wallvessel systemThe analysis of gene expression by use of wholegenome microarray and clustering showed that 882 geneswere downregulated and 505 genes were upregulated afterexposure to s-v A multitude of genes belonging to cell com-partment biological process and signaling pathway clusterswere modulated as identified by gene ontology clusteringSignificant reductions in osteogenic and chondrogenic geneexpression and an increase in adipogenic gene expressionwere shown and could be validated by a parallel adipogenicdifferentiation assay In order to investigate the effects ofs-120583119892 on chondrogenic differentiation of human adipose-derived MSCs (ADSCs) Yu et al [22] cultured cells on


a RCCS in pellets with or without the chondrogenic growthfactor TGF-120573

1 Analysis of real-time PCR and histological

results demonstrated that s-120583119892 has a synergistic effect onchondrogenesis with TGF-120573

1 The p38 MAPK pathway was

activated by TGF-1205731alone and was additionally stimulated

by s-120583119892 Inhibition of p38 activity with SB203580 suppressedchondrocyte-specific gene expression andmatrix productionThis indicates that the p38 MAPK signal mediates s-120583119892-induced chondrogenesis of ADSCs InMSCs cultured duringchondrogenic induction in a rotating culture combined withpolyglycolic acid (PGA)mRNAand proteins of collagen typeII and aggrecan were significantly more expressed in the s-120583119892 rotating culture group than the static culture group asreported by Wu et al [25] Zhang et al [26] described thatMSCs spread out in a spindle shape when cultured in normalgravity while they become unspread and round under s-120583119892 Also under s-120583119892 their cytoskeleton fibers are beingreorganized The function of MSCs was affected by thesemorphological changes transmitted through the activity ofRhoA To test the hypothesis that s-120583119892 has the capacity tooffer a novel choice in the stimulation of neovascularizationMSCs were cultured under s-120583119892 stimulation followed byVEGFdifferentiationThe responses revealed thatMSCsweredifferentiated into endothelial-like cells after 72 h incubationand were able to form a capillary network Their endothelialdifferentiation potential improved compared with the staticcontrol group

Another approach of modeling 120583119892 in hMSCs is the use ofa large gradient high magnetic field (LGHMF) produced bya superconducting magnet Shi et al [64] analyzed the effectsof LGHMF-120583119892 on survival cytoskeleton and osteogenicpotential of hMSCs Results showed that the LGHMF-120583119892treatment disrupted the cytoskeleton of hMSCs a LGHMF-120583119892 treatment for 24 h led to cell death LGHMF-120583119892 treat-ments in early stages of osteogenic induction resulted in sup-pression of osteogenesis of hMSCsThe suppression intensitywas reduced gradually as the treatment stage of LGHMF-MG was postponed A LGHMF-120583119892 treatment during theending-stage of osteogenic induction had no visual effecton osteogenesis of hMSCs which indicates that LGHMF-120583119892affects the initiation of osteogenesis

Furthermore a study of Uddin and Qin [23] examinedthe effects of low intensity pulsed ultrasound (LIPUS) onthe osteogenic differentiation of adipose-derived human stemcells (Ad-hMSC) under s-120583119892 conditions Microgravity wassimulated in a 1D clinostat and treated with LIPUS at30mWcmminus2 for 20min dayminus1 Hypothetically the applica-tion of LIPUS to s-120583119892 cultures would restore osteogenesisin Ad-hMSCs The results showed significant increases inALP OSX RANKL and RUNX2 and decreases in OPGgene expression in LIPUS treated SMG cultures of Ad-MSCcompared to nontreated cultures LIPUS also restored OSXRUNX2 and RANKL gene expression in osteoblast cells s-120583119892 significantly reduced ALP positive cells by 70 (119875 lt 001)and ALP activity by 22 (119875 lt 001) while LIPUS treatmentrestored ALP positive cell number and activity to equivalencewith normal gravity controls Extracellular matrix collagenand mineralization was assessed by Sirius red and Alizarinred staining respectively s-120583119892 cultures showed little or no

Table 1 Comparative methods of 3D cell culture systems usingsimulated 120583119892

Device Working principle

Free fall machine FFM Free fall for 800ms ldquobouncerdquoof 20 g for 50ms

Levitating magnets LM A high gradient magnetic fieldprevents sedimentation

2D-clinostat Rotation along one axisRandom positioningmachine RPM Two frames with randomized

movement

Rotating wall vessel RWV Constant rotation preventscells from settling

collagen or mineralization but LIPUS treatment restoredcollagen content to 50 (119875 lt 0001) and mineralization by45 (119875 lt 0001) relative to s-120583119892mdashonly cultures

7 Multicellular Tumor Spheroids

3D growth of tumor cells creating MCTS in vitro has beenobserved in various tumor cell lines including thyroid andcolorectal cancer [31] MCTS mimic the growth of solidtumors and represent a simple model approaching someof the characteristics found in vivo including physiologicalcharacteristics such as multicellular architecture and naturalbarriers of mass transportation Therefore the use of MCTSas an in vitro tool for testing anticancer drugs has gainedsignificant interest as MCTS potentially provide a morereliable model for drug testing compared to single layeradherent cell cultures During the approval process of drugsbefore clinical testing in trials the mechanisms of deliveryand the effectiveness of the drugs must be determined Thefirst steps of preclinical drug testing are typically carriedout using adherent cell formats growing in two dimensions[157] However the outcome of such investigations in two-dimensional cell systems is often very different from what isobserved in a whole-body situation This makes it difficult todraw clear conclusions of the drug properties anticipated invivo In terms of drug delivery a spheroid test platform hasinherent advantages providing a natural barrier resemblingthe natural tumor environment Spheroids of a particularsize exhibit certain gradients of oxygen and nutrition [35ndash37 158 159] Spheroids larger than 400ndash500120583m in diametershow characteristics of layered structures with a hypoxiccore consisting of necrotic cells surrounded by quiescentcells and an outer layer of proliferating cells [38 160 161]Hence 3D tumor cell systems are a valuable tool for studyingdrug delivery and the response and metabolism of hypoxictumor cells to cancer therapy Fang et al reported thatmulticellular spheroids of primary human colon cancercells were resistant to chemotherapy-induced apoptosis andretained the expression of colon cancer marker CD133mimicking colorectal cancer [162] Were these cells grownunder normal conditions they did not retain these char-acteristics Size control of MCTS is a major challenge inobtaining uniform and reliable high throughput test systems


Table 2 Overview of various cell types used for tissue engineering and 120583119892-conditions involved

Cell type Engineered tissue Method ReferencesOsteoblasts Bone RWV bioreactor (s-120583119892) [8ndash12]Osteoprogenitor cells Bone RWV bioreactor (s-120583119892) [13 14]Mesenchymal stem cells Bone RWV bioreactor (s-120583119892) [12 15ndash23]Mesenchymal stem cells Divers RWV bioreactor (s-120583119892) [18 21 24ndash26]Mesenchymal stem cells Divers RPM (s-120583119892) [27ndash30]Several cell types MCTS RWV bioreactor (s-120583119892) [31ndash34]Several cell types MCTS Spinner flask (s-120583119892) [35ndash38]Several cell types MCTS RPM (s-120583119892) [39ndash44]Several cell types MCTS Space (r-120583119892) [45]Hepatocytes liver RWV bioreactor (s-120583119892) [34 46]

various techniques such as forced aggregation techniquesmicro textured surfaces and porous 3D scaffolds are beingemployed to solve these issues [163ndash166] There are severalconventional methods for generating 3D aggregates of cancercells including NASA rotary cell culture systems hangingdrops and culturing of cells using nonadherent surfaces[166ndash168] Conditions of s-120583119892 using the RPM (Figure 2(b)) orthe HARV have been shown to induce the growth of MCTSwithout the use of scaffolds in several types of human cancercells [31ndash33 39 169] The molecular switches initiating s-120583119892-induced spheroid formation are still unknown Severalchanges in morphology and gene expression profiles havebeen observed in follicular thyroid cancer cells grown unders-120583119892 conditions with the main features involving changes inthe ECM and early induction of apoptosis [39 40] Signalingbetween exogenous ECM and tumor surface receptors haslong been thought to be an essential component in reg-ulating the tumorigenic phenotype in 3D cultures Thesephenomena have been demonstrated in studies showing thatblocking specific ECM-integrin signaling can cause a shiftin the malignant potential of tumor cells leading to a morebenign phenotype [40 170] In an attempt to identify gravitysensitive genes responsible for MCTS formation Grosseet al [41] identified 487 transcripts which were differentlyregulated after 24 h of s-120583119892 in comparison to the groundcontrol Comparing adherent cells with MCTS under s-120583119892conditions revealed significant differences in terms of growthpatterns and signaling Interestingly the rate of apoptosis wasincreased in adherent cells compared to MCTS indicatingthat the early phase induction of apoptosis may be concomi-tant with the transition of cells shifting from2D to 3D growthSeveral NF-120581B-driven genes involved in the regulation oftumor invasion were upregulated by s-120583119892 in adherent cellshighlighting that s-120583119892 initiates distinct adaptive mechanismsin the cells

8 Summary

The development of tools like the RPM or RWV gave newimpacts in the field of tissue engineering (Table 1) Growingcells in simulated or real weightlessness for example on theRPMmight be a highly promising new technique to generate

tissue constructs in a scaffold-free manner Cultivation ofchondrocytes might lead to small cartilage particles whichcould be used to replace injured or outworn cartilageRestoring normal osteogenic differentiation of MSCs from s-120583119892 exposure by daily short-term stimulation could be helpfulso that tissue products may become commercially availablelike it has already happened for some bone tissues (BioSeed-Oral Bone codon osteotransplant and Osteocel)

At the moment studies analyzing the molecular mech-anisms behind spheroid formation of for example thyroidcells chondrocyte and others have increased the knowledgeof the complex regulation of 3D growth in 120583119892 [42ndash45 171]

To be able to use this new technique more efficientlyfurther studies are necessary to better understand the exactcellular changes specific to these conditions Tissue whichwas produced under s- or r-120583119892 conditions might be helpfulto better understand cell signaling intercellular contactand tissue growth as well as being sufficient for medicaltransplantationMCTS can be used as an alternative to animalexperiments

Although the 120583119892 environment is not a common fieldfor biologists and medical researchers recent studies haveclearly shown that the loss of gravity impacts the cells and itdramatically changes the genome proteome and secretomeof these cells [43 45] Therefore it is important to systemati-cally explore the advantages of this new research opportunityDifferent space flights have already demonstrated a 3D cellgrowth (Table 2) and similar results have been detectedwith the help of devices simulating 120583119892 in ground-basedlaboratories [34 44 46 172ndash174]



Acknowledgments

This paper was supported by the German Space Agency DLR(DanielaGrimm) (BMWiGrants 50WB0824 and 50WB1124)the European Space Agency (ESA CORA-GBF-2013-001


CORA-GBF-2013-004) and Aarhus University Denmark(Daniela Grimm) The authors would like to thank Mr PeterLindborg Los Angeles CA USA for reviewing the paper forlanguage and grammar errors

References

[1] R JWhite andMAverner ldquoHumans in spacerdquoNature vol 409no 6823 pp 1115ndash1118 2001

[2] A I Grigoriev V S Oganov A V Bakulin et al ldquoClinicaland physiological evaluation of bone changes among astronautsafter long-term space flightsrdquo Aviakosmicheskaia i Ekologich-eskaia Meditsina vol 32 no 1 pp 21ndash25 1998

[3] L Vico P Collet A Guignandon et al ldquoEffects of long-term microgravity exposure on cancellous and cortical weight-bearing bones of cosmonautsrdquoTheLancet vol 355 no 9215 pp1607ndash1611 2000

[4] S Tavella A Ruggiu A Giuliani et al ldquoBone turnover in wildtype and pleiotrophin-transgenicmice housed for threemonthsin the International Space Station (ISS)rdquo PLoS ONE vol 7 no3 Article ID e33179 2012



[7] D Grimm M Wehland J Pietsch J et al ldquoGrowing tissuesin real and simulated microgravity new methods for tissueengineeringrdquo Tissue Engineering Part B Reviews 2014

[8] C Granet N Laroche L Vico C Alexandre and M HLafage-Proust ldquoRotating-wall vessels promising bioreactors forosteoblastic cell culture comparison with other 3D conditionsrdquoMedical and Biological Engineering and Computing vol 36 no4 pp 513ndash519 1998

[9] K Song Z Yang T Liu et al ldquoFabrication and detection oftissue-engineered bones with bio-derived scaffolds in a rotatingbioreactorrdquo Biotechnology and Applied Biochemistry vol 45 no2 pp 65ndash74 2006

[10] K Song T Liu Z Cui X Li and X Ma ldquoThree-dimensionalfabrication of engineered bone with human bio-derived bonescaffolds in a rotating wall vessel bioreactorrdquo Journal of Biomed-ical Materials Research A vol 86 no 2 pp 323ndash332 2008

[11] K Kyriakidou G Lucarini A Zizzi et al ldquoDynamic co-seedingof osteoblast and endothelial cells on 3D polycaprolactonescaffolds for enhanced bone tissue engineeringrdquo Journal ofBioactive and Compatible Polymers vol 23 no 3 pp 227ndash2432008

[12] Q Lv L Nair and C T Laurencin ldquoFabrication charac-terization and in vitro evaluation of poly(lactic acid gly-colic acid)nano-hydroxyapatite composite microsphere-basedscaffolds for bone tissue engineering in rotating bioreactorsrdquoJournal of Biomedical Materials Research A vol 91 no 3 pp679ndash691 2009

[13] S R Facer R S Zaharias M E Andracki J Lafoon S KHunter and G B Schneider ldquoRotary culture enhances pre-osteoblast aggregation and mineralizationrdquo Journal of DentalResearch vol 84 no 6 pp 542ndash547 2005

[14] G B Schneider J K Boehrs J V Hoopes and D A SeaboldldquoUse of 3-dimensional environments to engineer osseous-liketissuerdquo Journal of Developmental Biology andTissue Engineeringvol 3 no 4 pp 42ndash47 2011

[15] D Turhani E Watzinger M Weiszligenbock et al ldquoAnalysis ofcell-seeded 3-dimensional bone constructs manufactured invitro with hydroxyapatite granules obtained from red algaerdquoJournal of Oral andMaxillofacial Surgery vol 63 no 5 pp 673ndash681 2005

[16] F Jin Y Zhang K Xuan et al ldquoEstablishment of three-dimen-sional tissue-engineered bone constructs under microgravity-simulated conditionsrdquo Artificial Organs vol 34 no 2 pp 118ndash125 2010

[17] J G Gershovich and L B Buravkova ldquoMorphofunctionalstatus and osteogenic differentiation potential of human mes-enchymal stromal precursor cells during in vitro modelingof microgravity effectsrdquo Bulletin of Experimental Biology andMedicine vol 144 no 4 pp 608ndash613 2007

[18] M Zayzafoon W E Gathings and J M McDonald ldquoModeledmicrogravity inhibits osteogenic differentiation of human mes-enchymal stem cells and increases adipogenesisrdquo Endocrinol-ogy vol 145 no 5 pp 2421ndash2432 2004

[19] R Saxena G Pan and J M McDonald ldquoOsteoblast andosteoclast differentiation in modeled microgravityrdquo Annals ofthe New York Academy of Sciences vol 1116 pp 494ndash498 2007


[21] D Sheyn G Pelled D Netanely E Domany and D GazitldquoThe effect of simulated microgravity on human mesenchymalstem cells cultured in an osteogenic differentiation system abioinformatics studyrdquo Tissue Engineering A vol 16 no 11 pp3403ndash3412 2010

[22] B Yu D Yu L Cao et al ldquoSimulated microgravity using arotary cell culture system promotes chondrogenesis of humanadipose-derived mesenchymal stem cells via the p38 MAPKpathwayrdquo Biochemical and Biophysical Research Communica-tions vol 414 no 2 pp 412ndash418 2011

[23] S M Uddin and Y X Qin ldquoEnhancement of osteogenicdifferentiation and proliferation in human mesenchymal stemcells by a modified low intensity ultrasound stimulation undersimulated microgravityrdquo PLoS ONE vol 8 no 9 Article IDe73914 2013


[25] X Wu S-H Li L-M Lou and Z-R Chen ldquoThe effect ofthe microgravity rotating culture system on the chondrogenicdifferentiation of bonemarrowmesenchymal stem cellsrdquoMolec-ular Biotechnology vol 54 no 2 pp 331ndash336 2013


[27] N VMerzlikina L B Buravkova and Y A Romanov ldquoThe pri-mary effects of clinorotation on cultured human mesenchymalstem cellsrdquo Journal of Gravitational Physiology vol 11 no 2 pp193ndash194 2004

[28] M Nishikawa H Ohgushi N Tamai et al ldquoThe effect ofsimulated microgravity by three-dimensional clinostat on bonetissue engineeringrdquoCell Transplantation vol 14 no 10 pp 829ndash835 2005


[29] L Yuge T Kajiume H Tahara et al ldquoMicrogravity potentiatesstem cell proliferation while sustaining the capability of differ-entiationrdquo Stem Cells and Development vol 15 no 6 pp 921ndash929 2006

[30] I G Gershovich and L B Buravkova ldquoInterleukine productionin culture of mesenchymal stromal cells of humans duringsimulation of the microgravity effectsrdquo Aviakosmicheskaia IEkologicheskaia Meditsina vol 43 no 3 pp 44ndash50 2009

[31] M IngramG B Techy R Saroufeem et al ldquoThree-dimensionalgrowth patterns of various human tumor cell lines in simulatedmicrogravity of a nasa bioreactorrdquo In Vitro Cellular and Devel-opmental Biology Animal vol 33 no 6 pp 459ndash466 1997

[32] B Marrero J L Messina and R Heller ldquoGeneration of atumor spheroid in a microgravity environment as a 3D modelof melanomardquo In Vitro Cellular and Developmental BiologymdashAnimal vol 45 no 9 pp 523ndash534 2009

[33] S M Konstantinov M M Mindova P T Gospodinov andP I Genova ldquoThree-dimensional bioreactor cultures a usefuldynamic model for the study of cellular interactionsrdquo Annals ofthe New York Academy of Sciences vol 1030 pp 103ndash115 2004

[34] T T Chang and M Hughes-Fulford ldquoMonolayer and spheroidculture of human liver hepatocellular carcinoma cell line cellsdemonstrate distinct global gene expression patterns and func-tional phenotypesrdquo Tissue Engineering A vol 15 no 3 pp 559ndash567 2009

[35] A Bredel-Geissler U Karbach S Walenta L Vollrath and WMueller-Klieser ldquoProliferation-associated oxygen consumptionand morphology of tumor cells in monolayer and spheroidculturerdquo Journal of Cellular Physiology vol 153 no 1 pp 44ndash52 1992

[36] L A Kunz-Schughart C Groebe and W Mueller-KlieserldquoThree-dimensional cell culture induces novel proliferative andmetabolic alterations associated with oncogenic transforma-tionrdquo International Journal of Cancer vol 66 no 4 pp 578ndash5861996

[37] M Wartenberg F Donmez F C Ling H Acker J Hes-cheler and H Sauer ldquoTumor-induced angiogenesis studied inconfrontation cultures of multicellular tumor spheroids andembryoid bodies grown from pluripotent embryonic stemcellsrdquoThe FASEB Journal vol 15 no 6 pp 995ndash1005 2001

[38] R-Z Lin and H-Y Chang ldquoRecent advances in three-dimensional multicellular spheroid culture for biomedicalresearchrdquo Biotechnology Journal vol 3 no 9-10 pp 1172ndash11842008






[44] J Pietsch A Sickmann G Weber et al ldquoA proteomic approachto analysing spheroid formation of two human thyroid cell linescultured on a random positioning machinerdquo Proteomics vol 11no 10 pp 2095ndash2104 2011


[46] T T Chang and M Hughes-Fulford ldquoMolecular mechanismsunderlying the enhanced functions of three-dimensional hep-atocyte aggregatesrdquo Biomaterials vol 35 no 7 pp 2162ndash21712014

[47] M Y Kapitonova N Salim S Othman et al ldquoAlteration of cellcytoskeleton and functions of cell recovery of normal humanosteoblast cells caused by factors associated with real spaceflightrdquo The Malaysian Journal of Pathology vol 35 no 2 pp153ndash163 2013

[48] E A Blaber N Dvorochkin C Lee et al ldquoMicrogravityinducespelvic bone loss throughosteoclastic activity osteocytic osteoly-sis and osteoblastic cell cycle inhibition by CDKN1120572p21rdquo PLoSONE vol 8 no 4 Article ID 61372 2013



[51] R Hemmersbach M Krause R Braucker and K IvanovaldquoGraviperception in ciliates steps in the transduction chainrdquoAdvances in Space Research vol 35 no 2 pp 296ndash299 2005

[52] D A M Mesland ldquoNovel ground-based facilities for researchin the effects of weightrdquo ESA Microgravity News vol 9 1996

[53] D A M Meslana A H Anton H Willemsen and H vanden Ende ldquoThe free fall machinemdasha ground-based facility formicrogravity research in life sciencesrdquoMicrogravity Science andTechnology vol 9 no 1 pp 10ndash14 1996

[54] M Schwarzenberg P Pippia M A Meloni G Cossu MCogoli-Greuter and A Cogoli ldquoSignal transduction in Tlymphocytesmdasha comparison of the data from space the freefall machine and the random positioning machinerdquo Advancesin Space Research vol 24 no 6 pp 793ndash800 1999

[55] M V Berry and A K Geim ldquoOf flying frogs and levitronsrdquoEuropean Journal of Physics vol 18 no 4 pp 307ndash313 1997

[56] J M Valles Jr K Lin J M Denegre and K L Mowry ldquoStablemagnetic field gradient levitation ofXenopus laevis toward low-gravity simulationrdquo Biophysical Journal vol 73 no 2 pp 1130ndash1133 1997

[57] M J A Moes J C Gielen R-J Bleichrodt J J W A van LoonP C M Christianen and J Boonstra ldquoSimulation of micro-gravity by magnetic levitation and random positioning effecton human A431 Cell morphologyrdquo Microgravity Science andTechnology vol 23 no 2 pp 249ndash261 2011



[59] T Higashi A Yamagishi T Takeuchi et al ldquoOrientation oferythrocytes in a strong static magnetic fieldrdquo Blood vol 82 no4 pp 1328ndash1334 1993

[60] K A Mirica F Ilievski A K Ellerbee S S Shevkoplyasand G M Whitesides ldquoUsing magnetic levitation for threedimensional self-assemblyrdquo Advanced Materials vol 23 no 36pp 4134ndash4140 2011

[61] J S Brooks J A Reavis R A Medwood et al ldquoNew opportu-nities in science materials and biological systems in the low-gravity (magnetic levitation) environment (invited)rdquo Journal ofApplied Physics vol 87 no 9 pp 6194ndash6199 2000


[63] C E Dijkstra O J Larkin P Anthony et al ldquoDiamagneticlevitation enhances growth of liquid bacterial cultures byincreasing oxygen availabilityrdquo Journal of the Royal SocietyInterface vol 8 no 56 pp 334ndash344 2011

[64] D Shi R Meng W Deng et al ldquoEffects of microgravitymodeled by large gradient highmagnetic field on the osteogenicinitiation of humanmesenchymal stem cellsrdquo Stem Cell Reviewsand Reports vol 6 no 4 pp 567ndash578 2010


[66] F Castro-Chavez K C Vickers J S Lee C Tung and J DMorrisett ldquoEffect of lyso-phosphatidylcholine and Schnurri-3on osteogenic transdifferentiation of vascular smooth musclecells to calcifying vascular cells in 3D culturerdquo Biochimica etBiophysica Acta vol 1830 no 6 pp 3828ndash3834 2013

[67] F G J R von Sachs ldquoUber Ausschliessung der geotropischenund heliotropischen Krummungen wahrend des WachstumsrdquoWurzburger Arbeiten vol 2 pp 209ndash225 1879


[69] S Murakami and M Yamada ldquoArchitecture of statocytes andchloroplasts under the microgravity environmentrdquo BiologicalSciences in Space vol 2 no 4 p 301 1988

[70] J J W A van Loon ldquoEffects of spaceflight on biologicalsystemsrdquo in Biology in Space and Life on Earth E BrinckmannEd vol 17 Wiley-VCH 2007


[72] T Hoson S Kamisaka Y Masuda M Yamashita and BBuchen ldquoEvaluation of the three-dimensional clinostat as asimulator of weightlessnessrdquo Planta vol 203 pp S187ndashS1971997


[74] C A D Leguy R Delfos and M J B M Pourquie ldquoFluidmotion for microgravity simulations in a random positioningmachinerdquoGravitational and Space Biology vol 25 no 1 pp 36ndash39 2011

[75] S Wakayama Y Kawahara C Li K Yamagata L Yuge andT Wakayama ldquoDetrimental effects of microgravity on mousepreimplantation development in vitrordquo PLoS ONE vol 4 no 8Article ID e6753 2009

[76] M Z Luo R Meng S S Li et al ldquoWeightlessness simulatedwith random positioning machine influences the cytoskeletonand migration of MC3T3-E1 cellsrdquo Journal of the Japan Societyof Microgravity Application vol 28 no 2 pp S41ndashS45 2011

[77] W A Loesberg X F Walboomers E M Bronkhorst J JW A Van Loon and J A Jansen ldquoThe effect of combinedsimulated microgravity and microgrooved surface topographyon fibroblastsrdquo Cell Motility and the Cytoskeleton vol 64 no 3pp 174ndash185 2007

[78] G Aleshcheva J Sahana X Ma et al ldquoChanges inmorphologygene expression and protein content in chondrocytes culturedon a Random Positioning Machinerdquo PLoS ONE vol 8 no 11Article ID e79057 2013

[79] L Prodanov J J W A van Loon J te Riet J A Jansenand X F Walboomers ldquoNanostructured substrate conforma-tion can decrease osteoblast-like cell dysfunction in simulatedmicrogravity conditionsrdquo Journal of Tissue Engineering andRegenerative Medicine 2012

[80] J Ichigi and M Asashima ldquoDome formation and tubulemorphogenesis by Xenopus kidney A6 cell cultures exposed tomicrogravity simulated with a 3D-clinostat and to hypergrav-ityrdquo In Vitro Cellular Developmental Biology Animal vol 37 no1 pp 31ndash44 2001

[81] Y D Jiang W N Li L F Wang Z Y Zhang B M Zhang andH J Wu ldquoSeveral new types of clinostatsrdquo Space Medicine ampMedical Engineering vol 21 no 4 pp 368ndash371 2008


[83] L Xiang F Qi D Dai C Li and Y Jiang ldquoSimulatedmicrogravity affects growth of Escherichia coli and recombi-nant 120573-D-glucuronidase productionrdquoApplied Biochemistry andBiotechnology vol 162 no 3 pp 654ndash661 2010

[84] R Hemmersbach S M Strauch D Seibt and M SchuberldquoComparative studies on gravisensitive protists on ground (2Dand 3D clinostats) and in microgravityrdquo Microgravity Scienceand Technology vol 18 no 3-4 pp 257ndash259 2006

[85] T J Goodwin J M Jessup and D A Wolf ldquoMorphologic dif-ferentiation of colon carcinoma cell lines HT-29 andHT-29KMin Rotating-Wall Vesselsrdquo In Vitro Cellular and DevelopmentalBiology Animal vol 28 no 1 pp 47ndash60 1992


[87] T Liu X Li X Sun X Ma and Z Cui ldquoAnalysis on forcesand movement of cultivated particles in a rotating wall vesselbioreactorrdquo Biochemical Engineering Journal vol 18 no 2 pp97ndash104 2004

[88] T J Goodwin T L Prewett D A Wolf and G F SpauldingldquoReduced shear stress a major component in the ability ofmammalian tissues to form three-dimensional assemblies insimulated microgravityrdquo Journal of Cellular Biochemistry vol51 no 3 pp 301ndash311 1993

[89] E A Nauman C M Ott E Sander et al ldquoNovel quantitativebiosystem formodeling physiological fluid shear stress on cellsrdquoApplied and EnvironmentalMicrobiology vol 73 no 3 pp 699ndash705 2007

[90] XGuo Y ZhaoHChang et al ldquoCreation of engineered cardiactissue in vitro from mouse embryonic stem cellsrdquo Circulationvol 113 no 18 pp 2229ndash2237 2006


[91] D L Tucker C M Ott S Huff et al ldquoCharacterization ofEscherichia coli MG1655 grown in a low-shear modeled microgravity environmentrdquo BMCMicrobiology vol 7 article 15 2007

[92] M N Cinbiz R S Tıgli I G Beskardes M Gumusdereliogluand U Colak ldquoComputational fluid dynamics modeling ofmomentum transport in rotating wall perfused bioreactor forcartilage tissue engineeringrdquo Journal of Biotechnology vol 150no 3 pp 389ndash395 2010

[93] A Puca G Russo and A Giordano ldquoProperties of mechano-transduction via simulated microgravity and its effects onintracellular trafficking of VEGFRrsquosrdquo Oncotarget vol 3 no 4pp 426ndash434 2012


[95] Y C Wang S Zhang T Y Du B Wang and X Q SunldquoClinorotation upregulates inducible nitric oxide synthase byinhibiting AP-1 activation in human umbilical vein endothelialcellsrdquo Journal of Cellular Biochemistry vol 107 no 2 pp 357ndash363 2009


[97] P I Lelkes N Akhtar E Lelkes et al ldquoNeuroendocrine tissueengineering in rotating wall vessel bioreactors under simulatedmicrogravity conditionsrdquo in Proceedings of the 23rd AnnualInternational Conference of the IEEE Engineering in Medicineand Biology Society vol 3 pp 2987ndash2990 October 2001

[98] A Abbott ldquoCell culture biologyrsquos new dimensionsrdquoNature vol424 no 6951 pp 870ndash872 2003






[104] X Ma M Wehland H Schulz et al ldquoGenomic Approach toidentify factors that drive the formation of three-dimensionalstructures by EAhy926 endothelial cellsrdquo PLoS ONE vol 8 no5 Article ID e64402 2013

[105] C Ulbrich K Westphal J Pietsch et al ldquoCharacterizationof human chondrocytes exposed to simulated microgravityrdquoCellular Physiology and Biochemistry vol 25 no 4-5 pp 551ndash560 2010

[106] L E Freed and G Vunjak-Novakovic ldquoCultivation of cell-polymer tissue constructs in simulated microgravityrdquo Biotech-nology and Bioengineering vol 46 no 4 pp 306ndash313 1995


[108] V Stamenkovic G Keller D Nesic A Cogoli and S P GroganldquoNeocartilage formation in 1 g simulated and microgravityenvironments implications for tissue engineeringrdquo Tissue Engi-neering A vol 16 no 5 pp 1729ndash1736 2010

[109] L Kock C C van Donkelaar and K Ito ldquoTissue engineering offunctional articular cartilage the current statusrdquoCell and TissueResearch vol 347 no 3 pp 613ndash627 2012

[110] C H Chang F H Lin T F Kuo and H C Liu ldquoCartilagetissue engineeringrdquo Biomedical Engineering Applications Basisand Communications vol 17 no 2 pp 61ndash71 2005

[111] E Ozcivici Y K Luu B Adler et al ldquoMechanical signals asanabolic agents in bonerdquo Nature Reviews Rheumatology vol 6no 1 pp 50ndash59 2010

[112] R Yang W Lin Y Chen et al ldquoRegulation by ultrasoundtreatment on the integrin expression and differentiation ofosteoblastsrdquo Bone vol 36 no 2 pp 276ndash283 2005

[113] H C Hsu Y C Fong C S Chang et al ldquoUltrasound inducescyclooxygenase-2 expression through integrin integrin-linkedkinase Akt NF-120581B and p300 pathway in human chondrocytesrdquoCellular Signalling vol 19 no 11 pp 2317ndash2328 2007

[114] Y I Arfat W Z Xiao S Iftikhar et al ldquoPhysiological effects ofmicrogravity on bone cellsrdquo Calcified Tissue International vol94 no 6 pp 569ndash579 2014

[115] D A Starr ldquoCommunication between the cytoskeleton and thenuclear envelope to position the nucleusrdquoMolecular BioSystemsvol 3 no 9 pp 583ndash589 2007


[117] S B Doty E R Morey-Holton G N Durnova and A SKaplansky ldquoCosmos 1887 morphology histochemistry andvasculature of the growing rat tibiardquo FASEB Journal vol 4 no1 pp 16ndash23 1990

[118] N V Rodionova O V Polkovenko and V S Oganov ldquoInter-actions of cells in zones of bone resorption under microgravityand hypokinesiardquo Journal of Gravitational Physiology vol 11no 2 pp P147ndash151 2004

[119] O D Kennedy B C Herman D M Laudier R J MajeskaH B Sun and M B Schaffler ldquoActivation of resorption infatigue-loaded bone involves both apoptosis and active pro-osteoclastogenic signaling by distinct osteocyte populationsrdquoBone vol 50 no 5 pp 1115ndash1122 2012

[120] C Lin X Jiang Z Dai et al ldquoSclerostinmediates bone responseto mechanical unloading through antagonizing Wnt120573-cateninsignalingrdquo Journal of Bone and Mineral Research vol 24 no 10pp 1651ndash1661 2009


[122] A Sundaresan S F Clarke and N R Pellis Production ofBoneMorphogenic Proteins (BMPS)Using aNovel Tissue CulturePlatform 2011 US Patent 8076136


[123] A Sundaresan S F Clarke and N R Pellis ldquoConstructionsosseuses mineralisees en trois dimensionsrdquo European PatentNumber EP2013730 2011

[124] A Sundaresan M S F Clarke andM Brinker Development ofa human colloidal bone graft material US 8506982 B2 2013

[125] J R Milstead S J Simske and T A Bateman ldquoSpaceflightand hindlimb suspension disuse models in micerdquo BiomedicalSciences Instrumentation vol 40 pp 105ndash110 2004

[126] H M Frost and W S Jee ldquoOn the rat model of humanosteopenias and osteoporosesrdquo Bone and Mineral vol 18 no3 pp 227ndash236 1992


[128] S M Smith M EWastney K O OrsquoBrien et al ldquoBonemarkerscalcium metabolism and calcium kinetics during extended-duration space flight on the Mir Space Stationrdquo Journal of Boneand Mineral Research vol 20 no 2 pp 208ndash218 2005

[129] JMVogel ldquoBonemineralmeasurement Skylab experimentM-078rdquo Acta Astronautica vol 2 no 1-2 pp 129ndash139 1975

[130] S M Smith M E Wastney B V Morukov et al ldquoCalciummetabolism before during and after a 3-mo spaceflight kineticand biochemical changesrdquo American Journal of Physiology vol277 no 1 part 2 pp R1ndashR10 1999

[131] J E Zerwekh L A Ruml F Gottschalk and C Y C PakldquoThe effects of twelve weeks of bed rest on bone histology bio-chemicalmarkers of bone turnover and calciumhomeostasis ineleven normal subjectsrdquo Journal of Bone and Mineral Researchvol 13 no 10 pp 1594ndash1601 1998

[132] T Lang A LeBlanc H Evans Y Lu H Genant and A YuldquoCortical and trabecular bone mineral loss from the spine andhip in long-duration spaceflightrdquo Journal of Bone and MineralResearch vol 19 no 6 pp 1006ndash1012 2004

[133] A LeBlanc C Lin L Shackelford et al ldquoMuscle volume MRIrelaxation times (T2) and body composition after spaceflightrdquoJournal of Applied Physiology vol 89 no 6 pp 2158ndash2164 2000

[134] M Iki E Kajita Y Dohi et al ldquoAge menopause bone turnovermarkers and lumbar bone loss in healthy Japanese womenrdquoMaturitas vol 25 no 1 pp 59ndash67 1996

[135] J Sirola H Kroger R Honkanen et al ldquoFactors affecting boneloss around menopause in women without HRT a prospectivestudyrdquoMaturitas vol 45 no 3 pp 159ndash167 2003

[136] J D Sibonga H J Evans H G Sung et al ldquoRecovery ofspaceflight-induced bone loss bone mineral density after long-duration missions as fitted with an exponential functionrdquo Bonevol 41 no 6 pp 973ndash978 2007

[137] T F Lang A D Leblanc H J Evans and Y Lu ldquoAdaptationof the proximal femur to skeletal reloading after long-durationspaceflightrdquo Journal of Bone andMineral Research vol 21 no 8pp 1224ndash1230 2006

[138] J J W A Van Loon D-J Bervoets E H Burger et alldquoDecreased mineralization and increased calcium release inisolated fetal mouse long bones under near weightlessnessrdquoJournal of Bone andMineral Research vol 10 no 4 pp 550ndash5571995

[139] R Langer ldquoTissue engineering a new field and its challengesrdquoPharmaceutical Research vol 14 no 7 pp 840ndash841 1997

[140] G M Crane S L Ishaug and A G Mikos ldquoBone tissueengineeringrdquoNatureMedicine vol 1 no 12 pp 1322ndash1324 1995

[141] M J Yaszemski R G Payne W C Hayes R Langer and A GMikos ldquoEvolution of bone transplantation molecular cellular

and tissue strategies to engineer human bonerdquoBiomaterials vol17 no 2 pp 175ndash185 1996

[142] E M Younger and M W Chapman ldquoMorbidity at bone graftdonor sitesrdquo Journal of Orthopaedic Trauma vol 3 no 3 pp192ndash195 1989

[143] S Gronthos and P J Simmons ldquoThe biology and applicationof human bone marrow stromal cell precursorsrdquo Journal ofHematotherapy and Stem Cell Research vol 5 no 1 pp 15ndash231996

[144] N Jaiswal S E Haynesworth A I Caplan and S P BruderldquoOsteogenic differentiation of purified culture-expandedhuman mesenchymal stem cells in vitrordquo Journal of CellularBiochemistry vol 64 no 2 pp 295ndash312 1997

[145] J E Aubin ldquoOsteoprogenitor cell frequency in rat bonemarrowstromal populations role for heterotypic cell-cell interactions inosteoblast differentiationrdquo Journal of Cellular Biochemistry vol72 no 3 pp 396ndash410 1999

[146] J Rauh F Milan K Gunther and M Stiehler ldquoBioreactorsystems for bone tissue engineeringrdquo Tissue Engineering BReviews vol 17 no 4 pp 263ndash280 2011

[147] S Kern H Eichler J Stoeve H Kluter and K BiebackldquoComparative analysis of mesenchymal stem cells from bonemarrow umbilical cord blood or adipose tissuerdquo StemCells vol24 no 5 pp 1294ndash1301 2006

[148] M Kassem and BM Abdallah ldquoHuman bone-marrow-derivedmesenchymal stem cells biological characteristics and poten-tial role in therapy of degenerative diseasesrdquo Cell and TissueResearch vol 331 no 1 pp 157ndash163 2008

[149] J E Dennis S E Haynesworth R G Young and A I CaplanldquoOsteogenesis in marrow-derived mesenchymal cell porousceramic composites transplanted subcutaneously effect offibronectin and laminin on cell retention and rate of osteogenicexpressionrdquo Cell Transplantation vol 1 no 1 pp 23ndash32 1992

[150] B D Boyan T W Hummert D D Dean and Z SchwartzldquoRole of material surfaces in regulating bone and cartilage cellresponserdquo Biomaterials vol 17 no 2 pp 137ndash146 1996

[151] A L Olivares E Marsal J A Planell and D Lacroix ldquoFiniteelement study of scaffold architecture design and culture con-ditions for tissue engineeringrdquo Biomaterials vol 30 no 30 pp6142ndash6149 2009

[152] S B Vangordon R S Voronov T B Blue R L ShambaughD V Papavassiliou and V I Sikavitsas ldquoEffects of scaffoldarchitecture on preosteoblastic cultures under continuous fluidshearrdquo Industrial amp Engineering Chemistry Research vol 50 no2 pp 620ndash629 2011

[153] E Canalis A N Economides and E Gazzerro ldquoBonemorpho-genetic proteins their antagonists and the skeletonrdquo EndocrineReviews vol 24 no 2 pp 218ndash235 2003

[154] B L Eppley J E Woodell and J Higgins ldquoPlatelet quantifi-cation and growth factor analysis from platelet-rich plasmaimplications for wound healingrdquo Plastic and ReconstructiveSurgery vol 114 no 6 pp 1502ndash1508 2004

[155] E A Botchwey S R Pollack E M Levine and C T LaurencinldquoBone tissue engineering in a rotating bioreactor using amicrocarrier matrix systemrdquo Journal of Biomedical MaterialsResearch vol 55 no 2 pp 242ndash253 2001

[156] J Chen R Liu Y Yang et al ldquoThe simulated microgravityenhances the differentiation of mesenchymal stem cells intoneuronsrdquo Neuroscience Letters vol 505 no 2 pp 171ndash175 2011

[157] D N Karunaratne P S Silverstein V Vasadani et al ldquoCellculture models for drug transport studiesrdquo in Drug Delivery


Principles and Applications B Wang T Siahaan and R SolteroEds pp 103ndash124 JohnWiley amp Sons Hoboken NJ USA 2005

[158] M Gassmann J Fandrey S Bichet et al ldquoOxygen supply andoxygen-dependent gene expression in differentiating embry-onic stem cellsrdquo Proceedings of the National Academy of Sciencesof the United States of America vol 93 no 7 pp 2867ndash28721996

[159] R M Sutherland ldquoCell and environment interactions in tumormicroregions the multicell spheroid modelrdquo Science vol 240no 4849 pp 177ndash184 1988

[160] L A Kunz-Schughart J P Freyer F Hofstaedter and R EbnerldquoThe use of 3-D cultures for high-throughput screening themulticellular spheroid modelrdquo Journal of Biomolecular Screen-ing vol 9 no 4 pp 273ndash285 2004

[161] F Hirschhaeuser H Menne C Dittfeld J West W Mueller-Klieser and L A Kunz-Schughart ldquoMulticellular tumorspheroids an underestimated tool is catching up againrdquo Journalof Biotechnology vol 148 no 1 pp 3ndash15 2010

[162] D D Fang Y J Kim C N Lee et al ldquoExpansion of CD133+colon cancer cultures retaining stem cell properties to enablecancer stem cell target discoveryrdquo British Journal of Cancer vol102 no 8 pp 1265ndash1275 2010

[163] S M Ong C Zhang Y Toh et al ldquoA gel-free 3D microfluidiccell culture systemrdquo Biomaterials vol 29 no 22 pp 3237ndash32442008

[164] Y-S Torisawa B-H Chueh D Huh et al ldquoEfficient formationof uniform-sized embryoid bodies using a compartmentalizedmicrochannel devicerdquo Lab on a Chip vol 7 no 6 pp 770ndash7762007

[165] L Y Wu D Di Carlo and L P Lee ldquoMicrofluidic self-assemblyof tumor spheroids for anticancer drug discoveryrdquo BiomedicalMicrodevices vol 10 no 2 pp 197ndash202 2008

[166] Y S Torisawa A Takagi Y Nashimoto T Yasukawa HShiku and T Matsue ldquoA multicellular spheroid array to realizespheroid formation culture and viability assay on a chiprdquoBiomaterials vol 28 no 3 pp 559ndash566 2007

[167] J M Kelm and M Fussenegger ldquoMicroscale tissue engineeringusing gravity-enforced cell assemblyrdquo Trends in Biotechnologyvol 22 no 4 pp 195ndash202 2004

[168] J Friedrich R Ebner and L A Kunz-Schughart ldquoExperimentalanti-tumor therapy in 3-D spheroidsmdashold hat or new chal-lengerdquo International Journal of Radiation Biology vol 83 no11-12 pp 849ndash871 2007

[169] J M Kelm and M Fussenegger ldquoScaffold-free cell delivery foruse in regenerative medicinerdquoAdvanced Drug Delivery Reviewsvol 62 no 7-8 pp 753ndash764 2010

[170] V M Weaver O W Petersen F Wang et al ldquoReversionof the malignant phenotype of human breast cells in three-dimensional culture and in vivo by integrin blocking antibod-iesrdquoThe Journal of Cell Biology vol 137 no 1 pp 231ndash245 1997


[172] Y Naito T Shinoka D Duncan et al ldquoVascular tissue engi-neering towards the next generation vascular graftsrdquo AdvancedDrug Delivery Reviews vol 63 no 4-5 pp 312ndash323 2011


[174] S M Grenon M Jeanne J Aguado-Zuniga M S Conteand M Hughes-Fulford ldquoEffects of gravitational mechanical

unloading in endothelial cells association between caveolinsinflammation and adhesion moleculesrdquo Scientific Reports vol3 article 1494 2013

Review ArticleMultisensory Integration and Internal Models forSensing Gravity Effects in Primates

Francesco Lacquaniti123 Gianfranco Bosco123 Silvio Gravano13 Iole Indovina13

Barbara La Scaleia3 Vincenzo Maffei3 and Myrka Zago3

1 Centre of Space Bio-Medicine University of Rome Tor Vergata Via Montpellier 1 00133 Rome Italy2 Department of Systems Medicine University of Rome Tor Vergata Via Montpellier 1 00133 Rome Italy3 Laboratory of Neuromotor Physiology IRCCS Santa Lucia Foundation Via Ardeatina 306 00179 Rome Italy

Correspondence should be addressed to Francesco Lacquaniti lacquanitimeduniroma2it

Received 2 May 2014 Accepted 26 May 2014 Published 1 July 2014


Copyright copy 2014 Francesco Lacquaniti et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Gravity is crucial for spatial perception postural equilibrium and movement generation The vestibular apparatus is the mainsensory system involved in monitoring gravity Hair cells in the vestibular maculae respond to gravitoinertial forces but theycannot distinguish between linear accelerations and changes of head orientation relative to gravityThe brain deals with this sensoryambiguity (which can cause some lethal airplane accidents) by combining several cues with the otolith signals angular velocitysignals provided by the semicircular canals proprioceptive signals from muscles and tendons visceral signals related to gravityand visual signals In particular vision provides both static and dynamic signals about body orientation relative to the vertical butit poorly discriminates arbitrary accelerations of moving objects However we are able to visually detect the specific accelerationof gravity since early infancy This ability depends on the fact that gravity effects are stored in brain regions which integrate visualvestibular and neck proprioceptive signals and combine this information with an internal model of gravity effects

1 Introduction

Intuitively sensing gravity effects should be a trivial problemfor a complex nervous system such as our own On the onehand direction and magnitude of gravity are quasi-constanton Earth Thus gravitational acceleration varies by lt1 bychanging latitude or altitude while the vertical deflection islt005∘ On the other hand our nervous system is compu-tationally high-powered being endowed with asymp1011 neuronsinterconnected via asymp1015 synapses All axons pieced togetherwould cover the distance between the Earth and the Moon(about 400000 km) One would assume that we are able tomonitor gravity directly by means of our sensory systemsbut this is not the case As we shall review in this papergravity effects are only extrapolated indirectly by the brain bycombining multisensory information with internal modelsthat is with neural processes which mimic a physical event

Sensing and coping with gravity is crucial for spaceperception control of upright posture and generation of

movements Indeed gravity provides a unique reference axisto which we can anchor body orientation and monitor orien-tation changes Gravity effects on limb and body movementsare two-sided insofar as gravity acts both as a perturbingforce that must be counteracted to avoid falling down and asa facilitating force which allows walking and running via theground contact forces

2 Vestibular Information

The vestibular receptors lie inside the labyrinth of thetemporal bone Somewhat similar sensors evolved first ininvertebrates and then in vertebrates about 500 Myrs ago[1] The vestibular apparatus acts as an inertial navigationsystem including in each ear three semicircular canalsoriented roughly orthogonal to each other and two otolithicorgans the sacculus and utriculus with sensory epitheliaoriented roughly vertically and horizontally respectively [2]



The vestibular sensors function as accelerometers the semi-circular canals transducing angular accelerations (roll yawand pitch) and the otoliths transducing linear accelerationsHead acceleration bends the cilia of the hair cells in thesensory organs resulting in a change of the membranepotential and synaptic transmission of the neurons of thevestibular ganglion innervating the receptors The signalsfrom the vestibular neurons carry information about headvelocity and acceleration to the vestibular nuclei in the brainstem In turn signals from these nuclei are relayed andprocessed in several regions of the brain and spinal cordgiving rise to sensations and movements [3]

Thewidely distributed polarities of response of the otolithreceptors in the maculae allow monitoring acceleration vec-tors in any arbitrary direction (Figure 1(a)) These receptorsare extremely sensitive being able to detect displacementsof the cilia as small as 03 nm (typical atomic diameter) andcorrespondingly small accelerations In fact the receptorsin the maculae respond to the projection of an appliedforce (or acceleration) Thus the component of gravitationalacceleration projected on the saccular macula is 119892 cos120572where 120572 is the angle of tilt of the head relative to thegravity direction whereas the component of gravitationalacceleration projected on the utriclemacula is119892 sin120572 Gravityaccelerates the body downwards and is opposed by theground contact forcesThese contact forces are transmitted toall body segments and to the head where they are monitoredby the otolith receptors These receptors respond to a tiltof the head relative to gravity but in general they cannotprovide a unique measurement of gravity effects As anyaccelerometer also the sacculus and utricle respond to netgravitoinertial accelerations and they cannot distinguishbetween the gravitational and the inertial component Forinstance otolith afferents cannot distinguish whether weare accelerating backward (Figure 1(b)) or tilting the headforward (Figure 1(c)) This is because the effect of gravity islocally indistinguishable from the effect of a linear accelera-tion of the reference system [8] In fact the otolith afferentssignal the net gravitoinertial acceleration (a) resulting fromthe vector difference between the gravity vector (g) and thelinear acceleration vector (119891)

119886 = 119892 minus 119891 (1)

All vectors are time-varying referred to head-fixed coordi-nates of the vestibular sensors

This intrinsic ambiguity can give rise to perceptualillusions which become extremely dangerous under somespecific conditions For instance during takeoff an airplanepilot may sense an erroneously high value of upward pitchbecause the resultant of the vector sum of gravity andbackward inertial acceleration is misperceived as the actualorientation relative to the vertical Under conditions of lowvisibility and without the aid of instruments the pilot maythen attempt to correct the aircraft attitude by pitchingdownward with the risk of impacting the ground Spatialdisorientations originating from sensory ambiguities of thiskind are often involved in severe aviation accidents [9]

Under usual conditions however we have no difficultyin sensing the orientation of the head relative to the vertical

even with the eyes closed and in the presence of appreciableaccelerations provided the latter have short duration (asthose of a car or train) This is because the ambiguity can besolved by the brain using a variety of ldquotricksrdquo First the brainfilters the otolith signals so that the low-frequency (longerlasting) signals are automatically interpreted as a change inthe tilt angle of the head relative to gravity [10] Converselyhigh-frequency (shorter lasting) signals are interpreted asrelated to a linear acceleration Also the perceptual illusionof the aircraft pilot mentioned above is consistent withthe frequency segregation hypothesis A prolonged linearacceleration (such as that at airplane takeoff) is a very rareevent when it occurs it is interpreted erroneously as a tiltrelative to gravity Notice however that unlike the outputof a simple low-pass filter the phase of perceived tilt hasbeen shown to be relatively constant across a broad frequencyrange [3 11]

A second ldquotrickrdquo used to disambiguate gravitoinertialacceleration consists in combining the otolith signals withthose of the semicircular canals [11ndash13] just as the man-made inertial systems which combine accelerometers andgyroscopesWhen we turn our heads the semicircular canalsintegrate the angular acceleration and signal the correspond-ing angular velocity for frequencies above about 005Hz [14]Information about angular head velocity can then be usedto keep track of changes in orientation of the gravity vectorrelative to the head [15 16] Formally

119892 = 119892 times 120596 (2)

where 119892 denotes the time derivative of the gravity vec-tor 120596 denotes the angular velocity and times denotes vectorcross-product An internal three-dimensional estimate of thegravity vector in head coordinates can then be obtained byintegrating (2) if the initial conditions for 119892 are known

119892 = int119892 times 120596119889119905 (3)

Notice that the angular velocity that needs to be integratedin (3) is represented by the component parallel to the Earthhorizon because this component changes the orientation ofthe head relative to gravity Given the estimate of 119892 providedby (3) gravitoinertial accelerations can be disambiguated bysolving (1)

A potential problemwith themodel outlined above is thatthe semicircular canals do not provide a reliable estimate ofangular velocity at steady-state [14] Errors in the estimate of120596 would determine an error in the estimate of tilt relativeto gravity provided by (3) A solution consists in correctingthe errors by means of the so-called somatogravic feedback(Figure 2) which tends to align the estimate of the gravita-tional acceleration with the gravitoinertial acceleration [4 517] In other words the time-average of the gravitoinertialacceleration over several seconds yields an estimate of gravityorientation at low frequencies The somatogravic effect canbe incorporated in the model of (2) by including a low-pass


Endolymph fluid

Otolith membrane

Gel layer

Hair cells

(a)

Translation

120572

a = f

(b)

Tilt

120572

a = g

(c)

Figure 1 (a)Otolith organs Left utricle Center saccule Arrows indicate the local on-directions of the hair cells thick black lines indicate thestriola Right cross-section through the otolith membrane showing the different layers Licensed under the Creative Commons Attribution-Share (authorsThomasHaslwanter andRudi Jaeger) (b)-(c) Tilt-translation ambiguity of otolith receptors (b)Theupright head is acceleratedbackward (c) The head is tilted forward These two gravitoinertial accelerations cannot be discriminated by the otolith sensory neurons

filtered term to the tilt estimate thereby canceling any drift[4] The resulting equation is

119892 = 119892 times 120596 minus119892 minus 119886

120591 (4)

The time constant 120591 controls the gain and phase of the 119892estimate when the otolith organs alone are activated for

example by pure translation Alternatively the somatogravicfeedback effects can be substituted by a Bayesian prior atzero translational acceleration [4 17] This prior is also com-patible with the aviation illusion mentioned above Indeedwhile the correction due to the feedback or the prior isbeneficial under normal conditions it can result in the so-called somatogravic illusion [4 9] During translation the tilt


Somatogravicfeedback

IGIA-GAccelerationA

G

A

= G-GIA

GIA

Retinal flow

Canals

Gra

vity

Gravityestimator

+ Ω

k

ko

go

kFmiddotGIAxG

minus1Tvs

intdt

intGx Ω

(a) Complete vestibular information processing model

AccelerationA = G-GIA

GIA

Gra

vity

Zero translation prior

Zero velocity prior

+

minus

Ωintdt

intGx Ω

(b) Bayesian model

Figure 2 Model of visuovestibular processing proposed by Laurens and Angelaki [4] (a) Blue lines vestibular pathways grey lines visualpathways green lines inertial pathways (b) Schematicmodel of Bayesian inference for vestibular processing Black lines deterministicmodellightning bolts sources of noise question marks points of error accumulation blue lines influence of the zero velocity prior green linesinfluence of the zero translation prior grey lines incorporation of visual information (reproduced with permission from [4])

estimate increases over time as the estimated gravitymoves towards the gravitoinertial acceleration This causes adecrease in the translation estimate and in an aftereffect at theend of the translational acceleration

Neural correlates of the operations described above havebeen discovered in the monkey by Angelaki and colleagueswho found that the neural computation of translation (119891)occurs in the so-called Vestibular-only neurons of thevestibular nuclei in the rostral portion of cerebellar fastigiumand nodulus [18] Neurons in these regions combine tem-porally processed signals from the canals and otoliths aspredicted by the internalmodel hypothesis Recently neuronsextracting gravity have been discovered in the cerebellum [5]Laurens et al identified a group of Purkinje cells in the caudalcerebellar vermis with responses that reflect an estimate ofhead tilt (Figure 3) These tilt-selective cells are complemen-tary to the translation-selective Purkinje cells mentionedabove such that their population activities sum to the netgravitoinertial acceleration encoded by the otolith organs

3 Multisensory Integration

As we remarked in the previous section vestibular sensationsresult from composite signals because the otolith signalsare centrally combined with those of the semicircular canalsalready at the level of second-order sensory neurons in the

vestibular nuclei of the brainstem As far as gravity transduc-tion is concerned the vestibular signals are centrally com-bined with other sensory information such as proprioceptivesignals from muscle and tendon receptors visceral signals(from the kidneys vena cava etc) and visual signals Visionin particular provides both static and dynamic (eg opticflow) signals about the orientation of the body relative to thevertical Finally also the so-called efference copy ofmotor sig-nals (ie a copy of the motor commands sent by higher braincenters) and internal estimates of the body axis orientation[19] contribute to an estimate of body orientation All thesesignals are centrally combined yielding accuratemultisensoryestimates about gravity direction Indeed in darkness an erectperson makes errors lt2∘ when aligning an initially tiltedluminous bar with the expected direction of gravity [20]

Under normal light conditions of daily life there areseveral visual cues which point to the direction of gravity[21] Thus trees are rooted downwards and grow verticallyupwards and the walls of the houses are also vertical asare the chandeliers hanging from the ceiling The visualreference to gravity is so strong that there exist touristattraction places (so-called mystery spots) where someanomaly of the environment is exploited to provide theillusion that the gravity law is violated For instance in someplaces there is a strong slope of the terrain and trees growslanted Similar effects can be obtained with tilted walls in


Translation Roll

100

0

100

04 s

OVAR beginning OVAR steady-state

Tran

slatio

n ce

llTi

lt ce

llFi

ring

rate

(spi

kes

s)

180∘s

62∘s

04

G

GA GIA

(a) (b)

(c)(l) (m) (n) (o)

(h) (i) (j) (k)

(d) (e) (f) (g)

TAFB axis

Roll Pitch

LR axisYaw

OVAR tiltTranslation Roll tilt

Yaw signal

Roll signalGIA

(OTO)

Figure 3 (a) Equivalence principle the otolith organs are sensitive to the gravitoinertial acceleration (GIA) equal to the difference betweenthe gravity vector (GA) and the translational acceleration (TA) (b) Naming conventions of the headrsquos translation and rotation axes FBforward-backward LR leftward-rightward (c) Representation of the motion protocols used by Laurens et al [9] GIA along the LR axisrepresented by a swinging pendulum (bottom) is identical in the 3 protocols (translation tilt and off-vertical axis rotation [OVAR]) ((d)ndash(o))Responses from a translation-selective cell (red) and a tilt-selective cell (green) during left-right (LR) translation ((d) and (h)) roll tilt ((e)and (i)) and constant velocity OVAR ((f) (g) (j) and (k)) (l) (m) (n) and (o) show the corresponding yaw velocity (detected by horizontalcanals blue) roll velocity (detected by vertical canals cyan) and GIA along the LR axis (detected by otolith organs [OTO] black) Graycurves fit to the LR translation response (shown in (d) translation cell) or the roll tilt response (shown in (i) tilt cell) (reproduced withpermission from [5])

houses built for the purpose of creating such illusions Theslope angle distorts the perspective of the observer and mayeven create the perceptual illusion that a ball can roll upwardsby itself In Italy such illusions can be felt inside the leaninghouse designed by Vicino Orsini at Bomarzo (see httpenwikipediaorgwikiFileBomarzo parco mostri casa pen-dentejpg)

Except when some cue is so strong as to drive spaceperception by itself (a winner-take all situation) neuralestimates of gravity direction normally are computed by thecentral nervous system as a weighted average of multicueinformation including vestibular visual neck and truncalsignals plus a prior distribution about head and bodyorientation based on experience [20 22ndash24] In Bayesianterms the posterior estimate is obtained by combining noisysensory measurements with a prior each term being weighedinversely to its variance (noise [23])

4 Visual Perception ofGravitational Acceleration

So far we considered the problemofmonitoring the directionof gravity A different problem concerns monitoring its

magnitudeHowdowe estimate the gravitational accelerationof an object in a visual scene This situation occurs quitefrequently as when we experience the vision of objects infree-fall projectile or pendulum motion In addition toobject motion also self-motion may involve visual stimuli(optic flow) accelerated by gravity as when we fall or jumpfrom a height When confronted with gravity effects thevisual system faces a unique challenge In contrast withbody graviceptors (such as those of the vestibular systemmuscle and tendon proprioceptors and visceral organs) thevisual system does not deal with physical gravity directlybut only with the acceleration of the retinal image Whereasgravitational acceleration is constant at a given location thecorresponding retinal acceleration varies inversely with theviewing distance (distance between the observer and thescene) Therefore the visual estimate of gravity effects ona target motion requires accurate estimates of both imageacceleration and viewing distance Both types of estimatesare potentially problematic Indeed while the visual systemis very accurate in velocity estimates it is rather poor inacceleration estimates In fact the visual discrimination ofacceleration is about 5 times worse than that of velocity[25] Also viewing distance may be difficult to assess Eyevergence accommodation and stereo-disparity contribute


Figure 4 Test with curved tubes in preschoolers (reproduced withpermission of Prof Bruce Hood University of Bristol)

to estimating viewing distance of target motion in three-dimensional space but these cues are ineffective when thetarget is far (because of trigonometry) or when it moves on atwo-dimensional video display (as in a videogame) Pictorialinformationmdashsuch as that provided by the presence of objectsof known size (people trees houses etc) in the visual scenemdashalso aids recovering an environmental reference and scale andthus allows the calibration of the retinal image [26]

An internal model of gravity effects represents a criticalcomponent of the visual estimates in addition to raw sensorysignals Indeed gravity represents a special case of visualacceleration to which we are exposed since birth Thereforeit is very likely that it has been internalized in the brain Infact it has been shown that gravity effects on a visual objectare detected early in life [27] Between 5 and 7 months of ageinfants expect that an object moving down an inclined planeaccelerates and an upwardly moving object decelerates andare surprised to see the effects of an artificial reversed gravity(ie objects decelerating while moving downwards andaccelerating while moving upwards) Implicit expectationof gravity effects can generate striking judgment errors inpreschoolers Children around 2 years of age believe that adescending object always falls vertically downwards Thuswhen they are asked to find a ball that is dropped along acurved tube they search directly under the point of fall ratherthan at the exit of the tube [28] However if the ball motion isartificially reversed so that the ball seems to rise upwards allchildren solve the task perfectly (Figure 4) Notice that falsebeliefs about free-fall can still persist in adulthood (so-callednaıve physics) For instance several people without formalscientific background believe that heavier objects fall fasterthan lighter objects of the same size [6]

Strikingly however the motor system has an implicitknowledge of physics which ismuch better than that availableto the cognitive system [6] Daily life offers several examplesof behavior demonstrating the implicit knowledge of physicsand the anticipation of the effects of gravitational and cen-trifugal forces for instance when we try to keep equilibriumwhile riding a bike Also the automatic motor responses

Elbow angle

Wrist angle

Biceps EMG

02 kg

06 kg

130ms

Figure 5 Catching balls of different weight In different trials asubject caught a 02 kg ball (upper panel) and a 06 kg ball (lowerpanel) dropped from a 12m height In each panel traces from topto bottom correspond to elbowflexion angle wrist flexion angle andrectified electrical activity (EMG) of biceps muscle The right-mostvertical line denotes the time of impact of the ball on the hand Theleft-most vertical line denotes the time of onset of the anticipatoryEMG activity (reproduced with permission from [6])

evoked by seeing a falling object are programmed by the brainby taking into account the law of free-fall first formulatedby Galileo Galilei In a laboratory experiment (Figure 5)subjects were asked to catch with the hand a ball that wasdropped vertically from 12m height relative to the hand [6]The ball could weigh 200 g or 600 g in different trials andfell in about 05 s Subjects prestiffened their arm muscles toabsorb the impact at about 130ms irrespective of the specificmass of the ball Instead the amplitude of muscle activationscaled in proportion to the ball mass because a stronger forceis required to counteract a stronger ball momentum [29]It has also been shown that the time of muscle contractionalways leads the impact time by the same amount irrespectiveof the height of fall (Figure 6) [29] Therefore the motorsystem is accurately tuned to the effects of Earth gravity

This tuning persists at the beginning of orbital flightdespite the sensory and cognitive evidence of weightlessnessand despite the motor responses being inappropriate tothe new conditions [30] These results are compatible witha Bayesian interpretation of the estimate of gravitationalacceleration if one assumes that the variance in the prior of1 g acceleration is very small compared with the variance inthe sensory likelihood If so the 1 g prior would bias stronglythe estimate until it is updated with prolonged exposure toweightlessness [31]

Gravity effects are taken into account not only by themotor system but also perceptually as when people judge theduration of motion of a falling target [32 33] or the periodof oscillation of a pendulum [34] Thus in experiments inwhich a pendulum oscillates faster or slower than normalthe observers rate the oscillations violating the physicallength-period relation as less natural than the oscillationscomplying with physics [34] The implicit bias toward grav-itational motion when viewing an oscillating pendulum is


Time-course of anticipatory biceps activity

Fall duration (s) Height (m)

16

12

08

06

04

057

049

040

029

04 02 00

Time-to-contact (s)

Figure 6 Time course of the EMG anticipatory responses ofbiceps Traces correspond to the results obtained for catches of ballsdropped from the heights indicated on the right (fall durations areindicated on the left) EMG traces have been scaled in amplitudeto their maximum and aligned relative to collision time Time axisindicates the time remaining prior to collision (reproduced withpermission from [6])

also revealed by the observation that harmonic motion isperceived as uniform [35] Also the perceptual judgment ofpassive egomotion along the vertical directionmdashsimulated bymeans of immersive visual stimulimdashis based on the internalmodel of gravity [36]

Just as in the case of the estimates of the directionof gravity also those of visual gravitational accelerationgenerally depend on a combination of multiple cues andsuch combination may obey Bayesrsquo rules The internal modelprovides the prior while various sensory cues provide thelikelihood of the estimate with a reliability that depends onthe context In one study observers were asked to judgethe duration of motion of a target accelerating in one offour different directions downwards upwards leftwards andrightwards relative to a visual scene [33] Downward motioncomplied with the gravity constraint whereas motion in theother directions violated this constraint Observers watchedeither a pictorial or an empty scene while being upright or

tilted by 45∘ relative to the monitor and Earthrsquos gravity Inanother condition observers were upright and the scene wastilted by 45∘ Discrimination precisionwas significantly betterfor downwardmotion than for the other directions Howeverthe difference in precision was not constant across conditionsbut was highest when both the observer and the pictorialscenewere upright and lowestwhen the target direction in theempty scene was tilted by 45∘ relative to an upright observerThus the behaviour observed in the study was consistentwith the combination of pictorial cues orientation of theobserver relative to the physical vertical and orientation oftarget motion relative to the physical vertical

The interaction of the visual signals with vestibularsignals about subject orientation relative to physical gravitywas shown in a study performed during a parabolic flightcampaign [37] During each parabola a 20 s weightless (0 g)phase was preceded and followed by 20 s of hypergravity(15ndash18 g) Strikingly the timing of interception of a visualtarget moving along the visual vertical reversed sign duringthe weightless phases compared with the responses at normalgravity [37] This reversal depends on the reversal of theotolith responses during the transition from hypergravity tohypogravity which was sensed as a negative gravity that isas a gravitational pull in the upward direction (comparableto when we are suspended upside-down)

5 Neural Substrates of the Internal Model ofGravity Effects on Visual Motion

The hypothesis that the effects of gravity on a target motionare taken into account by combining multisensory informa-tion including visual and vestibular cues is supported byhuman neuroimaging studies In a series of fMRI studies[7 38ndash40] visual gravitational acceleration (involving eitherobject motion or simulated egomotion) engaged a network ofbrain regions located within and around the Sylvian fissureclose to the temporoparietal junction (TPJ) posterior insularcortex retroinsula parietal operculum supramarginal gyrustemporal operculum and superior and middle temporalgyri In addition gravitational motion engaged primarysomatosensory and motor cortex ventral premotor cortexSMA cingulate cortex visual cortex including the lingualgyrus and several subcortical structures posterior thalamusputamen cerebellum and vestibular nuclei A causal linkbetween TPJ activity and the processing of visual gravita-tional motion has been demonstrated by transiently disrupt-ing the activity of TPJ by means of transcranial magneticstimulation (TMS) [41]

As reviewed earlier electrophysiological studies in themonkey showed that a population of Purkinje cells in thecaudal cerebellar vermis encodes head tilt thus reflectingan estimate of gravity direction based on vestibular infor-mation [5] Interestingly human posterior cerebellar vermis(a homologue region of that studied by Laurens et al [5]in monkeys) and vestibular nuclei appear to be involved incombining pictorial information with the internal model ofgravity to extract gravitational motion from visual scenes(Figure 7) [7]


(a) (b)

(c)

6

4

2

0

minus2

minus4a b c

T-v

alue

PictorialNonpictorial

(gminusrg)

(d)

Figure 7 Functionalmagnetic resonance imaging of the cerebellum and brainstem in a task of interception of a ballmoving along the verticalBrain areas showing preferential activation for natural gravity motion specifically for the pictorial visual context (a) (b) (c) Activations inthe midline cerebellum ((a) axial section (c) medial sagittal section) and vestibular nuclei ((b) axial section) Roman numerals in (c) denoteLarsell lobules White circles are centered on maximal statistical activation peaks (a) Lobules IXX (b) Left vestibular nuclei (c) LobulesVIIVIII (d) Bar-graphs of the difference (plusmn between-subjects sem) t-values for natural gravity (119892) and artificial reversed gravity (119903119892) trialsin pictorial (white) and nonpictorial (black) context for the activity peaks circled in (a) (b) and (c) (reproduced with permission from [7])

In sum the neuroimaging studies reviewed above indi-cate that the effects of gravity on visual motion are encodedin a highly distributed cortical-subcortical network Sev-eral regions of this network colocalize with the regionsindependently activated by vestibular caloric stimuli [38]These regions then presumably belong to the multimodalvestibular network which also responds to visual and neckproprioceptive stimuli [42 43] Lesions of vestibular cortexcan lead to a tilt of the perceived visual vertical and rotationalvertigounsteadiness [44] while focal electrical stimulationor epileptic discharges can elicit sensations of self-motion oraltered gravity [45 46]

6 Conclusions

We argued that an apparently simple problem such as that ofmonitoring gravity effects on our body and on the external

environment is in fact computationally very demandingeven for a high-powered brain such as that of primatesMeasurements derived from individual sensory organs areoften ambiguous (due to the intrinsic constraints of physicallaws) and noisy (due to biological limitations) Howeverthe combination of multisensory signals (visual vestibularproprioceptive and visceral) and the reliance on internalmodels of physics yield estimates which are very accurateunder normal conditions but which can fail badly underanomalous conditions (such as the early phases of spaceflight) Central processing of multisensory information andinternal models occurs in a widely distributed network ofcortical and subcortical regions The extensive integrationof sensory and motor information in this network makesgravity-related information available to many vital functionsof the organism



The research reviewed in this paper was conducted in theabsence of any commercial or financial relationships thatcould be construed as a potential conflict of interests

Acknowledgments

The authorsrsquo work was supported by the Italian Ministryof University and Research (PRIN grant) and Italian SpaceAgency (CRUSOE COREA SLINK andARIANNA grants)

References

[1] K W Beisel Y Wang-Lundberg A Maklad and B FritzschldquoDevelopment and evolution of the vestibular sensory appa-ratus of the mammalian earrdquo Journal of Vestibular ResearchEquilibrium and Orientation vol 15 no 5-6 pp 225ndash241 2005

[2] M E Goldberg M F Walker and A J Hudspeth ldquoThevestibular systemrdquo inPrinciples ofNeural Science E KKandel JH Schwartz T M Jessell S A Siegelbaum and A J HudspethEds pp 917ndash934 McGraw-Hill New York NY USA 2013

[3] D E Angelaki and K E Cullen ldquoVestibular system the manyfacets of a multimodal senserdquo Annual Review of Neurosciencevol 31 pp 125ndash150 2008

[4] J Laurens and D E Angelaki ldquoThe functional significance ofvelocity storage and its dependence on gravityrdquo ExperimentalBrain Research vol 210 no 3-4 pp 407ndash422 2011

[5] J Laurens HMeng andD E Angelaki ldquoNeural representationof orientation relative to gravity in the macaque cerebellumrdquoNeuron vol 80 no 6 pp 1508ndash1518 2013

[6] M Zago and F Lacquaniti ldquoCognitive perceptual and action-oriented representations of falling objectsrdquo Neuropsychologiavol 43 no 2 pp 178ndash188 2005

[7] W L Miller V Maffei G Bosco et al ldquoVestibular nuclei andcerebellum put visual gravitational motion in contextrdquo Journalof Neurophysiology vol 99 no 4 pp 1969ndash1982 2008

[8] C Fernandez and J M Goldberg ldquoPhysiology of peripheralneurons innervating otolith organs of the squirrel monkey IResponse to static tilts and to long duration centrifugal forcerdquoJournal of Neurophysiology vol 39 no 5 pp 970ndash984 1976

[9] R Gibb B Ercoline and L Scharff ldquoSpatial disorientationdecades of pilot fatalitiesrdquo Aviation Space and EnvironmentalMedicine vol 82 no 7 pp 717ndash724 2011

[10] R Mayne ldquoA systems concept of the vestibular organsrdquo inVestibular System Psychophysics Applied Aspects and GeneralInterpretations H Kornhuber Ed vol 6 part 2 pp 493ndash580Springer Berlin Germany 1974

[11] S Glasauer ldquoInteraction of semicircular canals and otoliths inthe processing structure of the subjective zenithrdquo Annals of theNew York Academy of Sciences vol 656 pp 847ndash849 1992

[12] D E Angelaki M Q McHenry J D Dickman S D Newlandsand B J M Hess ldquoComputation of inertial motion neuralstrategies to resolve ambiguous otolith informationrdquo Journal ofNeuroscience vol 19 no 1 pp 316ndash327 1999

[13] D M Merfeld L Zupan and R J Peterka ldquoHumans useinternal models to estimate gravity and linear accelerationrdquoNature vol 398 no 6728 pp 615ndash618 1999

[14] J M Goldberg and C Fernandez ldquoVestibular mechanismsrdquoAnnual Review of Physiology vol 37 pp 129ndash162 1975

[15] D M Merfeld and L H Zupan ldquoNeural processing of grav-itoinertial cues in humans III Modeling tilt and translationresponsesrdquo Journal of Neurophysiology vol 87 no 2 pp 819ndash833 2002

[16] AM Green andD E Angelaki ldquoAn integrative neural networkfor detecting inertial motion and head orientationrdquo Journal ofNeurophysiology vol 92 no 2 pp 905ndash925 2004

[17] J Laurens and J Droulez ldquoBayesian processing of vestibularinformationrdquo Biological Cybernetics vol 96 no 4 pp 389ndash4042007

[18] D E Angelaki A G Shaikh A M Green and J D DickmanldquoNeurons compute internal models of the physical laws ofmotionrdquo Nature vol 430 no 6999 pp 560ndash564 2004

[19] HMittelstaedt ldquoA new solution to the problem of the subjectiveverticalrdquo Naturwissenschaften vol 70 no 6 pp 272ndash281 1983

[20] R A A Vingerhoets M de Vrijer J A M van Gisbergen andW PMedendorp ldquoFusion of visual and vestibular tilt cues in theperception of visual verticalrdquo Journal of Neurophysiology vol101 no 3 pp 1321ndash1333 2009

[21] I P Howard Human Visual Orientation John Wiley amp SonsNew York NY USA 1982

[22] L H Zupan D M Merfeld and C Darlot ldquoUsing sensoryweighting to model the influence of canal otolith and visualcues on spatial orientation and eye movementsrdquo BiologicalCybernetics vol 86 no 3 pp 209ndash230 2002

[23] P R MacNeilage M S Banks D R Berger and H H BulthoffldquoABayesianmodel of the disambiguation of gravitoinertial forceby visual cuesrdquo Experimental Brain Research vol 179 no 2 pp263ndash290 2007

[24] D E Angelaki E M Klier and L H Snyder ldquoA vestibular sen-sation probabilistic approaches to spatial perceptionrdquo Neuronvol 64 no 4 pp 448ndash461 2009

[25] P Werkhoven H P Snippe and A Toet ldquoVisual processing ofoptic accelerationrdquo Vision Research vol 32 no 12 pp 2313ndash2329 1992

[26] S E Palmer Vision Science Photons to Phenomenology MITPress Cambridge UK 1999

[27] I K Kim and E S Spelke ldquoInfantsrsquo sensitivity to effectsof gravity on visible object motionrdquo Journal of ExperimentalPsychology Human Perception and Performance vol 18 no 2pp 385ndash393 1992

[28] B M Hood ldquoGravity does rule for falling eventsrdquo Developmen-tal Science vol 1 no 1 pp 59ndash63 1998

[29] F Lacquaniti and C Maioli ldquoThe role of preparation in tuninganticipatory and reflex responses during catchingrdquoThe Journalof Neuroscience vol 9 no 1 pp 134ndash148 1989



[32] M A Grealy C M Craig C Bourdin and S G ColemanldquoJudging time intervals using a model of perceptuo-motorcontrolrdquo Journal of Cognitive Neuroscience vol 16 no 7 pp1185ndash1195 2004

[33] A Moscatelli and F Lacquaniti ldquoThe weight of time gravita-tional force enhances discrimination of visualmotion durationrdquoJournal of Vision vol 11 no 4 pp 1ndash17 2011

[34] J B Pittenger ldquoDetection of violations of the law of pendulummotion observersrsquo sensitivity to the relation between periodand lengthrdquo Ecological Psychology vol 2 no 1 pp 55ndash81 1990


[35] B La Scaleia M Zago A Moscatelli F Lacquaniti and PViviani ldquoImplied dynamics biases the visual perception ofvelocityrdquo PloS ONE vol 9 no 3 Article ID e93020 2014

[36] I IndovinaVMaffei andF Lacquaniti ldquoAnticipating the effectsof visual gravity during simulated self-motion estimates oftime-to-passage along vertical and horizontal pathsrdquo Experi-mental Brain Research vol 229 no 4 pp 579ndash586 2013

[37] P Senot M Zago A le Seacrsquoh et al ldquoWhen up is down in 0ghow gravity sensing affects the timing of interceptive actionsrdquoJournal of Neuroscience vol 32 no 6 pp 1969ndash1973 2012

[38] I Indovina V Maffei G Bosco M Zago E Macaluso and FLacquaniti ldquoRepresentation of visual gravitational motion inthe human vestibular cortexrdquo Science vol 308 no 5720 pp416ndash419 2005

[39] VMaffei EMacaluso I Indovina G Orban and F LacquanitildquoProcessing of targets in smooth or apparent motion alongthe vertical in the human brain an fMRI studyrdquo Journal ofNeurophysiology vol 103 no 1 pp 360ndash370 2010

[40] I Indovina V Maffei K Pauwels E Macaluso G A Orbanand F Lacquaniti ldquoSimulated self-motion in a visual gravityfield sensitivity to vertical and horizontal heading in the humanbrainrdquo NeuroImage vol 71 pp 114ndash124 2013

[41] G Bosco M Carrozzo and F Lacquaniti ldquoContributions of thehuman temporoparietal junction and MTV5+ to the timingof interception revealed by transcranial magnetic stimulationrdquoThe Journal of Neuroscience vol 28 no 46 pp 12071ndash120842008

[42] S Bense T Stephan T A Yousry T Brandt and M DieterichldquoMultisensory cortical signal increases and decreases duringvestibular galvanic stimulation (fMRI)rdquo Journal of Neurophys-iology vol 85 no 2 pp 886ndash899 2001

[43] G Bottini H O Karnath G Vallar et al ldquoCerebral representa-tions for egocentric space functional-anatomical evidence fromcaloric vestibular stimulation and neck vibrationrdquo Brain vol124 no 6 pp 1182ndash1196 2001

[44] T Brandt andMDieterich ldquoThe vestibular cortex its locationsfunctions and disordersrdquo Annals of the New York Academy ofSciences vol 871 pp 293ndash312 1999

[45] O Blanke S Ortigue T Landis and M Seeck ldquoStimulatingillusory own-body perceptionsrdquo Nature vol 419 no 6904 pp269ndash270 2002

[46] D K Nguyen D B Nguyen R Malak et al ldquoRevisiting the roleof the insula in refractory partial epilepsyrdquo Epilepsia vol 50 no3 pp 510ndash520 2009

Research ArticleIntegration Analysis of MicroRNA andmRNA Expression Profiles in Human Peripheral BloodLymphocytes Cultured in Modeled Microgravity

C Girardi1 C De Pittagrave1 S Casara1 E Calura1 C Romualdi1 L Celotti12 and M Mognato1

1 Dipartimento di Biologia Universita degli Studi di Padova Via U Bassi 58B 35131 Padova Italy2 Laboratori Nazionali di Legnaro INFN Viale dellrsquoUniversita 2 Legnaro 35020 Padova Italy

Correspondence should be addressed to L Celotti luciacelottiunipdit and M Mognato maddalenamognatounipdit

Received 16 April 2014 Revised 22 May 2014 Accepted 22 May 2014 Published 23 June 2014


Copyright copy 2014 C Girardi et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

We analyzed miRNA and mRNA expression profiles in human peripheral blood lymphocytes (PBLs) incubated in microgravitycondition simulated by a ground-based rotating wall vessel (RWV) bioreactor Our results show that 42miRNAswere differentiallyexpressed in MMG-incubated PBLs compared with 1 g incubated ones Among these miR-9-5p miR-9-3p miR-155-5p miR-150-3p and miR-378-3p were the most dysregulated To improve the detection of functional miRNA-mRNA pairs we performedgene expression profiles on the same samples assayed for miRNA profiling and we integrated miRNA and mRNA expressiondata The functional classification of miRNA-correlated genes evidenced significant enrichment in the biological processes ofimmuneinflammatory response signal transduction regulation of response to stress regulation of programmed cell death andregulation of cell proliferation We identified the correlation of miR-9-3p miR-155-5p miR-150-3p and miR-378-3p expressionwith that of genes involved in immuneinflammatory response (eg IFNG and IL17F) apoptosis (eg PDCD4 and PTEN) andcell proliferation (eg NKX3-1 and GADD45A) Experimental assays of cell viability and apoptosis induction validated the resultsobtained by bioinformatics analyses demonstrating that in human PBLs the exposure to reduced gravitational force increases thefrequency of apoptosis and decreases cell proliferation

1 Introduction

Exposure to spaceflight environment is known to cause inhumansmany adverse physiological changes including skele-tal muscle atrophy [1ndash3] cardiovascular and microvasculardisorders [4ndash6] bone deterioration [7 8] and impairedimmune system function [9 10] Immune systemdysfunctiondue to exposure to microgravity has been documented aswell in terms of reduced activationproliferation alteredcytokine production and altered signal transduction [11 12]Alterations in global gene expression patterns have beenalso observed in space-flown human cells involving mainlygenes of immune system activation [13 14] cytoskeleton [15]and cell cycle [16 17] However due to the difficulty andlimitations of performing experiments in the real micro-gravity of space many investigations have been conductedunder simulated microgravity conditions in which cells are

cultured in ground-based machines such as clinostats androtating wall vessel bioreactors that generate a residual 10minus3ndash10minus6 g force that approximates microgravity [5 18ndash24] Theresults indicate that similar to space microgravity simulatedmicrogravity affects both cell structure and function as wellas gene expression inmammalian cells [14 19 25] in bacteria[26 27] or in other living organisms [28ndash30]

Since molecular changes at the gene level may compro-mise cell function it is important to understand the cellularresponse to reduced gravity at the molecular level For thispurpose a class of noncoding RNAs called microRNAs(miRNAs) plays a key role miRNAs are a large family ofsmall RNAs of 18ndash24 nucleotides that are involved in post-transcriptional regulation of gene expression by interactingwith 31015840-untranslated regions (UTR) of target genes The reg-ulatory process is complex and occurs posttranscriptionallythrough miRNA interaction with a target site in the mRNA



that has partial or complete complementarity to the miRNAThe binding of miRNAs to complementary sequence of theirtarget mRNAs may repress translation or induce degradationofmRNAs [31] Recently the destabilization of targetmRNAsinstead of translational repression has been shown to be thepredominant mechanism for reduced protein output [32]Less often dsRNA formed by miRNA target complexes cantarget gene promoters and actually enhance transcription oftarget genes sometimes termed RNAa (RNA activation) [33]A single miRNA may have broad effects on gene expressionnetworks such as regulating cell lineage specificity cellularfunctions or stress response Besides a physiological roleof miRNAs in a variety of important biological processesincluding differentiation apoptosis [34] and fat metabolism[35] the miRNA-mediated gene regulation operates alsoduring viral infection [36] stress response pathway [37] andpathological processes such as tumorigenesis [38ndash42]

The present study is addressed to identify alterationsin miRNA profiles of human peripheral blood lympho-cytes (PBLs) incubated in modeled microgravity (MMG)with respect to those incubated in gravity 1 g To simulatemicrogravity we used a specialized bioreactor developed atthe NASA-Johnson Space Center (Houston TX USA) therotating wall vessel which represents a valid ground modelto simulate as far as possible a condition of reduced gravityTo identify miRNA-correlated genes whose expression levelwas significantly altered as a function ofMMGwe performedgene expression profiling on the same PBL samples assayedfor miRNA profiling and we integrated microRNAome andtranscriptome by using MAGIA2 [43] a web tool for theintegrative analysis of miRNA and genes expression dataincorporating transcriptional regulation A group ofmiRNA-mRNA pairs related to immunity cell proliferation andapoptosis was identified in PBLs incubated in MMG Thedifferences between MMG and 1 g on correlated miRNA-mRNA pairs involved in cell proliferation and apoptosiswere investigated by in vitro assays of clonogenic ability andapoptosis induction in PBLs incubated in MMGwith respectto those incubated in 1 g conditions


21 Lymphocytes Isolation and Microgravity SimulationHuman peripheral blood lymphocytes (PBLs) were obtainedfrom freshly collected ldquobuffy coatsrdquo from blood samples oftwelve healthy anonymous donors at the Blood Centre ofthe City Hospital of Padova (Italy) This study obtainedthe Ethics Approval from the Transfusion Medicine (TM)Ethics Committee of Blood Centre of the City Hospital ofPadova PBLs were isolated by separation on Biocoll densitygradient (BIOCHROM Berlin Germany) After isolationPBLs were preincubated overnight at a concentration of 3times 106mL in basal medium RPMI 1640 containing Gluta-MAX I (Invitrogen Life Technologies Carlsbad CA USA)124UmL penicillin 63120583gmL streptomycin sulfate and 10fetal bovine serum (FBS BIOCHROM Berlin Germany)After the overnight incubation PBLs consisting of peripheralmononuclear cells depleted of monocytes were suspended

at 1 times 106mL in basal medium and subjected to modeledmicrogravity simulated by the rotating wall vessel (RWV)bioreactor (Synthecon Cellon) placed inside a humidifiedincubator vertically rotating at 23 rpm [44] In the rotatingsystem the gravity is balanced by equal and oppositemechan-ical forces (centrifugal Coriolis and shear components)and the gravitational vector is reduced to about 10minus3 g Inthese conditions single cells are nearly always in suspensionrotating in quasi-stationary manner with the fluid in a low-shear culture environment [19 45] Ground-based (1 g) PBLswere kept at the same cell density in 75 cm2 flasks (FALCON)in the same medium After 24 48 or 72 h of incubationin MMG and 1 g PBLs were activated to enter cell cycle tomeasure cell proliferation by incubation in culture medium(CM) containing phytohaemagglutinin (PHA BIOCHROMBerlin Germany) and interleukin 2 (IL2 Chiron Siena Italy)as stimulating factors [46]

22 Total RNA Isolation Total RNA was isolated from107 PBLs at the end of 24 h incubation in MMG and 1 g byusing Trizol Reagent (Invitrogen Life Technologies CarlsbadCA USA) according to the manufacturerrsquos protocol TotalRNA was quantified using the ND-1000 spectrophotometer(Nanodrop Wilmington DE USA) and RNA integrity andthe content of miRNAs were assessed by capillary elec-trophoresis using the Agilent Bioanalyzer 2100 as previouslydescribed [47] Only total RNA samples with RNA integritynumber (RIN) values ge6 and with miRNA lt20 were usedfor microarray analysis

23 miRNA and Gene Expression Profiling MicroRNAs pro-filing was carried out in PBL samples incubated in MMGversus 1 g Analyses were performed by using the ldquoHumanmiRNA Microarray kit (V2)rdquo (Agilent Technologies) thatallows the detection of 723 known human (miRBase v101)and 76 human viral miRNAs Total RNA (200 ng) waslabeled with pCp Cy3 according to the Agilent protocoland unincorporated dyes were removed with MicroBioSpin6columns (BioRad) [48] Probes hybridization and slideswashing were performed as previously reported [47] AgilentFeature Extraction software version 10511 was used forimage analysis

Gene expression profiling was carried out in MMG-incubated PBLs versus 1 g incubated PBLs on total RNAextracted from the same PBL samples assayed for miRNAprofiling We used the ldquoWhole Human Genome Oligo Mi-croarrayrdquo (Agilent) consisting of sim41000 (60-mer) oligonu-cleotide probes which span conserved exons across thetranscripts of the targeted full-length genes 800 ng of totalRNAwas labeled with ldquoAgilent One-Color Microarray-BasedGene Expression protocolrdquo according to the manufacturerrsquosinstructions The method uses T7 RNA polymerase whichsimultaneously amplifies target material and incorporatescyanine 3-labeled CTP The Cy3-labeled cRNAs were puri-fied using Qiagenrsquos RNeasy mini spin columns (Qiagen)and quantified using the ND-1000 spectrophotometer (Nan-odrop Wilmington DE USA) Probes hybridization andslides washing were performed as previously reported [47]


Slides were scanned on an Agilent Microarray Scanner Sys-tem (model G2565CA) and Agilent Feature Extraction soft-ware version 10511 was used for image analysis Raw data areavailable on the Gene Expression Omnibus (GEO) website(httpwwwncbinlmnihgovgeo) using SuperSeries acces-sion number GSE57418 that groups microRNA (GSE57400)and mRNA expression profiles (GSE57408)

24 Statistical Analysis of miRNA and Gene Expression DataInterarray normalization of expression levels was performedwith cyclic Lowess formiRNA experiments and with quantilefor gene expression profiling [49] to correct possible exper-imental distortions Normalization function was applied toexpression data of all experiments and then values of spotreplicates within arrays were averagedThemodalities of spotquality measures and hybridization are reported previously[47] The identification of differentially expressed genes andmiRNAs was performedwith one- and two-class SignificanceAnalysis of Microarray (SAM) program [50] with defaultsettingsThe expression level of eachmiRNA andmRNAwascalculated as the log2 (MMG1 g) PBLs of the same donorPathway analysis on differentially expressed genes has beenperformed using Graphite web [51] hypergeometric test onReactome Pathways considering significant those categorieswith FDR lt 01

25 Identification of miRNA Target Genes and CorrelationAnalysis of miRNA and mRNA Expression Data To predictmiRNA targets we performed a computational analysis inte-grating mRNA and miRNA expression measurements fromthe same donor usingMAGIA2 web tool [43] freely availableat httpgencompbiounipditmagia2We used httpwwwmicrornaorg predictions and Pearson correlation (119903 gt 04)to estimate the degree of correlation between any putativepairs of miRNA and mRNA [52 53] To identify the bio-logical processes most involved in target prediction we haveperformed an enrichment analysis on Gene Ontology (GO)usingDAVIDweb tool v67 [54] considering significant thosecategories with FDR lt 02

Intraclass analyses have been performed consideringMMGand 1 g samples separately In order to have comparableresults the intraclass analysis has been performed usingMAGIA2 software with the same parameters predictors andcutoff described for the previous analysis Specific inter-actions for MMG and 1 g networks have been identifiedand a GO enrichment analysis (FDR lt 02) was performedseparately on the nodes belonging to the specific MMG and1 g networks using DAVID web tool

26 Quantitative Real-Time PCR (qRT-PCR) Assay In orderto verify the expression data generated by miRNA andmRNA microarrays we performed qRT-PCR experimentsfor miRNAs and genes which showed significant expressionchanges in MMG The following miRNAs were subjectedto the RT-qPCR validation miR-9-5p miR-378a miR-155-5p and miR-150-3p Reverse transcription of 10 ng of totalRNA with primers corresponding to each miRNA and toU48 small nuclear RNA (RNU48) as endogenous control

was performed as directed by the protocol of the two-stepTaqMan MicroRNA Assay kit (Applied Biosystems FosterCity CA USA) that incorporates a target-specific stem-loopreverse transcription primer to provide specificity for themature miRNA target For the PCR reaction 1 120583L of the RTreaction was combined with 05 120583L of TaqMan MicroRNAAssay 20x and 5 120583L of TaqMan Universal PCR Master Mixin a 10 120583L final volume The reactions were incubated ina Mastercycler EP gradient 119878 (Eppendorf) in 02mLPCRtubes for 30min at 16∘C and 30min at 42∘C followed by5min at 85∘C and then held at 4∘CThe resulting cDNA wasquantitatively amplified in 40 cycles on an ABI 7500 Real-Time PCR System using TaqMan Universal PCRMaster Mixand TaqMan MicroRNA Assays

For mRNA detection 1 120583g of total RNA was retro-transcribed with ImProm-II Reverse Transcription System(Promega) qRT-PCR was performed with the GoTaq qPCRMaster Mix (Promega) and gene-specific primers for IFNGIL17F TLR4 HLA-DRB1 and BCL6 genes and for GAPDHas reference qRT-PCR reactions were performed in quad-ruplicates in PBL samples from 6 to 8 donors Real-timePCR was performed using an Applied Biosystems 7500 FastReal-Time PCR System with cycling conditions of 95∘C for10min followed by 95∘C for 15 sec and 60∘C for 60 sec 45cycles in total The relative expression levels of miRNAsand mRNAs between samples were calculated using thecomparative delta CT (threshold cycle number) method(2minusΔΔCT) implemented in the 7500 Real-Time PCR Systemsoftware [55] Primersrsquo pairs used are as follows GAPDH(glyceraldehyde-3-phosphate dehydrogenase) fw 51015840-TCC-TCTGACTTCAACAGCGA-31015840 rev 51015840-GGGTCTTACTCC-TTGGAGGC-31015840 IFNG (interferon gamma) fw 51015840- GGC-ATTTTGAAGAATTGGAAAG-31015840 rev 51015840-TTTGGATGC-TCTGGTCATCTT-31015840 IL17F (interleukin 17) fw 51015840-GGC-ATCATCAATGAAAACCA-31015840 rev 51015840- TGGGGTCCCAAG-TGACAG-31015840 TLR4 (Toll-like receptor 4)fw 51015840-CCTGCG-TGAGACCAGAAAG-31015840 rev 51015840-TTCAGCTCCATGCAT-TGATAA-31015840 HLA-DRB1 (major histocompatibility com-plex class II DR beta 1) fw 51015840-ACAACTACGGGGTTG-TGGAG-31015840 rev 51015840-GCTGCCTGGATAGAAACCAC-31015840BCL6 (B-cell CLLlymphoma 6) fw 51015840-CGAATCCACACA-GGAGAGAAA-31015840 rev 51015840-ACGCGGTATTGCACCTTG-31015840

27 Cell Proliferation and Apoptosis Induction Cell viabilitywas determined by the T-cell cloning assay [44 46] at theend of 24 h 48 h and 72 h incubation in MMG and 1 gBriefly four 96-well U-bottomed microtiter plates with twoviable lymphocyteswell were seeded in medium CM in thepresence of 1 times 104 feeder cellswell (TK6 lymphoblastoidcells lethally irradiatedwith 40Gy of 120574-rays) Twoweeks laterthe plates were scored for growing colonies to calculate thecloning efficiency (CE) from the proportion of negative wellsassuming a Poisson distribution (CE = minusIn P

0N where P

0is

the fraction of wells without cells andN is the number of cellsseeded into wells) [56]

Apoptotic index was determined in PBLs incubated for24 and 48 h in MMG and in parallel in 1 g For detectionof apoptotic morphology PBLs were fixed and stained with


hsa-miR-9-5phsa-miR-9-3p

hsa-miR-155-5phsa-miR-125a-5p

hsa-let-7e-5phsa-miR-376c-3phsa-miR-99b-5p

hsa-miR-29b-1-5phsa-miR-132-3p

hsa-let-7i-3phsa-miR-376a-3phsa-miR-193b-3p

hsa-let-7i-5phsa-miR-146b-5p

hsa-miR-221-5phsa-miR-505-5phsa-miR-10a-5p

hsa-miR-7-5phsa-miR-625-5phsa-miR-629-5p

hsa-miR-200a-3phsa-miR-342-5phsa-miR-192-5phsa-miR-7-1-3phsa-miR-185-5p

hsa-miR-107hsa-miR-103a-3p



hsa-miR-135a-3phsa-miR-378a-5p

hsa-miR-663ahsa-miR-940


hsa-miR-34b-5phsa-miR-150-3p

hsa-miR-378a-3p

Expression valueminus4 minus2 0 2 4 6

(a)

hsa-miR-376a-3phsa-miR-376c-3phsa-miR-192-5phsa-miR-7-1-3phsa-miR-625-5phsa-miR-342-5phsa-miR-10a-5phsa-miR-505-5phsa-miR-221-5phsa-miR-7i-5phsa-miR-7-5phsa-miR-629-3p5phsa-miR-146b-5phsa-miR-200a-3phsa-miR-145-5phsa-miR-193b-3phsa-miR-7i-5phsa-miR-29b-1-5phsa-miR-1323phsa-miR-99b-5phsa-miR-125a-5phsa-miR-7e-5phsa-miR-155-5phsa-miR-9-3phsa-miR-9-5phsa-miR-150-3phsa-miR-378a-3p5phsa-miR-378a-5phsa-miR-181a-3phsa-miR-34a-5phsa-miR-34b-5phsa-miR-362-5phsa-miR-532-5phsa-miR-423-3phsa-miR-103a-3phsa-miR-107hsa-miR-185-5phsa-miR-223-3phsa-miR-940

G D E F I C L A P B H Mminus371 00 307

(b)

Figure 1 Differentially expressed miRNAs in human PBLs incubated in MMG (a) The expression level of each miRNA indicated as foldchange is the mean of the expression values obtained from the transformed log2 ratio (MMG1 g) (b) Dendrogram of miRNAs differentiallyexpressed in MMG The range of expression value is from minus37 (green downregulation) to 307 (red upregulation) Grey boxes correspondto not available (NA) fluorescent signal from the microarray platform

2 120583gmL 46-diamino-2-phenylindol (DAPI Roche) in anantifade solution (Vectashield Vector Lab) as previouslydescribed [57] At least 2000 cells were scored for each time-point by fluorescence microscopy (1000x magnification)The activation of caspase-3 was measured by the caspase-3fluorescent assay kit (Clontech BD Biosciences) at the endof incubation in 1 g or MMG as previously described [58]The fluorescent emission at 505 nm (excitation at 400 nm)of cleaved 7-amino-4-trifluoromethyl coumarin (AFC) wasmeasured with a PerkinElmer LS-50 B spectrofluorimeter

3 Results

31 Identification of miRNAs Affected by MicrogravitymiRNA expression profiling was performed on total RNAextracted from PBLs of twelve healthy donors at the end of

24 h incubation time in MMG and in 1 g conditions By com-paring the expression profile of MMG-incubated versus 1 gincubated PBLs of the same donor we found 42 differentiallyexpressed miRNAs 25 upregulated and 17 downregulatedfor which raw data and means of miRNA expression valuesare available at Supplementary Table S1 (see SupplementaryTable S1 in the Supplementary Material available online athttpdxdoiorg1011552014296747) miR-9-5p miR-9-3pand miR-155-5p were the most upregulated (46- 35- and24-fold resp) whereas miR-378a and miR-150-3p were themost downregulated (sim2-fold) (Figure 1)

32 Effect of Microgravity on Gene Expression Profile Geneexpression analysis was performed in PBLs incubated for 24 hin MMG and in 1 g By comparing the expression profilesof MMG-incubated and 1 g incubated PBLs we identified1581 differentially expressed genes inMMG versus 1 g among


0

200

400

600

800

1000

Num

ber o

f diff

eren

tially

expr

esse

d ge

nes

1200

Upregulated Downregulated

(a)

57

1919

19 19

19

ImmunityHemostasisLipid metabolic processSignal transductionMetabolismCell migrationdevelopment

Cell surface interactionsDiseaseNeuronal systemTransmembrane transportof small moleculesMuscle contraction

283

188151

132

94

(b)

Figure 2 Results of gene expression analysis Differentially expressed genes (a) and pie chart of biological process () containing pathwayssignificantly enriched in PBLs incubated in MMG (b)

which 465 (29) genes were upregulated whereas 1116 (71)genes were downregulated (Figure 2(a) and SupplementaryTable S2) By selecting a 2-fold cut-off threshold we iden-tified 312 (197) genes in MMG 157 genes (10) showedalterations in expression level with a fold change greater than40 and among these 20 genes showed a fold change ge160(Supplementary Table S3) To identify sets of genes withexpression changes in MMG condition we used Graphite[51] a novel web tool for topological-based pathway analysesbased on high-throughput gene expression data analysesPathway analysis on differentially expressed genes has beenperformed by using hypergeometric test on Reactome Path-ways as implemented inGraphiteweb considering significantthose categories with FDR lt 01 We evidenced biologicalpathways significantly enriched in MMG 15 (283) wererelated to immunity 10 (188) to hemostasis 7 (132) tolipid metabolic process and signal transduction 5 (94) tometabolism 3 (57) to cell migrationdevelopment and 1(19) tocell surface interactions disease neuronal systemtransmembrane transport of small molecules and musclecontraction (Figure 2(b)) The list of pathways is reported inSupplementary Table S4

Among the immune pathways significantly enriched inMMG those of ldquoMHC class II antigen presentationrdquo ldquoTollReceptor Cascadesrdquo and ldquoDAP12 signalingrdquo showed the high-est number of differentially expressed genes (30 20 and 20genes resp) Also the ldquointerferon gamma signalingrdquo pathwaywas significantly enriched in MMG with 13 differentiallyexpressed genes Ten genes codifying for proteins of MHCclass II (HLA-DRA HLA-DRB1 HLA-DRB3 HLA-DRB4

HLA-DRB5 HLA-DQA1 HLA-DQA2 HLA-DPA1 HLA-DPB1 and HLA-DQB1) were common to nine pathways(Figure 3) CD4 codifying for a membrane glycoprotein of Tlymphocytes that interacts with the major histocompatibilitycomplex class II antigens was common to eight immunepathways RELA PTEN and PAK1 were included in threepathways whereas HLA-DOA andHLA-DMBwere includedin two pathways

33 Target Prediction and Integration Analysis of miRNAand mRNA Expression Profiles We examined the regulatoryeffects of miRNAs on global gene expression under modeledmicrogravity (MMG) condition in comparison with groundgravity (1 g) To predict the target genes of differentiallyexpressed miRNAs in MMG we performed a computationalanalysis using TargetScan tool which predicts biologicaltargets of miRNAs by searching for the presence of conserved8mer and 7mer sites that match the seed region of eachmiRNA [59] However all available software for target pre-diction is characterized by a large fraction of false positivethus the integration of target predictions with miRNA andgene target expression profiles has been proposed to refinemiRNA-mRNA interactions The correlation analyses on thedifferentially expressed miRNAs and mRNAs were carriedout with MAGIA2 software [43] by microRNA Pearson pre-diction analysis which allowed the identification of miRNA-mRNA interactions (Supplementary Table S5) To discoverfunctional relationships between miRNAs and the transcrip-tome and uncover the gene pathways that are regulated


Gene symbol description Fold change Toll

rece

ptor

casc

ade

HLA-DPA1 Major histocompatibility complex class II DP alpha 1 HLA-DPB1 Major histocompatibility complex class II DP beta 1 HLA-DQA1 Major histocompatibility complex class II DQ alpha 1HLA-DQA2 Major histocompatibility complex class II DQ alpha 2HLA-DQB1 Major histocompatibility complex class II DQ beta 1 HLA-DRA Major histocompatibility complex class II DR alphaHLA-DRB1 Major histocompatibility complex class II DR beta 1 HLA-DRB3 Major histocompatibility complex class II DR beta 3 HLA-DRB4 Major histocompatibility complex class II DR beta 4 HLA-DRB5 Major histocompatibility complex class II DR beta 5 HLA-DOA Major histocompatibility complex class II DO alphaHLA-DMB Major histocompatibility complex class II DM betaCD4 CD4 moleculeRELA v-rel reticuloendotheliosis viral oncogene homolog APTEN Phosphatase and tensin homologPAK1 p21 protein (Cdc42Rac)-activated kinase 1

minus178minus183minus151minus164minus14minus155minus161minus185minus174minus159minus095minus147minus078051

minus026minus132

DA

P12

signa

ling

Cos

timul

atio

n by

the C

D28

fam

ily

TCR

signa

ling

Dow

nstre

am T

CR si

gnal

ing

MH

C cla

ss II

antig

en p

rese

ntat

ion

Gen

erat

ion

of se

cond

mes

seng

er m

olec

ules

Tran

sloca

tion

of Z

AP-70

to im

mun

olog

ical

syna

pse

Phos

phor

ylat

ion

of C

D3

and

TCR

zeta

chai

ns

Inte

rfero

n ga

mm

a sig

nalin

g

PD-1

signa

ling

Figure 3 Differentially expressed genes common to immune-related pathways identified by Reactome database in PBLs incubated inMMGThe expression value of each gene indicated as fold change is the mean of expression levels calculated as the log2 ratio (MMG1 g) on PBLsamples (see Supplementary Table S2)

by miRNAs in MMG we performed Gene Ontology (GO)analysis using DAVID [54]

In our analysis we used high classification stringency andconsidered only GO terms that have 119875 lt 01 after permu-tation corrections (Benjamini) (Table 1) Several GO termsbelonged to immune system function (ie ldquoinnate immuneresponserdquo ldquoinflammatory responserdquo ldquoregulation of cytokineproductionrdquo ldquopositive regulation of immune system processrdquoand ldquoresponse to bacteriumrdquo) in accordance with the resultsof pathway analysis on transcriptome (see SupplementaryTable S4) GO terms of ldquocell developmentrdquo ldquoregulation ofcell differentiationrdquo ldquoregulation of cell communicationrdquo ldquocellmotilityrdquo and ldquocell migrationrdquo were significantly enriched inMMG together with the category ldquoorgan developmentrdquo Inaddition the biological categories of ldquoregulation of signaltransductionrdquo ldquoregulation of response to stressrdquo ldquoregulationof cell deathrdquo and ldquoregulation of cell proliferationrdquo wereenriched in PBLs incubated in MMG

To determine whether different miRNAs within a GOcategory interact with the same target genes we performednetwork analysis using MAGIA2 [43] a software platformfor the visualization of complex miRNA-mRNA interactionsWe focused on miRNAs that correlated both positively andnegatively with the GO categories of immuneinflammatory

response regulation of programmed cell death and reg-ulation of cell proliferation As shown in Figure 4 mosttranscripts are associated with more than one miRNA as inthe case of TLR4 transcript correlated with eight differentmiRNAs (miR-10a-5p miR-7-5p miR-135a-3p miR-103a-3pmiR-7-1-3p miR-107 miR-629-5p and miR-362-5p)

By using Cytoscape [60] we visualized the func-tional interactions between miRNAs whose expression levelschanged the most in PBLs incubated in MMG such as miR-9-5p miR-9-3p miR-155-5p miR-150-3p and miR-378a-3pand correlated target genes involved in GO categories ofimmuneinflammatory response regulation of programmedcell death and regulation of cell proliferation (Figure 5)

Our results show that miR-155-5p correlates with IFNGIL17F BCL6 and RELA involved in immuneinflammatoryresponse with PTEN BNIP3L APAF1 and PDCD4 involvedin regulation of programmed cell death and with NKX3-1 involved in regulation of cell proliferation miR-150-3pcorrelates with immune-related genes (IFNG IL1A andHLA-DRB1) and with proapoptotic gene PDCD4 miR-9-3p correlates with genes regulating cell proliferation (NKX3-1 GADD45A and TP53BP1) apoptosis (APAF1 BNIP3L)and immunity (CCL7 CXCL5 and BCL6) Among genesenriched within the three functional categories miR-9-5p


Table 1 Selected GO terms of biological processes significantly affected by microgravity The complete list of GO terms can be found inSupplementary Table S6

GOID Term Count 119875 value Fold enrichment FDRGO0045087 Innate immune response 25 242 times 10minus5 2566978193 0040412626GO0009966 Regulation of signal transduction 88 167 times 10minus4 1471577879 027973741GO0048468 Cell development 62 171 times 10minus4 1610338849 0285253218GO0045595 Regulation of cell differentiation 54 268 times 10minus4 1648 0448057857GO0006954 Inflammatory response 41 323 times 10minus4 1787513228 0539044451GO0080134 Regulation of response to stress 37 543 times 10minus4 1806696296 090478009GO0010646 Regulation of cell communication 95 822 times 10minus4 1380599647 1366092587GO0048513 Organ development 139 822 times 10minus4 1290910116 1366713032GO0042060 Wound healing 26 838 times 10minus4 2025910165 1393334889GO0007165 Signal transduction 198 0001398603 1211224944 2313943814GO0001817 Regulation of cytokine production 27 0001406003 1926233766 2326052387GO0048514 Blood vessel morphogenesis 26 0001709187 193009009 2820918727GO0001817 Regulation of cytokine production 15 0001406003 1926233766 2326052387GO0007399 Nervous system development 88 0002097026 1359812471 3450524099GO0022603 Regulation of anatomical structure morphogenesis 28 0002408369 1831111111 3953172097GO0048870 Cell motility 33 0002838826 1710188679 464407324GO0048522 Positive regulation of cellular process 159 0003551259 1221594406 5777298149GO0002684 Positive regulation of immune system process 31 0003855111 1711490787 6256755641GO0050865 Regulation of cell activation 26 0003880839 1819447983 6297247375GO0051174 Regulation of phosphorus metabolic process 51 0003964604 1486259947 6428964883GO0001944 Vasculature development 28 0003981783 1767969349 6455956993GO0043066 Negative regulation of apoptosis 41 0004040803 156952381 6548633436GO0010941 Regulation of cell death 83 0004115909 1341019608 6666445839GO0009617 Response to bacterium 22 0004991824 19032021 8030143455GO0043067 Regulation of programmed cell death 82 000552374 1328770895 8849108723GO0008285 Negative regulation of cell proliferation 40 000576007 154741784 9210770389GO0010557 Positive regulation of macromolecule biosynthetic process 62 0006543793 139014966 104004889GO0016477 Cell migration 30 0006706505 1664646465 1064564555GO0042127 Regulation of cell proliferation 71 0009848433 1331149033 152576907GO0048523 Negative regulation of cellular process 139 0012514364 119870225 189945349

correlates with BCL6 miR-378a-3p is correlated with HLA-DRB1 GPNMB and NKX3-1 the same transcripts togetherwith IL17F are correlated also with miR-378a-5p

The microarray data from miRNA and gene expressionprofiling were validated by real-time qPCR experiments forfour miRNAs (miR-9-5p miR-155-5p miR-378a and miR-150-3p) and five mRNAs (IFNG IL17F BCL6 HLA-DRB1and TLR4) whose expression level was significantly alteredby MMG incubation (Figure 6) miR-9-5p and miR-155-5p together with IFNG IL17F and BCL6 transcripts wereupregulated in MMG whereas miR-378a and miR-150-3ptogether with HLA-DRB1 and TLR4 transcripts were down-regulated in MMG

34 Intraclass Integrated Analysis Recently Censi and col-leagues [61] observed a significant increase in the numberand strength of genes correlation under stress conditionssuch as disease and environmental or physiological changesTo evaluate whether the stress induced by MMG increasesthe amount of correlation of the system with respect to

1 g control condition we integrated mRNAs and miRNAsdata separately for MMG and 1 g using MAGIA2 By com-paring the two regulatory networks we observed a similarnumber of interactions between MMG (190 interactions)and 1 g (218 interactions) (Figure S1) indicating that therewas no significant connectivity enrichment under modeledmicrogravity By contrast GeneOntology analysis performedon MMG- and 1 g-specific interactions reported 50 GOcategories significantly enriched in only MMG condition(119875 lt 02 after Benjamini corrections Supplementary TableS7) 10 out of these were previously described in Table 1 Inparticular the GO categories ldquoregulation of cellular processrdquoand ldquocell differentiationrdquo were significantly affected byMMGWith intraclass analysis the GO categories of immunity andcell death were not enriched in MMG probably because inour study the number of PBL samples available for suchanalysis was relatively small On the whole the intraclassanalysis shows that modeled microgravity does not increasethe general connectivity of miRNA-gene interaction networkbut rather increases the transcriptome plasticity with respect


hsa-let-7i-3p

hsa-miR-200a-3p

TP73

CXCL10

hsa-miR-34a-5p

IRAK2

hsa-miR-663aIL18RAP

CCL19

hsa-miR-150-3p

BMP6

IL6

CXCL11CD180

CCR7

hsa-miR-378a-5p

FOS

FCN2

hsa-miR-125a-5p

hsa-miR-132-3pPDPN

hsa-miR-505-5phsa-miR-34b-5p

hsa-miR-155-5p

hsa-miR-9-3p PRDX2

ATRN

FN1

SPP1HMOX1

RELAhsa-miR-7-5p

hsa-let-7i-5p

LYZ

CLEC7AVSIG4

hsa-miR-10a-5p

hsa-miR-940

hsa-miR-362-5p

hsa-miR-532-5pCD55

hsa-miR-9-5p

hsa-miR-146b-5p

IL1RAPhsa-miR-7-1-3p

hsa-miR-103a-3p

CXCL1

CIITA

TLR4

TLR7

hsa-miR-629-5p

hsa-miR-107

MEFV

hsa-miR-376a-3p

hsa-miR-376c-3p

IL17F

IL1A

hsa-let-7e-5phsa-miR-221-5p

hsa-miR-185-5p

CCL7

hsa-miR-135a-3p

SIGLEC1 TLR5

LIPA AIF1

CFP


hsa-miR-1225-5p

C1QCF2R

C5

C1QB

hsa-miR-378a-3p

(a) Immuneinflammatory response

GIMAP5

DDAH2 HMOX1

SOX4PRDX1

hsa-miR-505-5p

hsa-miR-223-3p

SNCA hsa-miR-221-5p

hsa-miR-192-5p

UNC13B

PTPRFRNF7

TXNIPAPAF1

RUNX3

MGMTPTENNOTCH2IFNG

MPOETS1

NRG1

PRDX2

ANXA4

TNFSF13

NEFL GPX1NAIP

BNIP3L hsa-miR-185-5p

STK17ATLR4


hsa-miR-135a-3p

hsa-miR-99b-5p

IL12A

ITGA1CHD8

RAG1

UACASH3RF1

IL1A

HGF

KCNMA1

HTATIP2

SMOCAMK1D

IL6NOL3

AVEN

APOE

hsa-miR-663a

hsa-miR-200a-3p

hsa-miR-378a-5phsa-let-7i-5phsa-miR-532-5p

hsa-miR-376c-3phas-miR-362-5p

TAF9B

LTB

RASGRF2

LGALS1

TRIO

MEF2CBCL6

NQO1

ERBB3

AIFM3

TRAF1

FURIN

NR4A2

SERPINB2

FGD2

PIM2

NR4A1

TNFSF14

TIMP3

CEBPG

FGD4CARD6

RAB27A

ARHGEF3NF1

PPT1

F2RRELA

VAV2

SMAD6

ESR1

BMF ITSN1

MUC2

FOXO3

hsa-miR-132-3p

hsa-miR-423-3p

hsa-miR-9-5p

hsa-miR-146b-5p


hsa-miR-378a-3p

hsa-miR-629-5p

SLC25A4

hsa-let-7e-5p

MAP3K5

hsa-miR-181a-3p

TP73hsa-let-7i-3p

PPP1R13Bhsa-miR-9-3p

hsa-miR-940

hsa-miR-7-1-3p

hsa-miR-625-5p


hsa-miR-342-5p

hsa-miR-7-5p

hsa-miR-155-5p

hsa-miR-29b-1-5p

hsa-miR-10a-5p

(b) Regulation of programmed cell deathLTB

VSIG4

hsa-a-423-3p

SMO

hsa-let-7i-5p

hsa-miR-629-3p

hsa-let-7e-5p

VEGFB

NCK2

hsa-miR-505-5pHMOX1

IL12RB2 IL6CEBPA

CD86

CYP27B1

hsa-miR-7-5p

RARRES1

hsa-miR-940

hsa-miR-629-5p

hsa-miR-146b-5p

KLF11

RAC2

APOEPDCD1LG2

IFI30

F2R

hsa-miR-342-5p

hsa-miR-625-5pIL1A

SYKKLF4

HHEX

BCL6

AIF1

PDGFRB

IL12A

IFNGCTTNBP2

GPNMB

EREG

HES1

ETS1

hsa-miR-378a-5p

hsa-miR-155-5p

hsa-miR-378a-3p

NKX3-1

PPARG

CD33

NOTCH4

hsa-miR-221-5p

hsa-miR-135a-3p

ERBB3

CXCL10

COMTTNFRS13B

hsa-miR-34a-5p

hsa-miR-376c-3p

TNFSF13

hsa-miR-1225-5p

GPX1

hsa-miR-132-3p

hsa-miR-185-5p CXCL5CXCL1

KLF5FIGF

SOX2

KIFAP3 DLG3

hsa-miR-532-5p

hsa-miR-192-5p

hsa-miR-193b-3pPLAU

PTGS1PTEN

PDGFBIRS1

hsa-miR-10a-5p

SOX4NDN

hsa-miR-223-3p

hsa-miR-29b-1-5p

NOTCH2

hsa-miR-7-1-3pTXNIP

hsa-miR-107

CD9

NF1

NRG1

SLAMF1

RUNX3

PTPRF

MUC2

hsa-miR-125a-5p

hsa-miR-103a-3p

CHRNA10

RELA

PDGFCIL13RA1

hsa-miR-9-3p

hsa-miR-34b-5p hsa-miR-663a

GRN

hsa-miR-9-5p

hsa-miR-362-5pTIMP2

(c) Regulation of cell proliferation

Figure 4 Network analysis on correlated miRNA-mRNA pairs in PBLs incubated in MMG Network analyses were performed by MAGIA2software using miRNAs correlating both positively and negatively with transcripts involved in immuneinflammatory response (a) inregulation of programmed cell death (b) and in regulation of cell proliferation (c) Circles represent transcripts and triangles representmiRNAs

to 1 g gravity condition as evidenced by the enriched GOcategories

35 In Vitro Validation of GO Analysis Effects of MMG onCell Proliferation and Apoptosis Experimental assays wereperformed to validate the results obtained by bioinformaticsanalyses on the GO categories ldquoregulation of cell prolifera-tionrdquo and ldquoregulation of programmed cell deathrdquo associatedwith variations in miRNA expression under MMG To mea-sure cell proliferation quiescent (119866

0) PBLs from the same

donorwere incubated for different times (24 h 48 h and 72 h)in 1 g and inMMGAt the end of incubation times the colony

forming ability has been determined by the T-cell cloningassay [44] in which cells were incubated in medium contain-ing mitogen factors (ie PHA and IL2) to trigger their cellcycle entry Our results showed that cloning efficiency (CE)decreased with time in both gravity conditions howeverMMG incubation affected the ability of PBLs to form colonies(119875 lt 005 at 24 h Figure 7(a)) To investigate whether MMGincubation increased the frequency of apoptotic cells PBLswere scored for the presence of apoptotic bodies Apoptoticindex was very similar at 24 and 48 h and significantlyhigher in PBLs incubated in MMG than in 1 g (Figure 7(b)119875 lt 005) In the same PBL samples caspase-3 activationassayed by the cleavage of the peptide substrate DEVD-AFC


Expression value

Regu

latio

n of

cell

prol

ifera

tion

Regu

latio

n of

pro

gram

med

cell

deat

hIm

mun

ein

flam

mat

ory

resp

onse

hsa-miR-155-5pIFNG

hsa-miR-125a-5p hsa-miR-940

hsa-miR-150-3p

hsa-miR-150-3p

hsa-miR-9-3p

CCL7hsa-let-7e-5p hsa-miR-7-5p

hsa-miR-135a-3p

hsa-miR-378a-5p

hsa-miR-135a-3p

IL17F

hsa-miR-155-5p

hsa-miR-9-3p

CXCL5

IL1A

hsa-miR-7-5pRELA

hsa-miR-155-5p hsa-miR-155-5p

hsa-miR-378a-5p hsa-miR-150-3p

HLA-DRB1hsa-miR-378a-3p hsa-miR-629-5p

hsa-miR-9-3p hsa-miR-9-5pBCL6

hsa-miR-629-5p

hsa-miR-34a-5p

hsa-miR-185-5p

hsa-miR-223-3p

hsa-miR-107

hsa-miR-185-5pPTEN

hsa-miR-10a-5p hsa-miR-7-1-3p

hsa-miR-29b-1-5p


hsa-miR-7-5p

hsa-miR-9-3p

hsa-miR-7-1-3p

hsa-miR-29b-1-5p

hsa-let-7e-5p

hsa-miR-505-5pAPAF1


hsa-miR-221-5p

hsa-miR-376a-3p

hsa-miR-9-3pTP53BP1

hsa-miR-9-3p

hsa-miR-29b-1-5p

GADD45A


hsa-miR-9-3p

hsa-miR-378a-3p

hsa-miR-378a-3p

hsa-miR-155-5p

hsa-miR-1225-5p

hsa-miR-185-5phsa-miR-185-5pNKX3-1

hsa-miR-107GPNMB


PDCD4

hsa-miR-200a-3p

hsa-miR-135a-3p

hsa-miR-532-5p

hsa-miR-629-5p

hsa-miR-362-5p

minus51 0 75

hsa-miR-378a-5p

BNIP3Lhsa-miR-155-5p

hsa-let-7i-5p

Figure 5 Cytoscape visualization of miRNA-mRNA correlations in PBLs incubated 24 h in MMG Relationships between miRNAs andcorrelated target genes involved in ldquoimmuneinflammatory responserdquo (IFNG CCL7 IL17F HLA-DRB1 IL1A CXCL5 RELA and BCL6)ldquoregulation of programmed cell deathrdquo (PTEN BNIP3L APAF1 and PDCD4) and ldquoregulation of cell proliferationrdquo (GPNMB NKX3-1TP53BP1 and GADD45A) Circles represent transcripts and triangles represent miRNAs the expression levels of each feature are representedas color scale


0

05

1

15

2

25

3

35

4

45

miR-9-5p miR-155-5p

MMG

Relat

ive e

xpre

ssio

n

lowastlowastlowast

lowastlowast

1g

0

02

04

06

08

1

12

miR-378a miR-150-3p

Relat

ive e

xpre

ssio

n

lowastlowastlowast lowastlowastlowast

MMG1g

(a)

0

5

10

15

20

25

IFNG IL17F BCL6

80

120

140

Relat

ive e

xpre

ssio

n

lowast

lowastlowastlowast

MMG1g

0

02

04

06

08

1

12

TLR4 HLA-DRB1

Relat

ive e

xpre

ssio

nlowastlowastlowast lowastlowastlowast

MMG1g

(b)

Figure 6 Microarray data validation by quantitative real-time PCR (qRT-PCR) Validation of microarray data by qRT-PCR in MMG-incubated versus 1 g incubated PBLs The results are consistent with the cumulative microarray data of miRNAs (a) and mRNAs (b) Values(fold change dark grey bars) are means plusmn SE of expression levels calculated as the log2 (MMG1 g) on PBL samples from 4 to 6 differentdonors The value ldquo1rdquo of control 1 g PBLs (light grey bars) is arbitrarily given when no change is observed ( lowastlowastlowast119875 lt 0001 lowastlowast119875 lt 001 andlowast

119875 lt 005 t-test)

increased significantly in PBLs incubated 48 h in MMG withrespect to those in 1 g (Figure 7(c) 119875 lt 005)

4 Discussion

In the present study we evaluated the effects of modeledmicrogravity (MMG) on human PBLs by analyzing miRNAand gene expression profiles in comparison with PBLs cul-tured in Earth gravity condition (1 g) Our results reported 42differentially expressed miRNAs in PBLs cultured for 24 h inMMGwith respect to 1 g of which 14 (miR-34a-5p miR-34b-5p miR-663a miR-135a-3p miR-1225-5p miR-940 miR-221-5p miR-29b-1-5p miR-10a-5p let-7i-3p miR-200a-3p miR-7-5p miR-7-1-3p and miR-505-5p) were found altered also

by 120574-irradiation as assessed in our previous study [47] Themost dysregulatedmiRNAs identified in the present work arethe upregulated miR-9-5p miR-9-3p and miR-155-5p andathe downregulated ones are miR-150-3p and miR-378a-3pSuch miRNAs have been found altered in human tumors inparticular miR-9 is an oncogenic miRNA overexpressed inmixed lineage leukemia- (MLL-) rearranged acute myeloidleukemia [62] in muscle-invasive bladder cancer [63] andin osteosarcoma cell lines [64] miR-155 is commonly upreg-ulated in hematological malignancies [65 66] and has beenlinked to the development of breast lung and stomachtumors [67ndash70] miR-150 is significantly downregulated inmost cases of acute myeloid leukemia [71] and colorectalcancer [72] in addition miR-150 has an important role in


25

20

15

10

5

0

lowast

24 48 72

Incubation time (hrs)

Clon

ing e

fficie

ncy (

)

1gMMG

(a)

)

5

4

3

2

1

024 48


Apop

totic

inde

x (

1gMMG

lowast lowast

(b)

60

50

40

30

20

10

0

24 48

70


Casp

ase-3

activ

ation

(au

)

lowast

1gMMG

(c)

Figure 7 Cell proliferation and apoptosis induction in human PBLs incubated in MMG and in 1 g (a) T-cell cloning assay performed atthe end of 24 h 48 h and 72 h of incubation in the two gravity conditions Data are means plusmn SE from thirteen independent experiments(b) Apoptotic index at the end of incubation for 24 h and 48 h in MMG and 1 g determined by nuclear chromatin condensation with DAPIstaining (c) Caspase-3 activation at the end of 24 h and 48 h incubation in MMG and 1 g assessed by fluorimetric assay (au arbitrary units)Data in (b) and (c) are means plusmn SE from 3-4 independent experiments (lowast119875 lt 005 t-test)

normal hematopoiesis and its aberrant downregulation is asensitivemarker indicative of lymphocyte depletion and bonemarrow damage [73] miR-378 (actually annotated as miR-378a) is significantly downregulated in colorectal cancer [74]in cutaneous squamous cell carcinoma [75] and in renalcell carcinoma [76] The effects of microgravity on miRNAexpression profile are currently reported in only one studycarried out in human lymphoblastoid TK6 cells incubatedunder simulated microgravity for 72 h [77] Among thedysregulated miRNAs only two were common to our datamiR-150 and miR-34a although the direction and intensityof fold change were different demonstrating the cell typespecific signature of miRNA profile

miRNAs modulate gene expression by interacting withthe 31015840UTR of target genes and since a single miRNA couldhave hundreds to thousands of predicted target genes [25]it is difficult to determine the true target regulated by themiRNA which affects a biological function Moreover bind-ing of multiple miRNAs to one target could further increasethe complexity of target prediction The identification ofmiRNA target genes is usually performed by bioinformaticprediction algorithms based on (i) sequence similarity searchpossibly considering target site evolutionary conservationand (ii) thermodynamic stability However it is known thatthe results of target prediction algorithms are characterizedby very low specificity [78] For this purpose the integrationof target predictions with miRNA and gene expression pro-files has been recently proposed to improve the detection offunctional miRNA target relationships [79 80] Therefore toidentify the most likely target genes of miRNAs differentiallyexpressed in MMG we defined gene expression signatureon the same samples of PBLs assayed for miRNA profilingthen we integrated expression profiles from both miRNAsand mRNAs with in silico target predictions to reducethe number of false positives and increase the number of

biologically relevant targets [81ndash83] Our results of geneexpression profiling reported the downregulation of multiplegenes in MMG (71) in accordance with previous findingsin activated human T lymphocytes incubated for 24 h insimulated microgravity [21] Moreover we found that about20 of genes responded to MMG by more than 2-foldchange in expression level and twenty genes showed a ge16-fold change in expression Most of these top dysregulatedgenes were immune-related such as those codifying forinflammatory cytokines (CCL1 CCL7 CXCL5 CXCL11 andIL1A) and for proteins with a role in immunoregulatoryfunctions (IFNG TNIP3 TREM1 APOC1 FCN1 FCN2and CPVL) (Supplementary Table S3) Biological pathwaysenriched in PBLs exposed to MMG were mainly involved inimmunity (Figure 2(b)) including adaptive immune systemresponse (ie PD-1 signaling phosphorylation of CD3 andTCR zeta chains translocation of ZAP-70 to immunologicalsynapse MHC class II antigen presentation TCR signalingand costimulation by the CD28 family) innate immunesystem response (ie Toll Receptor Cascades) and cytokinesignaling in immune system (ie interferon gamma signal-ing) (Supplementary Table S4) All these pathways includedten downregulated genes codifying for MHCmolecules classII (HLA-DPA1 HLA-DPB1 HLA-DQA1 HLA-DQA2 HLA-DQB1 HLA-DRA HLA-DRB1 HLA-DRB3 HLA-DRB4and HLA-DRB5) which are expressed in antigen presentingcells (APC) and play a central role in the immune system bypresenting peptides derived from extracellular proteins [84]Therefore the downregulation of these genes suggests that thedisplay of antigens at the cell surface of APCmay be disturbedby gravity reduction affecting the efficiency of immuneresponse as observed in astronauts during spaceflight andimmediately afterwards [85ndash87] Moreover our data are inaccordance with the inhibition of immediate early genesin T-cell activation observed in space microgravity [14]


and with alterations of gene expression in human activatedT-cells incubated in modeled microgravity including thedownregulation of HLA-DRA gene [21]

By integrating the transcriptome and microRNAomewe detected significant miRNA-mRNA relationships underMMG Since miRNAs act prevalently through target degra-dation expression profiles of miRNAs and target genesare generally expected to be inversely correlated Neverthe-less since miRNA activity is part of complex regulatorynetworks and gene expression profiles are the result ofdifferent levels of regulation also positive correlation (ieupregulated miRNAupregulated mRNA or downregulatedmiRNAdownregulated mRNA) is expected Indeed in an invivo mouse model the activated expression of miRNAs hasbeen shown to correlate with activated expression of mRNAsrather thanwithmRNAdownregulation [88] GeneOntology(GO) analysis conducted on the significantly correlatedmiRNA-mRNA pairs evidenced the biological categoriessignificantly overrepresented in MMG (Table 1) Many GOterms of immune response were enriched such as ldquoinnateimmune responserdquo ldquoinflammatory responserdquo ldquoregulation ofcytokine productionrdquo ldquopositive regulation of immune systemprocessrdquo and ldquoresponse to bacteriumrdquo Notably the mostdysregulated miRNAs detected in the present study miR-378a-3p miR-150-3p miR-155-5p miR-9-3p and miR-9-5pare significantly correlated with immune-related genes Inparticular miR-378a-3p is positively correlated with tran-scripts of MHC molecules class II such as HLA-DRB1(Figure 5) and HLA-DOA HLA-DRB5 and HLA-DQA2(not shown) miR-150-3p is negatively correlated with HLA-DRB1 IFNG and IL1Atranscripts whereas miR-155-5p ispositively correlated with IFNG and IL17F and negativelycorrelated with RELA and BCL6 (Figure 5) IL1A IL17Fand IFN-120574 are proinflammatory cytokines acting during theimmune response IFN-120574 is a soluble cytokine having broaderroles in activation of immune responses in part throughupregulating transcription of genes involved in antigen pro-cessingpresentation in cell cycle regulation and apoptosisand its correlation with miR-155 has been recently validated[89] BCL6 which encodes a nuclear transcriptional repres-sor has a role not only in regulation of lymphocyte functionbut also in cell survival and differentiation Similarly thepleiotropic transcription factor RELA has a role in immunebiological process and it is also involved in cell growth andapoptosis Besides miR-155-5p BCL6 is correlated with fourmiRNAs including miR-9 (3p and 5p) in addition miR-9-3p is positively correlated with CCL7 and CXCL5 (Figure 5)Recent evidences show that miR-9 is highly involved inimmunity and inflammatory diseases [90ndash92] by enhancingIFN-120574 production in activated human CD4(+) T-cells [91]Moreover Gao et al [90] have shown that miR-9 disturbsthe display of antigens at the cell surface by suppressing theexpression of MHC class I gene transcription

Together with the categories of immune response GOanalysis reported that also the categories of regulation of cellproliferation and regulation of programmed cell death weresignificantly enriched in MMG as previously reported in 120574-irradiated PBLs [47] Interestingly such categories were notenriched from pathway analysis conducted on transcriptome

and the reason could be that integrated analysis of miRNAsand mRNAs expression profiles evidences the posttran-scriptional effect mediated by miRNAs on gene expressionAmong genes involved in cell proliferation the transcriptionfactor NKX3-1 (24-fold upregulated) which mediates non-cell autonomous regulation of gene expression and inhibitscell proliferation is correlated with miR-9-3p miR-155-5pmiR-378a-3p and miR-378-5p (Figure 5) TP53BP1 (14-foldupregulated) encoding for a chromatin-associated factorinvolved in cell cycle checkpoint and growth is correlatedwith miR-9-3p GADD45A (15-fold upregulated) regulatingcell cycle arrest DNA repair cell survival senescence andapoptosis is also correlated with miR-9-3p GPNMB (32-fold downregulated) expressed in a wide array of normaltissues such as bone hematopoietic system and skin whereit influences cell proliferation adhesion differentiation andsynthesis of extracellular matrix proteins [93] is targetedby five miRNAs including miR-378a-3p Among miRNA-correlated genes involved in apoptosis we identified PDCD4(proapoptotic 14-fold upregulated) and found out that it iscorrelated with seven miRNAs including miR-155-5p andmiR-150-3p BNIP3L (proapoptotic 15-fold downregulated)also correlated with seven miRNAs including miR-155-5pandmiR-9-3p APAF1 (proapoptotic 13-fold downregulated)correlated with four miRNAs including miR-155-5p andmiR-9-3p and PTEN (proapoptotic) correlated with sevenmiRNAs including miR-155-5p Notably many transcripts(ie BCL6 PTEN BNIP3L PDCD4 NKX3-1 and GPNMB)are targeted by multiple miRNAs indicating a pleiotropiceffect in gene regulation by coexpressed endogenousmiRNAsin MMG Moreover our results suggest that under MMGcondition a small group of miRNAs regulates transcriptomeby modulating the same transcripts within one pathway

To validate the results of Gene Ontology analysis weevaluated whether the GO categories ldquoregulation of cell pro-liferationrdquo and ldquoregulation of programmed cell deathrdquo wereaffected byMMG incubation by performing biological assaysof T-cell cloning and apoptosis induction Our results showthat cloning ability of PBLs was lower after 24 h incubationin MMG than in 1 g in accordance with the suppression ofproliferative response of human lymphocytes to mitogenicstimulation in microgravity [94 95] The ability to originateclones in PBLs incubated for 48h and 72 h decreased in bothgravity conditions probably because the longer119866

0-phase con-

dition experienced by PBLs affected their responsiveness toenter into cell cycleThe antiproliferative effect of micrograv-ity has been recently reported also in human thyroid cancercells [96] and in human lung adenocarcinoma cells [97]Our results of apoptosis induction demonstrated that boththe formation of apoptotic bodies activation and caspase-3activation increased significantly in PBLs incubated inMMGthan in 1 g The activation of apoptotic process seems relatedto the overexpression of PDCD4 and RELA rather than tothe underexpression of BNIP3L and APAF1 indicating theexistence of complex regulatory networks between miRNAsand mRNAs that occur at different levels of regulation IFN-120574 besides having an important role in activating innateand adaptive immune responses plays important roles in


inhibiting cell proliferation and inducing apoptosis Its over-expression in MMG mediated by miR-9-3p and miR-155-5p could thus mediate the antiproliferative effect and theapoptosis induction In addition the correlation betweenmiR-9-3p and TP53BP1 could explain the clonogenicitydecrease and apoptosis increase in PBLs incubated in MMGIndeed overexpression of TP53BP1 has been demonstrated todecrease the clonogenicity and induce apoptosis in ovariancancer cells [98]

5 Conclusions

Our results show that MMG leads to changes in expressionlevel of a considerable fraction of microRNAome and tran-scriptome in human PBLs miRNAs differentially expressedin MMG are correlated with immuneinflammatory-relatedgenes as IFNG accordingly with the important role ofmiRNAs in immune function regulation Since inflammationworks as a tumor-promoting agent the abnormal expressionof such miRNAs under microgravity condition could influ-ence the carcinogenic process by affecting cancer cell immuneescape Moreover miRNAs mostly dysregulated in MMGsuch as miR-9 miR-155 and miR-150 are oncogenic sug-gesting that their abnormal expression can influence the car-cinogenic process The results of miRNA-mRNA integrationanalysis demonstrate that MMG increases the transcriptomeplasticity compared with 1 g condition and that categories ofregulation of cell proliferation and programmed cell deathare affected by MMG as confirmed by in vitro experimentalvalidation Taken togetherour results of high-throughputexpression analysis and miRNA-mRNA integration analysisgive new insight into the complex genetic mechanisms of cellresponse to stress environment under reduced gravity




M Mognato and C De Pitta conceived and designed theexperiments C Girardi S Casara and M Mognato per-formed the experiments C De Pitta C Romualdi E CaluraL Celotti and M Mognato analyzed the data M Mognatoand L Celotti wrote the paper Cristina Girardi and CristianoDe Pitta contributed equally to this work

Acknowledgments

The authors gratefully acknowledge M De Bernard forcritical discussion and R Mazzaro for graphical supportThis work was done with the support of the Italian SpaceAgency (ASI XMAB fromMolecules to Man 1014060) toL Celotti and of the University of Padova (CPDA061783) toMMognatoThe authors also apologize to the authors whosework could not be cited due to space limitations

References

[1] R H Fitts D R Riley and J J Widrick ldquoFunctional and struc-tural adaptations of skeletal muscle to microgravityrdquo Journal ofExperimental Biology vol 204 no 18 pp 3201ndash3208 2001

[2] M Narici B Kayser P Barattini and P Cerretelli ldquoEffects of 17-day spaceflight on electrically evoked torque and cross-sectionalarea of the human triceps suraerdquo European Journal of AppliedPhysiology vol 90 no 3-4 pp 275ndash282 2003

[3] S Trappe D Costill P Gallagher et al ldquoExercise in spacehuman skeletal muscle after 6 months aboard the InternationalSpace Stationrdquo Journal of Applied Physiology vol 106 no 4 pp1159ndash1168 2009

[4] S I M Carlsson M T S Bertilaccio E Ballabio and J AMMaier ldquoEndothelial stress by gravitational unloading effectson cell growth and cytoskeletal organizationrdquo Biochimica etBiophysica Acta vol 1642 no 3 pp 173ndash179 2003






[10] B E Crucian R P Stowe D L Pierson and C F SamsldquoImmune system dysregulation following short- vs long-duration spaceflightrdquo Aviation Space and Environmental Medi-cine vol 79 no 9 pp 835ndash843 2008

[11] G Sonnenfeld J S Butel and W T Shearer ldquoEffects of thespace flight environment on the immune systemrdquo Reviews onEnvironmental Health vol 18 no 1 pp 1ndash17 2003

[12] G Sonnenfeld ldquoEditorial space flight modifies T cellactivationmdashrole of microgravityrdquo Journal of Leukocyte Biologyvol 92 no 6 pp 1125ndash1126 2012

[13] A Semov N Semova C Lacelle et al ldquoAlterations in TNF-and IL-related gene expression in space-flown WI38 humanfibroblastsrdquoTheFASEB Journal vol 16 no 8 pp 899ndash901 2002

[14] T T Chang I Walther C-F Li et al ldquoThe RelNF-120581B pathwayand transcription of immediate early genes in T cell activationare inhibited bymicrogravityrdquo Journal of Leukocyte Biology vol92 no 6 pp 1133ndash1145 2012


[16] S J PardoM J PatelMC Sykes et al ldquoSimulatedmicrogravityusing the Random Positioning Machine inhibits differentiationand alters gene expression profiles of 2T3 preosteoblastsrdquoAmerican Journal of Physiology vol 288 no 6 pp C1211ndashC12212005


[17] M Monticone Y Liu N Pujic and R Cancedda ldquoActivationof nervous system development genes in bone marrow derivedmesenchymal stem cells following spaceflight exposurerdquo Jour-nal of Cellular Biochemistry vol 111 no 2 pp 442ndash452 2010




[21] N E Ward N R Pellis S A Risin and D Risin ldquoGeneexpression alterations in activated human T-cells induced bymodeledmicrogravityrdquo Journal of Cellular Biochemistry vol 99no 4 pp 1187ndash1202 2006

[22] J Q Clement S M Lacy and B L Wilson ldquoGene expres-sion profiling of human epidermal keratinocytes in simulatedmicrogravity and recovery culturesrdquo Genomics Proteomics andBioinformatics vol 6 no 1 pp 8ndash28 2008

[23] R Kumari K P Singh and J W DuMond Jr ldquoSimulatedmicrogravity decreases DNA repair capacity and induces DNAdamage in human lymphocytesrdquo Journal of Cellular Biochem-istry vol 107 no 4 pp 723ndash731 2009


[25] J Krutzfeldt M N Poy andM Stoffel ldquoStrategies to determinethe biological function of microRNAsrdquoNature Genetics vol 38no 1 pp S14ndashS19 2006

[26] C A Nickerson C M Ott J W Wilson et al ldquoLow-shearmodeled microgravity a global environmental regulatory sig-nal affecting bacterial gene expression physiology and patho-genesisrdquo Journal of Microbiological Methods vol 54 no 1 pp1ndash11 2003

[27] K Arunasri M Adil K Venu Charan C Suvro S HimabinduReddy and S Shivaji ldquoEffect of simulated microgravity on Ecoli K12 MG1655 growth and gene expressionrdquo PLoS ONE vol8 no 3 Article ID e57860 2013

[28] O Marcu M P Lera M E Sanchez et al ldquoInnate immuneresponses of Drosophila melanogaster are altered by space-flightrdquo PLoS ONE vol 6 no 1 Article ID e15361 2011

[29] Y Honda A Higashibata Y Matsunaga et al ldquoGenes down-regulated in spaceflight are involved in the control of longevityin Caenorhabditis elegansrdquo Scientific Reports vol 2 article 4872012


[31] D P Bartel ldquoMicroRNAs target recognition and regulatoryfunctionsrdquo Cell vol 136 no 2 pp 215ndash233 2009

[32] H Guo N T Ingolia J S Weissman and D P BartelldquoMammalian microRNAs predominantly act to decrease targetmRNA levelsrdquo Nature vol 466 no 7308 pp 835ndash840 2010

[33] V Huang Y Qin J Wang et al ldquoRNAa is conserved inmammalian cellsrdquo PLoS ONE vol 5 no 1 Article ID e88482010

[34] L B Frankel N R Christoffersen A Jacobsen M Lindow AKrogh and A H Lund ldquoProgrammed cell death 4 (PDCD4)is an important functional target of the microRNA miR-21 inbreast cancer cellsrdquoThe Journal of Biological Chemistry vol 283no 2 pp 1026ndash1033 2008

[35] M N Poy M Spranger and M Stoffel ldquomicroRNAs and theregulation of glucose and lipid metabolismrdquo Diabetes Obesityand Metabolism vol 9 no 2 pp 67ndash73 2007

[36] N Stern-Ginossar N Elefant A Zimmermann et al ldquoHostimmune system gene targeting by a viral miRNArdquo Science vol317 no 5836 pp 376ndash381 2007

[37] C J Marsit K Eddy and K T Kelsey ldquoMicroRNA responsesto cellular stressrdquo Cancer Research vol 66 no 22 pp 10843ndash10848 2006

[38] P M Voorhoeve C le Sage M Schrier et al ldquoA geneticscreen implicates miRNA-372 and miRNA-373 as oncogenes intesticular germ cell tumorsrdquo Cell vol 124 no 6 pp 1169ndash11812006

[39] E A C Wiemer ldquoThe role of microRNAs in cancer no smallmatterrdquo European Journal of Cancer vol 43 no 10 pp 1529ndash1544 2007

[40] W C S Cho ldquoOncomiRs the discovery and progress ofmicroRNAs in cancersrdquo Molecular Cancer vol 6 article 602007

[41] S Mi J Lu M Sun et al ldquoMicroRNA expression signaturesaccurately discriminate acute lymphoblastic leukemia fromacute myeloid leukemiardquo Proceedings of the National Academyof Sciences of the United States of America vol 104 no 50 pp19971ndash19976 2007

[42] S M Hammond ldquoMicroRNAs as tumor suppressorsrdquo NatureGenetics vol 39 no 5 pp 582ndash583 2007

[43] A Bisognin G Sales A Coppe S Bortoluzzi andC RomualdildquoMAGIA2 from miRNA and genes expression data integrativeanalysis to microRNA-transcription factor mixed regulatorycircuits (2012 update)rdquoNucleic Acids Research vol 40 no 1 ppW13ndashW21 2012

[44] M Mognato and L Celotti ldquoModeled microgravity affects cellsurvival and HPRT mutant frequency but not the expressionof DNA repair genes in human lymphocytes irradiated withionising radiationrdquoMutation Research vol 578 no 1-2 pp 417ndash429 2005


[46] S-M Hou F J Van Dam F De Zwart et al ldquoValidation ofthe human T-lymphocyte cloning assaymdashring test report fromthe EU concerted action on HPRT mutation (EUCAHM)rdquoMutation Research vol 431 no 2 pp 211ndash221 1999

[47] C Girardi C de Pitta S Casara et al ldquoAnalysis of miRNAandmRNA expression profiles highlights alterations in ionizingradiation response of human lymphocytes under modeledmicrogravityrdquo PLoS ONE vol 7 no 2 Article ID e31293 2012

[48] H Wang R A Ach and B O Curry ldquoDirect and sensitivemiRNA profiling from low-input total RNArdquo RNA vol 13 no1 pp 151ndash159 2007



[50] V G Tusher R Tibshirani and G Chu ldquoDiagnosis of multiplecancer types by shrunken centroids of gene expressionrdquo Pro-ceedings of the National Academy of Sciences of the United Statesof America vol 98 pp 5116ndash5121 2001

[51] G Sales E Calura P Martini and C Romualdi ldquoGraphite webweb tool for gene set analysis exploiting pathway topologyrdquoNucleic Acids Research vol 41 pp 89ndash97 2013

[52] F Xin M Li C Balch et al ldquoComputational analysis ofmicroRNA profiles and their target genes suggests significantinvolvement in breast cancer antiestrogen resistancerdquo Bioinfor-matics vol 25 no 4 pp 430ndash434 2009

[53] HWang andW-H Li ldquoIncreasingMicroRNA target predictionconfidence by the relative R2 methodrdquo Journal of TheoreticalBiology vol 259 no 4 pp 793ndash798 2009

[54] D W Huang B T Sherman and R A Lempicki ldquoSystematicand integrative analysis of large gene lists using DAVID bioin-formatics resourcesrdquo Nature Protocols vol 4 no 1 pp 44ndash572009

[55] K J Livak and T D Schmittgen ldquoAnalysis of relative geneexpression data using real-time quantitative PCR and the2minusΔΔ119862T methodrdquoMethods vol 25 no 4 pp 402ndash408 2001

[56] R J Albertini K L Castle and W R Borcherding ldquoT-cellcloning to detect the mutant 6-thioguanine-resistant lympho-cytes present in human peripheral bloodrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 79 no 21 I pp 6617ndash6621 1982

[57] MMognato CGirardi S Fabris and L Celotti ldquoDNA repair inmodeledmicrogravity double strand break rejoining activity inhuman lymphocytes irradiated with 120574-raysrdquoMutation Researchvol 663 no 1-2 pp 32ndash39 2009

[58] S Canova F Fiorasi M Mognato et al ldquoldquoModeled micrograv-ityrdquo affects cell response to ionizing radiation and increasesgenomic damagerdquo Radiation Research vol 163 no 2 pp 191ndash199 2005

[59] B P Lewis C B Burge and D P Bartel ldquoConserved seedpairing often flanked by adenosines indicates that thousandsof human genes are microRNA targetsrdquo Cell vol 120 no 1 pp15ndash20 2005

[60] M S Cline M Smoot E Cerami et al ldquoIntegration ofbiological networks and gene expression data using CytoscaperdquoNature Protocols vol 2 no 10 pp 2366ndash2382 2007

[61] F Censi A Giuliani P Bartolini and G Calcagnini ldquoA multi-scale graph theoretical approach to gene regulation networks acase study in atrial fibrillationrdquo IEEETransactions onBiomedicalEngineering vol 58 no 10 pp 2943ndash2946 2011

[62] P Chen C Price Z Li et al ldquomiR-9 is an essential onco-genic microRNA specifically overexpressed in mixed lineageleukemia-rearranged leukemiardquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 110 no28 pp 11511ndash11516 2013

[63] G Pignot G Cizeron-Clairac S Vacher et al ldquoMicroRNAexpression profile in a large series of bladder tumors identifi-cation of a 3-miRNA signature associated with aggressivenessof muscle-invasive bladder cancerrdquo International Journal ofCancer vol 132 no 11 pp 2479ndash2491 2013

[64] H M Namloslashs L A Meza-Zepeda T Baroslashy et al ldquoModulationof the osteosarcoma expression phenotype by microRNAsrdquoPLoS ONE vol 7 no 10 Article ID e48086 2012

[65] P S Eis W Tam L Sun et al ldquoAccumulation of miR-155and BIC RNA in human B cell lymphomasrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 102 no 10 pp 3627ndash3632 2005

[66] Y Pan M Meng G Zhang H Han and Q Zhou ldquoOncogenicmicroRNAs in the genesis of leukemia and lymphomardquoCurrentPharmaceutical Design 2014

[67] M V Iorio M Ferracin C-G Liu et al ldquoMicroRNA geneexpression deregulation in human breast cancerrdquo CancerResearch vol 65 no 16 pp 7065ndash7070 2005

[68] Z Lu Y Ye D Jiao J Qiao S Cui and Z Liu ldquoMiR-155 andmiR-31 are differentially expressed in breast cancer patientsand are correlated with the estrogen receptor and progesteronereceptor statusrdquo Oncology Letters vol 4 no 5 pp 1027ndash10322012

[69] F Gao J Chang H Wang and G Zhang ldquoPotential diagnosticvalue ofmiR-155 in serum from lung adenocarcinoma patientsrdquoOncology Reports vol 31 no 1 pp 351ndash357 2014

[70] G Higgs and F Slack ldquoThe multiple roles of microRNA-155 inoncogenesisrdquo Journal of Clinical Bioinformatics vol 3 no 1 p17 2013

[71] H Fayyad-Kazan N BitarM Najar et al ldquoCirculatingmiR-150andmiR-342 in plasma are novel potential biomarkers for acutemyeloid leukemiardquo Journal of TranslationalMedicine vol 11 no1 article 31 2013

[72] M Yanlei P Zhang F Wang et al ldquomiR-150 as a potentialbiomarker associated with prognosis and therapeutic outcomein colorectal cancerrdquo Gut vol 61 no 10 pp 1447ndash1453 2012

[73] N K Jacob J V Cooley T N Yee et al ldquoIdentification of Sensi-tive SerummicroRNABiomarkers for Radiation BiodosimetryrdquoPLoS ONE vol 8 no 2 Article ID e57603 2013

[74] G J Zhang H Zhou H X Xiao Y Li and T Zhou ldquoMiR-378 isan independent prognostic factor and inhibits cell growth andinvasion in colorectal cancerrdquo BMC Cancer vol 14 no 1 p 1092014

[75] M Sand M Skrygan D Georgas et al ldquoMicroarray analysis ofmicroRNA expression in cutaneous squamous cell carcinomardquoJournal of Dermatological Science vol 68 no 3 pp 119ndash1262012

[76] S Hauser L M Wulfken S Holdenrieder et al ldquoAnalysisof serum microRNAs (miR-26a-2lowast miR-191 miR-337-3p andmiR-378) as potential biomarkers in renal cell carcinomardquoCancer Epidemiology vol 36 no 4 pp 391ndash394 2012

[77] L S Mangala Y Zhang Z He et al ldquoEffects of simulatedmicrogravity on expression profile of microRNA in humanlymphoblastoid cellsrdquo The Journal of Biological Chemistry vol286 no 37 pp 32483ndash32490 2011

[78] P Alexiou MMaragkakis G L Papadopoulos M Reczko andA G Hatzigeorgiou ldquoLost in translation an assessment andperspective for computational microrna target identificationrdquoBioinformatics vol 25 no 23 pp 3049ndash3055 2009

[79] J Nunez-Iglesias C-C Liu T E Morgan C E Finch and XJ Zhou ldquoJoint genome-wide profiling of miRNA and mRNAexpression in Alzheimers disease cortex reveals alteredmiRNAregulationrdquo PLoS ONE vol 5 no 2 Article ID e8898 2010

[80] LMa YHuangWZhu et al ldquoAn integrated analysis ofmiRNAand mRNA expressions in non-small cell lung cancersrdquo PLoSONE vol 6 no 10 Article ID e26502 2011

[81] J C Engelmann and R Spang ldquoA least angle regression modelfor the prediction of canonical and non-canonical miRNA-mRNA interactionsrdquo PLoS ONE vol 7 no 7 Article ID e406342012

[82] N Bossel Ben-Moshe R Avraham M Kedmi et al ldquoContext-specific microRNA analysis identification of functionalmicroRNAs and their mRNA targetsrdquo Nucleic Acids Researchvol 40 no 21 pp 10614ndash10627 2012


[83] S Artmann K Jung A Bleckmann and T Beiszligbarth ldquoDetec-tion of simultaneous group effects inmicroRNA expression andrelated target gene setsrdquo PLoS ONE vol 7 no 6 Article IDe38365 2012

[84] R N Germain ldquoMHC-dependent antigen processing andpeptide presentation providing ligands for T lymphocyte acti-vationrdquo Cell vol 76 no 2 pp 287ndash299 1994

[85] I V Konstantinova E N Antropova V I Legenkov and VD Zazhirey ldquoStudy of the reactivity of blood lymphoid cells increw members of Soyuz 6 7 and 8 before and after space flightrdquoKosmicheskaia Biologiia iMeditsina vol 7 no 6 pp 35ndash40 1973(Russian)

[86] A Cogoli and A Tschopp ldquoLymphocyte reactivity duringspaceflightrdquo Immunology Today vol 6 no 1 pp 1ndash4 1985

[87] N Gueguinou C Huin-Schohn M Bascove et al ldquoCouldspaceflight-associated immune system weakening preclude theexpansion of human presence beyond Earths orbitrdquo Journal ofLeukocyte Biology vol 86 no 5 pp 1027ndash1038 2009

[88] Y O Nunez J M Truitt G Gorini et al ldquoPositively correlatedmiRNA-mRNA regulatory networks in mouse frontal cortexduring early stages of alcohol dependencerdquo BMCGenomics vol14 p 725 2013

[89] R P Sullivan L A Fogel J W Leong et al ldquoMicroRNA-155 tunes both the threshold and extent of NK cell activationvia targeting of multiple signaling pathwaysrdquo The Journal ofImmunology vol 191 no 12 pp 5904ndash5913 2013

[90] F Gao Z-L Zhao W-T Zhao et al ldquoMiR-9 modulates theexpression of interferon-regulated genes and MHC class Imolecules in humannasopharyngeal carcinoma cellsrdquoBiochem-ical and Biophysical Research Communications vol 431 no 3pp 610ndash616 2013

[91] F Bazzoni M Rossato M Fabbri et al ldquoInduction andregulatory function of miR-9 in human monocytes and neu-trophils exposed to proinflammatory signalsrdquo Proceedings of theNational Academy of Sciences of theUnited States of America vol106 no 13 pp 5282ndash5287 2009

[92] S Thiele J Wittmann H-M Jack and A Pahl ldquomiR-9enhances IL-2 production in activated human CD4+ T cells byrepressing Blimp-1rdquo European Journal of Immunology vol 42no 8 pp 2100ndash2108 2012

[93] M Singh F Del carpio-Cano J Y Belcher et al ldquoFunctionalroles of osteoactivin in normal and disease processesrdquo CriticalReviews in Eukaryotic Gene Expression vol 20 no 4 pp 341ndash357 2010

[94] M Cogoli-Greuter M A Meloni L Sciola et al ldquoMovementsand interactions of leukocytes in microgravityrdquo Journal ofBiotechnology vol 47 no 2-3 pp 279ndash287 1996


[96] X Ma J Pietsch M Wehland et al ldquoDifferential gene expres-sion profile and altered cytokine secretion of thyroid cancer cellsin spacerdquoThe FASEB Journal vol 28 no 2 pp 813ndash835 2014

[97] D Chang H Xu Y Guo et al ldquoSimulated microgravity altersthe metastatic potential of a human lung adenocarcinoma celllinerdquo In Vitro Cellular and Developmental BiologymdashAnimal vol49 no 3 pp 170ndash177 2013

[98] S Hong X Li Y ZhaoQ Yang and B Kong ldquo53BP1 suppressestumor growth and promotes susceptibility to apoptosis ofovarian cancer cells through modulation of the Akt pathwayrdquoOncology Reports vol 27 no 4 pp 1251ndash1257 2012




Guest Editors Mariano Bizzarri Monica Moniciand Jack J W A van Loon

Copyright copy 2015 Hindawi Publishing Corporation All rights reserved

This is a special issue published in ldquoBioMed Research Internationalrdquo All articles are open access articles distributed under the CreativeCommons Attribution License which permits unrestricted use distribution and reproduction in any medium provided the originalwork is properly cited

Contents

HowMicrogravity Affects the Biology of Living Systems Mariano Bizzarri Monica Moniciand Jack J W A van LoonVolume 2015 Article ID 863075 4 pages

Simulated Microgravity Critical Review on the Use of Random Positioning Machines for MammalianCell Culture Simon L Wuest Stephane Richard Sascha Kopp Daniela Grimm and Marcel EgliVolume 2015 Article ID 971474 8 pages

Regulation of ICAM-1 in Cells of the MonocyteMacrophage System in Microgravity Katrin PaulsenSvantje Tauber Claudia Dumrese Gesine Bradacs Dana M Simmet Nadine Golz Swantje HauschildChristiane Raig Stephanie Engeli Annett Gutewort Eva Hurlimann Josefine Biskup Felix UnverdorbenGabriela Rieder Daniel Hofmanner Lisa Mutschler Sonja Krammer Isabell Buttron Claudia PhilpotAndreas Huge Hartwin Lier Ines Barz Frank Engelmann Liliana E Layer Cora S Thiel and Oliver UllrichVolume 2015 Article ID 538786 18 pages

Genes Required for Survival in Microgravity Revealed by Genome-Wide Yeast Deletion CollectionsCultured during Spaceflight Corey Nislow Anna Y Lee Patricia L Allen Guri Giaever Andrew SmithMarinella Gebbia Louis S Stodieck Jeffrey S Hammond Holly H Birdsall and Timothy G HammondVolume 2015 Article ID 976458 10 pages

RhoGTPases as Key Players in Mammalian Cell Adaptation to Microgravity Fiona LouisChristophe Deroanne Betty Nusgens Laurence Vico and Alain GuignandonVolume 2015 Article ID 747693 17 pages

A Tissue Retrieval and Postharvest Processing Regimen for Rodent Reproductive Tissues Compatiblewith Long-Term Storage on the International Space Station and Postflight Biospecimen SharingProgram Vijayalaxmi Gupta Lesya Holets-Bondar Katherine F Roby George Enders and Joseph S TashVolume 2015 Article ID 475935 12 pages

Large Artery Remodeling and Dynamics following Simulated Microgravity by Prolonged Head-DownTilt Bed Rest in Humans Carlo Palombo Carmela Morizzo Martino Baluci Daniela Lucini Stefano RicciGianni Biolo Piero Tortoli and Michaela KozakovaVolume 2015 Article ID 342565 7 pages

Space Flight Effects on Antioxidant Molecules in Dry TardigradesThe TARDIKISS ExperimentAngela Maria Rizzo Tiziana Altiero Paola Antonia Corsetto Gigliola Montorfano Roberto Guidettiand Lorena RebecchiVolume 2015 Article ID 167642 7 pages

Identification of Reference Genes in HumanMyelomonocytic Cells for Gene Expression Studies inAltered Gravity Cora S Thiel Swantje Hauschild Svantje Tauber Katrin Paulsen Christiane RaigArnold Raem Josefine Biskup Annett Gutewort Eva Hurlimann Felix Unverdorben Isabell ButtronBeatrice Lauber Claudia Philpot Hartwin Lier Frank Engelmann Liliana E Layer and Oliver UllrichVolume 2015 Article ID 363575 20 pages

AWhole-GenomeMicroarray Study of Arabidopsis thaliana Semisolid Callus Cultures Exposed toMicrogravity and Nonmicrogravity Related Spaceflight Conditions for 5 Days on Board of Shenzhou 8Svenja Fengler Ina Spirer Maren Neef Margret Ecke Kay Nieselt and Rudiger HamppVolume 2015 Article ID 547495 15 pages

RCCS Bioreactor-Based Modelled Microgravity Induces Significant Changes on In Vitro 3D NeuroglialCell Cultures Caterina Morabito Nathalie Steimberg Giovanna Mazzoleni Simone GuarnieriGiorgio Fano-Illic and Maria A MariggioVolume 2015 Article ID 754283 14 pages

The Impact of Microgravity and Hypergravity on Endothelial Cells Jeanette A M MaierFrancesca Cialdai Monica Monici and Lucia MorbidelliVolume 2015 Article ID 434803 13 pages

A Functional Interplay between 5-Lipoxygenase and 120583-Calpain Affects Survival and Cytokine Profile ofHuman Jurkat T Lymphocyte Exposed to Simulated Microgravity Valeria Gasperi Cinzia RapinoNatalia Battista Monica Bari Nicolina Mastrangelo Silvia Angeletti Enrico Dainese and MauroMaccarroneVolume 2014 Article ID 782390 10 pages

HowMicrogravity Changes Galectin-3 inThyroid Follicles Elisabetta Albi Francesco CurcioAndrea Lazzarini Alessandro Floridi Samuela Cataldi Remo Lazzarini Elisabetta Loreti Ivana Ferriand Francesco Saverio Ambesi-ImpiombatoVolume 2014 Article ID 652863 5 pages

The Influence of Simulated Microgravity on Purinergic Signaling Is Different between IndividualCulture and Endothelial and Smooth Muscle Cell Coculture Yu Zhang Patrick Lau Andreas PanskyMatthias Kassack Ruth Hemmersbach and Edda TobiaschVolume 2014 Article ID 413708 11 pages

Human Locomotion under Reduced Gravity Conditions Biomechanical and NeurophysiologicalConsiderations Francesca Sylos-Labini Francesco Lacquaniti and Yuri P IvanenkoVolume 2014 Article ID 547242 12 pages

Conditioned Media fromMicrovascular Endothelial Cells Cultured in Simulated Microgravity InhibitOsteoblast Activity Alessandra Cazzaniga Sara Castiglioni and Jeanette A M MaierVolume 2014 Article ID 857934 9 pages

Phenotypic Switch Induced by Simulated Microgravity on MDA-MB-231 Breast Cancer CellsMaria Grazia Masiello Alessandra Cucina Sara Proietti Alessandro Palombo Pierpaolo ColucciaFabrizio DrsquoAnselmi Simona Dinicola Alessia Pasqualato Veronica Morini and Mariano BizzarriVolume 2014 Article ID 652434 12 pages

Oxidative Stress and NO Signalling in the Root Apex as an Early Response to Changes in GravityConditions Sergio Mugnai Camilla Pandolfi Elisa Masi Elisa Azzarello Emanuela MonettiDiego Comparini Boris Voigt Dieter Volkmann and Stefano MancusoVolume 2014 Article ID 834134 10 pages

Cytoskeleton Modifications and Autophagy Induction in TCam-2 Seminoma Cells Exposed toSimulated Microgravity Francesca Ferranti Maria Caruso Marcella Cammarota Maria Grazia MasielloKatia Corano Scheri Cinzia Fabrizi Lorenzo Fumagalli Chiara Schiraldi Alessandra CucinaAngela Catizone and Giulia RicciVolume 2014 Article ID 904396 14 pages

Gravity Affects the Closure of the Traps inDionaea muscipula Camilla Pandolfi Elisa Masi Boris VoigtSergio Mugnai Dieter Volkmann and Stefano MancusoVolume 2014 Article ID 964203 5 pages

The Impact of Simulated and Real Microgravity on Bone Cells and Mesenchymal Stem CellsClaudia Ulbrich Markus Wehland Jessica Pietsch Ganna Aleshcheva Petra Wise Jack van LoonNils Magnusson Manfred Infanger Jirka Grosse Christoph Eilles Alamelu Sundaresan and Daniela GrimmVolume 2014 Article ID 928507 15 pages

Multisensory Integration and Internal Models for Sensing Gravity Effects in PrimatesFrancesco Lacquaniti Gianfranco Bosco Silvio Gravano Iole Indovina Barbara La Scaleia Vincenzo Maffeiand Myrka ZagoVolume 2014 Article ID 615854 10 pages

Integration Analysis of MicroRNA and mRNA Expression Profiles in Human Peripheral BloodLymphocytes Cultured in Modeled Microgravity C Girardi C De Pitta S Casara E CaluraC Romualdi L Celotti and M MognatoVolume 2014 Article ID 296747 16 pages

EditorialHow Microgravity Affects the Biology of Living Systems

Mariano Bizzarri1 Monica Monici2 and Jack J W A van Loon3

1Department of Experimental Medicine Systems Biology Group University La Sapienza 00161 Rome Italy2ASAcampus Joint Laboratory ASA Research Division Department of Experimental and Clinical Biomedical Sciences University ofFlorence 50121 Florence Italy3Department of Oral and Maxillofacial SurgeryOral Pathology VU-University Medical Center 1081 HZ Amsterdam Netherlands


Received 20 November 2014 Accepted 20 November 2014

Copyright copy 2015 Mariano Bizzarri et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Gravity has constantly influenced both physical and biologi-cal phenomena throughout Earthrsquos history The gravitationalfield has played a major role in shaping evolution whenlife moved from water to land even if for a while it hasbeen generally deemed to influence natural selection onlyby limiting the range of acceptable body sizes accordingto Galileirsquos principle Indeed to counteract gravity livingorganisms would need to develop systems to provide cellmembrane rigidity fluid flow regulation and appropriatestructural support for locomotion However gravity mayinfluence in a more deep and subtle fashion the way the cellsbehave and build themselves

The first empirical experiments mostly done by Russianscientists in the 60s were unable to unravel major changesafter exposure tomicrogravity thus nurturing the false notionfor which near weightlessness does not get any appreciableeffects on living organisms [1 2] However as fundamentalinvestigations began in the space environment it became evi-dent that biological properties change as gravitational force isdiminished underscoring the relationship between physicalforce and biological function Cells exposed to microgravitycan indeed be profoundly affected by the physical changesthat occur in this unique environment which include theloss of gravity-dependent convection negligible hydrody-namic shear and lack of sedimentation [3ndash5] Cell-substrateadhesions as well as cell-to-cell junctions are consequentlyprofoundly affected at Earthrsquos gravity impairing multicellularaggregates and tissue formation while such structures can bemore easily sustained for days or months in microgravity [6]These modifications eventually lead to a significant change

in the way the cell mechanosensor apparatus responds to awide array of environmental and internal biophysical stresses[7] As a consequence enzymatic genetic and epigeneticpathways change in concert leading to several modificationsin cells and tissues shape function and behavior [8 9]Fruitful insights about the involvement of several molecularpathways during microgravity exposure are reported in thisissue by the studies of V Gasperi et al (unravelling newpathways involved in immune function impairment duringspaceflight) and E Albi et al (overexpression of Galectin-3in thyroid follicles due to microgravity-induced membraneremodelling) Namely a sophisticated analysis of mRNAexpression in human blood lymphocytes carried out byC Girardi et al confirmed that microgravity induces ageneralized inhibition of proliferation and a contemporaryincrease in apoptosis rate

Indeedmdashand unfortunatelymdashnear weightlessness dra-matically impairs biological functions and thereby contrarytowhatwas previously thought [2] cells cannot be consideredldquoblindrdquo with respect to gravity

The microgravity space environment may result in achallenging threat for living beings as aptly documented bythe paper from C Nislow et al showing that spaceflighthas subtle but significant effects on core cellular processesincluding growth control via RNA and ribosomal biogen-esis metabolism modification and decay pathways It isnoteworthy that despite the fact that some reference-genesremain stable during microgravity exposure several othersinvestigated in the study of C S Thiel et al change quitedramatically thus reinforcing the concept that exposure to



near weightlessness may have a profound impact on livingprocesses Namely it seems that genes involved in ROSdetoxification are especially impaired in such conditionas reported by the paper from S Fengler et al thereforesuggesting how relevant could be the role sustained by theredox status in counteracting at least some downstreamconsequences of microgravity Yet as reported in the articleof S Mugnai et al both nitric oxide and ROS are likely toplay a previously unrecognized role as messengers during thegravitropic response inmany root tips Relevance of oxidativeprocesses during microgravity exposure was also reported bythe study of A M Rizzo et al in which a significant increasein oxidative stress has been observed in tardigrades exposedto spaceflight

Cells may ldquosenserdquo changes in the microgravitational fieldthrough (a) an indirect mechanism (mainly based on themodification of physical properties of their microenviron-ment) (b) the development of specialized structures forthe mechanical perception and transduction of gravitationalforces (like the cytoskeleton) and (c) changes in the dynamicsof enzymes kinetics or protein network self-assembly It isworth noting that the latter two processes are dramaticallyaffected by nonequilibrium dynamics Nonlinear dynamicalprocesses far from equilibrium involve an appropriate com-bination of reaction and diffusion and the pattern arisingfrom those interactions is tightly influenced by evenminimalchanges in reactant concentrations or modification in thestrength of the morphogenetic field [10] Processes of thiskind are called Turing or dissipative structures given that aconsumption of energy is required to drive and maintain thesystem far from equilibrium That prerequisite is needed inorder to allow the system to promptly change its configura-tion according to the systemrsquos needs In turn the dissipativeenergy provides the thermodynamic driving force for theself-organization processes Some experimental evidence hasalready been provided that change of the gravitational fieldmay significantly affect some nonlinear reactions occurringwithin cells and tissues [11 12] Herein a further confirmationis provided by the article of M G Masiello et al in whichthe near weightlessness condition is shown to drive thesystems towards different attractor states thus enabling cellsto acquire new and unexpected phenotypes in the courseof a true phase transition [13] According to such resultsgravity seems to be an ldquoinescapablerdquo constraint that obligesliving beings to adopt only a few configurations amongmanyothers By ldquoremovingrdquo the gravitational field living structureswill be free to recover more degrees of freedom thusacquiring new phenotypes and new functionspropertiesThat statement raises several crucial questions Some of theseentail fundamentals of theoretical biology as they questionthe gene-centered paradigm according to which biologicalbehavior can be explained by solely genetic mechanisms [14]

What are the mechanism(s) through which microgravitymay so profoundly modify cell function and structureSeveral studies included in this issue deal with that topiccalling into the question the pivotal role sustained by thecytoskeleton in mediating several microgravity-based effects

A common outcome in nearly all cell types exposed tomicrogravity is indeed the alteration of cytoskeletal elements

actin microfilaments and microtubules [15 16] Disorga-nization of basic cellular architecture can affect activitiesranging from cell signalling and migration to cell cycling andapoptosis In this issue K Paulsen and colleagues investigatedhow surface expression of ICAM-1 protein and expressionof ICAM-1 mRNA in cells of the monocytemacrophagesystem change inmicrogravity Given that ICAMproteins areessential for cell-to-cell adhesion as well as for cytoskeletonproper functioning such results outline the involvement ofthe cytoskeleton system in mediating at least some effectsdue to microgravity That statement is further reinforced bythe paper from F Louis et al in which dramatic decreasein RhoGTPases activity has been documented RhoGTPasesrepresent a unique hub for integration of biochemical andmechanical signals As such they are probably very rapidlyinvolved in a cellrsquos adaptation to microgravity-related con-ditions Additionally RhoGTPases activity is tightly andmechanistically bound to alterations of the cytoskeletonadhesion and fibrillogenesis as well as to an enhancementof ROS delivery As a result RhoGTPases may be consideredtrue mechanosensitive switches responsible for cytoskeletaldynamics and cells commitment Relevant modification ofthe cytoskeleton architecture and microtubule organizationin testicular cells has been also reported in the study byF Ferranti et al where a significant correlation betweencytoskeleton abnormalities induced by simulated micrograv-ity and enhanced autophagy was recorded Yet cytoskeletonchanges affect different cell types including endothelial cellsIn the paper of J Maier et al it is shown that endothelial cellsare highly sensitive to gravitational stress as microgravityleads to changes in the production and expression of vasoac-tive and inflammatory mediators and adhesion moleculeswhich mainly results from changes in the remodelling of thecytoskeleton and the distribution of caveolae In addition bykeeping in mind that the cytoskeleton dynamics is a funda-mental player in cell proliferation and migration it is notsurprising that microgravity significantly affects the flytrapclosure a process involving not only the actin dynamics butalso the ion channels and aquaporin activities as evidencedin the article from C Pandolfi et al

Cytoskeleton changes have also profound consequenceson both cell shape and tissue modelling Simulated nearweightlessness in human volunteers is associated with asignificant change in arterial geometry flow stiffness andshear rate as documented by C Palombo et al Microgravityis acting on endothelial cells also through modulation of P2-receptor and the release of several cytokines as reportedby the study from Y Zhang et al Given that P2-receptorartificial ligands are applied as drugs it is reasonable toassume that they might be promising candidates againstthe cardiovascular deconditioning the astronauts experienceduring spaceflight

Overall the alterations occurring in microgravity haveundoubtedly significant backwashes on the physiologicalhomeostasis of the whole organism Such aspect is high-lighted by two papers from the group of F Lacquaniti et aldealingwith the effects of nearweightlessness on nervous sys-tem function Gravity is indeed crucial for spatial perceptionpostural equilibrium and movement generation The brain


may deal with the gravitational field by integrating a widearray of different signals thus enabling the system to triggerthe most appropriate response F Lacquaniti et al providecompelling evidence that this ability depends on the fact thatgravity effects are stored in brain regions which integratevisual vestibular and neck proprioceptive signals where thenervous system combines this information with an internalmodel of gravity effects The second study evidenced inturn the beneficial effect of the neurophysiologic adaptationto near weightlessness and how knowledge acquired onthis field may even enhance the development of innovativetechnologies for gait rehabilitation

Research on microgravity and hypergravity effectivelyadvances our knowledge on physiology and biochemistrythus providing valuable data and models for the understand-ing for some important human diseases Moreover space-based research has played and presumably will continuouslyplay an important role in reformulating the theoreticalframework in biology and physiology and may serve as anovel paradigm for innovation Namely microgravity-relatedresearch fostered the development of new tools-like forculturing cells in three dimensions It is now well understoodthat 3D growth environments that facilitate unrestricted cell-cell interactions are mandatory for defining the biology ofcancer cells and tissues including tumour formation tumourmicroenvironment and tumour progression [17 18] Indeedthree-dimensional culture in real and simulatedmicrogravityallows a more precise appreciation of the role the biophysicalconstraints play in shaping cell phenotypes and functions Inturn such devices may help in improving tissue-engineeringtechniques Experimental models of cellstissues culturesin both simulated and real microgravity need howeverto be further improved in order to obtain more reliableand reproducible data and to minimize the impact of con-founding factors Such studies may indeed provide valu-able information about modulations in signal transductioncell adhesion or extracellular matrix induced by alteredgravity conditions These systems also facilitate the analysisof the impact of growth factors hormones or drugs onthese tissue-like constructs in order to better address issueslike pharmacokinetics and pharmacodynamics Paradigmaticexamples of such studies are reported in this issue by thearticles of several groups (C Ulbrich et al C Morabitoet al V Gupta et al) some of which (S L Wuest et al)critically reviewed the reliability of available technical tools(like the Random Positioning Machine) These facilities mayalso allow investigating developmental and organogenesisprocesses

The motivation for this focussed issue of the BiomedResearch International Journal is to take stock of the state ofresearch and identify possible areas for future developmentThere is an urgent need for this as the last comprehensivecollection of studies devoted to space biomedicine researchdates back to the 90s [19]

As editors we have collected an eclectic mix of arti-cles provided by research groups fully involved in spacebiomedicine research and actively participating in studiescarried out both on the International Space Station andon the ground by means of different techniques enabling

performing conditions of simulated near weightlessness andincreased gravityThis is not a ldquoone view fits allrdquo approach It israther one to ldquolet a hundred flowers bloomrdquo Yet they providea fruitful overview on what is going to come from spacebiomedicine research Overall studies reported in the issuedemonstrated how relevant physical cues may be in shapingbiological phenotypes and function influencing so in depthmolecular and genetic pathways It is regrettable to noticethat such influences have been for so long overlooked bythe scientificmainstream [20 21] Furthermore microgravitystudies forced us to develop new technological solutions andmore appropriate experimental models Thereby knowledgegathered in space research has offered an invaluable supportin understanding both human physiology and pathologyfostering technological innovation and the development ofpriceless medical and experimental devices

This is why it has been argued that the ultimate reason forhuman space exploration is precisely to enable us to discoverourselves Undoubtedly the microgravity and space relatedresearch present an unlimited horizon for investigation anddiscovery Controlled studies conducted in microgravity canadvance our knowledge providing amazing and unforeseeninsights into the biological mechanism underlying physiol-ogy as well as many relevant diseases like cancer [22]

Mariano BizzarriMonica Monici

Jack J W A van Loon

References

[1] P O Montgomery Jr J E Cook R C Reynolds et al ldquoTheresponse of single human cells to zero gravityrdquo In Vitro vol 14no 2 pp 165ndash173 1978

[2] M G Tairbekov G P Parfyonov E Y Shepelev and F VSushkov ldquoExperimental and theoretical analysis of the influenceof gravity at the cellular level a reviewrdquo Advances in SpaceResearch vol 3 no 9 pp 153ndash158 1983

[3] J J W van Loon ldquoThe gravity environment in Space exper-imentsrdquo in Biology in Space and Life on Earth Effects ofSpaceflight on Biological Systems E Brinckmann Ed pp 17ndash32Wiley-VCH 2007

[4] T G Hammond and J M Hammond ldquoOptimized suspen-sion culture the rotating-wall vesselrdquo American Journal ofPhysiologymdashRenal Physiology vol 281 no 1 pp F12ndashF25 2001

[5] P Todd ldquoGravity-dependent phenomena at the scale of thesingle cellrdquo ASGSB Bulletin vol 2 pp 95ndash113 1989


[7] J Klein-Nulend R G Bacabac J P Veldhuijzen and J J WA Van Loon ldquoMicrogravity and bone cell mechanosensitivityrdquoAdvances in Space Research vol 32 no 8 pp 1551ndash1559 2003

[8] S J PardoM J PatelMC Sykes et al ldquoSimulatedmicrogravityusing the Random Positioning Machine inhibits differentiationand alters gene expression profiles of 2T3 preosteoblastsrdquoAmerican Journal of PhysiologymdashCell Physiology vol 288 no6 pp C1211ndashC1221 2005


[9] M Monici F Fusi M Paglierani et al ldquoModeled gravitationalunloading triggers differentiation and apoptosis in preosteo-clastic cellsrdquo Journal of Cellular Biochemistry vol 98 no 1 pp65ndash80 2006

[10] G Nicolis and I Prigogine ldquoIntroductionrdquo in Self-Organizationin Nonequilibrium Systems FromDissipative Structures to OrderThrough Fluctuations John Wiley amp Sons New York NY USA1977

[11] P J Stiles and D F Fletcher ldquoThe effect of gravity on the rateof a simple liquid-state reaction in a small unstirred cylindricalvesselrdquo Physical Chemistry Chemical Physics vol 3 no 9 pp1617ndash1621 2001


[13] M Bizzarri and A Giuliani ldquoRepreseenting cancer cell trajec-tories in a phase-space diagram switching cellular states bybiological phase transitionsrdquo in Applied Statistics for NetworkBiology Methods in Systems Biology M Dehmer F Emmert-Streib A Graber and A Salvador Eds pp 377ndash403 WileyNew York NY USA 2011

[14] M Bizzarri A Cucina A Palombo and M Grazia MasielloldquoGravity sensing by cellsMechanisms and theoretical groundsrdquoRendiconti Lincei vol 25 no 1 pp 29ndash38 2014

[15] M L Lewis ldquoThe cytoskeleton in spaceflown cells an overviewrdquoGravitational and Space Biology Bulletin vol 17 pp 1ndash11 2004

[16] D Vorselen W H Roos F C MacKintosh G J L Wuite andJ J W A van Loon ldquoThe role of the cytoskeleton in sensingchanges in gravity by nonspecialized cellsrdquo FASEB Journal vol28 no 2 pp 536ndash547 2014

[17] D Grimm M Wehland J Pietsch et al ldquoGrowing tissuesin real and simulated microgravity new methods for tissueengineeringrdquo Tissue Engineering Part B Reviews vol 20 no 6pp 555ndash566 2014


[19] D Schmitt ldquoWorkshop purpose and structurerdquo The FASEBJournal vol 13 supplement S1 no 9001 1999


[21] M Monici and J van Loon Cell Mechanochemistry Bio-logical Systems and Factors Inducing Mechanical Stress Suchas Light Pressure and Gravity Trivandrum Research Sign-postTransword Research Network 2010

[22] J L Becker and G R Souza ldquoUsing space-based investigationsto informcancer research onEarthrdquoNature ReviewsCancer vol13 no 5 pp 315ndash327 2013

Review ArticleSimulated Microgravity Critical Review on the Use ofRandom Positioning Machines for Mammalian Cell Culture

Simon L Wuest1 Steacutephane Richard1 Sascha Kopp2 Daniela Grimm2 and Marcel Egli1

1 Lucerne University of Applied Sciences and Arts School of Engineering and ArchitectureCC Aerospace Biomedical Science and Technology Space Biology Group Lucerne University of Applied Sciences and ArtsSeestraszlige 41 6052 Hergiswil Switzerland

2 Institute of Biomedicine Pharmacology Aarhus University Wilhelm Meyers Alle 4 8000 Aarhus C Denmark

Correspondence should be addressed to Marcel Egli marceleglihsluch



Copyright copy 2015 Simon L Wuest et alThis is an open access article distributed under the Creative CommonsAttribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Random Positioning Machines (RPMs) have been used since many years as a ground-based model to simulate microgravity Inthis review we discuss several aspects of the RPM Recent technological development has expanded the operative range of theRPM substantially New possibilities of live cell imaging and partial gravity simulations for example are of particular interestFor obtaining valuable and reliable results from RPM experiments the appropriate use of the RPM is of utmost importanceThe simulation of microgravity requires that the RPMrsquos rotation is faster than the biological process under study but not sofast that undesired side effects appear It remains a legitimate question however whether the RPM can accurately and reliablysimulate microgravity conditions comparable to real microgravity in space We attempt to answer this question by mathematicallyanalyzing the forces working on the samples while they are mounted on the operating RPM and by comparing data obtained underreal microgravity in space and simulated microgravity on the RPM In conclusion and after taking the mentioned constraintsinto consideration we are convinced that simulated microgravity experiments on the RPM are a valid alternative for conductingexaminations on the influence of the force of gravity in a fast and straightforward approach

1 Introduction

Gravity is an omnipresent force on Earth and all livingorganisms have evolved under the influence of constantgravity Some organisms have learned to take advantage ofthe force of gravity by using it as a reference for orientationThe condition ofmicrogravity (or nearweightlessness) and itseffects on living organisms on the other hand have alwayspresented a fascinating scenario in biology and medicineWith the first manned space flights it became clear thatthe human organism reacts with a series of adaptations tomicrogravity Interestingly some of the symptoms observedin space such as wasting muscle mass and decreasing bonedensity are typically diagnosed in the elderly as well [1ndash3]This is one important factor that fostered scientistsrsquo interestin doing space research

Numerous studies on mammalian organisms for exam-ple have demonstrated that the absence of gravity has severe

effects not only on a systemic level but also on a cellular levelShort-term effects of microgravity (on the order of seconds)can be studied on research platforms such as drop towers orairplanes that fly in parabolic maneuvers In contrast long-term effects can only be studied on board sounding rockets(on the order of minutes) and space vehicles in flight Due tothe extensive preparation effort safety constraints and rareflight opportunities however access to space experimentsis limited For many years the random positioning machine(RPM) besides other tools has been successfully usedto simulate microgravity for screening studies pre- andpostflight experiments and hardware testing The principleof the RPM (a specialized two-axis form of the clinostat)is based on gravity vector averaging to zero [4] The typicalRPM system comprises two gimbal-mounted frames whichare each driven by independent motors Through dedicatedalgorithms the samples placed on the inner frame are con-stantly reoriented such that the gravity vector is distributed



in all directions over time Thus from the samplersquos point ofview the constantly reorienting gravity vectorrsquos trajectoryaveraged over time shall converge toward zero However 1 g isalways acting on the sample at any given instant It is assumedthat the gravity vector needs to point in a specific directionfor a minimal period of time in order to allow biologicalsystems like cells to adapt to the gravity vector But if thegravity vector constantly changes its orientation the cellswill lose the sense of direction and thus experience a statesimilar to microgravity (removed gravity vector) Thereforethe rotation of the frames shall be faster than the biologicalprocess studied [5] However the rotation cannot be too fastas centrifugal forces will become effective [6] Therefore theRPM is typically used to examine slow processes which areobserved at least on the timescale of hours It remains a legiti-mate question whether the RPM can reliably simulate micro-gravity In this review we attempt to provide an answer to thatquestion by comparing data of mammalian cells obtainedat real microgravity in space and at simulated microgravitygenerated by using the RPM In the first part however asummary is provided on the latest technical development aswell as new applications of the RPM

2 RPM Development and Technology

21 RPM Systems Todayrsquos RPMs were introduced byJapanese plant researchers for conducting their particularstudies [7 8] Later on a similar machine was developed inThe Netherlands (Dutch Space) [5] Although both systemswere commercialized [6] their range of use for doing space-related experiments was limited For instance scientificstudies with mammalian cells that are very sensitive totemperature fluctuations were difficult to carry out becauseof a missing temperature control unit Thus these kinds ofexperiments had to be operated in a temperature-controlledroom (eg a growth chamber) One approach to overcomethis limitation was to miniaturize the RPM to fit into anordinary cell culture incubator (max size 50 times 50 times 50 cm)that offers precisely controlled temperatures (also referred toas desktop RPM) [4] Through this RPM modification theinstallation of large climate chambers around RPMs becameunnecessary We have recently reported another approach toupgrading the RPM by installing a commercial CO

2incu-

bator onto the rotating frames This RPM called ldquorandompositioning incubatorrdquo (RPI) [9 10] has the advantage ofbeing independent of large laboratory incubators (Figure 1)Furthermore the closed chamber of the incubator isolates theenvironment of the culture flasks and thus prevents exposureof biological samples to vapor and wear from the machineryfor example [10] Besides differences in the design of thethree RPM types (regular RPM small desktop RPM andRPI) there appear to be slightly different concepts of how toaverage the gravity vectorThe algorithm implemented on theJapanese RPM (referred to as a regular RPM) lets the RPMrun with random rotational speeds and changes the velocityafter two possible predefined periods (eg 30 or 60 s) [8]TheDutch systems (referred to as regular and desktop RPMs)rotate with random speeds that are varied at random timepoints [6] In contrast our RPI rotates with constant velocity

Figure 1 Randompositioning incubator (RPI) featuring a fully inte-grated CO

2incubator (developed by the Institute for Automation

University of Applied Science Northwestern Switzerland)

but the rotation direction is inverted at random time pointsThe transition from forward to backward takes place at apredefined rotational acceleration [10] All three algorithmsemployed by the different RPM types are reported to bereliable in averaging gravity To our current knowledge thesealgorithms are equivalent from a biological point of view

22 Live Cell Imaging on the RPM Microscopy is a commonanalytical tool used in cell biology Even though microscopesare used on clinostats (rotation around one horizontal axis)[11 12] until recently live cell imaging was not successful onan operating RPM To date most of the optical microscopytechniques applied under simulated microgravity conditionshave been realized in the field of physical sciences For suchexperiments microscopes with a low numerical apertureand poor imaging performances were used because of theirintrinsic robustness to environmental disturbances such asvibrations In life science however high magnification isneeded to detect modifications at the cellular or subcellu-lar level Because most of the ground-based microgravityresearch platforms are not vibration-free high-performancemicroscopy has not been applicable Thus studies involvingcell imaging have been conducted in ground laboratories afterchemical fixation of the cell in microgravity This approachimplies a series of static shots which cannot truly reveal thedynamic processes and labile cellular events occurring in cellsin response to microgravity exposure

Until recently there was no system available that allowedhigh-quality real-time images taken at cellular or subcellularlevel under (real or simulated) microgravity The break-through came with the use of a digital holographic micro-scope (DHM) that we have combined with an epifluorescentmicroscope In this dual-mode microscope the two imagingmodes (DHM and fluorescent) operate sequentially TheDHM is an innovative interferometric microscope that is lesssensitive to vibrations The technological advantages of theDHMwhich comprise continuous and fast digital autofocus-ingwith a short exposure time allowhigh-resolution imaging[13ndash16] We tested the DHM on the RPM as well as duringparabolic flights and in both cases we obtained good data


[13ndash16] For instance we followed reorganization of the actincytoskeleton and fluctuations of the intracellular calciumconcentration under simulated microgravity (unpublisheddata)

3 Partial Earth Gravity Load

During past years RPM development was focused on theimprovement of the hardware We have also been workingon an upgrade of the software responsible for controllingthe motion of the RPM Three different algorithms wereintroduced recently that simulated partial Earth gravity (0to 06 g) allowing simulation of moon- or Mars-like gravityconditions [9] All algorithms are adaptations of the randomwalk algorithm originally designed to simulate microgravity[10] As described above simulated microgravity is achievedby rotating both frames with constant velocity and invertingthe rotation direction at random times Partial gravity isachieved by altering the randomwalk in a way that the Earthrsquosgravity vector is not completely randomized anymore andpoints (from the samplersquos point of view) for a prolongedtime in a specific direction In one algorithm version thisis accomplished by slowing down the rotational velocitywhile the gravity vector (considered in the sample frame)is pointing downwards Otherwise the frames rotate withthe predefined velocity The ratio of the two velocities finallydetermines the mean gravity (gravity vector averaged overtime) In the other two algorithm versions the random walkis interleaved with static intervals in which the frames standstill in a predefined orientation However the timing of thesestatic intervals (start point and duration) is handled differ-ently In one case the timing is flexible and adjusted online asthe experiment runs In the other case the timing is strictlyperiodic and predefined before the experiment starts [9]

All three algorithms were tested on suspended humanT cells and adherent mice myoblasts Chemically activatedT cells showed a decreased activation rate that correlatesstrongly to the decreasing simulated mean gravity values[9] The results were similar for all algorithms The adheredmyoblast (C2C12 cell line) showed a decreased proliferationrate with decreasing mean gravity [9] Interestingly thiseffect is algorithm dependent The correlation between meangravity and proliferation was reduced or disappeared in thetwo algorithms involving static intervals [9] Ideally thesetypes of partial gravity experiments are carried out in spaceby using a centrifuge To our knowledge no comparablespace experiments have been conducted so far except duringparticular parabolic flight campaigns of the European SpaceAgency (ESA) Therefore a direct comparison to spaceis not possible at this time However these experimentsdemonstrate that simulation of partial gravity opens a newfield of scientific questions that attracts other research groupsDutch Space was attracted by the new topic as well andthus recently introduced a modified desktop RPM (pre-sented at the ELGRA meeting 2013) allowing partial gravitysimulations Partial Earth gravity enabling RPMs increasethe application range substantially allowing investigation ofthe influence of gravitymdashlike on the moon or Mars forexamplemdashon cells and small organisms at affordable cost

Figure 2 Mouse myoblasts (C2C12 cell line) were cultured untilnear confluence and subsequently exposed to a frequently passingair bubble The culture chamber filled with medium was swingingupside down such that the intentional air bubble frequently passedthe same trajectory The sample was fixed and stained for actin(green) and DNA (blue) thereafter The cells in the trajectory of theair bubble got detached from the substrate (dark central area) whilecells in the unaffected area kept proliferating (lateral green areas)Interestingly detached cells could reattach to the opposite side of theculture chamber Measuring bar 200120583m (Due to the limited field ofview this image has been stitched together from five images)

These results may help to estimate the biological response ofcells or even whole organisms when exposed to the gravityloads of other planets or moons

4 RPM Use and Experiment Quality

41 Cultivation Method of Mammalian Cells In order toobtain comparable data it is important to standardize cellculture methods One of the most important aspects of doingso is a stable cultivating environment When cultivatingcells on the moving RPM additional aspects have to beconsidered such as avoiding air bubbles in the culturechambers [4] Experiments have shown that an air bubblepassing by adherent cells at the same trajectory repetitively (asthe culture chamber moves in a ldquoswinging motionrdquo) the cellscan detach from the substrate (Figure 2 unpublished obser-vation) Interestingly these cells often reattach at the oppositeside of the culture chamber wall Using air- and gas-tight cul-ture chambers on the RPM has the advantage of being moreindependent of the culture environmentHowever a gas-tightculture chamber requires a culture medium that does notrequire CO

2for pH buffering which reduces the overall cul-

tivation period in which the culture flasks do not have to bemanipulated Gas-tight chambers in turn can cause problemswhen cultivating gas-producing cells such as yeast cells

42 Artifacts through Kinematic Rotation In addition toa standardized cultivation method artifacts caused by thekinematic rotation need to be considered While the Earthrsquosgravity vector is distributed in a way that the mean gravityconverges to zero over time the accelerations caused by theRPMrsquos kinematics are not well controlled In order to avoidartifacts the rotational velocity the samplersquos distance to thecenter of rotation and the rotational acceleration (duringvelocity transitions) have to be chosen appropriately Sincethere has been no systematic study on acceptable limitsscientists have to rely on their common sense The followingconsiderations can be used as guidelines For explanatoryreasons we also refer here to the somewhat simpler caseof clinorotation around one axis Clinorotation and therelated rotating wall vessel (RWV) bioreactor are alternativemethods commonly used in many laboratories to simulate


0 5 10 15 20 25 30

0

002

004

006

008

01

012

014

016

018

Peak centrifugal acceleration worst case


Cen

trifu

gal a

ccel

erat

ion

(g)

90

80

70

60

50

40

30

Rota

tiona

l vel

ocity

(deg

s)

Figure 3 The worst-case peak centrifugal acceleration on an RPMdepending on the distance to the center of rotation (119886pc asymp 241sdot120596

2 sdot119903)For example a moderate rotational velocity of 60 degs (cyan line)and a distance of 10 cm from the center of rotation (vertical dashedline) results in a peak centrifugal acceleration of approximately003 g (horizontal dashed line)

microgravity on the ground These methods simulate micro-gravity by rotating samples around a horizontal axis (Select-ing the appropriate rotation velocity for suspended cells inclinostat experiments has been discussed elsewhere [17])

To minimize centrifugal acceleration the rotationalvelocity and the samplersquos distance to the center of rotationshould be set as low as the experiment allows As mentionedearlier the rotation shall be clearly faster than the biologicalprocesses investigated [5] For mammalian cell experimentsmany scientists have used a rotational velocity of 60 degs[4] In the case of chemically activated T cells (as discussedfurther below) we could also create a microgravity-likeenvironment with a rotational velocity of 40 degs [10] Forrotation around one axis as in a clinostat or centrifuge thecentrifugal acceleration (in ms2) is time independent and iscomputed by 119886

119888= 1205962 sdot 119903 where 120596 is the rotation velocity (in

rads) and 119903 is the distance from the center of rotation (inmeters) For rotations around two perpendicular axes as isthe case for RPMs the centrifugal acceleration becomes timedependentThus the centrifugal acceleration depends now onthe two rotation velocities the samplersquos position in space andtime It is no longer trivial tomake a statement on the effectivecentrifugal acceleration at the samples within the cultivationchamber For the simplified case where both velocities areequal and constant the centrifugal acceleration becomesperiodically oscillating By focusing on a worst-case scenarioin terms of centrifugal acceleration the analysis provideseasy equations in such a scenario the peak centrifugalacceleration (inms2) can be approximated to 119886pc asymp 241sdot120596

2 sdot119903

(Figure 3) where 120596 is the rotation velocity of both frames(in rads) and 119903 is the distance from the center of rotation(in meters) As the equation indicates all cells are ideallyplaced at the center of rotation Therefore the scientist is

0 5 10 15 20 25 300

0005

001

0015

002

0025

003

0035

004

0045

Tangential acceleration (velocity transition) worst case


Tang

entia

l acc

eler

atio

n (g

)

40

30

20

10

5

Rota

tiona

l acc

eler

atio

n (d

egs2)

Figure 4 The worst-case tangential acceleration depending onthe distance from the center of rotation (119886

119905= 2 sdot 120572 sdot 119903) For a

smooth velocity transition of for example 10 degs2 (green line)and 10 cm distance from the center of rotation (vertical dashedline) a tangential acceleration of approximately 0004 g is expected(horizontal dashed line)

responsible for compactly placing the samples around thecenter of rotation By using the distance to the center ofrotation from the sample farthest away from this point (worstcase) the largest expected centrifugal acceleration can beestimated For a moderate velocity of typically 60 degs [4]and a moderate distance from the center of rotation (eg10 cm) the centrifugal acceleration is in the order of 10minus2 gSuch small forces are detectable by some specialized cells[18] Since at any instance in time the Earthrsquos gravity vector(which is averaged to zero over time) is present as well thecentrifugal acceleration is two orders of magnitude smallerand we therefore consider it to be negligible In additionthe transitions of the framesrsquo rotational velocities introduceadditional accelerations and thus should be smooth byselecting a small rotational acceleration For the clinostatthis tangential acceleration (in ms2) is 119886

119905= 120572 sdot 119903 where

120572 is the rotational acceleration (in rads2) For the RPMthe tangential acceleration becomes 119886

119905= 2 sdot 120572 sdot 119903 in

the worst case when both frames accelerate simultaneously(Figure 4) For a smooth velocity transition of 10 degs2 anda moderate distance from the center of rotation (eg 10 cm)the tangential acceleration is well below 10minus2 g

Besides these parasitic accelerations rotation introducesfluid motion in the culture flask leading to shear forcesand enhanced convection (Figure 5)This condition is unlikespace conditions where no convection is present Thereforethe nutrition supply on the RPM is enhanced as comparedto static or space experiments In order to avoid additionalmechanical stimulation such as shear stress a moderaterotational velocity needs to be chosen and the velocitytransitions have to be smooth [19] Because the behavior offluid motion has not been fully elucidated yet the acceptablelimits for rotation velocity and acceleration are not clarified


0 s

(a)

12 s

(b)

24 s

(c)

40 s

(d)

Figure 5 The RPM rotation introduces fluid motion in the culture flask leading to shear forces and enhanced convection Therefore amoderate rotational velocity needs to be chosen and the velocity transitions have to be smooth in order to minimize the introduction ofadditional mechanical stimulation of the samples In this numerical illustration the fluid motion is shown if both frames rotate at 60 degsThis results in a periodic motion of 6 seconds The four images indicate snapshots of the velocity at 0 s (a) 12 s (b) 22 s (c) and 4 s (d)

However the values provided above are a good starting pointand have been successfully used in previous experiments[9 10]

5 Experiment Reporting

As new and innovative technologies expand the range ofpossible experiments it is becoming important to documentthe used hardware precisely In accordance with good labo-ratory practice (GLP) any researcher who is using RPMs orclinostats should follow the ldquoBonnCriteriardquo In this documentit is stated that ldquoExperimental reporting should include theproperties of the culture vessel culture media and carrierbeads These should also include dimensions and rotationspeed of vessels chemical consistency including density andviscosity of media size density and porosity of beads sizedensity and porosity of cells whether cells are motile or

non-motile density of beads with cells attached as well astime of rotation nature of controls operating temperatureand gas content [20]rdquo As described above improper use ofthe RPM can introduce additional forces leading to unwantedmechanical stimulation of the sample cells Interpretingresults from such experiments could lead to wrong conclu-sions and could thus jeopardize a whole study

6 RPM Application in MammalianCell Culture

61 Can the RPM Reproduce Microgravity Conditions Des-pite the long history of RPM usage the difference betweensimulated and real microgravity in space shall be criticallyexamined when interpreting experimental results Particu-larly for adhered cells the rotation generated by the RPMcould provide an unwanted source of mechanical stimuli


[6] Unfortunately only a few researches have systematicallycompared experiments performed in a real microgravityenvironment and on an RPM Most of these comparativestudies have been done on leukocytes for which the RPMshowed good agreement with space experiments it is wellknown that T lymphocytes fail to activate in microgravityafter being exposed to the activator ConA [21]This effect wasreproduced numerous times on an RPM [9 10 22 23] Simi-larly Villa and colleagues have shown slower proliferation ofthe human leukemic myelomonocytic cell line U937 exposedto simulated microgravity on the RPM [24] The samephenomenon was previously observed on a space shuttleexperiment [25] In a study on cell mobility under micro-gravity with the human leukemic monocytemacrophage cellline the RPM predicted real microgravity results Monocytelocomotion ability was clearly reduced in real as well as insimulatedmicrogravityThe authors suggest that this is linkedto changes in the cytoskeletal structures since they observedreduced density of actin filaments and disruption of the 120573-tubulin architecture [26 27] Furthermore peripheral bloodmononuclear cells cultured for 48 hours onboard the Inter-national Space Station (ISS) showed remarkably increasedapoptotic hallmarks which could also be reproduced undersimulated microgravity [28]

In recent years two investigators directly compared theresults from RPM experiments to results obtained in spaceconditions performed simultaneously in the first experi-ment primary porcine chondrocytes from articular cartilagewere flown for 16 days aboard the ISS Cells exposed tomicrogravity showed higher collagen III ratio and reducedaggrecanversican ratio at the mRNA level In addition celldensity was significantly reduced and the extracellular mat-rix straining was weaker on the ISS samplesThe samples thatwere simultaneously exposed to simulated microgravity onan RPM generally showed results that were similar to thoseof the space samples but not as prominent [29] In the secondexperiment cells from the human thyroid carcinoma cellline FTC-133 were flown aboard the Shenzhou-8 spacecraftand fixed after 10 days in space Cells exposed to spaceflightappeared to form three-dimensional tumor spheroids whilethe inflight 1 g controls remained in two-dimensional mono-layers The FTC-133 cells exposed to simulated microgravityon the RPM also formed three-dimensional spheroids eventhough the spheroids appeared to be smaller than thoseformed in space [30] In addition EGF and CTGF geneexpressionwas upregulated in both real and simulatedmicro-gravity Interestingly EGF expression was lower and CTGFexpression was higher in the RPM samples than the spacesamples [30] The reason the RPM sample showed inter-mediate effects between the 1 g control and the space samplesis not clear at this point Since the RPM can only be used forslow processes one possible speculation is that some of theunderlying molecular processes might be too fast for RPM-simulated microgravity

In conclusion the RPM has been shown to mimic micro-gravity responses reliably for several but not all experimentalconditions Particularly for leukocytes several effects seenin space were reproduced on the RPM Particular stud-ies designed to investigate differences in cellular responses

between space samples and samples exposed to simulatedmicrogravity elucidated an underestimation or overestima-tion of simulated versus real microgravity Overall the RPMgenerally seems to underestimate the spaceflight effectsTherefore results from RPM experiments need to be inter-preted with caution and if possible more directly comparedto experiments under realmicrogravity in order to fully assesstheir capability to support gravitational biology studies

62 Novel Applications of the RPM The exact mechanismsby which mechanical stimuli initiate cellular modificationshave still not been fully elucidated [31]This is the motivationof mechanobiologists to expose cells to various mechani-cal stimuli such as distinct patterns of shear flow tensilestretch or mechanical compression at various parametriccombinations of magnitude duration or frequency [31] TheRPM can be regarded as an additional mechanical devicefor reducing the long-term effects of the mechanical force ofgravity Due to the constant reorientation of samples on theRPM gravity-dependent intracellular responses will not betriggered anymore Thus one can say that the RPM generatesa state of a mechanically unloaded environment in which thelonger-lasting impact of gravity can be studied

Monolayer (two-dimensional) cell cultures have beensuccessfully used for many decades allowing a better under-standing of many cellular and molecular processes Theyactually represent an important source of information priorto animal experimentation Despite numerous advantagesthe monolayer model cannot simulate organs or tissuesrealistically Therefore three-dimensional cell culturing hasemerged over the last decades as an alternative to mimicbetter tissue-like organization with the idea of closing thegap of uncertainty between tissue-like and monolayer cellculture The RPM in that context appears as an alternativeapproach to generating a three-dimensional culture [32]The random repositioning of the cells around the gravityvector over time allows constant redistribution of gravityforces which thus leads to the formation of cell aggregatesthat can form microspheroids (Figure 6) [32ndash34] Spheroidsorganized as multilayers are closer to in vivo tissue situationthan monolayer cells [32] Such samples are therefore moreaccurate as a model integrating the three-dimensional realsurroundings of a cell in an in vivo tissue Thus spheroidstructures open a new field of applications such as testsystems for drug therapies or diagnosis [35] The spheroidstructure is actually a good model to screen for penetrationcharacteristics of drugs or antibodies through tissue

7 Conclusion

Several RPMs have evolved during the past years that featuredifferent designs functions and motion patterns They allhave reliably proven to simulate microgravity conditionsDevelopments to RPMhardware and software have expandedthe experimental possibilities substantially The successfuloperation of digital holographic microscopy (DHM) on theRPM and the implementation of partial gravity algorithmshave opened new fields in gravitational research particularlyin mechanobiology


Figure 6Thyrocytes cultured for seven days on the RPM organizedto spheroid structures (arrow)

In order to obtain reliable and comparable data theappropriate use of the RPM and application of standardizedcultivation methods are of central importance The RPM hasbeen established as a reliable tool supporting ground-basedmicrogravity studies Effects seen in real microgravity werereproducedwith good agreement on RPMs Some RPM stud-ies however also showed cellular effects that were betweenthose of the real microgravity and 1 g ground control resultsThe RPM is furthermore an ideal tool for preliminary micro-gravity tests screening studies inwhich simulatedmicrograv-ity effects are checked on various organisms and hardwaretesting Particularly for suggesting live science experimentsfor the conduction under real microgravity in space thepresentation of preliminary data showing modificationsunder simulated microgravity is becoming very importantAdvances in RPM engineering and live science qualify theRPM as an interesting tool for novel applications such asthree-dimensional cell culturing as well as tissue engineering


The authors have no conflict of interests regarding thepublication of this paper

Acknowledgments

The authors thank their coworkers at the CC AerospaceBiomedical Science and Technology and especially NicoleWittkopf for the support and critical discussions Further-more they would like to thank Adrian Koller and MarianaReyes Perez from the CC of Mechanical Systems LucerneSchool of Engineering and Architecture for interesting dis-cussions and close collaboration Special thanks also go toJorg Sekler and his coworkers at the Institute for AutomationUniversity of Applied Science Northwestern Switzerland forthe fruitful collaboration

References

[1] R H Fitts S W Trappe D L Costill et al ldquoProlongedspace flight-induced alterations in the structure and function of

human skeletalmuscle fibresrdquo Journal of Physiology vol 588 no18 pp 3567ndash3592 2010

[2] D A Riley J L W Bain J L Thompson et al ldquoDecreased thinfilament density and length in human atrophic soleus musclefibers after spaceflightrdquo Journal of Applied Physiology vol 88no 2 pp 567ndash572 2000

[3] S W Trappe T A Trappe G A Lee J J Widrick D L Costilland R H Fitts ldquoComparison of a space shuttle flight (STS-78)and bed rest on human muscle functionrdquo Journal of AppliedPhysiology vol 91 no 1 pp 57ndash64 2001


[5] D Mesland ldquoNovel ground-based facilities for research in theeffects of weightrdquo ESAMicrogravity News vol 9 pp 5ndash10 1996



[8] T Hoson S Kamisaka Y Masuda M Yamashita and BBuchen ldquoEvaluation of the three-dimensional clinostat as asimulator of weightlessnessrdquo Planta vol 203 no 1 pp S187ndashS197 1997

[9] T Benavides Damm I Walther S L Wuest J Sekler and MEgli ldquoCell cultivation under different gravitational loads usinga novel random positioning incubatorrdquo Biotechnology and Bio-engineering vol 111 no 6 pp 1180ndash1190 2014

[10] S L Wuest S Richard I Walther et al ldquoA novel micrograv-ity simulator applicable for three-dimensional cell culturingrdquoMicrogravity Science and Technology vol 26 no 2 pp 77ndash882014

[11] M Cogoli ldquoThe fast rotating clinostat a history of its use ingravitational biology and a comparison of ground-based andflight experiment resultsrdquo ASGSB Bulletin vol 5 no 2 pp 59ndash67 1992

[12] R Hemmersbach M von der Wiesche and D Seibt ldquoGround-based experimental platforms in gravitational biology andhuman physiologyrdquo Signal Transduction vol 6 no 6 pp 381ndash387 2006

[13] M F Toy J Kuhn S Richard J ParentM Egli andCDepeurs-inge ldquoAccelerated autofocusing of off-axis holograms usingcritical samplingrdquo Optics Letters vol 37 no 24 pp 5094ndash50962012

[14] C Pache J Kuhn K Westphal et al ldquoDigital holographicmicroscopy real-time monitoring of cytoarchitectural alter-ations during simulated microgravityrdquo Journal of BiomedicalOptics vol 15 no 2 Article ID 026021 2010

[15] M F Toy S Richard J Kuhn A Franco-Obregon M Egliand C Depeursinge ldquoEnhanced robustness digital holographicmicroscopy for demanding environment of space biologyrdquoBiomedical Optics Express vol 3 no 2 pp 313ndash326 2012

[16] M F Toy C Pache J Parent J Kuhn M Egli and CDepeursinge ldquoDual-mode digital holographic and fluorescencemicroscopy for the study of morphological changes in cellsunder simulated microgravityrdquo inThree-Dimensional and Mul-tidimensional Microscopy Image Acquisition and ProcessingXVII pp 7570ndash7573 2010


[17] D M Klaus P Todd and A Schatz ldquoFunctional weightlessnessduring clinorotation of cell suspensionsrdquo Advances in SpaceResearch vol 21 no 8-9 pp 1315ndash1318 1998

[18] D Driss-Ecole V Legue E Carnero-Diaz and G PerballdquoGravisensitivity and automorphogenesis of lentil seedlingroots grown on board the International Space Stationrdquo Physi-ologia Plantarum vol 134 no 1 pp 191ndash201 2008

[19] C A Leguy R Delfos M J B M Pourquie et al ldquoFluidmotionformicrogravity simulations in a randompositioningmachinerdquoGravitational and Space Biology Bulletin vol 25 no 1 2011

[20] T Hammond and P Allen ldquoThe Bonn criteria minimal experi-mental parameter reporting for clinostat and random position-ing machine experiments with cells and tissuesrdquo MicrogravityScience and Technology vol 23 no 2 pp 271ndash275 2011

[21] M Cogoli-Greuter ldquoThe lymphocyte storymdashan overview ofselected highlights on the in vitro activation of human lympho-cytes in spacerdquoMicrogravity Science and Technology vol 25 no6 pp 343ndash352 2014

[22] M Schwarzenberg P Pippia M A Meloni G Cossu MCogoli-Greuter and A Cogoli ldquoSignal transduction in T lym-phocytesmdasha comparison of the data from space the free fallmachine and the random positioning machinerdquo Advances inSpace Research vol 24 no 6 pp 793ndash800 1999


[24] A Villa S Versari J A Maier and S Bradamante ldquoCell behav-ior in simulated microgravity a comparison of results obtainedwith RWV and RPMrdquo Gravitational and Space Biology Bulletinvol 18 no 2 pp 89ndash90 2005

[25] J P Hatton F Gaubert M L Lewis et al ldquoThe kinetics oftranslocation and cellular quantity of protein kinaseC in humanleukocytes aremodified during spaceflightrdquoTheFASEB Journalvol 13 supplement pp S23ndashS33 1999


[27] M A Meloni G Galleri P Pippia and M Cogoli-GreuterldquoCytoskeleton changes and impaired motility of monocytes atmodelled low gravityrdquo Protoplasma vol 229 no 2-4 pp 243ndash249 2006

[28] N Battista M A Meloni M Bari et al ldquo5-Lipoxygenase-dependent apoptosis of human lymphocytes in the interna-tional space station data from the ROALD experimentrdquo TheFASEB Journal vol 26 no 5 pp 1791ndash1798 2012

[29] V Stamenkovic G Keller D Nesic A Cogoli and S P GroganldquoNeocartilage formation in 1 g simulated and microgravityenvironments implications for tissue engineeringrdquo Tissue Engi-neeringmdashPart A vol 16 no 5 pp 1729ndash1736 2010


[31] J Wang D Lu D Mao and M Long ldquoMechanomics anemerging field between biology and biomechanicsrdquo Protein ampCell vol 5 no 7 pp 518ndash531 2014

[32] J Pietsch A Sickmann G Weber et al ldquoA proteomic approachto analysing spheroid formation of two human thyroid cell linescultured on a random positioning machinerdquo PROTEOMICSvol 11 no 10 pp 2095ndash2104 2011

[33] D Grimm J Bauer C Ulbrich et al ldquoDifferent responsivenessof endothelial cells to vascular endothelial growth factor andbasic fibroblast growth factor added to culture media undergravity and simulated microgravityrdquo Tissue Engineering Part Avol 16 no 5 pp 1559ndash1573 2010

[34] C Ulbrich ldquoCharacterization of human chondrocytes exposedto simulated microgravityrdquo Cellular Physiology and Biochem-istry vol 25 no 4-5 pp 551ndash560 2010

[35] A Ivascu and M Kubbies ldquoRapid generation of single-tumorspheroids for high-throughput cell function and toxicity analy-sisrdquo Journal of Biomolecular Screening vol 11 no 8 pp 922ndash9322006

Research ArticleRegulation of ICAM-1 in Cells of the MonocyteMacrophageSystem in Microgravity

Katrin Paulsen1 Svantje Tauber12 Claudia Dumrese13 Gesine Bradacs1

Dana M Simmet12 Nadine Goumllz1 Swantje Hauschild12 Christiane Raig1

Stephanie Engeli1 Annett Gutewort12 Eva Huumlrlimann1 Josefine Biskup1

Felix Unverdorben2 Gabriela Rieder1 Daniel Hofmaumlnner1 Lisa Mutschler1

Sonja Krammer1 Isabell Buttron1 Claudia Philpot4 Andreas Huge5 Hartwin Lier6

Ines Barz7 Frank Engelmann67 Liliana E Layer1 Cora S Thiel12 and Oliver Ullrich1289

1 Institute of Anatomy Faculty of Medicine University of Zurich Winterthurerstraszlig 190 8057 Zurich Switzerland2Department of Machine Design Engineering Design and Product Development Institute of Mechanical EngineeringOtto-von-Guericke-University Magdeburg Universitatsplatz 2 39106 Magdeburg Germany3Flow Cytometry Facility University of Zurich Winterthurerstraszlig 190 8057 Zurich Switzerland4German Aerospace Center Space Agency Konigswinterer Straszlige 522-524 53227 Bonn Germany5Integrated Functional Genomics (IFG) University of Muenster Roentgenstraszlig 21 48149 Muenster Germany6KEK GmbH Kemberger Straszlige 5 06905 Bad Schmiedeberg Germany7University of Applied Science Jena Carl-Zeiss-Promenade 2 07745 Jena Germany8Zurich Center for Integrative Human Physiology (ZIHP) University of Zurich Winterthurerstraszlig 190 8057 Zurich Switzerland9Study Group ldquoMagdeburger Arbeitsgemeinschaft fur Forschung unter Raumfahrt- und Schwerelosigkeitsbedingungenrdquo (MARS)Otto-von-Guericke-University Magdeburg Universitatsplatz 2 39106 Magdeburg Germany




Copyright copy 2015 Katrin Paulsen et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Cells of the immune system are highly sensitive to altered gravity and themonocyte as well as themacrophage function is proven tobe impaired under microgravity conditions In our study we investigated the surface expression of ICAM-1 protein and expressionof ICAM-1 mRNA in cells of the monocytemacrophage system in microgravity during clinostat parabolic flight sounding rocketand orbital experiments In murine BV-2 microglial cells we detected a downregulation of ICAM-1 expression in clinorotationexperiments and a rapid and reversible downregulation in the microgravity phase of parabolic flight experiments In contrastICAM-1 expression increased inmacrophage-like differentiated humanU937 cells during themicrogravity phase of parabolic flightsand in long-termmicrogravity provided by a 2D clinostat or during the orbital SIMBOXShenzhou-8 mission In nondifferentiatedU937 cells no effect of microgravity on ICAM-1 expression could be observed during parabolic flight experiments We concludethat disturbed immune function in microgravity could be a consequence of ICAM-1 modulation in the monocytemacrophagesystem which in turn could have a strong impact on the interaction with T lymphocytes and cell migration Thus ICAM-1 can beconsidered as a rapid-reacting and sustained gravity-regulated molecule in mammalian cells

1 Introduction

Several limiting factors for human health and performancein microgravity have been clearly identified arising fromthe immune system and substantial research activities are

required in order to provide the basic information for appro-priate integrated risk management The gravity-sensitivenature of cells of the immune system renders them anideal biological model in search for general gravity-sensitivemechanisms to understandhow the architecture and function



of human cells are related to the gravitational force andtherefore adapted to life on Earth Cells of the immune systemare highly sensitive to altered gravity (for review see [1ndash4]) T lymphocytes as well as monocytes and macrophagesare impaired severely in their functions under microgravityconditions [2ndash4] T cell activation is severely disturbed undermicrogravity conditions as shown in the blood of astronautsduring and after space flight [5] and in numerous in vitroexperiments (reviewed by [6]) In monocytes the secretionof the cytokines IL-1 IL-6 TNF-alpha and IL-10 is alteredunder microgravity conditions [7 8] Substantial changes ingene expression of monocytes and in gene induction associ-ated with the differentiation of monocytes into macrophageshave been observed [8]

Migration and adhesion of immune competent cells atareas of infection inflammation or structural disordersare indispensable for the immune response [9] For theseprocesses the communication and connection between cellsare essential The integrins of the LeuCAM family (LFA-1 and MAC-1) and their ligands the intercellular adhesionmolecules (ICAMs) are receptors that mediate the attach-ment between cells (cell-cell contact) and of cells and theextracellular matrix (cell-matrix contact) [10] ICAMs aretransmembrane proteins that are expressed on epithelial cellsendothelial cells and cells of the immune system includingT cells and macrophages Binding of ICAM-1 (CD54) toreceptors on endothelia of blood vessels enables leucocytesto attach and migrate through the endothelia to sites ofinflammation [11] Later on in the immune reaction closeand strong interaction between ICAM-1 and LFA-1 is indis-pensable for the immunological synapse formation betweenT cells and antigen-presenting cells such as monocytes[12]

ICAM-1 expression is known to be upregulated duringmechanical stress [13] in a long-term microgravity envi-ronment [14] in the NASA-developed Rotary Cell CultureSystems (RCCS) as well as during short-term microgravityin parabolic flights [15] in endothelial cells While thesestudies show gravity sensitivity of ICAM-1 in endothelialcells less is known about the effects of microgravity on cellsof the 2 monocytemacrophage system (MMS) Thereforein this study we investigate whether the ICAM-1 surfaceexpression is regulated by altered gravity in these cell typesThe MMS belongs to the innate immune system and rep-resents the bodyrsquos first line of defense The innate immunesystem is characterized by a fast but unspecific immunereaction and it activates the adaptive immune responseThisactivation occurs through interaction of antigen-presentingcells (APCs)mdashdendritic cells and macrophages [16]mdashwithT lymphocytes Macrophages are relatively long-lived carrya variety of surface receptors and reside in many tissuesincluding the gastrointestinal tract the respiratory tract theliver the spleen bones and connective tissues [17]Microglialcells are the brain-resident macrophage population whichcrucially controls and regulates immune reactions inside thecentral nervous system (CNS)

In our study we investigated the surface expression ofICAM-1 protein and expression of ICAM-1 mRNA in cellsof the monocytemacrophage system in microgravity As cell

models we used primary cells (macrophages T cells) as wellas cell lines (U937 myelomonocytic cells macrophage-likedifferentiated U937 cells and BV-2 microglial cells) We con-ducted experiments with different durations of microgravityin clinorotation parabolic flight sounding rocket and orbitalflight experiments

2 Methods

21 U937 Cell Culture and Macrophage-Like DifferentiationU937 cells (ATCC CRL-15932) are a human monocyticcell line that preserves the main monoblastic character-istics of monocytes including the ability to differentiateinto a macrophage-like phenotype U937 cells were culturedin RPMI 1640 medium with or without 20mM HEPES(Biochrom Berlin Germany) supplemented with 10 fetalcalf serum (FCS Biochrom) or 10 human serum (HSBiowest) 2mMglutamine (PromoCell) 100UmLpenicillinand 100 120583gmL streptomycin (Gibco) Subcultivation wasdone at a cell density of 1 times 106 cellsmL Stimulationand differentiation were performed by adding 25 nM phor-bolmyristylacetate (PMA) (Sigma-Aldrich) in dimethylsul-foxid (DMSO 01) (Sigma-Aldrich) at a cell density of05times 106 cellsmLDifferentiationmediumwas supplementedwith 10 HS 2mM glutamine 100UmL penicillin and100 120583gmL streptomycin Cells were differentiated on poly-carbonate (Makrolon) slides (SIMBOXShenzhou-8) or incell culture flasks (parabolic flights and clinorotation exper-iments) for 72 h at 37∘C with 5 CO

2into a macrophage-

like phenotype U937macrophage-like cells were detached byMacrophage Detachment Solution DXF (PromoCell) follow-ing manufacturerrsquos protocol After detaching cells were filledimmediately into Nutrimix bags (B BraunMelsungen) (1-2 times107 cells in 10mL medium each bag) for parabolic flight orin serological pipettes (1mL cell suspension with 025ndash05 times106mL) for clinorotation experiments [18 19]

22 Primary Human Macrophages Human primary M2macrophages (PromoCell) were cultivated with M2-Macrophage Generation Medium DXF (PromoCell) Cellswere detached by Macrophage Detachment Solution DXF(PromoCell) following the manufacturerrsquos protocol Afterdetaching cells were filled immediately into Nutrimix bags(B Braun Melsungen) (2 times 106 cells in 10mL medium eachbag) for parabolic flight or in serological pipettes (1mLcell suspension with 025ndash05 times 106mL) for clinorotationexperiment [18 19]

23 BV-2 Microglia Cell Culture Since primary microglia arenot available in quantities required for parabolic flight exper-iments we used the murine cell line BV-2 whose functionresembles that of tissue macrophages and which share manyproperties with both peripheral macrophages andmonocytes[20] BV-2 microglial cells were cultured in DMEMmedium(Biochrom Berlin Germany) supplemented with 10 FCSand without antibiotics 72 hours before parabolic flightexperiment the cells were set on only 2 FCS (serumstarved) for the transport and were supplemented again


(a) (b)

(c) (d)

Figure 1 Technology for cell culture experiments in different microgravity research platforms (a) Fast-rotating two-dimensional (2D)clinostat manufactured by the German Aerospace Center (DLR Cologne Germany) was used to provide simulated microgravity Underthe chosen experimental conditions (60 rpm 4mm pipette diameter) a maximal residual acceleration of 4 times 10minus3 g is achieved at the outerradius of the pipette and decreases towards the center (b) Experimental hardware structure which consists of an incubator rack to store thecell containers temporarily before the experiment at 37∘C (left) an experimental rack in which all active aggregates are accommodated andwhere the living cells are handled during altered gravity (right) and a cooling rack to temporarily store all cell containers after the injection ofthe stopfixation liquid at 4∘C until landing (front) (c) Payload of TEXUS-49 sounding rocket tempered and vacuum-resistant container withexperiment syringe systems (d) Plunger unit EUE for SIMBOX (Science in Microgravity Box) incubator system support structure (housingmade of PEEK) which includes three culture chambers and six supply units two for each culture compartment Each culture chamberrepresents an independent loop The culture chambers filled with medium are closed on the top of the housing by means of polycarbonatespecimen window slides where the adherent cells are attached beforehand The housing is tightened by silicon sealing and covered by analuminum plate (cover) fixed with screws

with 10 FCS after arrival Before transport to Bordeaux-Merignac airport BV-2 cells were transferred into 200mLNutrimix bags (B Braun Melsungen Melsungen Germany)at a density of 3 times 106 cells in 15mL medium

24 Experiments in Simulated Microgravity (2D Clinorota-tion) A fast-rotating two-dimensional (2D) clinostat man-ufactured by the German Aerospace Center (DLR CologneGermany) was used to provide simulated microgravity(Figure 1(a)) The principle of clinorotation-induced micro-gravity is the rotation of a cell suspension in a serologicalpipette perpendicular to the Earthrsquos gravityThemicrogravityproduced is an averaging of the gravity vector if the clinostatrotates with 40ndash100 rpm Under the chosen experimental

conditions (60 rpm 4mmpipette diameter) amaximal resid-ual acceleration of 4 times 10minus3 g is achieved at the outer radiusof the serological pipette and decreases towards the centerThe clinostat device was placed in an incubator providingconstant 37∘C Fifteen serological pipettes rotated at the sametime with 60 rpm 1 g controls were placed at the ground plateof the clinostat without rotation but the same environmentalconditions are as the 120583g samples The density of the cellsuspension was 05 times 106mL (U937 macrophage-like cells)025ndash05 times 106mL (human primary macrophages) or 075 times106mL (BV-2 microglial cells) in 1mL volume each Theduration of the filling procedure was not longer than 10minfor all 30 serological pipettes Cells were cultured in theserological pipettes for 24 hndash96 h After clinorotation cells


were fixed by the addition of 500 120583L of 3 PFA (Sigma-Aldrich)2 sucrose (Sigma-Aldrich) solution for 30minwashed with PBS and analyzed after immunocytochemicalstaining by flow cytometry

25 Parabolic Flights asMicrogravity Research Platform Dur-ing a parabolicmaneuver an aircraft is weightless by flying ona Keplerian trajectory described as an unpropelled body inideally frictionless space subjected to a centrally symmetricgravitational field [21] During this free-fall trajectory theresultant of all forces acting on the aircraft other thangravity is zeroed During a flight campaign which normallyconsists of three individual flights 31 parabolas are flownon each flight with 93 parabolas in total On each parabolathere is a period of increased gravity (18 g) which lastsfor 20 seconds immediately prior to and following the 20-second period of reduced gravity (acceleration in 119909- 119910-and 119911-axis was below 2 times 10minus3 g at all times during the 120583gparabola Figure 1(a)) During the parabolic flight maneuverthe aircraft gradually pulls up its nose and starts climbingat an angle of approximately 45 degrees This phase lastsfor about 20 seconds during which the aircraft experiencesan acceleration of around 18 g The engine thrust is thenreduced to the minimum required to compensate for air-drag and the aircraft is then in a free-fall condition lastingapproximately for 20 seconds during which weightlessness isachieved At the end of this phase the aircraft must pull outthe parabolic arc a maneuver which gives rise to another 20second period to 18 g on the aircraft after which it returnsto normal level flight attitude Special designated flight areaswere above the Atlantic Ocean and the Mediterranean SeaExperiments were conducted during the 10th and 19th DLRparabolic flight campaigns of the German Aerospace Center(ldquoDeutsches Zentrum fur Luft- und Raumfahrtrdquo DLR) inBordeaux France The campaign used the only large aircraftthat is licensed in Europe to perform parabolic flights forresearch purposes the Airbus A300 ZERO-GThis aircraft isa specially configured test aircraft operated by NOVESPACE(Bordeaux France) according to the standing orders ofNOVESPACE (A300 ZERO-G Rules and Guidelines RG-2001-1 RG-2008-1 RG-2008-2 RG-2009-1 andRG-2009-02)and the CEV (Centre drsquoessai en vol)

26 In-Flight Hardware for Parabolic Flight Experiments Acustom-made hardware meeting the requirements for exper-iments with human cell culture on board the Airbus A300ZERO-G was developed in collaboration with KEK GmbHGermany (Figure 1(b)) The system has already been usedsuccessfully for cell culture experiments during 9 parabolicflight campaigns [18 19]The system consists of double-sealedcell containers holding the cells of themonocyte-macrophagesystem and three experimentalmodules that supply storage ofsamples before the experiment half-automated performanceof the experiment and storage of the processed samplesThe first module holds the cell containers at 365∘C in ahanging position From there containers are transferred intothe second module manually In this module cells were fixedby the addition of fixation reagent upon triggering Triggering

was done manually at defined time intervals (20 sec) afterthe onset of the gravitational condition of interest Thethird experimental module served as in-flight storage for thefixated samples at 4∘C Three samples could be processedin parallel Sample exchange required approximately oneminute of a defined procedure by three trained persons

27 Procedures during Parabolic Flight Experiments Trans-port of in-flight cell culture bags in in-flight-configurationand of fixed samples after the parabolic flight was providedby the Swiss Air Force from Zurich to Bordeaux during eachflight day of the 13th DLR Parabolic Flight Campaign or bytrain during each flight day of the 19th DLR Parabolic FlightCampaign After arrival at the flight location on the eveningbefore the flight cells were incubated overnight at 37∘C andhandled very carefully in order to avoid any mechanical ortemperature cell stress All steps of the entire cell preparationand transport procedure had been tested extensively withrespect to cell viability and function beforehand All proce-dures during the parabolic flight campaign had been testedseveral times and highly standardized following an extensiveand detailed standard protocol During the campaign allprocedures were documented and double-checked In-flight120583g and 1 g control experiments were performed in 200mLNutrimix bags [18 19] used as in-flight cell culture bags con-taining 3times 107 cells in 15mLDuring the onset of 120583g or during1 g (in-flight control experiments) 10 ng PMAmL (with 001residual DMSO) or 10 ng TNF-120572mL or plain cell culturemediumwere added to the cells After 20 sec of 120583g or 1 g cellswere fixed by addition of 1 formaldehyde (Sigma-Aldrich)(for cytometry analysis) or lysed by RLT buffer (Qiagen)(for RNA analysis) and cooled immediately (4∘C) during theremaining flight Experiments were performed at least threetimes during independent flights and separate flight daysAfter the flight fixed cells were transported to the laboratorieson the same day harvested and subjected to analysis

28 TEXUS-49 Sounding Rocket Experiment For theTEXUS-49 campaign at ESRANGE (Kiruna Sweden) U937cells were cultivated in the fully installed laboratories on siteCells were seeded with a density of 02 times 106 cellsmL and themedium was exchanged every 48 hours as described aboveOn the launch day cells were visually inspected harvestedcounted and pooled to a concentration of 5 times 107 cellsmL05mL of this cell suspension was filled into a sterile 3mLplastic syringe shortly before the launch Additionallyone syringe was filled with 03mL of cell culture mediumand another one with 1mL Trizol LS (Life TechnologiesGermany) The three syringes were mounted on a plasticblock with a tubing system connecting them This unit wasfinally integrated into the automatically operated experimentsystem (Figure 1(c)) In total 35 of these experiment unitswere prepared and kept at 37∘C until the integration intothe payload of the rocket During the experimental runfirst the 03mL of medium as a potential placeholder foran activation solution and then the 1mL of Trizol LS wereinjected to the cell suspension at defined time points to lysethe cells and preserve the current status of differential gene


expression Injections were performed at 75 sec after launchto monitor a so-called baseline (BL) directly before the 120583gphase and at 375 sec after launch at the end of the 120583g phaseA group of 1 g ground controls was treated immediatelyafter the 120583g sample group TEXUS-49 consisted of a VSB-30engine (S-30 solid rocket stage with an S-31 second stage)and of the payload The rocket was launched on March 292011 at 0601 am from the ESRANGE Space Center nearKiruna Sweden During the ballistic suborbital flight analtitude of 268 km and 378 sec of microgravity with a qualityof 10minus5 g were achieved

29 SIMBOX Incubator System with Plunger ExperimentInsert SIMBOX (Astrium GmbH Friedrichshafen Ger-many Kayser Italia Livorno Italy) is a programmable space-qualified incubator for biological research in space equippedwith a 1 g in-flight centrifuge for 1 g control experiments Theincubator allows for fully automatic execution of biologicalexperiments with limited use of commands during orbitalflight in a controlled thermal environment The SIMBOXincubator (internal volume 34 liters dimensions 461 times 551 times273mm empty mass 16 kg fully integrated mass 345 kgmax power 130W) accommodates 40 experiment uniqueequipment (EUEs) with 24 EUEs on the 120583g-platform and16 EUEs on the 1 g-centrifuge The plunger experimentinsert (Figure 1(d)) was developed by Astrium GmbH andis described in the Astrium Space Biology Product Catalog[22] It allows medium exchange and chemical fixation ofadherent cell cultures There are two plungers which can befilled with any liquid and automatically activated to injectit into the experimental volume The EUEs consisted of asupport structure (housing made of PEEK) which includesthree culture chambers (CCs) and six supply units (SUplungers) two per culture compartment Each CC has twoSUs and represents an independent loop The CCs are closedon the top of the housing by Specimen Slides (SS) made ofpolycarbonate on which the adherent cells were attachedThe chamber (covered by the window slide) contained themedium The housing is tightened by silicon sealing andcovered with an aluminum plate (cover) which is fixedwith screws The container lid of the Biorack standardtype I container is mounted onto the housing The Biorackstandard is based on the accommodation of various EUEsinto experiment containers which provide the interface tofacilities and support infrastructure [22] The plunger unit isqualified for an unmanned capsulemission and for use on theInternational Space Station (ISS)

The unmanned Shenzhou-8 spacecraft was launched onOctober 31 2011 at 2158 UTC (November 1 2011 0558LT) on board of a Long March 2F (CZ-2F) rocket from theJiuquan Satellite Launch Center (JSLC) in Inner MongoliaOn November 17 the capsule was autonomously deorbitedand landed at 1238 UTC (2038 LT) around 500 km northof Beijing The SIMBOX was recovered immediately andtransported by helicopter and jet aircraft to the PITCBeijing Total early retrieval time was 6 hours On arrivalat the PITC the SIMBOX was opened and the EUEswere removed and inspected The samples were recovered

and stored in cold (4∘C) PBS until arrival in Zurich foranalysis

210 SIMBOX Experiment Execution Medium was changedbefore integration of the slides into the EUEs Insidethe EUEs the slides were bedded in 05mL fully CO

2

saturated RPMI 1640 medium with 10 HS 2mM glu-tamine 100UmL penicillin 100 120583gmL streptomycin and250 ngmL amphotericin B (PromoCell) Bellow 1 3 and 5 ofthe EUEs [23] were filled with RPMI 1640 medium 10 HS100UmL penicillin 100120583gmL streptomycin 250 ngmLamphotericin B 2mM glutamine 1 PFA and 06 sucroseBellow 2 4 and 6 of the EUEs [23] were filled withPBS 100UmL penicillin 100 120583gmL streptomycin and250 ngmL amphotericin B Two EUEs (6 chambers) wereprepared for the 120583g-position and one EUE (3 chambers) wasprepared for the 1 g position The gravity vector of the 1 gposition was perpendicular to the surface of sample slides (119911-axis) During the unpowered transport from the laboratory tothe spacecraft and installation in the spacecraft (total time 3 h06min) the temperature was always above 21∘C Shenzhou-8launch was on October 31 2011 2158 UTC and the spacecraftattained orbit at 2208 UTC The SIMBOX timeline startedat 2234 UTC Active temperature control was set to 23∘CThe centrifuge speed for the 1 g reference centrifuge was7440 rpm Plungers 1 3 and 5 of all three EUEs wereactivated between 1205000 and 1205520 (hoursminsec)of the timeline sequences of 40 seconds Plungers 2 4 and6 of all three EUSs were activated between 1225000 and1225522 (hoursminsec) of the timeline sequences of 40seconds Human macrophage-like U937 cells were cultivatedfor 5 days inside the SIMBOX hardware on board of theShenzhou-8 spacecraft in 120583g and 1 g conditions fixed with1 PFAsucrose solution for 2 h and stored in PBS on boardat 23∘C until landing After landing the polycarbonate slideswere removed washed and then stored in PBS at 4∘C for2 weeks until analysis The ground control experiment wasexecuted analogously to the flight scenario Details about theexperiment were published previously [23]

211 Quantification of ICAM-1 by Flow Cytometry Surfaceexpression of ICAM-1 on BV-2 microglial cells and U937monocytic and macrophage-like cells as well as primaryhuman macrophages was analyzed by flow cytometry Cellswere collected from the Nutrimix bags (parabolic flight)or standardized serological pipettes (clinorotation) fixatedin PFAsucrose solution After the washing procedure (PBSwithout CaMg Biochrom) cells were stained with ICAM-1monoclonal antibody (BV2 Invitrogen FITC labeled U937and primarymacrophages cell signaling PE labeled) Analy-sis was performed using a flow cytometer (FACSCanto II BDBiosciences Heidelberg Germany) collecting at least 20000cells per sampleMean fluorescence intensity Ratio (MFI)wascalculated as MFI of sampleMFI of isotype control

212 ICAM-1 Analysis in BV-2 Cells from Parabolic FlightExperiments Cells were quadruple stained for ICAM-1apoptosis (TUNEL) cell delineation (HCS cell mask) and


DNA (DAPI) In brief cells were cytospinned onto glassslides washed 3x with PBS permeabilized for 1min with01 Triton-X 100 (Sigma-Aldrich Buchs Switzerland) andwashed again 3x with PBS and incubated with the TUNELlabeling mix (Boehringer Mannheim Germany) accordingto themanufacturerrsquos instructions For TUNEL staining rho-damine coupled dUTP was used Subsequently to overnightincubation cells were washed again 3x with PBS blockedwith 05 BSA and stained with FITC labeled ICAM-1antibody (BD Pharmingen San Jose USA) at a concentrationof 005mgmL for 2 h After additional washing cells werestained entirely with HCS cell mask deep red cytoplasmicand nuclear stain (Invitrogen Basel Switzerland) using adilution of 1 20000 and nuclei were labeled with DAPI(Invitrogen) at 1 120583gmL for 10min Labeled cells were imagedusing a Leica microscope DMI 6000 and LAS AF software(Leica Microsystems Wetzlar Germany) For automatedimaging the unified random sampling module was utilized63 randomized images of each sample were recorded and atleast 500 single cells from 3 independent experiments from3 different parabolas were analyzed From each image cellswere identified according to the following criteria nucleusof a predefined size and brightness being TUNEL negativeand containingHCS staining over a certain threshold Surfacecalculation of these cells was performed with Imaris andautomated for all images using batch coordinator (BitplaneAG Zurich Switzerland) Therefore the mean intensity ofthe ICAM-1 signal was analyzed in living cells exclusivelyand binned into ICAM-1 intensity categories of 50 graylevels Statistical analysis was carried out using GraphPadPrism software (GraphPad Software Inc La Jolla USA) andStudentrsquos 119905-test was applied for all analyzed data

213 ICAM-1 Analysis in Differentiated U937 Cells fromthe SIMBOX Experiment Polycarbonate slides were cut bya water jet method into 16 T-shaped pieces Each piecewas stained individually In order to differentiate betweendead (necroticapoptotic) and living cells before fixationslides were stained with CellMask-deep red plasma mem-brane stain (Invitrogen) andTUNEL reagent (Fluorescein-12-dUTP Roche) In addition cells were labeled with differentmono- and polyclonal primary antibodies directed againstthe cytoskeleton components and immunological relevantsurface molecules (reported in [17]) and ICAM-1 in con-centrations according to the manufacturersrsquo protocols Afterblocking with 1 BSA in PBS for 1 h primary antibodies weredetected by species specific secondary antibodies used in adilution of 1 1000 in 05 BSA in PBS Secondary antibodieswere labeled with Alexa-Fluor405 or Alexa-Fluor568 (Invit-rogen) Slide pieces were analyzed by confocal laser scanningmicroscopy (Leica SP5) Only cells positive for CellMask andnegative for TUNEL were subjected to further analysis sincethese represent the living cell population in the experimentDigital image analysis was performed using Imaris software(Bitplane)

214 RNA Isolation from the Parabolic Flight ExperimentsAfter the return of the aircraft and transport of

the samples to the on-site laboratory facilities the containerswere disassembled the Nutrimix bags were gently agitatedand the lysed cell solution from each bag was filled into aT75 straight neck cell culture flask The cell solution wasvortexed for 10 sec and passed four times through a Oslash08 times 120mm needle (B Braun Melsungen Germany) fittedto a 50mL syringe 50mL of absolute ethanol was addedand precipitates were resuspended by vigorous shakingA valve and a sterile connective piece were placed on aQiavac 24 plus vacuum system (Qiagen Germany) and anRNA maxi column (Qiagen Germany) was attached to theconnective piece A vacuum of minus200mbar was adjusted andthe column was loaded with the lysed cell suspension Thenthe valve was closed and the column was centrifuged at4000 g for 3min 15mL of buffer RW1 (Qiagen Germany)was applied for washing membrane bound RNA Aftercentrifugation at 4000 g for 7min the flow was discardedand two washing steps with 10mL RPE buffer (QiagenGermany) followed each with centrifugation at 4000 g for3min and 10min respectively The column bound RNA waseluted by application of 600 120583L of RNase-free water (QiagenGermany) incubation for 1min at room temperature andcentrifugation for 4min at 4000 g The elution step wasrepeated with the first eluate The RNA was transported atapproximately minus150∘C in a Cryo Express dry shipper (CX-100 Taylor-Wharton USA) prepared with liquid nitrogenand stored at minus80∘C until the processing of the RNA for themicroarray analysis

215 RNA Isolation during the TEXUS-49 Sounding RocketCampaign Directly after landing localization and recoveryof the payload the experiment modules were dismantledand handed over to the scientists The cell suspension wassheared three times with a 20G needle (B BraunMelsungenGermany) and distributed in two 20mL tubes 01mL ofchloroform (Sigma-Aldrich Germany) was added and thesolution was vortexed for 15 sec and incubated for 5minat room temperature before a 15min centrifugation step at11000 g and 4∘C The upper phase of both 20mL tubes wastransferred into a 15mL tube and 4mL of RLT buffer and3mL of absolute ethanol were added and mixed 4mL ofthis solution was pipetted on an RNAMidi column (QiagenGermany) and centrifuged for 30 sec at 3000 g at roomtemperatureTheflowwas discarded and the residual 4mLofRNA solution was loaded on the column and centrifuged for5min at 3000 g at room temperatureThen the columns werewashed twice with 25mL of RPE buffer and centrifuged for2min and 5min respectively at 3000 g and room tempera-ture The RNA was eluted by the addition of 250120583L RNasefree water (Qiagen Germany) to the column incubationfor 1min at room temperature and centrifugation for 3minat 3000 g and room temperature The eluate was loadedagain onto the column followed by a 1min incubation andcentrifugation for 5min at 3000 g and room temperatureThe isolated RNAwas transferred into sterile Cryo-tubes andstored until the return transport at approximately minus150∘Cin a Cryo Express dry shipper (CX-100 Taylor-WhartonUSA) preparedwith liquid nitrogen After arrival in the home


laboratory samples were stored at minus80∘C until processing theRNA for the microarray analysis

216 RNA Processing and Microarray Analysis for ParabolicFlight and TEXUS-49 Sounding Rocket Campaign SamplesRNA quantity and purity were analyzed spectrophotomet-rically using a Nanodrop 1000 (Thermo Scientific) IsolatedRNA samples were all of high quality with 260280 nmratios between 19 and 21 The RNA integrity was measuredusing an Agilent 2100 Bioanalyzer (Agilent TechnologiesUSA) Only RNA with an RNA Integrity Number (RIN) gt87 was used for the following microarray analysis 400 ngtotal RNA was applied to Cy3-labeling with the ldquoLow RNAInput Linear Amplification Kit PLUS One-Colorrdquo (AgilentTech-nologies) and hybridized for 175 h to a NimbleGenexpression microarray (12 times 135000 features) employing theldquoGene Expression Hybridization Kitrdquo (Agilent TechnologiesUSA) Afterwards arrays were washed and scanned by theMicro Array Scanner G2505B (Agilent Technologies USA)

The image files of the scanner were analyzed with theNimbleScan Software 26 using the RobustMulti-ArrayAnal-ysis (RMA) with the default parameters RMA a probe-levelsummarization method identifies probes that are outliersin the overall behavior of the expression measured for agiven gene The contribution of outlier probes is reducedin the reported gene expression level which has beendemonstrated to improve the sensitivity and reproducibilityof microarray results In addition to screening outlier probesNimbleScan softwarersquos implementation of RMAused quantilenormalization and background correction The normalizedmicroarray data were analyzed using Partek Genomics Suite66 Statistical analysis was performed using the one-wayANOVA and the false discovery rate (FDR) for multipletesting corrections Further the coefficient of variation (CV)expressed in percent was calculated also known as ldquorelativevariabilityrdquo which equals the standard deviation divided bythe mean Genes of interest were identified and the log2values of the measured fluorescent intensities returned bythe Partek software were back calculated to linear valuesThen means of all values of the same gene generated bydifferent probes were calculated if at least three values existedexcluding outliers Subsequently standard deviations werecalculated for the means and an unpaired 119905-test with Welchcorrection was performed to test statistical significance

217 Pathway Enrichment Analysis The pathway enrichmentanalysis was performed using Partek Genomics Suite 66 andthe KEGG human pathway library [24 25]The119875 values werecalculated by the Fisher exact test Enrichment analysis wasapplied on the genes showing differential expression with 119875values of lt005 and fold change gt+15 or lt minus15

218 Statistical Analysis Data are expressed as median oras median plusmn SE Groups contain the analysis of 200ndash1000cells (SIMBOX shown in box-plots) or data of three inde-pendent experiments with 1ndash5 samples (119899 = 3ndash15 shown incolumns) Data were analyzed by one-way ANOVA followedby Wilcoxon or unpaired 119905-test using GraphPad Prism 5

lowast119875 lt 01 was considered to be significant lowastlowast119875 lt 005 assignificant and lowastlowastlowast119875 lt 001 as very significant

3 Results

31 Clinorotation of Downregulated ICAM-1 Expression inBV-2 Microglial Cells First we analyzed ICAM-1 expressionin BV-2 microglial cells after 24 h clinorotation (60 rpm4mm pipette diameter maximal residual acceleration of 4 times10minus3 g at the outer radius of the pipette) The clinostat devicewas placed in an incubator which provides constant 37∘CFifteen serological pipettes rotated at the same time with60 rpm 1 g controls were placed at the ground plate of theclinostat without rotation but with the same environmentcondition like 120583g samples A 1 g control group of BV-2 cellswas filled into 1mL serological pipettes in the same way asthe clinorotation cell group but was not clinorotated Anothercontrol group was kept at regular cell culture conditionsin the incubator (37∘C 5 CO

2) Cells were subsequently

stained for cell surface ICAM-1 apoptosis (TUNEL) celldelineation (HCS CellMask) and DNA (DAPI) (Figure 2)ICAM-1 expression analysis by flow cytometry revealed twodistinct subtypes of cells in the clinorotated group (120583ggroup) compared to the 1 g control group and the incubatorcontrol group consisting of only one subtype respectively(Figure 2(a)) The first of the two subtypes was small andstronger granulated (subtype 1) than the second subtypewhich appears taller but less granulated (subtype 2) Apop-totic cells were excluded from the analysis by TUNELstaining Subtype 1 could possibly represent an activated stateSubtype 1 was found in the 120583g group as well as in the 1 gcontrol group whereas the incubator control did virtuallynot contain this subtype Subtype 2 was represented in allthree cell groups 120583g 1 g control and incubator control cellgroup However it was primarily present in the 120583g andin the incubator control group and less present in the 1 gcontrol groupThe population distribution within cell groupsis illustrated in Figure 2(b) showing the relative cell numbersof each population in each cell group Since the incubatorcontrol group consisted almost exclusively of cells in subtype2 this number was nearly 100 whereas subtype 1 was closeto 0 The 120583g group had almost as many cells in subtype2 as in subtype 1 with a slight predominance in subtype 2In Figure 2(c) the mean fluorescence intensity of the cellsubtypes in the different cell groups was depicted Whilethe ICAM-1 expression in the incubator control group wasstable in both subtypes (2158 plusmn 2344 RFU versus 2082 plusmn171 RFU) and the 120583g cell group displayed significantly lessexpression of ICAM-1 in subtype 1 compared to subtype 2ICAM-1 expression was significantly reduced in the 120583g groupcompared to the 1 g control and the incubator control groupCells in the 1 g control group exhibited a similar ICAM-1expression distribution as cells from the 120583g group The meanfluorescence intensities between subtype 2 of different groupsdid not change dramatically except a significant differencebetween the 1 g control group and the incubator controlgroup In summary we suppose that ICAM-1 expression wasdownregulated in microglia cells in simulated microgravity



Subtype 1Subtype 1

Subtype 2

Subtype 1

Subtype 2 Subtype 2

0

0

SSC-

A

FSC-A

120583g group 1g control group250K

200K

200K

150K

150K

100K

100K

50K

0

SSC-

A

250K

200K

150K

100K

50K

0

SSC-

A

250K

200K

150K

100K

50K

50K 250K 0

FSC-A200K150K100K50K 250K 0

FSC-A200K150K100K50K 250K

(a)

200

150

100

50Cel

ls in

pop

ulat

ion

()

120583g groupIncubator control group

lowastlowast

lowastlowast

lowastlowast

lowastlowast

1 2

Subtype

1g control group

(b)

5000

4000

3000

2000

1000Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

lowastlowast

lowastlowast

lowastlowastlowast

lowastlowastlowast

lowastlowast


All 1 2

Subtype

(c)

Figure 2 Cytometry analysis of ICAM-1 expression in BV-2 microglial cells in simulated microgravity (2D clinorotation) BV-2 microglialcells were exposed to either clinorotation (120583g) placed in the clinostat but not rotated (1 g control group) or cultured under standard cellculture conditions (incubator control) for 24 h Cells were stained for ICAM-1 surface expression and analyzed by flow cytometry The levelof ICAM-1 surface expression is represented by the mean fluorescent intensity assessed by flow cytometry (a) In forwardsideward scatterdetection mode of flow cytometry two gates were set to separate two subtypes of BV-2 microglial cells that appeared different in size andgranulation (subtypes 1 and 2 in dot plots) (b) Distribution of BV-2 microglial cells in subtypes 1 and 2 after exposure to different gravityconditions (c) Quantification of ICAM-1 expression after exposure to different gravity conditions within subtypes 1 and 2 Data are given asmedian plusmn SE (lowast119875 lt 01 lowastlowast119875 lt 005 lowastlowastlowast119875 lt 001 119899 = 3 according to one-way ANOVA followed by Wilcoxon or unpaired 119905-test)

32 Rapid and Reversible Downregulation of ICAM-1 SurfaceExpression in BV-2 Microglial Cells in Real Microgravity Inthe next step we investigated the cell surface expression ofICAM-1 in real microgravity provided by parabolic flightsin murine BV-2 microglial cells During parabolic flightexperiments cells were activated at the onset of 120583g (or during1 g for in-flight control experiments) by the addition of PMAor TNF-120572 or not activated by the addition of medium onlyAfter a 20 sec period of altered gravity cells were fixed by theaddition of formaldehyde

During the 13th DLR parabolic flight campaign we alsoaddressed the issue that during parabolic flight experi-ments cells are generally subjected to irregular stress bycell preparation and handling and by the in-flight situationitself This combination of interference factors always leadsto a significant degree of damaged or dead cells whichcould affect the experiment results and mask a possiblemicrogravity-related effect even under presence of internal

controls For this reason we developed an automated analysismethod which allows for the specific analysis of alive andmorphologically intact cells at the moment of fixation

Experiments from different parabolas (1 g and 120583g resp)and different flights were analyzed The experiments wereperformed in a sequence of three consecutive 120583g and 1 gphases A quadruple fluorescent staining was performedusing TUNEL (rhodamine) for detection of apoptotic cellsDAPI for the nuclei high content screening (HCS) CellMaskdeep red for the delineation of cells and FITC-labeled antiICAM-1 antibody for identification of cell surface expressionof ICAM-1 Cells were imaged with a widefield microscope(Leica Microsystems Wetzlar Germany) using the uniformrandom sampling module and identified by an iso-surfacecalculation (Imaris Bitplane AG Zurich Switzerland) Thisquadruple staining allowed the exclusion of apoptotic cellsin a highly reliable fashion An example of an apoptotic celland a living cell is depicted in Figure 3(a)Themean intensity


(A) (B) (C) (D) (E) (F)

(a)

Con100

80

60

40

20

0

0 250 500 750 1000

Rela

tive f

requ

ency

MIF of ICAM-1

(b)

lowastlowastlowast

1600

1400

1200

1000

800

600

400

200

0

ICA

M-1

inte

nsity

120583g 1g

(c)

lowastlowastlowast

120583g 1g

700

600

500

400

300

200

100

0

MIF

ICA

M-1

(d)

PMA100

80

60

40

20

0

0 250 500 750 1000

Rela

tive f

requ

ency

MIF of ICAM-1

(e)

lowastlowastlowast

120583g 1g

1000

800

600

400

200

0

ICA

M-1

inte

nsity

(f)

lowastlowastlowast

120583g 1g

400

300

200

100

0M

IF IC

AM

-1

(g)

TNF-120572100

80

60

40

20

0

0 250 500 750 1000

Rela

tive f

requ

ency

MIF of ICAM-1

120583g1g

(h)

120583g 1g TNF-120572

1000

800

600

400

200

0

ICA

M-1

inte

nsity

(i)

120583g 1g TNF-120572

300

200

100

0

MIF

ICA

M-1

(j)

Figure 3 ICAM-1 surface expression reacts rapidly and reversibly to microgravity (a) Microscopy of ICAM-1 TUNEL HCS CellMask andDAPI including surface calculation for HCS In order to identify nuclei cells were stained with DAPI (A) Apoptotic cells were identifiedby TUNEL reaction (B) and HCS CellMask label (C) which can be retained to a higher extend in nonapoptotic cells ICAM-1 intensity isdepicted in (D) A merge of TUNEL DAPI and ICAM-1 (E) shows an apoptotic cell (998819) and a living cell (rarr ) The automated calculation ofan iso-surface is exclusively done for living cells using the HCS CellMask channel as shown in the merge with TUNEL DAPI and ICAM-1(F) (b)ndash(j) BV-2 microglial cells were treated with PMA ((e) (f) and (g)) or TNF-120572 ((h) (i) and (j)) at the onset of microgravity or duringthe 1 g in-flight control phase or left untreated ((b) (c) and (d)) Cells were fixed in flight after 20 sec normogravity (1 g) (-e-) or 20 secmicrogravity (120583g) (-I-) Cells were stained imaged and analyzed as described above The mean intensity of the ICAM-1 signal was binnedintomean intensity fluorescence (MIF) categories and the number of cells (frequency) is plotted against these intensity categories ((b) (e) and(h)) ICAM-1 fluorescence intensity of all analyzed cells ((c) (f) and (i)) is depicted for normogravity (triangles) and microgravity (squares)Mean ICAM-1 fluorescence intensity of all analyzed cells ((d) (g) and (j)) was pooled for normogravity (black bar) and microgravity (openbar) For automated imaging the unified random sampling module was utilized and 63 randomized images of each sample were recordedand at least 500 single cells from 3 independent experiments from 3 different parabolas were analyzed Mean intensity and SEM are shownand studentrsquos 119905-test showed highly significant difference of the fluorescence values of lowastlowastlowast119875 lt 00001 119899 = 3



3000

2000

1000

Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

lowastlowastlowast

lowastlowastlowast

lowastlowastlowast

lowastlowast

1 3 5

Time (d)


(a)


3000

2000

1000Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

lowastlowastlowast

lowast

lowast

lowastlowast

1 3 5

Time (d)

Primary macrophages

(b)

Figure 4 Cytometry analysis of ICAM-1 expression in macrophage-like differentiated U937 cells or primary human macrophages insimulated microgravity (2D clinorotation) Macrophage-like differentiated U937 cells (a) or primary human macrophages (b) were exposedto either clinorotation (120583g) placed in the clinostat but not rotated (1 g control group) or cultured under standard cell culture conditions(incubator control) Cells were stained for ICAM-1 surface expression and analyzed by flow cytometryThe level of ICAM-1 surface expressionis represented by the mean fluorescent intensity assessed by flow cytometry (a) Quantification of ICAM-1 expression in macrophage-likedifferentiated U937 cells after exposure to different gravity conditions for 1 h 3 h or 5 h 119899 = 6 (b) Quantification of ICAM-1 expression inprimary humanmacrophages after exposure to different gravity conditions for 1 h 3 h or 5 h 119899 = 3 Data are given as median plusmn SE (lowast119875 lt 01lowastlowast119875 lt 005 lowastlowastlowast119875 lt 001 according to one-way ANOVA followed by Wilcoxon or unpaired 119905-test)

of the ICAM-1 signal was analyzed in nondamaged andnonapoptotic cells only and binned into intensity categoriesThe relative frequency of these cells was plotted against thefluorescence intensity (Figures 3(b) 3(e) and 3(h))

We found a rapid and reversible downregulation ofICAM-1 on the surface of BV-2 microglial cells after 20 secof microgravity apparent by the frequency of cells expressingICAM-1 in various intensities (Figures 3(b) and 3(c)) andthe mean of ICAM-1 expression intensity (Figure 3(d)) beingonly 70 in microgravity compared to normogravity In thepresence of PMA ICAM-1 expression was upregulated (Fig-ures 3(e) 3(f) and 3(g)) whereas the presence of the proin-flammatory cytokine TNF-120572 abrogated the microgravity-induced ICAM-1 downregulation (Figures 3(h) 3(i) and3(j)) Statistical analysis of all pooled data revealed downregu-lation of ICAM-1 expression in unstimulated microglia uponmicrogravity to be highly significantly different with 119875 lt0001 (Figure 3(d)) Changes of ICAM-1 expression in PMAstimulated cells were as well highly significant (Figure 3(g))whereas TNF-120572 stimulation slightly ameliorated the gravity-dependent changes in ICAM-1 expression (Figure 3(j))Thuswe found a rapid and reversible disappearance of ICAM-1protein from the cell surface in microgravity

33 Increase of ICAM-1 Expression in U937 HumanMacrophage-Like Cells and Human Primary Macrophagesin Simulated Microgravity To corroborate the relevanceof the results obtained with murine BV-2 microglial cellswe investigated a human macrophage-like cell systemTherefore human monocytic U937 cells were differentiated

into macrophage-like cells [23] and humanM2 macrophageswere differentiated from blood mononuclear cells Beforethe experiment differentiated macrophage-like cells weredetached resuspended in fresh medium and filled into1mL standardized serological pipettes for the clinostatClinorotation was performed for 1 d 3 d and 5 d The 1 gcontrol group of differentiated U937 cells was filled into 1mLserological pipettes in the same way as the clinorotation cellgroup but was not rotated Cells were subsequently fixed andstained for cell surface ICAM-1 and apoptosis (TUNEL) toexclude apoptotic cells from the analysis and subjected toflow cytometry We detected a highly significant increaseof ICAM-1 expression in the clinorotated cells (120583g group)compared to the nonrotated cells (1 g control group) after 1 dand 5 d in differentiated U937 cells (Figure 4(a)) and primarymacrophages (Figure 4(b)) However this increase recededafter 3 and 5 days of clinorotation Therefore we concludethat ICAM-1 expression is increased in human macrophagesafter 1 and 5 days of simulated microgravity

34 Increased ICAM-1 Expression in Differentiated U937Cells in Real Microgravity during Parabolic Flight Duringparabolic flight experiments we investigated rapid effectsof real microgravity on nondifferentiated and differenti-ated U937 cells and on primary human M2-differentiatedmacrophages Nondifferentiated and differentiated myelom-onocytic U937 cells were cultured and seeded into Nutrimixbags as described During the parabolicmaneuvers cells wereactivated at the onset of 120583g or during 1 g for in-flight controlexperiments by the addition of PMA with medium in the


case of nondifferentiated U937 cells or not activated by theaddition of medium only in the case of differentiated U937cells and primary macrophages After 20 sec microgravitycells were fixed by the addition of paraformaldehyde A groupof ground control cells was left in Nutrimix bags in thelaboratory incubator and activated and fixed after landing inthe same experimental equipment Experiments from differ-ent parabolas (1 g and 120583g resp) and different flights wereanalyzed A quadruple fluorescent staining was performedusing TUNEL (rhodamine) for detection of apoptotic cellsDAPI for the nuclei high content screening (HCS) CellMaskdeep red for the delineation of cells and FITC-labeled antiICAM-1 antibody for identification of cell surface expressionof ICAM-1

Nondifferentiated U937 Cells Differentiation of U937 mono-cytic cells into macrophage-like cells significantly increasedthe cell surface expression of ICAM-1 (Figure 5(a)) Non-differentiated U937 did not demonstrate differential expres-sion of ICAM-1 in microgravity neither in PMA-stimulatedmyelomonocytic U937 cells nor in non-stimulated cellsany significant alteration of ICAM-1 expression could bedetected in comparison between microgravity and 1 g condi-tions (Figure 5(b)) The only significant difference could beobserved in nonstimulated U937 cells between the groundcontrol group the 120583g group and the 1 g control groupDifferences between 1 g ground and 1 g in-flight controls canbe attributed to the flight itself (eg vibrations handling ofcell containers) and not to an altered gravity

Differentiated U937 Cells In contrast to nondifferentiatedU937 cells macrophage-like U937 cells displayed a highlysignificant gravity-dependent change in ICAM-1 expression(Figure 5(c)) In flight cell surface ICAM-1 was reduced dras-tically compared to the ground control In the microgravitygroup ICAM-1 expression was enhanced This finding isconsistent with our experiments in simulated microgravityWe suppose that ICAM-1 is upregulated in differentiatedmacrophage-like cells in microgravity

PrimaryMacrophages For the analysis of primary humanM2macrophages double fluorescence staining was performedusing TUNEL (rhodamine) for detection of apoptotic cellsand FITC-labeled anti ICAM-1 antibody for identificationof cell surface expression of ICAM-1 Unfortunately the 1 gincubator control was lost during the experiment proce-dures Between the 1 g in-flight control and the microgravitygroup no differences in ICAM-1 surface expression couldbe detected in primary human macrophages However dueto the technical problems and low detected expression levelscompared to primary macrophages in clinostat experiments(see Figure 4(b)) the informative value of these results maybe limited and it is planned to repeat the parabolic flightexperiment with primary human macrophages

35 Increased ICAM-1 Expression in Differentiated U937 Cellsduring Long-Term Microgravity in the SIMBOX ExperimentDuring the SIMBOX (Science in Microgravity Box) mis-sion on Shenzhou-8 we investigatedmicrogravity-associated

long-term alterations inmacrophage-like differentiated U937cells and analyzed the effect of long-term microgravityon the cytoskeleton and immunologically relevant surfacemolecules [23] Human U937 cells were differentiated intoa macrophage-like phenotype and exposed to micrograv-ity or 1 g on a reference centrifuge on orbit for 5 daysThe unmanned Shenzhou-8 spacecraft was launched with aLong March 2F (CZ-2F) rocket from the Jiuquan SatelliteLaunch Center (JSLC) and landed after a 17-day missionAfter on-orbit fixation the samples were analyzed withimmunocytochemical staining and confocalmicroscopy afterlanding Double fluorescent staining was performed usingHCS CellMask deep red for the delineation of cells andFITC-labeled anti ICAM-1 antibody for identification ofcell surface expression of ICAM-1 Cells were analyzed asdescribed above We detected a significant higher expressionof ICAM-1 in long-term microgravity in comparison tothe in-flight 1 g control group (Figure 6) Similar to theparabolic flight experiments incubation of the macrophage-like differentiated U937 cells in the experiment hardwarecaused a significant downregulation of ICAM-1 expressionThus it can be excluded that the microgravity effects onICAM-1 were caused by the experiment system itself

36 No Influence of Altered Gravity on ICAM-1 mRNA RNAsamples were analyzed for their quantity and quality and fur-ther processed for the microarray hybridization on 12 times 135 KRoche NimbleGen arrays Data from 46 single microarrays(19th DLR PFC 8x 120583g 6x HW 8x 1 g and 6x 18 g TEXUS-49 7x120583g 6xHW and 5xBL)were collected normalized andfurther analyzed The data tables were screened for ICAM-1values and mean fluorescence intensities including standarddeviations were calculated for all samples of one conditionICAM-1 shows stable expression for all gravity conditionsduring the 19th DLR PFC and the TEXUS-49 campaignas well as for the HW controls (Figure 7) indicating thatmicrogravity and hypergravity conditions did not have aninfluence on mRNA ICAM-1 level in the range of 20 secondsuntil 6 minutes

37 PathwayAnalysis Reveals an Influence of RealMicrogravityon the Natural Killer Cell Mediated Cytotoxicity of MonocyticU937 Cells Due to the wealth of data microarray analysisprovides we were able to perform a GeneSet enrichmentanalysis to identify any affected pathways or biologicalnetworks in connection with ICAM-1 (see Supplement 1 inthe Supplementary Material available online at httpdxdoiorg1011552015538786) For the experiments performedduring the 19th DLR PFC with monocytic U937 cellswe identified one significantly influenced ICAM-1 relatedpathway during 20 sec of microgravity compared to thein-flight 1 g control namely the natural killer cell mediatedcytotoxicity (enrichment 119875 value 00203328) For theexperiments performed on TEXUS-49 during 6min ofmicrogravity with monocytic U937 cells we found twoweakly altered pathways Specifically the NF-kappa Bsignaling pathway (enrichment 119875 value 00632651) andthe Epstein-Barr virus infection (enrichment 119875 value


8000

6000

4000

2000

Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

Mon U937 Mac U937


lowastlowastlowast

(a)


4000

3000

2000

1000

Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

+minus +minus +minus

PMA (10120583M)

lowastlowastlowast

lowastlowastlowast

U937 cells

(b)


8000

6000

4000

Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

lowastlowastlowast

lowastlowastlowast

lowastlowastlowast


(c)


Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

500

400

300

200

100

Primary macrophages

(d)

Figure 5 ICAM-1 expression in U937 cells macrophage-like differentiated U937 cells and primary human macrophages in different gravityconditions during parabolic flight experiment ICAM-1 expression was assessed by flow cytometry and fluorescent microscopy followingimmunocytochemical staining Cells were cultured under standard cell culture conditions (incubator control) or exposed to different gravityconditions during the 19th DLR parabolic flight campaign U937 cells were fixed either after PMA-activation in microgravity (120583g group) orin 1 g (1 g control group) Differentiated U937 and primary macrophages were fixed after the microgravity phases (120583g group) or after the 1 gphases before and after the120583g phase (1 g control group)The level of ICAM-1 surface expression is represented by themeanfluorescent intensityassessed by flow cytometry (a) ICAM-1 surface expression in myelomonocytic U937 cells (mon U937) and macrophage-like differentiatedU937 cells (max U937) under standard cell culture conditions (b) ICAM-1 surface expression in U937 cells with and without activationby PMA in different gravity conditions (c) ICAM-1 surface expression in macrophage-like differentiated U937 cells in different gravityconditions (d) ICAM-1 surface expression in primary macrophages in different gravity conditions Data are given as median plusmn SE (lowast119875 lt 01lowastlowast119875 lt 005 and lowastlowastlowast119875 lt 001 according to one-way ANOVA followed by Wilcoxon or unpaired 119905-test)

00641782) appeared sensitive to microgravity compared tobaseline

4 Discussion

In our study we investigated the surface expression ofICAM-1 protein and expression of ICAM-1 mRNA in cells

of the monocytemacrophage system in microgravity dur-ing clinostat parabolic flight sounding rocket and orbitalexperiments In murine BV-2 microglial cells we founda downregulation of ICAM-1 expression in clinorotationexperiments and a rapid and reversible downregulation inthe microgravity phase of parabolic flight experiments Incontrast ICAM-1 expression increased in macrophage-like



1000

800

600

400

200

Mea

n flu

ores

cenc

e int

ensit

y (R

FU)

lowastlowastlowast

lowastlowastlowast

(a)

CD54

conOverlay

10120583m

10120583m 10120583m 10120583m

10120583m10120583m

1g 120583g

(b)

Figure 6 ICAM-1 expression in macrophage-like differentiated U937 cells after long-term exposure to microgravity during theSIMBOXShenzhou-8 mission Cells were cultured under standard cell culture conditions (incubator control) or exposed to different gravityconditions during the SIMBOXShenzhou-8 mission Differentiated U937 cells were fixed in microgravity (120583g group) or in 1 g (1 g controlgroup) after 5 days Only CellMask-positive and TUNEL-negative cells were analyzed (a) Each group represents analysis of the meanfluorescence of 200ndash1000 individual cells from one recovered slide Data are expressed as the median of mean single cell fluorescenceintensities with the smallest observation (sample minimum) lower quartile median upper quartile and largest observation (samplemaximum) Statistical analysis was performed with GraphPad Prism 5 Wilcoxon test lowast119875 lt 005 lowastlowast119875 lt 001 and lowastlowastlowast119875 lt 0001 (b) Standardcell culture control (con) 1 g hardware control (1 g) and the microgravity sample (120583g)

10000

8000

6000

4000

2000

0

Fluo

resc

ence

inte

nsity

19th DLR PFC ICAM-1

HW 1g 18 g 120583g

(a)

TEXUS-49 ICAM-110000

8000

6000

4000

2000

0

HW BL 120583g

Fluo

resc

ence

inte

nsity

(b)

Figure 7 Influence of altered gravity during parabolic flight and sounding rocket flight on ICAM-1 mRNA expression levels (a) ICAM-1mRNA expression levels are demonstrated for samples of the 19th DLR parabolic flight campaign after 1 g (light gray) 18 g (dark gray) 120583g(black) and hardware ground controls (HW striped) exposure and (b) for samples of the TEXUS-49 campaign after launch and acceleration(BL dark gray) 120583g (black) and hardware ground controls (HW striped) ICAM-1 fluorescence intensities do not show any significantdifferences for all compared conditions in both experimental setups The number of analyzed arrays 19th DLR PFC 1 g (119899 = 8) 18 g (119899 = 6)120583g (119899 = 8) and HW (119899 = 6) TEXUS-49 HW (119899 = 6) 120583g (119899 = 7) BL (119899 = 5)

differentiated human U937 cells during the micrograv-ity phase of parabolic flights and in long-term micro-gravity provided by a 2D clinostat or during the orbitalSIMBOXShenzhou-8 mission

In nondifferentiated U937 cells no effect of microgravityon ICAM-1 expression could be observed during parabolicflight experiments A summarizing table which presents anoverview about the cell types tested the platforms usedthe experiment durations analysis method the number ofexperiments and the detected effects on ICAM-1 is demon-strated in Table 1 In our study and according to previous

investigations [26] we detected effects of the experimentalhardware whichwere controlled by the appropriate hardwarecontrol experiments

In clinostat experiments subtype of BV-2 microglial cellsappeared in the FACS analysis of all clinostat samples (120583g or1 g controls) but not at all in the incubator controlsThis sub-type consisted of smaller cells and we suppose an ldquoactivatedrdquophenotype of BV-2 microglial cells It is well established thatmicroglial form and function are linked and that cells cancycle reversibly from a simple rounded (activated and amoe-boid) to a complex branched form (ramified and resting) [27]


Table1Re

gulatio

nofICAM-1in

cells

ofthem

onocytemacroph

agesystem

inmicrogravityO

verviewabou

tcelltypesm

icrogravity

platform

sthee

xperim

entd

urationsanalysis

metho

ds

numbero

fexp

erim

ents

andeffectson

ICAM-1

Experim

ental

platform

Celltypes

Experim

entalgroup

sAnalysis

metho

dNum

bero

freplicates

Timeo

f120583gexpo

sal

Regu

latio

nof

ICAM-1

120583g1g

1gin

cubatorc

ontro

l

Clinorotation

BV2

microgliacells

120583ggrou

p1g

controlgroup

FACS

(2000

0events

sample)

119873=3

24h

darr

Prim

ary

macroph

ages

120583ggrou

p1g

controlgroup

FACS

(5000

events

sample)

119873=3

1ndash5d

uarr

U937

macroph

ages

120583ggrou

p1g

controlgroup

FACS

(1000

0events

sample)

119873=6

1ndash5d

uarr

Parabo

licflight

BV2

microgliacells

120583ggrou

p1g

controlgroup

Con

focalm

icroscop

yrand

omsamplingmod

ule

(gt500cellssam

ple)

119873=3(w

o)

uarr

119873=3(w

ithPM

A)

20sec

darr

119873=3(w

ithTN

F-a)

mdash

U937

macroph

ages

120583ggrou

p1g

controlgroup

Groun

dcontrol

FACS

(1000

0events

sample)

119873=9

20sec

uarrdarr

119873=9

119873=6

U937

mon

ocytes

120583ggrou

p1g

controlgroup

Groun

dcontrol

FACS

(1000

0events

sample)

119873=9(w

oPM

A)

20sec

mdashmdash

119873=5(w

ithPM

A)

119873=12

(wo

PMA)

119873=6(w

ithPM

A)

119873=2(w

oPM

A)

119873=1(with

PMA)

120583ggrou

p1g

controlgroup

18gcontrolgroup

Hardw

arec

ontro

l

Microarray

119873=8

20sec

mdashmdash

119873=8

119873=6

119873=6

Prim

ary

macroph

ages

120583ggrou

p1g

controlgroup

FACS

(1000

0events

sample)

119873=9

20sec

mdash119873

=9

SIMBO

XShenzhou

8U937

Macroph

ages

120583ggrou

p1g

controlgroup

Incubatorc

ontro

l

Con

focalm

icroscop

y(300

cellssam

ple)

(650

cellssam

ple)

(100

0cellssam

ple)

119873=1

5duarr

darr

TEXU

S-49

U937

mon

ocytes

120583ggrou

pbaselin

egroup

Hardw

arec

ontro

lMicroarray

119873=7

10min

mdashmdash

119873=5

119873=6


Thus we assume that microgravity activates microglias cellsbut downregulates ICAM-1 expression (Figure 2) Controlexperiments revealed no influence of the serological pipetteincubation system on the ICAM-1 expression compared toldquonormalrdquo cell culture conditions between 1 and 5 d (data notshown)

In this study we also developed a method that alloweda randomized screening of only those cells that were aliveat the fixation time point after the parabola (Figure 3)This is of particular importance because of the damagecaused to all cells subjected to a flight experiment Untilnow a method for analyzing only viable and nondamagedcells obtained from flight experiments was lacking Dueto the very limited number of samples during the exper-iment with BV-2 cells during the 13th DLR PFC FACSanalysis could not be utilized We therefore developed amicroscopy based method to analyze exclusively the livingcell portion Samples were imaged using the uniform ran-dom sampling module of Leica LAS AF software in orderto fulfill all statistically necessary criteria of randomizedsampling Surface calculation of cells negative for TUNELlabel under a certain threshold and positive for HCS Cell-Mask allowed the exclusion of all apoptotic cells The meanICAM-1 intensity value of each analyzed cell was taken intoaccount

Modulation of the expression of surface adhesionmolecules such as ICAM-1 has been reported as the conse-quence of long-term microgravity [28 29] In our study wefound that ICAM-1 surface expression responds to gravitychanges in BV-2 microglial cells within 20 seconds Therapid and reversible changes of ICAM-1 on the cell surfacesuggest a direct gravity-sensitive effect on the membranecompartment or on protein folding whereas transcriptionalor proteolytic processes are rather unlikely as they wouldbe too slow Interestingly ICAM-1 cell surface expressionin microgravity was upregulated in macrophage-like dif-ferentiated human U937 cells (Figures 4 5 and 7) butdownregulated in murine BV-2 microglial cells (Figure 2) Inprimary humanmacrophages no clear conclusion is possiblebecause of the very low fluorescence levels in the analysisof parabolic flight samples (Figure 5) However the clinostatexperiments with primary human macrophages (Figure 4)suggest an upregulation of ICAM-1 in microgravity Thedifferent ICAM-1 regulation between macrophage-like dif-ferentiated human U937 and murine BV-2 microglial cellsin microgravity could be the consequence of the differentspecies (murine and human) or differentmolecular and func-tional features of peripheral macrophages and CNS macro-phages

In our study we detected no effect of microgravity onICAM-1 mRNA expression neither in a parabolic flightexperiment nor during the sounding rocket experiment(Figure 7) In a previous study also no effect of simulatedmicrogravity on ICAM-1 mRNA expression in endothelialcells could be found [15] However performing pathwayanalyses on ICAM-1 related pathways we identified thenatural killer cell mediated cytotoxicity being influencedsignificantly after 20 sec of microgravity After 6min ofmicrogravity this effect appeared to be reversed and we

found the NF-kappa B signaling pathway and the Epstein-Barr virus infection close to significant alteration Thisobservation is in line with findings in astronauts after long-term space missions where latent viruses persisting in adormant state after primary infection were reactivated [3031] Therefore we hypothesize that these two pathways maybe stronger affected over a longer period of microgravity Aswe were not able to find an influence on the natural killercell mediated cytotoxicity after 6min of microgravity wesuppose this is one of the short-term reversible processesthat can recover after an adaptation phase to micrograv-ity

Related to the regulation of surface ICAM-1 expressioninternalization and receptor recycling of ICAM-1 are highlydynamic processes [32 33] and linked to cytoskeletal function[34 35]

Multiple investigators have reported that this complexnetwork of fibers is sensitive to environmental factors suchas microgravity and altered gravitational forces [36ndash38]Several studies demonstrated modifications of the actin andmicrotubule cytoskeleton in real and simulated microgravityin lymphocytes astrocytes neurons glial cells mesenchymalstem cells and thyroid carcinoma cells [36ndash41] Morphologi-cal differences of both themicrotubule and actin componentsof the cytoskeleton have been observed in cells grown inreal and simulatedmicrogravity [39ndash43] During space flightactin reorganization in response to the gravity level andabnormal assembly of actin stress fibers has been reported[44ndash46]

We conclude that disturbed immune function in micro-gravity could be a consequence of ICAM-1 modulation inthe monocytemacrophage system which in turn could havea strong impact on the cellsrsquo interaction with T lympho-cytes and migration An experiment under real microgravityconditions on board of the ISS was conducted by Italianand Swiss investigators to test the hypothesis that lack ofinteraction might be the reason for the loss of activity ofT cells in microgravity [14] The investigation consistedof analyzing the cap formation of the adhesion proteinsLFA-1 on T cells and ICAM-1 on monocytes The datashowed that LFA-1ICAM-1 interactions occur in space butare dependent on activation time they show differences innumber arrangement and fluorescence intensityThus LFA-1 and ICAM-1 adhesion proteins seem to be sensitive to realmicrogravity without being altered in their interaction Lossof functional ICAM-1 in the brain-resident microglial cellsbears the risk of a significant impairment of the CNS immunesystem Indeed reactivation and shed of varicella-zoster virus(VZV) have been reported in astronauts [30 31] a viruswhich becomes latent in the nervous system after primaryinfection but is reactivated frequently in immune suppressedindividuals

In conclusion we found that ICAM-1 can be downreg-ulated rapidly and reversibly in BV-2 microglial cells andupregulated in macrophage-like differentiated U937 cells inresponse to microgravity In both cell types long-term effectsup to several days could be detected Thus ICAM-1 canbe considered as a rapid-reacting and sustained gravity-regulated molecule in mammalian cells


Abbreviations

BL BaselineCC Cell culture controlDLR German Aerospace CenterESA European Space AgencyESRANGE European Space and Sounding Rocket RangeEUE Experiment Unique EquipmentFACS Fluorescence activated cell sortingFCS Fetal calf serumGC Ground controlHW HardwareICAM-1 Intercellular adhesion molecule 1JSLC Jiuquan Satellite Launch CenterLFA-1 Lymphocyte function-associated antigen 1MAC-1 Myelomonocytic leukocyte integrin CD11b120583g MicrogravityMMS Monocyte-macrophage-systemPBMC Peripheral blood mononuclear cellPEEK Polyether ether ketonePFA ParaformaldehydePFC Parabolic flight campaignPITC Payload integration and test centerPMA 12-O-Tetradecanoylphorbol-13-acetateRCCS Rotary Cell Culture SystemsRFI Relative fluorescence intensityRIN RNA integrity numberRPM Random positioning machineSSC Swedish Space CorporationSIMBOX Science in Microgravity BoxTUNEL TdT-mediated dUTP-biotin nick end labeling


The authors declare that they have no competing interests


Oliver Ullrich developed the study idea concept and theoverall study design in addition to planning coordinatingand supervising the study Katrin Paulsen and Oliver Ullrichwrote the paper Svantje Tauber Cora S Thiel and Liliana ELayer contributed to the paper Oliver Ullrich Cora S ThielGesine Bradacs Sonja Krammer Josefine Biskup GabrielaRieder Lisa Mutschler Daniel Hofmanner Isabell Buttronand Hartwin Lier performed the experiments during the13th DLR parabolic flight campaign Liliana E Layer CoraS Thiel Oliver Ullrich Svantje Tauber Swantje HauschildClaudia Philpot Annett Gutewort Eva Hurlimann Jose-fine Biskup and Hartwin Lier performed the experimentsduring the 19th DLR parabolic flight campaign AndreasHuge performed the pathway analysis Svantje Tauber CoraS Thiel and Oliver Ullrich performed the experimentsduring the TEXUS-49 mission and Katrin Paulsen SvantjeTauber Dana M Simmet Oliver Ullrich Eva Hurlimannand Nadine Golz during the SIMBOXShenzhou-8 missionKatrin Paulsen was responsible for the sample analysis fromthe 13th DLR and 19th DLR parabolic flight campaigns andthe SIMBOXShenzhou-8 mission and Cora S Thiel was

responsible for sample analysis from the TEXUS-49 missionNadine Golz and Svantje Tauber contributed to the sampleanalysis Frank Engelmann contributed to and supervised thetechnical procedures during the 13th and 19th DLR parabolicflight campaigns

Acknowledgments

The authors gratefully acknowledge financial support by theGerman Aerospace Center DLR (Grant nos 50WB0912 and50WB1219) They also gratefully acknowledge the support of(in alphabetic order) Markus Braun Miriam Christen Gio-vanni Colacicco Ulrike Friedrich Andre Hilliger AndreasHuge Schirin Ibrahim Otfried Joop Andre Melik ShirinMilani Brice Moutett Marianne Ott Irina Rau Frank RuhliChen Sang Burkhard Schmitz Andreas Schutte JohannaStahn Marc Studer Susanne Wolf and Fengyuan ZhuangThey would like to thank the Swiss Air Force for theoutstanding support of their study by providing daily airtransports of in-flight cell samples from Zurich to Bordeauxand back during the 13th DLR parabolic flight campaign Itsreliable and rapid transport system guaranteed the recoveryof flown samples in an outstanding quality for analysis

References

[1] G Sonnenfeld ldquoThe immune system in space andmicrogravityrdquoMedicine and Science in Sports and Exercise vol 34 no 12 pp2021ndash2027 2002



[4] O Ullrich and C S Thiel ldquoGravitational force triggeredstress in cells of the immune systemrdquo in Stress Challengesand Immunity in Space From Mechanisms to Monitoring andPreventive Strategies A Chouker Ed chapter 14 pp 187ndash202Springer Berlin Germany 2012

[5] I V Konstantinova E N Antropova V I Legenrsquokov and VD Zazhireı ldquoReactivity of lymphoid blood cells in the crew ofldquoSoiuz-6rdquo ldquoSoiuz-7rdquo and ldquoSoiuz-8rdquo spacecraft before and afterflightrdquo Kosmicheskaia Biologiia i Aviakosmicheskaia Meditsinavol 7 no 6 pp 35ndash40 1973

[6] S Hauschild S Tauber B Lauber C SThiel L E Layer andOUllrich ldquoT cell regulation in microgravitymdashthe current knowl-edge from in vitro experiments conducted in space parabolicflights and ground-based facilitiesrdquo Acta Astronautica vol 104no 1 pp 365ndash377 2014

[7] B Crucian R Stowe H Quiriarte D Pierson and C SamsldquoMonocyte phenotype and cytokine production profiles aredysregulated by short-duration spaceflightrdquo Aviation Spaceand Environmental Medicine vol 82 no 9 pp 857ndash862 2011

[8] M Hughes-Fulford T Chang and C F Li ldquoEffect of gravity onmonocyte differentiationrdquo in Proceedings of the 10th ESA LifeSciences Symposium29th Annual ISGP Meeting24th AnnualASGSB MeetingELGRA Symposium ldquoLife in Space for Life onEarthrdquo Angers France June 2008


[9] B Bechler A Cogoli M Cogoli-Creuter O Muller E Hun-zinger and S B Criswell ldquoActivation of microcarrier-attachedlymphocytes in microgravityrdquo Biotechnology and Bioengineer-ing vol 40 no 8 pp 991ndash996 1992

[10] J-L Wautier H Setiadi D Vilette D Weill and M-P WautierldquoLeukocyte adhesion to endothelial cellsrdquo Biorheology vol 27no 3-4 pp 425ndash432 1990

[11] W A Muller ldquoMechanisms of leukocyte transendothelialmigrationrdquoAnnual Review of PathologyMechanisms of Diseasevol 6 pp 323ndash344 2011

[12] A Grakoui S K Bromley C Sumen et al ldquoThe immunologicalsynapse a molecular machine controlling T cell activationrdquoScience vol 285 no 5425 pp 221ndash227 1999

[13] T Nagel N Resnick W J Atkinson C F Dewey Jr and M AGimbrone Jr ldquoShear stress selectively upregulates intercellularadhesion molecule-1 expression in cultured human vascularendothelial cellsrdquo The Journal of Clinical Investigation vol 94no 2 pp 885ndash889 1994

[14] M A Meloni G Galleri G Pani A Saba P Pippia and MCogoli-Greuter ldquoEffects of real microgravity aboard interna-tional space station onmonocytes motility and interaction withT-lymphocytesrdquo in Proceedings of the 10th ESA Life SciencesSymposium29th Annual ISGP Meeting24th Annual ASGSBMeetingELGRA Symposium ldquoLife in Space for Life on EarthrdquoAngers France 2008


[16] K P Murphy Janewayrsquos Immunology Garland Science Taylor ampFrancis Group LLC New York NY USA 8th edition 2012

[17] B Alberts A Johnson J Lewis M Raff K Roberts and PWalter Molecular Biology of the Cell Garland Science Tayloramp Francis Group LLC New York NY USA 5th edition 2008

[18] C S Thiel K Paulsen G Bradacs et al ldquoRapid alterationsof cell cycle control proteins in human T lymphocytes inmicrogravityrdquo Cell Communication and Signaling vol 10 no 1article 1 2012

[19] K Paulsen C Thiel J Timm et al ldquoMicrogravity-inducedalterations in signal transduction in cells of the immunesystemrdquo Acta Astronautica vol 67 no 9-10 pp 1116ndash1125 2010

[20] E Blasi R Barluzzi V Bocchini R Mazzolla and F BistonildquoImmortalization of murine microglial cells by a v-raf v-myccarrying retrovirusrdquo Journal of Neuroimmunology vol 27 no2-3 pp 229ndash237 1990

[21] S J Gerathewohl Ed Zero-G Devices and WeightlessnessSimulators Report for the Armed Forces-NAS-NRC Committeeon Bioastronautics Panel on Acceleration National ResearchCouncil Publication 781 National Academy of Sciences Wash-ington DC USA 1961

[22] U Kuebler ldquoSIMBOX Plungerrdquo in Astrium Space Biology Prod-uct Catalog chapter 718 pp 153ndash157 AstriumSpaceTransporta-tion Department of New Business Friedrichshafen Germany2012

[23] K Paulsen S Tauber N Goelz et al ldquoSevere disruption of thecytoskeleton and immunologically relevant surface moleculesin a human macrophageal cell line in microgravitymdashresultsof an in vitro experiment on board of the Shenzhou-8 spacemissionrdquo Acta Astronautica vol 94 no 1 pp 277ndash292 2014

[24] M Kanehisa S Goto Y Sato M Kawashima M Furumichiand M Tanabe ldquoData information knowledge and principleback to metabolism in KEGGrdquo Nucleic Acids Research vol 42no 1 pp D199ndashD205 2014

[25] M Kanehisa and S Goto ldquoKEGG kyoto encyclopedia of genesand genomesrdquo Nucleic Acids Research vol 28 no 1 pp 27ndash302000

[26] S Tauber S Hauschild C Crescio et al ldquoSignal transductionin primary human T lymphocytes in altered gravitymdashresults ofthe MASER-12 suborbital space flight missionrdquo Cell Communi-cation and Signaling vol 11 no 1 article 32 2013

[27] A Karperien H Ahammer and H F Jelinek ldquoQuantitatingthe subtleties of microglial morphology with fractal analysisrdquoFrontiers in Cellular Neuroscience vol 7 no 3 2013

[28] L Buravkova Y Romanov M Rykova O Grigorieva andN Merzlikina ldquoCell-to-cell interactions in changed gravityground-based and flight experimentsrdquo Acta Astronautica vol57 no 2-8 pp 67ndash74 2005

[29] Y A Romanov L B Buravkova M P Rikova E N AntropovaN N Savchenko and N V Kabaeva ldquoExpression of celladhesion molecules and lymphocyte-endothelium interactionunder simulated hypogravity in vitrordquo Journal of GravitationalPhysiology vol 8 no 1 pp 5ndash8 2001

[30] R J Cohrs S K Mehta D S Schmid D H Gilden and DL Pierson ldquoAsymptomatic reactivation and shed of infectiousvaricella zoster virus in astronautsrdquo Journal of Medical Virologyvol 80 no 6 pp 1116ndash1122 2008

[31] S K Mehta R J Cohrs B Forghani G Zerbe D H Gildenand D L Pierson ldquoStress-induced subclinical reactivation ofvaricella zoster virus in astronautsrdquo Journal of Medical Virologyvol 72 no 1 pp 174ndash179 2004

[32] S Muro R Wiewrodt A Thomas et al ldquoA novel endocyticpathway induced by clustering endothelial ICAM-1 or PECAM-1rdquo Journal of Cell Science vol 116 no 8 pp 1599ndash1609 2003

[33] S Muro C Gajewski M Koval and V RMuzykantov ldquoICAM-1 recycling in endothelial cells A novel pathway for sustainedintracellular delivery and prolonged effects of drugsrdquoBlood vol105 no 2 pp 650ndash658 2005

[34] O Carpen P Pallai D E Staunton and T A SpringerldquoAssociation of intercellular adhesion molecule-1 (ICAM-1)with actin-containing cytoskeleton and 120572-actininrdquo Journal ofCell Biology vol 118 no 5 pp 1223ndash1234 1992

[35] E VandenBerg M D Reid J D Edwards and H W DavisldquoThe role of the cytoskeleton in cellular adhesion moleculeexpression in tumor necrosis factor-stimulated endothelialcellsrdquo Journal of Cellular Biochemistry vol 91 no 5 pp 926ndash937 2004

[36] H Schatten M L Lewis and A Chakrabarti ldquoSpaceflight andclinorotation cause cytoskeleton andmitochondria changes andincreases in apoptosis in cultured cellsrdquo Acta Astronautica vol49 no 3-10 pp 399ndash418 2001




[40] B M Uva F Strollo F Ricci M Pastorino J I Mason andM A Masini ldquoMorpho-functional alterations in testicular andnervous cells submitted to modelled microgravityrdquo Journal ofEndocrinological Investigation vol 28 no 11 pp 84ndash91 2005



[42] V E Meyers M Zayzafoon J T Douglas and J M McDonaldldquoRhoA and cytoskeletal disruption mediate reduced osteoblas-togenesis and enhanced adipogenesis of human mesenchymalstem cells in modeled microgravityrdquo Journal of Bone andMineral Research vol 20 no 10 pp 1858ndash1866 2005


[44] M Hughes-Fulford ldquoReview of the biological effects of weight-lessness on the human endocrine systemrdquo Receptor vol 3 no3 pp 145ndash154 1993

[45] R Gruener R Roberts and R Reitstetter ldquoReduced receptoraggregation and altered cytoskeleton in cultured myocytes afterspace-flightrdquo Uchu Seibutsu Kagaku vol 8 no 2 pp 79ndash931994


Research ArticleGenes Required for Survival in MicrogravityRevealed by Genome-Wide Yeast Deletion CollectionsCultured during Spaceflight

Corey Nislow1 Anna Y Lee2 Patricia L Allen3 Guri Giaever1

Andrew Smith2 Marinella Gebbia2 Louis S Stodieck4 Jeffrey S Hammond5

Holly H Birdsall67 and Timothy G Hammond3689

1Faculty of Pharmaceutical Sciences The University of British Columbia Vancouver BC Canada V6T 1Z32Donnelly CCBR University of Toronto Toronto ON Canada M5S 3E13Durham VAMedical Center Research amp Development Service Durham NC 27705 USA4Bioserve Space Technologies University of Colorado Boulder CO 80309 USA5The Institute for Medical Research Durham NC 27705 USA6Department of Veterans Affairs Office of Research and Development Washington DC 20420 USA7Departments of Otorhinolaryngology Immunology and Psychiatry Baylor College of Medicine Houston TX 77030 USA8Nephrology Division Department of Internal Medicine Duke University School of Medicine Durham NC 27705 USA9Nephrology Section Department of Internal Medicine George Washington University School of MedicineWashington DC 20052 USA

Correspondence should be addressed to Timothy G Hammond grumpy70115yahoocom



Copyright copy 2015 Corey Nislow et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Spaceflight is a unique environment with profound effects on biological systems including tissue redistribution andmusculoskeletalstresses However the more subtle biological effects of spaceflight on cells and organisms are difficult to measure in a systematicunbiasedmanner Here we test the utility of the molecularly barcoded yeast deletion collection to provide a quantitative assessmentof the effects of microgravity on a model organism We developed robust hardware to screen in parallel the complete collectionof sim4800 homozygous and sim5900 heterozygous (including sim1100 single-copy deletions of essential genes) yeast deletion strainseach carrying unique DNA that acts as strain identifiers We compared strain fitness for the homozygous and heterozygous yeastdeletion collections grown in spaceflight and ground as well as plus and minus hyperosmolar sodium chloride providing a secondadditive stressor The genome-wide sensitivity profiles obtained from these treatments were then queried for their similarity to acompendium of drugs whose effects on the yeast collection have been previously reported We found that the effects of spaceflighthave high concordance with the effects of DNA-damaging agents and changes in redox state suggesting mechanisms by whichspaceflight may negatively affect cell fitness

1 Introduction

Physical effects of microgravity during spaceflight can oftenbe described by equations that allow their quantification[1 2] For example microgravity has well-defined effects onsedimentation in association with reduced terminal velocityand shear in suspension culture [3] and reduced gravity-dependent convection of gases [4] Biological effects of

spaceflight on cells and organisms on the other hand aremuch harder to define [1 2] For example spaceflight alsoentails radiation exposure which has been studied in diversesystems but whose effects are not fully understood [5] Whatis needed is a robust unbiased quantifiable system that isrelevant for translation to ground-based applications andthat is able to clearly distinguish spaceflight effects It isour premise that yeast deletion collections are ideally suited



for this type of analysis as yeast can be precisely controlledgenetically and readily grown under spaceflight conditionsBiological responses of yeast strains during spaceflight canbe quantified and compared to well-established databasesof ground-based stressors and the comparisons can revealfeatures that are unique to microgravity as well as featuresthat are sharedwith ground-based perturbationsWhile yeastcannot completely reflect the complexities of mammaliancells organized into tissues the high degree of homologyshared with human (sim70 of all essential yeast genes havea significant human homolog) provides hypotheses for themechanism of many responses of interest [6 7]

Previous studies have attempted to identify isolate andoffset the various physical factors changing during spaceflightto demonstrate their effects in an iterative fashion [8ndash10]Earlier studies on the effects of space radiation on yeast failedto find any change in point mutation rates DNA replicationandor repair heritable damage or colony morphology [58 11 12] However those studies were limited by assaysensitivity Here we applied the yeast deletion collection asa biological reporter to understand the metabolic pathwaysaffecting survival during culture in spaceflight In this fash-ion we are able to make genome-wide comparisons and testfor concordance against an extensive library of more than3200 physical and pharmacological stressors [13]

Yeast is the first and to date only organism for whicha complete genome-wide knockdown collection is availableThis collection is comprised of a genome-wide set of strainswhere each strain carries a precise deletion of a single gene[14] Assembled over a four-year period by a consortium of35 laboratories this collection has been used by hundredsof laboratories to test thousands of different environmentalstressors to define the genes required for survival in thoseconditions (see [15] for review) The molecular barcodespresent in each strain allow the yeast deletion collection tobe grown as a pool in the presence or absence of the stressorof interest after which the relative abundance of each strainis subsequently quantified [16] Strains carrying a deletion ofa gene required for survival in the presence of the stressorgrow more slowly and thus exhibit a fitness defect reflectedby their reduced abundance at the end of the culture periodIn this manner all genes required for growth can be readilyidentified in a single experiment revealing the genes andassociated pathways affected by the stressor

To identify the metabolic and genomic pathways affectedby spaceflight the homozygous and heterozygous yeastdeletion collections were grown in spaceflight and groundcontrol conditions with and without hyperosmolar sodiumchloride providing a second stressor In spaceflight alonethe homozygous deletion collection revealed the importanceof processes linked to mitochondria while the heterozygouscollection highlighted genes involved in regulating transla-tion and ribosomal RNA transport Both homozygous andheterozygous collections highlighted DNA repair With theaddition of NaCl the homozygous collection also revealedthe importance of RNA-related processes including ribosomeassembly and biogenesis andmRNA processing and decay aswell as modification of tRNAs Moreover the NaCl additionhighlighted replication processes more clearly (compared to

the homozygous collection without NaCl) suggesting thatspaceflight has measurable effects on these core and evolu-tionarily conserved processes With the heterozygous collec-tion the addition of NaCl led to the identification of a nuclearpore organization gene potentially providing additionalinsight into how RNA transport is affected by spaceflightTaken together the deletion collections identified severalbiological processes associatedwith spaceflight and the addi-tional hyperosmolar stress emphasized the importance ofrelated processes

In a follow-up analysis we queried the effects of space-flight against a database of drug effects on yeast to search forthose that are most concordant thereby suggesting similarmechanisms of perturbation Not only do the effects ofspaceflight have relatively high concordance with the effectsof DNA-damaging agents but also there is tight agreementamongst multiple therapeutic agents in this drug classproviding additional support for these findings


21 Overall Design The Opticell Processing Moduledescribed below was used to perform a series of sim21 genera-tion pooled growth experiments on two yeast deletioncollections (i) sim4800 homozygous strains and (ii) sim5900heterozygous strains (including sim1100 single-copy deletionsof essential genes) each carrying unique DNA barcodes thatact as strain identifiers Experiments were performed in bothrich media and rich media supplemented with 05M NaClto assess the additional effect of osmotic stress on survivalThe samples flew sortie on space shuttle mission STS-135to the International Space Station (ISS) Parallel controlexperiments were performed in static 1G terrestrial controlsin the Orbital Environmental Simulator at Kennedy SpaceCenter to match temperature humidity air compositionand volatile organic compounds Ground controls wereconducted in a 24-hour asynchronous fashion to allowmatching of the experimental timelines on ISS as relayedthrough air-to-ground communication by the flight crew Atthe end of the growth period the fitness of each strain in eachexperimental pool was assessed as described [17] Brieflygenomic DNA was extracted from each sample the barcodesin each pool were amplified by PCR and the abundance ofeach barcode was quantified by next generation sequencingA barcode count reflects the abundance of the correspondingstrain at the end of the experiment that is a quantification ofthe relative requirement of the deleted gene for growth in thetested condition In total the experiment results in a countfor each gene resulting in a gene list rank ordered by theirimportance for growth in the tested condition

22 Yeast Deletion Pool Construction The yeast deletion col-lections were stored as individual strains in YPD containing7 DMSO at minus80∘C in 96-well plates The plates werethawed mixed and robotically pinned onto YPD agar platesas an array of 384 strains After two days of growth at 30∘Ccolonies were consolidated (four plates of 384 to one plateof 1536 colonies) and robotically pinned in triplicate Cellswere grown in 30∘C for 2-3 days until colonies formed Slow


growing strainswere grown separately for 2-3 additional daysAll plates were then flooded with 5ndash7mL of media scrapedand pooled in YPD + 7 DMSO to a final concentration ofOD600

= 084 and frozen atminus80∘Cuntil use as described [17]

23 Construction of Opticell Culture System and SpaceflightExperiment In this study we designed the Opticell Pro-cessing Module or OPM (Figure 1) that was capable ofmaintaining the yeast deletion collection as a pool grownin liquid culture for at least 20 generations in microgravityThe hardware comprised a liquid-sealed system of growthchambers (Opticells) that allowed for gas exchange acrosspolystyrene membranes Each OPM consisted of three NuncOpticells held together with a common manifold and valvesystem that is autoclaved and attached with watertight O-ring seals A 3mL syringe connected to the manifold witha Luer fitting is used to transfer liquid between chambersandmixwithout breaking sterility andwithminimal operatorinterventionThe valve on the manifold has four settings thatconnect the syringe to the following port locations 1 Offposition 2 Opticell A 3 Opticell B or 4 Opticell C TheOPM allows propagation of each deletion collection for acombined sim21 generations of growth when three chambersare used and the inoculum and transfer volumes are 05mL

To perform a growth assay in the OPM each of thethree chambers was prefilled with 7mL of sterile growthmedia Deletion collection aliquots were preloaded intoeach syringe and shipped to Kennedy Space Center frozenin media containing 7 DMSO (vv) as a cryoprotectantDuring final integration at Kennedy Space Center the OPMswere prechilled to 4∘C Each deletion collection aliquotwas thawed attached to an OPM manifold injected andmixed into chamber A Cultures were maintained at 4∘C andflown to the International Space Station (ISS) The growthexperiment was initiated on orbit by warming the OPMs to30∘C After 16ndash24 h at 30∘C a 05mL sample was removedfrom chamber A using the same syringe and inoculated intochamber B The process was repeated 16ndash24 hours later toinoculate 05mL of sample from chamber B into chamberC After an additional 16ndash24 hours the OPMs were cooledback down to 4∘C to greatly reduce any further growthand preserve the samples for return to Earth and postflightanalysis Exponential yeast growth leads to early depletionof growth media nutrients and significant retardation offurther growth well before 16ndash24 hours Growth is limitedby media volume and strain distribution within the yeastdeletion library reaches a steady state within that Opticell

24 Next Generation Sequencing The flight samples thatreturned from the ISS were handled in parallel with theground control set The OPMwas disassembled into its threeOpticells and the entire contents were transferred to a storagetube using a blunt needle connected to a 20mL syringe OnemL of each sample (at a final OD

600of 10-20) was pro-

cessed to extract genomic DNA Purified deletion pool DNAwas amplified in two separate PCR reactions as described[17] and the amplicons purified prior to sequencing on anIllumina HiSeq2000 Each purified amplicon library wassequenced to a minimum depth of 500 countsstrainsample

Syringe with yeast

3-way stopcock

Opticell A

Opticell B

Opticell C

Figure 1 The Opticell Processing Module (OPM) designed forpropagation of each deletion collection for sim21 generations ofgrowth The OPM comprises three commercially available opticallyclear chambers (Opticells Nunc) that are joined by a manifold andscaffold that can be autoclaved and assembled rapidlyThemanifoldcontains a multiway valve unit which mates to each Opticell or toan the off position using O-ring seals The opposite side of the valvecontains a Luer fitting into which a standard 3cc syringe is attachedTo perform a growth assay in theOPM each of the three chambers isfilled with 7mL of sterile growth media Deletion pools are loadedinto the inoculation syringe and then injected into Chamber A ofthe OPM precooled to 4∘C Growth is initiated by warming theunit to 30∘C After 16ndash24 h 05mL is removed fromChamber A andinjected and mixed into Chamber B using the same syringe This isrepeated to continue multigenerational growth in Chamber C

as described [18] Duplicate experiments were performed forall conditions to ensure that at least one complete time coursewas collected for each pool (heterozygote and homozygote)and each condition Due to failures in sample processingseveral time points were not recovered or did not meetour in-house quality metrics (eg if sequence countsstrainwere below threshold values) Accordingly we focused onevaluating each experimental condition using singleton dataas described in Table 1

25 Data Analysis All computational analyses were per-formed using 119877 [19] unless otherwise indicated

251 Normalization of Sequence Counts Sequence countsfor each strain in each experiment were quantified and nor-malized according to [18] Briefly each 20-mer barcode wasamplified with primers comprised of the common barcodeprimers in addition to the sequences required for clusterformation on the Illumina flow cell Formultiplexed Illuminasequencing 5-mer tag sequences were incorporated into eachprimer between the Illumina and barcode primer sequencesThis multiplexing tag allowed postsequencing assignment ofeach amplicon to a particular experiment Results for the7-generation time point of the heterozygous deletion poolgrown in spaceflight without NaCl did not pass our quality


Table 1 Experimental samples collected and available for analysis

Condition Zygosity GenerationsGround Homozygous 7 14 21Ground + 05M NaCl Homozygous 7 14 21Flight Homozygous 7 14 21Flight + 05M NaCl Homozygous 7 14 21Ground Heterozygous 14 21lowast

Ground + 05M NaCl Heterozygous 7 14 21Flight Heterozygous 14 21lowast

Flight + 05M NaCl Heterozygous 7 14 21lowast7-generation samples from the indicated condition were not available foranalysis due to failures in sample processing or failure to meet in-housequality metrics as described in Methods

control and consequently that time point was omitted fromanalyses of the heterozygous pool without NaCl All othercounts were mean-normalized between experiments suchthat each experiment had the same mean count We addedten pseudocounts to all sequence tag tallies (and thus allsubsequent gene tallies) to prevent division by zero duringdata analyses (see Table S1 and Table S2 for mean normalizedcounts in Supplementary Materials available online at httpdxdoiorg1011552014976458)

252 Barcode Selection for Each Strain For each strain weused signal from only the upstream or the downstreambarcode (relative to the deletion site) First we assumed thatbarcode counts close to zero represent backgroundnoise (egpossibly due to incorrect mapping of reads to barcodes) Wethus selected a background threshold (bgThreshold = 100see Parameter Selection) assuming that counts below it donot accurately reflect strain abundance Then for each timecourse we filtered out barcodes where the average (normal-ized) count for the first 14 generations (the earliest time pointwith usable data in all experiments) was below bgThresholdThis filtering removed all barcodes for sim650ndash3200 strains(depending on the time course) and these strains wereomitted from subsequent analyses see Supplementary TablesS3 (homozygous strains) and S4 (heterozygous strains)

For time courses that only had two time points (andthus an inefficient number to compute fits) strains thatstill had two barcodes after filtering were represented bytheir upstream barcodes due to their overall better behaviorobserved in a previous study [20] For other time courseslinear fits (with and without the time logged) were computedfor each remaining barcode We defined the best fit as the fitwith the lowest residual sum of squares (RSS) and used the119865-test to compute a 119875 value estimating the significance withwhich the fitted model is better than the null model (of a flatline at the average count value)TheBenjamini andHochbergmethod was used to correct the 119875 for multiple comparisonsand generate FDR values [21] Strains with two remainingbarcodes were represented by the barcode with the higher 1198772(ameasure of the amount of variation in the data explained bythe fitted model) because they manifest less noise and betterfit the data

253 Parameter Selection The selected normalizationmethod (tested mean and quartile normalization) andbgThreshold (tested 50 100 and 150) is the combination thatresulted in the most significant enrichment of slow growingstrains identified in the heterozygous deletion pool sampledevery two generations for 20 generations (data not shown)with slow growers identified in a previous study [22] Brieflywe defined slow growers as those exhibiting sizable decreasesin abundance over time The significance of the decrease wasestimated with FDR values (see Barcode Selection for EachStrain) and the magnitude was estimated with ΔAUC =(⟨area under the growth curve⟩ minus ⟨area under the flat growthcurve⟩)(119905max minus 1199050) where the flat growth curve is fixed at the1199050abundance level and the area under a curve is estimated

using the trapezoid method Also if at some time point theabundance of a strain is less than or equal to bgThreshold andremains at negligible levels for the rest of the time course weidentified the strain as slow growing

254 Identification of Significant Fitness Defects in TimePoint Comparisons To identify strains that exhibited sig-nificant fitness defects at a later time point (14 or 21 gen-erations) compared to the first time point (7 generations)normalized counts less than bgThreshold were first forcedto equal bgThreshold Then for each strain we comput-ed log

2ratio = log

2(abundance

7Gabundance14G21G) whereabundanceyG is the count of the strain at 119910 generationsFor a given time point robust 119885 scores were computedfrom the set of log

2ratios for example 119885

119894= (log

2ratio119894minus

⟨log2ratio median⟩)⟨log

2ratio MAD⟩ for strain 119894 Each 119885

119894

was then used to obtain 119875119894from the standard normal

distribution and we assume that strains with low 119875 valuesare outliers in the distribution of log

2ratios Moreover strains

with counts above bgThreshold at the first time point andcounts equivalent to bgThreshold at the later time point ofinterest are defined as having dropped out Taken togetherwe define strains with significant fitness defects at a specifictime point as strains with log

2ratio ge 1 and 119875 le 0001 andor

strains that dropped out (Table S5 Table S6)

255 Spaceflight versus Ground Comparisons For compar-isons involving specific time points we identified the set ofstrains that exhibited significant fitness defects (relative tothe first time point) in the flight condition but not in theground conditionThis set is then further restricted to the setof strains with useable data in both conditions

256 GeneOntology (GO) Enrichment Analysis Weobtainedgene ontology (GO) annotations of yeast genes from theSaccharomyces Genome Database (downloaded on May 262012) GO biological processes that were too specific (con-taining less than five genes) or too general (containing greaterthan 300 genes) were excluded from the analysis

Given a query set of genes (eg genes deleted from a setof (flight-ground) strains) we used the hypergeometric test toobtain a119875 value estimating the significancewithwhich the setis enrichedwith genes annotated to a given biological processrelative to a gene universe defined as the set of genes withusable data for both flight and ground conditions Due to a


limited number of significantly enriched processes followingcorrection for multiple comparisons (FDR le 01) here wereport significantly enriched processes prior to the correction(119875 le 001)

We visualized GO enrichment results with enrichmentmaps shown in Figures 2 and 3 that were generated using anapproach similar to the Enrichment Map Cytoscape Pluginv11 [23 24] In contrast to the plugin the nodes in eachmap were clustered with MCL (inflation = 2) using theoverlap coefficient computed by the plugin as the similaritymetric (coefficients less than 05 were set to zero) Nodes inthe same cluster were assigned the same node color and acluster label was determined based on common themes inthe processes within the cluster Moreover the size of a nodewas made to be proportional to the significance with whichthe corresponding process is enriched [minuslog

10(119875)] Each bar

plot summarizes the genes that contribute the most to theenrichment of processes with the same node color as the plotborder Specifically a plot shows the flight-associated genesthat are annotated to the largest number of relevant processes(if more than 10 genes only the top 10 are shown) For eachgene the bar length is proportional to a fitness defectmeasure(ie log

2ratio)

The enrichment maps also combine two sets of enrich-ment results with the processes enriched in one set shownwith circle nodes the processes enriched in the second setshown with square nodes and the processes enriched in bothsets shown with diamond nodes

257 Similarity between Flight-Associated Genes and Com-pound-Associated Genes We previously treated pools ofyeast deletion strains with sim3200 compounds separately[13] Each compound was subsequently associated with aset of genes deleted from strains that exhibited significantfitness defects induced by the compound Like sets of flight-associated genes in this study sets of compound-associatedgenes were assessed for enrichment of genes annotated tospecific biological processes (as described above) resultingin an ldquoenrichment profilerdquo for each condition of interest Ineach profile each process is associated with a 119875 measuringthe significance of enrichment Similarity between a pairof enrichment profiles was computed by concordance ofminuslog10(119875) across all processes common to both profiles

where concordance is like Pearson correlation except thatscale is not ignored Compounds with enrichment profilesthat are most similar to a given flight enrichment profile mayinduce cellular responses that aremost similar to the responseinduced by flight


Because we cannot distinguish the individual parameters thatinclude flight lack of gravity and increased radiation for thepurposes of this paper these are referred to collectively asldquospaceflightrdquo throughout the text To measure the effects ofspaceflight on the rate of yeast growth in the Opticell weinoculated 05mL of a yeast deletion pool into 7mL of YPDresulting in a starting OD

600of sim006mL and incubated at

30∘C Following growth forsim24 hr (sim7 generations) 05mLof

the saturated culturewas inoculated into the second chamberThis process was repeated for the final growth phase in thethird chamber Population doubling time was sim100min inmicrogravity compared tosim90min in ground-based controlsEach sample was grown for seven generationsOpticell for atotal of 21 generations (Table 1) Doubling times were back-calculated using the OD

600of the samples collected at each

time pointThe morphology of Opticell-grown yeast in spaceflight

was indistinguishable from static controls when observedby light microscopy for example budding pattern overallshape and size were not detectably different in the twoconditions On scanning electron microscopy there werebudding polarity and ruffling changes in every field but therewere no consistent differences (data not shown)

We assessed the yeast deletion collection samples forpatterns of strain sensitivity in the followingmanner barcodecounts for each strain in each sample were measured andnormalized as described inMethodsThe counts were used torank each strain in each sample in order of their importancefor growth Four different samples were available from bothspaceflight and ground cultures (1) homozygous deletion col-lection in YPD (2) homozygous deletion collection in YPDplus 05MNaCl (3) heterozygous deletion collection in YPDand (4) heterozygous deletion collection in YPD plus 05MNaCl Each culturewas sampled at three different time points7 generations 14 generations and 21 generations and shownin Table 1 Samples from ground controls were compared tothe corresponding samples grown inmicrogravity on the ISS

We analyzed changes in strain abundance by comparingeach time point to a later time point Using this approachallowed us to capture those strains that became depleted inany seven-generation interval Strains with sizable decreasesin measured abundance or with abundances that drop tobackground levels (and remain there) were identified asexhibiting fitness defects (FDs) Moreover strains with flight-specific FDs were identified by subtracting the strains withFDs in the ground condition

For the purposes of our gene ontology (GO httpamigogeneontologyorg) enrichment analysis we considered thehomozygous and heterozygous data separately Based on awealth of published data [14 15] the homozygous nonessen-tial deletion collection tends to reveal a similar set of genesinvolved in pathways required for resistance tosurvival inmultiple environmental conditions whereas the heterozy-gous collection of all strains tends to be more specificidentifying essential proteins uniquely required for growth ina specific condition [13]

For the homozygous deletion collection strains thatwere depleted from the pool specifically in spaceflight con-ditions are significantly enriched for genes in biologicalprocesses related to different aspects of RNA metabolismand catabolism including ribosome biogenesis regulationof ribosomal protein transcription cytoplasmic RNA trans-lation rRNA processing tRNA modification and mRNAdecay (Table 2 Figure 2 and Table S5) We also found thatprocesses related to DNA integrity were required for survivalin spaceflight In particular the linked processes of DNArepair and DNA recombination and replication as well as


00 04 08 12

GIR2RPL27ARPS12RPL20ARPL21A

FD score

xx

xx

x

14G

00 10 20

PEX19

MON1AFG3SAM37VPS9MMM1VPS3

FD score

xxx

xx

x

21G00 10 20

TRM7SIT4NCS2ELP6TGS1ELP2URM1

FD score

xxx

21G

00 10 20

MGM1MMM1RPO41MRPL8GGC1

FD score

x

x

21G

00 10 20

RPS11ARPS17ARPS11BSSF1RPS27B

FD score

xx

xx

21G

00 10 20

SRB2HMO1

FD score

x 21G

00 10 20

BOI1BEM4SLG1

FD score

xx

21G

00 10 20

PEX19VPS9VPS3MDM31

FD score

x 21G

00 10 20

MON1

FD score

x 21G

00 10 20

SNF2MUS81MMS1IRC19RAD54RAD27RAD51ARP8

FD score

xxx

xx

x

21G

00 10 20

NOP12RPL35A

FD score

xx

21G

00 10 20

RPS9BRPS11A

RPS17ASLX9RPS11BRPS16BHCR1RPS21ARPS27B

FD score

xx

xx

x

21G00 10 20

SNF2PHO2NHP10RAD54ARP8

FD score

xx

x

21G

Ribosomeassembly

DNA repairrecombination

Cytoplasmictranslation

rRNA processing

Rho proteinsignal

transductionOrganelle inheritance

Regulation ofribosomal proteingene transcription

Mitochondrialgenome maintenance

Chromatin remodeling

Maturation oflarge subunit rRNA

Vesicle docking

tRNA modification Protein localizationin mitochondrion

SAC3

TOF1

BUD21

VAM3

SAC3LOA1

TOM5

Figure 2 Biological processes enriched amongst genes associatedwith flight-specific fitness defects at different time points in the homozygousdeletion series Each node represents a significantly enriched gene ontology (GO) biological process (hypergeometric test 119875 le 001) A circlenode indicates enrichment at 14 generations compared to 7 generations (the first time point) a square node indicates enrichment at 21generations compared to 7 generations and a diamond node indicates enrichment at both 14 generations and 21 generations (see Methods)Node size is proportional to the significance of enrichment [minuslog

10(119875)] Node color indicates processes that share genes (see Methods) and

summary labels are shown for nodes of the same color Edges indicate ge 50 gene overlap between connected processes width is proportionalto the degree of overlap Each bar plot provides fitness defect (FD) scores for genes that contribute to the enrichment of processeswith the samenode color as the plot border Specifically the length of a bar is proportional to the log

2(abundance

7Gabundance14G21G) where abundanceyGrepresents the abundance of the corresponding gene deletion strain at 119910 generations (see Methods) An ldquo119909rdquo on the bar indicates that theabundance of the strain lowers to background level at later time point

chromatin remodeling were all required for resistance to theeffects of spaceflight Finally these DNA repair requirementsextend to the mitochondria which by virtue of its smallgenome is hypersensitive to DNA damage Consistent withthis we found that genes required for both mitochondrialmaintenance and proper protein localization to the mito-chondria were enriched in the homozygous samples

The enrichment of these particular processes is consistentwith a general induction of DNA damage which in turn

perturbs RNA biogenesis [25] Interestingly we have previ-ously observed this phenomenon with a class of therapeuticsthat act as nucleotide analogs such as 5-fluorouridine andfluorocytosine (described in detail below) Additionally it isparticularly noteworthy that although we do see evidenceof a requirement for RNA and DNA processing genesin spaceflight alone the requirement is exacerbated whenspaceflight is combined with the additional hyperosmoticstress imposed by the addition of 05M NaCl (Table S6)


00 10 20

SPE1HYP2

FD score

x

00 10 20

EPT1SPE1

FD score

xx

14G

14G

00 05 10 15

EPT1KAP95PSD1

FD score

xxx

21G

00 05 10 15

KAP95NUP57BIK1POM33

FD score

xxxx

00 05 10 15

MSH3

FD score

xx

21G

21G

00 05 10 15

RAP1PLC1YKU80

FD score

xx

x

21G

Modifiedamino acidbiosynthesis

Phosphatidylcholinemetabolism

Biogenic aminemetabolism

Nuclear poreorganization DNA catabolism

Proteinlocalization

to chromosome

MAG1

Figure 3 Biological processes enriched amongst genes associated with flight-specific fitness defects in the presence of NaCl at differenttime points in heterozygous deletion samples Each node represents a significantly enriched gene ontology (GO) biological process(hypergeometric test 119875 le 001) Nodes edges and plots are as specified for Figure 2

Table 2 Effects of spaceflight on yeast genome responses identifiedwith the homozygous deletion series

General pathway GO biological process

RNA metabolismand catabolism

(i) Ribosome biogenesis(ii) Regulation of ribosomal proteintranscription(iii) Cytoplasmic RNA translation(iv) rRNA processing(v) tRNA modification(vi) mRNA decay

DNA integrity

(i) DNA repair(ii) Recombination and replication(iii) Chromatin remodeling(iv) Mitochondrial maintenance(v) Proper protein localization to themitochondria

We speculate that the added salt stress potentiates the DNA-damaging effects of spaceflight via the induction of reactiveoxygen species (ROS)The ability of salt stress to induce ROSand subsequent DNA damage has been previously reported[26] and in particular the yeast mitochondria appears to behypersensitive to this type of stress consistent with its smallgenome being susceptible to the effects of DNA damage [27]Furthermore mitochondrial protein abundance has beenshown to rapidly increase upon osmotic shock and therefore

the enrichment for mitochondrial protein localization weobserve may reflect this requirement

To gain further insight into the pathways that modulatethe response to bothmicrogravity stress and combined space-flight and salt stress we used the GO enrichment profiles toquery a database of over 3200 distinct drug treatments of theyeast deletion collections [13] Specifically we quantified thesimilarity between the GO enrichments by computing theconcordance of minuslog

10(119875) between any two profiles where

119875 measures the significance of enrichment of a single GOcategory These concordance values are similar to Pearsoncorrelation values that is values closer to one indicate greatersimilarity between profiles except that high concordance alsorequires the scale of values to be similar between the profilesWhen calculating concordance we focused on GO biologicalprocess enrichment profiles (Table S7)

One of the strong concordances was observed with 5-fluorouridine (042) an FDA-approved anticancer drug thatis thought to also act by two mechanisms (i) inhibitingthymidylate synthetase and (ii) through metabolism intocytotoxic ribonucleotides and deoxyribonucleotides that canbe incorporated into DNA and RNA (Table 3) [14] Inaddition to being incorporated in DNA and RNA we andothers have shown that the drug has been shown to inhibit theessential ribonuclease activity of the exosome complex [28]Similarly carmofur a derivative of 5-fluorouracil displaysa concordance of 034 A similar concordance is seen with


Table 3 Concordance between drug effects and spaceflight effectson yeast genome responses identified with the homozygous deletionseries (+NaCl)

Drug (concordance) Biological function

5-Fluorouridine (042)5-Fluorouracil (036)Carmofur (034)5-Fluorocytosine (035)

Pyrimidine analogs that inhibitthymidylate synthase and aremetabolized into cytotoxicribonucleotides anddeoxyribonucleotides that can beincorporated into DNA and RNA

8-Methoxypsoralen(032) DNA-damaging agent

Diallyl disulfide (04)Increased glutathione-S-transferasechanges redox state by bindingelectrophilic toxins

5-fluorocytosine (5-FC) whose activity is identical to 5-fluorouracil (5-FU) Finally 8-methoxypsoralen a DNA-damaging agent that upon photoactivation conjugates andforms covalent bonds with DNA shows a congruence of 032This compound causes the formation of bothmonofunctional(addition to a single strand ofDNA) and bifunctional adducts(crosslinking of psoralen to both strands of DNA) thatultimately result in cell death

We also found high concordance to the diallyl disulfideprofile (040) an agent that has been demonstrated to beefficient for detoxification of a variety of cells Diallyl disulfideand related garlic derivatives have been shown to signifi-cantly increase the production of the enzyme glutathione S-transferase (GST) which binds electrophilic toxins in thecell Overloading the cell with inhibitory doses of diallyldisulfide reveals genes required for survival in the presenceof increased reactive oxygen species (ROS) [29]

In the case of the heterozygous collection we foundsignificant GO enrichments for the following categories lipidmetabolism DNA catabolism and regulation of transla-tion and posttranslational modification (specifically proteinphosphorylation) (Figure 3) As expected (based on previousstudies of the heterozygous collection) both the number ofgenes associated with FDs and the number of enriched cate-gories are considerably smaller than those derived from thehomozygous collection [30] This likely reflects two relatedphenomena first genes that when deleted in heterozygotesare sensitive to spaceflight encode proteins that participate inthe pathways identified in the homozygous collection wherethe fitness defect is stronger because the gene is completelyabsent Second none of these heterozygote strains encode adirect target of the perturbation

Interestingly when we searched for drug profiles withhigh concordance with the spaceflight profiles derived fromthe heterozygous collection we detected modest concor-dance with two human chemotherapeutics mitoxantrone(concordance = 019) and Epirubicin (congruence 0142)Both of these agents damage DNA by intercalating intothe DNA double helix and also by stabilizing the cleavablecomplex that is the substrate of topoisomerase II [31ndash33]

4 Conclusions

The experiments presented here represent a proof of prin-ciple for conducting full genome environmental screens inspaceflight using robust hardware that can recapitulate a fullautomation suite with environmental control in the space of asmall suitcaseThe performance of this platform is significantfor spaceflight studies and promises to enable terrestrialexperiments in extreme environments that will have directapplication to microbial bioprocessing for manufacturingalternative fuel development and basic research The resultsfrom these experiments suggest that spaceflight has subtle butsignificant effects on core cellular processes including growthcontrol via RNA and ribosomal biogenesis metabolismmodification and decay pathways Furthermore significantroles for DNA repair and replication response to pH signal-ing control of gene expression and mitochondrial functionwere observedThe yeast chemogenetic analysis of spaceflightsamples presented here strongly implicates DNA and RNAdamage as the major ground-based analogs of spaceflightstress Given the unique and substantial radiation exposurein space this is consistent with major radiation-mediatedeffects Unfortunately a 1 g control on ISS that might haveallowed better discrimination between the contributions ofspace radiation versus the effects of microgravity on yeastresponses was not available to us at this time Current on-going experiments are designed explore these effects anddissect them from other potentially confounding variablesThe high concordance to the profile induced by diallyl disul-fide suggests increased glutathione S-transferase binding ofelectrophilic toxins increased reactive oxygen species andchange in redox stateThese pathways which are required forsurvival in spaceflight can guide future experiments in twofundamental ways firstly by suggesting environmental mod-ifications that can bolster cellular and organismal integrity byavoiding further stress to these pathways and secondly byidentifying drug stresses that can exacerbate these pathwayrequirements in an effort to control pathological cell growthin the case of proliferative diseases

Abbreviations

AUC Area under the curveBP Biologic processesDMSO Dimethyl sulfoxideFD Fitness defectFDR False discovery rateGO Gene ontologyGST Glutathione S-transferaseISS International Space StationMAD Median absolute deviationMCL Markov Clustering AlgorithmOD Optical densityOPM Opticell Processing ModulePCR Polymerase chain reactionROS Reactive oxygen speciesRSS Residual sum of squaresYPD Yeast peptone dextrose



None of the authors have any commercial associations thatmight create a conflict of interests

Acknowledgments

These studies were supported by NASA Grant noNNX10AP01G The authors thank NASA for spaceflightaccess under the auspices of the International Space StationNational Lab Pathfinder program This material is the resultof work supported with resources and the use of facilities atthe Durhan Veterans Affairs Medical Center and the Officeof Research and Development Department of VeteransAffairs Veterans Health Administration Sequencing wasperformed in part at UBCSeq Vancouver Contents do notrepresent the views of the Department of Veterans Affairs orthe United States of America

References


[2] J J van Loon E H T E Folgering C V C Bouten J PVeldhuijzen and T H Smit ldquoInertial shear forces and the useof centrifuges in gravity research What is the proper controlrdquoJournal of Biomechanical Engineering vol 125 no 3 pp 342ndash346 2003

[3] T G Hammond and J M Hammond ldquoOptimized suspensionculture the rotating-wall vesselrdquo American Journal of Physiol-ogy Renal Physiology vol 281 no 1 pp F12ndashF25 2001

[4] M R Benoit R B Brown P Todd E S Nelson and D MKlaus ldquoBuoyant plumes from solute gradients generated bynon-motile Escherichia colirdquo Physical Biology vol 5 no 4Article ID 046007 2008

[5] J Kiefer and H D Pross ldquoSpace radiation effects and micro-gravityrdquoMutation Research vol 430 no 2 pp 299ndash305 1999

[6] D Botstein S A Chervitz and J M Cherry ldquoYeast as a modelorganismrdquo Science vol 277 no 5330 pp 1259ndash1260 1997

[7] D Botstein and G R Fink ldquoYeast an experimental organismfor 21st century biologyrdquo Genetics vol 189 no 3 pp 695ndash7042011

[8] P Todd ldquoOverview of the spaceflight radiation environmentand its impact on cell biology experimentsrdquo Journal of Gravi-tational Physiology vol 11 no 1 pp 11ndash16 2004

[9] J J W A van Loon ldquoMicro-gravity and mechanomicsrdquo Gravi-tational and Space Biology vol 20 no 2 pp 3ndash18 2007

[10] M Hughes-Fulford ldquoTo infinity and beyond Human space-flight and life sciencerdquo FASEB Journal vol 25 no 9 pp 2858ndash2864 2011

[11] T Fukuda K Fukuda A Takahashi et al ldquoAnalysis of deletionmutations of the rpsL gene in the yeast Saccharomyces cerevisiaedetected after long-term flight on the Russian space stationMirrdquoMutation Research Genetic Toxicology and EnvironmentalMutagenesis vol 470 no 2 pp 125ndash132 2000

[12] A Takahashi K Ohnishi S Takahashi et al ldquoThe effectsof microgravity on induced mutation in Escherichia coli andSaccharomyces cerevisiaerdquo Advances in Space Research vol 28no 4 pp 555ndash561 2001

[13] A Y Lee R P StOnge M J Proctor et al ldquoMapping thecellular response to small molecules using chemogenomicfitness signaturesrdquo Science vol 344 no 6186 pp 208ndash211 2014

[14] G Giaever P Flaherty J Kumm et al ldquoChemogenomic profil-ing identifying the functional interactions of small moleculesin yeastrdquo Proceedings of the National Academy of Sciences of theUnited States of America vol 101 no 3 pp 793ndash798 2004

[15] G Giaever and C Nislow ldquoThe yeast deletion collection adecade of functional genomicsrdquoGenetics vol 197 no 2 pp 451ndash465 2014

[16] T Roemer J Davies G Giaever and C Nislow ldquoBugs drugsand chemical genomicsrdquo Nature Chemical Biology vol 8 no 1pp 46ndash56 2012

[17] S E Pierce R W Davis C Nislow and G Giaever ldquoGenome-wide analysis of barcoded Saccharomyces cerevisiae gene-deletion mutants in pooled culturesrdquo Nature Protocols vol 2no 11 pp 2958ndash2974 2007

[18] A M Smith L E Heisler J Mellor et al ldquoQuantitativephenotyping via deep barcode sequencingrdquo Genome Researchvol 19 no 10 pp 1836ndash1842 2009

[19] Development Core Team R A Language and Environment forStatistical Computing R Foundation for Statistical ComputingVienna Austria 2011

[20] A C Douglas A M Smith S Sharifpoor et al ldquoFunctionalanalysis with a barcoder yeast gene overexpression systemrdquo G3vol 2 no 10 pp 1279ndash1289 2012

[21] Y Benjamini and Y Hochberg ldquoControlling the false discoveryrate a practical and powerful approach to multiple testingrdquoJournal of the Royal Statistical Society Series B vol 57 no 1 pp289ndash300 1995

[22] A M Deutschbauer D F Jaramillo M Proctor et al ldquoMecha-nisms of haploinsufficiency revealed by genome-wide profilingin yeastrdquo Genetics vol 169 no 4 pp 1915ndash1925 2005

[23] D Merico R Isserlin O Stueker A Emili and G DBader ldquoEnrichment map a network-basedmethod for gene-setenrichment visualization and interpretationrdquo PLoS ONE vol 5no 11 Article ID e13984 2010

[24] M E Smoot K Ono J Ruscheinski P-L Wang and T IdekerldquoCytoscape 28 new features for data integration and networkvisualizationrdquo Bioinformatics vol 27 no 3 pp 431ndash432 2011

[25] S Llanos and M Serrano ldquoDepletion of ribosomal protein L37occurs in response to DNA damage and activates p53 throughthe L11MDM2pathwayrdquoCell Cycle vol 9 no 19 pp 4005ndash40122010

[26] G F Ribeiro M Corte-Real and B Johansson ldquoCharacteriza-tion of DNA damage in yeast apoptosis induced by hydrogenperoxide acetic acid and hyperosmotic shockrdquo MolecularBiology of the Cell vol 17 no 10 pp 4584ndash4591 2006

[27] N A Doudican B Song G S Shadel and P W DoetschldquoOxidative DNA damage causes mitochondrial genomic insta-bility in Saccharomyces cerevisiaerdquoMolecular and Cellular Biol-ogy vol 25 no 12 pp 5196ndash5204 2005

[28] F Fang J Hoskins and J S Butler ldquo5-fluorouracil enhancesexosome-dependent accumulation of polyadenylated rRNAsrdquoMolecular and Cellular Biology vol 24 no 24 pp 10766ndash107762004

[29] Y-T Lin J-S Yang S-Y Lin et al ldquoDiallyl disulfide (DADS)induces apoptosis in human cervical cancer Ca Ski cells viareactive oxygen species and Ca2+-dependent mitochondria-dependent pathwayrdquo Anticancer Research vol 28 no 5 pp2791ndash2799 2008


[30] M E Hillenmeyer E Fung J Wildenhain et al ldquoThe chemicalgenomic portrait of yeast uncovering a phenotype for all genesrdquoScience vol 320 no 5874 pp 362ndash365 2008

[31] L Capolongo G Belvedere and M DrsquoIncalci ldquoDNA dam-age and cytotoxicity of mitoxantrone and doxorubicin indoxorubicin-sensitive and -resistant human colon carcinomacellsrdquo Cancer Chemotherapy and Pharmacology vol 25 no 6pp 430ndash434 1990

[32] B Bellosillo D Colomer G Pons and J Gil ldquoMitoxantronea topoisomerase II inhibitor induces apoptosis of B-chroniclymphocytic leukaemia cellsrdquo British Journal of Haematologyvol 100 no 1 pp 142ndash146 1998

[33] P Vejpongsa and E T H Yeh ldquoTopoisomerase 2120573 a promisingmolecular target for primary prevention of anthracycline-induced cardiotoxicityrdquo Clinical Pharmacology and Therapeu-tics vol 95 no 1 pp 45ndash52 2014

Review ArticleRhoGTPases as Key Players in MammalianCell Adaptation to Microgravity

Fiona Louis1 Christophe Deroanne2 Betty Nusgens2

Laurence Vico1 and Alain Guignandon1

1 INSERM U1059 Laboratoire de Biologie du Tissu Osseux Universite Jean Monnet 42023 Saint-Etienne Cedex France2Laboratoire de Biologie des Tissus Conjonctifs GIGA Universite de Liege 4000 Sart Tilman Belgium

Correspondence should be addressed to Alain Guignandon guignanduniv-st-etiennefr

Received 25 April 2014 Revised 14 August 2014 Accepted 9 September 2014


Copyright copy 2015 Fiona Louis et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

A growing number of studies are revealing that cells reorganize their cytoskeleton when exposed to conditions of microgravityMost if not all of the structural changes observed on flown cells can be explained by modulation of RhoGTPases which aremechanosensitive switches responsible for cytoskeletal dynamics control This review identifies general principles defining cellsensitivity to gravitational stresses We discuss what is known about changes in cell shape nucleus and focal adhesions and try toestablish the relationship with specific RhoGTPase activities We conclude by considering the potential relevance of live imagingof RhoGTPase activity or cytoskeletal structures in order to enhance our understanding of cell adaptation to microgravity-relatedconditions

1 Introduction

Microgravity has been demonstrated to have profound effectson both cellular and molecular levels including changes incell morphology [1 2] alterations of proliferation growthor differentiation [3 4] modification of gene expression [5ndash7] and changes in signal transduction cascades [5 8] Singleundifferentiated cells in vitro respond to altered conditionsof gravity but not all sensors and upstream regulators areknown which limits our understanding of cell sensitivity tomicrogravity-related conditions and evenmore tomicrograv-ity per se

There are numerous observations strengthening the ideathat cytoskeletal structures and cell surface receptors con-nected to them play an important role in the regulation ofthe differentiation potential of stem cells [9] As changes ofshape and of the inner cytoskeletal architecture are com-mon cell responses under conditions of real or simulatedmicrogravity [2] the idea of cytoskeletal involvement in thecellular response to microgravity seems obvious Moreoverstem cells or multipotent cells are recognized as being sen-sitive to mechanical stresses which are known to influence

cell commitment [10 11] The idea that not only terminallydifferentiated cells but also multipotent cells are sensitive tomicrogravity explains why even limited effects on cell com-mitment could have dramatic consequences Small GTPasesof the Rho family are known to control several aspects of celldynamics (vesicular transport traffic cytoskeleton turnover)[12 13] and appear to be the key players when trying togain a better understanding of the effects of microgravity ondifferentiated and multipotent cells

This review first attempts to highlight the fact that struc-tures involved inmechanotransduction pathways are respon-sible for adaptation to microgravity it will be explained thatstructural changes observed in cells exposed to real and sim-ulatedmicrogravitymay result from specific RhoGTPase reg-ulationsThen the degree to which the effects ofmicrogravityare important controllers of multipotent cell commitmentwill be discussed highlighting the critical role of RhoGT-Pases in these regulations The monitoring of RhoGTPaseactivities in conditions of microgravity is still a challenge asit is a dynamic process that controls other highly dynamicprocesses such as actin polymerization or focal adhesionturnover In order to decipher cell adaptation in conditions



of microgravity the community is in need of a live imagingtechnology like the one from Pache et al [15] but that can beset up in flightWe are conscious of all the difficulties of usingForster resonance energy transfer- (FRET-) based biosensorsdedicated to RhoA (Ras homolog gene family member A)and Rac1 (Ras-related C3 botulinum toxin substrate 1) twoimportant actors of this GTPases family under conditionsof microgravity and we are convinced that research groupsthat are successful with these types of sensors will providevery exciting results that will eliminate many confoundingfactors related to conditions of microgravity such as launchvibrations We predict that many specific GAP and GEF(resp RhoGTPases inhibitors and stimulators) will turn outto be key players in cell adaptation to microgravity-relatedconditions in the future

2 Mechanotransductors as Gravity Sensors

Discussions of whether an in vitro single cell or a cell pop-ulation can sense changes in the gravitational field are verycontroversial The currently most unknown research areainvolves the mechanism by which the physical event of g-force susception (by invagination sedimentation or buoy-ancy) becomes the biological process of g-force percep-tion Despite this an enormous body of experimental dataundoubtedly indicates that several types of cultured cellsare sensitive to gravity [16 17] If in fact cells do not fall(collapse) it is because they are supported in some wayThis support takes the form of a mechanical stress set upby the intermolecular forces in response to the distortionproduced by gravity In conditions where gravity is limited(microgravity) (such as those found in an orbiting vehicle)there is thus no distortion produced and consequently thereis no (limited) mechanical stress

It seems that undifferentiated cells have structural ele-ments that may play the role of ldquogravitational sensorsrdquo andldquosenserdquo the intensity of a mechanical tension and that severalintracellular processes can depend on the value of the gravita-tional forceTheoretical considerations suggest that the forcesinvolved are too small to trigger any response to the changedenvironment Several research teams think that these effectsaremostly caused by changes at the tissue and organ level [17]and that such environmental changes are stronger and morediverse [18] (eg lung heart and kidney become larger whilespleen or pancreas get smaller in rats [19]) In conclusiongravitational effects have been considered significant for cellswith a diameter of no less than 10 120583m[20]Thusmicrogravityseems to alter mammalian cells as compared to bacterial cellswhich are normally too small

Actors in the mechanotransduction chain represent keyelements involved in microgravity adaptation Nature pro-vides clear examples of defined mechanoreceptors in eukary-otes such as the statoliths in plants and the otoliths of theinner ear in most species of vertebrates Similar specializedcells of the sense organs detect pressure (touch) and vibra-tions and communicate these physical stimulations to thenerves of the afferent pathway up to the brain

It thus seems that undifferentiated mammalian cells doindeed have structural elements that may play the role of

a ldquogravitational sensorrdquo and ldquosenserdquo the intensity of amechan-ical tension and that many intracellular processes (adhesionproliferation survival contractility migration extracellularmatrix (ECM) architecture gene expression etc) can dependon the intensity of the gravitational forceThe identification ofcell structures capable of acting as gravisensors in in vitro cellsstill remains a problemThe general view of mechanosensingis that the overall cell is sensitive and is not a particularelement

In our opinion the most significant element (primummovens) that may impact on cytoskeletal dynamics undermicrogravity is the displacement of the nucleus The locationof the nucleus is probably dictated by a tension equilibriumbetween the cyto- and nucleoskeletons and we can imaginethat these tensions are constantly changing (in response tosignals) and that the nucleus probably oscillates continuously[21] A microgravity environment may influence the oscil-lating behavior of the nucleus [22] and then trigger a seriesof mechanical adjustments that may modulate cell shapeand structures as well as functions by way of transcriptionactivities

In response to changes in nucleus location cytoskeletalstructures and integrinsmight be solicited for cell adaptationThe cytoskeleton is a network of three interconnected systemsof filaments the actin microfilaments the microtubules andthe intermediate filaments They condition the shape of thecells and the major mechanical functions such as adhesionpolarization directional migration as well as proliferationsurvival or apoptosis gene expression and architecturalorganization of their supporting scaffold [12]

Experiments in real and simulated conditions of micro-gravity have shown that cytoskeletal modulations can occurquickly after variations in gravity have taken place Numer-ousarticles have reported on changes within 30min of theonset of a microgravity simulation affecting from focal adhe-sions to signal transduction Nevertheless cell response canbe observed only after few seconds following gravitationalchanges for example in parabolic flight experiments Afteronly 22 seconds of microgravity ML-1 thyroid cancer cellsshowed no sign of apoptosis or necrosis but the F-actin andcytokeratin cytoskeleton was altered [23] Endothelial cellsalso demonstrated no signs of death (after 31 parabolas of 22seconds) but had a cytoplasmic rearrangement and an alter-ation of cytoskeleton gene expressions [24] Concerningmes-enchymal stem cellsmorphologic characteristics of apoptosiscells (cell shrinkage membrane blebbing nuclear chromatincondensation etc) and decreased cell viability (rate of apop-tosis up to 5695) were reported 12 h after parabolic flightexperiment The F-actin stress fibers and microtubules weredisrupted and the expression of p53 (mRNA and proteinlevels) was upregulated [25] So gravity-induced response ofcells can occur very early within seconds

The reorganization of the cytoskeleton is believed togovern the modifications in size and shape of cells and nucleias well as the patterning number and maturation of focaladhesions The structures of the cytoskeleton nuclei andintegrins may claim to varying degrees to fulfill the role ofgravisensors [26]


The most likely candidates to assume the role of thesestructures are various elements of the cytoskeleton thenucleus intracellular organelles and also certain cell surfacereceptors (integrins) which interact both with cytoskeletalstructures and the extracellular matrix These structures areable to sense constraints and deformations in the matrixwhich are caused either by a gravitational or mechanicalfield and convert this signal into intracellular messengerswhich then give rise to a cellular response to the changes ingravity [21 27] It is also noteworthy that the cytoskeletonand integrins are not the primary sensors but react inresponse to their regulatory proteins (controllers of polymer-izationdestabilization agent)

Numerous cellular processes are controlled by gravity forexample calcium signaling mechanotransduction ligand-receptors interactions and cell-cell communications whichare all linked [28] During these mechanisms cell density isimportant because force transmission is greatest at cell-celland cell-substrate focal contacts where signaling moleculesare concentrated or clustered (ie integrin clustering) [17]Indeed transmission of forces from outside the cell throughcell-matrix and cell-cell contacts appears to control thematu-ration or disassembly of these adhesions which rearrange theorganization and contractile activity of the cytoskeleton Thecytoskeletal tensions formed at adhesions mediate mechan-ical signalling [29] Thus vinculin phosphorylation deter-mines whether cadherins transmit force and can producebiologically distinct functions [30]

In microgravity gravity-induced breakage of cell-celladhesions is reduced So cell-cell interaction was shown tobe promoted in absence of gravity [31] Cell adhesion proteinexpression specifically proteins found in tight junctions andadherens junctions was upregulated resulting in enhancedcell-cell contact between cells (endothelial cells [32]) Alsoincreased levels of E-cadherin were observed in 3D tumorconstructs cultured in simulated microgravity [33]

In osteoblasts a downregulation of cell-cell adhesion pro-teins such as catenin is observed [34] and also a reduc-tion in adhesion proteins such as vinculin and extracellularmatrix proteins such as fibronectin [35] To explain this phe-nomenon Levenberg et al showed that there is an autoreg-ulatory pathway that is activated by the presence of cell-cellor cell-substrate adhesion sites So when cell-cell adhesionis enhanced cell-matrix adhesion is decreased [36] Theseadhesion processes are also dependent on Ca2+ signalingpathways such as cell-cell adhesion via E-cadherinThis Ca2+dependence is through activation of the protein kinase C(PKC) second messenger system as well as activation ofphospholipase C (PLC) which in turn activates a signalingcascade resulting in the release of intracellular Ca2+ [37]This release of intracellular calcium facilitating the bindingof cadherins and 120573-catenin to the actin filaments comprisingthe cytoskeleton resulted in increased strength of cell-cellcontacts [38]

And several teams actually found a calcium release in vas-cular smooth muscle cells after 14 days of hindlimb unload-ing [39] and a downregulation of Calcium channel after

Microgravity

Cytoskeleton disruptionnuclear shape changesion channel activation

Rho-GTPases

Celladhesion

ECM

CSKdynamic

Integrated cell responses cell survival proliferation cell differentiation and stem cell commitment

ROS

Intracelltension

Wnt120573-catsignaling

Cell-cellcontact

Figure 1 Central role of the RhoGTPases in the integrated responseof mammalian cell to microgravity-related conditions A growingnumber of studies are revealing that cells reorganize their cytoskele-ton modulate intracellular tension and initiate nuclear shapeschanges when exposed to conditions ofmicrogravityMost if not allof the structural changes observed on flown cells can be explained bymodulation of RhoGTPases which are mechanosensitive switchesRhoGTPases are known for cytoskeletal dynamics control never-theless they are also involved in many other aspects as discussed inthis review We identify general principles dependent on RhoGT-Pases and define cell sensitivity to gravitational stresses such asoxidative stress intracellular tension cell-cell and cell-ECM adhe-sions and Wnt120573-catenin pathways We will try to establish thatintegrated cellular responses in microgravity are related to specificRhoGTPase activities

28 days [40] Also a reduction in intracellular calcium con-centration is observed after 2 days of simulated microgravityin chondrocytes [41] as well as in neurons [42] Moreoverin neutrophils PKC pathway is inhibited under microgravityleading to a decrease in intracellular concentration of Ca2+[43]

All the structural changes observed in cells subjectedto microgravity-related conditions are dictatedcontrolled bydynamic molecular switches of the GTPase family (Figure 1)Small RhoGTPases mainly control the regulation of intra-cellular traffic and are responsible for cytoskeletal dynamics[44]


RhoA

Rac1

Stabilisation microtubulesLinear actin polymerization in filopodia

Fibrillar actin stabilization

Actinmyosin fibers polymerization

Actinmyosin fibers depolymerization

Branching actin polymerization in lamellipodia+ end elongation microtubules

Fibrillogenesis

Migration

GEF

GAP

GDI

Sensing adhesion

cell-cell interactions

shape and polarity

Cell membrane

ECM

Integrins

Figure 2 RhoGTPase actions on the cytoskeleton and cell dynamics (modified from [14]) Integrins are necessary for translating themechan-ical properties of the extracellular environment into intracellular RhoGTPase-signaling pathways RhoA influences filopodia formation andfocal adhesion assembly and maturation in addition to controlling stress fiber formation and intracellular tension Rac1 primarily controlsactin assembly and formation of lamellipodia to ensure cell migration Fibrillogenesis is controlled positively by RhoA and negatively byRac1 Both RhoA and Rac1 are controlled by specific activators (GEF) and inhibitors (GAP GDI) Cell adaptation to mechanicalgravitationalchallenges triggers activation of pathways integrated by RhoGTPases

3 RhoGTPases MechanosensitiveMolecular Switches

RhoGTPases found in all eukaryotic cells are key regulatorymolecules which link surface receptors to the organizationand turnover of the cytoskeleton govern the formation ofcell-matrix adhesions and uphold the transcriptional controlof gene expression cell survival and proliferation [45] Theyare members of the Ras superfamily of small GTP-bindingproteins and are divided into three major classes RhoARac1 and Cdc42 GTPases are molecular switches that usea simple biochemical strategy to control complex cellularprocesses They switch between two conformational states aguanosine triphosphate- (GTP-) bound (ldquoactiverdquo) state andanother (ldquoinactiverdquo) state related to guanosine diphosphate(GDP) In their inactive forms RhoGTPases are sequestratedin the cytoplasm while upon signaling identified by integrinsand growth factor receptors they switch to their activeforms and translocate to the cell membrane [46] Therethey activate distinct and specific effector molecules whichin turn regulate the organization of the cytoskeleton and cell-matrix adhesions thus controlling cellular activities such asadhesion and also affect cell proliferation and the expressionof specific genes (Figure 2) [12] The cycle between theactive and inactive forms is under the direct control ofthree groups of regulatory proteins The guanine nucleotideexchange factors (GEFs) catalyze the exchange of GDP forGTP to activate Rho proteins The Rho proteins are thendeactivated by GTPase-activating proteins (GAPs) whichincrease the intrinsic GTPase activity of the Rho proteinleading to the hydrolysis of GTP to GDP The third groupof proteins involved in the cycle of Rho signaling is guaninedissociation inhibitors (RhoGDI) which hide the isoprenyl

groups of GTPases an action that promotes the sequestrationof inactive GTPases in the cytosol The RhoGDIs also inhibitthe release of GDP from the GTPase and contribute tothe maintenance of GTPases in an inactive state The Rhoprotein cycle is stimulated by agonists acting through Gprotein-coupled receptors (GPCRs) tyrosine kinase recep-tors cytokine receptor activation and mechanical stressesthat mainly govern the activity of the GEFs [47] The bestknown actions of the RhoGTPases onmechanical parametersof the cytoskeleton can be underscored by the expression ofconstitutively active RhoA and Rac1 in cell lines These mod-elsshow that RhoAactivation leads to better cell spreading butlower mechanical properties while Rac1 activation inducesmechanotransduction [48] As we assume that exposureto gravitational stress is a mechanical stimulation Rac1might be rapidly induced in microgravity-related conditionsThese results reveal the importance of RhoGTPases onmechanosensing cell shape adaptation or signal transduc-tionWewill summarize below the different controls they canhave on cellular mechanisms and metabolism

4 RhoGTPases Control Cytoskeleton Dynamic

In microgravity a qualitative and quantitative analysis of thestructures of F-actin 120573-tubulin and vinculin has revealeda higher density of filamentous actin and a decreased orga-nization in stress fibers Exposing mesenchymal stem cells(MSCs) to low gravity affected the distribution of the differentfilaments and more specifically led to a significant reductionof the F-actin fibers [49 50] extended filopodia increasedperinuclear distribution and decreased density [15 51]Moreover other research groups have found evidence of anaccumulation of actin at the cell border [52 53] This loss of


stress fibers is accompanied by an increase in monomeric G-actin content within the cells The preceded alterations maybe explained by a preferential reduction of RhoA activity

Indeed the activation of RhoAor Rac1 leads to the assem-bly of contractile actinmyosin filaments protrusive actin-rich lamellipodia and protrusive actin-rich filopodia whichin turn give rise to both the formation (actin polymerization)and the organization (filament bundling) of actin filamentsThus a number of studies (eg [54]) have shown that Rhokinase (ROCK) modulates the nonmuscle myosin II (NMM-II) activity by phosphorylation Another protein cofilinregulates actin polymerization and filament elongation Itsphosphorylation leads to inactivation and occurs primarilythrough LIM kinases (LIMK) which are activated by Rac1-dependent kinases Moreover LIMK-dependent phosphory-lation of cofilin can also be induced by RhoA acting throughits target ROCK whichmay be an important event in the sta-bilization of actinmyosin filaments [55] Microgravity leadsto an alteration of the actin cytoskeleton and consequentlyto a decrease of integrin signaling that may be caused by theinhibition of RhoA activity The absence of gravity increasestheG-actin form which reduces cofilin phosphorylation andis consistent with a decrease in focal adhesions and thus stressfibers [56]

Finally if a constitutively active RhoA is overexpresseda recovery stress of the fibers is enabled similar to what canbe observed under normal gravity and integrin signaling isrestored as shown in MSCs [57]

Microtubules play critical roles in eukaryotic cells Theyare key structural elements of the mitotic spindle apparatusduringmitosis and interphase and serve as tracks uponwhichmotor proteins transport vesicles and other componentsmove throughout the cell [58] Several studies have men-tioned perinuclear clustering in the microtubular networkduringmicrogravity [50 59] Also the loss of the radial struc-ture of microtubules has been observed after long stretches oftime (4 h) in microgravity [60]

Microgravity has also been proposed to influence micro-tubules by affecting the self-organization of filaments Accord-ing to the theory on self-organization and in a series of invitro studies with a change in gravity direction [61 62] andmicrogravity [61] it was clearly shown that microtubule self-organization is sensitive to the direction and the magnitudeof gravity which may explain the results obtained undermicrogravity Furthermore the observed disorganization ofmicrotubules may lead to a reduced rate of chromosome seg-regation during mitosis while alterations of actin microfila-ments and focal adhesions may also slow down cytokinesisand thus cell proliferation

RhoGTPases regulate microtubule dynamics in differ-ent ways Rac1 can phosphorylate at Ser16 of the microtu-bule plus-end-binding proteins (stathmins) which occurs inresponse to a number of extracellular stimuli [63] The effectof RhoA on microtubule dynamics is likely to be context-dependent For instance in migrating fibroblasts RhoA pro-motes the formation of stabilized microtubules Also micro-tubules play a major role in defining cell shape and polaritythrough the specific interaction of their plus-ends with pro-teins at the cell cortex This plus-end capture of microtubules

has been attributed to a number of plus-end-binding pro-teins whose activities are influenced by RhoGTPases [12]Altogether results onmicrotubules observed in conditions ofmicrogravity may be explained by an alteration of the RhoAand Rac1 activities

Microgravity has also had an impact on intermediatefilaments which after 12min in microgravity appeared aslarge bundles and aggregates in the vimentin network thatis the most distributed of all intermediate filament proteins[64] ROCK phosphorylates intermediate filament proteinsspecifically at the cleavage furrow during cytokinesis Thiscleavage furrow-specific phosphorylation plays an importantrole in the breakdown of local intermediate filaments andenables an efficient separation of intermediate filament net-works [65] In fact RhoA and Rac1 induce phosphorylationand reorganization of vimentin through kinases such asRhoA-associated protein kinase 2 (ROCK2) p21-activatedkinase (PAK) Src kinase (family of nonreceptor tyrosinekinases) and tyrosine kinases [66]

Concerning lamins which are nuclear intermediate fil-aments Uva et al showed DNA fragmentation in glial cellsafter 30min of microgravity and explained the phenomenonby caspases causing lamina to collapse and chromatin to con-dense [67] Proteins linking nucleoskeleton and cytoskeletoncomplexes (LINC) thus connecting lamina to the cytoskele-ton have been foundWhen it comes to laminopathymodelsin which this LINC complex is disrupted they lead mostlyto RhoA inhibition and lowered cytoplasmic elasticity whileactin and focal adhesion structures are mildly affected [68]Changes in nuclear structures that we identified earlier as animportant initiator ofmicrogravity effects [22] might explainthe RhoA activity inhibition and changes in cell tensionevoked under microgravity

Rac1 was shown to accumulate in the nuclear envelope inaddition to being expressed in the nucleoplasm and seemedto have the same pattern as that reported for lamin B [69]This Rac1 accumulation was proven to promote cell divisionIn microgravity the altered proliferation observed by Dai etal orDammet al [70 71] is controversial since Yuge et al [72]rather found an increased proliferation in human mesenchy-mal stem cellsWe thus suggest based on our results obtainedon rat osteosarcoma [73] that the lower proliferation mightbe explained by a reduced Rac1 activity in conditions ofmicrogravity

5 RhoGTPases as Regulators of Cell Adhesionand Matrix Remodeling

Integrins are transmembrane receptors that mediate theattachment between a cell and its surroundings such as othercells or the ECM In signal transduction integrins conveyinformation about the chemical composition andmechanicalstatus of the environment into the cell Therefore in additionto transmitting mechanical forces they are involved in cellsignaling and the regulation of cell cycles shapes andmotility [74]

Among the ligands of integrins can be mentionedfibronectin vitronectin collagen and lamininThen adapter


proteins such as talin and vinculin link the cytoskeleton tointegrins which attach the cell to the substrate forming afocal adhesion A variety of signaling proteins are associatedwith focal adhesions including focal adhesion kinase (FAK)which is an important mediator of signaling at these centersForces are also transmitted to the substrate at these sitesIn fibroblasts local forces correlate with the area of focaladhesions and actomyosin contractility blocking results in arapid disruption of focal adhesions [75]

In conditions of microgravity a reduced focal adhesion-related area (frequently reported [35 76]) can be explained bythe lower tension applied to the cytoskeleton This situationcan be associated with an inactivation of RhoA and as aresult by decreased fibrillogenesis (fibronectin collagen) dra-matically limiting integrin signaling The proof of a reducedintegrin signaling is thatMSCs have been observed to displaychanges in the expression levels of collagen-specific integrinsafter 7 days of cultivation in a rotational bioreactor [77] Infact activated expression of the 1205722-integrin has been seenduring the course ofMSC differentiation to osteogenesis [53]In addition Loesberg et al found a downregulation of 1205721 1205731and 1205733 integrins after 48 h of simulated microgravity [78]1205731 integrin has been shown to be important formediating

the response of MSCs to mechanical stimulation [79] Uponapplication of fluid shear stress an increase in alkaline phos-phatase (ALP) activity and expression of osteogenic markersis observed along with the activation of FAK and extra-cellular signal-regulated kinase 12 (ERK12) But when 1205731integrins are blocked FAK and ERK12 activation becomesinhibited [79] Phosphorylation of FAKhas also been demon-strated to be important for osteogenic differentiation ofhuman MSCs in response to tension [80] In microgravity-related conditions the limitation of integrin signaling can bea plausible explanation for the reduced osteogenesis

In addition limitation of the integrin-mediated responsecan also reduce important negative regulatory pathwaysThus growth of preadipocytes on a fibronectin matrix inhib-its adipocyte differentiation and this effect is overcome whenactin filaments are disrupted and promotes a rounding-upof cells [81] However 1205731 in association with 1205725 binds tofibronectin and Liu et al [82] reported the presence of anexpression switch from 1205725 to 1205726 at the growth arrest stageof differentiation which is consistent with an ECM changeobserved during adipogenesis This switch is necessary inorder to go from proliferation to differentiation of preadi-pocytes and can be explained by integrins 12057261205731 that bindto laminin and can thus interfere with chromatin and generegulation

These two integrins 1205725 and 1205726 are coordinately regulatedby cyclic adenosine monophosphate (cAMP) InterestinglycAMP has been shown to be activated in microgravity [83ndash85] RhoA and cAMP have antagonistic roles in regulatingcellular morphology [86] Thus the excessive productionof cAMP in microgravity may explain the limitation of RhoAactivation during adipogenesis followed by the integrinswitch of 1205725 to 1205726 to promote adipogenesis Also it is wellestablished that cAMP enhances the expression of bothCCAAT-enhancer-binding proteins (CEBP) 120572 and 120573 [87 88]

and initiates adipogenesis via the transcription factor CREB(cAMP response element binding protein) [89]

Concerning Rac1 cell adhesion to fibronectin (1205725 inte-grin) but not to laminin (1205726 integrin) is particularly effi-cient in activating Rac1 [90] leading to osteogenesis via 120573-cateninWnt pathways [91] In microgravity fibrillogenesis israpidly limited [92 93] which explains the delay or absence ofosteogenesis in multipotent cells The extracellular domainsof cadherins and 120573-catenin provide a link to 120572-catenin andthe actin cytoskeleton [94] Upon tyrosine phosphorylation120573-catenin also plays a significant role in signaling whentranslocated to the nucleus to regulate cell proliferation [95]

Noritake et al [96] have explained the increase in Rac1during osteogenesis until subconfluence cell adhesions accu-mulate E-cadherins at the sites of cell-cell contacts whichinduce Rac1 and thus actin-meshwork formation and 120573-catenin leading to osteogenesis In fact before E-cadherin-mediated cell-cell adhesion is establishedGDP-Rac1 is seques-tered in the cytosol by Rho GDIWhen E-cadherins accumu-late GDP-Rac1 is converted to GTP-Rac1 through the actionof a GEF and is targeted to the plasma membrane releasing120573-catenin linked to E-cadherin which can go to the nucleus[97]

In addition cell-to-cell physical contact via N-cadherinalso plays a crucial role in regulating osteoblastic activitysuch as alkaline phosphatase activity and 120573-catenin signaling[98 99] Consequently reduced cell-cell adhesion observedin microgravity due to limited proliferation may induce adecrease in Rac1 action and osteogenesis

Moreover it has been largely described that matrix rigid-ity affects osteogenesis MSCs grown on collagen-I stiff gels(linking to 1205721 or 1205722-1205731 integrins) have demonstrated acti-vated osteogenesis whereas softer collagen-I gels primeMSCs for a myogenic lineage [100] However cytoskeletonand the dynamicmechanical balance that exists between cellsand their ECM support appear as major players in severalmechanotransduction pathways [74] Microgravity decreasesthe expression of collagen I [101ndash103] induces matrix met-alloproteinases (MMP) production and reduces the levelof fibrillar collagen Thus it could be expected that alteredconditions of gravity may change the mechanical propertiesof ECM (ie the stiffness) Several studies for exampleMcBeath et al or Shih et al [104 105] have shown that oste-ogenic differentiation becomes increased on stiffer matri-ces as evident by type-I collagen osteocalcin Runx2 geneexpressions ROCK FAK and ERK12 induction and alizarinred S staining for mineralization Consequently FAK affectsosteogenic differentiation through ERK12 whereas RhoAand ROCK regulate both FAK and ERK12 [105]

In microgravity an initial modification of cytoskeletaldynamics might be at the origin of the following vicious cir-cle remodeling of a cytoskeleton is associated with a reducedinternal tension (contractility) leading to the dispersion ofFA With such a reduction in FA the cell tension cannotbe restored and fibrillogenesis might be limited Matrixdeposition limitation and MMP activation (Rac1 dependentprocess [106 107]) may further reduce the matrix stiffnessthus amplifying the dispersion of FAand reducing cell tensionand fibrillogenesis After a short exposure (from minutes to


hours) to microgravity-related conditions (before fibrilloge-nesis MMP production) the matrix stiffness is not modifiedWe can thus speculate that the ability of the cells to detect thestiff matrix they are normally seeded on has become rapidlyimpaired Mechanical information is normally conveyed byECM and cells by FA adaptation following tensegrity prin-ciples (equilibrium of internal and external tension) [21] inmicrogravity it seems that the displacement of the nucleus(sensitive to G) conveys the mechanical stimulus and from atensegrity perspective the cell adapts to the reduced tensionby lowering the ECM tension (interruption of fibrillogen-esis and MMP production) The short-term adaptation ofthe cell to microgravity that we have described up to nowseems to be characterized by a rapid reduction of RhoA andan increased Rac1 activity Altogether these studies revealedthat the control of cytoskeleton remodeling by RhoGTPasesimpacts on cell adhesion signaling limiting internal cel-lular tension as well as ECM fibrillogenesis and triggersMMP production thus limiting cell-matrix adhesion andsurvival

6 RhoGTPases in Stem Cell Commitment

In simulated microgravity cellular morphology is drasticallychanged after 7 days The MSCs appear rounder and lessfirmly attached to their substrate than under conditions ofnormal gravity Rather they are very spread out and display afibroblastic morphology [53]

Since the work by McBeath et al we know that theshape of a cell affects its differentiation potential [104] ThusMSCs that have been allowed to adhere over a larger areaare able to differentiate towards the osteogenic lineage whilecells adhering to a smaller area are restricted to the adi-pogenic lineage These impacts on lineage commitment bymesenchymal stem cells seem to be regulated by shape-induced changes in the RhoGTPase activity and cytoskeletaltension [108] Yao et al [109] showed that the cell shape itselfis an inherent cue to regulate stem cell differentiation bothwith and without external chemical induction factors Thusaccording to McBeath et al [104] expressing dominant-negative RhoA causes MSCs to become adipocytes whileconstitutively active RhoA induces osteoblastic or myocyticdifferentiation [110 111]

Concerning Rac1 it has been shown to promote cell adhe-sion and spreading and thereby to prevent the cell shapechange and the establishment of the cortical actin structurenecessary for adipocyte formation [109] Adhering cells arecharacterized by an elaborate network of stress fibers andfocal adhesions and are thus more prone to adopt a fibroblas-tic cell shape reflecting cytoskeleton tension [112 113] whichseems to be altered in conditions of microgravity

The cell shape may also depend on the available substratearea and hence the cell density However if cellular growth isreduced in microgravity the cell density will also be alteredGao et al [110] found that levels of RhoA activity did notvary substantially but that the Rac1 activity was significantlyhigher in well-spread cells during early differentiation than inhigh-density cells

They also demonstrated that Rac1 is necessary for osteo-genesis and that constitutively active Rac1 inhibited adipoge-nesis even if it is important for adipose commitment Liuet al [82] showed that an increase in preadipocyte densityinhibited the RhoA activity and that a downregulation of theRhoA-ROCK pathway was required for both adipose lineagecommitment and maturation [104 111] An increased celldensity thus appeared to be critically important

GTPases have also been shown to act in the cell cyclemitosis and cytokinesis RhoGTPases influence the cyclin-dependent kinase (cdk) activity during the G1-Phase of thecell cycle Thus RhoGTPases control the organization ofthe microtubule and actin fibers during cell cycling Aninhibition of RhoA or Rac1 blocks the G1 progression in avariety of mammalian cell types [114 115] Also Rac1 (but notRhoA) stimulates cyclin D1 transcription mediated by NF-120581B (nuclear factor kappa-light-chain-enhancer of activated Bcells) [116 117]Thus the necessity to downmodulate the Rac1activity in adipogenesis is that Rac1may prolong proliferationof preadipocytes which is consistent with the reported effectsof Rac1 on cyclin D1 [90 118 119] In fact Rac1 accumulates inthe nucleus during theG2phase of the cell cycle andpromotescell division [69] Concerning the cell division itself it hasbeen shown that actinmyosin filaments under the controlof ROCK are required at the cortex to allow positioning ofthe centrosomes [120] RhoA also plays a crucial role in thecontractile ring function [121]

Microgravity affects the growth proliferation and dif-ferentiation of osteoblasts Since the inhibition of RhoAobserved undermicrogravity blocksG1 progression [114 115]this may explain the altered proliferation and differentiationof osteoblastic cells and increased adipogenesis as summa-rized in Figure 3

Furthermore several cytoskeletal components includingRac1 GTPase activating protein 1 (Rac-GAP1) and Tropo-modulin 1 segregate asymmetrically during stem cell divi-sion and overexpression of these proteins may enhanceMSCcommitment as already proven with asymmetrical divisionsof hematopoietic stem cells to progenitor cells [122]

7 RhoGTPases and Wnt120573-CateninSignaling Crosstalk

Three Wnt signaling pathways have been characterized thecanonical Wnt pathway the noncanonical planar cell polar-ity pathway and the noncanonical Wntcalcium pathwayThe canonical Wnt pathway leads to regulation of genetranscription the noncanonical planar cell polarity pathwayregulates the cytoskeleton via a RhoGTPase regulation thatis responsible for the shape of the cell and the noncanonicalWntcalcium pathway regulates calcium inside the cell [123]

Mellor et al found that Wnt signaling was inhibited inconditions of microgravity [124] and mouse osteoblasts sub-jected to simulated microgravity were found to have lowerlevels of several components of the Wnt120573-catenin signalingpathway This may indicate even indirectly the activation ofan adipogenic program under microgravity [125] MoreoverWan et al [126] recently demonstrated a changed RhoA and


Microgravity

cAMP

RhoA

Rac1

E-cadherins N-cadherins

120573-Catenin

FAK ERK12

OsteoblastogenesisAdipogenesis

On MSC

12057251205731

12057261205731

CEBP120572et 120573

Figure 3 Role of AMPc on RhoGTPases activities and commitment of multipotent cells Microgravity affects the growth proliferation anddifferentiation ofmultipotent cells by increasingAMPc production AMPc contributes to cytoskeleton reorganization as it regulates negativelyRhoA activity Limitation of osteoblastogenesis might be linked to the ability of microgravity to reduce RhoA and Rac1 activities RhoAand Rac1 activations support osteoblasts differentiation for their respective role in ERK activation and beta-catenin nuclear translocationSustained adipogenesis observed in microgravity-related condition might be linked to ability of AMPc to trigger integrin a5b1a6b1 switchSignaling through a6b1 integrin is known to support adipogenesis A direct activation of adipogenic transcription factors (cEBPs) by AMPchas been also described

120573-catenin signaling after 1 and 25 h respectively in clinoro-tated osteoblasts They revealed that both the RhoA activityand the TCFLEF (T-cell factor-1 and lymphoid enhancingfactor-1) activity a 120573-catenin recruiter were downregulatedby unloading However the inhibition of 120573-catenin signalingblocked the unloading-inducedRhoA suppression and dom-inant negative RhoA inhibited the TCFLEF suppressionrevealing a regulation loop between 120573-catenin RhoA andTCFLEF Furthermore while 120573-catenin signaling seemedto be required for microgravity regulation of RhoA thisresponse was not mediated by the actin cytoskeleton or intra-cellular tension [126] The same was observed for Rac1120573-catenin signaling [91]

The Wnt canonical pathway involves the translocationof 120573-catenin to the nucleus and 120573-catenin has been shownto promote osteogenic differentiation in early osteoblastprogenitors in vivo [127] In contrast other studies have sug-gested that canonical Wnt signaling may actually promotestem cell renewal and inhibit osteogenic differentiation ofosteoprogenitor cells in vivo [128] as well as promoting stemcell renewal in human MSCs derived from bone marrow[129] Arnsdorf and colleagues [130] investigated the role ofnoncanonical Wnt signaling in mechanically induced osteo-genic differentiation of MSCs Exposure of MSCs to oscilla-tory fluid flow resulted in a translocation of 120573-catenin [131]and an upregulation of Wnt5a which is capable of inducingboth canonical and noncanonical pathways Wnt5a is alsonecessary for the flow-induced activation of RhoA Howeverthe inhibition of Wnt5a did not affect the 120573-catenin translo-cation which may instead be regulated by cadherin-catenin

signaling In addition Santos et al [132] showed that theactivation of the RhoAROCK pathway by Wnt5a induceda downregulation of adipogenic markers It was furtherreported that RhoA could also be activated by Wnt3a one ofthe canonical Wnt family members [133] and that an inhi-bition of intracellular 120573-catenin decreased the RhoA activity[134]

Kim et al [135] also found that Wnt signaling regulatedthe MSC differentiation into cardiomyocyte-like cells with aconcomitant downregulation of RhoA and upregulation ofRac1 Concerning Rac1 it was shown to be a critical regulatorin shear stress-driven 120573-catenin signaling in osteoblasts [91]and constitutively active Rac1 mutant caused a significantenhancement of the TCFLEF activity

These studies demonstrate that Wnt signaling is impor-tant for mechanically induced differentiation through RhoAor Rac1 pathways So in conditions of microgravity reducedRhoA cell shape and migratory behaviors can be explainedby Wnt and 120573-catenin signaling Finally RhoGTPases areregulated by Wnt signaling but in return 120573-catenin location(translocation) is dependent on RhoGTPases This intricateinterplay between both regulatory elements makes themparticularly important for the interpretation of microgravityeffects

8 RhoGTPases and Oxidative Stress

One of the first targets of Rac1 to be identified was p67phoxan essential structural component of the NADPH oxidasecomplex [136] Since then Rac1 has been reported to promote


reactive oxygen species (ROS) production in many cellsand to mediate the activity of the Nox family [137 138]Consequently Rac1 activation leads to the generation ofROS enabling adipogenesis commitment [139] and reducingosteoblastogenesis [140 141] Moreover GTPases act on theantioxidant master gene Nrf2 (nuclear factor-like 2) whichactivates a protective adaptive response to oxidative stressthrough transcriptional activation of antioxidant defensegenes [142]

RhoA is involved in Nrf2 phosphorylation which is nec-essary for its activation [143] Nrf2 is a transcription factorfor Hace1 (HECT domain and ankyrin repeat containing E3ubiquitin protein ligase 1) and Hace1 binds and ubiquitylatesRac1 when the latter is associated with NADPH oxidase thusblocking ROS generation byNOX [143 144] So RhoA activa-tion may limit ROS production and adipogenesis while Rac1activation may support it However several research groupshave reported that ROS causes RhoA activation [145 146]while Nimnual et al demonstrated that Rac1-mediated ROSproduction results in the downregulation of theRhoAactivity[147] This is also required for Rac1-induced formation ofmembrane ruffles and integrin-mediated cell spreading TheGTPase regulation by oxidative cell status thus still remainsunclear

In line with these papers several research groups suchas Versari et al have found increased oxidative stress duringspace flight due to microgravity [148 149] and cosmic radia-tions [150] As RhoA is decreased in microgravity this couldexplain the increased production of reactive oxygen speciesAccording to this paper we can assume that Rac1 activitiesare increased in microgravity An upregulated Rac1 activityfits well with enhanced ROS production and improved adi-pogenesis

However a higher Rac1 activity is also consistent witha higher ability for cell migration [151 152] Neverthelessresults of migration in space are controversial Bone marrowcells from rats and human embryonic brain cells show a facil-itated cell migration [153 154] while bone marrow CD34+cells have a lower migration potential in simulated micro-gravity [155] We can interpret the apparent discrepancies inmigration results based on the time spent in microgravityfor short-term exposure (from minutes to hours) there areseveral reasons to believe that RhoA is decreased and Rac1increased in line with their reciprocal inhibition [156] butfor longer exposure (from hours to days) the Rac1-inducedROS productionmay increase RhoA activation [145 146] andreduce the Rac1 activity limiting migration capabilities Themissing information in microgravity is related to the lack ofmeasurements of specific RhoGTPase activities

9 RhoGTPases Activities Monitoringin Microgravity

Meyers et al showed a reduction in active RhoA (minus88(plusmn2)) and a decrease in phosphorylation of cofilin after7 days in microgravity in addition to the absence of stressfibers [56] If overexpression of active RhoA is carried outthis enables a recovery of stress fibers and restored integrin

signals similar to those observed in normal gravity in MSC[57] In simulated microgravity a decrease in RhoA activitywas also observed after 72 h [157 158] Unfortunately nothingis known about Rac1 activity Zayzafoon et al thus proposedamodel in which the cytoskeleton is actually not the first sen-sor but a secondary step affected by a gravity-sensitive sensorIn this model it is the RhoA inactivation that is followedby cytoskeletal changes and transduction at FAs [57] whichexplains the alterations on MSC differentiations observedin microgravity To our knowledge our team is the first toperform RhoA and Rac1 monitoring during osteogenesisand adipogenesis in simulatedmicrogravity using embryonicmesenchymal stem cells C3H10T12 multipotent cells werecultured in modeled microgravity using NASArsquos rotatingwall vessels (RWV) or in control cultures under conditionsof earth gravity for up to 8 days seeded on collagen-coatedmicrobeads (Cytodex 3 Sigma) The results presented inFigure 4 show significant decreases in both RhoA and Rac1after long-term exposure to simulated microgravity To ourknowledge no comparison can bemade with data obtained inreal microgravity unfortunately Regardless of the limitationof themodel when it comes to simulatedmicrogravity-relatedconditions these results clearly showed that downregulationsof RhoA and Rac1 were compatible with enhanced adipoge-nesis and limited osteogenesis

As preservation of active RhoGTPases in flight conditionmight be challenging the recent validation of biosensorsfor imaging of active RhoA Rac1 and Cdc42 represents animportant step in understanding cell responses tomicrograv-ity Despite the critical role of RhoGTPases that we describein this review there is a dramatic lack of data concerning themonitoring of their activities during exposure to micrograv-ity particularly in real microgravity These data are of crucialimportance since cell adaptation is a dynamic process weneed to use available technologies such as fused fluorescentproteins and biosensors dedicated to following RhoGTPaseactivities in order to decipher cell adaptation in conditions ofmicrogravity On ground experiments extensive biochemicaland profiling studies on mechanotransduction pathways canbe performed In an automated spaceflight the use of biosen-sors specific to molecules integrating many pathways such asRhoGTPases should be presented as a simplified and inte-grated view of cell mechanics The community is in needof a live imaging data (already validated on ground [159])that can be now used in flight conditions We believe thatgroups that are successful in providing this type of integrateddata will surprise our community whose thinking is limitedby analysis of fixed images of cells and the monitoring ofindividual parameters

10 Conclusion

RhoGTPases represent a unique hub for integration of bio-chemical and mechanical signals As such they are probablyvery rapidly involved in a cellrsquos adaptation to microgravity-related conditions Published data describing this adaptationhave reported on alterations of the cytoskeleton adhesionand fibrillogenesis as well as an enhancement of the ROS


RhoA active assay-AD media

D6 120583G D6 1G0

10

20

30

40

Time (days)

( o

f pos

itive

cont

rol120583

g A

DN

)

lowast

(a)

RhoA active assay-OB media

D6 120583G D6 1G0

20

40

60

80

100P = 005

Time (days)

( o

f pos

itive

cont

rol120583

g A

DN

)

(b)

Rac1 active assay-AD media

D6 120583G D6 1G0

20

40

60

80

Time (days)

( o

f pos

itive

cont

rol120583

g A

DN

)

lowastlowast

(c)

Rac1 active assay-OB media

D6 120583G D6 1G0

100

200

300P = 005

Time (days)

( o

f pos

itive

cont

rol120583

g A

DN

)

(d)

Figure 4 RhoA and Rac1 activities are downregulated after 6 days of culture in simulated-microgravity conditions Cultures were performedwith C3H10T12 (multipotent embryonic cells) on collagen-coated microbeads (Cytodex 3 Sigma) for adipogenic induction and on Cytodex3 beads coated with apatite minerals complexed to collagen for an osteogenic oneThe adipogenic media contained 1 120583Mof rosiglitazone andthe osteogenic media 5mgmL of L-ascorbic acid 120573-glycerophosphate at 10minus3M and retinoic acid at 10minus5M in 120572MEMMicrobeads with cellswere cultured for 2 days in 90mm petri dishes (untreated for culture) with 10mL of proliferation media (120572MEM) after which the cells wereswitched 2 days in differentiated media and finally left for 6 days in a NASA rotating wall vessel (RWV) In parallel controls were realized byculturing beads in petri dishes RhoA and Rac1 active assays were performed with specific G-LISA kits (cytoskeleton) The positive controlswere pure active proteins of RhoA and Rac1 provided with the kit The results are expressed as percentage of the positive controls they showstandard error of themean (SEM) of samples extracted from three independent experiments and are compared with Studentrsquos statistical 119905-test

production and migration that can be explained by the spe-cific regulation of RhoGTPases To summarize the literaturewe can speculate that after a short exposure of a cell tomicrogravity the RhoA activity is depressed and the Rac1activity increased For long-term exposure osteogenesis hasbeen reported to be impaired and adipogenesis promotedThese changes in multipotent cell commitment fit nicelywith prolonged depressed activities of both RhoA and Rac1(Figure 5)

As we are convinced that focal adhesion and F-actinfibers are not the primary sensors of microgravity-relatedsignals (but rather transducers or effectors of the response)many specific GAP and GEF (resp RhoGTPase inhibitorsand stimulators) will emerge as new players in the adaptation

of cells to microgravity-related conditions What are themechanisms that explain the activation or inhibition of theseGTPases regulators As we try to establish that mechanosen-sors are involved in cell adaptation to microgravity we canpredict that critical players identified in these extreme con-ditions will in return be recognized in the mechanobiologyfield

Abbreviations

ALP Alkaline phosphataseCEBP CCAAT-enhancer-binding proteinscAMP Cyclic adenosine monophosphateCREB cAMP response element-binding protein


ECM

RhoATension Rac1

Migration

Earth gravity

(a)

ECM (Stiff)

RhoATension Rac1Smallernucleus

Migration

Microgravity

Short term(minutes to hours)

ROS

TranscriptionFocal

adhesions

(b)

ECM

RhoA Rac1

Migration

Focal adhesions

(Soft)

Anoikis

Fibrillo-genesis

Microgravity

Long term(hours to days) Osteogenesis

myogenesis Adipogenesis

Actin stress fibersMicrotubules

Intermediate filamentsPerinuclear actin

(c)

Figure 5 Proposedmodels describing the regulations of RhoA andRac1 activities in space-related conditions On EarthMSCs are well spreadand exhibit a tensed cytoskeleton in particular of microtubules intermediate filaments and actin stress fibers associated with stable focaladhesions within the extracellular matrixThese elements are controlled by GTPases RhoA and Rac1 We hypothesize that during short-termexposure to microgravity RhoAmight be inhibited to allow cytoskeleton reorganization in respect to the newmechanical status Cell tensionreduction might be mandatory during this adaptation At the same time Rac1 is activated to control peripheral actin polymerization andinduces ROS production All these events lead rapidly to a rounder cell shape with disorganization of microtubules stress fibers intermediatefilaments and focal adhesions Transcription may be also altered as nucleus shape is changed In these conditions cell is still able to migrateWhen exposure to microgravity is prolonged both RhoA and Rac1 might be inhibited explaining decreases in osteogenesis and myogenesisand enhancement of adipogenesis of MSCs In addition RhoA inhibition limits fibrillogenesis (a tension-dependent process) extracellularmatrix is not properly synthesized and lost its mechanical properties appearing softer for MSCs reinforcing adipogenesis At that timemigration is inhibited consistent with cytoskeleton alterations and Rac1 decrease MSCs become very round have low adhesion and mayinitiate anoikis

CSK CytoskeletonECM Extracellular matrixERK12 Extracellular signal-regulated kinase 12FAK Focal adhesion kinaseFRET Forster resonance energy transferGAPs GTPase-activating proteinsGDIs Guanine dissociation inhibitorsGDP Guanosine diphosphate

GEFs Guanine nucleotide exchange factorsGPCR G protein-coupled receptorGTP Guanosine triphosphateHace1 HECT domain and ankyrin repeat

containing E3 ubiquitin protein ligase 1LIMK LIM kinasesLINC Proteins linking nucleoskeleton and

cytoskeleton complexes


MMPs Matrix metalloproteinasesMSC Mesenchymal stem cellNF-120581B Nuclear factor

kappa-light-chain-enhancer of activatedB cells

NMM-II Nonmuscle myosin IINrf2 Nuclear factor (erythroid-derived 2-)

like 2PAK p21-activated kinaseRac1 Ras-related C3 botulinum toxin

substrate 1RhoA Ras homolog gene family member AROCK Rho kinaseROCK2 RhoA-associated protein kinase 2ROS Reactive oxygen speciesRWV Rotating wall vesselsSEM Standard error of the meanSrc family kinase Family of nonreceptor tyrosine kinasesTCFLEF T-cell factor-1 (Tcf-1) and lymphoid

enhancing factor-1 (Lef-1)


The authors declare that there is no conflict of interestsregarding to the publication of this paper

Acknowledgment

Thestudywas partially funded by the European SpaceAgency(Microgravity Application ProgramMAP ldquoERISTOrdquo) (Euro-pean Research in Space and Terrestrial Osteoporosis Con-tract no 1423200NLSH) and the French Centre NationaldrsquoEtudes Spatiales (CNES)

References

[1] P J Rijken R P de GrootW Briegleb et al ldquoEpidermal growthfactor-induced cell rounding is sensitive to simulated micro-gravityrdquo Aviation Space amp Environmental Medicine vol 62 no1 pp 32ndash36 1991


[3] A Cogoli ldquoFundamentals of space biology research on cellsanimals and plants in spacerdquo inCell Biology G Clement andKSlenzka Eds pp 121ndash170 Springer New York NY USA 2006


[5] R P de Groot P J Rijken J Boonstra A J Verkleij S W deLaat andW Kruijer ldquoEpidermal growth factor-induced expres-sion of c-fos is influenced by altered gravity conditionsrdquo Avia-tion Space and Environmental Medicine vol 62 no 1 pp 37ndash401991

[6] T G Hammond F C Lewis T J Goodwin et al ldquoGene expres-sion in spacerdquo Nature Medicine vol 5 no 4 p 359 1999

[7] Y Liu and EWang ldquoTranscriptional analysis of normal humanfibroblast responses tomicrogravity stressrdquoGenomics Proteom-ics and Bioinformatics vol 6 no 1 pp 29ndash41 2008


[9] P S Mathieu and E G Loboa ldquoCytoskeletal and focal adhesioninfluences on mesenchymal stem cell shape mechanical prop-erties and differentiation down osteogenic adipogenic andchondrogenic pathwaysrdquo Tissue EngineeringmdashPart B Reviewsvol 18 no 6 pp 436ndash444 2012

[10] J C Chen and C R Jacobs ldquoMechanically induced osteogeniclineage commitment of stem cellsrdquo Stem Cell Research andTherapy vol 4 article 107 no 5 2013

[11] J Eyckmans G L Lin and C S Chen ldquoAdhesive and mechan-ical regulation of mesenchymal stem cell differentiation inhuman bonemarrow and periosteum-derived progenitor cellsrdquoBiology Open vol 1 no 11 pp 1058ndash1068 2012

[12] A B Jaffe and A Hall ldquoRho GTPases biochemistry andbiologyrdquo Annual Review of Cell and Developmental Biology vol21 pp 247ndash269 2005

[13] A Hall ldquoG proteins and small GTpases distant relatives keep intouchrdquo Science vol 280 no 5372 pp 2074ndash2075 1998

[14] M Raftopoulou and A Hall ldquoCell migration rho GTPases leadthe wayrdquoDevelopmental Biology vol 265 no 1 pp 23ndash32 2004

[15] C Pache J Kuhn K Westphal et al ldquoDigital holographicmicroscopy real-time monitoring of cytoarchitectural altera-tions during simulated microgravityrdquo Journal of BiomedicalOptics vol 15 no 2 Article ID 026021 2010

[16] L B Buravkova ldquoProblems of the gravitational physiology of acellrdquo Human Physiology vol 36 no 7 pp 746ndash753 2010


[18] I D Pestov ldquoFundamentals of gravitational biologyrdquo Kosmich-eskaia Biologiia i Meditsina vol 2 no 1 9 pages 1997

[19] M Miyake M Yamasaki A Hazama S Nielsen and TShimizu ldquoEffects of microgravity on organ development of theneonatal ratrdquo Uchu Seibutsu Kagaku vol 18 no 3 pp 126ndash1272004

[20] E C Pollard ldquoTheoretical studies on living systems in theabsence ofmechanical stressrdquo Journal ofTheoretical Biology vol8 no 1 pp 113ndash123 1965

[21] D E Ingber N Wang and D Stamenovic ldquoTensegrity cellularbiophysics and the mechanics of living systemsrdquo Reports onProgress in Physics vol 77 no 4 2014

[22] R G Bacabac T H Smit J J W A van Loon B Z DoulabiM Helder and J Klein-Nulend ldquoBone cell responses to high-frequency vibration stress does the nucleus oscillate within thecytoplasmrdquo The FASEB Journal vol 20 no 7 pp 858ndash8642006



[25] R Meng H-Y Xu S-M Di et al ldquoHuman mesenchymal stemcells are sensitive to abnormal gravity and exhibit classic apop-totic featuresrdquo Acta Biochimica et Biophysica Sinica vol 43 no2 pp 133ndash142 2011


[26] M G Tairbekov ldquoMolekulyarnye i kletochnye osnovy gravitat-sionnoi chuvstvitelrsquonosti (Molecular andCellular Fundamentalsof Gravitational Sensitivity)rdquo 2002

[27] C A Lambert C M Lapiere and B V Nusgens ldquoBiology ofadherent cells in microgravityrdquo in Biology in Space and Life onEarth R N Enno Brinckmann Ed pp 123ndash155 Wiley-VCHNew York NY USA 2007

[28] T B Damm and M Egli ldquoCalciumrsquos role in mechanotransduc-tion during muscle developmentrdquo Cellular Physiology and Bio-chemistry vol 33 no 2 pp 249ndash272 2014

[29] C S Chen J Tan and J Tien ldquoMechanotransduction at cell-matrix and cell-cell contactsrdquo Annual Review of BiomedicalEngineering vol 6 pp 275ndash302 2004

[30] J L Bays X Peng C E Tolbert et al ldquoVinculin phosphory-lation differentially regulates mechanotransduction at cell-celland cell-matrix adhesionsrdquo Journal of Cell Biology vol 205 no2 pp 251ndash263 2014

[31] N L Cowger E Benes P L Allen and T G HammondldquoExpression of renal cell proteinmarkers is dependent on initialmechanical culture conditionsrdquo Journal of Applied Physiologyvol 92 no 2 pp 691ndash700 2002

[32] G L Sanford D Ellerson C Melhado-Gardner A E Sroufeand S Harris-Hooker ldquoThree-dimensional growth of endothe-lial cells in the microgravity-based rotating wall vessel bioreac-torrdquo In Vitro Cellular amp Developmental Biology Animal vol 38no 9 pp 493ndash504 2002

[33] M IngramG B Techy R Saroufeem et al ldquoThree-dimensionalgrowth patterns of various human tumor cell lines in simulatedmicrogravity of a NASA bioreactorrdquo In Vitro Cellular amp Devel-opmental BiologymdashAnimal vol 33 no 6 pp 459ndash466 1997

[34] H L Nichols N Zhang and X Wen ldquoProteomics and genom-ics of microgravityrdquo Physiological Genomics vol 26 no 3 pp163ndash171 2006

[35] A Guignandon M H Lafage-Proust Y Usson et al ldquoCellcycling determines integrin-mediated adhesion in osteoblasticROS 1728 cells exposed to space-related conditionsrdquo TheFASEB journal vol 15 no 11 pp 2036ndash2038 2001

[36] S Levenberg B-Z Katz K M Yamada and B Geiger ldquoLong-range and selective autoregulation of cell-cell or cell-matrixadhesions by cadherin or integrin ligandsrdquo Journal of CellScience vol 111 no 3 pp 347ndash357 1998

[37] J A Felix V V Chaban M L Woodruff and E R DirksenldquoMechanical stimulation initiates intercellular Ca2+ signaling inintact tracheal epitheliummaintained under normal gravity andsimulated microgravityrdquo American Journal of Respiratory Celland Molecular Biology vol 18 no 5 pp 602ndash610 1998

[38] K S Ko P D Arora V Bhide A Chen and C A McCullochldquoCell-cell adhesion in human fibroblasts requires calcium sig-nalingrdquo Journal of Cell Science vol 114 part 6 pp 1155ndash11672001

[39] P N Colleran B J Behnke M K Wilkerson A J Donatoand M D Delp ldquoSimulated microgravity alters rat mesentericartery vasoconstrictor dynamics through an intracellular Ca2+releasemechanismrdquoAmerican Journal of Physiology RegulatoryIntegrative and Comparative Physiology vol 294 no 5 ppR1577ndashR1585 2008

[40] M-J Xie Y-G Ma F Gao et al ldquoActivation of BKCa channel isassociated with increased apoptosis of cerebrovascular smoothmuscle cells in simulated microgravity ratsrdquo American Journalof Physiology Cell Physiology vol 298 no 6 pp C1489ndashC15002010

[41] X Li S Yang S Li P Jiang and Z Lin ldquoEffects of simulatedmicrogravity on the alkaline phosphatase activity and intracel-lular calcium concentration of cultured chondrocytesrdquo ChineseScience Bulletin vol 44 no 3 pp 218ndash221 1999

[42] K Meissner J R Piqueira and W Hanke ldquoFluorescent anddispersion experiments on biological membranes under micro-gravityrdquo Journal of Gravitational Physiology vol 11 no 2 ppP195ndashP196 2004

[43] A Sundaresan D Risin and N R Pellis ldquoLoss of signal trans-duction and inhibition of lymphocyte locomotion in a ground-basedmodel ofmicrogravityrdquo InVitro CellularampDevelopmentalBiologymdashAnimal vol 38 no 2 pp 118ndash122 2002

[44] A Hall ldquoRho GTpases and the actin cytoskeletonrdquo Science vol279 no 5350 pp 509ndash514 1998

[45] S Etienne-Manneville and A Hall ldquoRho GTPases in cell biol-ogyrdquo Nature vol 420 no 6916 pp 629ndash635 2002

[46] A J Ridley and A Hall ldquoThe small GTP-binding protein rhoregulates the assembly of focal adhesions and actin stress fibersin response to growth factorsrdquo Cell vol 70 no 3 pp 389ndash3991992

[47] M Chiariello J P Vaque P Crespo and J S GutkindldquoActivation of Ras and Rho GTPases and MAP Kinases by G-protein-coupled receptorsrdquo Methods in Molecular Biology vol661 pp 137ndash150 2010

[48] S Servotte Z Zhang C A Lambert et al ldquoEstablishmentof stable human fibroblast cell lines constitutively expressingactive Rho-GTPasesrdquo Protoplasma vol 229 no 2ndash4 pp 215ndash220 2006

[49] MHughes-Fulford andM L Lewis ldquoEffects ofmicrogravity onosteoblast growth activationrdquo Experimental Cell Research vol224 no 1 pp 103ndash109 1996


[51] S I M Carlsson M T S Bertilaccio E Ballabio and J AMMaier ldquoEndothelial stress by gravitational unloading effectson cell growth and cytoskeletal organizationrdquo Biochimica etBiophysica ActamdashMolecular Cell Research vol 1642 no 3 pp173ndash179 2003

[52] L B Buravkova P M Gershovich J G Gershovich and A IGrigorrsquoev ldquoMechanisms of gravitational sensitivity of osteogenicprecursor cellsrdquo Acta Naturae vol 2 no 1 pp 28ndash36 2010


[54] F Matsumura ldquoRegulation of myosin II during cytokinesis inhigher eukaryotesrdquo Trends in Cell Biology vol 15 no 7 pp 371ndash377 2005

[55] K Ohashi K Nagata M Maekawa T Ishizaki S Narumiyaand K Mizuno ldquoRho-associated kinase ROCK activates LIM-kinase 1 by phosphorylation at threonine 508 within the activa-tion looprdquo The Journal of Biological Chemistry vol 275 no 5pp 3577ndash3582 2000

[56] V E Meyers M Zayzafoon J T Douglas and J M McDonaldldquoRhoA and cytoskeletal disruption mediate reduced osteoblas-togenesis and enhanced adipogenesis of human mesenchymalstem cells in modeled microgravityrdquo Journal of Bone and Min-eral Research vol 20 no 10 pp 1858ndash1866 2005


[57] M Zayzafoon V E Meyers and J M McDonald ldquoMicrograv-ity the immune response and bonerdquo Immunological Reviewsvol 208 no 1 pp 267ndash280 2005

[58] C E Walczak ldquoMicrotubule dynamics and tubulin interactingproteinsrdquo Current Opinion in Cell Biology vol 12 no 1 pp 52ndash56 2000

[59] F Yang ZDai Y Tan andY Li ldquoEffects of altered gravity on thecytoskeleton of neonatal rat cardiocytesrdquo Microgravity Scienceand Technology vol 22 no 1 pp 45ndash52 2010


[61] J Tabony N Rigotti N Glade and S Cortes ldquoEffect of weight-lessness on colloidal particle transport and segregation in self-organising microtubule preparationsrdquo Biophysical Chemistryvol 127 no 3 pp 172ndash180 2007


[63] H Daub K Gevaert J Vandekerckhove A Sobel and A HallldquoRacCdc42 and p65PAK regulate the microtubule-destabiliz-ing protein stathmin through phosphorylation at serine 16rdquoTheJournal of Biological Chemistry vol 276 no 3 pp 1677ndash16802001

[64] L Sciola M Cogoli-Greuter A Cogoli A Spano and P PippialdquoInfluence of microgravity on mitogen binding and cytoskele-ton in Jurkat cellsrdquo Advances in Space Research vol 24 no 6pp 801ndash805 1999

[65] H Goto H Kosako and M Inagaki ldquoRegulation of interme-diate filament organization during cytokinesis possible roles ofRho-associated kinaserdquoMicroscopyResearch andTechnique vol49 no 2 pp 173ndash182 2000

[66] L Chang and R D Goldman ldquoIntermediate filaments mediatecytoskeletal crosstalkrdquo Nature Reviews Molecular Cell Biologyvol 5 no 8 pp 601ndash613 2004

[67] BMUvaM AMasiniM Sturla et al ldquoMicrogravity-inducedapoptosis in cultured glial cellsrdquoEuropean Journal ofHistochem-istry vol 46 no 3 pp 209ndash214 2002

[68] C M Hale A L Shrestha S B Khatau et al ldquoDysfunctionalconnections between the nucleus and the actin andmicrotubulenetworks in laminopathic modelsrdquo Biophysical Journal vol 95no 11 pp 5462ndash5475 2008

[69] D Michaelson W Abidi D Guardavaccaro et al ldquoRac1 accu-mulates in the nucleus during the G2 phase of the cell cycle andpromotes cell divisionrdquo Journal of Cell Biology vol 181 no 3 pp485ndash496 2008


[71] T B Damm A Franco-Obregon and M Egli ldquoGravitationalforce modulates G

2M phase exit in mechanically unloaded

myoblastsrdquo Cell Cycle vol 12 no 18 pp 3001ndash3012 2013[72] L Yuge T Kajiume H Tahara et al ldquoMicrogravity potentiates

stem cell proliferation while sustaining the capability of differ-entiationrdquo Stem Cells and Development vol 15 no 6 pp 921ndash929 2006

[73] AGuignandonCGenty LVicoM-H Lafage-Proust S Palleand C Alexandre ldquoDemonstration of feasibility of automated

osteoblastic line culture in space flightrdquo Bone vol 20 no 2 pp109ndash116 1997

[74] F J Alenghat and D E Ingber ldquoMechanotransduction allsignals point to cytoskeleton matrix and integrinsrdquo SciencersquosSTKE Signal Transduction Knowledge Environment vol 2002no 119 article PE6 2002

[75] N Q Balaban U S Schwarz D Riveline et al ldquoForce and focaladhesion assembly a close relationship studied using elasticmicropatterned substratesrdquoNature Cell Biology vol 3 no 5 pp466ndash472 2001

[76] N Nabavi A Khandani A Camirand and R E HarrisonldquoEffects of microgravity on osteoclast bone resorption and oste-oblast cytoskeletal organization and adhesionrdquoBone vol 49 no5 pp 965ndash974 2011

[77] J Gebken B Luders H Notbohm et al ldquoHypergravity stimu-lates collagen synthesis in human osteoblast-like cells evidencefor the involvement of p4442 MAP-kinases (ERK 12)rdquo TheJournal of Biochemistry vol 126 no 4 pp 676ndash682 1999

[78] W A Loesberg X FWalboomers J JW A Van Loon and J AJansen ldquoSimulated microgravity activates MAPK pathways infibroblasts cultured on microgrooved surface topographyrdquo CellMotility and the Cytoskeleton vol 65 no 2 pp 116ndash129 2008

[79] L Liu C Zong B Li et al ldquoThe interaction between1205731 integrinsand ERK12 in osteogenic differentiation of human mesenchy-mal stem cells under fluid shear stress modelled by a perfu-sion systemrdquo Journal of Tissue Engineering and RegenerativeMedicine vol 8 no 2 pp 85ndash96 2014

[80] D FWard Jr W AWilliams N E Schapiro et al ldquoFocal adhe-sion kinase signaling controls cyclic tensile strain enhanced col-lagen I-induced osteogenic differentiation of humanmesenchy-mal stem cellsrdquoMolecular and Cellular Biomechanics vol 4 no4 pp 177ndash188 2007

[81] B M Spiegelman and C A Ginty ldquoFibronectin modulation ofcell shape and lipogenic gene expression in 3T3-adipocytesrdquoCell vol 35 no 3 part 2 pp 657ndash666 1983

[82] J Liu S M DeYoung M Zhang M Zhang A Cheng andA R Saltiel ldquoChanges in integrin expression during adipocytedifferentiationrdquo Cell Metabolism vol 2 no 3 pp 165ndash177 2005

[83] P Barbe J Galitzky I de Glisezinski et al ldquoSimulated micro-gravity increases 120573-adrenergic lipolysis in human adipose tis-suerdquo The Journal of Clinical Endocrinology amp Metabolism vol83 no 2 pp 619ndash625 1998

[84] H Maass J Transmontano and F Baisch ldquoResponse of adren-ergic receptors to 10 days head-down tilt bedrestrdquo Acta Physi-ologica Scandinavica Supplement vol 144 no 604 pp 61ndash681992

[85] V A Convertino J L Polet K A Engelke G W HofflerL D Lane and C G Blomqvist ldquoEvidence for increased 120573-adrenoreceptor responsiveness induced by 14 days of simulatedmicrogravity in humansrdquo American Journal of Physiology vol273 no 1 part 2 pp R93ndashR99 1997

[86] J-M Dong T Leung E Manser and L Lim ldquocAMP-inducedmorphological changes are counteracted by the activated RhoAsmall GTPase and the Rho kinase ROK120572rdquoThe Journal of Biolog-ical Chemistry vol 273 no 35 pp 22554ndash22562 1998

[87] Z Cao RM Umek and S L McKnight ldquoRegulated expressionof three CEBP isoforms during adipose conversion of 3T3-L1cellsrdquo Genes amp Development vol 5 no 9 pp 1538ndash1552 1991

[88] Q-Q Tang M-S Jiang and M D Lane ldquoRepressive effect ofSp1 on the CEBP120572 gene promoter role in adipocyte differen-tiationrdquoMolecular and Cellular Biology vol 19 no 7 pp 4855ndash4865 1999


[89] J E Reusch L A Colton and D J Klemm ldquoCREB activationinduces adipogenesis in 3T3-L1 cellsrdquo Molecular amp CellularBiology vol 20 no 3 pp 1008ndash1020 2000

[90] A Mettouchi S Klein W Guo et al ldquoIntegrin-specific activa-tion of Rac controls progression through theG

1phase of the cell

cyclerdquoMolecular Cell vol 8 no 1 pp 115ndash127 2001[91] QWan E ChoH Yokota and SNa ldquoRac1 andCdc42GTPases

regulate shear stress-driven 120573-catenin signaling in osteoblastsrdquoBiochemical and Biophysical Research Communications vol 433no 4 pp 502ndash507 2013

[92] M Hughes-Fulford and V Gilbertson ldquoOsteoblast fibronectinmRNA protein synthesis andmatrix are unchanged after expo-sure tomicrogravityrdquoFASEB Journal vol 13 no 8 pp S121ndashS1271999

[93] A Guignandon C Faure T Neutelings et al ldquoRac1 GTPasesilencing counteracts microgravity-induced effects on osteo-blastic cellsrdquo The FASEB Journal vol 28 no 9 pp 4077ndash40872014

[94] F H Brembeck M Rosario andW Birchmeier ldquoBalancing celladhesion andWnt signaling the key role of 120573-cateninrdquo CurrentOpinion in Genetics and Development vol 16 no 1 pp 51ndash592006

[95] F M van Roy and P D McCrea ldquoA role for kaiso-p120ctncomplexes in cancerrdquoNature Reviews Cancer vol 5 no 12 pp956ndash964 2005

[96] J Noritake M Fukata K Sato et al ldquoPositive role of IQGAP1an effector of Rac1 in actin-meshwork formation at sites of cell-cell contactrdquo Molecular Biology of the Cell vol 15 no 3 pp1065ndash1076 2004

[97] M Fukata and K Kaibuchi ldquoRho-family GTPases in cadherin-mediated cell-cell adhesionrdquo Nature Reviews Molecular CellBiology vol 2 no 12 pp 887ndash897 2001

[98] CHMCastro C S Shin J P Stains et al ldquoTargeted expressionof a dominant-negative N-cadherin in vivo delays peak bonemass and increases adipogenesisrdquo Journal of Cell Science vol117 no 13 pp 2853ndash2864 2004

[99] S L Ferrari K Traianedes M Thorne et al ldquoRole for N-cadherin in the development of the differentiated osteoblasticphenotyperdquo Journal of Bone andMineral Research vol 15 no 2pp 198ndash208 2000

[100] A J Engler S Sen H L Sweeney and D E Discher ldquoMatrixelasticity directs stem cell lineage specificationrdquo Cell vol 126no 4 pp 677ndash689 2006

[101] T P Stein and C E Wade ldquoMetabolic consequences of muscledisuse atrophyrdquo The Journal of Nutrition vol 135 no 7 pp1824Sndash1828S 2005

[102] B Nusgens G Chometon A Guignandon et al ldquoRole ofthe RhoGTPases in the cellular receptivity and reactivity tomechanical signals including microgravityrdquo Journal of Gravita-tional Physiology vol 12 no 1 pp 269ndash270 2005

[103] Z-G Zhang C A Lambert S Servotte et al ldquoEffects of con-stitutively active GTPases on fibroblast behaviorrdquo Cellular andMolecular Life Sciences vol 63 no 1 pp 82ndash91 2006

[104] R McBeath D M Pirone C M Nelson K Bhadriraju and CS Chen ldquoCell shape cytoskeletal tension and RhoA regulatestem cell lineage commitmentrdquo Developmental Cell vol 6 no4 pp 483ndash495 2004

[105] Y-R V Shih K-F Tseng H-Y Lai C-H Lin and O KLee ldquoMatrix stiffness regulation of integrin-mediated mechan-otransduction during osteogenic differentiation of humanmes-enchymal stem cellsrdquo Journal of Bone andMineral Research vol26 no 4 pp 730ndash738 2011

[106] D L Long J S Willey and R F Loeser ldquoRac1 is requiredfor matrix metalloproteinase 13 production by chondrocytes inresponse to fibronectin fragmentsrdquo Arthritis and Rheumatismvol 65 no 6 pp 1561ndash1568 2013

[107] D C Radisky D D Levy L E Littlepage et al ldquoRac1b andreactive oxygen species mediate MMP-3-induced EMT andgenomic instabilityrdquo Nature vol 436 no 7047 pp 123ndash1272005

[108] J Settleman ldquoTension precedes commitmentmdasheven for a stemcellrdquoMolecular Cell vol 14 no 2 pp 148ndash150 2004

[109] X Yao R Peng and J Ding ldquoEffects of aspect ratios of stem cellson lineage commitments with and without induction mediardquoBiomaterials vol 34 no 4 pp 930ndash939 2013

[110] L Gao R McBeath and C S Chen ldquoStem cell shape regulatesa chondrogenic versus myogenic fate through Rac1 and N-cadherinrdquo Stem Cells vol 28 no 3 pp 564ndash572 2010

[111] R Sordella W Jiang G-C Chen M Curto and J Settle-man ldquoModulation of Rho GTPase signaling regulates a switchbetween adipogenesis and myogenesisrdquo Cell vol 113 no 2 pp147ndash158 2003

[112] S Huang C S Chen and D E Ingber ldquoControl of cyclin D1p271198701198941199011 and cell cycle progression in human capillary endothe-lial cells by cell shape and cytoskeletal tensionrdquo MolecularBiology of the Cell vol 9 no 11 pp 3179ndash3193 1998

[113] SHuang andD E Ingber ldquoThe structural andmechanical com-plexity of cell-growth controlrdquo Nature Cell Biology vol 1 no 5pp E131ndashE138 1999

[114] M F Olson A Ashworth and A Hall ldquoAn essential role forRho Rac and Cdc42 GTPases in cell cycle progression throughG1rdquo Science vol 269 no 5228 pp 1270ndash1272 1995

[115] MYamamotoNMarui T Sakai et al ldquoADP-ribosylation of therhoA gene product by botulinum C3 exoenzyme causes Swiss3T3 cells to accumulate in the G1 phase of the cell cyclerdquo Onco-gene vol 8 no 6 pp 1449ndash1455 1993

[116] D Joyce B Bouzahzah M Fu et al ldquoIntegration of Rac-dependent regulation of cyclin D1 transcription through anuclear factor-120581B-dependent pathwayrdquoThe Journal of BiologicalChemistry vol 274 no 36 pp 25245ndash25249 1999

[117] J K Westwick Q T Lambert G J Clark et al ldquoRac regulationof transformation gene expression and actin organization bymultiple PAK-independent pathwaysrdquo Molecular amp CellularBiology vol 17 no 3 pp 1324ndash1335 1997

[118] M L Coleman and C J Marshall ldquoA family outing smallGTPases cyclinrsquo through G1rdquo Nature Cell Biology vol 3 no 11pp E250ndashE251 2001

[119] A J Ridley ldquoCyclinrsquo round the cell with Racrdquo DevelopmentalCell vol 1 no 2 pp 160ndash161 2001

[120] J Rosenblatt L P Cramer B Baum and KMMcGee ldquoMyosinII-dependent cortical movement is required for centrosomeseparation and positioning during mitotic spindle assemblyrdquoCell vol 117 no 3 pp 361ndash372 2004

[121] M Glotzer ldquoAnimal cell cytokinesisrdquoAnnual Review of Cell andDevelopmental Biology vol 17 pp 351ndash386 2001

[122] S B Ting E Deneault K Hope et al ldquoAsymmetric segregationand self-renewal of hematopoietic stem and progenitor cellswith endocytic Ap2a2rdquo Blood vol 119 no 11 pp 2510ndash25222012

[123] R Nusse and H Varmus ldquoThree decades of Wnts a personalperspective on how a scientific field developedrdquo The EMBOJournal vol 31 no 12 pp 2670ndash2684 2012


[124] L Mellor T Bake M Hiremath E G Loboa and J T OxfordldquoSimulated microgravity affects Wnt signaling in articularcartilage possible implications for crosstalk between cartilageand subchondral bonerdquo inProceedings of the 2014NASAHumanResearch Program InvestigatorsrsquoWorkshop Galveston Tex USAFebruary 2014

[125] M Capulli A Rufo A Teti and N Rucci ldquoGlobal transcrip-tome analysis in mouse calvarial osteoblasts highlights setsof genes regulated by modeled microgravity and identifies Aldquomechanoresponsive osteoblast gene signaturerdquordquo Journal of Cel-lular Biochemistry vol 107 no 2 pp 240ndash252 2009

[126] Q Wan E Cho H Yokota and S Na ldquoRhoA GTPase interactswith beta-catenin signaling in clinorotated osteoblastsrdquo Journalof Bone andMineralMetabolism vol 31 no 5 pp 520ndash532 2013

[127] S J Rodda and A P McMahon ldquoDistinct roles for Hedgehogand caronicalWnt signaling in specification differentiation andmaintenance of osteoblast progenitorsrdquo Development vol 133no 16 pp 3231ndash3244 2006

[128] J-B Kim P Leucht K Lam et al ldquoBone regeneration is regu-lated by Wnt signalingrdquo Journal of Bone and Mineral Researchvol 22 no 12 pp 1913ndash1923 2007

[129] D Baksh and R S Tuan ldquoCanonical and non-canonical Wntsdifferentially affect the development potential of primary isolateof human bone marrow mesenchymal stem cellsrdquo Journal ofCellular Physiology vol 212 no 3 pp 817ndash826 2007

[130] E J Arnsdorf P Tummala and C R Jacobs ldquoNon-canonicalWnt signalling andN-cadherin related120573-catenin signalling playa role in mechanically induced osteogenic cell faterdquo PLoS ONEvol 4 no 4 Article ID e5388 2009

[131] N Case M Ma B Sen Z Xie T S Gross and J Rubin ldquo120573-Catenin levels influence rapid mechanical responses in osteo-blastsrdquoThe Journal of Biological Chemistry vol 283 no 43 pp29196ndash29205 2008

[132] A Santos A D Bakker J M A De Blieck-Hogervorst andJ Klein-Nulend ldquoWNT5A induces osteogenic differentiationof human adipose stem cells via rho-associated kinase RockrdquoCytotherapy vol 12 no 7 pp 924ndash932 2010

[133] J Rossol-Allison L N Stemmle K I Swenson-Fields et alldquoRho GTPase activity modulates Wnt3a120573-catenin signalingrdquoCellular Signalling vol 21 no 11 pp 1559ndash1568 2009

[134] L Peng Y Li K Shusterman M Kuehl and C W GibsonldquoWnt-RhoA signaling is involved in dental enamel develop-mentrdquo European Journal of Oral Sciences vol 119 supplementS1 pp 41ndash49 2011

[135] M-H Kim M Kino-oka N Maruyama A Saito Y SawaandM Taya ldquoCardiomyogenic induction of human mesenchy-mal stem cells by altered Rho family GTPase expression ondendrimer-immobilized surface with d-glucose displayrdquo Bio-materials vol 31 no 30 pp 7666ndash7677 2010

[136] D Diekmann A Abo C Johnston A W Segal and A HallldquoInteraction of Rac with p67phox and regulation of phagocyticNADPHoxidase activityrdquo Science vol 265 no 5171 pp 531ndash5331994

[137] J D Lambeth ldquoNoxDuox family of nicotinamide adenine din-ucleotide (phosphate) oxidasesrdquo Current Opinion in Hematol-ogy vol 9 no 1 pp 11ndash17 2002

[138] R Takeya and H Sumimoto ldquoMolecular mechanism for activa-tion of superoxide-producingNADPHoxidasesrdquoMolecules andCells vol 16 no 3 pp 271ndash277 2003

[139] M Almeida E Ambrogini L Han S C Manolagas andR L Jilka ldquoIncreased lipid oxidation causes oxidative stress

increased peroxisome proliferator-activated receptor-120574 expres-sion and diminished pro-osteogenicWnt signaling in the skele-tonrdquo The Journal of Biological Chemistry vol 284 no 40 pp27438ndash27448 2009

[140] C-L Kao L-K Tai S-H Chiou et al ldquoResveratrol promotesosteogenic differentiation and protects against dexamethasonedamage in murine induced pluripotent stem cellsrdquo Stem Cellsand Development vol 19 no 2 pp 247ndash257 2010

[141] S W Lane S de Vita K A Alexander et al ldquoRac signaling inosteoblastic cells is required for normal bone development butis dispensable for hematopoietic developmentrdquo Blood vol 119no 3 pp 736ndash744 2012

[142] M-K Kwak K Itoh M Yamamoto T R Sutter and T WKensler ldquoRole of transcription factor Nrf2 in the induction ofhepatic phase 2 and antioxidative enzymes in vivo by the cancerchemoprotective agent 3H-1 2-dimethiole-3-thionerdquo Molecu-lar Medicine vol 7 no 2 pp 135ndash145 2001

[143] M K Cho W D Kim S H Ki et al ldquoRole of G12057212and G120572

13

as novel switches for the activity of Nrf2 a key antioxidativetranscription factorrdquo Molecular amp Cellular Biology vol 27 no17 pp 6195ndash6208 2007

[144] M Daugaard R Nitsch B Razaghi et al ldquoHace1 controlsROS generation of vertebrate Rac1-dependent NADPH oxidasecomplexesrdquo Nature Communications vol 4 article 2180 2013

[145] A Y Chi G B Waypa P T Mungai and P T SchumackerldquoProlonged hypoxia increases ros signaling and RhoA activa-tion in pulmonary artery smooth muscle and endothelial cellsrdquoAntioxidants and Redox Signaling vol 12 no 5 pp 603ndash6102010

[146] D Kondrikov R B Caldwell Z Dong and Y Su ldquoReactiveoxygen species-dependent RhoA activation mediates collagensynthesis in hyperoxic lung fibrosisrdquo Free Radical Biology andMedicine vol 50 no 11 pp 1689ndash1698 2011

[147] A S Nimnual L J Taylor and D Bar-Sagi ldquoRedox-dependentdownregulation of Rho by Racrdquo Nature Cell Biology vol 5 no3 pp 236ndash241 2003

[148] S Versari G Longinotti L Barenghi J A M Maier and SBradamante ldquoThe challenging environment on board the Inter-national Space Station affects endothelial cell function by trig-gering oxidative stress through thioredoxin interacting proteinoverexpression the ESA-SPHINX experimentrdquo The FASEBJournal vol 27 no 11 pp 4466ndash4475 2013

[149] T P Stein ldquoSpace flight and oxidative stressrdquo Nutrition vol 18no 10 pp 867ndash871 2002

[150] I Testard M Ricoul F Hoffschir et al ldquoRadiation-inducedchromosome damage in astronautsrsquo lymphocytesrdquo InternationalJournal of Radiation Biology vol 70 no 4 pp 403ndash411 1996

[151] M FukataMNakagawa andKKaibuchi ldquoRoles of Rho-familyGTPases in cell polarisation and directionalmigrationrdquoCurrentOpinion in Cell Biology vol 15 no 5 pp 590ndash597 2003

[152] C D Lawson and K Burridge ldquoThe on-off relationship of RhoandRac during integrin-mediated adhesion and cellmigrationrdquoSmall GTPases vol 5 no 1 Article ID e27958 2014

[153] TMitsuharaM Takeda S Yamaguchi et al ldquoSimulatedmicro-gravity facilitates cell migration and neuroprotection after bonemarrow stromal cell transplantation in spinal cord injuryrdquo StemCell Research andTherapy vol 4 no 2 article 35 2013

[154] P A Plett R Abonour S M Frankovitz and C M OrschellldquoImpact of modeled microgravity onmigration differentiationand cell cycle control of primitive human hematopoietic pro-genitor cellsrdquo Experimental Hematology vol 32 no 8 pp 773ndash781 2004


[155] A Espinosa-Jeffrey P M Paez V T Cheli V Spreuer IWanner and J De Vellis ldquoImpact of simulated microgravity onoligodendrocyte development implications for central nervoussystem repairrdquo PLoS ONE vol 8 no 12 Article ID e76963 2013

[156] K Burridge and K Wennerberg ldquoRho and Rac take centerstagerdquo Cell vol 116 no 2 pp 167ndash179 2004

[157] A Higashibata M Imamizo-Sato M Seki T Yamazaki andN Ishioka ldquoInfluence of simulated microgravity on the activa-tion of the small GTPase Rho involved in cytoskeletal forma-tionmdashmolecular cloning and sequencing of bovine leukemia-associated guanine nucleotide exchange factorrdquo BMC Biochem-istry vol 7 article 19 2006


[159] K Hamamura G Swarnkar N Tanjung et al ldquoRhoA-mediatedsignaling in mechanotransduction of osteoblastsrdquo ConnectiveTissue Research vol 53 no 5 pp 398ndash406 2012

Research ArticleA Tissue Retrieval and Postharvest ProcessingRegimen for Rodent Reproductive Tissues Compatible withLong-Term Storage on the International Space Station andPostflight Biospecimen Sharing Program

Vijayalaxmi Gupta1 Lesya Holets-Bondar1 Katherine F Roby23

George Enders2 and Joseph S Tash1

1Department of Molecular amp Integrative Physiology University of Kansas Medical Center Mail Stop 3050 3901 Rainbow BoulevardHLSIC 3098 Kansas City KS 66160 USA2Department of Anatomy and Cell Biology University of Kansas Medical Center Kansas City KS 66160 USA3Institute for Reproductive Health and Regenerative Medicine University of Kansas Medical Center Kansas City KS 66160 USA

Correspondence should be addressed to Joseph S Tash jtashkumcedu

Received 6 June 2014 Revised 18 September 2014 Accepted 20 October 2014


Copyright copy 2015 Vijayalaxmi Gupta et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Collection and processing of tissues to preserve space flight effects from animals after return to Earth is challenging Specimensmust be harvestedwithminimal time after landing tominimize postflight readaptation alterations in protein expressiontranslationposttranslational modifications and expression as well as changes in gene expression and tissue histological degradation aftereuthanasia We report the development of a widely applicable strategy for determining the window of optimal species-specific andtissue-specific posteuthanasia harvest that can be utilized to integrate into multi-investigator Biospecimen Sharing Programs Wealso determined methods for ISS-compatible long-term tissue storage (10 months at minus80∘C) that yield recovery of high qualitymRNA and protein for western analysis after sample return Our focus was reproductive tissues The time following euthanasiawhere tissues could be collected and histological integrity was maintained varied with tissue and species ranging between 1 and3 hours RNA quality was preserved in key reproductive tissues fixed in RNAlater up to 40min after euthanasia Postfixationprocessing was also standardized for safe shipment back to our laboratory Our strategy can be adapted for other tissues underNASArsquos Biospecimen Sharing Program or similar multi-investigator tissue sharing opportunities

1 Introduction

With the current paucity of opportunities for studying wholeanimal mammalian physiology in space flight the Biospec-imen Sharing Program (BSP) for postflight tissue collectionoffers the opportunity to broaden access to biological samplesshortly after return and maximize the data generated fromflight animal payloads The logistics of space flight experi-ments involving live animals often requires harvesting tissuesat a remote site followed by shipping the specimens to thePrinciple Investigatorsrsquo laboratories for detailed analysis Fur-thermore as the capabilities to house rodent andother animalson the International Space station (ISS) and to conduct long-term space flight experiments using animals are enabled the

need to harvest and fix tissues for long-term storage on ISSthat will retain high quality RNA and protein for subsequentanalysis in laboratories on Earth is also required Theseapproaches to live animal experimentation in space flightthat include tissue harvest for multiple investigators requiredetermination of (1) tissue-specific windows of time betweeneuthanasia and tissue fixation that retain quality of histologyand (2) tissue-specific windows of time for tissue fixation thatretain high quality protein and RNA for subsequent analysisDetermination of these windows provides a quantitativelogical approach to generate appropriately prioritized andoptimized tissue harvest and fixation logistics in a multi-investigator Biospecimen Sharing Program scenario be it onEarth or during tissue harvest on the ISS In addition tissue



storage methods should retain high sample quality underlong-term storage as samplesmay be harvested and stored onthe ISS but may not be returned to Earth for many monthsdepending on ISS to Earth flight frequency and payloadcapacities This issue has been addressed for preserving plantmaterial for gene expression analysis [1] but there is nodata available for animal tissues Our participation in theBSP program involved tissue harvest from female mice atKennedy SpaceCenter Florida USA (KSC) for animals flownfor 12ndash15 days in orbit on three space shuttle flights STS-131STS-133 and STS-135 In addition we harvested tissues frommale mice at the Institute for Biomedical Problems (IMBP)laboratory in Moscow RU for animals flown for 30 daysin orbit on the BION M1 satellite Both the STS and BIONseries of flight experiments involved age- and time-matchedground control groups of animals During the early flightplanning phase of BION there were possibilities that maleor female mice would be flown and that male gerbil tissuesmay also be provided to the BSPThus to be prepared for anyof these possibilities we undertook to determine the optimaltissue harvest windows for all of the species and reproductivetissues that we might be able to obtain Our participationin these multi-investigator specimen sharing efforts coveringfour different primary flight PI experimental designs andharvest logistics necessitated a determination of the windowof time between euthanasia and harvest and preservation ofour tissues of interest that would allow us sufficient flexibilityto obtain the highest possible quality of tissue for histopathol-ogy RNA for gene-transcription analysis and protein forexpression and posttranslational modification analysis Adetermination of these time windows is essential since thetissues that are made available to the BSP investigators areprovided after the primary flight PIrsquos have obtained their tis-sues Knowledge of the optimal windows for all of the tissuesof interest aids in the preparation of targeted tissue harvestflow logistics that can provide each of the BSP teammembersthe highest possible quality tissue respectively Therefore aswe report here we developed a strategy to determine thewindowof time between euthanasia and fixation for retentionof high quality histology for male and female reproductiveorgans We also determined methods for long-term tissuestorage for 10 months that provide for recovery of both highquality protein and RNA These strategies are adaptable andcan be applied to harvest and storage of other time-sensitivelabile tissues from animals and plants Furthermore thesemethods can be used to optimize logistics and data collectionundermulti-investigator tissue harvest and sharing programsoperated by any space agency commercial entity or flightplatform


21 Animals Approximately 8 wk old male and femaleC57Bl6J (Jackson Lab Bar Harbor ME) and sim10-month oldmale and female Mongolian gerbils (Charles River Wilming-ton MA) were used throughout this study All animal useprotocols were approved by the University of Kansas Insti-tutional Animal Care andUse Committee (IACUC) Animals

were maintained in standard cages with 12 12 h dark lightcycle and standard food and water were provided ad libitumAll animals were euthanized using CO

2asphyxiation fol-

lowed by cervical dislocation prior to tissue harvest

22 Determination of the Limit of Time andTemperature between Euthanasia and Tissue Fixation

221 Male Mice Our protocol consisted of three groups ofsix mice each with one mouse for each time point In groupA all mice were euthanized at one time and the testes andepididymides were harvested and separated and then imme-diately placed in Hamrsquos F-10 medium (Sigma Aldrich StLouis MO) on ice At time intervals of 0 05 10 15 20 and25 hr each testis (one mouse per time point) and epididymis(2 mice per time point) were separated and then transferredfrom Hamrsquos F-10 medium to Bouinrsquos fixative (Sigma AldrichSt Louis MO) at room temperature (RT 21∘C) In group Bthe same procedure was followed except that the testes andepididymides separated were placed in Hamrsquos F-10 mediumat RT until transfer to Bouinrsquos at the same time interval asabove In group C all mice were euthanized at one time thecarcasses were maintained at (RT) and then at time 0 05 1015 20 and 25 hr the testis and epididymis were harvestedfrom the carcasses respectively At each of the time pointsabove groupC testes were placed in Bouinrsquos solution and pro-cessed as per standard histology protocols as detailed belowFor all groups the tissues fixed in Bouinrsquos were processed asdetailed in Section 23 below At the same time point groupC epididymides were processed to obtain cauda sperm toassess sperm motility by computer assisted sperm analysis(CASA) [2] For animals used for collection of cauda epi-didymal sperm formotility analysis two animals were used ateach time point

222 Female Mice Female mice were euthanized (carcassesmaintained at RT as per groupC above) and their ovaries anduteri were harvested at 0 05 10 15 20 25 and 30 hr aftereuthanasia (one animal per time point) At the times indi-cated the tissues were fixed at RT in Bouinrsquos overnight andthen processed for histology as detailed below (Section 23)

223 Mongolian Gerbils The six male gerbils were euth-anized at one time and the carcasses were kept at roomtemperature At 0 05 10 15 20 and 30 hr after euthanasiatestis and epididymis were harvested from the carcasses andseparated respectively (one animal per time point) Similarlyall six female gerbils were euthanized and the carcasses werekept at RT Ovaries and uterine horns were harvested fromeach carcass at the same time interval as the males At thetimes indicated above the tissues were placed in Bouinrsquos fixa-tive at RT and then processed for histology as detailed below(Section 23) Sperm were immediately harvested from caudaepididymis as mentioned above and sperm motility analysiswas carried out using CASA as described above [2]

23 Postharvest Processing of Mouse Testicular Tissues Testesand epididymides from the mature mice were harvested as


detailed above Unless indicated otherwise all procedureswere done at RT Tissues were fixed in Bouinrsquos solution for48 h washed in 70 ethanol (ETOH) until the yellow colorof Bouinrsquos disappeared (sim48 hr with frequent changes of 70ETOH and gentle agitation) and divided into four groupsControl group tissue was stored in 70 ETOH until paraffinembedding Rapid transition 70 ETOH was immediatelyreplaced with PBS (Sigma Aldrich St Louis MO) pH 74for one wk and then rapidly replaced with 70 ETOHSlow Re-ETOH only 70 ETOH was immediately replacedwith PBS for one wk and then sequentially replaced at 2 hrintervals with each of 10 30 50 and 70 ETOH Slowrehydration-dehydration 70 ETOH was sequentially (7050 30 10 then PBS at 2 hr each) substituted with PBSfor one wk and then replaced at 2 hr intervals with each of10 30 50 and 70 ETOH To prevent contaminationwe kept tissue in PBS at 4∘C and ETOH replacement wascompleted at RT with 2 hr intervals between changes asdetailed above After each of the respective final dehydrationsteps above the tissues were paraffin-embedded and pro-cessed for histology and hematoxylin and eosin staining (HE)using standard methods as performed previously [3]

24 Total RNA Extraction and Preservation Freshly harves-ted mouse testes were immediately stabilized in 10 volumesof TRIzol reagent (Invitrogen Carlsbad CA) or RNAlatersolution (Ambion Austin TX) The samples placed inRNAlater were stored for (1) two wks at 4∘C (2) one wk at RTand one wk at 4∘C or (3) two wks at RT Before RNAextraction tissues were retrieved from RNAlater solutionwith sterile forceps and then submerged in TRIzol reagentRNAwas isolatedwithTRIzol reagent according to themanu-facturerrsquos instructions Time of placement into RNAlater wasnoted to determine if RNA quality was related to the durationof window from euthanasia to placement in RNAlater RNAintegrity and quantity were determined using Agilent RNAkit andAgilent Bioanalyzer 2100 (Santa Clara CA) One 120583g ofRNA was subjected to RT PCR with the primers specific formouse GAPDH (Forward 51015840CCTTCATTGACCTCAAC-TAC Reverse 51015840ATGACAAGCTTCCCATTCTC) and inte-rleukin-1alpha (IL-1120572) (Forward 51015840ACTTGTTTGAAGAC-CTAAAG Reverse 51015840GTTTCAGAGGTTCTCAGAG) Pri-mers were designed using online Primer Design Tool Primer3 Uteri and ovaries from STS-135 were harvested in RNAlatersolution and stored at minus80∘C for 10 months Total RNA wasisolated from ovaries and uterine horns with Gene EluteMammalian RNA kit (Sigma Aldrich St Louis MO) PCRproducts were verified on 30 agarose gels using standardprocedures

25 Protein Extraction from Ovaries and Uterus Tissue Pre-served in RNAlater Ovaries stabilized in RNAlater for onewk at RT and two wks at 4∘C were used for protein extrac-tion The tissue was removed from RNAlater briefly rinsedwith ice-cold PBS pH 74 and then homogenized in RIPAbuffer containing protease inhibitor cocktail (all from SigmaAldrich St Louis MO) or ProteoJet Mammalian Cell LysisReagent (Fermentas Pittsburgh PA) Uteri from STS-135

mission mice were preserved in RNAlater for 10 months atminus80∘C prior to processing in RIPA as described above Pro-tein concentration was determined using the DC Assay (Bio-Rad Hercules CA) and 15 120583g protein was electrophoresedunder denaturing conditions on 4ndash15 polyacrylamide geland transferred to nitrocellulose membrane (Bio-Rad Her-cules CA) The membranes were blocked for 1 hr with 5nonfat milk in TBS-T (Tris buffered saline with Tween-20SigmaAldrich St LouisMO) and probedwith 1120583gmL rabbitanti-mouse estrogen receptor alpha (ER120572) antibody (SantaCruz Biotechnology CA) To verify equal loading of proteinsmembranes were stripped and reprobed with a goat anti-120573-actin antibody (Santa Cruz Biotechnology CA) Incubationswith primary antibodies were carried out overnight at 4∘CAfter washing in TBS-T and probing with the correspondinghorseradish peroxidase-labeled secondary antibody (PierceBiotechnology Rockford IL) bound antibodies were iden-tified using AmershamACL PlusWestern Blotting DetectionReagents (GEHealthcare Pittsburgh PA) and luminographyWestern blots were quantitated by densitometric analysis

3 Results

31 Effect of Delayed Processing on Quality of Mouse andGerbil Reproductive Tissue Histology

311 Male Mouse Testes harvested any time between time0 after euthanasia up to 25 hr after euthanasia showedhistological properties comparable to the time 0-harvestedmice For comparative purposes the 0min harvested testesrepresent the control for subsequent time points for eachstorage treatment respectively Histological quality of thetestes was excellent in all postharvest treatment proceduresnamely tissues were kept in the carcass until fixation or fixedin Bouinrsquos immediately or stored in Hams F10 on ice or RTbefore fixation Testis tubules appeared normal in all treat-ment groups compared to the 0min controls with noshrinkage of tissue and complete retention of histologic archi-tectural details (Figures 1(a) 1(b) and 1(c))

Total motility and progressive motility of cauda epididymalmouse sperm at times 0 05 10 15 20 and 25 hr were notsignificantly different between any of the tissue harvest timepoints and also not significantly different between the threetissue harvest scenarios Table 1 shows the percent progressivemotility for each time point under all three tissue storageconditions Table 2 shows the percent total motility for eachtime point under all three tissue storage conditions Based oncomparable results between the three tissue harvest regimenswe focused on the ldquotissue in carcassrdquo at RT regimen forsubsequent experimentswith femalemice aswell asmale andfemale gerbils

312 Male Gerbil Testis showed normal distinct histologicaldetails with all spermatogenic cells arranged in a normal pat-tern in the tubule when collected up to 25 hr after euthanasia(Figure 2) Total motility of the gerbil cauda epididymalsperm harvested at each at time point was analyzed and ispresented in Table 3 Since the data represent a single animal


(a) (b)

(c)

Figure 1 Effect of delayed processing on C57Bl6J testicular morphology Each panel represents light microscopy (40x objective) of sectionsof adult mouse testis stained with hematoxylin and eosin (HE) (magnification bar is 100 120583m) (a) Tissues kept on ice for 0 to 25 hr afterharvesting (b) tissues allowed to remain in the carcass for 0 to 25 hr (c) tissues removed and kept at RT for 0 to 25 hr in Hamrsquos F10 afterharvesting


Figure 2Gerbil testicularmorphology (40x objective) at various time points after tissue harvest from the carcass (magnification bar is 50 120583m)All gerbils were euthanized at once and tissues were harvested from the carcass at 05 hr interval from 0 to 25 hr HE staining demonstratedthe retention of histological features at every time point

Table 1 Percent progressivemotility (plusmnSD) ofmouse cauda epididy-mal sperm after exposure of epididymis to various conditions



0 h 23 plusmn 18 25 plusmn 26 26 plusmn 1905 h 28 plusmn 13 22 plusmn 13 27 plusmn 1210 h 31 plusmn 3 32 plusmn 19 31 plusmn 915 h 27 plusmn 6 47 plusmn 9 30 plusmn 1120 h 26 plusmn 7 18 plusmn 9 28 plusmn 1425 h 26 plusmn 9 23 plusmn 12 25 plusmn 16lowastTissue was submerged in Hamrsquos F-10 medium in a 15mL tissue culture tubewhich was placed on ice or at room temperature Values are mean plusmn standarddeviation (119899 = 2mice at each time point)There was no significant differencein motility between the three testing conditions at each time point

statistical analysis cannot be done Given the variation inmotilitywith time the data suggest thatmotilitywas relativelystable at all time points except with perhaps a drop at 25 hr

313 Female Mouse The ovaries (Figure 3(a)) and uteri(Figure 3(b)) harvested from female mice up to 3 hr aftereuthanasia showed excellent histological properties devoid ofapparent tissue degradation

314 FemaleGerbil Gerbil ovaries (Figure 3(c)) harvested upto 1 hr after euthanasia showed normal healthy follicles and

Table 2 Percent total motility (plusmnSD) of mouse cauda epididymalsperm after exposure of epididymis to various posteuthanasiaconditions



0 h 48 plusmn 18 49 plusmn 29 45 plusmn 2105 h 51 plusmn 12 47 plusmn 11 50 plusmn 1510 h 53 plusmn 6 60 plusmn 21 40 plusmn 915 h 57 plusmn 12 67 plusmn 16 54 plusmn 1120 h 51 plusmn 14 46 plusmn 14 43 plusmn 1425 h 50 plusmn 17 49 plusmn 18 52 plusmn 12lowastTissue was submerged in Hamrsquos F-10 medium in a 15mL culture tube whichwas placed on ice or at room temperature as indicated Values are mean plusmnstandard deviation (119899 = 2mice at each time point) There was no significantdifference inmotility between the three testing conditions at each time point

Table 3 Percent total motility of gerbil cauda epididymal spermrecovered from the epididymis after storage in the carcass at RT forthe times indicated (119899 = 1 at each time point)

Time aftereuthanasia

Total motility()

Progressivemotility ()

0 h 983 85005 h 824 58010 h 826 61015 h 952 86220 h 879 73225 h 756 475Since we used one animal per time point standard deviation could not bedetermined


(a)

(b)

(c)

(d)

(e)Figure 3 Mouse and gerbil ovarian and uterine horn histology HE staining was used to evaluate quality of oocyte and follicles Rowsrepresent HE staining of (a) mouse ovary up to 3 hr after euthanasia (4x objective magnification bar is 500120583m) (b) mouse uteri up to3 hr after euthanasia (10x objective magnification bar is 500120583m) (c) gerbil ovary up to 15 hr after euthanasia (4x objective magnificationbar is 500 120583m) (d) significantly high number of vacuoles are indicated in the yellow circles in gerbil ovaries from 15 h after euthanasia (40xobjective magnification bar is 100 120583m) (e) gerbil uteri up to 25 hr after euthanasia (10x objective magnification bar is 500 120583m)

healthy oocytes devoid of signs of tissue degradation how-ever at 15 hr after euthanasia high numbers of unhealthy fol-licles and shrunken oocytes were seen suggesting tissue dete-rioration due to the delay in fixation process after euthanasiaSignificantly high numbers of ldquovacuoles-likerdquo structures werealso seen at 15 hr after euthanasia (Figure 3(d)) which isindicative of tissue degradation (indicated by the yellow cir-cles) Control (0min) gerbil ovaries had negligible ldquovacuolesrdquo(indicated by two black arrows) The ovary harvested after

15 hr did not sustain the histological processing as they weretoo fragile and degraded implying that gerbil ovaries weremore sensitive and should be harvested within 1 hr of euth-anasia Uterine histology indicated tissues were intact andcomparable to control at all tested time points (Figure 3(e))

32 Effect of Different Postfixation Procedures on Quality ofTesticular Histology Testicular and epididymis morphology


60x

(a)

60x

(b) (c)

(d)

Figure 4 Morphological analysis of mouse testis after different postfixative manipulations Sections of adult mouse testis were stained withHE (all at 60x magnification bar is 40 120583m) (a) Control (b) slow rehydration-dehydration stepwise replacement of ETOH-PBS-ETOH (c)slow Re-ETOH only stepwise replacement (d) rapid ETOH-PBS-ETOH transition Sertoli cell (Se) spermatogonia (Sp) spermatocytes (Sc)spermatids (Sd) and Leydig cell (L) Black arrowsmdashabnormal open spaces in seminiferous epithelium yellow arrowsmdashabnormal wavy andthinner basement membrane

was evaluated for histological changes after different post-fixation processing (Figures 4 and 5) In control testiculartissue (Figure 4(a)) all types of spermatogenic cells sper-matogonia (Sp) Sertoli cells (Se) spermatocytes (Sc) andspermatids (Sd) were evident Lymphatic spaces betweenseminiferous tubules and adjacent to Leydig cells (L) clustersare clearly defined After the slow rehydration-dehydrationstepwise replacement with ETOH testis tubules appearednormal with histological architecture similar to the controlgroup (Figure 4(b)) Although all types of spermatogenic cellwere identified after slow rehydration-dehydration stepwisereplacement spermatogonia and Sertoli cell nuclei weremoredifficult to distinguish compared to control The quality oftesticular histology observed after slow Re-ETOH only (Fig-ure 4(c)) was similar to that observed in the slow rehydration-dehydration stepwise replacement group However openspaces (black arrows) were observed within portions ofseminiferous tubules in the slow Re-ETOH only group andthe seminiferous tubule basement membrane in this groupoften appeared wavy and thinner (yellow arrows) comparedto the control (Figure 4(a))The histologic quality was poor inthe group treated by single step (rapid transition) change ofsolution (Figure 4(d)) as evidenced by a diffuse appearance ofthe tissue (not due to focus) and limited clarity of nucleardetails in spermatogonia and Sertoli cells In addition theseminiferous tubule basement membrane was occasionallyindistinct and some spermatid artifactual loss of residualbodies is also observed

We found epididymal tissue to be sensitive to dehy-drationrehydration shock (Figure 5) Slow rehydration-dehydration ETOHreplacement had no visible negative effect

on the quality of epididymal morphology (Figure 5(b)) how-ever histological examination revealed differences in epi-didymal morphology after slow Re-ETOH only (Figure 5(c))and rapid transition procedures (Figure 5(d)) compared tocontrol (Figure 5(a)) The differences included alterations inthe thickness of columnar epithelium and basal and principalcells are not very sharp The slow rehydration-dehydrationcaused less destruction of testicular and epididymal tissuethan the rapid single-step changes of solution and has anoverall better morphological detail preservation compared tothe rapid ETOH-PBS solution change

33 Total RNA and Protein Evaluation after TissuePreservation in RNAlater under Long-Term Storage

331 RNA Stability We compared RNA integrity afterpreservation testicular tissue in TRIzol reagent and RNAlater(Figure 6(a)) Total RNA integrity analysis demonstrated highquality RNA after storing tissue in RNAlater for 1-2 wk at RTor 4∘C compared to TRIzol preservation Distinct 28S and 18Sribosomal RNA and absence of degraded RNAwere observedon the gel RT-PCR analysis of RNAwith primers forGAPDHand IL-1120572 indicated high quality of expected PCR products inall analyzed samples (Figure 6(b)) Integrity analysis of totalRNA isolated from STS-131 ground controls (G10 G11 andG12) and flight (F10 F11 and F12) mouse uteri after 30 to40min after euthanasia demonstrate high RNA quality inthese samples (Figure 6(c)) We found no differences in RNAquality between all analyzed uteri and ovaries samples placedinto RNAlater from 15min to 40min after euthanasia


40x

(a)

40x

(b) (c)

(d)

Figure 5 Morphological analysis of mouse epididymis after different postfixative modifications Sections of adult mouse epididymis werestained with HE (all at 40x magnification bar is 50 120583m) (a) Control (b) slow rehydration-dehydration stepwise replacement of ETOH-PBS-ETOH (c) slow Re-ETOH only replacement (d) rapid ETOH-PBS-ETOH transition Epithelium (E) sperm (S) basal cell (B) and principalcell (P)

332 Long-Term Storage for RNA Analysis Using ovary anduteri harvested from STS 135 mice we also determined theeffect of long-term (10 months) preservation in RNAlater onRNA and protein quality (Figure 7) This time frame waschosen to mimic a possible storage scenario that could occuron the ISS Ovaries and uterus from STS-135 ground controlsstabilized in RNAlater for 10 months showed excellent RNAquality (Figure 7(a)) and yield in range 6ndash8 120583g

333 Protein Stability Although not commonly done tissuestabilized in RNAlater can be used for subsequent proteinextraction Protein obtained from samples stored inRNAlateris suitable for western blotting or 2D gel electrophoresis butnot for applications that require native protein (AmbionGuideline for RNAlater) We optimized the protocol for pro-tein extraction from ovaries preserved in RNAlater for onewk at RT or two wks at 4∘C by using RIPA buffer or ProteoJetLysis reagent (LR) Western analysis of 120573-actin integritydemonstrated that one wk ambient storage of testicular tissuein RNAlater did not affect actin integrity as evidenced by theabsence of proteolytic fragments and consistent signal inten-sity in replicate samples (Figure 6(d))

334 Long-Term Storage for Protein Analysis Protein fromuteri stabilized in RNAlater for 10 months after STS-135mission was extracted with RIPA buffer Immunoblot forER120572 and 120573-actin verified excellent expression levels of bothproteins with no evidence of proteolytic degradation of bothproteins (Figure 7(b)) Quantitative densitometry analysis ofwestern blots indicated reduced levels ER120572 in mostly flight

animals These results demonstrate that RNAlater is aneffective sample collection and stabilization reagent for pro-tecting both RNA and protein under long-term conditionscompatible for the ISS

4 Discussion

We report here a logical method to determine the optimaltime window of tissue harvest and fixation after euthanasiafor use in multi-investigator tissue harvest programs that iscompatible with processing of tissue sample obtained fromspace flight animals We also demonstrate here a long-termstorage regimen for animal tissues compatible with recoveryof high quality RNA and protein under conditions similar tothat on the ISS when there may be many months betweensample collection and return to Earth The procedures setforth also include methods for tissue harvest at the site ofreturn and for safe shipment to external laboratories forfurther processing for histopathology and recovery of proteinand RNA Several fixationpreservation studies have beencarried out for plant samples [1 4 5] One European SpaceAgency report included a brief discussion on fixation ofmammalian cells in tissue culture for microscopy [6] Freidinet al [7] have demonstrated significant alterations on geneexpression in lung carcinoma tissues collected about 30minutes after harvest Durrenberger et al [8] reported thatin human brain samples collected from several brain banksantemortem events appeared to negatively affect the RNAquality but postmortem delays caused no significant dete-rioration This observation supported earlier report that


Ladd

er

1 2 3

(s)

70

65

60

55

50

45

40

35

30

25

20

4

120573-Actin

RIPA-R RIPA-4∘C LR-RT LR-4∘C

IL-1120572 GAPDH1 2 3 4 1 2 3 4

Ladd

er

G10

ut

G11

ut

G12

ut

F10

ut

F11

ut

F12

ut

4000

2000

1000

500

200

25

(nt)

(a)

(b) (d)

(c)

Figure 6 Effect of different extraction and storage methods on RNA and protein quality in ovaries and testes extracted andor stabilized inTRIzol or RNAlater (a) Total RNA integrity analysis Total RNA was isolated from mouse testis and stabilized in TRIzol reagent (RNA yieldand quality control (lane 1) RNAlater for 2wks at 4∘C (lane 2) RNAlater for 1 wk at room temperature and 1 wk at 4∘C (lane 3) RNAlater for2 wks at RT (lane 4) (b) Agarose gel electrophoresis of PCR products with primers for IL-1120572 and GAPDH For RT PCR total RNA was usedafter stabilization in TRIzol reagent (lane 1) RNAlater for 2wks at 4∘C (lane 2) RNAlater for 1 wk at RT and 1wk at 4∘C (lane 3) RNAlater for2 wks at RT (lane 4) (c) Total RNA integrity analysis STS-131 ground (G10 G11 and G12) and flight (F10 F11 and F12) uteri fixed in RNAlaterafter 30ndash40min after euthanasia (d) Comparison of buffers to remove RNAlater for subsequent western analysis of 120573-actin integrity inovaries Mouse ovaries were stored at RNAlater for 1 wk at RT or 2wk at 4∘C tissues were homogenized in RIPA lysis buffer (lanes 1 2) orProteoJET Lysis reagent (LR) (lanes 3 4) Total cell lysates were prepared and subjected to SDS-PAGE Western for 120573-actin is presented

postmortem delay had negligible effect on RNA quality [9]Human stomach has been described as the tissue showingthe earliest sign of postmortem [10] Presnell and Cinadescribed stomach and pancreas as the earliest human tissuesto deteriorate following death [11] The significance of quickprocessing of histopathological specimen in a clinical settinghas been identified by Rohr et al [12] However we havenot come across similar studies for animal tissues used inbiomedical research Prior to our study reported here therewas a significant knowledge gap in the literature for methodsto process animal tissues compatible for multi-investigatorBiospecimen Sharing Programs for space flight logistical sce-narios For our flight studies using male and female mice onthree different space shuttle flights and the BIONM1 flight itwas critical to determine the optimum conditions of tissueharvest and processing for tissues of our interest namelytestis epididymis ovary and uteri

Space flight studies usually comprise remotely dispersedmulti-investigator collaborationsThus there is a need to ship

tissue samples from the site of collection at the return-to-Earth laboratory facility for initial tissue harvest to the site offinal processing and detailed data collection With respect tocollection of RNA samples TRIzol reagent gives excellentRNA integrity however it is a phenol-based solution andtissue preserved in TRIzol cannot be shipped internationallydue to airline safety restrictions Lyophilization of fresh tissuespecimen has been shown to preserve RNA and protein qual-ity and levels by Wu et al [13] Though lyophilization makesshipping easier especially across international borders itis not a viable option for preservation of highest qualityhistological analysis To obtain optimum RNA and proteinquality our results demonstrate that mouse testicular tissuecan be submerged in RNAlater and stored successfully foranalysis at a later time point at least 10 months at minus80∘C Itshould be noted that 10 months represents a minimum limitand longer storage intervals would still need to be directlyassessed Given the current estimate for ISS SpaceX Dragonflights at approximately 3-month intervals this would span


(nt)La

dder

G16

OV

G18

OV

G20

OV

G22

OV

G24

OV

G26

OV

G16

utG18

ut

G20

ut

G22

ut

G24

ut

G26

ut

L 1 2 3 4 5 6 7 8 9 10 11 12

4000

20001000

500200

25

(a)

16

14

12

10

08

06

04

02

00

G16

G18

G20

G22

G24

G26

G28

G32

G34

G40

G42

G44

F46

F50

F52

F54

F56

F58

F60

F62

F64

F66

F68

F70

Ground control group Flight group

Rela

tive E

R120572120573

-act

in p

rote

in ex

pres

sion

G1618 20 22 24 26 G28 32 38 40 42 44 F46 F6050 52 54 56 58 62 64 66 68 70

ER120572

120573-Actin

(b)

Figure 7 RNA and protein quality of STS-135 uteri and ovaries stabilized in RNAlater for 10 months at minus80∘C (a) STS-135 ground control(G16ndashG26) ovarian and uterine RNA integrity analysis Total RNA was extracted and examined for RNA quality (b) Western blot analysis ofER120572 and actin in STS-135 mouse uterus Total cell lysates were prepared and subjected to SDS-PAGE (15120583glane) Western blot analysis wasperformed using the corresponding antibodies to check expression levels of the proteins Representative Immunoblot (top) and its graphicalpresentation (bottom) Densitometric intensities of specific protein bands were digitally obtained and normalized to 120573-actin

three opportunities for sample return after an experimentis terminated and ensure maintenance of RNA and proteinsample quality It is known that in order to isolate high qualityRNA and protein from mammalian tissue the tissue mustbe processed directly after harvest We determined for ourtissues of interest that excellent RNA stability was achieved ifthe tissue samples were placed into RNAlater up to 40minafter euthanasia RNAlater is a popular reagent that inac-tivates all cellular enzymes including RNAses thus RNAexpression profiles can be preserved in situations whenimmediate RNA isolation is not feasible The tissue can bestored in RNAlater for a long time without nucleic acid deg-radation RNAlater has been used by investigators for col-lection of human tissue [14] and used in RNA expressionmicroarrays [15] Our results indicated that RNAlater enableslong-term tissue preservation for RNA and protein extractioncompatible with delayed sample return from flight Thisprocess should be evaluated for use on other tissues to max-imize optimal histology and gene transcription data collec-tion in the primary flight experiments as well as the Biospec-imen Sharing Programs (BSP) investigators Freidin et al [7]have confirmed the significance of RNAlater as a medium topreserve gene expression of lung tissues Tissue processingmethods should be standardized for best storage and analysisof harvested tissues Our methods for tissue fixation long-term storage and recovery of protein and RNA are compati-ble for planned in-flight tissue harvest on ISS

With regard to obtaining the best possible tissue fixationfor histology collected under a multi-investigator Biospec-imen Sharing Program obtaining many tissues of interest

to the participating investigators has to be considered in anintegrated way to accommodate the scientific requirementsof the overarching flight project as well as the natural degra-dation process that tissues undergo as soon as euthanasia hasoccurred Depending on the tissue and species differentwindows of time from euthanasia to fixationmay exist withinwhich histological architecture (as well as RNA and proteinintegrity) is stable In the study design reported here we havedetermined the window of time and temperature for optimalpostharvest maintenance of male and female tissue quality(testis epididymis ovary and uterine horn) and sperm via-bility in mice and gerbils These harvest protocols provide alogical method for integrating the tissue flow logistics forpostflight animals for any project involving multiple inves-tigators Protocols compatible with investigators who requiremore rapid tissue retrieval can be identified and prioritized toensure data preservation We determined that mouse testeswere able to retain excellent histological details when pro-cessed up to 3 hr after euthanasia Sperm motility showedgradual decline with time The authors would reiterate asmentioned in the Results section that sperm motility is ahighly sensitive and variable parameter It is normal to seemajor differences inmotility of sperm obtained fromnot onlyone mouse to another but also within samples obtained fromthe same mouse or gerbil Nonetheless even within the spanof variability seen our results indicate that in case of mousethere was no major difference in sperm motility between anyof the time points whereas in case of gerbil we saw sug-gestions of a slight drop at 25 hr after harvest Future studiesfor flight will require analysis of larger 119899rsquos during definition


phase if gerbils will be used Of the tissues studied hereMongolian gerbil ovaries appear most sensitive to delay inprocessing and require more rapid posteuthanasia process-ing than mouse ovaries Determining optimum conditionsfor tissue handling after harvest is very crucial and can help inmaximizing tissue retrieval form animal models therebymaximizing data output Based on the time-sensitivity inves-tigators may be able to plan the sequence in which the tissuesare harvested starting with the most sensitive tissue to leastsensitive

Finally with regard to the requirement to ship tis-sues fixed for histopathology airline and ground transportproviders (especially international carriers) have specificsafety regulations that prohibit shipment of samples contain-ing many widely used fixatives and preservatives of tissuehistologic integrity Science requirements may present chal-lenges in using alternative fixatives In this regard Bouinrsquosfixative (which contains picric acid) is the best fixative for thetestis if the tissue is to be embedded in paraffin [16] Howeverin space flight experiments safety regulations prevent its useon flight platforms and its presence in tissue samples beingshipped Thus we optimized protocols for fixative removaland storage of tissues in PBS for safe shipping as well asreconstitution protocols for storing tissues in 70 ethanol toretain excellent histology

5 Conclusion

Optimal time frames for harvesting testis epididymis ovaryand uteri without compromising the histological qualitysperm motility and RNA quality have been determinedDifferences in tissue-specific optimal fixation time windowswere noted between mice and gerbils We provide here newmethods for (1) fixative removal and transfer of tissues intoaqueousmedia for safe shipping and (2) reconstitution proto-cols into 70 ethanol that retains excellent histologyWe con-clude that stepwise replacement of ETOH-PBS-ETOHcausedless degradation of histological quality of tissue than a single-step change of solution Our results demonstrate that maleand female mouse reproductive tissues stored in RNAlatersolution were stable and gave high quality RNA and proteinafter 10months of storage atminus80∘CThuswe have determinedmethods for postharvest tissue processing to replace Bouinrsquoswith 70 ethanol for safe shipping across USA and alsoreplace 70 ethanol with PBS to enable shipping of tissuesacross international borders These protocols will facilitateintegration of tissue harvest logistics in multi-investigatorBiospecimen Sharing Programs for optimal tissue histologyand retention of high quality RNA and protein recovery fromanimal tissues on long-term space flight experiments on ISSas well as other flight platforms

Abbreviations

TBS-T Tris buffered saline with 01 Tween 20BSP Biospecimen Sharing ProgramETOH Ethyl alcoholSTS Space Transport System

PBS Phosphate buffered salineISS International Space StationCASA Computer assisted sperm analysisWk WeekRT Room temperatureIACUC Institutional Animal Care and Use CommitteeKSC Kennedy Space CenterRNA Ribonucleic acidCO2 Carbon dioxide



Acknowledgments

This study is supported by NASA Grant NNX09AP04G toJoseph S TashThe authors wish to acknowledge the excellentsupport for the project from Richard Boyle Paula DumarsVera Vizir Gwo-Shing and Kenny Vassigh from AmesResearch Center NASA and from Stacy Engel AshleighRuggles and Ramona Bobber at Kennedy Space CenterFlorida The authors thank Stanton Fernald of the KUMCImaging Core for assistance in preparation of final figures

References

[1] A-L Paul H G Levine W McLamb et al ldquoPlant molecularbiology in the space station era utilization of KSC fixation tubeswith RNAlaterrdquo Acta Astronautica vol 56 no 6 pp 623ndash6282005

[2] J Luo V Gupta B Kern et al ldquoRole of FYN kinase inspermatogenesis defects characteristic of FYN-null sperm inmicerdquo Biology of Reproduction vol 86 no 1 article 22 2012

[3] J S Tash B Attardi S A Hild R Chakrasali S R Jakkaraj andG I Georg ldquoA novel potent indazole carboxylic acid derivativeblocks spermatogenesis and is contraceptive in rats after a singleoral doserdquo Biology of Reproduction vol 78 no 6 pp 1127ndash11382008

[4] M Braun B Buchen and A Sievers ldquoFixation procedure fortransmission electronmicroscopy ofChara rhizoids under mic-rogravity in a Spacelab (IML-2)rdquo Journal of Biotechnology vol47 no 2-3 pp 245ndash251 1996

[5] V D Kern F D Sack N J White K Anderson W Wells andCMartin ldquoSpaceflight hardware allowing unilateral irradiationand chemical fixation in petri dishesrdquo Advances in SpaceResearch vol 24 no 6 pp 775ndash778 1999

[6] F J Medina A Cogoli C Dournon et al ldquoPreservation ofsamples during space experimentsrdquo in Topical Teams in theLife amp Physical Sciences Towards New Research Applications inSpace pp 200ndash208 European Space Agency 2005

[7] M B Freidin N Bhudia E Lim A G Nicholson W OCookson andM FMoffatt ldquoImpact of collection and storage oflung tumor tissue on whole genome expression profilingrdquoJournal ofMolecular Diagnostics vol 14 no 2 pp 140ndash148 2012

[8] P F Durrenberger S Fernando S N Kashefi et al ldquoEffects ofantemortem and postmortem variables on human brainmRNAquality a brainNet Europe studyrdquo Journal of Neuropathology ampExperimental Neurology vol 69 no 1 pp 70ndash81 2010


[9] J F Ervin E L Heinzen K D Cronin et al ldquoPostmortemdelay has minimal effect on brain RNA integrityrdquo Journal ofNeuropathology and Experimental Neurology vol 66 no 12 pp1093ndash1099 2007

[10] A R Thomas A Practical Guide for Making Post-MortemExaminations BiblioBazaar 2009

[11] S E Presnell and S J Cina ldquoPostmortem changesrdquo MedscapeDrugs amp Diseases 2013 httpemedicinemedscapecomarticle1680032-overview

[12] L R Rohr L J Layfield DWallin andDHardy ldquoA comparisonof routine and rapid microwave tissue processing in a surgicalpathology laboratory quality of histologic sections and advan-tages of microwave processingrdquoThe American Journal of Clini-cal Pathology vol 115 no 5 pp 703ndash708 2001

[13] YWuMWuY Zhang et al ldquoLyophilization is suitable for stor-age and shipment of fresh tissue samples without altering RNAand protein levels stored at room temperaturerdquo Amino Acidsvol 43 no 3 pp 1383ndash1388 2012

[14] S R Florell C M Coffin J A Holden et al ldquoPreservation ofRNA for functional genomic studies a multidisciplinary tumorbank protocolrdquo Modern Pathology vol 14 no 2 pp 116ndash1282001

[15] G L Mutter D Zahrieh C Liu et al ldquoComparison of frozenand RNALater solid tissue storage methods for use in RNAexpressionmicroarraysrdquoBMCGenomics vol 5 article 88 2004

[16] L D Russell R A Ettlin H A Sinha and E D Clegg EdsHistological andHistopathological Evaluation of the Testis CacheRiver Press Clearwater Fla USA 1990

Research ArticleLarge Artery Remodeling and Dynamics following SimulatedMicrogravity by Prolonged Head-Down Tilt Bed Rest in Humans

Carlo Palombo1 Carmela Morizzo1 Martino Baluci1 Daniela Lucini2 Stefano Ricci3

Gianni Biolo4 Piero Tortoli3 and Michaela Kozakova1

1Department of Surgical Medical Molecular and Critical Area Pathology University of Pisa 56124 Pisa Italy2Department of Medical Biotechnologies and Translational Medicine University of Milan 20129 Milan Italy3Department of Information Engineering University of Florence 50139 Florence Italy4Department of Medicine Surgery and Health Sciences University of Trieste 34127 Trieste Italy

Correspondence should be addressed to Carlo Palombo carlopalombomedunipiit

Received 16 May 2014 Revised 26 October 2014 Accepted 27 October 2014


Copyright copy 2015 Carlo Palombo et alThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

The effects of simulated microgravity on the static and dynamic properties of large arteries are still mostly unknown The presentstudy evaluated using an integrated vascular approach changes in structure and function of the common carotid and femoralarteries (CCA and CFA) after prolonged head-down tilt bed rest (HDTBR) Ten healthy men were enrolled in a 5-week HDTBRstudy endorsed by the Italian Space Agency (ASI) Arterial geometry flow stiffness and shear rate were evaluated by ultrasoundLocal carotid pulse pressure and wave reflection were studied by applanation tonometry After five weeks of HDTBR CFA showeda decrease in lumen diameter without significant changes in wall thickness (IMT) resulting in an inward remodeling Local carotidpulse pressure decreased and carotid-to-brachial pressure amplification increasedThe ratio of systolic-to-diastolic volumetric flowin CFA decreased whereas in CCA it tended to increase Indices of arterial stiffness and shear rate did not change during HDTBReither in CCA or CFA In summary prolonged HDTBR has a different impact on CCA and CFA structure and flow probablydepending on the characteristics of the vascular bed perfused

1 Introduction

Prolonged head-down tilt bed rest (HDTBR) representsan established experimental model allowing investigatingthe physiologic adaptations to microgravity conditions onthe ground [1] Studies evaluating the effect of simulatedmicrogravity on cardiovascular system have demonstratedthat the prolonged HDTBR is followed by a significantdecrease in left ventricular (LV) mass accompanied by areduction in LV performance [2] Our group has previouslydemonstrated that a reduction in echocardiographic indicesof LV systolic and diastolic performance after a 5-weekperiod of HDTBR does not reflect an impairment in intrinsicmyocardial function but simply an adaptive response tocirculatory unloading [3] Data regarding response of thearterial system to bed rest are less clear Prolonged unloading

has been shown to induce an inward remodeling of thefemoral artery with time-dependent decrease in arterial sizereaching 17 after 52 days of bed rest [4] Eight weeks ofphysical inactivity have been also shown to increase carotidand femoral arterywall thickness andwall-to-lumen ratio [5]Pathophysiologic mechanisms underlying these structuralchanges are supposed to include inactivity-related muscleatrophy associated with a reduced metabolic demand ofthe downstream muscle tissue [6] as well as an impact ofaltered hemodynamic stimuli on the arterial wall It hasbeen demonstrated that arteries are capable to respond tochanges in hemodynamic stimuli (flow and shear rate) andmechanical forces (circumferential and pulsatile stress) bymodification of their geometry [7]However previous studiesdid not provide definite evidence on bed rest induced changesin flow shear rate or wall stress and data regarding impact



of deconditioning on arterial stiffness and wave reflection aresporadic [8] In the present study the common carotid andfemoral arteries were investigated at baseline and after a 5-week HDTBR by an integrated vascular approach allowingevaluating impact of deconditioning on different structuraland functional properties of the arterial system

2 Methods

21 Subjects Ten healthy young volunteers all men meanage 23 plusmn 2 years were enrolled in amultidisciplinary HDTBRstudy endorsed by the Italian Space Agency (ASI) and takingplace at the Orthopedic Hospital Valdoltra Ankaran Slove-nia None of the volunteers was a smoker Medical historyphysical examination laboratory examinations resting andstress ECG and echocardiography have excluded any acuteor chronic medical problem The National Committee forMedical Ethics of the Slovene Ministry of Health (LjubljanaSlovenia) approved the study All participants were informedabout the aim of the investigation the procedures andthe methods and signed a written informed consent formaccording to the Declaration of Helsinki

22 Study Protocol All participants underwent a 5-weekperiod of bed rest in a 6∘ head-down tilt position (HDTBR)During the bed rest period participants were kept strictlyin bed for 24 hours a day and none of them took anymedication or underwent any physical or pharmacologicalcountermeasure Dietary intake was 2300 kcalday and waterintakewas 10ndash15 LdayDiuresiswasmonitored daily andBPand heart rate were measured every 4 hours during daytimeBody composition and hematocrit were measured before andat the end of the bed rest Carotid and femoral ultrasoundcarotid applanation tonometry carotid-femoral pulse wavevelocity (PWV) and cardiac ultrasound were performed theday before entering bed rest and within 24 hours after itstermination Vascular and cardiac examinations were per-formed in a quiet room three hours after a light breakfast andafter an acclimatization period of 30min in supine positionAll vascular acquisitions and readings were performed by asingle operator (CM)

23 Measurements

231 BodyCompositionAssessment Bodyweight and fat-freemass were measured by electrical bioimpedance (BioScan916S Maltron International Ltd Essex UK)

232 Carotid and Femoral Ultrasound On the right com-mon carotid and femoral artery (CCA CFA) two sequentialacquisitions were performed using a modified commerciallyavailable equipment (MyLab30 Esaote Firenze Italy with a75ndash12MHz broadband linear transducer LA435) in orderto obtain the following measures (a) intima-media thick-ness (IMT) systolic diastolic and mean arterial luminaldiameters (b) centerline blood flow velocity (by conventionalDuplex ultrasound) (c) shear rate values directlymeasured atthe near and far arterial wall (by multigate Doppler system)

For all ultrasound acquisitions the angle of inclination forDoppler velocity measurements was consistently adjustedto 60∘ whereas the vessel lumen was set parallel to thetransducer

(a) Longitudinal B-mode images of the right CCA andCFA with well-defined intima-media complex of the nearand far wall were obtained and a loop over 5 cardiac cycleswas stored Brachial pressure and heart rate were mea-sured during loop acquisition (Omron 705 Tokyo Japan)Vascular ultrasound scans were analyzed by the computer-driven image analysis system MIP (Medical Image Pro-cessing Institute of Clinical Physiology CNR Pisa Italy)end-diastolic and end-systolic frames of the CCA or CFAwere selected end-diastolic far-wall IMT and minimumand maximum luminal diameters were measured within aregion of interest Arterial remodeling was assessed as aratio of end-diastolic IMT and luminal radius (IMTradius)where radius was calculated as minimum diameter2 End-diastolic wall stress (kPa) was calculated as follows diastolicBP (in kPa) lowast end-diastolic radiusIMT Delta diameter (Δdiameter) was calculated as the difference betweenmaximumand minimum diameter and the stiffness index beta wascalculated as minimum diameter lowast ln(Systolic BPDiastolicBP)Δ diameter The values reported represent the averageof three cardiac cycles Intraindividual variability of IMTand arterial diameter measurement byMIP in our laboratoryis 48 plusmn 28 and 31 plusmn 19 respectively To estimatemean volumetric flow per beat CCA and CFA diameteraveraged over the entire cardiac cycle was measured from theradiofrequency signal processed by a dedicated software tool(QIMT Esaote EuropeMaastricht Netherlands) in a 1 regionof interest placed at the same area as flow-velocity integral wasmeasured

(b) In spectral Doppler recordings peak systolic anddiastolic velocities as well as systolic diastolic and systo-diastolic flow-velocity integrals weremeasured both for CCAand CFA Resistive index was calculated as follows (peak sys-tolic velocity minus peak diastolic velocity)peak systolic velocitySystolic and diastolic volumetric flows per beat were calcu-lated as systolic and diastolic arterial area (Πlowastdiameter24)multiplied by the corresponding flow-velocity integral Meanvolumetric flow over cardiac cycle was calculated as systo-diastolic flow-velocity integral multiplied by area of luminaldiameter averaged over cardiac cycle as obtained from radio-frequency signal (see above) All values are reported as theaverage of 3 cardiac cycles

(c) Shear rate was assessed by a validated multigateDoppler system determining a flow velocity profile from amatrix of 128-point power spectral densities correspondingto 128 different depths along the Doppler beam [9] A customPC board based on a high-speed digital signal processor wasused to process the quadrature demodulated echo signalsderived from theMyLab30 and to display results in real timeA polynomial least-square fit is applied off-line on the 128experimental velocity points and the resulting profile is usedto evaluate the gradient with respect to radius The localpeak shear rate at the near and far blood-wall interfaces wascalculated


233 Carotid Applanation Tonometry Carotid applanationtonometry was performed on the right CCAusing a validatedsystem (PulsePen Diatecne Milan Italy) [10] Carotid pres-sure waveforms were calibrated according to brachial meanand diastolic pressure as previously described [11] In thecarotid pressure waveform the following parameters weremeasured local systolic BP local pulse pressure and aug-mentation index (AIx) Pulse pressure index was calculatedas local pulse pressure divided bymean BP and pulse pressureamplification as the ratio of brachial to carotid pulse pressure[12] The mean of 3 measurements was used for statisticalanalysis

234 Carotid-Femoral Pulse Wave Velocity Carotid-femoralPWV was measured according to current guidelines [13]using the Complior device (Alam Medical VincennesFrance) Briefly arterial waveforms were obtained transcu-taneously over the right CCA and femoral artery and thetime delay (t) was measured between the feet of the twowaveforms The distance (D) covered by the waves wasestablished as the distance between the two recording sitesPWV was then calculated as D (meters)t (seconds) Themeasurement was performed three times and the mean valuewas used for statistical analysis Simultaneous BP measure-ment was performed at the left brachial artery (OmronKyoto Japan) In our laboratory intraindividual variability ofPWVmeasurement is 45 plusmn 28

235 Cardiac Ultrasound Cardiac ultrasound was per-formed as previously described [3] Stroke volume was mea-sured as a product of aortic area and flow-velocity integralin aortic orifice [14] Flow-velocity integral was obtained alsoin ascending aorta from the suprasternal notch Results onchanges in LV mass performance and loading conditionsobserved in the same study group were previously publishedin detail [3]

24 Statistical Analysis Quantitative data are expressed asmean plusmn sd Paired 119905-test was used to compare the measure-ments obtained before and after HDTBR Linear univariateregression analysis was used to test the relationships betweenbed rest-induced changes in arterial diameter or flow and inFFM or stroke volume Statistical significance was set at avalue of119875 less than 005 Statistical analysis was performed byJMP software version 802 (SAS Institute Inc Cary NorthCarolina USA)

3 Results

During the bed rest period body weight BMI fat FFMand Doppler-derived stroke volume and flow-velocity inte-gral in ascending aorta diminished peripheral BP did notchange significantly and heart rate and hematocrit increased(Table 1)

After 5 weeks of HDTBR no significant changes wereobserved in CCA geometry and stiffness (Table 2) CFAdiameter significantly decreased (minimum diameter by 10plusmn 4) CFA intima-media thickness did not change and

Table 1 Main anthropometric and hemodynamic characteristicsand hematocrit in 10 healthy volunteers before and after HDTBR

Before After 119875

Weight (kg) 75 plusmn 10 73 plusmn 9 lt001BMI (kgm2) 233 plusmn 20 228 plusmn 16 lt005Fat-free mass (kg) 64 plusmn 5 61 plusmn 5 lt00001Hematocrit () 444 plusmn 29 479 plusmn 21 0001Systolic BP (mmHg) 115 plusmn 17 113 plusmn 10 051Diastolic BP (mmHg) 62 plusmn 7 65 plusmn 4 033Pulse pressure (mmHg) 53 plusmn 11 48 plusmn 10 019Heart rate (bpm) 60 plusmn 10 71 plusmn 7 lt0005Stroke volume (mL) 76 plusmn 11 63 plusmn 10 lt001FVI ascending aorta (cm) 217 plusmn 21 192 plusmn 28 001BMI body mass index FFM fat-free mass BP blood pressure FVI flow-velocity integral

therefore the ratio end-diastolic CFA IMTradius increasedand circumferential wall stress decreased (Table 2) Thechanges in CFA minimum diameter showed a trend tocorrelate with changes in fat-free mass (119903 = 049 119875 = 015)CFA beta stiffness index remained unchanged after HDTBR(Table 2)

Responses in flow velocities and volumes differedbetween CCA and CFA In CCA peak systolic and diastolicvelocity did not change significantly during the bed restperiod In CFA both peak systolic and diastolic velocitiesincreased but the increase was higher for diastolic velocityand consequently the resistive index decreased Systolicvolumetric flow per beat remained stable both in CCA andin CFA In contrast diastolic volumetric flow showed a trendto decrease in CCA whereas it increased in CFA (Table 2)Consequently the ratio of systolic-to-diastolic flow in CCAtended to increase while in CFA it significantly decreasedThe relationships between volumetric flow per beat in CCAand stroke volume or ascending aorta flow-velocity integral(estimated by Doppler echocardiography) as well as therelationship between volumetric flow per beat in CFA andstroke volume were tested In CCA the mean and diastolicflow per beat at baseline were strongly related to baselinestroke volume (119903 = 075 119875 = 001 and 119903 = 082 119875 lt 0001)and the changes in mean and diastolic flow per beat duringHDTBR were related to changes in stroke volume (119903 = 070119875 lt 005 and 119903 = 054 119875 = 010) as well as to changes inascending aorta flow-velocity integral (119903 = 078 119875 lt 001and 119903 = 048 119875 = 015) None of these relationships wereobserved for CFA

Wall shear rate at near and far arterial wall did not changeduring HDTBR either in CCA or in CFA (Table 2) In CCAthe mean luminal diameter was positively related to wallshear rate at anterior (119903 = 062 119875 = 005) and posterior wall(119903 = 063 119875 = 005) however this correlation was lost afterthe period of bed rest No relationship between shear rate andluminal diameter was observed for CFA

Carotid femoral PWV and AIx did not change after 5weeks ofHDTBRwhile local carotid pulse pressure and pulsepressure index decreased and pressure amplification index


Table 2 Common carotid artery and common femoral artery structure stiffness and flow before and after HDTBR in 10 healthy volunteers

CCA119875

CFA119875

Before After Before AfterIMT (120583m) 503 plusmn 48 520 plusmn 36 027 515 plusmn 79 523 plusmn 57 058Diameter minimum (mm) 51 plusmn 03 50 plusmn 03 012 74 plusmn 09 67 plusmn 10 lt001Diameter maximum (mm) 59 plusmn 03 58 plusmn 03 009 81 plusmn 10 74 plusmn 10 lt001Δ diameter (mm) 080 plusmn 013 078 plusmn 011 071 074 plusmn 022 072 plusmn 023 072End-diastolic IMTradius 018 plusmn 002 019 plusmn 002 014 014 plusmn 003 016 plusmn 002 lt001End-diastolic wall stress (kPa) 424 plusmn 59 415 plusmn 56 070 607 plusmn 109 549 plusmn 79 005Beta index 32 plusmn 07 29 plusmn 07 029 65 plusmn 19 56 plusmn 21 020Peak velocity systolic (cms) 124 plusmn 25 125 plusmn 22 083 89 plusmn 15 116 plusmn 35 009Peak velocity diastolic (cms) 25 plusmn 5 26 plusmn 6 066 6 plusmn 3 10 plusmn 5 lt005Resistive index 079 plusmn 004 079 plusmn 004 088 094 plusmn 002 091 plusmn 002 005Mean flow per beat (mL) 93 plusmn 16 86 plusmn 13 042 94 plusmn 21 93 plusmn 24 090Systolic flow per beat (mL) 49 plusmn 09 50 plusmn 06 074 78 plusmn 21 71 plusmn 23 037Diastolic flow per beat (mL) 34 plusmn 07 29 plusmn 06 008 15 plusmn 05 19 plusmn 07 005Ratio systdiast flow per beat 15 plusmn 05 17 plusmn 04 007 58 plusmn 20 39 plusmn 06 001WSR peak near wall (sminus1) 524 plusmn 80 575 plusmn 120 029 569 plusmn 177 557 plusmn 197 089WSR peak far wall (sminus1) 460 plusmn 107 494 plusmn 95 049 357 plusmn 52 326 plusmn 69 034IMT intima-media thickness WSR wall shear rate

Table 3 Carotid-femoral pulse wave velocity and carotid pressurewaveform analysis before and after HDTBR in 10 healthy volunteers

Before After 119875

C-F PWV (ms) 69 plusmn 10 69 plusmn 07 053Local SBP (mmHg) 106 plusmn 11 101 plusmn 7 023Local PP (mmHg) 44 plusmn 11 36 plusmn 7 lt005PPI 055 plusmn 011 046 plusmn 009 005AIx 66 plusmn 59 54 plusmn 44 050Pressure amplification 124 plusmn 011 131 plusmn 010 lt005C-F PWV carotid-femoral pulse wave velocity SBP systolic blood pressurePP pulse pressure PPI pulse pressure index AIx augmentation index

increased (Table 3) Changes in hematocrit were not relatedto changes in vascular measures

4 Discussion

The present study compares the response of large elasticand muscular artery to prolonged HDTBR and providessome novel information about arterial mechanics and flowdynamics during deconditioning that are summarized inFigure 1 A complex vascular approach integrating establishedinvestigative modalities with new advanced techniques wasexploited to this purpose

41 Bed Rest Deconditioning and Vascular Geometry In ouryoung healthy volunteers an inward remodeling of femoralartery due to luminal diameter reduction and a diminutionof circumferential wall stress was observed after a 35-day bedrest Carotid geometry on the other hand was not signif-icantly influenced by deconditioning a finding confirming

the differences in response of carotid and femoral artery tobed rest Observed reduction in femoral artery diameter isin agreement with results of the Berlin Bed Rest (BBR) study[4] and may reflect structural andor functional changesextensively discussed in a review paper of Thijssen et al[7] In our study the changes in CFA diameter showed atrend to correlate directly with changes in fat-free massSuch a correlation might suggest that the reduction in CFAlumen reflects a reduced metabolic demand in a downstreammuscle tissue as the gravitational unloading involves bothartery and muscle Yet similar to the BBR study femoralartery volumetric flow did not decrease after bed rest Thisapparent discrepancy could be explained by the fact thatconduit arteries adapt primarily to peak blood flow andoxygen demand during exercise [15]The association betweenconduit artery diameter and muscle work has been suggestedalso in a recent study in which a reduction in femoralartery diameter was demonstrated in subjects wearing amechanical device (HEPHAISTOS) allowing an ldquounloadedorthosisrdquo [16] that is a reduction of muscle work withunchanged gravitational acceleration

In contrast with results of the second BBR study [5]reporting an increment in CCA and femoral artery IMTafter a 60-days bed rest period we did not observe asignificant change in carotid or femoral wall thickness ashorter duration of bed rest in our study could explain thediscrepancy

42 Bed Rest Deconditioning and Blood Flow Previous stud-ies evaluating the effect of unloading on the blood flow inthe lower extremity have produced inconclusive evidenceIn the BBR study [4] the mean blood flow did not changein CFA and superficial femoral artery after bed rest in theHEPHAISTOS study [16] blood flow volume in superficial


5-week HDTBR

mass and work

inward remodeling

of flow in CFA

CCA CFA

Systemic hemodynamics

vasodilation

dArr Lower limb muscle

dArr Metabolic demand

dArr CFA diameter

dArr Wall stress

uArr Diastolic component

dArr Resistive index

dArr Stroke volume

dArr Aortic flow

dArr CCA diastolic flow

dArr Vascular resistance

dArr Sympathetic tone

dArr Wave reflection from peripherydArr Pulsatile component of flow

dArr Local pulse pressuredArr Pulse pressure indexuArr Carotid-brachial pressure amplification

Figure 1 Schematic representation of changes observed at common carotid level at femoral artery level and in central hemodynamics after35-days head-down tilt bed rest in 10 young healthy volunteers

femoral artery remained unaffected by a reduction in mus-cle work while flow velocity increased by 17 in studiesusing plethysmography the blood flow at the arteriolarlevel decreased [17] In a HDTBR study a large portionof blood flow reduction as measured by plethysmographywas observed already after the first day of unloading [18]However plethysmographic and Doppler measurements arehardly comparable

Our study is the first to look separately at the systolicand diastolic flow velocities and volumes both at CCA andCFA levels In CCA the diastolic component of volumetricflow after HDTBR showed trend to decrease and the changesin both diastolic and mean volumetric flow were directlyrelated to changes in stroke volume and in flow-velocityintegral in ascending aorta This observation suggests thatcarotid artery flow simply mirrors the changes occurring inaortic flow In contrast in CFA the diastolic component oflocal blood flow significantly increased and consequentlythe resistive index and the ratio of systolic to diastolic flowvolume decreased This behavior may reflect a decrease inlocal vascular resistance at arteriolar level of the leg Basedupon evidence from previous HDTBR studies a reductionof sympathetic firing to lower limb vessels could explain ourfinding Stout et al reported that during simulated micro-gravity a cutaneous microcirculatory vasodilation is moremarked in the lower than in the upper part of the body beingrelated to a baroreflex-mediated withdrawal of a sympathetictone [19] More recently in healthy volunteers maintained for90 days in HDTBR Ferretti et al demonstrated a significantreduction of the efferent muscle sympathetic nerve activity inthe leg [20]

Wall shear rate describing the tangential force exerted bythe flow stream on the arterial wall did not change duringthe study either in elastic or muscular artery A role of shearrate in arterial diameter control was suggested by a directrelationship between near- and far-wall shear rate and CCAluminal diameter at baseline conditions [21] However such arelationship was not observed at femoral artery level

43 Bed Rest Deconditioning Large Artery Stiffness and Cen-tral Pressure The lack of bed rest induced changes in indicesof either local carotid and femoral stiffness or segmentalaortic stiffness (Table 3) further supports the premise thatthe changes in large artery geometry and flow depend uponfunctional instead of structural vascular changes The signif-icant reduction in the local carotid pulse pressure a goodsurrogate of aortic pressure [22] together with the reductionin the pulse pressure index and the increase in the carotid-to-brachial pressure amplification estimated by means ofcarotid waveform analysis reflect a significant decrease inthe pulsatile component of central pressure compared to thesteady one and possibly a reduction of wave reflection froma vasodilated periphery

5 Study Limitations

This study has several limitations First the population stud-ied is small and consists only of men Second all participantswere young and thus the results do not provide informationon the effect of bed rest on arterial structure and function inolder subjects Third vascular examinations were performed


only one day before and one day after HDTBR consequentlywe could not evaluate a sequence of changes over the bedrest period or after termination of bed rest Fourth plasmaviscosity was not measured so that only wall shear rate butnot wall shear stress could be assessed Furthermore theexperimental model used in this study although establishedfor simulating unloading conditions related to microgravitydoes not allow separating the effects of a reduced muscleactivity from those of a reduced gravitational accelerationFinally AIx values were not adjusted for heart rate Dueto the significant increase in heart rate observed after bedrest we could have overestimated AIx and underestimated areduction in wave reflection

6 Conclusion

An integrated vascular approach combining established andexperimental ultrasound arterial contour wave analysis andpulse wave velocity assessment was exploited to investigatethe adaptation of large arteries to microgravity conditionssimulated by HDTBR Prolonged HDTBR showed a differentimpact on CCA and CFA structure and flow probablydepending on the characteristics of the vascular bed perfusedChanges in CCA blood flow seem to reflect bed rest induceddecrease in stroke volume and aortic flow and did notalter CCA geometry Reduction in CFA luminal diameterand inward remodeling may result from reduced metabolicdemand in a downstream unloaded muscle tissue andchanges in CFA flow may reflect decrease in local vascularresistance secondary to withdrawal of a sympathetic toneObserved changes in systemic hemodynamics that includeddecrease in local pulse pressure and pulse pressure index andincrease in carotid-brachial pressure amplification suggest areduction of wave reflection from a vasodilated periphery(Figure 1) Therefore 5-week HDTBR results in a relativereduction of the pulsatile versus the steady component ofblood flow and arterial pressure possibly reflecting changesin systemic hemodynamics and in sympathetic control of thearteriolar tone

In the prospect of improving the management of subjectsundergoing real microgravity conditions data obtained inthis study confirm the indication to active counter-measure-ments aimed to prevent unloading-related sarcopenia as wellas the possible usefulness of common carotid artery as aldquowindowrdquo to monitor central hemodynamic changes



Acknowledgments

The authors are grateful to Dr Bostjan Simunic Institute ofKinesiology Research University of Primorska Koper andto the personnel at the Valdoltra Orthopaedic Hospital inAnkaran (Slovenia) for their valuable medical assistance andtechnical support This study was partly supported by grants

of the Italian SpaceAgency (ASI) projectsDisorders ofMotorand Cardio-Respiratory Control (DMCR) and Osteoporosisand Muscle Atrophy (OSMA) and a Grant (PRIN 2010-2011)of the Italian Ministry of University and Research (MIUR)

References

[1] J-O Fortrat D Sigaudo R L Hughson A Maillet YYamamoto and C Gharib ldquoEffect of prolonged head-downbed rest on complex cardiovascular dynamicsrdquo AutonomicNeuroscience Basic and Clinical vol 86 no 3 pp 192ndash201 2001

[2] M A Perhonen J H Zuckerman and B D Levine ldquoDeteri-oration of left ventricular chamber performance after bed restldquocardiovascular deconditioningrdquo or hypovolemiardquo Circulationvol 103 no 14 pp 1851ndash1857 2001

[3] M Kozakova EMalshi CMorizzo et al ldquoImpact of prolongedcardiac unloading on left ventricular mass and longitudinalmyocardial performance an experimental bed rest study inhumansrdquo Journal ofHypertension vol 29 no 1 pp 137ndash143 2011

[4] M W P Bleeker P C E de Groot G A Rongen et alldquoVascular adaptation to deconditioning and the effect of anexercise countermeasure results of the Berlin Bed Rest studyrdquoJournal of Applied Physiology vol 99 no 4 pp 1293ndash1300 2005

[5] N T L vanDuijnhovenD J GreenD FelsenbergD L BelavyM T E Hopman and D H J Thijssen ldquoImpact of bed rest onconduit artery remodeling effect of exercise countermeasuresrdquoHypertension vol 56 no 2 pp 240ndash246 2010

[6] P C E de Groot M W P Bleeker and M T E HopmanldquoMagnitude and time course of arterial vascular adaptations toinactivity in humansrdquo Exercise and Sport Sciences Reviews vol34 no 2 pp 65ndash71 2006

[7] D H J Thijssen D J Green and M T E Hopman ldquoBloodvessel remodeling and physical inactivity in humansrdquo Journalof Applied Physiology vol 111 no 6 pp 1836ndash1845 2011

[8] E V Nosova P Yen K C Chong et al ldquoShort-term phys-ical inactivity impairs vascular functionrdquo Journal of SurgicalResearch vol 190 pp 672ndash682 2014

[9] P Tortoli T Morganti G Bambi C Palombo and K V Ram-narine ldquoNoninvasive simultaneous assessment of wall shearrate and wall distension in carotid arteriesrdquo Ultrasound inMedicine amp Biology vol 32 no 11 pp 1661ndash1670 2006

[10] P Salvi G Lio C Labat E Ricci B Pannier and A BenetosldquoValidation of a new non-invasive portable tonometer fordetermining arterial pressure wave and pulse wave velocity thePulsePen devicerdquo Journal of Hypertension vol 22 no 12 pp2285ndash2293 2004

[11] L M Van Bortel E J Balkestein J J van der Heijden-Spek et al ldquoNon-invasive assessment of local arterial pulsepressure comparison of applanation tonometry and echo-trackingrdquo Journal of Hypertension vol 19 no 6 pp 1037ndash10442001

[12] A P Avolio LM Van Bortel P Boutouyrie et al ldquoRole of pulsepressure amplification in arterial hypertension expertsrsquo opinionand review of the datardquoHypertension vol 54 pp 375ndash383 2009

[13] P Boutouyrie and S J Vermeersch ldquoDeterminants of pulsewavevelocity in healthy people and in the presence of cardiovascularrisk factors establishing normal and reference valuesrdquo Euro-pean Heart Journal vol 31 no 19 pp 2338ndash2350 2010

[14] J F Lewis L C Kuo J G Nelson M C Limacher and M AQuinones ldquoPulsed Doppler echocardiographic determinationof stroke volume and cardiac output clinical validation of two


new methods using the apical windowrdquo Circulation vol 70 no3 pp 425ndash431 1984

[15] F A Dinenno H Tanaka K D Monahan et al ldquoRegularendurance exercise induces expansive arterial remodelling inthe trained limbs of healthymenrdquoThe Journal of Physiology vol534 no 1 pp 287ndash295 2001

[16] T Weber M Ducos E Mulder et al ldquoThe specific role ofgravitational accelerations for arterial adaptationsrdquo Journal ofApplied Physiology vol 114 no 3 pp 387ndash393 2013

[17] V A Convertino D F Doerr K L Mathes S L Stein andP Buchanan ldquoChanges in volume muscle compartment andcompliance of the lower extremities in man following 30 daysof exposure to simulated microgravityrdquo Aviation Space andEnvironmental Medicine vol 60 no 7 pp 653ndash658 1989

[18] F Louisy P Schroiff and A Guell ldquoChanges in leg vein fillingand emptying characteristics and leg volumes during long-termhead-down bed restrdquo Journal of Applied Physiology vol 82 no6 pp 1726ndash1733 1997

[19] M S Stout D E Watenpaugh G A Breit and A R HargensldquoSimulated microgravity increases cutaneous blood flow in thehead and leg of humansrdquo Aviation Space and EnvironmentalMedicine vol 66 no 9 pp 872ndash875 1995

[20] G Ferretti F Iellamo P Pizzinelli et al ldquoProlonged head downbed rest-induced inactivity impairs tonic autonomic regulationwhile sparing oscillatory cardiovascular rhythms in healthyhumansrdquo Journal of Hypertension vol 27 no 3 pp 551ndash5612009

[21] S K Samijo J M Willigers R Barkhuysen et al ldquoWall shearstress in the human common carotid artery as function of ageand genderrdquoCardiovascular Research vol 39 no 2 pp 515ndash5221998

[22] T Weber S Wassertheurer B Hametner et al ldquoReferencevalues for central blood pressurerdquo Journal of the AmericanCollege of Cardiology vol 63 no 21 article 2299 2014

Research ArticleSpace Flight Effects on Antioxidant Molecules inDry Tardigrades The TARDIKISS Experiment

Angela Maria Rizzo1 Tiziana Altiero2 Paola Antonia Corsetto1 Gigliola Montorfano1

Roberto Guidetti3 and Lorena Rebecchi3

1Department of Pharmacological and Biomolecular Sciences Universita degli Studi di MilanoVia D Trentacoste 2 20134 Milano Italy2Department of Education and Human Sciences University of Modena and Reggio Emilia Via A Allegri 9 42121 Reggio Emilia Italy3Department of Life Sciences University of Modena and Reggio Emilia Via G Campi 213D 41125 Modena Italy

Correspondence should be addressed to Angela Maria Rizzo angelamariarizzounimiitand Lorena Rebecchi lorenarebecchiunimoreit

Received 10 July 2014 Accepted 22 September 2014


Copyright copy 2015 Angela Maria Rizzo et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

The TARDIKISS (Tardigrades in Space) experiment was part of the Biokon in Space (BIOKIS) payload a set of multidisciplinaryexperiments performed during the DAMA (Dark Matter) mission organized by Italian Space Agency and Italian Air Force in2011 This mission supported the execution of experiments in short duration (16 days) taking the advantage of the microgravityenvironment on board of the Space Shuttle Endeavour (its last mission STS-134) docked to the International Space StationTARDIKISS was composed of three sample sets one flight sample and two ground control samples These samples provided thebiological material used to test as space stressors includingmicrogravity affected animal survivability life cycle DNA integrity andpathways of molecules working as antioxidants In this paper we compared the molecular pathways of some antioxidant moleculesthiobarbituric acid reactive substances and fatty acid composition between flight and control samples in two tardigrade speciesnamely Paramacrobiotus richtersi and Ramazzottius oberhaeuseri In both species the activities of ROS scavenging enzymes thetotal content of glutathione and the fatty acids composition between flight and control samples showed few significant differencesTARDIKISS experiment together with a previous space experiment (TARSE) further confirms that both desiccated and hydratedtardigrades represent useful animal tool for space research

1 Introduction

As the interest in space exploration grows it becomes ofgreat importance to predict and know the response of uni-and multicellular organisms to unfavourable space condi-tions including microgravity This allows us to elaboratethe opportune countermeasures to avoid the risks imposedby space environmental stressors To date many studiesfor understanding physiological biochemical and molecu-lar mechanisms against space stressors are performed onunicellular organisms or cultivated cells of multicellularorganisms [1] Although the experiments on cell culturesare useful it is equally clear that cell cultures representonly the first level of life organization and they cannot be

compared to the response of an entire multicellular livingorganism The use of animals in space research allows us toconduct experiments with organisms characterized by a highlevel of hierarchical biological complexity and physiologicalprocesses comparable to those of humans [2]

Even though animals could be useful models in spaceresearch their use is often limited by the fact that many ofthem need specific rearing bioreactors of large volume [1 3]Tardigrades or water bears are little known and neglectedanimals that allow overcoming this problem Their use inspace research is supported by several reasons (i) they areminiaturized animals (from 200 to 1000120583m in length) thatcan be kept and reared in small facilitiesbioreactors (ii)while having tissues and organs they are simpler than several



other animals having a limited cell number (about 1000)(iii) they can be easily reared under lab conditions (iv)many of them are parthenogenetic often apomictic so clonallineages can be obtained [1 2] Although all tardigrades areaquatic animals they thrive in terrestrial habitats subjectedto periodic desiccation thanks to their ability to enter ahighly stable state of suspended metabolic activity calledanhydrobiosis [4] Entering in this physiological state tardi-grades lose up to 97 of their body water and shrivel into adesiccated structure about one-third of its original sizeWhenrehydrated tardigrades can return to their active metabolicstate in a few minutes to a few hours [4 5] Desiccatedtardigrades can persist in anhydrobiosis for several years anda remarkable resilience to physical and chemical extremeshas been documented [4ndash6] By possessing the abilities towithstand complete desiccation severe cold microgravityvacuum and high levels of ionizing and UV radiationsanhydrobiotic tardigrades fulfill the most important criteriafor tolerating exposure to natural space conditions includingopen space [2]

Tardigrades have already been exposed to space stressorson Low Earth Orbit during the FOTON-M3 mission in2007 with different projects (TARDIS [7] TARSE [1 8]RoTaRad [9]) With the TARSE (Tardigrade Resistance toSpace Effect) project we analyzed the responses of bothdesiccated and hydrated physiological state of the tardigradeParamacrobiotus richtersi to spaceflight conditions withinthe spacecraft [1 8] Microgravity and radiation had noeffect on animal survival and life history traits even thougha higher number of laid eggs a shorter egg developmenttime and a higher number of flight-born juveniles wererecordedwith respect to tardigrades reared onEarth [1 10] Inaddition spaceflight induced in active tardigrades an increaseof glutathione content an increase of glutathione peroxidaseactivity and a decrease of catalase superoxide dismutase andglutathione reductase activities [1] Lastly no change in thio-barbituric acid reactive substances was detected On the basisof these results we developed the new project TARDIKISS(Tardigrades in Space) with the aim to deepen the study ofsurvivorship life history traits and regulation of antioxidantdefences on alive desiccated tardigrades under space stressorsincluding microgravity exposure The flight tardigrades ofthe project TARDIKISS have had a very high survival (morethan 91) and females laid eggs which were able to hatchproducing normal newborns able to reproduce in adulthood[11] In this paper we compared the molecular pathwaysof molecules with antioxidant activity thiobarbituric acidreactive substances and fatty acid composition between flighttardigrades and ground control ones with the final aim toprovide news about the biochemical mechanisms underlyingresistance to space stress conditions


21 TARDIKISS Project The TARDIKISS project was partof the BIOKIS (Biokon in Space) payload a set of multidis-ciplinary experiments in the field of biology and dosimetryperformed in microgravity condition during the DAMA(Dark Matter) mission organized by Italian Space Agency

(ASI) and Italian Air Force in 2011 This mission sup-ported the execution of experiments in short duration (16days) taking the advantage of the microgravity environmenton board of the last mission (STS-134) of Space ShuttleEndeavour docked to the International Space Station (ISS)[11]

TARDIKISSwas composed of three sample sets one flightsample (F) and two ground control samples The formercontrol (temperature control TC) was a postflight controlin which samples were exposed to the temperature profileexperienced by tardigrades the days immediately beforeduring and just after the flight mission the latter (laboratorycontrol LC) was maintained in Modena laboratory for theduration of the flight at constant temperature These samplesprovided the biological material used to test as space stres-sors including microgravity affected animal survivabilitylife cycle DNA integrity and changes of the pathways ofmolecules working as antioxidants

Two anhydrobiotic eutardigrade species were consid-ered namely Paramacrobiotus richtersi (Murray 1911) (Mac-robiotidae) and Ramazzottius oberhaeuseri (Doyere 1840)(Ramazzottiidae) Paramacrobiotus richtersi is the modelspecies already used in the FOTON mission [1] P richtersiwas extracted fromahazel leaf litter (sample codeC3499) it iscarnivorous white in colour and the population here consid-ered is bisexual and amphimictic R oberhaeuseri (Figure 1)was extracted from the lichen Xanthoria parietina (L) ThFr (1860) (sample code C3282) it is herbivorous brownredin colour and the population considered in this study isunisexual and parthenogenetic To extract tardigrades fromtheir substrates leaf litter and lichen were sprinkled with tapwater and after 15min submerged in water for 15min at roomtemperature Later each substrate was sieved (mesh size ofsieves 250 120583mand 37 120583m)under runningwater then animalswere picked up from the sieved sediments with a glass pipetteunder a stereomicroscope

For both tardigrade species animals in desiccated (anhy-drobiotic) physiological state were used To obtain desiccatedspecimens tardigrades were dehydrated in lab under con-trolled air relative humidity (RH) and temperature Afterextraction from their substrates tardigrades were kept inwater for 24 h at 15∘C without any food source Then theywere forced into anhydrobiosis by placing groups of animalson a square (1 cm2) blotting paper with natural mineral water(30 120583L) The paper with tardigrades was initially exposedto 80 RH and 18∘C for 4 h then to 50 RH at 18∘C for4 h in a climatic chamber and finally to 0ndash3 RH at roomtemperature for 12 h [1]

Papers with desiccated tardigrades were stored in twelvesmall plastic Petri dishes (18 cm times 10 cm) enveloped withparafilm and integrated within the Biokon facility (KayserItalia) where a radiation dosimeter for neutrons and i-buttondata logger recorded temperature were also present [11]During the entire flight mission the temperature profile wasrelatively constant ranging from 21∘C to 25∘C [11] while thedose equivalent rates due to space radiation exposure were320 120583Sv (measured by TLD 100 and TLD 700) and 360 120583Sv(measured by TLD 600) [11]


(a) (b)

Figure 1 Micrographs by scanning electronmicroscopy of the tardigrade Ramazzottius oberhaeuseri showing its two physiological states (a)Hydrated and metabolically active specimen (b) Desiccated and metabolically inactive specimen Bars a = 10 120583m b = 5 120583m

22 Biochemical Assays Biochemical assays were performedon desiccated tardigrades comparing F samples with TCsamples

The activities of the enzymes superoxide dismutase(SOD total activity) catalase (CAT) glutathione peroxidase(GPx) and glutathione reductase (GR) were evaluated Thetotal glutathione (GSH) content thiobarbituric acid reactivesubstances (TBARS) and fatty acid composition were alsodetermined as previously described [12]

Substrates and reagents for enzyme determinations wereNAD(P)H DTNB GSH GSSG glutathione reductase andtert-butyl hydroperoxide all of them were purchased fromSigma-Aldrich (St Louis Missouri USA) For each sampleset and each species 6 or 8 (with the exception of SOD)replicates each made up by 10 in toto tardigrades werehomogenized in water on ice with potter using 3 cycles of30 sec each The homogenate was assayed for protein content(according to [13]) and used for enzyme determination Foreach enzyme homogenates were analyzed in duplicate

Briefly the activity of the enzyme superoxide dismutasewas assayed using the method based on NAD(P)H oxidationinhibition (according to [14]) the inhibition of NADPHoxidation by superoxide which was chemically generatedwas measured at 340 nm for 20min in the presence of tissueextracts The incubation mixture included 213 120583L of TDB(triethanolaminediethanolamine 100mM pH 74) 10 120583L ofNADPH 75 120583M 7 120583L of EDTA-MnCl

2(100mMndash50mM)

and 20 120583L of sample or blank One unit of SOD activity wasdefined as the amount of enzyme required to inhibit the rateof NADPH oxidation by 50

To evaluate the activity of catalase samples were assayedby measuring the consumption of H

2O2(according to [15])

Consumption of hydrogen peroxide by the tissue extracts wasdetermined at 240 nm for 1min at 30∘CThe incubation mix-ture included 10 120583L of H

2O2200mM 20 120583L of homogenate

and 170 120583L of Na-phosphate buffer (50mM pH 70) One unitofCATactivitywas defined as the amount of enzyme requiredto catalyze the decomposition of 1mmol of H

2O2minminus1

The activity of the glutathione reductase was assayedfollowing the oxidation of NADPH (according to [16])Briefly GSSG reduction and NADPH consumption werefollowed at 340 nmThe incubation mixture included 5 120583L ofGSSG 125mM 3120583L of NADPH 11mM animal homogenatefrom 20 to 50 120583L and K-phosphate buffer (100mM pH 70)to reach a final volume of 025mL One unit of GR activity

was defined as the amount of enzyme required to catalyze theoxidation of 1 120583mol NADPHminminus1

To evaluate the activity of selenium-dependent glu-tathione peroxidase the enzyme activitywas assayed (accord-ing to [17]) following the decrease in the absorbance at340 nm for 3min which corresponds to the rate of GSHoxidation to GSSG in the presence of NADPH and glu-tathione reductase The incubation mixture included 5 120583L ofGSH 100mM 3 120583L of NADPH 22mM GR 1 unit 5 120583L oftert-butyl hydroperoxide 20mM from 20 to 50 120583L of animalhomogenate and EDTA-K phosphate buffer (3mMndash100mMpH 70) to reach a final volume of 025mL One unit of GPxactivity was defined as the amount of enzyme required tocatalyze the oxidation of 1 120583mol of NADPHminminus1

To measure the total glutathione tardigrades werehomogenized on ice in 5 metaphosphoric acid thehomogenatewas centrifuged at 5000timesg for 10min at 4∘C andthe supernatant was assayed (according to [18]) with someslight modifications Briefly the sulfhydryl group of GSHalso generated from GSSG by adding GR reacts with DTNB(551015840-dithiobis-2-nitrobenzoic acid) and produces a yellow-coloured 5-thio-2-nitrobenzoic acid (TNB) The rate of TNBproduction is directly proportional to this reaction whichin turn is directly proportional to the concentration of GSHin the sample The measurement of the absorbance of TNBat 412 nm provides an accurate estimation of the GSH levelpresent in the sample

To evaluate the thiobarbituric acid reactive substances(TBARS) tardigrade samples standards (from 25 to100 pmol TEP 11-33 tetraethoxypropane) and blanks wereassayed (according to [19]) both before and after inductionof lipid peroxidation by FeSO

4and ascorbic acid TBARS

were determined using a fluorescence spectrophotometer(Carly Eclipse Varian CA USA) at an excitation wavelengthof 517 nm and an emission wavelength of 550 nm For eachsample set (F and TC) and species (R oberhaeuseri and Prichtersi) 2 or 4 replicates were analyzed

To evaluate the fatty acid composition lipids wereextracted from groups of 10 desiccated tardigrades withchloroformmethanol (according to [20]) The total extractwas used for derivatization with sodium methoxide inmethanol 333wv to obtain the fatty acid methylesters(FAME) FAME were injected into a gas chromatograph(Agilent Technologies 6850 Series II) equipped with a flameionization detector (FID) under the following experimental


Table 1 Percentage of fatty acid composition in the tardigrades Paramacrobiotus richtersi and Ramazzottius oberhaeuseri

Fatty acid Paramacrobiotus richtersi Ramazzottius oberhaeuseriTC F TC F

C160 2886 (156) 2941 (353) 2965 (184) 3264 (115)C161 844 (198) 891 (079) 656 (168) 977 (023)C180 1453 (268) 1786 (487) 1622 (176) 1856 (455)C181 1985 (343) 1713 (455) 2145 (171) 2004 (621)C182 n-6 975 (324) 1318 (299) 1211 (659) 1297 (169)C183 n-3 261 (205) 225 (169) 403 (348) 150 (030)C203 n-6 111 (089) 103 (067) 022 (015) 024 (032)C204 n-6 978 (756) 505 (520) 587 (174) 291 (202)C205 n-3 117 (046) 199 (078) 130 (155) 057 (077)C225 n-3 051 (051) 014 (019) 014 (023) 023 (033)C226 n-3 400 (127) 303 (036) 245 (036) 056 (060)lowast

PUFA 2892 (749) 2668 (445) 2612 (325) 1899 (073)lowast

TBARS basal (pmoles120583g proteins) 281 (104) 251 (055) 277 (056) 260 (108)TBARS induced (pmoles120583g proteins) 2606 (365) 2825 (127) 4365 (161) 3291 (258)TC = ground temperature control samples F = flight samples PUFA = polyunsaturated fatty acids TBARS = thiobarbituric reactive substances lowast119875 lt 005 inbrackets SD

conditions capillary column AT Silar length 30m filmthickness 025120583m gas carrier helium temperatures injector250∘C detector 275∘C oven 50∘C for 2min and rate of10∘Cminminus1 until 200∘C for 20min A standard mixture con-taining methyl ester fatty acids was injected for calibrationFor each sample set and species 2 or 4 replicates wereanalyzed

23 Statistical Analysis Data were analyzed with Mann-Whitney test and expressed as mean plusmn SD using the pro-gramme SPSS

3 Results

The results of the enzyme activities in the tardigradesParamacrobiotus richtersi and Ramazzottius oberhaeuseri arealways indicated in relation to 120583g of proteins It is worthnoting that R oberhaeuseri contains a lower amount ofproteins compared to P richtersi (Figure 2)

In both species the comparative analysis of the enzymeactivities and other antioxidant molecules between flight (F)and temperature control samples (TC) showed few significantdifferences (Figures 3 and 4) In particular a significantdecrease (119875 lt 005) of the glutathione reductase activitywas detected in R oberhaeuseri F samples with respect toTC samples (Figure 4(b)) Although not statistically sup-ported in this species a tendency to decrease catalasesuperoxide dismutase and glutathione peroxidase activityand in glutathione content was detected In P richtersi atendency to decrease catalase superoxide dismutase and glu-tathione reductase activities and to increase the glutathioneperoxidase activity was detected Noteworthy differenceswere recorded in the activities of ROS scavenging enzymesbetween the two species

The total percentage fatty acid composition of F and TCsamples is reported in Table 1 In R oberhaeuseri a significant

000

020

040

060

080

100

120


TCFlight

(120583g

prot

eins

aa)

Figure 2 Total protein content in flight and ground temperaturecontrol (TC) samples in the tardigrades Paramacrobiotus richtersiand Ramazzottius oberhaeuseri The bars show the mean with SD

decrease (119875 lt 005) was recorded for the fatty acidC22-6 n-3 and polyunsaturated fatty acids (PUFA) in theF samples with respect to the TC samples Moreover Roberhaeuseri has significantly lower amount of C22-6 n-3compared to P richtersi The amount of thiobarbituric acidreactive substances (TBARS) in tardigrades both before andafter induction of peroxidation in vitro is also reported inTable 1 No differences were detected between F and TCsamples in both species for basal levels and after inductionof peroxidation


0

005

01

015

02

025

03

035

04


Superoxide dismutase (SOD)

TCFlight

(Un

g pr

otei

ns)

(a)

0

10

20

30

40

50

60

70

80


Catalase

TCFlight

(mU

120583 g

prot

eins

)

(b)

Figure 3 Superoxide dismutase (a) and catalase (b) activities in flight and ground temperature control (TC) samples in the tardigradesParamacrobiotus richtersi and Ramazzottius oberhaeuseri The bars show the mean with SD

0

05

1

15

2

25

3


Glutathione peroxidase

TCFlight

(mU

120583g

prot

eins

)

(a)

0010203040506070809

1


Glutathione reductase

lowast

TCFlight

(mU

120583g

prot

eins

)

(b)

0

02

04

06

08

1


Glutathione (GSH)

TCFlight

(mm

oles

120583g

prot

eins

)

(c)

Figure 4 Glutathione peroxidase (a) glutathione reductase (b) and total glutathione content (c) in flight and ground temperature control(TC) samples in the tardigrades Paramacrobiotus richtersi and Ramazzottius oberhaeuseri The bars show the mean with SD lowast119875 lt 005


4 Discussion

Exposure to space stress conditions induces oxidative stressOxidative stress resulting from an imbalance between theexcessive production of reactive oxygen species (ROS) andlimited action of antioxidant defences is implicated inthe development of many important human pathologiesincluding atherosclerosis hypertension inflammation can-cer Parkinson and Alzheimer diseases [21] Oxidative stressmay be highly destructive also in anhydrobiotic organismseven if the lower cellular water content decreases the produc-tion of ROS [21 22] Under normal conditions antioxidantsystems minimize the adverse effects caused by ROS butdesiccation stress could cause the loss or reduction of thesedefence control mechanisms since the metabolic activity isabsent or reduced [21ndash24]

The ability of some animals tardigrades among themto survive extreme desiccation involves a complex arrayof factors working at structural physiological and molec-ular level From a molecularbiochemical point of viewanhydrobiotic organisms synthesize molecules working asbioprotectants during entering permanence and leaving ina desiccated state [25] For example trehalose and sucrosestabilise the biological membrane avoiding protein unfoldingand membrane disturbances late embryogenesis abundantproteins and heat shock proteins work as chaperone systemsrepairing or eliminating damaged molecules while antiox-idant molecules counteract the negative effects of oxidativestress [25]

Since it is known that both hydrated and desiccatedtardigrades have a good natural capability to overcomeoxidative stress [26] they have been used in TARDIKISSexperiments to evaluate the role of antioxidant defence inovercoming oxidative stress induced by exposure to spacestress conditions such as ionizing and UV radiations

The first space experiment (TARSE) conducted withhydrated starved specimens of the tardigrade P richtersidemonstrated that some of the enzymes involved in antiox-idant defences were significantly influenced by the flightstresses [1] In particular there was a significant decrease incatalase and SOD activities the more active enzymes in Prichtersi In addition the glutathione system the less activesystem in not stressed specimens of this species [26] wassignificantly induced during space flight [1] These resultscould be related to the stresses experienced by the hydratedand metabolically active animals (microgravity starvationand radiations) during the flight On the contrary the analysisof antioxidant defences in desiccated tardigrades of theTARDIKISS experiment showed fewer differences relatedto space flight even if the tendency was similar to thatrecorded in hydrated metabolically active animals of theTARSE experiment A similar trend between TARSE andTARDIKISS experiments was also detected in regard to tardi-grade survival since flight animals did not show significantdifferences in survival from temperature laboratory controlones [1 11] Only inR oberhaeuseri (TARDIKISS experiment)a significant decrease in survival rate was recorded between Fand TC samples the species in which a significant decrease ofthe C226 n-3 fatty acid and of glutathione reductase activity

and even though not significant of the activity of the otherROS scavenging enzymes were detected

In conclusion TARDIKISS experiment together withprevious space experiments using tardigrades [1 7ndash9] furtherconfirms that both desiccated and hydrated physiologicalstates of tardigrades represent useful animal tool for spaceresearch To further develop the space research using tardi-grades the setup of experimentswith the possibility to changethe exposition condition of metabolically hydrated animalsas well as the possibility to expose desiccated tardigrades toopen space is necessary Experiments under true space con-dition provide a realistic evaluation of the mechanisms thatcould allowmulticellular organisms including tardigrades tosurvive the combined and synergic effects of space stressorsNevertheless experiments on ground using simulators ofmicrogravity radiation temperature and other space stressesare an essential part of space research complementing exper-iments under true space conditions The comparisons of twodifferent sets of data (ground and space data) will allow betterunderstanding of the physiological and molecular pathwaysof living organisms under space environment



Acknowledgments

The authors are very grateful to the Italian Space Agency(ASI) and the Italian Air Force (AM) which funded theDAMA mission The authors are also very grateful to KayserItalia (KI) which developed and manufactured the hardwareinvolved in the BIOKIS payload They are grateful to anony-mous reviewers for their constructive suggestions

References

[1] L Rebecchi T Altiero R Guidetti et al ldquoTardigrade resistanceto space effects First results of experiments on the LIFE-TARSEMission onFOTON-M3 (September 2007)rdquoAstrobiology vol 9no 6 pp 581ndash591 2009

[2] R Guidetti A M Rizzo T Altiero and L Rebecchi ldquoWhatcan we learn from the toughest animals of the Earth Waterbears (tardigrades) asmulticellularmodel organisms in order toperform scientific preparations for lunar explorationrdquoPlanetaryand Space Science vol 74 no 1 pp 97ndash102 2012

[3] H Marthy ldquoDevelopmental biology of animal models undervaried gravity conditions a reviewrdquo Vie et Milieu vol 52 no4 pp 149ndash189 2002

[4] N Moslashbjerg K A Halberg A Joslashrgensen et al ldquoSurvival inextreme environmentsmdashon the current knowledge of adapta-tions in tardigradesrdquo Acta Physiologica vol 202 no 3 pp 409ndash420 2011

[5] R Guidetti T Altiero and L Rebecchi ldquoOn dormancy strate-gies in tardigradesrdquo Journal of Insect Physiology vol 57 no 5pp 567ndash576 2011

[6] T Altiero R Guidetti V Caselli M Cesari and L RebecchildquoUltraviolet radiation tolerance in hydrated and desiccated


eutardigradesrdquo Journal of Zoological Systematics and Evolution-ary Research vol 49 supplement 1 pp 104ndash110 2011

[7] K I Jonsson E Rabbow R O Schill M Harms-Ringdahl andP Rettberg ldquoTardigrades survive exposure to space in low Earthorbitrdquo Current Biology vol 18 no 17 pp R729ndashR731 2008

[8] L Rebecchi T Altiero M Cesari et al ldquoResistance of the anhy-drobiotic eutardigrade Paramacrobiotus richtersi to space flight(LIFE-TARSE mission on FOTON-M3)rdquo Journal of ZoologicalSystematics and Evolutionary Research vol 49 supplement 1 pp98ndash103 2011

[9] D Persson K A Halberg A Joslashrgensen C Ricci N Moslashbjergand R M Kristensen ldquoExtreme stress tolerance in tardigradessurviving space conditions in low earth orbitrdquo Journal of Zoolog-ical Systematics and Evolutionary Research vol 49 supplement1 pp 90ndash97 2011

[10] T Altiero L Rebecchi and R Bertolani ldquoPhenotypic varia-tions in the life history of two clones of Macrobiotus richtersi(Eutardigrada Macrobiotidae)rdquo Hydrobiologia vol 558 no 1pp 33ndash40 2006

[11] M Vukich P L Ganga D Cavalieri et al ldquoBIOKIS amodel payload for multisciplinary experiments in micrograv-ityrdquo Microgravity Science and Technology vol 24 pp 397ndash4092012

[12] A M Rizzo L Adorni G Montorfano F Rossi and B BerraldquoAntioxidant metabolism of Xenopus laevis embryos duringthe first days of developmentrdquo Comparative Biochemistry andPhysiologymdashB Biochemistry and Molecular Biology vol 146 no1 pp 94ndash100 2007

[13] O H Lowry N J Rosebrough A L Farr and R J RandallldquoProtein measurement with the Folin phenol reagentrdquo TheJournal of Biological Chemistry vol 193 no 1 pp 265ndash275 1951

[14] F Paoletti and A Mocali ldquoDetermination of superoxide dis-mutase activity by purely chemical system based on NAD(P)HoxidationrdquoMethods in Enzymology vol 186 pp 209ndash220 1990

[15] H Aebi ldquoCatalase in vitrordquoMethods in Enzymology vol 105 pp121ndash126 1984

[16] M C Pinto A M Mata and J Lopez-barea ldquoReversible inacti-vation of Saccharomyces cerevisiae glutathione reductase underreducing conditionsrdquo Archives of Biochemistry and Biophysicsvol 228 no 1 pp 1ndash12 1984

[17] J R Prohaska and H E Ganther ldquoSelenium and glutathioneperoxidase in developing rat brainrdquo Journal of Neurochemistryvol 27 no 6 pp 1379ndash1387 1976

[18] O W Griffith ldquoGlutathione and glutathione disulphiderdquo inMethods of Enzymatic Analysis H U Bergmeyer Ed vol 3 pp521ndash529 Academic Press New York NY USA 1984

[19] H E Wey L Pyron and M Woolery ldquoEssential fatty acid defi-ciency in cultured human keratinocytes attenuates toxicity dueto lipid peroxidationrdquo Toxicology and Applied Pharmacologyvol 120 no 1 pp 72ndash79 1993

[20] J Folch M Lees and G H S Stanley ldquoA simple method for theisolation and purification of total lipides from animal tissuesrdquoThe Journal of Biological Chemistry vol 226 no 1 pp 497ndash5091957

[21] M B Franca A D Panek and E C A Eleutherio ldquoOxidativestress and its effects during dehydrationrdquoComparative Biochem-istry and PhysiologymdashA Molecular and Integrative Physiologyvol 146 no 4 pp 621ndash631 2007

[22] R Cruz de Carvalho M Catala J Marques da Silva CBranquinho and E Barreno ldquoThe impact of dehydration rateon the production and cellular location of reactive oxygen

species in an aquatic mossrdquo Annals of Botany vol 110 no 5 pp1007ndash1016 2012

[23] I Kranner and S Birtic ldquoA modulating role for antioxidants indesiccation tolerancerdquo Integrative and Comparative Biology vol45 no 5 pp 734ndash740 2005

[24] R Cornette and T Kikawada ldquoThe induction of anhydrobiosisin the sleeping chironomid current status of our knowledgerdquoIUBMB Life vol 63 no 6 pp 419ndash429 2011

[25] L Rebecchi ldquoDry up and survive the role of antioxidantdefences in anhydrobiotic organismsrdquo Journal of Limnology vol72 no 1 pp 62ndash72 2013

[26] A M Rizzo M Negroni T Altiero et al ldquoAntioxidant defencesin hydrated and desiccated states of the tardigrade Paramac-robiotus richtersirdquo Comparative Biochemistry and Physiology BBiochemistry and Molecular Biology vol 156 no 2 pp 115ndash1212010

Research ArticleIdentification of Reference Genes in Human MyelomonocyticCells for Gene Expression Studies in Altered Gravity

Cora S Thiel123 Swantje Hauschild123 Svantje Tauber123 Katrin Paulsen12

Christiane Raig1 Arnold Raem4 Josefine Biskup12 Annett Gutewort12 Eva Huumlrlimann1

Felix Unverdorben1 Isabell Buttron1 Beatrice Lauber1 Claudia Philpot5 Hartwin Lier6

Frank Engelmann67 Liliana E Layer1 and Oliver Ullrich1238

1 Institute of Anatomy Faculty of Medicine University of Zurich Winterthurerstraszlige 190 8057 Zurich Switzerland2Department of Machine Design Engineering Design and Product Development Institute of Mechanical EngineeringOtto-von-Guericke-University Magdeburg Universitatsplatz 2 39106 Magdeburg Germany3Study Group ldquoMagdeburger Arbeitsgemeinschaft fur Forschung unter Raumfahrt- und Schwerelosigkeitsbedingungenrdquo (MARS)Otto-von-Guericke-University Magdeburg Universitatsplatz 2 39106 Magdeburg Germany4Arrows Biomedical Deutschland GmbH Center for Nanotechnology at the Westfalische Wilhelms-Universitat MunsterHeisenbergstraszlige 11 48149 Munster Germany5German Aerospace Center Space Agency Konigswinterer Straszlige 522-524 53227 Bonn Germany6KEK GmbH Kemberger Straszlige 5 06905 Bad Schmiedeberg Germany7University of Applied Science Jena Carl-Zeiss-Promenade 2 07745 Jena Germany8Zurich Center for Integrative Human Physiology (ZIHP) University of Zurich Winterthurerstraszlige 190 8057 Zurich Switzerland


Received 14 May 2014 Accepted 4 September 2014


Copyright copy 2015 Cora S Thiel et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Gene expression studies are indispensable for investigation and elucidation of molecular mechanisms For the process ofnormalization reference genes (ldquohousekeeping genesrdquo) are essential to verify gene expression analysis Thus it is assumed thatthese reference genes demonstrate similar expression levels over all experimental conditions However common recommendationsabout reference genes were established during 1 g conditions and therefore their applicability in studies with altered gravity has notbeen demonstrated yetThemicroarray technology is frequently used to generate expression profiles under defined conditions andto determine the relative difference in expression levels between two ormore different states In our study we searched for potentialreference genes with stable expression during different gravitational conditions (microgravity normogravity and hypergravity)which are additionally not altered in different hardware systems We were able to identify eight genes (ALB B4GALT6 GAPDHHMBS YWHAZ ABCA5 ABCA9 and ABCC1) which demonstrated no altered gene expression levels in all tested conditions andtherefore represent good candidates for the standardization of gene expression studies in altered gravity

1 Introduction

Since several limiting factors for human health and perfor-mance in microgravity have been clearly identified [1] ithas been concluded that substantial research and develop-ment activities are required in order to provide the basicinformation for appropriate integrated risk managementincluding efficient countermeasures and tailored life supportsystems [2] In particular bone loss during long stays in

weightlessness still remains an unacceptable risk for long-term and interplanetary flights [3] and serious concernsarose whether spaceflight-associated immune system weak-ening ultimately precludes the expansion of human presencebeyond Earthrsquos orbit [4] The immune and skeletal systemsare tightly linked by cytokine and chemokine networks anddirect cell-cell interactions [5 6] and the immune systeminfluences metabolic structural and functional changes inbones directly [6] Both systems share common cellular



players such as the osteoclasts which are bone-residentmacrophages and derivatives of monocytic cells Thereforeknowing the cellular and molecular mechanisms of howgravity influences cell function is a valuable requirementto provide therapeutic or preventive targets for keepingimportant physiological systems fully functional during long-term space missions

Since the first pioneering in vitro studies that revealed thatcells of the immune system are sensitive to changes of grav-itational force [7ndash10] several studies in real and simulatedmicrogravity have confirmed microgravity-induced alter-ations in the molecular mechanisms and signal transductionprocesses in leukocytes including themonocytemacrophagesystem (MMS) [11 12] The MMS belongs to the innateimmune system and is characterized by a fast but nonspecificimmune reaction the first line of defense against invadingpathogens Cells of the MMS in microgravity demonstrateddisturbed cytokine release [13ndash15] reduced oxidative burst[16 17] alteration of the cytoskeleton [18] and reductionin their locomotion ability [19] Importantly analysis ofgene expression of monocytes during an ISS experimentrevealed significant changes in gene expression associatedwith macrophageal differentiation [20]

Differential gene expression analyses are a widely usedmethod to investigate the influence of different treatmentsor conditions on a cell system The resulting changes on themolecular level can be investigated either by reverse tran-scription quantitative real-time PCR (RT-qPCR) as majortechnique for the sensitive and robust analysis of expressionlevels of specific genes [21ndash27] and microarrays for wholegenome or transcriptome analyses [28]

After the genome sequencing era when numerousgenomes were completely decoded the focus of interest shi-fted towards genome wide expression level analyses so that asnapshot of the whole genome expression profile is obtainedin a single experiment [28ndash30] offering also a possibility toobtain an insight into networks and pathways of biomolecularinteractions on a large scale [29ndash33] The technology behindmicroarray analysis developed fast and different suppliersused different protocols for for example hybridization anddata normalization Therefore it was and still is difficultto establish standards for the experimental procedure andprocessing of the raw data obtained [30] Consequently aconcept for the development of standards for microarrayexperiments and data has been presented by the microarraygene expression database group (MGED) describing themin-imum information about a microarray experiment (MIAME[34]) This compilation covers (1) the experimental design(2) the array design (3) samples (4) hybridizations (5)measurements and (6) normalization controls [34] Also forRT-qPCR technique [35ndash38] standard guidelines (MIQE =minimum information for the publication of quantitativereal-time PCR) were developed [27 39ndash41]

One of the most crucial requirements of standardizationare suitable internal controls so called reference genes that areused for data normalization which are important to accountfor differences in the amount and quality of starting materialas well as reaction efficiency [42] GAPDH HPRT 120573-actintubulin and ribosomal RNA genes are typical examples for

frequently used reference genes [43ndash45] However referencegenes have to be tested for their suitability as an endogenouscontrol in each case prior to the experiment This is ofhigh importance substantiated by many studies reportingexpression effects of classical ldquohousekeeping genesrdquo uponexperimental treatments [46ndash49] A selection of severalreference genes used simultaneously can also be a good wayto further increase reliability of the resulting data [50 51] Infact recommendations state to identify three stable referencegenes for each planned assay to assure a reliable outcome[50 52]The identification of stably expressed reference genescan be performed in a pilot study using dedicated algorithmslike geNorm or BestKeeper or a combination hereof wherea minimum of eight potential reference genes are tested andranked according to their stability being an indication fortheir suitability as control genes for normalization [50 51]Candidate reference genes for such a study may be forexample chosen from the literature or from experimentaldata obtained from microarray analysis [27 51]

However common recommendations about referencegenes were established during 1 g conditions and there-fore their applicability in studies with altered gravity condi-tions has not been intensively demonstrated so far Altho-ugh there are numerous publications describing differentialgene expression analyses under simulated and real microgra-vity conditions in various cells types and tissues (suppleme-ntary Table 1 available online at httpdxdoiorg1011552014363575) a systematic research on reference genes stableunder altered gravity conditions has not been published yet

In our study we used microarray analyses to investigatethe differential gene expression in U937 cells a myelomono-cytic human cell line exposed to short-term (20 seconds) andmiddle-term (6 to 7 minutes) microgravity and hypergravityduring parabolic flights and sounding rocket flights twoplatforms commonly used by researchers to investigate theeffects of real microgravity Our experimental goal was toidentify potential reference genes that can be recommendedto the community of gravitational biology for differentialexpression analysis performed with cells of the immunesystem on those two frequently used platforms Thereforewe chose 22 reference genes widely used throughout theliterature and screened our microarray data for these partic-ular genes evaluating their stability for possible applicationas control genes Besides the highly conserved ribosomalRNA genes and others ABC transporter and tRNA genesbelong to evolutionary well-conserved genes as well Sinceribosomal RNA and tRNA genes are not represented onthe array we decided to adhere to tRNA related genes liketRNA synthetases as these play a central role in basal cellularfunctions and should be robustly expressed to ensure cellsurvival Therefore our study comprised published referencegenes ABC transporters and tRNA related genes


21 Cell Culture U937 cells (ATCC CRL15932) originatingfrom a diffuse histiocytic lymphoma displayingmanymono-cytic characteristics were used as a model cell line to inves-tigate the differential gene expression under altered gravity


conditions in monocyticmacrophageal cells U937 cells werecultured in RPMI 1640medium (BiochromMerckMilliporeGermany) supplemented with 10 fetal bovine serum (FBSSuperior BiochromMerck Millipore Germany) 2mM glu-tamine (GibcoLife Technologies Germany) and 100UmLpenicillin as well as 100 120583gmL streptomycin (GibcoLifeTechnologies Germany) Cells were seeded with a densityof 02 times 106 cellsmL and the medium was exchanged every48 hours Cells were harvested by centrifugation at 300 g for5min at room temperature resuspended in fresh mediumand an aliquot was used for an adequate dilution with trypanblue to count the vital cell number Cells were reseeded infresh medium at a concentration of 02 times 106 cellsmL

22 Parabolic Flight Experiments We designed and constru-cted an experiment module suitable to perform cell cultureexperiments with living mammalian cells during parabolicflights on board the Airbus A300 ZERO-G During the 19thDLR parabolic flight campaign (PFC) we focused on theanalysis of differential gene expression in U937 cells consi-dering the different gravity conditions in-flight 1 g 18 gand 0 g Experiments were only performed during the firstparabola to assure that the investigated differential geneexpressions are generated by a direct effect of gravitationalchange and not an accumulated long-term effect Duringthe 19th DLR PFC experiments were reproduced on twoindependent flight days

In search of rapidly responsive molecular alterationsin mammalian cells short-term microgravity provided byparabolic flight maneuvers is an ideal instrument to eluci-date initial and primary effects without the influence andinterference of secondary signal cascades Parabolic flightsprovide 1 g 18 g and microgravity (120583g) with a quality ofapproximately 10minus2 to 10minus3 g For the 19th DLR PFC 1 times 107U937 cells in 10mL medium (RPMI 1640 supplemented with100UmL penicillin 100 120583gmL streptomycin 250 ngmLamphotericin B (GibcoLife Technologies Germany) 2mMglutamine and 2 FBS (ie serum starved)) were filledinto 200mL Nutrimix bags (B Braun Melsungen Germany)and transported from the home laboratory to the preflightpreparation laboratories at the NOVESPACE premises inBordeaux France After arrival cells were destarved byaddition of 08mL FBS per Nutrimix bag and used for theflight experiment on the following day For the flight daythe Nutrimix bags were placed in a solid plastic containerto create a double containment to prevent spillage of fluidsin the aircraft in case of leakage which is strictly prohibitedby the NOVESPACE regulations The rapid preservation ofthe effects of altered gravity on the gene expression in theU937 cells was achieved by injection of 50mL of RLT buffer(Qiagen Germany) a lysis buffer immediately lysing cellsand tissues prior to RNA isolation The 1 g in-flight controlswere performed 5min before the first parabola and the 18 gsample directly before the microgravity phase of the firstparabola The 120583g samples were fixed directly at the end ofthe microgravity phase of the first parabola Samples weretransported to the laboratory immediately after landing 1 gground controls were performed immediately after landing

using the experimental module in the aircraft In total 30samples were obtained during two parabolic flight days 6x 1 gground controls 9x 1 g in-flight controls 6x 18 g and 9x 120583g

23 RNA Isolation after the Parabolic Flight ExperimentsAfter landing of the aircraft and transport of the samples tothe laboratory on site facilities the containers were disassem-bled the Nutrimix bags were gently agitated and the lysedcell solution was filled into a T75 straight neck cell cultureflaskThe cell solutionwas vortexed for 10 sec and passed fourtimes through aOslash 08times 120mmneedle (B BraunMelsungenGermany) fitted to a 50mL syringe 50mL of absoluteethanol was added and precipitates were resuspended byvigorous shaking A valve and a sterile connective piecewere placed on a QIAvac 24 plus vacuum system (QiagenGermany) and an RNAmaxi column (Qiagen Germany) wasattached to the connective piece A vacuum of minus200mbarwas adjusted and the column was loaded with the lysed cellsuspension Then the valve was closed and the column wascentrifuged at 4000 g for 3min 15mL of buffer RW1 (QiagenGermany) was applied for washing membrane bound RNAAfter centrifugation at 4000 g for 7min the flow throughwas discarded and two washing steps with 10mL RPE buffer(Qiagen Germany) followedwith centrifugation at 4000 g for3min and 10min respectively The column bound RNA waseluted by application of 600 120583L of RNase-free water (QiagenGermany) incubation for 1min at room temperature andcentrifugation for 4min at 4000 g The elution step wasrepeated with the first eluate The RNA was transported atapproximately minus150∘C in a Cryo Express dry shipper (CX-100 Taylor-Wharton USA) prepared with liquid nitrogenand stored at minus80∘C until the processing of the RNA for themicroarray analysis

24 Experiments during the TEXUS-49 Sounding Rocket Cam-paign For theTEXUS-49 campaign at ESRANGE (EuropeanSpace and Sounding Rocket Range Kiruna Sweden) U937cells were cultured in the fully installed laboratories on siteCells were seeded with a density of 02 times 106 cellsmL andthe medium was exchanged every 48 hours as describedabove On the launch day cells were visually inspectedharvested counted and pooled to a concentration of 5 times 107cellsmL 05mL of this cell suspension was filled in a sterile3mL plastic syringe shortly before the launch Additionallyone syringe was filled with 03mL of cell culture mediumand another one with 1mL Trizol LS (Life TechnologiesGermany) The three syringes were mounted on a plasticblock with a tubing system connecting them This unit wasfinally integrated into the automatically operated experimentsystem In total 35 of these experiment units were preparedand were kept at 37∘C until the integration into the payloadof the rocket

During the experimental run firstly the 03mL ofmedium as a potential placeholder for an activation solutionand secondly the 1mL of Trizol LS were injected to thecell suspension at defined time points to lyse the cells andpreserve the current status of differential gene expressionThis sequential injection of fluids was performed at 75 secafter launch to monitor the so-called baseline (BL) directly


before the 120583g phase and at 375 sec after launch shortly beforethe end of the 120583g phase A group of 1 g ground controls werekept on ground in the incubator simultaneously to the 120583gsample group

TEXUS-49 consisted of a VSB-30 engine (S-30 solidrocket stage with an S-31 second stage) and of the payloadThe rocket was launched on March 29 2011 at 0601 amfrom the ESRANGE Space Center near Kiruna SwedenDuring the ballistic suborbital flight an altitude of 268 kmand 378 sec of microgravity with a quality of 10minus5 g wereachieved Further parameters include first stage peak thrustacceleration 63 g mean thrust acceleration 503 g burnout at123 sec and engine separation at 136 sec second stage peakthrust acceleration 135 g mean thrust acceleration 730 gburnout at 430 sec yo-yo despin at 560 sec and engineseparation at 590 sec

25 RNA Isolation after the TEXUS-49 Sounding RocketCampaign Directly after landing localization and recoveryof the payload the experiment modules were dismantledand handed over to the scientists The cell suspension wassheared three times with a 20G needle (B BraunMelsungenGermany) and distributed in two 20mL tubes 01mL ofchloroform (Sigma-Aldrich Germany) was added and thehomogenate was vortexed for 15 sec and incubated for 5minat room temperature before a 15min centrifugation step at11000 g and 4∘C The upper phase of both 20mL tubeswas transferred into a 15mL tube and 4mL of RLT bufferas well as 3mL of absolute ethanol was added and mixed4mL of this solution was pipetted on an RNA Midi column(Qiagen Germany) and centrifuged for 30 sec at 3000 g androom temperature The flow through was discarded and theresidual 4mL of RNA solution was loaded on the column andcentrifuged for 5min at 3000 g at room temperature Thenthe columns were washed twice with 25mL of RPE bufferand centrifuged for 2min and 5min respectively at 3000 gat room temperature The RNA was eluted by addition of250120583L RNase-free water (Qiagen Germany) to the columnincubation for 1min at room temperature and centrifugationfor 3min at 3000 g and room temperature The eluate wasloaded again onto the column followed by a 1min incubationand centrifugation for 5min at 3000 g and room temperatureThe isolated RNA was transferred into sterile cryotubes andstored until the return transport at approximately minus150∘Cin a Cryo Express dry shipper (CX-100 Taylor-WhartonUSA) preparedwith liquid nitrogen After arrival in the homelaboratory samples were stored at minus80∘C until the processingof the RNA for the microarray analysis

26 RNA Processing and Microarray Analysis RNA quantityand purity were analyzed spectrophotometrically using aNanoDrop 1000 (Thermo Scientific USA) Isolated RNAsamples were all of high quality with 260280 nm ratiosbetween 19 and 21 The RNA integrity was measured usingan Agilent 2100 Bioanalyzer (Agilent Technologies USA)Only RNA with an RNA integrity number (RIN) gt 87 wasused for the following microarray analysis 400 ng total RNAwas applied to Cy3-labeling with the ldquoLow RNA Input LinearAmplification Kit PLUS One-Colorrdquo (Agilent Technologies)

and hybridized for 175 h to a NimbleGen expressionmicroar-ray (12 times 135000 features) employing the ldquoGene Expres-sion Hybridization Kitrdquo (Agilent Technologies) Afterwardsarrays were washed and scanned by the Microarray ScannerG2505B (Agilent Technologies)

The image files of the scanner were analyzed with theNimbleScan Software 26 using the robustmultiarray analysis(RMA) with the default parameters RMA a probe-levelsummarization method identifies probes that are outliers inthe overall behavior of the expression measured for a givengene The contribution of outlier probes is reduced in thereported gene expression level which has been demonstratedto improve the sensitivity and reproducibility of microarrayresults In addition to screening outlier probes NimbleScansoftwarersquos implementation of RMA [53] used quantile nor-malization and background correction

The normalized microarray data were analyzed usingPartek Genomics Suite 66 Statistical analysis was performedusing the one-wayANOVAand the false discovery rate (FDR)[54] for multiple-testing correction Further the coefficientof variation (CV) expressed in percent was calculated alsoknown as ldquorelative variabilityrdquo It equals the standard devi-ation divided by the mean An integration tool (availableat httpwwwleonxiecomreferencegenephp) [50 51 55]of four algorithms (geNorm NormFinder BestKeeper andthe comparative delta-CT method) was used to evaluate theexpression stability of the reference genes On the basis ofthe resulted rankings from the four algorithms an overallranking of the candidate genes was achieved

27 Statistical Analysis of Selected Genes Genes of interestwere identified and the log 2 values of the measured fluo-rescent intensities returned by the Partek software were backcalculated to linear values Then means of all values of thesame gene generated by different probes were calculated ifat least three values existed excluding outliers Subsequentlystandard deviations were calculated for the means and anunpaired t-test with Welch correction was performed usingExcel 2011 (119905-test tails 2 type 3) to obtain statistical signifi-cance

3 Results

The aim of our study was to identify a group of genes thatshow a stable nonchanging expression profile in immunecells under altered gravity conditions over a time rangeof seconds until several minutes Therefore we performedexperiments on the 19th DLR PFC and the sounding rocketmission TEXUS-49 two platforms that offer microgravitytimes of 20 seconds and 6 minutes respectively During bothmissions U937 cells a model for monocyticmacrophagealcells of the human immune system were exposed to differentgravity conditions for various time periods (see Table 1)During the 19th DLR PFC cells were exposed only to the firstparabola with the following sequence 1 g in-flight control18 g and microgravity (120583g) Cells were subjected to alteredgravity conditions of 18 g and 120583g for 20 seconds in each caseand were immediately fixed and stored cooled until RNAisolation In case of theTEXUS-49 campaign cells underwent


134

12

105

9

75

6

461 3 5 7 9 11 13 15 17 19 21 23 25 27

Sample IDGravity condition

Log

inte

nsiti

es19th DLR PFC

120583gHW

1g18 g

(a)

1 2 3 4 5 6 7 8 9 10 11 12 16 17Sample ID

13 14 15 18

134

12

105

9

75

6

46

Log

inte

nsiti

es

TEXUS-49

Gravity condition120583gHW

BL

(b)Figure 1 Boxplots showing the log expression values of individual microarrays The central line represents the 50th percentile or medianwhereas the upper and lower boundaries of the box display the 75th and 25th percentile respectively The upper and lower bars represent the9th and the 91st percentile Two experimental data sets are displayed (a) 28 microarrays hybridized with samples from the 19th DLR PFC (8x120583g 6x HW 8x 1 g 6x 18 g) and (b) 18 microarrays hybridized with samples originating from the TEXUS-49 campaign (7x 120583g 6x HW 5x1 g) The expression data show an even distribution for the displayed log intensities

Table 1 Gravity conditions 19th DLR PFC and TEXUS-49

Gravity condition 19th DLR PFC TEXUS-491 g ground controls(hardware HW) HW HW

Microgravity 120583g (20 sec) 120583g (378 sec)1 g in-flight control 1 g mdashIn-flight baseline (hyper-gphase directly before 120583gBL)

18 g (20 sec) BL (1 gmdashmax135 g 75 sec)

the following sequence of altered gravity hypergravity up to135 g during the first 75 seconds after liftoff and 120583g for 378seconds Hypergravity is defined as the baseline (BL) becausesamplesmirror the vibration and hypergravity effects directlybefore the microgravity phase In both experimental setupson ground 1 g hardware controls (HW) were performed tobe able to differentiate between the effects caused by theconditions experienced before hypergravity and 120583g and thealtered gravity conditions themselves After the campaignsthe RNA samples were analyzed for quantity and quality byNanoDrop spectrophotometry and a bioanalyzer analysisand only samples with an RNA integrity number (RIN)higher than 87 were chosen for subsequent microarrayanalysis 12 times 135 K Roche NimbleGen arrays were hybridizedand data were collected after the normalization procedure Intotal we obtained data from 46 single microarrays (19th DLRPFC 8x 120583g 6x HW 8x 1 g and 6x 18 g TEXUS-49 7x 120583g6x HW and 5x BL)

Data tables were compiled individually for the 19th DLRPFC and TEXUS-49 including all gravity conditions listedin Table 1 and a first overview of the datasets was providedby a boxplot diagram (Figures 1(a) and 1(b)) Boxplots are auseful tool to visualize the variation within a microarray and

between microarrays The central line shows the position ofthe median while the upper and the lower boundaries repre-sent the upper (75th percentile) and lower (25th percentile)quartile The ends of the tails display the 9th and the 91stpercentile The boxplots of the microarray data show thatthere is only little variation within a single array and betweenthe arrays that belong to the same gravity condition Figure 1shows that the quality of both data sets (19th DLR PFC andTEXUS-49) is sufficient to proceed with further analyses

In search of potential reference genes for gravitationalstudies in this monocyticmacrophageal cell system we firstperformed PubMed database search to identify commonlyused reference genes in RNA expression analyses in humancells We found 22 genes that were used in several reversetranscription quantitative real-time PCR (RT-qPCR) studiesas control genes for normalization (Table 2 supplementaryTable 2) The microarray data tables were screened for these22 widely used reference genes and 20 of them could belocated on the Roche NimbleGen 12 times 135 K array that wasused in our experiments Two genes coding for 5s and 18srRNAs could not be identified since they are not spotted onthe array The PFC and TEXUS data sets were screened forthose 20 selected potential reference genes and fluorescenceintensities were compiled for each gene and each gravitycondition in heatmaps (Figures 2(a) and 2(b)) Overall flu-orescence intensities for all samples showed only minor dif-ferences in the heatmaps A more detailed visual inspectionrevealed completely equal fluorescence intensities for ACTBALB B4GALT6 HMBS HPRT1 PPIA RPLP0 and YWHAZfor the gravity conditions prevailing during the 19thDLRPFC(Figure 2(a)) The gravity conditions investigated during theTEXUS-49 campaign showed stable expression values for thegenes ACTB ALB B4GALT6 GUSB PLA2G4A POLR2APPIA TBP UBC and YWHAZ (Figure 2(b)) For furthercharacterization and identification of stable reference genes


Table 2 List of potential reference genes

Potential reference gene Genesymbol Citation

5s rRNA [85]18s rRNA [86]120573-Actin ACTB [49 51]Albumin ALB [49]120573-2 microglobin B2M [49 51 56]UDP-GalbGlcNAcb14-galactosyl-transferasepolypeptide 6

B4GALT6 [86]

Glucose 6-phosphatedehydrogenase G6PD [49]

Glyceraldehyde-3-phosphatedehydrogenase GAPDH [49 51 56 85

87 88]Glucuronidase beta GUSB [86]Hydroxymethylbilane synthase(porphobilinogen deaminase) HMBS [49 51 56]

Hypoxanthinephosphoribosyltransferase 1 HPRT1 [49 51 56 86]

Heat shock protein 90 kDa HSP90AA1 [86]Phospholipase A2 PLA2G4A [49]RNA polymerase II POLR2A [49 86]Peptidylprolyl isomerase A(Cyclophilin A) PPIA [49 86]

Ribosomal protein L13 RPL13A [49 51 56 86]Acidic ribosomalphosphoprotein P0 RPLP0 [89]

Succinate dehydrogenasecomplex subunit A SDHA [51 56 86]

TATA box binding protein TBP [49 51 56 86]120572-Tubulin TUBA1 [49]Ubiquitin C UBC [51 56]Tyrosine 3-monooxygenasetryptophan 5-monooxygenaseactivation protein

YWHAZ [51 56 86]

we performed a geNorm pilot study [51] and calculated thecoefficient of variation (CV) for all 20 potential referencegenes (Figure 3) For homogeneous groups CV values below25 and for heterogeneous groups CV values below 50 areacceptable [56] Rapid and extreme changes in gravity inducestrong changes in cellular functions Therefore we classifiedour samples as heterogeneous groups According to the setcriteria all analyzed potential reference genes showed CVvalues below 50 for the PFC and TEXUS data sets (Figures3(a) and 3(b)) In the sample set of the 19th DLR PFC allgenes but HMBS fulfill even the more stringent criterion of aCVbelow 25 (Figure 3(a)) For the samples collected duringthe TEXUS-49 campaign all genes butHPRT1 andPLA2G4Adisplay CV values below 25 (Figure 3(b))

To increase the number of potential reference genes thatcan be used as standards for differential expression anal-yses in gravitational studies we extended our analysis toevolutionary highly conserved genes We hypothesized that

ACTB

ALB

B2M

B4GALT

6GAPD

HG

USB

HMBS

HPR

T1H

SP90

AA

1H

SP90

AB1

PLA

2G4A

POLR

2APP

IARP

L13A

RPLP

0SD

HA

TBP

TUBA

1U

BCYW

HAZ

10000

19th DLR PFC

120583g

1g

HW

108 g

1755000 3500000

(a)

ACTB

ALB

B2M

B4GALT

6GAPD

HG

USB

HMBS

HPR

T1H

SP90

AA

1H

SP90

AB1

PLA

2G4A

POLR

2APP

IARP

L13A

RPLP

0SD

HA

TBP

TUBA

1U

BCYW

HAZ

TEXUS-49

120583g

HW

BL

10000 1755000 3500000

(b)

Figure 2 Heatmaps for selected reference genes The graph illus-trates fluorescent intensity levels of the 20 potential reference genesfromTable 2 between the three and four different gravity conditionsrespectively Each gene is represented in one column and eachgravity condition is represented in one row (a) 120583g HW 1 g and18 g (19th DLR PFC) and (b) HW BL and 120583g (TEXUS-49) Theheatmap shows large variation in fluorescence intensities for thedifferent genes However within the same gene expression levels aresimilar for all tested conditions The lower bar with the graduatedred colors is the measure for the different fluorescence intensities

genes stable over time and taxonomic kingdoms should havevery fundamental functions within a cell and thus be largelyindependent from external influences to ensure basic cellularfunctions Besides ribosomal RNA genes which are notrepresented on the microarray applied in this study ABCtransporters and tRNA genes are also evolutionary highlyconserved over a wide variety of organisms Unfortunatelythe 12 times 135 K Roche NimbleGen array does also not containprobes for tRNA Therefore we had a look at expressionprofiles of ABC transporters and tRNA related genes (sup-plementary Tables 2 and 3) Since almost all fluorescencevalues of tRNA related genes showed a high variance making


43387344301258215172129

9643

Coe

ffici

ent o

f var

iatio

n (C

V) (

) 19th DLR PFC

ACTB

B2M

GU

SB

HPR

T1H

SP90

AA1

HSP

90

AB1

PLA2

G4

APO

LR2

APP

IARP

L13

A

SDH

A

TUBA

1

UBCALB

B4GALT

6GAPD

H

HMBS

RPLP

0

TBP

YWHAZ

0

(a)

43387344301258215172129

9643

0Coe

ffici

ent o

f var

iatio

n (C

V) (

)

ACTB

ALB

B2M

B4GALT

6GAPD

HG

USB

HMBS

HPR

T1H

SP90

AA1

HSP

90

AB1

PLA2

G4

APO

LR2

APP

IARP

L13

ARP

LP0

SDH

ATB

PTU

BA1

UBC

YWHAZ

TEXUS-49

(b)Figure 3 Coefficient of variation calculation for the potential reference genes This bar chart displays the coefficient of variation (CV) in of the 20 potential reference genes across the gravity conditions for the 19th DLR PFC (HW 1 g 18 g 120583g) and TEXUS-49 (HW BL 120583g) Alower value corresponds to higher stability in gene expression (a) 19th DLR PFC All calculated CV values are below the threshold of 50(b) TEXUS-49 all CV values are below 50 but in total more genes show higher coefficients of variation

19th DLR PFC

ABC

C1

ABC

C4

ABC

D4

40ABC

F2

TAP2

1 3 5 7 9 25 27 33 35 4511 13 15 17 19 21 31 39 41 4323 29 37 47

10000

120583g

HW

130000 250000

1g

18 g

(a)

HW

BL

120583g

TEXUS-49

ABC

C1

ABC

C4

ABC

D440

ABC

F2

TAP2

1 3 5 7 9 25 27 33 35 4511 13 15 17 19 21 31 39 41 4323 29 37 47

10000 130000 250000

(b)

Figure 4 Heatmaps for highly conserved ABC transporters The fluorescent intensity levels of the 47 ABC transporter genes shown insupplementary Table 2 were quantified for the different gravity conditions Each gene is represented in one column and each gravity conditionis represented in one row (a) 120583g 1 g HW and 18 g (19thDLRPFC) and (b) 120583g BL andHW (TEXUS-49)The heatmaps show large variationin fluorescence intensities for the different genes However within the same gene expression levels are mostly similar for all tested conditionsThe lower bar with the graduated red colors is the measure for the different fluorescence intensities

reasonable analysis impossible we concentrated on the ABCtransporters Heatmap analyses were carried out to obtain afirst impression on the gene stability (Figure 4) The samplesfrom the 19th DLR PFC and TEXUS-49 mission also showa rather high variation in fluorescence intensities (Figures4(a) and 4(b)) The calculation of the CV for these samples(Figure 5) displays higher values compared to the potentialreference genes however taken together all analyzed samplesof the 19th DLR PFC fulfill the criterion of CV values lessthan 50 in case of ABC transporter signals (Figure 5(a))Out of 47 samples 36 are also below 25CV Althoughthree samples from TEXUS-49 showed values above 50CV

(Figure 5(b)) 37 samples stayed below the 25 threshold(Figure 5(b)) Selected reference genes and ABC transporters(marked in bold Figures 4 and 5) were chosen for furtherdetailed analysis of differential gene expression under alteredgravity conditions

For nine of the potential reference genes from the litera-ture there were at least three values returned by themicroarray generated by independent probes targeting thesame gene Two of these genes were excluded from furtheranalysis due to high variance between their single values(HSP90AA1 and PPIA) and the remaining seven genes (ALBB4GALT6 GAPDH HMBS RPLP0 TBP and YWHAZ (see


401 3 5 7 9 25 27 33 35 4511 13 15 17 19 21 31 39 41 4323 29 37 47

ABC

C1

ABC

C4

ABC

D4

ABC

F2

TAP2

69

621

552

483

414

345

276

207

138

69

0

Coe

ffici

ent o

f var

iatio

n (C

V) (

)

19 th DLR PFC

(a)

401 3 5 7 9 25 27 33 35 4511 13 15 17 19 21 31 39 41 4323 29 37 47

ABC

C1

ABC

C4

ABC

D4

ABC

F2

TAP2

69621552483414345276207138

690

Coe

ffici

ent o

f var

iatio

n (C

V) (

)

TEXUS-49

(b)Figure 5 Coefficient of variation calculation for the ABC transporter genes This bar chart displays the coefficient of variation (CV) in ofthe 47 ABC transporter genes across the gravity conditions for the 19th DLR PFC (HW 1 g 18 g 120583g) and TEXUS-49 (HW BL 120583g) A lowervalue corresponds to higher stability in gene expression (a) 19th DLR PFC all calculated CV values are below the threshold of 50 and fulfillthe criterion (b) TEXUS-49 three genes showCV values higher than 50 andwere excluded from further analysesThe numbers correspondto the ABC transporters listed in supplementary Table 2 Genes that were further analyzed are labeled and marked in bold (ABCC1 ABCC4ABCD4 ABCF2 and TAP2)


5000

10000

15000

20000

25000

30000 19th DLR PFC potential reference genes

Fluo

resc

ence

inte

nsity

lowast

lowast lowast

lowastlowastlowast

HW1g

18 g120583g

(a)

Fluo

resc

ence

inte

nsity

ALB B4GALT60

20

40

60

HW1g

18 g120583g

19th DLR PFC potential reference genes

(b)

Figure 6 Influence of altered gravity during parabolic flight on potential reference genes RNA expression levels after 1 g (light gray) 18 g(dark gray) and 120583g (black) conditions during the 19th DLR parabolic flight campaign Hardware ground controls (HW striped) are shownfor each experimental group RNA expression levels are shown as fluorescence intensities (a) The expression values for GAPDH HMBSRPLP0 TBP and YWHAZ are displayed (b) ALB and B4GALT6 show low but stable fluorescent intensities GAPDH HMBS YWHAZALB and B4GALT6 show no significant change in RNA levels upon altered gravity for 20 sec while RPLP0 displays 120583g sensitivity comparedto 1 g and TBP reacts sensitively to all g conditionsMean values of at least threemeasurements with standard deviations are shown lowast119875 lt 005lowastlowast119875 lt 0005

Table 3)) were subjected to further statistics The calculationof the mean fluorescence intensity levels revealed that dif-ferent ranges of transcript abundance are present in bothexperimental setups While ALB and B4GALT6 seem to beexpressed rather low GAPDH HMBS RPLP0 TBP andYWHAZ are represented inmuch higher abundance (Figures6 and 7) The comparison of mean fluorescence intensities ofone gene under different g conditions revealed that GAPDHHMBS YWHAZ ALB and B4GALT6 are stably expressedwith respect to all investigated gravity conditions duringparabolic flight of the 19th DLR PFC (Figures 6(a) and 6(b))

RPLP0 is significantly upregulated by 120583g compared to 1 gwhile TBP is initially downregulated by 18 g and recoversduring 120583g (Figure 6(a)) Furthermore comparison of in-flight 1 g controls to 1 g ground controls (HW) shows asignificantly reduced mRNA level of TBP portending thatduring the preexperimental phase a certain kind of stress wasaccumulated in the cells influencing its expression level

The data analysis of the TEXUS-49 sounding rocketexperiment reveals stable RNA expression levels throughoutthe different g levels for GAPDH HMBS RPLP0 YWHAZALB and B4GALT6 (Figures 7(a) and 7(b) Table 4 and


Table3Selected

potentialreference

genes(19th

DLR

PFC)

Gene

symbo

lAc

cession

number

19th

DLR

PFC

potentialreference

genes

HW

1g18

g120583g

119875values

MeanFI

SDMeanFI

SDMeanFI

SDMeanFI

SDHW

versus

1g1g

versus

18g

18gversus120583g

1gversus120583g

ALB

NM

000477

213

204

4028

122

4062

088

4010

071

02627

07154

04724

08383

BC035969

BC034023

B4GALT

6NM

004775

4055

765

3930

762

4341

1138

3992

762

08506

06342

06847

09253

AF0

69054

BC074835

GAPD

HNM

002046

2179123

91423

2116

232

144799

1912

458

131950

2082321

200500

05652

01466

02971

08250

BC001601

BC00

9081

HMBS

NM

000190

218420

97377

161283

6774

999873

44339

122698

5315

904563

02695

05994

04833

NM

001024382

BC008149

RPLP

0

NM

001002

1135

234

172826

1163377

42619

1138089

206

050

1271495

83710

07397

08004

02346

004

25BC

001127

BC008594

BC00

0087

BC070194

TBP

NM

003194

415856

31025

32714

32172

622872

01610

9277682

17076

00192

000

4200226

00390

X54993

BC109053

YWHAZ

NM

003406

913631

207039

944522

199027

984330

212525

985306

218444

07808

07239

09934

07214

BC06

8456

BC108281

BC101483

BC083508

BC072426

BC003623



5000

10000

15000

20000

25000

30000 TEXUS-49 potential reference genes

Fluo

resc

ence

inte

nsity

lowast lowast

HWBL120583g

(a)

Fluo

resc

ence

inte

nsity

ALB B4GALT60

20

40

60

HWBL120583g

TEXUS-49 potential reference genes

(b)Figure 7 Influence of altered gravity during sounding rocket flight on potential reference genes GAPDHHMBS RPLP0 TBP and YWHAZ(a) ALB and B4GALT6 (b) RNA expression levels after launch and acceleration (BL dark gray) and 120583g (black) conditions of TEXUS-49Hardware ground controls (HW striped) are shown for each experimental group RNA levels are displayed as fluorescence intensitiesGAPDH HMBS RPLP0 YWHAZ ALB and B4GALT6 show no significant change in RNA levels upon altered gravity while TBP reactssensitively to all g conditions Mean values of at least three measurements with standard deviations are shown lowast119875 lt 005

supplementary Table 1) TBP RNA levels were reduced in 120583gsamples compared to the in-flight BL Interestingly compar-isons between the HW ground controls and BL revealed apostlaunch increase in RNA expression most likely inducedby the launch vibrations or hypergravity (Figure 7(a))

Only a very low number of tRNA related genes fulfilledour criterion of being represented by at least three probes(four out of 32) For three out of those four genes fluorescentintensity showed a great variance between the single valuesas mentioned above Only one tRNA synthetase (SARS)yielded reasonable resultsThe exposure of the cells to alteredgravity conditions during the parabolic flight resulted ina decreased SARS expression in 1 g in-flight control and18 g samples compared to the HW ground control and 1 gcontrol respectively (supplementary Table 3) Although notsignificant there is a visible increase of SARS mRNA upon120583g compared to 18 g arguing for an immediate expressionrecovery after termination of 18 g This is in line with theresults from TEXUS-49 flight campaign where SARS showsno significant expression change in in-flight baseline controlor in 120583g compared to HW ground controlThis could be dueto fast expression recovery of the gene during g alterationshyper-g phase and 120583g

The highly conserved ABC transporters were representedas a large group of genes on the applied microarray We ana-lyzed a total of 47 ABC transporters belonging to nine differ-ent sub-families (supplementary Table 2) 19 ABC transportergenes were represented by three or more individual probeson the microarray and 11 of them had similar fluorescentintensities meeting the requirements for a statistical analysisExemplarily five of those 11 ABC transporters are shown inFigures 8 and 9

During the short-term gravity alterations achieved byparabolic flights ABCC1 and ABCF2 displayed no significantdifferential expression between all g conditions analyzedTAP2 showed a significant reduction of RNA expression


2000

4000

6000

8000

10000

Fluo

resc

ence

inte

nsity

19th DLR PFC ABC transporter genes


lowast

lowastlowast lowastlowast

HW1g

18 g120583g

Figure 8 Influence of altered gravity during parabolic flight onABC transporter genes ABCC1 ABCC4 ABCD4 ABCF2 andTAP2 RNA levels after 1 g (light gray) 18 g (dark gray) and 120583g(black) conditions during the 19th DLR parabolic flight campaignHardware ground controls (HW striped) are shown for eachexperimental group RNA expression levels are displayed as fluo-rescence intensities ABCC1 and ABCF2 show no significant changein RNA expression levels upon altered gravity while ABCC4 andTAP2 display 120583g sensitivity compared to 18 g and to 18 g and 1 grespectively ABCD4 reacts sensitively to 18 g compared to 1 g andABCD4 and TAP2 show vibration sensitivity comparing 1 g to HWMean values of at least threemeasurements with standard deviationsare shown lowast119875 lt 005 lowastlowast119875 lt 0005

comparing 120583g samples to 18 g samples while ABCC4 showedan increase ABCD4 revealed hyper-g sensitivity by reducingits RNA level during 18 g compared to 1 g And ABCD4 andTAP2 displayed reduced expression during preflight phasecompared to HW control (Figure 8 Table 5)

A prolonged exposure of the cells to 120583g (378 sec ver-sus 20 sec) during TEXUS-49 experiment led to significantreduction of mRNA levels of ABCC4 ABCD4 ABCF2 and


Table4Selected

potentialreference

genes(TE

XUS-49)

Gene

symbo

lAc

cessionnu

mber

TEXU

S-49potentia

lreference

genes

HW

BL120583g

119875values

MeanFI

SDMeanFI

SDMeanFI

SDHW

versus

BLBL

versus120583g

HW

versus120583g

ALB

NM

000477

1802

129

1747

118

1788

188

06184

07708

09215

BC035969

BC034023

B4GALT

6NM

004775

1715

426

1767

605

1595

326

09093

06933

07197

AF0

69054

BC074835

GAPD

HNM

002046

2338582

172178

2486207

22710

22545117

169290

04238

07383

02126

BC001601

BC00

9081

HMBS

NM

000190

29214

010

8402

30874

910

9422

256823

8996

508609

05612

06872

NM

001024382

BC008149

RPLP

0

NM

001002

802710

123345

903611

153820

740641

262246

02871

02728

06499

BC001127

BC008594

BC00

0087

BC070194

TBP

NM

003194

373825

14390

446

601

25759

35176

95387

00215

002

0101037

X54993

BC109053

YWHAZ

NM

003406

397064

149810

573874

211820

377601

159671

00993

00758

08180

BC06

8456

BC108281

BC101483

BC083508

BC072426

BC003623


Table5Selected

ABC

transporters(19

thDLR

PFC)

No

Gene

symbo

lAc

cession

Num

ber

19th

DLR

PFC

ABC

transporterg

enes

HW

1g18

g120583g

119875values

MeanFI

SDMeanFI

SDMeanFI

SDMeanFI

SDHW

versus

1g1g

versus

18g

18gversus120583g

1gversus120583g

23ABC

C1AB2

09120

314046

7517

133092

974399

225845

32589

293479

49693

07959

0119

301312

05141

NM

019900

NM

004996

29ABC

C4BC

041560

149737

1997

514

3784

2794

58570

53065

106701

9126

07805

00676

004

5901380

AY133678

NM

005845

37ABC

D4

NM

005050

171937

16462

125413

24026

76594

23575

115589

25753

00222

00273

00673

05972

BC012815

NM

020326

NM

020325

40ABC

F2NM

005692

751096

66499

677422

4078

5602209

52591

652551

51517

01916

01263

03019

05496

BC001661

AF0

91073

47TA

P2NM

018833

336561

13624

261483

9454

248522

19087

20278

47461

000

2303711

00394

000

14BC

002751

AF105151


Table6Selected

ABC

transporters(TEX

US-49)

No

Gene

symbo

lAc

cession

number

TEXU

S-49A

BCtransporterg

enes

HW

BL120583g

119875values

MeanFI

SDMeanFI

SDMeanFI

SDHW

versus

BLBL

versus120583g

HW

versus120583g

23ABC

C1AB2

09120

240

502

68683

309607

63460

202076

39849

02701

00797

04598

NM

019900

NM

004996

29ABC

C4BC

041560

76534

29367

138237

29464

5378

714092

00620

002

2803164

AY133678

NM

005845

37ABC

D4

NM

005050

7599

219032

84389

9439

56511

13898

04697

00194

01539

BC012815

NM

020326

NM

020325

40ABC

F2NM

005692

838241

75110

796343

7214

6637322

57328

05244

004

3000238

BC001661

AF0

91073

47TA

P2NM

018833

490284

1208

400

089

3526

343097

4927

000

01000

02000

02BC

002751

AF105151



2000

4000

6000

8000

10000TEXUS-49 ABC transporter genes

Fluo

resc

ence

inte

nsity

lowast

lowast lowast

lowast

lowastlowastlowast

lowastlowastlowast

lowastlowastlowast

HWBL120583g

Figure 9 Influence of altered gravity during sounding rocket flighton ABC transporter genes ABCC1 ABCC4 ABCD4 ABCF2 andTAP2 RNA expression levels after launch and acceleration (BL darkgray) and 120583g (black) conditions of TEXUS-49 Hardware groundcontrols (HW striped) are shown for each experimental groupRNA levels are depicted as fluorescence intensities Only ABCC1expression is stable over all g conditions ABCC4 ABCD4 ABCF2and TAP2 display 120583g sensitivity compared to BL and to HW inthe case of ABCF2 and TAP2 TAP2 also shows vibration sensitivitycomparing BL to HW Mean values of at least three measurementswith standard deviations are shown lowast119875 lt 005 lowastlowast119875 lt 0005 andlowastlowastlowast119875 lt 00005

TAP2 in 120583g compared to in-flight BL Furthermore TAP2expression already decreases in the first phase after launch(BL versus HW) while the other ABC transportersrsquo mRNAlevels appeared stable (Figure 9 Table 6)

Taken together in this study we identified eight genesas nonchanging reference genes suitable for studies underaltered gravity conditions and nine genes as candidates forg-sensitivity and 83 genes could not be assigned to eithergroup due to low probe number on the microarray or to greatvariance between the probe values (Table 7)

4 Discussion

Microarray expression data are intensely used to analyzedifferential gene expression in cells tissues and organismsthat are exposed to various conditions [29 30] Even inthe field of gravitational biology gene expression analysesare utilized with increasing frequency Recently an articlewas released giving an overview of all published microarraybased microgravity studies [57] describing the difficulties tocombine and overlay the data from different experimentsstudy objects microgravity platforms (simulated micro-gravity sounding rocket space shuttle and ISS missions)and different microarray experimental designs [58ndash64]The different analyses were done mostly in simulated micro-gravity and investigated various organisms and cell types likeArabidopsis thaliana Salmonella enterica and rat and mousetissues as well as human osteoblasts and T-cells [58ndash61 65ndash70] (for the complete list see [57]) The goal was to screen thevast amount of data to identify a list of major ldquospace genesrdquo

that are sensitive to microgravity throughout all involvedplatforms The data inspection revealed a huge number ofdifferentially expressed genes butwith only little or no overlapbetween closely related studies on the level of single genesIn contrast on the level of pathway analysis it was possibleto define major pathways like ECM-receptor interactionfocal adhesion TGF-beta signaling and glycolysis beingaffected in many species (human mouse rat and Xenopusin different combinations) by the exposure to microgravity[57] Moreover major ldquospace genesrdquo sensitive to microgravitywere defined if they were found to be differentially expressedin at least four of the examined studies The results showedin total eight potential space genes (CD44 MARCKS FN1TUBA1 CTGF CYR61 MT2 and MT1) which are involvedin T-cell development cell motility extracellularmatrix com-ponents cytoskeleton and oxidative stress protection [57]The study describes in detail the difficulties of combininggene expression data from different groups due to varyingexperimental setups and conditions It elucidates that it is ofhigh relevance to be able to standardize gene expression datathat arose from RT-qPCR or microarray studies A key com-ponent for standardization within a single experiment andbetween different experiments is normalization An impo-rtant factor for normalization is the use of stable referencegenes There are numerous studies describing that commo-nly used reference genes could represent a pitfall becausethey are often differentially expressed under specific experi-mental conditions and that they have to be considered carefu-lly before the experiment [46ndash49] Different guidelineshave been published to facilitate standardized experimentaldesign and increase comparability between analyses (MIQEMIAME) [34 39 40] It is for example highly recommendedto perform a pilot study with programs like geNorm or Best-Keeper prior to the experiment to identify several stable refe-rence genes that can be used simultaneously as controls fornormalization in the differential gene expression analysis[50 51]

Alternatively tomicroarrays a novel technique RNA-Seqis under development for whole genome expression analysesIt is reported that thismethod has advantages in detecting lowabundance transcripts genetic variants and splice isoformsof genes as well as distinguishing biologically critical isoforms[71] Despite the described technical advantages of RNA-Seq microarrays remain popular for some reasons Themicroarray platforms have a proven track record spanningnearly two decades in the lab The arrays are generallyconsidered easier to use with less complicated and less labor-intensive sample preparation than RNA-Seq The same holdstrue for the data storage and data analysis Moreover despitethe rapid drop in the cost associated with next-generationsequencing (NGS) arrays are still more economical andyield higher throughput providing significant advantageswhen working with a large number of samples Thereforemicroarray analyses are still more commonly used for tran-scriptional profiling experiments [71]

Taking into account that many other studies throughoutthe last few years have reported a considerable portion of thetraditionally used reference genes not being stably expressedunder various experimental conditions it becomes rather


Table 7 Overview of g-stable (+) and g-sensitive genes (minus)

Gene symbol Accession number 19th DLR PFC TEXUS-49(HW

versus 1 g)(1 g versus

18 g)(18 g

versus 120583g)(1 g versus120583g)

(HWversus BL)

(BL versus120583g)

(HWversus 120583g)

Potential reference genes

ALBNM 000477

+ + + + + + +BC035969BC034023

B4GALT6NM 004775

+ + + + + + +AF069054BC074835

GAPDHNM 002046

+ + + + + + +BC001601BC009081

HMBSNM 000190

+ + + + + + +NM 001024382BC008149

RPLP0

NM 001002

+ + + minus + + +BC001127BC008594BC000087BC070194

TBPNM 003194

minus minus minus minus minus minus +X54993BC109053

YWHAZ

NM 003406

+ + + + + + +

BC068456BC108281BC101483BC083508BC072426BC003623

ABC transporter genes

ABCA5NM 018672

+ + + + + + +AJ275973AY028897

ABCA9NM 080283

+ + + + + + +BC062472NM 172386

ABCC1AB209120

+ + + + + + +NM 019900NM 004996

ABCC4BC041560

+ + minus + + minus +AY133678NM 005845

ABCC12

AK127951

minus + + + minus minus +NM 145187NM 033226BC036378

ABCD4

NM 005050

minus minus + + + minus +BC012815NM 020326NM 020325

ABCF2NM 005692

+ + + + + minus minusBC001661AF091073

TAP2NM 018833

minus + minus minus minus minus minusBC002751AF105151


apparent that a natural constant as 1 g might have evenmore an effect on the expression of genes than other testcircumstances Therefore we focused in this study on theinvestigation of the expression qualities of several potentialreference genes under 1 g compared to altered gravity con-ditions generated by two widely used platforms parabolicflights and sounding rockets These two platforms are ofspecial interest because of the rather easy access comparedto the extremely limited accessibilities of long-term micro-gravity experiments on satellites and the ISS We present amicroarray based analysis identifying stable reference genesin cells of the immune system exposed to short-term (severalseconds) and middle-term (several minutes) altered gravityconditions on the twowidely used platforms parabolic flightsand sounding rockets

Our analyses of commonly used reference genes ABCtransporters and tRNA related genes revealed that nineof the 17 genes suspected to be ubiquitously expressed areg-sensitive and therefore inappropriate for our purposesamongst them being TATA box binding protein (TBP) afundamental transcription factor for many genes and seryl-tRNA synthetase (SARS) an essential enzyme for mRNAtranslation also regulating vascular development (Table 7)

Two of the g-sensitive genes that we identified in thisstudy are involved in multidrug resistance processes likethe ABC transporters ABCC4 and transporter associatedwith antigen presenting 2 (TAP2) ABCC4 is of particularinterest because it has the ability to provide resistance toantiviral and anticancer nucleotide analogs andmethotrexate[72 73] acts as an independent regulator of intracellularcAMP mediates cAMP dependent signal transduction tothe nucleus and controls human and rat smooth musclecell (SMC) proliferation [74] It is known that cAMP haslargely inhibitory effects on components of macrophageactivation and elevation of cAMP levels which suppressesFcgammaR-mediated phagocytosis [75] Therefore it wouldbe interesting to look at this multidrug resistance-associatedprotein (ABCC4) in more detail in microgravity exposedcells to elucidate its role in the signaling cascades importantfor immune cell action and reaction under space conditionsas ABCC4 proved to be 120583g-sensitive during parabolic andsounding rocket flight (Figures 8 and 9) Furthermore TAP2seems to be evenmore g-sensitive because it shows significantdifferential gene expression under 120583g and hypergravity con-ditions during parabolic and sounding rocket flight (Figures8 and 9) It will be interesting to further analyze the potentialeffects of differential gene expression of TAP2 because it isa key player in endogenous pathways for antigen presenta-tion and involved in the cellular transport of antigens forsubsequent association with MHC class I molecules [76] Animbalance in its gene expression could lead to an impairedreactivity of cells of the immune system under altered gravityconditions

Further standard genes as well as ABC transporterslike RPLP0 ABCD4 and ABCF2 also turned out in ouranalysis to be g-sensitive RPLP0 encodes for a ribosomalprotein that is a component of the 60S subunit and interactswith P1 and P2 to form pentameric complexes [77] It isinvolved for example in Chagas disease [78] as well as mixed

connective tissue disease [79]TheABC transporters ABCD4and ABCF2 are involved in transport of molecules acrossextra- and intracellular membranes like in peroxisomalimport of fatty acids andor fatty acyl-CoAs in the organelle[80] and play a role in suppression of volume-sensitiveoutwardly rectifying Cl channel (VSOR) respectively [81]Altered expression levels of those genes by microgravity orhypergravity could have an impact on the translational levelor the supply of the cell with essential resources important forproper cellular function Recently it was shown that duringparabolic flights the activity of the MRP2-ABC-transporterwas significantly reduced [82] Furthermore under shortduration spaceflight missions certain ABC transporter genesin the medically relevant species Salmonella sp and Candidasp were upregulated [83 84]

Interestingly we identified many of the g-sensitive genesnot only reacting on 120583g but also on hypergravity indicatingthat not only the experimental g-conditions should be takeninto account when selecting an appropriate reference genebut also the accompanying g-conditions prevailing usuallybefore 120583g is achieved A detailed differential gene expressionanalysis of the parabolic flight and sounding rocket flight datasets for g-sensitive genes is currently ongoing

Genes that proved to be stable over all g-conditions testedwere

(i) albumin (ALB) a protein comprising about one halfof blood serum protein

(ii) UDP-GalbetaGlcNAc beta 14-galactosyltransferasepolypeptide 6 (B4GALT6) a type II membrane-bound glycoprotein important for glycolipid biosyn-thesis

(iii) glyceraldehyde-3-phosphate dehydrogenase(GAPDH) a protein with several distinct functionsfor example the reversible oxidative phosphorylationof glyceraldehyde-3-phosphate

(iv) hydroxymethylbilane synthase (HMBS) a proteincatalyzing the head to tail condensation of four por-phobilinogen molecules into the linear hydroxyme-thylbilane

(v) tyrosine 3-monooxygenasetryptophan 5-monooxy-genase activation protein zeta (YWHAZ) a gene pro-duct belonging to the 14-3-3 family of proteins thatinteracts with IRS1 suggesting a role in regulatinginsulin sensitivity

(vi) ATP-binding cassette subfamily A member 5(ABCA5) a membrane-associated protein belongingto the only major ABC subfamily found exclusivelyin multicellular eukaryotes with unknown function

(vii) ATP-binding cassette subfamily A member 9(ABCA9) another ABC1 family member inducedduring monocyte differentiation into macrophagesand

(viii) ATP-binding cassette subfamily C member 1(ABCC1) a member of the MRP subfamily of ABCtransporters involved in multidrug resistance and


functioning as a multispecific organic anion transpo-rter

Taken together the compilation of genes that we presentin Table 7 gives an overview about which genes are stablyexpressed during all investigated gravitational conditionslasting from seconds to minutes and can therefore be con-sidered as suitable reference genes Furthermore Table 7 canbe regarded as a tool for the community that can be easilyadapted to select potential control genes in the design phaseof a new immune cell based experiment on parabolic flightsand sounding rocket flights because it provides valuableinformation about gene expression levels in 120583g as well as in18 g in-flight 1 g and hardware ground control Our resultsalso allow for the identification of adaptation mechanisms bycomparing short (parabolic flight) and intermediate (sound-ing rocket) microgravity periods and spot those genes thatconvert from sensitive into stable and vice versa Our workshould considerably facilitate identification of appropriatereference genes for individual experiments performed duringparabolic flight and sounding rocket campaignswith immunecells especially of the monocytemacrophage system inaltered gravity

Abbreviations

BL BaselineCEV Centre drsquoessai en volCV Coefficient of variationESA European Space AgencyFBS Fetal bovine serumFDR False discovery rateDLR German Aerospace CenterHW Hardware ground controlsHKG Housekeeping geneshyper-g HypergravityIL InterleukinISS International Space StationLPS Lipopolysaccharide120583g MicrogravityMMS Monocyte-macrophage systemPFC Parabolic flight campaignqPCR Quantitative real-time PCRRIN RNA integrity numberRPM Random positioning machineROS Reactive oxygen speciesRWV Rotating wall vesselSD Standard deviationTNF-120572 Tumor necrosis factor-alpha



Authorsrsquo Contributions

Oliver Ullrich and Cora S Thiel developed the study ideaconcept and the overall study design in addition to planningcoordinating and supervising the study Cora SThiel Liliana

E Layer and Swantje Hauschild wrote the paper OliverUllrich edited the paper Beatrice Lauber contributed tothe paper Liliana E Layer Cora S Thiel Oliver UllrichSvantje Tauber Swantje Hauschild Claudia Philpot AnnettGutewort Eva Hurlimann Josefine Biskup andHartwin Lierperformed the experiments during the 19th DLR parabolicflight campaign Svantje Tauber Cora S Thiel Felix Unver-dorben and Oliver Ullrich performed the experiments dur-ing the TEXUS-49 mission Cora S Thiel was responsible forsample analysis from the 19th DLR parabolic flight campaignand TEXUS-49 mission Liliana E Layer contributed tothe sample analysis Frank Engelmann contributed to andsupervised the technical procedures during the 19th DLRparabolic flight campaign

Acknowledgments

The authors gratefully acknowledge financial support byGerman Aerospace Center DLR (grants nos 50WB0912 and50WB1219) They also gratefully acknowledge the support of(in alphabetic order) Gesine Bradacs Markus Braun MiriamChristen Giovanni Colacicco Ulrike Friedrich NadineGolzAndre Hilliger AndreasHuge Schirin IbrahimOtfried JoopSonja Krammer Andre Melik Shirin Milani Brice MouttetMarianne Ott Irina Rau Frank Ruhli Chen Sang BurkhardSchmitz Brita Scholte Andreas Schutte Johanna StahnMarcStuder Susanne Wolf and Fengyuan Zhuang

References

[1] B Comet ldquoLimiting factors for human health and performancemicrogravity and reduced gravity In study on the survivabilityand adaptation of humans to long-duration interplanetary andplanetary environmentsrdquo Technical Note 2 Critical Assess-ments of the Limiting Factors for Human Health and Perfor-mance and Recommendation of Countermeasures HUMEX-TN-002 2001

[2] G Horneck R Facius M Reichert et al ldquoHUMEX a studyon the survivability and adaptation of humans to long-durationexploratory missions part II missions to Marsrdquo Advances inSpace Research vol 38 no 4 pp 752ndash759 2006

[3] G Horneck and B Comet ldquoGeneral human health issues forMoon and Mars missions results from the HUMEX studyrdquoAdvances in Space Research vol 37 no 1 pp 100ndash108 2006


[5] H Takayanagi ldquoOsteoimmunology shared mechanisms andcrosstalk between the immune and bone systemsrdquo NatureReviews Immunology vol 7 no 4 pp 292ndash304 2007

[6] J Caetano-Lopes J E Canhao and H Fonseca ldquoOsteoimmu-nologymdashthe hidden immune regulation of bonerdquo Autoimmu-nity Reviews vol 8 no 3 pp 250ndash255 2009

[7] A Cogoli B Bechler O Muller and E Hunzinger ldquoEffect ofmicrogravity on lymphocyte activationrdquo in Biorack on SpacelabD1 pp 89ndash100 European Space Agency Paris France 1988

[8] A Cogoli and A Tschopp ldquoLymphocyte reactivity during spa-ceflightrdquo Immunology Today vol 6 no 1 pp 1ndash4 1985


[9] A Cogoli A Tschopp and P Fuchs-Bislin ldquoCell sensitivity togravityrdquo Science vol 225 no 4658 pp 228ndash230 1984

[10] A Cogoli ldquoGravitational physiology of human immune cellsa review of in vivo ex vivo and in vitro studiesrdquo Journal ofGravitational Physiology vol 3 no 1 pp 1ndash9 1996


[12] J W Armstrong R A Gerren and S K Chapes ldquoThe effectof space and parabolic flight onmacrophage hematopoiesis andfunctionrdquo Experimental Cell Research vol 216 no 1 pp 160ndash168 1995

[13] M Limouse S Manie I Konstantinova B Ferrua and LSchaffar ldquoInhibition of phorbol ester-induced cell activation inmicrogravityrdquo Experimental Cell Research vol 197 no 1 pp 82ndash86 1991

[14] D A Schmitt J P Hatton C Emond et al ldquoThe distribution ofprotein kinase C in human leukocytes is altered in micrograv-ityrdquo FASEB Journal vol 10 no 14 pp 1627ndash1634 1996

[15] C-L Hsieh P-D L Chao and S-H Fang ldquoMorin sulphatesglucuronides enhance macrophage function in microgravityculture systemrdquo European Journal of Clinical Investigation vol35 no 9 pp 591ndash596 2005

[16] I Kaur E R Simons V A Castro C M Ott and D L PiersonldquoChanges inmonocyte functions of astronautsrdquoBrain Behaviorand Immunity vol 19 no 6 pp 547ndash554 2005

[17] A Adrian K Schoppmann J Sromicki et al ldquoThe oxidativeburst reaction in mammalian cells depends on gravityrdquo CellCommunication and Signaling vol 11 no 1 article 98 2013


[19] MAMeloniGGalleri G Pani A Saba P Pippia andMCog-oli-Greuter ldquoSpace flight affects motility and cytoskeletal struc-tures in human monocyte cell line J-111rdquo Cytoskeleton vol 68no 2 pp 125ndash137 2011

[20] M Hughes-Fulford T Chang and C-F Li ldquoEffect of gravity onmonocyte differentiationrdquo in Proceedings of the 10th ESA LifeSciences Symposium29th Annual ISGP Meeting24th AnnualASGSB MeetingELGRA Symposium ldquoLife in Space for Life onEarthrdquo pp 22ndash27 2008

[21] C Porcher M-C Malinge C Picat and B Grandchamp ldquoAsimplified method for determination of specific DNA or RNAcopy number using quantitative PCR and an automatic DNAsequencerrdquo BioTechniques vol 13 no 1 pp 106ndash114 1992

[22] R Higuchi C Fockler G Dollinger and R Watson ldquoKineticPCR analysis real-time monitoring of DNA amplificationreactionsrdquo Nature Biotechnology vol 11 no 9 pp 1026ndash10301993

[23] P-W Chiang W-J Song K-Y Wu et al ldquoUse of a fluorescent-PCR reaction to detect genomic sequence copy number andtranscriptional abundancerdquo Genome Research vol 6 no 10 pp1013ndash1026 1996

[24] U E Gibson C A Heid and P M Williams ldquoA novel methodfor real time quantitative RT-PCRrdquoGenome Research vol 6 no10 pp 995ndash1001 1996

[25] C A Heid J Stevens K J Livak and PMWilliams ldquoReal timequantitative PCRrdquoGenome Research vol 6 no 10 pp 986ndash9941996

[26] H D VanGuilder K E Vrana and W M Freeman ldquoTwenty-five years of quantitative PCR for gene expression analysisrdquo Bio-Techniques vol 44 no 5 pp 619ndash626 2008

[27] R Biassoni and A Raso Eds Quantitative Real-Time PCRMethods and Protocols Humana Press New York NY USA2014

[28] W M Freeman D J Robertson and K E Vrana ldquoFundamen-tals of DNA hybridization arrays for gene expression analysisrdquoBioTechniques vol 29 no 5 pp 1042ndash1055 2000

[29] S Draghici Data Analysis for DNA Microarrays Chapman ampHallCRC Boca Raton Fla USA 2003

[30] H C Causton J Quackenbush and A Brazma MicroarrayGene Expression Data Analysis A Beginnerrsquos Guide BlackwellPublishing Malden Mass USA 2003

[31] ldquoThe chipping forecastrdquo Nature Genetics vol 21 pp 1ndash60 1999[32] P O Brown and D Botstein ldquoExploring the new world of the

genome with DNA microarraysrdquo Nature Genetics vol 21 no 1pp 33ndash37 1999

[33] D J Lockhart and E A Winzeler ldquoGenomics gene expressionand DNA arraysrdquoNature vol 405 no 6788 pp 827ndash836 2000

[34] A Brazma PHingamp J Quackenbush et al ldquoMinimum infor-mation about a microarray experiment (MIAME)mdashtowardstandards for microarray datardquo Nature Genetics vol 29 no 4pp 365ndash371 2001

[35] R D Canales Y Luo J C Willey et al ldquoEvaluation of DNAmicroarray results with quantitative gene expression platformsrdquoNature Biotechnology vol 24 no 9 pp 1115ndash1122 2006

[36] S Lefever J Vandesompele F Speleman and F Pattyn ldquoRTPri-merDB the portal for real-time PCR primers and probesrdquoNucleic Acids Research vol 37 no 1 pp D942ndashD945 2009

[37] F Pattyn P Robbrecht A de Paepe F Speleman and J Vande-sompele ldquoRTPrimerDB the real-time PCR primer and probedatabase major update 2006rdquo Nucleic Acids Research vol 34supplement 1 pp D684ndashD688 2006

[38] X Wang and B Seed ldquoA PCR primer bank for quantitativegene expression analysisrdquoNucleic Acids Research vol 31 no 24article e154 2003

[39] S A Bustin V Benes J A Garson et al ldquoTheMIQE guidelinesminimum information for publication of quantitative real-timePCRexperimentsrdquoClinical Chemistry vol 55 no 4 pp 611ndash6222009

[40] S Taylor MWakem G Dijkman M Alsarraj andM NguyenldquoA practical approach to RT-qPCR-Publishing data that con-form to theMIQE guidelinesrdquoMethods vol 50 no 4 pp S1ndashS52010

[41] V Marx ldquoPCR living life amplified and standardizedrdquo NatureMethods vol 10 no 5 pp 391ndash395 2013

[42] S A Bustin ldquoAbsolute quantification of mrna using real-timereverse transcription polymerase chain reaction assaysrdquo Journalof Molecular Endocrinology vol 25 no 2 pp 169ndash193 2000

[43] T Suzuki P J Higgins and D R Crawford ldquoControl selectionfor RNAquantitationrdquoBioTechniques vol 29 no 2 pp 332ndash3372000

[44] O Thellin W Zorzi B Lakaye et al ldquoHousekeeping genes asinternal standards use and limitsrdquo Journal of Biotechnology vol75 no 2-3 pp 291ndash295 1999

[45] N Tanic M Perovic A Mladenovic S Ruzdijic and SKanazir ldquoEffects of aging dietary restriction and glucocorticoidtreatment on housekeeping gene expression in rat cortex andhippocampusmdashevaluation by real time RT-PCRrdquo Journal ofMolecular Neuroscience vol 32 no 1 pp 38ndash46 2007


[46] T D Schmittgen and B A Zakrajsek ldquoEffect of experimentaltreatment on housekeeping gene expression validation byreal-time quantitative RT-PCRrdquo Journal of Biochemical andBiophysical Methods vol 46 no 1-2 pp 69ndash81 2000

[47] J A Warrington A Nair M Mahadevappa and M Tsygan-skaya ldquoComparison of human adult and fetal expression andidentification of 535 housekeepingmaintenance genesrdquo PhysiolGenomics vol 2 no 3 pp 143ndash147 2000

[48] K Dheda J F Huggett S A Bustin M A Johnson G Rookand A Zumla ldquoValidation of housekeeping genes for norma-lizing RNA expression in real-time PCRrdquo BioTechniques vol 37no 1 pp 112ndash119 2004

[49] A Radonic SThulke IMMackayO LandtW Siegert andANitsche ldquoGuideline to reference gene selection for quantitativereal-time PCRrdquo Biochemical and Biophysical Research Commu-nications vol 313 no 4 pp 856ndash862 2004

[50] M W Pfaffl A Tichopad C Prgomet and T P NeuviansldquoDetermination of stable housekeeping genes differentiallyregulated target genes and sample integrity bestKeepermdashexcel-based tool using pair-wise correlationsrdquo Biotechnology Lettersvol 26 no 6 pp 509ndash515 2004

[51] J Vandesompele K De Preter F Pattyn et al ldquoAccurate nor-malization of real-time quantitative RT-PCR data by geometricaveraging of multiple internal control genesrdquo Genome Biologyvol 3 no 7 Article ID RESEARCH0034 2002

[52] J H Cai S Deng SW Kumpf et al ldquoValidation of rat referencegenes for improved quantitative gene expression analysis usinglow density arraysrdquo BioTechniques vol 42 no 4 pp 503ndash5122007


[54] Y Benjamini and Y Hochberg ldquoControlling the false discoveryrate a practical and powerful approach to multiple testingrdquoJournal of the Royal Statistical Society Series B Methodologicalvol 57 no 1 pp 289ndash300 1995

[55] N Silver S Best J Jiang and S L Thein ldquoSelection of house-keeping genes for gene expression studies in human reticu-locytes using real-time PCRrdquo BMC Molecular Biology vol 7article 33 2006

[56] J Hellemans G Mortier A de Paepe F Speleman and JVandesompele ldquoqBase relative quantification framework andsoftware for management and automated analysis of real-timequantitative PCR datardquo Genome biology vol 8 no 2 p R192007

[57] J Q Clement ldquoGene expression microarrays in microgravityresearch toward the identification of major space genesrdquo inInnovations in Biotechnology E C Agbo Ed pp 319ndash348InTech 2012

[58] A-I Kittang J J van Loon O Vorst R D Hall K Fossumand T-H Iversen ldquoGround based studies of gene expressionin Arabidopsis exposed to gravity stressesrdquo Journal of Gravita-tional Physiology vol 11 no 2 pp P223ndashP224 2004


[60] V Chopra A A Fadl J Sha S Chopra C L Galindo andA K Chopra ldquoAlterations in the virulence potential of ente-ric pathogens and bacterial-host cell interactions under sim-ulated microgravity conditionsrdquo Journal of Toxicology and

Environmental Health Part A Current Issues vol 69 no 14 pp1345ndash1370 2006

[61] S Yamada T Ganno N Ohara and Y Hayashi ldquoChitosanmonomer accelerates alkaline phosphatase activity on humanosteoblastic cells under hypofunctional conditionsrdquo Journal ofBiomedical Materials Research Part A vol 83 no 2 pp 290ndash295 2007


[63] M A Meloni G Galleri S Carta et al ldquoPreliminary studyof gene expression levels in human T-cells exposed to cosmicradiationsrdquo Journal of Gravitational Physiology vol 9 no 1 ppP291ndashP292 2002

[64] S J PardoM J PatelMC Sykes et al ldquoSimulatedmicrogravityusing the Random Positioning Machine inhibits differentiationand alters gene expression profiles of 2T3 preosteoblastsrdquo TheAmerican Journal of PhysiologymdashCell Physiology vol 288 no 6pp C1211ndashC1221 2005

[65] J W Wilson R Ramamurthy S Porwollik et al ldquoMicroarrayanalysis identifies Salmonella genes belonging to the low-shearmodeled microgravity regulonrdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 99 no21 pp 13807ndash13812 2002

[66] M Wittwer M Fluck H Hoppeler S Muller D Desplanchesand R Billeter ldquoProlonged unloading of rat soleus musclecauses distinct adaptations of the gene profilerdquoThe FASEB Jou-rnal vol 16 no 8 pp 884ndash886 2002


[68] K M Fridley I Fernandez M-T A Li R B Kettlewell and KRoy ldquoUnique differentiation profile of mouse embryonic stemcells in rotary and stirred tank bioreactorsrdquo Tissue EngineeringPart A vol 16 no 11 pp 3285ndash3298 2010

[69] A Qian S Di X Gao et al ldquocDNA microarray reveals thealterations of cytoskeleton-related genes in osteoblast underhigh magneto-gravitational environmentrdquo Acta Biochimica etBiophysica Sinica vol 41 no 7 pp 561ndash577 2009

[70] N E Ward N R Pellis S A Risin and D Risin ldquoGene expre-ssion alterations in activated humanT-cells induced bymodeledmicrogravityrdquo Journal of Cellular Biochemistry vol 99 no 4 pp1187ndash1202 2006

[71] S Zhao W-P Fung-Leung A Bittner K Ngo and X LiuldquoComparison of RNA-Seq and microarray in transcriptomeprofiling of activated T cellsrdquo PLoS ONE vol 9 no 1 ArticleID e78644 2014

[72] Z-S Chen K Lee and G D Kruh ldquoTransport of cyclicnucleotides and estradiol 17-120573-D-glucuronide by multidrugresistance protein 4 Resistance to 6-mercaptopurine and 6-thioguaninerdquo The Journal of Biological Chemistry vol 276 no36 pp 33747ndash33754 2001

[73] Z-S Chen K Lee S Walther et al ldquoAnalysis of methotrex-ate and folate transport by multidrug resistance protein 4(ABCC4) MRP4 is a component of the methotrexate effluxsystemrdquo Cancer Research vol 62 no 11 pp 3144ndash3150 2002

[74] Y Sassi L Lipskaia G Vandecasteele et al ldquoMultidrug resi-stance-associated protein 4 regulates cAMP-dependent signal-ing pathways and controls human and rat SMC proliferationrdquo


The Journal of Clinical Investigation vol 118 no 8 pp 2747ndash2757 2008

[75] D M Aronoff C Canetti C H Serezani M Luo and MPeters-Golden ldquoCutting edge macrophage inhibition by cyclicAMP (cAMP) differential roles of protein kinase A andexchange protein directly activated by cAMP-1rdquoTheThe Journalof Immunology vol 174 no 2 pp 595ndash599 2005

[76] E Procko and R Gaudet ldquoAntigen processing and presentationTAPping into ABC transportersrdquo Current Opinion in Immunol-ogy vol 21 no 1 pp 84ndash91 2009

[77] K-M Lee C W Yu D S Chan et al ldquoSolution structure of thedimerization domain of ribosomal protein P2 provides insightsfor the structural organization of eukaryotic stalkrdquoNucleic AcidsResearch vol 38 no 15 pp 5206ndash5216 2010

[78] I Ferrari M J Levin G Wallukat et al ldquoMolecular mimicrybetween the immunodominant ribosomal protein P0 of Try-panosoma cruzi and a functional epitope on the human 1205731-adrenergic receptorrdquo Journal of Experimental Medicine vol 182no 1 pp 59ndash65 1995

[79] R W Hoffman and M E Maldonado ldquoImmune pathogenesisof Mixed Connective Tissue Disease a short analytical reviewrdquoClinical Immunology vol 128 no 1 pp 8ndash17 2008

[80] S Kemp and R J A Wanders ldquoX-linked adrenoleukodys-trophy very long-chain fatty acid metabolism ABC half-transporters and the complicated route to treatmentrdquoMolecularGenetics and Metabolism vol 90 no 3 pp 268ndash276 2007

[81] Y Ando-Akatsuka T Shimizu T Numata and Y OkadaldquoInvolvements of the ABC protein ABCF2 and 120572-actinin-4 inregulation of cell volume and anion channels in human epithe-lial cellsrdquo Journal of Cellular Physiology vol 227 no 10 pp3498ndash3510 2012

[82] S Vaquer E Cuyas A Rabadan A Gonzalez F Fenollosa andR de la Torre ldquoActive transmembrane drug transport in micro-gravity a validation study using an ABC transporter modelrdquoF1000Research vol 3 article 201 2014

[83] JWWilson CMOtt KHoner Zu Bentrup et al ldquoSpace flightalters bacterial gene expression and virulence and reveals a rolefor global regulator Hfqrdquo Proceedings of the National Academyof Sciences of the United States of America vol 104 no 41 pp16299ndash16304 2007

[84] A Crabbe S M Nielsen-Preiss C M Woolley et al ldquoSpace-flight enhances cell aggregation and random budding in Can-dida albicansrdquo PLoS ONE vol 8 no 12 Article ID e80677 2013

[85] C K Mantri J P Dash J V Mantri and C C V DashldquoCocaine Enhances HIV-1 Replication in CD4+ T Cells byDown-Regulating MiR-125brdquo PLoS ONE vol 7 no 12 ArticleID e51387 2012

[86] F Jacob R Guertler S Naim et al ldquoCareful selection ofreference genes is required for reliable performance of RT-qPCR in human normal and cancer cell linesrdquo PLoS ONE vol8 no 3 Article ID e59180 2013

[87] A Marcant A Denys A Melchior et al ldquoCyclophilin B attenu-ates the expression of TNF-120572 in lipopolysaccharide-stimulatedmacrophages through the induction of B cell lymphoma-3rdquoTheJournal of Immunology vol 189 no 4 pp 2023ndash2032 2012

[88] B P Barna I Huizar A Malur et al ldquoCarbon nanotube-induced pulmonary granulomatous disease twist1 and alveolarmacrophage M1 activationrdquo International Journal of MolecularSciences vol 14 no 12 pp 23858ndash23871 2013

[89] J P Chou CM Ramirez J EWu andR B Effros ldquoAcceleratedaging in HIVAIDS novel Biomarkers of Senescent HumanCD8+ T Cellsrdquo PLoS ONE vol 8 no 5 Article ID e64702 2013

Research ArticleA Whole-Genome Microarray Study of Arabidopsis thalianaSemisolid Callus Cultures Exposed to Microgravity andNonmicrogravity Related Spaceflight Conditions for 5 Dayson Board of Shenzhou 8

Svenja Fengler1 Ina Spirer1 Maren Neef1 Margret Ecke1

Kay Nieselt2 and Ruumldiger Hampp1

1Physiological Ecology of Plants University of Tubingen Auf der Morgenstelle 1 72076 Tubingen Germany2Center for Bioinformatics University of Tubingen Sand 14 72076 Tubingen Germany

Correspondence should be addressed to Svenja Fengler svenjafengleruni-tuebingende

Received 8 May 2014 Revised 26 August 2014 Accepted 9 September 2014


Copyright copy 2015 Svenja Fengler et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

The Simbox mission was the first joint space project between Germany and China in November 2011 Eleven-day-old Arabidopsisthaliana wild type semisolid callus cultures were integrated into fully automated plant cultivation containers and exposed tospaceflight conditions within the Simbox hardware on board of the spacecraft Shenzhou 8 The related ground experiment wasconducted under similar conditions The use of an in-flight centrifuge provided a 1 g gravitational field in space The cells weremetabolically quenched after 5 days via RNAlater injectionThe impact on theArabidopsis transcriptomewas investigated bymeansof whole-genome gene expression analysis The results show a major impact of nonmicrogravity related spaceflight conditionsGenes that were significantly altered in transcript abundance are mainly involved in protein phosphorylation and MAPK cascade-related signaling processes as well as in the cellular defense and stress responses In contrast to short-term effects of microgravity(seconds minutes) this mission identified only minor changes after 5 days of microgravity These concerned genes coding forproteins involved in the plastid-associated translation machinery mitochondrial electron transport and energy production

1 Introduction

Gravitation biology is a field of research which hasmade con-siderable progress within the last years involving prokary-otes fungi plants and animals Plants are especially interest-ing because as sessile organisms they possess high versatilityin responding to environmental challenges and abiotic aswell as biotic ones In order to investigate responses toaltered gravitation a large range of methods is availablethat allows for modification of the Earthrsquos gravitational fieldThese involve centrifugation (hypergravity) clinorotationmagnetic levitation and random positioning (simulatedmicrogravity) or parabolic flights of planes and sound-ing rockets as well as satellites and spacecrafts (delivermicrogravity) Experiments with plants show that not onlytissues and organelles [1 2] but also single-cell systems

like characean rhizoids [3ndash7] as well as spores (Ceratopterisrichardii [8 9]) and protoplasts [10ndash12] or homogeneous cellcultures (Arabidopsis thaliana) exhibit gravisensitivity [2 13ndash16] Experimental approaches that analyze the response toaltered gravitation such as transcriptomics proteomics andmetabolomics dominate recently First molecular approacheswere using transcriptomics that is the search for genes whichchange their expression under altered gravitation In plantslike in other organisms the improvement of gene expressionquantification technologies together with growing databasessupports this development considerably To date databasesare available that exhibit plant datasets representing theirresponse to diverse experimental stimuli [17ndash20] They showthat external signals are translated into biochemical onesresulting in molecular signaling cascades which eventuallyresult in a life-sustaining adaptation process



For Arabidopsis (Arabidopsis thaliana) cell suspensioncultures the response to short-term microgravity was inves-tigated intensely in our group by means of parabolic flights[21] A combination of transcriptomics with phosphopro-teomics showed that changes in gene expression and proteinmodification occur within seconds The investigation ofeffects caused by longer-lasting microgravity depends onmuch scarcer availability of respective flight opportunitiesHowever data on cellular andmolecular long-term responsesof plants such as Brassicaceae (Arabidopsis) Fabaceae andPoaceae has recently been published [2 15 22ndash31] Withregard to long-term experiments on gene expression thereare conflicting reports Stutte et al [30] for example couldnot observe differentially expressed genes (DEGs) above a 2-fold cut-off in 24-day-old wheat leaves after a 21-day-spacemission In contrast Paul et al [15 24] detected many DEGsin nearly 20-day-old Arabidopsis callus cultures and 18-day-old seedlings after a nearly 13-day-space mission Further-more the set of altered genes detected in whole seedlingswas different from that in callus cultures [15] Thereby thespaceflight-mediated upregulated expression of heat shockproteins appeared to be an age-independent cell culturespecific response [15 16] Within the so-called TROPI-2experiment only 24 genes were altered in their abundance inArabidopsis seedlings [2] due to possible microgravity effectsafter 4 days In addition these authors reported differencesbetween the 1 g ground sample and the 1 g in-flight controlswith over 200 DEGs [2] Also Zhang et al [32] observed agreater difference between flight and ground samples withrespect to 1 g in-flight conditionsThese observations indicatethat the differing results could be related to the organismsinvestigated the time of exposure hardware experimentalparameters and set-up

In this study we report on results of a spaceflightexperimentThis experiment was part of the Simbox (Sciencein Microgravity Box) mission a joint project between thespace agencies from Germany (Deutsches Zentrum fur Luft-und Raumfahrt e V) and China (China Manned SpaceEngineering) in November 2011 As one out of 17 biologicalexperiments semisolid callus cultures of Arabidopsis wereexposed to a 17-day spaceflight on board of the Chinesespacecraft Shenzhou 8 Due to reduced viability after longerperiods of exposure within the flight hardware the calluscultures were metabolically quenched after 5 days in spaceResults of a whole-genomemicroarray screening (120583g exposedsamples 1 g in-flight samples kept in a reference centrifugeand 1 g ground samples) revealed major differences betweenboth 1 g controls but a minor impact of microgravity


21 Experiment-Specific Hardware (HW) The Simbox wasa modification of the Biobox-6 [33 34] which was devel-oped for unmanned recoverable capsules and space shuttlemissions Development and production were carried out byAstriumEADS Friedrichshafen Germany [35] This incu-bator (size of 461 times 551 times 273mm internal volume of34 L max power consumption of 130W and empty mass


Reference centrifuge


Position SP04 [FSGS]

Position C05 [FC]Inside of incubator housing

Figure 1 Photograph of the inside of the Simbox incubator usedwithin the flightground experiment (housing removed) The rotorof the reference centrifuge (position C05 for sample group FC) isindicated by a circle The static experimental platform is in themiddle and outside of the centrifuge rotor (position SP04 for samplegroup FS within the flight experiment and GS within the groundexperiment resp) (photograph DLRAstrium)

Slide with spikes

Window of CC biofoil and frame removed

Front CC

Substrate holder

Callus cultures

12a media with agar

Figure 2 Photograph of the inside of one culture chamber (CC)(experiment container (EC) window biofoil and frame removed)The semisolid callus cultures were positioned on substrate holders(slides) with plastic spikes on 12a agar containing culture media

of 17 kg) served as carrier for an experimentstatic platformwith an integrated centrifuge rotor (provides 1 g in-flightcontrol) The Simbox incubator (Figure 1) enabled samplecultivation at 22ndash24∘C (nominal temperature range) and 30ndash40 humidity throughout the mission A duplicate model ofthe Simbox was constructed for the ground experiment Ourbiological approach (experiment number 16) was realizedby means of three fully automated type V Experiment UnitEnvelopes (EUE plant cultivation unit without illumina-tion) EUEs consisted of support housing made of polyether-ketone with two culture chambers each (front and rear CC317 times 24 times 143mm plusmn 015mm) Our biological material waspositioned on substrate holders (slides) with plastic spikes(Figure 2) The latter were needed to keep the cultures inplace In order to allow gas exchange the CCs were sealedwith a biofoil made of polysulfone (Tecason S PolysulfoneEnsinger Inc Washington-Pennsylvania USA) In additiona peristaltic pump (flow rate of ge243mLmin) was used to


Window frame of CC with glued biofoil

Substrate holder slide with spikes

Rear CC Front CCType 5 EUE

housingCavity cover Type 1 EC

Fixativewaste tank Tubing system Liquid pump

Fittings for tube connection

Figure 3 Photograph of the fully automated plant cultivation unit type V EUE (left side) and EC removed (right side) (photographDLRAstrium)

connect the CC to a fixativewaste unit (volume 203mL plusmn05mL) EUEs were accommodated inside type I ExperimentContainers (ECs) (Figure 3) Via sensors parameters such astemperature humidity CO

2 and O

2content as well as acti-

vation of the pump system were recorded and transmitted

22 Cell Cultures Sterile cuttings (about 50mm long) ofstems of wild type Arabidopsis thaliana (cv Columbia Col-0) plants were used for callus formation on 12a media [36]containing 1 agar (Sigma-Aldrich Germany) Calli weretransferred to 500mL Erlenmeyer flasks with 200mL liquid12a medium and cultivated under sterile conditions at 23∘Cin the dark on a rotary shaker (130 rpm Infors BottmingenSwitzerland) as described previously [14] New medium wasadded every week to the resulting cell suspension Eightmonths before the Simbox mission an aliquot of this culture(3 g) was spread on 6 cm Petri dishes (Greiner Bio-OneFrickenhausen Germany) containing agar and 12a mediumCell cultures were mailed to the Institute of Physiologyand Ecology Shanghai (Laboratory of Prof Zheng) and thecultivation continued (as liquid suspension) as describedabove These suspension cultures were transferred to thePITC (Payload Integration and Test Center Beijing China)The cultivation was then continued on agar plates (see above)and finally these semisolid calli were brought to the launchsite (Jiuquan Satellite Launch Center Jiuquan China) byplane

23 Preparation of Final Experiment Configuration Oneday before the launch 11-day-old semisolid callus cultureswere transferred into the CCs with 2mL agar containingmedium (Figures 2 and 3) Two ECs were used for thespaceflight (flight models FM 16001 and FM 16002) and one

for the ground experiment (FM 16003) respectively One ofthe two ECs was contained in the centrifuge rotor and theother one was fixed at the experimentstatic platform (flightplatform) respectively (Figure 1) Metabolic quenching ofthe samples was by the injection of RNAlater (Ambion LifeTechnologiesDarmstadtGermany)This reagent is also usedto stabilize nucleic acids Twenty mL of this fixative wasfilled into the fixativewaste unit attached to the bottom ofthe EC Between handover and integration into the Simboxflightground incubator the ECs were stored at nominallaboratory temperature conditions (22ndash24∘C) The Simboxincubator was unpowered for about 3 hours during transportto the spacecraft During this time the lowest temperaturewas 21∘C (Figure 4)

24 The Experiment in Orbit The Simbox was launched onboard of the unmanned spacecraft Shenzhou 8 on Octo-ber 31 2011 at 2158 UTC (universal time coordinated)with a Long March 2F rocket from the cosmodrome inJSLC The precise mission timings including sample fixationtime points are illustrated in Figure 5 (for a gravity-levelprofile see Supplementary Material S1 available online athttpdxdoiorg1011552014547495) Experiment zero time(EZT) was set when the spacecraft reached the orbit At EZTthe centrifuge was activated to run with 7440 rpm Withinthe spacecraft the oxygen partial pressure ranged from1804 to 2732 kPa and the carbon dioxide partial pressurewas between minus003 and 046 kPa Radiation measurementsyielded a total dose of 593 to 81mSv and an equivalent doseof 037 to 051mSvd near the Simbox incubator (telemetrydata Chinese authorities personal communication) Thepump system was activated after 5 days in space and injectedthe fixative solution from the fixativewaste unit into the CCrsquos


Unpowered transport toShenzhou

Lowest temp

Launch215800(UTC) N

omin

al te

mp

rang

e

Time (hhmm)

Power supply on

Power supply off

2500

2450

2400

2350

2300

2250

2200

2150

2100

2050

22000000 0130 0300 0430 0600 0730 0900 1030 1200

rarr21∘C

TP1TP2TP3

Tem

pera

ture

(∘C)

3h 06min

Figure 4 Temperature profile as recorded by 3 temperature sensors (TP1-3) attached to the Simbox incubator during integration of ECs intothe incubator transport to Shenzhou and launch (data Astrium)

Figure 5 Precise mission timeline of the experiment in orbit (grey) and related ground experiment (white) Universal time coordinated(UTC) time units are given in hoursminutes seconds experimental zero time (EZT) Arrowheads (nabla) indicate sample fixation time pointsof sample groups FS FC and GS respectively

of FMs This yielded a final RNAlater concentration of about90 (vv) after mixing Temperature in CCs was kept at anominal range of 22 to 24∘C before during and after fixation(Figure 6) After 17 days in space the spacecraftwas separatedfrom Tjangong-1 and touched ground on November 17 2011

After landing and recovery of the capsule samples wereretrieved within 6 hours The ECs were disassembled andstored around 4∘C until they arrived in Tubingen on Novem-ber 25 2011 In the home laboratory calli were harvested andstored at minus80∘C until processing


Nom

inal

tem

p ra

nge

Transport to spacecraft (lowest

FC

FS2500

2450

2400

2350

2300

2250

2200

2150

21000000 4800 9600 14400 19200 24000 28800 33600 38400 43200

TX1

Time (hhmm)

TX2TX3TX4

EZT + 1202800

EZT + 1283830Te

mpe

ratu

re (∘

C)

temp 21∘C)

Figure 6 Temperature profile as recorded by 4 temperature sensors (TX1-4) attached to the Simbox incubator during the whole Simboxmission (data Astrium) Sample fixation time points for the spaceflight samples (FS and FC) are indicated by arrowheads (grey triangle)

25 Ground Control Immediately after the launch the labo-ratory equipment and cell cultures were brought back to thePITC by Chinese scientists The ground experiment startedwith a one-day delay on November 2 2011 (Figure 5) TheEUE was integrated into the Simbox duplicate accordingto the position in the flight incubator (experimentstaticplatform) and kept at 23∘C As in the experiment in orbitsamples weremetabolically quenched after 5 days (November7) The ground experiment ended on November 19 Thesamples were handled as described for the experiment inorbit

26 Experiment Conditions and Specification of GeneratedSamples During the Simbox mission the samples wereexposed to different experimental conditions In the experi-ment in orbit FM 16002 was attached to the static platform ofthe Simbox incubator and experienced 5 days ofmicrogravity(group FS Flight Static) FM 16001 was centrifuged resultingin a 1 g control (group FC in-flight centrifugation) In theground experiment the same experimental design was usedFM 16003 was fixed to the static platform (group GS groundstatic) In summary we obtained one biological sample perCC resulting in two replicates for each FM (front and rearCC) and for each experimental condition respectively

27 Isolation of Total RNA and High-Density OligonucleotideArrays Total RNA was extracted using the RNeasy Plus kit(Qiagen Hilden Germany) according to the manufacturerrsquos

instructions Quantity and quality controls were performedand samples were processed using theMessageAmp II-BiotinEnhanced Single Round aRNA Amplification Kit (AmbionLife Technologies Darmstadt Germany) as described earlier[21 37] Fragmented biotin-labeled aRNA was then submit-ted to a high throughputmicroarray analysis (GeneChipAra-bidopsis ATH1 Genome Array Ref 510690 LOT 4155830Affymetrix Santa Clara California USA) Hybridizationwasperformed according to the manufacturerrsquos instructions (fordetails see httpwwwaffymetrixcomsupporttechnicalmanualsaffx) The Affymetrix protocol EukGE-WS2 V4 wasused for washing and staining procedures

28 Gene Expression Analysis Expression data were calcu-lated from raw values of the detected signal intensity ofhybridization events of all spotted probe sets and savedas CEL data files Microarray data are available in theArrayExpress database (httpwwwebiacukarrayexpress[38]) under accession number E-MTAB-2518 For integrativedata analysis we used the open-source software Mayday[39] Normalization was performed using the robust multi-array average method of background-adjustment quantile-normalization andmedian-polish to ensure comparability ofarrays and estimate log

2expression values [40ndash42] Hierar-

chical clustering was performed by means of the neighbourjoining method [43] in order to reconstruct and visualizerelationships within expression values due to experimentconditions The Pearson Correlation coefficient was used to


Experiment in orbit [FS]

(a)

Experiment in orbit [FC]

(b)

Ground experiment [GS]

(c)

Figure 7 Photograph of Arabidopsis thaliana semisolidcallus cultures after a 5-day 120583g cultivation in orbit ((a) FS) 1 g in-flight cultivation((b) FC) or on ground ((c) GS) The photographs were taken after fixation by RNAlater and recovery of the spacecraft

calculate the distance between each experimental condition(FS FC andGS) and biological replicates (front and rearCC)The matrix of variant genes was filtered and subjected to aStudentrsquos 119905-test (119875 le 01) with combined false discovery rate(FDR) correction to identify significantly altered transcripts(119875 lt 01) between the sample groups FS and FC FS andGS and FC and GS respectively Differentially expressedgenes were determined by fold change (fc) calculation of log

2

transformed expression dataThereby the thresholdwas set atminus1ge log

2(fold change)ge 1 for at least 2-fold altered transcripts

[40 41 44] Additionally the Affymetrix probe identifierswere tested by Gene Set Enrichment Analysis (GSEA [45])for enrichment of functional ontologies usingGeneOntologyterms [46] within Mayday Thereby we focused on genesthat share their function in identical biological processes forinterpreting the genome-wide expression profiles

3 Results

The aim of this experiment was to characterize the transcrip-tome of Arabidopsis semisolid callus cultures after 5 daysin space Due to the availability of an in-flight centrifugeit was possible to compare expression data with (a) realmicrogravity samples (thought to yield the microgravityrelated alterations) and with (b) those from the groundcontrols (which should deliver effects of nonmicrogravityrelated spaceflight conditions) This was achieved with high-density oligonucleotide arrays

31 Performance of Hardware and Biological Material Thehardware was thoroughly tested in order to retain viabilityof the callus cultures for as long as possible These tests werefocused on the biocompatibility of the used materials gas-exchange properties of membranes and viability of the cellcultures under the cultivation conditions within the EC Wealso recorded the oxygen content within the CC [37] As this

declined from 8 to about 2mgL after 5 days automated sam-ple fixationwas set at day 5 after take-offMission parameterssuch as temperature were within nominal range during themission Radiation measurements recorded increased valuesAfter landing and return of the biological material to theUniversity of Tubingen (Germany) the samples were visuallychecked The fixed calli showed good morphology and hadwell grown during the initial culture of 5 days in space Thecalli from the 1 g controls (flight and ground experiment)were smaller compared to those exposed to microgravity(Figure 7)

32 Biology of Samples and Gene Expression Analysis Thequality of the extracted total ribonucleic acid was satisfyingfor GeneChip hybridization (for RNA quality see Supple-mentary Material S2) with clear bands representing the 28Sand 18S rRNA Whole-genome microarray screening wasperformed for each sample Due to the limited amount oftotal RNA the confirmation of expression data by quan-titative real-time PCR was not possible The data analysisrevealed experiment-specific properties of biological repli-cates which were visualized by hierarchical clustering on thebasis of the calculation of the Pearson Correlation coefficient(Figure 8) In this graph a relatively short distance impliesa high correlation between the samples As obvious fromFigure 8 the flight and ground experiment showed group-based clustering The short distance between FS and FC (FSand FC boxes) in contrast to GS (GS boxes) indicates thatnonmicrogravity related spaceflight conditions have majorimpact The transcriptome of the biological replicates withinthe experiment groups (front and rear chamber of FS FC GS119899 = 2) showed a high degree of similarity (Figure 8)This factwas confirmed by heat map generation based on calculatedcorrelations (Figure 9) The Pearson Correlation was about099 between front and rear CC for all three modules


FC re

ar

FC front

GS

front

GS rear

Flight experimentGround experiment

FS rear

FS front

Figure 8 Hierarchical clustering bymeans of the neighbour joiningmethod of generated sample groups (white ground experiment GSground static grey flight experiment FS flight space FC in-flightcentrifugation) Each EUE consisted of two culture chambers (frontand rear chambers illustrated by boxes)

(FS FC and GS 119899 = 2 Figure 9) Statistical (Studentrsquos 119905-test 119875 lt 01 and FDR correction) and comparative analysisshowed a relatively low response of semisolid callus cultures(Figure 10) Interestingly microgravity conditions did notinduce statistically significant changes (119875 lt 01) at thegene expression level although 298 genes were at least 2-fold differentially expressed (275 up- and 23 downregulated)within flight space (FS) samples In contrast nonmicro-gravity related spaceflight conditions interfered with geneexpression considerably Eight hundred ninety-seven geneswere significantly and differentially expressed (at least 2-fold119875 lt 01) when 1 g ground and 120583g exposed flight sampleswere compared Among them 463 were upregulated and 434geneswere downregulatedwithin FS (Figure 10) Comparisonbetween both 1 g controls (in-flight ground) resulted in 826significantly (119875 lt 01) differentially altered genes (543 up and283 downregulated Figure 10)Thereby 573 significant DEGs(119875 lt 01) were identical in both comparisons (Figure 10)

33 Identification of Altered Genes after Long-Term Micro-gravity For detection of gene expression changes due to 120583gexposure we compared data generated out of the samplegroups flight space (FS) and in-flight centrifugation (FC)Two hundred seventy-five genes were at least 2-fold differ-entially upregulated and 23 downregulated (Figure 10) Theapplication of statistics showed that there were no significant(119875 lt 01) alterations at the expression level after 5 daysin space By means of a Gene Ontology [46] based GeneSet Enrichment Analysis (GSEA) the DEGs were related tocommon biological processes In order to identify processeswhich are specifically influenced by microgravity conditionswe compared overrepresented processes that were identicalbetween sample group FS versus FC and FS compared toGS (Table 1) Most prominent were effects on the translationmachinery (Table 1 gene set number 24) Interestingly allgenes that were differentially upregulated and involved in

Ground static [GS front]

Ground static [GS rear]

Flight centrifugation [FC rear]

Flight centrifugation [FC front]

Flight space [FS rear]

Flight space [FS front]

Gro

und

stat

ic [G

S fro

nt]

Gro

und

stat

ic [G

S re

ar]

Flig

ht ce

ntrif

ugat

ion

[FC

rear

]

Flig

ht ce

ntrif

ugat

ion

[FC

front

]

Flig

ht sp

ace [

FS re

ar]

Flig

ht sp

ace [

FS fr

ont]

1 0994 0972 0973 0973 0974

0994 1 0981 098 0979 0983

0972 0981 1 0995 0987 099

0973 098 0995 1 0992 0991

0973 0979 0987 0992 1 0996

0974 0983 099 0991 0996 1

Figure 9 Pearson correlation heat map shows high degree ofsimilarity between front and rear culture chamber of each samplewithin each sample group Flight space (FS) in-flight centrifugation(FC) and ground static (GS)

translation processes were chloroplast-encoded This geneset comprises genes coding for several protein subunits andcomponents of ribosomes (eg ATCG00065 ATCG00660ATCG00770 andATCG00790) but also the nucleus-encodedtranslation initiation factor EIF-5A (AT1G13950) that is wellknown to regulate translation initiation and terminationwithin the cytoplasma of eukaryotes (Table 2) The otherpart of identified differentially upregulated genes is involvedin electron transport chains located within mitochondria(Table 1 gene sets number 4 8 and 11) such as subunits ofthe NADH dehydrogenase multi-enzyme complex of the res-piratory chain (ATMG00650 ATMG00070 ATMG00580)(Table 2) Mitochondrial electron transport is connectedto the production of adenosine triphosphate (ATP) Thusthe gene set representative for ATP biosynthesis was alsopart of the DEGs (ATCG00120 ATMG00410 ATCG00480and ATCG00150) (Table 2) Within the 23 downregulatedgenes (at least 2-fold) no special gene sets could befound but the largest group codes for heat shock pro-teins (AT4G27670 AT2G29500 AT5G12020 AT5G59720AT4G25200 AT1G53540 and AT5G12030)

34 Attempt to Distinguish between Effects of Microgravityand Nonmicrogravity Related Spaceflight Conditions on GeneExpression One aim of this investigation was to separateresponses to microgravity from those of nonmicrogravityrelated spaceflight conditions Until today onlymarginal dataexist about these effects on plants in spaceThus we screenedfor genes that were significantly (119875 lt 01) altered withinspaceflight samples (FS and FC) compared to the 1 g groundcontrol and were identical between FS and FC comparedto GS This overlap yielded 573 significantly altered (119875 lt01) DEGs (Figure 10) The GSEA of these genes representeddiverse biological processes (Table 1 bold font) The majority


Flig

ht

Gro

und

897 in total 20 in total 826 in total 155 in total

DEGs

298 in totalFlight space

FS front FS rear

Flight centrifugation

FC front FC rear

Ground static

GS front GS rear

573 in total 9 in total

Sign DEGs (P lt 01)

Uncorrected P value

Sign DEGs (P lt 01)

Uncorrected P value FDR-corrected P value

Sign DEGs (P lt 01)

Uncorrected P value FDR-corrected P valueFDR-corrected P value

463uarr434darr in FS 13uarr7darr in FS 543uarr283darr in FC 96uarr59darr in FC

275uarr23darr in FS

Figure 10 Overview of the number of differentially (fold change (fc) at least 2) and significantly expressed genes (DEGs 119875 lt 01) withinthe flight (grey) and ground (white) experiment The different sample groups are illustrated by boxes Up- and downregulated transcripts aresymbolized by arrows behind the number of altered genes Genes that are significantly (119875 lt 01) differentially expressed are shown in boxesframed in black (bold lines)

of these genes could be related to intracellular signalingpathways such as mitogen-activated protein kinase (MAPK)cascades and protein phosphorylation (Table 1 gene setnumber 6 and 12) Included were different MAP kinases (egAT1G01560 AT1G73500) serinethreoninetyrosine kinases(eg AT1G20650 AT5G16900 and AT4G38470) and manyother kinases (Table 3) Furthermore we identified genescoding for members of the calcium-binding EF-hand proteinfamily (AT3G01830 AT3G47480) and the WRKY transcrip-tion factors 54 70 and 38 (AT2G40750 AT3G56400 andAT5G22570) that have also transcription regulation activity(Table 3) Additionally the spaceflight environment otherthan microgravity had a significant (119875 lt 01) impact ongeneral stress-responsive (gene set number 20) and defense-related genes (3) especially those involved in the response tooxidative stress and respiratory burst responses (21) Theseare peroxidases 21 4 52 and 25 (AT2G37130 AT1G14540AT5G05340 and AT2G41480) catalase 3 (AT1G20620) andreceptor-like kinases (AT5G46330 AT2G19190) The lattercan be induced upon contact with the bacterial proteinflagellin which is an important elicitor of the plant defenseresponse These kinases are also important members ofthe MAP kinase signaling cascade Furthermore generalmetabolic processes (gene set number 7) protein targeting(13) and rRNA processing (21) were overrep-resented due tononmicrogravity related conditions in space

4 Discussion

The expression data of Arabidopsis semisolid callus culturesshow alterations in differential gene expression in responseto microgravity However the influence of the spaceflightenvironment in addition to microgravity is significant

41 Identification of Altered Genes after 5 Days of Micro-gravity Comparison between microgravity and 1 g spacecontrols revealed about 298 differentially (but not signifi-cantly) expressed genes This number is low in comparisonto short-term exposures to microgravity within a rangeof minutes (TEXUS 47 sounding rocket experiment [47])or seconds (14 DLR parabolic flight campaign [21]) Thisfinding could be due to the small number of biologicalreplicates (2 biological replicates only due to limited mate-rial and hardware) However similar observations are alsoreported by others After 4 days in space Arabidopsis plantsexhibited only 27 transcripts which were at least 2-foldaltered at their expression level [2] This might indicate thatplants respond immediately to a microgravity environmentbut then adapt to the new situation on the longer runAlso Zhang et al [32] could also identify only 45 proteinschanged in expression after 14 days in space (same mission)Genes with prolonged changes in expression could howeverprovide important information about the physiological needs


Table 1 Visualization of enriched Gene Ontology categorization terms (GSEA Gene Set Enrichment Analysis of biological processes) Genesets identical in FSFC and FSGS are not colored the ones identical in FSGS FCGS and the overlap of both are in bold font (FS = flightspace FC = flight centrifugation and GS = ground static)

Number Enriched gene set (biological process) FSFC FSGS FCGS OverlapGene set size

1 ATP catabolic process 0 8 8 72 ATP biosynthetic process 10 9 0 03 Defense response 0 20 26 144 Mitochondrial electron transport chain 7 7 0 05 Lipid metabolic process 0 8 7 66 MAPK cascade 0 29 36 277 Metabolic process 0 25 20 168 Mitochondrial electron transport 11 11 0 09 Oxidation-reduction process 0 13 12 810 Photosynthesis light harvesting 0 5 5 511 Photosynthetic electron transport chain 5 5 0 012 Protein phosphorylation 0 23 31 1813 Protein targeting to membrane 0 12 13 10

14 Regulation of transcriptionDNA-dependent 0 12 11 8

15 Respiratory burst involved in defenseresponse 0 22 26 21

16 Response to chitin 0 7 6 617 Response to ethylene stimulus 0 5 6 518 Response to hypoxia 0 6 9 619 Response to oxidative stress 0 15 13 920 Response to stress 0 9 9 621 rRNA processing 0 16 15 1422 Toxin catabolic process 0 7 7 623 Transition metal ion transport 0 10 12 824 Translation 27 28 0 0

25 Two-component signal transductionsystem 0 6 5 5

after a few days in space These include an upregulatedgroup of genes which code for proteins that constitute theribosomal complex within plastids These are necessary fortranslation of mRNAThe upregulation of the mitochondrialelectron transport chain could indicate an increased need forATP The upregulated expression of NADH dehydrogenasecould have the same reason Interestingly gene productsinvolved in processes like the response to stress proteindegradation or programmed cell death appeared not to bealtered in expression The involvement of a series of geneswith still unknown functions (not shown) suggests that thespace environment induces also unknown cellular processesTogether with the fact that there were no significant changesin gene expression detectable after 5 days of microgravity letsus suggest that at this stage the impact of a lack of gravitationon cell physiology was not too heavyThe space environmentper se however causes possibly an increased energy demandas shown by the upregulation of respiratory componentsThisaspect should be taken into consideration when plants will

be used to provide nutrients oxygen and energy on longduration space missions

Heat shock proteins (HSPs) dominate the group oftranscripts which are reduced in amount (not shown) Theseproteins are involved in many forms of stress response Theyenable the folding and membrane translocation of proteinsand are thought to reconstitute the tertiary structure ofproteins affected by stress events This way they can increasethe stress tolerance A decreased expression (our study)should thus indicate a lower number of proteins affectedin their structure and was also reported for Arabidopsis invitro callus cultures under simulatedmicrogravity conditions(magnetic levitation magnetic field strength 101 Tesla) [48]as well as for the single-cell system of the fern Ceratopterisrichardii [9] There are however also reports on increasedexpression of HSPs [15 16 21 24]

A group of plant genes which are always affected byaltered gravity are those involved in cell wall modification[2 49ndash51] This reflects the need for increased stability


Table 2 Differentially expressed genes (fold change (fc) at least 2) within the sample group flight space (FS frontrear CC) compared toin-flight centrifugation (FC) Samples taken after 5-day cultivation at microgravity and sorted according to the overrepresented biologicalprocesses identified by GSEA to be the most prominent

Number ATG number Gene namedescription log (fc) Enriched Gene set (biological process)1 ATCG00065 Ribosomal protein S12 236 Translation2 ATCG00660 Ribosomal protein L20 214 Translation3 ATCG00770 30S ribosomal protein S8 196 Translation4 ATCG00160 Ribosomal protein S2 184 Translation5 ATCG00790 Ribosomal protein L16 18 Translation6 ATCG00780 Ribosomal protein L14 163 Translation7 AT1G13950 Eukaryotic translation initiation factor 5A-1 114 Translation8 ATCG01120 Ribosomal protein S15 111 Translation9 ATCG00750 Ribosomal protein S11 105 Translation10 ATCG00800 Ribosomal protein S3 104 Translation11 ATMG00650 NADH dehydrogenase subunit 4L 23 Mitochondrial electron transport12 ATMG00060 NADH dehydrogenase subunit 5 184 Mitochondrial electron transport13 AT2G07751 NADH-ubiquinoneplastochinone oxidoreductase 175 Mitochondrial electron transport14 ATCG01050 Subunit of NAD(P)H dehydrogenase complex 174 Mitochondrial electron transport15 ATMG00160 Cytochrome c oxidase subunit 2 166 Mitochondrial electron transport16 ATMG00070 NADH dehydrogenase subunit 9 15 Mitochondrial electron transport17 ATCG00420 NADH dehydrogenase subunit J 143 Mitochondrial electron transport18 ATCG01250 NADH dehydrogenase ND2 125 Mitochondrial electron transport19 ATMG00510 NADH dehydrogenase subunit 7 124 Mitochondrial electron transport20 ATMG00270 NADH dehydrogenase subunit 6 124 Mitochondrial electron transport21 ATMG00580 NADH dehydrogenase subunit 4 119 Mitochondrial electron transport22 ATCG01070 NADH dehydrogenase ND4L 113 Mitochondrial electron transport23 ATCG00120 ATPase 120572-subunit 215 ATP biosynthesis24 ATCG00140 ATPase III subunit 159 ATP biosynthesis25 ATMG00410 ATPase subunit 6 156 ATP biosynthesis26 ATCG00130 ATPase F subunit 147 ATP biosynthesis27 ATCG00480 120573-Subunit of ATP synthase 133 ATP biosynthesis28 ATCG00150 Subunit of ATPase complex CF0 112 ATP biosynthesis

(hypergravity) or more flexibility (microgravity) In thepresent study expression of expansins (cell wall loosening)is increased (not shown) This might be the reason for theenhanced size of the microgravity cultures when comparedto the 1 g controls (Figure 7)

42 Impact of the Nonmicrogravity Related Spaceflight Condi-tions on Gene Expression The availability of a 1 g referencecentrifuge enabled us to screen for genes affected by nonmi-crogravity related spaceflight conditions in that we comparedexpression data between 120583g exposed and 1 g space with 1 gground samples This resulted in a considerable numberof identical genes altered in mRNA abundance (573 genes)(Figure 10) We thus assume that this could be due to effectsof spaceflight-related environmental conditions includingspace radiation Radiation measurements inside the capsulein a position close to our samples yielded a total dose of 59to 81mSV (milliSieverts) and an equivalent dose of 037 to051mSVd (data Chinese authorities) This is considerablymore compared to terrestrial conditions (1 to 2mSVa) and

could be one of the reasons for the alterations at transcriptlevels obviously not related to 120583g Also Zhang et al [32]reported a greater difference on protein expression of non-120583gconditions Analysis showed that both experimental condi-tions (120583g and non-120583g spaceflight conditions) affect differentbiological processes (Table 1) Overrepresented processesshould not be regarded separately as they are closely linkedtogether within a plant cell For example the formation ofreactive oxygen species (ROS) is one of the initial responsesupon most kinds of stresses They are also produced asby-products of redox reactions They are important secondmessengers as well as toxic species and their cellular levelsare closely controlled by detoxification systems [52ndash54] Therole of ROS in response to environmental changes canhowever also be deduced from alterations in gene productsinvolved in ROS production and turnover In this studywe observed that many ROS-related genes are significantlyregulated (Table 3) These comprise peroxidases catalaseand a glutathione S-transferase (Table 3) These proteins aresuggested to be part of the stress-induced antioxidant sys-tem [55] Glutathione S-transferases also possess peroxidase


Table 3 Differentially (at least 2-fold) and significantly expressed genes (119875 lt 01 573 in total) that are identical between flight space (FS) aswell as in-flight centrifugation (FC) compared to ground static (GS) Changes are due to nonmicrogravity related spaceflight conditions Thegenes are sorted according to the overrepresented biological processes identified by GSEA to be most prominent



1 AT1G01560 MAP kinase 11 183 (0034) 213 (0027) MAPK cascade2 AT1G73500 MAP kinase 9 134 (0006) 137 (0038) MAPK cascade






8 AT3G15500 NAC-domain containingtranscription factor 3 298 (572119864 minus 4) 263 (0003) MAPK cascade

9 AT1G35670Calcium-dependentcalmodulin-independent proteinkinase 2

12 (0002) 123 (0003) Proteinphosphorylation

10 AT1G20650 Serinethreonine protein kinasesuperfamily protein minus14 (0024) minus15 (0018) Protein

phosphorylation

11 AT3G61160 Serinethreonine protein kinasefamily protein minus122 (0007) minus14 (0008) Protein

phosphorylation

12 AT1G78290 Serinethreonine protein kinasefamily protein 2C 171 (0019) 20 (0034) Protein

phosphorylation

13 AT4G18640 Serinethreonine protein kinasefamily protein 108 (0019) 107 (0014) Protein

phosphorylation

14 AT4G18950 Serinethreoninetyrosineprotein kinase family protein 253 (0031) 317 (002) Protein

phosphorylation


phosphorylation


phosphorylation


phosphorylation


phosphorylation

19 AT4G38470 Tyrosine kinase family protein 46 114 (0008) 134 (0015) Proteinphosphorylation

20 AT1G69790 Protein kinase superfamilyprotein 119 (0038) 112 (0009) Protein

phosphorylation

21 AT5G53450 Protein kinase 188 (0088) 189 (0075) Proteinphosphorylation

22 AT1G51620 Protein kinase family protein 18 (0052) 231 (0048) Proteinphosphorylation

23 AT3G04530 Phosphoenolpyruvatecarboxylase kinase 2 minus16 (006) minus119 (043) Protein

phosphorylation

24 AT5G63650 Protein kinase 25 minus126 (0028) minus101 (0032) Proteinphosphorylation

25 AT1G16260 Cell-wall associated proteinkinase family protein 173 (0006) 213 (0003) Protein

phosphorylation

26 AT1G68690 Proline-rich extension-likereceptor kinase family protein 104 (0002) 103 (004) Protein

phosphorylation


Table 3 Continued



27 AT5G46330 Flagellin 2-induced receptor-likekinase minus185 (0043) minus238 (0016) Defense response

28 AT2G19190 Flagellin 22-inducedreceptor-like kinase 248 (0065) 227 (0075) Defense response

29 AT2G15120 Disease-resistance family protein 268 (0035) 253 (004) Defense response30 AT1G59780 Disease resistance protein 137 (0092) 198 (0052) Defense response31 AT1G63880 Disease resistance protein minus181 (0003) minus179 (0016) Defense response

32 AT2G39200

Transmembranedomain-containing proteinsimilar to mildew resistanceprotein 12

26 (0059) 255 (0063) Defense response

33 AT1G19610 Pathogenesis-related protein 14 minus217 (0002) minus214 (0029) Defense response

34 AT3G20600 Nonrace specific diseaseresistance protein 105 (0037) 212 (0011) Defense response

35 AT1G02360 Chitinase family protein 26 (0026) 287 (0019) Defense response36 AT3G54420 Chitinase family protein class IV 173 (0055) 253 (0026) Defense response

37 AT4G21390 Serinethreonine protein kinasefamily protein 15 (0068) 186 (0031) Defense response

38 AT3G46280 Protein kinase family protein 183 (0074) 23 (0048) Defense response39 AT5G35750 Histidine kinase 2 minus121 (0042) minus135 (0026) Defense response




43 AT4G37530 Peroxidase family protein 217 (0035) 215 (0026) Response tooxidative stress


45 AT1G20620 Catalase 3 minus112 (007) minus139 (0052) Response tooxidative stress

46 AT2G29490 Glutathione S-transferase 19 classtau 1 175 (007) 17 (0073) Response to

oxidative stress

47 AT3G22370 Oxidase family protein 13 (0013) 10 (0089) Response tooxidative stress

48 AT4G37220 Stress-responsive protein 287 (0004) 187 (0049) Response to stress49 AT4G21870 Heat shock protein 265 minus131 (0002) minus143 (0012) Response to stress

50 AT2G38750 Calcium-dependentphospholipid binding protein 148 (002) 115 (0032) Response to stress

activity and can thus prevent cell damage by peroxides suchas hydrogen peroxide [56 57]The increase in detoxification-related transcripts appears reasonable as radiation in orbitconsists of highly energetic (HZE) particles from interplan-etary galactic sources or results from solar particle eventswhich could have an impact on cells [58ndash62] Wan et al [6364] showed that X-rays 120574-rays protons and heavy chargedparticles increased oxidative stress in different cell types andcountermeasures for space radiation effects are the use ofantioxidants [62] Similar responses are probable for plantcells Therefore the impact of long-term space radiation on

the transcriptome of Arabidopsis should be investigated inground-based studies in simulation testbeds for the spaceenvironments [59]

In addition a range of WRKY transcription factors andcomponents of signaling chains (Ca2+-dependent proteinsMAP kinases) were identified (Table 3) These responsivekinases (Table 3) are potentially also modulated by cytosolicfluctuations of H

2O2and can thus be part of signal trans-

duction chains starting from hydrogen peroxide (for defense-related genes in tomato see Orozco-Cardenas et al [65]) Incontrast to other observations to altered gravitation [2 15]


in this study genes which are defense- resistance- andpathogen-related are significantly altered due to non-120583grelated spaceflight conditions

5 Conclusions

In this study gene expression changes within Arabidopsiswild type semisolid callus cultures were investigated aftera 5-day spaceflight and compared to on-board and groundcontrols Faced with limited HW capacities (only 3 EUEs)and small amounts of biological material (119899 = 2 foreach sample group) high-density oligonucleotide arrays wereused to screen for changes at the gene expression level Forfuture investigations it would thus be desirable to have flightrepetitions and an adequate amount of samples for addi-tional analysis (eg qPCR) Unexpectedly the response ofcallus cultures to long-term microgravity was less prominentcompared to nonmicrogravity related spaceflight conditionsThe latter including space radiation induced differential andsignificant expression changes of transcripts that are involvedin the stress-induced antioxidant system signalling chainsand defense-resistance-related genes These findings clearlyhighlight that the use of an in-flight reference centrifuge (1 gin-flight control) should be mandatory during space flightmissions



Acknowledgments

Thisworkwas supported by a grant of theDeutsches Zentrumfur Luft- und Raumfahrt (DLR) (Grant no 50WB0723) toRudiger Hampp The authors are indebted to Dr MarkusBraun (DLR) for perfect campaign organization and toAchim Schwarzwalder Dr Astrid Horn and the EADSAstrium team for hardware construction and technical sup-port They thank the China Manned Space Engineering andthe Chinese scientists especially Professor Zheng for goodcooperation at launch site They are grateful to Margret Eckefor skilful production and maintenance of the cell culturesand Fabian Bergwitz for assistance in China as well as AnneHennig for ground-based experiments before the mission

References

[1] G Perbal and D Driss-Ecole ldquoMechanotransduction ingravisensing cellsrdquo Trends in Plant Science vol 8 no 10 pp498ndash504 2003

[2] M J Correll T P Pyle K D L Millar et al ldquoTranscriptomeanalyses of Arabidopsis thaliana seedlings grown in spaceimplications for gravity-responsive genesrdquo Planta vol 238 no3 pp 519ndash533 2013

[3] M Braun ldquoGravitropism in tip-growing cellsrdquo Planta vol 203no 1 pp S11ndashS19 1997

[4] M Braun B Buchen and A Sievers ldquoElectron microscopicanalysis of gravisensing Chara rhizoids developed undermicro-gravity conditionsrdquo The FASEB Journal vol 13 no 8 pp S113ndashS120 1999

[5] M Braun ldquoGravity perception requires statoliths settled onspecific plasma membrane areas in characean rhizoids andprotonematardquo Protoplasma vol 219 no 3-4 pp 150ndash159 2002

[6] M Braun J Hauslage A Czogalla and C Limbach ldquoTip-localized actin polymerization and remodeling reflected by thelocalization of ADF profilin and villin are fundamental forgravity-sensing and polar growth in characean rhizoidsrdquoPlantavol 219 no 3 pp 379ndash388 2004

[7] M Braun and C Limbach ldquoRhizoids and protonemata ofcharacean algae model cells for research on polarized growthand plant gravity sensingrdquo Protoplasma vol 229 no 2ndash4 pp133ndash142 2006

[8] M L Salmi T J Bushart and S J Roux ldquoAutonomous gravityperception and responses of single plant cellsrdquo Gravitationaland Space Biology vol 25 pp 6ndash13 2011

[9] M L Salmi and S J Roux ldquoGene expression changes inducedby space flight in single-cells of the fern Ceratopteris richardiirdquoPlanta vol 229 no 1 pp 151ndash159 2008

[10] O Rasmussen D A Klimchuk E L Kordyum et al ldquoTheeffect of exposure to microgravity on the development andstructural organisation of plant protoplasts flown onBiokosmos9rdquo Physiologia Plantarum vol 84 no 1 pp 162ndash170 1992

[11] E Hoffmann K Schonherr and R Hampp ldquoRegenerationof plant cell protoplasts under microgravity investigation ofprotein patterns by SDS-PAGE and immunoblottingrdquo Plant CellReports vol 15 no 12 pp 914ndash919 1996

[12] R Hampp E Hoffmann K Schonherr P Johann and L DeFilippis ldquoFusion and metabolism of plant cells as affected bymicrogravityrdquo Planta vol 203 pp S42ndashS53 1997


[14] M Martzivanou and R Hampp ldquoHyper-gravity effects on theArabidopsis transcriptomerdquo Physiologia Plantarum vol 118 no2 pp 221ndash231 2003

[15] A L Paul A K Zupanska D T Ostrow et al ldquoSpaceflighttranscriptomes Unique responses to a novel environmentrdquoAstrobiology vol 12 no 1 pp 40ndash56 2012

[16] A K Zupanska F C Denison R J Ferl and A-L PaulldquoSpaceflight engages heat shock protein and other molecularchaperone genes in tissue culture cells of Arabidopsis ThalianardquoThe American Journal of Botany vol 100 no 1 pp 235ndash2482013

[17] R Edgar M Domrachev and A E Lash ldquoGene ExpressionOmnibus NCBI gene expression and hybridization array datarepositoryrdquo Nucleic Acids Research vol 30 no 1 pp 207ndash2102002

[18] A Brazma H Parkinson U Sarkans et al ldquoArrayExpressmdashapublic repository for microarray gene expression data at theEBIrdquo Nucleic Acids Research vol 31 no 1 pp 68ndash71 2003

[19] J Kilian D Whitehead J Horak et al ldquoThe AtGenExpressglobal stress expression data set protocols evaluation andmodel data analysis of UV-B light drought and cold stressresponsesrdquo Plant Journal vol 50 no 2 pp 347ndash363 2007

[20] D Swarbreck C Wilks P Lamesch et al ldquoThe ArabidopsisInformation Resource (TAIR) Gene structure and function


annotationrdquo Nucleic Acids Research vol 36 no 1 pp D1009ndashD1014 2008

[21] N Hausmann S Fengler A Hennig M Franz-Wachtel RHampp and M Neef ldquoCytosolic calcium hydrogen peroxideand related gene expression and protein modulation in Ara-bidopsis thaliana cell cultures respond immediately to alteredgravitation parabolic flight datardquo Plant Biology vol 16 no 1pp 120ndash128 2014

[22] J Z Kiss W J Katembe and R E Edelmann ldquoGravitropismand development of wild-type and starch-deficient mutants ofArabidopsis during spaceflightrdquo Physiologia Plantarum vol 102no 4 pp 493ndash502 1998

[23] A L Paul C J Daugherty E A Bihn D K Chapman KL L Norwood and R J Ferl ldquoTransgene expression patternsindicate that spaceflight affects stress signal perception andtransduction in Arabidopsisrdquo Plant Physiology vol 126 no 2pp 613ndash621 2001

[24] A-L Paul M P Popp W B Gurley C Guy K L Norwoodand R J Ferl ldquoArabidopsis gene expression patterns are alteredduring spaceflightrdquo Advances in Space Research vol 36 no 7pp 1175ndash1181 2005

[25] K D L Millar P Kumar M J Correll et al ldquoA novelphototropic response to red light is revealed in microgravityrdquoNew Phytologist vol 186 no 3 pp 648ndash656 2010

[26] J Allen P A Bisbee R L Darnell et al ldquoGravity control ofgrowth form in Brassica Rapa and Arabidopsis Thaliana (Bras-sicaceae) consequences for secondary metabolismrdquo AmericanJournal of Botany vol 96 no 3 pp 652ndash660 2009

[27] M EMusgrave A Kuang Y Xiao et al ldquoGravity independenceof seed-to-seed cycling in Brassica rapardquo Planta vol 210 no 3pp 400ndash406 2000

[28] G Perbal and D Driss-Ecole ldquoSensitivity to gravistimulus oflentil seedling roots grown in space during the IML 1 Missionof Spacelabrdquo Physiologia Plantarum vol 90 no 2 pp 313ndash3181994

[29] B C Tripathy C S Brown H G Levine and A D KrikorianldquoGrowth and photosynthetic responses of wheat plants grownin spacerdquo Plant Physiology vol 110 no 3 pp 801ndash806 1996

[30] G W Stutte O Monje R D Hatfield A L Paul R J Ferl andC G Simone ldquoMicrogravity effects on leaf morphology cellstructure carbon metabolism and mRNA expression of dwarfwheatrdquo Planta vol 224 no 5 pp 1038ndash1049 2006

[31] J Ueda KMiyamoto T Yuda et al ldquoGrowth and developmentand auxin polar transport in higher plants under microgravityconditions in space BRIC-AUX on STS-95 space experimentrdquoJournal of Plant Research vol 112 no 1108 pp 487ndash492 1999

[32] Y Zhang L Wang J Xie and H Zheng ldquoDifferential proteinexpression profiling ofArabidopsis thaliana callus undermicro-gravity on board the Chinese SZ-8 spacecraftrdquo Planta In press

[33] M Vukich A Donati and V Zolesi ldquoKayser Italia hardware forradiation and microgravity experiments in spacerdquo RendicontiLincei vol 25 no 1 pp 7ndash11 2014

[34] E Brinckmann ldquoCentrifuges and their application for biologi-cal experiments in spacerdquoMicrogravity Science and Technologyvol 24 no 6 pp 365ndash372 2012

[35] Astrium GmbH Astrium Space Transportation Astrium SpaceBiology Product Catalog Friedrichshafen Germany 2012

[36] H Kleinig ldquoPflanzlicheGewebekultur Ein PraktikumrdquoBiologiein unserer Zeit vol 16 no 4 p 128 1986

[37] S Fengler M Neef M Ecke and R Hampp ldquoThe Simboxexperiment with Arabidopsis thaliana cell cultures hardware

tests and first results from theGerman-Chinese satellitemissionShenzhou 8rdquo in Proceedings of the Life in Space for Life on EarthSymposium vol ESA SP-706 2013

[38] G Rustici N Kolesnikov M Brandizi et al ldquoArrayExpressupdate-trends in database growth and links to data analysistoolsrdquo Nucleic Acids Research vol 41 no 1 pp D987ndashD9902013

[39] F Battke S Symons and K Nieselt ldquoMaydaymdashintegrativeanalytics for expression datardquo BMC Bioinformatics vol 11article 121 2010

[40] R A Irizarry BM Bolstad F Collin LMCope BHobbs andT P Speed ldquoSummaries of Affymetrix GeneChip probe leveldatardquo Nucleic Acids Research vol 31 no 4 2003



[43] M Simonsen T Mailund and C N Pedersen ldquoRapidneighbour-joiningrdquo inAlgorithms in Bioinformatics vol 5251 ofLecture Notes in Computer Science pp 113ndash122 Springer BerlinGermany 2008

[44] D M Mutch A Berger R Mansourian A Rytz and M-ARoberts ldquoThe limit fold change model a practical approach forselecting differentially expressed genes from microarray datardquoBMC Bioinformatics vol 3 article 17 2002

[45] A Subramanian P Tamayo V K Mootha et al ldquoGene setenrichment analysis a knowledge-based approach for inter-preting genome-wide expression profilesrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 102 no 43 pp 15545ndash15550 2005

[46] M Ashburner C A Ball J A Blake et al ldquoGene ontology toolfor the unification of biologyrdquoNature Genetics vol 25 no 1 pp25ndash29 2000

[47] M Babbick Z Barjaktarovic and R Hampp ldquoAlterations in theexpression of transcription factors in Arabidopsis thaliana cellcultures during sounding rocket 120583Grdquo in Proceedings of the 18thESA Symposium on European Rocket and Balloon Programmespp 473ndash477 June 2007


[49] HWang Q Z HuiW Sha R Zeng and C X Qi ldquoA proteomicapproach to analysing responses of Arabidopsis thaliana calluscells to clinostat rotationrdquo Journal of Experimental Botany vol57 no 4 pp 827ndash835 2006

[50] T Hoson K Soga R Mori et al ldquoStimulation of elongationgrowth and cell wall loosening in rice coleoptiles under micro-gravity conditions in spacerdquo Plant and Cell Physiology vol 43no 9 pp 1067ndash1071 2002

[51] A Nasir S M Strauch I Becker et al ldquoThe influence ofmicrogravity on Euglena gracilis as studied on Shenzhou 8rdquoPlant Biology vol 16 no supplement 1 pp 113ndash119 2014

[52] K Apel and H Hirt ldquoReactive oxygen species metabolismoxidative stress and signal transductionrdquo Annual Review ofPlant Biology vol 55 pp 373ndash399 2004


[53] S Davletova K Schlauch J Coutu and R Mittler ldquoThe zinc-finger protein Zat12 plays a central role in reactive oxygen andabiotic stress signaling in Arabidopsisrdquo Plant Physiology vol139 no 2 pp 847ndash856 2005

[54] R Mittler ldquoAbiotic stress the field environment and stresscombinationrdquo Trends in Plant Science vol 11 no 1 pp 15ndash192006

[55] P G Sappl A J Carroll R Clifton et al ldquoThe Arabidopsisglutathione transferase gene family displays complex stressregulation and co-silencing multiple genes results in alteredmetabolic sensitivity to oxidative stressrdquoThe Plant Journal vol58 no 1 pp 53ndash68 2009

[56] V P Roxas R K Smith Jr E R Allen and R D Allen ldquoOver-expression of glutathione S-transferaseglutathione peroxidaseenhances the growth of transgenic tobacco seedlings duringstressrdquo Nature Biotechnology vol 15 no 10 pp 988ndash991 1997

[57] V P Roxas S A Lodhi D K Garrett J R Mahan and RD Allen ldquoStress tolerance in transgenic tobacco seedlings thatoverexpress glutathione S-transferaseglutathione peroxidaserdquoPlant and Cell Physiology vol 41 no 11 pp 1229ndash1234 2000

[58] G Horneck ldquoRadiobiological experiments in space a reviewrdquoNuclear Tracks and Radiation Measurements vol 20 no 1 pp185ndash205 1992

[59] G Horneck ldquoAstrobiology studies of microbes in simulatedinterplanetary spacerdquo in Laboratory Astrophysics and SpaceResearch pp 667ndash685 1999

[60] C Baumstark-Khan C E Hellweg A Arenz and M MMeier ldquoCellular monitoring of the nuclear factor 120581B pathwayfor assessment of space environmental radiationrdquo RadiationResearch vol 164 no 4 pp 527ndash530 2005

[61] C E Hellweg and C Baumstark-Khan ldquoGetting ready forthe manned mission to Mars the astronautsrsquo risk from spaceradiationrdquoNaturwissenschaften vol 94 no 7 pp 517ndash526 2007

[62] A R Kennedy ldquoBiological effects of space radiation anddevelopment of effective countermeasuresrdquo Life Sciences inSpace Research vol 1 no 1 pp 10ndash43 2014

[63] X S Wan Z Zhou and A R Kennedy ldquoAdaptation of thedichlorofluorescein assay for detection of radiation-inducedoxidative stress in cultured cellsrdquo Radiation Research vol 160no 6 pp 622ndash630 2003

[64] X S Wan Z Zhou J H Ware and A R Kennedy ldquoStandard-ization of a fluorometric assay for measuring oxidative stress inirradiated cellsrdquo Radiation Research vol 163 no 2 pp 232ndash2402005

[65] M L Orozco-Cardenas J Narvaez-Vasquez and C A RyanldquoHydrogen peroxide acts as a second messenger for the induc-tion of defense genes in tomato plants in response to woundingsystemin and methyl jasmonaterdquo The Plant Cell vol 13 no 1pp 179ndash191 2001

Research ArticleRCCS Bioreactor-Based Modelled Microgravity InducesSignificant Changes on In Vitro 3D Neuroglial Cell Cultures

Caterina Morabito12 Nathalie Steimberg23 Giovanna Mazzoleni23 Simone Guarnieri12

Giorgio Fanograve-Illic12 and Maria A Mariggiograve124

1 Department of Neuroscience Imaging and Clinical Sciences Unit of Functional Biotechnology Aging Research Center (CeSI)ldquoG drsquoAnnunziordquo University of Chieti-Pescara Via dei Vestini 29 66100 Chieti Italy

2 Interuniversity Institute of Myology Italy3 Laboratory of Tissue Engineering Department of Clinical and Experimental Sciences School of Medicine University of BresciaViale Europa 11 25123 Brescia Italy

4 Section of Physiology and Physiopathology Department of Neuroscience Imaging and Clinical SciencesldquoG drsquoAnnunziordquo University of Chieti-Pescara Via dei Vestini 31 66013 Chieti Italy

Correspondence should be addressed to Maria A Mariggio mariggiounichit

Received 24 April 2014 Revised 10 September 2014 Accepted 10 September 2014


Copyright copy 2015 Caterina Morabito et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

We propose a human-derived neuro-glial cell three-dimensional in vitro model to investigate the effects of microgravity on cell-cell interactions A rotary cell-culture system (RCCS) bioreactor was used to generate a modelled microgravity environment andmorphofunctional features of glial-like GL15 and neuronal-like SH-SY5Y cells in three-dimensional individual cultures (monotypicaggregates) and cocultures (heterotypic aggregates) were analysed Cell survival was maintained within all cell aggregates over 2weeks of cultureMoreover compared to cells as traditional staticmonolayers cell aggregates cultured undermodelledmicrogravityshowed increased expression of specific differentiation markers (eg GL15 cells GFAP S100B SH-SY5Y cells GAP43) andmodulation of functional cell-cell interactions (eg N-CAM and Cx43 expression and localisation) In conclusion this culturemodel opens a wide range of specific investigations at the molecular biochemical and morphological levels and it represents animportant tool for in vitro studies into dynamic interactions and responses of nervous system cell components to microgravityenvironmental conditions

1 Introduction

Microgravity modulates numerous features and functionsof biological organisms through its effects on physical phe-nomena such as hydrostatic pressure in fluid-filled com-partments sedimentation of organelles and convection pro-cesses of flow and heat These physical parameters can inturn directly and indirectly influence cellular and tissuemorphology metabolism and signalling and consequentlya wide range of cell functions [1] Several years ago it wasproposed that gravity is involved in embryonic developmentthrough effects on morphogenesis and organogenesis of thecentral nervous system and on sensory organs in inverte-brates and vertebrates In particular when amphibian eggs

were fertilised in vivo or in vitro under microgravity con-ditions some abnormalities during embryonic developmentwere observed even if compensatory mechanisms producednearly normal larvae [2] Also during space flight signsof neurophysiological impairment have been observed forastronauts although few studies have been carried out toinvestigate such effects on the nervous system in particularat the cellular level [3]

Recently Pani and colleagues reported that neuronalmonolayers showed alterations in morphology and viabilitywhen exposed to short- and middle-term simulated micro-gravity in the random positioning machine while long-termexposures revealed high adaptation of single neurons to thenew gravity conditions [4] Also other neuronal cell models



showed morphological andor cytoskeletal alterations whenexposed to simulated weightlessness or during changinggravity [5 6] These effects appeared conditioned by thepresence of microgravity conditions and after short-termexposures under ground-conditions the cells were able tofully recover their features and the ability to form adherentmonolayer cultures [4 7]

Traditional monolayer cell cultures that are kept understatic conditions (two-dimensional (2D) cell culture) haveprovided great advances in our understanding of the phys-iological regulatory processes of single cells On the otherhand the intrinsic complexity of cell-cell extracellular sig-nalling and the remarkable plasticity in the compositionand structure of the extracellular matrix have made it verydifficult to study these interactions using conventional cell-culture techniques For these reasons advanced methods areneeded to grow cells while maintaining their native three-dimensional (3D) cytoarchitecture and the specific tissue-like microenvironment Interestingly 3D cultures have beenshown to favour the maintenance of tissue-specific pheno-types and tissue-like cytoarchitectureHowever an importantlimitation for long-term culture in three dimensions is thelow diffusion of oxygen and nutrients and the absence of ablood supply to the deeper parts of the tissue construct Thisis particularly the case for neural cells and it can result in theappearance of a central core of dead cells [8 9]

In the 1990s after the beginning of themany internationalspace programmes attempts were made to grow 3D cellcultures or tissue explants in particular microenvironmentsto test the effects of reduced gravity Major efforts have beenaddressed to the building of a system that can reproducea tissue-like microenvironment in vitro and to study thecytoskeletal and nuclear matrix protein interactions duringcell exposure to simulated microgravity as is present in space[10] Engineers at the US National Aeronautics and SpaceAdministration (NASA) devised a rotating bioreactor whichis a useful device for culturing cells on Earth as well as inspace Briefly thismonoaxial clinostat (the rotary cell-culturesystem (RCCS) bioreactor) is a horizontally rotating andfluid-filled culture vessel that is equippedwith a gas-exchangemembrane that optimises the oxygen supply to the biologicalsamples Without air bubbles or air-liquid interface the fluiddynamic conditions inside the culture chamber generate alaminar flow state that greatly reduces shear stress and turbu-lence which are hazardous for cell survival These dynamicconditions provided by the RCCS bioreactor favour spatialcolocalisation and three-dimensional assembly of single cellsinto aggregates [11] The rotational speed of the culturechamber can be modified to set conditions in which the 3Dcell constructsaggregates also rotate around their own axesfurther providing an efficient high mass transfer of nutrientand wastes When cultured cell aggregates grow in sizethe rotational speed of the culture vessel can be increasedto compensate for the increased sedimentation rates Theoperational conditions of the RCCS bioreactor can also beadjusted so that the gravitational vectors are randomised upto reach a modelled microgravity state [12 13] In this way3D biological samples can remain in a constant orientationwith respect to the chamber wall and move in near-solid

body rotation with the fluid thus fulfilling the requirementsneeded to successfully model microgravity conditions [14]

In the present study we aimed to develop a 3D dynamicin vitro neuroglial coculture system to evaluate the capacityof the cells to reproduce at least in part neuronal featuresTo this end we used two well-characterised cell lines GL15and SH-SY5Y cells which are astrocyte-like and neuronal-like cells respectively The human glioblastoma GL15 cellline is an established in vitro astrocyte model that hasbeen functionally characterised by our group and others[15 16] and these express a typical astroglial phenotypeand functions The human neuroblastoma-derived SH-SY5Ycells are a widely used and well-characterised neuronalcell model that has been extensively used for in vitroneurotoxicity testing and has been shown to differentiatetowards either adrenergic or cholinergic phenotypes [17ndash20] In addition the human origin of these cell lines makesthem an appealing model for basic in vitro research studiesThus to develop astrocyte-like or neuronal-like in vitromodels 3D monotypic cultures (GL15 cells only or SH-SY5Ycells only) were established in a RCCS bioreactor Of noteit has been demonstrated that cell-cell interactions as forexample those between glial cells and neurons are crucialfor both glial and neuronal differentiation and developmentalprocesses as well as for response to neural injury [21 22]For these reasons we also established 3D neuronalglialheterotypic cultures (cocultures) to more closely reproducethe in vivo microenvironment of the nervous tissue andto bridge the gap between in vitro systems and animalmodels These analyses were also performed under modelledmicrogravity when the 3D cell aggregates were sufficientlygrown in size to adjust to the operational conditions of theRCCS bioreactor so as to reach a state of vector-averagedmicrogravity Under such conditions their cell morphologyviability and functional features were analysed and com-pared


All of the reagents for cell culturewere fromLife Technologies(Milan Italy) The plasticware was from BD Falcon (SaccoMilan Italy)

21 Cell Culture The SH-SY5Y cell line (from the EuropeanCollection of Cell Cultures supplied through Sigma-AldrichUK) and the GL15 cell line were both cultured in Dulbeccorsquosmodified Eaglersquos medium (DMEM) with 10 foetal bovineserum 100 IUmL penicillin 100 120583gmL streptomycin and1mM glutamine The cells were amplified in monolayersand detached for subculturing using 005 trypsin and002 EDTA SH-SY5Y and GL15 cell cultures used forexperimental assays were prepared by seeding cells in T75Falcon flasks to form 2D static monolayer cultures or in theRCCS bioreactor to establish 3D cultures subjected to micro-gravity Both culture models were cultured in DMEM with10 foetal bovine serum 100 IUmL penicillin 100 120583gmLstreptomycin and 1mM glutamine and maintained in thesame incubator (5 CO

2 at 95 humidity) for the same

times and the medium was refreshed twice a week


22 3D Culture in the RCCS Bioreactor The RCCS biore-actor (Synthecon Houston USA) can generate a specialmicroenvironment where highmass transfer is achieved withlow shear stress It is equipped with a cylindrical growthchamber that contains an inner corotating cylinder with agas-exchange membrane (a 55mL autoclavable slow-turninglateral vessel) where specific hydrodynamic and physicalconditions are attained The culture of cell spheroids wasperformed in this device in a 5 CO

2incubator at 95

humidity The horizontally rotating culture vessel was filledwith the complete medium (without air-liquid interface toreduce the shear stress) After a defined rotational speed wasreached the cells were cultured under Earth gravity in a nearlaminar fluid flow environment (ie a free-fall state) Undersuch conditions the cells grew in the form of 3Dmulticellularaggregates [23 24]

The cell-density seeding for both GL15 and SH-SY5Ycells was approximately 15 times 106 cellsmL The medium wasrefreshed twice a week For cocultures the SH-SY5Y and theGL15 cells were each seeded at a density of 075times 106 cellsmLThe rotational speed of the culture chamberwas initially set atbetween 6 rpmand 8 rpm and then it was gradually increasedas the multicellular aggregates increased in size to maintainthe aggregates in constant equilibrium (ie under free-fallconditions)

At the indicated times the cells were harvested andaccording to the experimental conditions required themulticellular aggregates were either included in Tissue-TekOCT compound (VWR International Srl USA) (for in situanalysis) or centrifuged at 2300 rpm for 5min at 4∘C and theresulting cell pellets were kept atminus80∘CuntilWestern blottingwas carried out

For the embedded aggregates slices (6120583m to 10 120583m)wereprepared with a CM1900 cryostat (Leica Milan Italy) andprocessed for cell viability assays or frozen at minus20∘C forfurther investigations

23 Morphological Analysis The frozen sections were left towarm up to room temperature and were subsequently incu-bated for 12min in Harrisrsquo haematoxylin solution washedtwice in water and incubated for 15 s in Eosin solution Afterwashing the sections were dehydrated and mounted withEukitt mounting medium (Electron Microscopy Sciences)The sections were examined under a Vanox optical micro-scope (Olympus Opera Zerbo Italy)

24 Cell Viability Assay The sections were incubated for15min at room temperature in a solution containing recombi-nant Annexin V conjugated to the Alexa 488 fluorophore andpropidium iodide (Vybrant kit 2 Life Technologies Italy)as described by the manufacturer Moreover to quantify thetotal number of cells in the aggregates 41015840-6-diamidino-2-phenylindole (DAPI) was added to this solution at a finalconcentration of 010 120583gmL The sections were mounted inProlong antifade medium (Life Technologies) and examinedunder an inverted fluorescence microscope (Axiovert ZeissArese Italy) equipped with an image analyser Photomicro-graphs were analysed with the ProImage+ and Scion Imagesoftware (httpproimagesoftwareinformercom andhttp

scion-imagesoftwareinformercom) to determine the cellviability

25 Immunostaining Assay The frozen GL15 and SH-SY5Ycells in OCT sections were fixed with 37 paraformaldehydeat room temperature for 30min Slices were then perme-abilised with 01 Triton X-100 at room temperature for15min and incubated for 1 h in 10 bovine serum albumin atroom temperature and then for 1 h at 37∘C with the primaryantibody followed by 1 h at 37∘C with either an Alexa 488-or an Alexa 633-conjugated secondary antibody (MolecularProbesMilan Italy) For double staining the second primaryantibody was incubated with the constructs after removal ofthe first Alexa 488-conjugated secondary antibody After theantibody incubation the cells were washed three times with01 Tween 20 each for 5min at room temperature Finallythe nuclei were stained with 1 120583gmL propidium iodide for30min After three washes with phosphate-buffered salinethey were mounted on coverslips and examined

Primary monoclonal mouse antibodies neuronal celladhesion molecule (N-CAM) tyrosine hydroxylase growthassociated protein 43 (GAP43) glial fibrillary acidic protein(GFAP) and S100B were from Sigma-Aldrich (Milan Italy)and connexin 43 (Cx43) was from Chemicon InternationalInc (Temecula CA USA)

The fluorescence images were obtained using a ZeissLSM510 META confocal system (Jena Germany) connectedto an inverted Zeiss Axiovert 200 microscope equipped witha Plan Neofluar oil-immersion objective (40x13 NA)

26 Western Blotting Frozen pellets of the GL15 andSH-SY5Y cell aggregates were lysed in cell lysis buffer(50mM Tris-HCl 100mM NaCl 50mM NaF 40mM 120573-glycerophosphate 5mM EDTA 1 Triton X-100 200120583Msodium orthovanadate 100 120583gmL phenylmethylsulfonyl flu-oride 10 120583gmL leupeptin 5 120583gmL pepstatin A 10 120583gmLbenzamidine and pH 74) After vortexing for 5 min thesamples were centrifuged at 1000 rpm for 10min at 4∘C ina microcentrifuge The protein content of each supernatantwas quantified colorimetrically (Bio-Rad Laboratories SrlMilan Italy) and aliquots containing 40 120583g protein wereadded to Laemmli buffer (8 SDS 10 glycerol 5 120573-mercaptoethanol 25mM Tris-HCl 0003 bromophenolblue and pH 65) and applied to and separated by SDS-PAGEon 7 to 10 SDS polyacrylamide slab gels Proteins wereelectroblotted onto hydrophobic polyvinylidene difluoridemembranes (Immobilon Millipore Milan Italy) using atank transfer system (Bio-Rad Laboratories Srl) Transferefficiency was verified by Ponceau red staining of the blotsand Coomassie blue staining of the gels The SH-SY5Y cellblots were incubated with the following mouse monoclonalantibodies anti-N-CAM (1 100 dilution Sigma-Aldrich)anti-tyrosine hydroxylase (1 1000 dilution Sigma-Aldrich)anti-GAP43 (1 1000 dilution Sigma-Aldrich) andor anti-Cx43 (1 1000 dilution Chemicon) The GL15 cell blots wereincubated with the following mouse monoclonal antibodiesanti-glial fibrillary acidic protein (GFAP 1 500 dilutionSigma-Aldrich) anti-S100B (1 500 dilution Sigma-Aldrich)andor also Cx43 (1 1000 dilution Chemicon) These were


then detected by chemiluminescence (ECL plus GE Health-care) Moreover after a membrane-stripping procedure theGL15 and SH-SY5Y cell membranes were immunostainedwith a mouse monoclonal anti-actin antibody (1 1000 dilu-tion Sigma-Aldrich)

3 Results

31 Cell Aggregates in the RCCS BioreactorMorphology and Viability

311 GL15 Cells Initial experiments were performed toestablish the most suitable protocol to prepare the cell aggre-gates GL15 cells were incubated in the RCCS bioreactor aspreinduced cell clusters or as homogeneous cell suspensionsthat were left to spontaneously aggregate The preinducedaggregates were obtained using the hanging drop method(see [24]) Both types of aggregates were maintained underconditions of microgravity in the RCCS bioreactor for upto 2 weeks The single cells spontaneously aggregated within48 h of culture although some features of the spontaneouscell aggregates were different compared to the preinducedaggregates

The preinduced aggregates provided relatively uniformclusters while the spontaneously formed aggregates appearedmore irregular in shape In addition after 2 weeks inthe RCCS bioreactor the spontaneously formed aggregatesshowed a trend (not significant) towards a greater meanarea (298 plusmn 026mm2) compared to that of the preinducedaggregates (189 plusmn 128mm2) (Figure 1)

The cell viability in the preinduced aggregates and thespontaneously formed aggregates was also assessed after 2weeks in the RCCS bioreactor to determine the apoptotic orthe necrotic cells (Figure 2) To this aim the cells were testedto measure early apoptosis by detecting phosphatidylserineexpression revealed by Annexin V binding or necrosis bymembrane permeability to the propidium iodide (PI) vitaldye Cells positive to Annexin V green fluorescence signalare known to be apoptotic cells while those positive to PIred fluorescence signal are necrotic cells the absence of greenor red signal and the nuclear staining with DAPI revealedviable cells The image analyses of stained cells revealedthat some apoptotic cells (Figure 2(a) green fluorescence)were evident at a similar extent in preinduced and spon-taneous GL15 aggregates (Figure 2(b)) A relevant amountof necrotic cells (Figure 2(a) red fluorescence) was presentin preinduced aggregates compared to spontaneous ones inwhich necrotic cells were nearly absent (Figure 2(b)) Thesedata revealed the presence of possible stress conditions inthe preinduced aggregates while the spontaneously formedaggregates showed cells that were in a more healthy stateThis cell stress might be the result of hypoxic processes inthe central core of the preinduced aggregates in particularpotentially due to the static conditions encountered in thehanging drops For this reason the rest of the investigationsused only the spontaneously formed cell aggregates andthose formed by the GL15 cells are henceforth referred to asthe G-aggregates

312 SH-SY5Y Cells Following the same proceduredescribed above for the formation of the G-aggregatesthe SH-SY5Y cells were cultured to form spontaneousaggregates in the RCCS bioreactor for up to 2 weeks andtheir cell morphology and viability were then assessedTheseaggregates formed by the SH-SY5Y cells are henceforthreferred to as the S-aggregates

At the end of the incubation period the S-aggregatesshowed variable and irregular shapes with a mean area of268 plusmn 013mm2 In addition there were very low levels ofapoptotic and necrotic cells (Figure 2(b)) which indicatedthat these 3D dynamic culture conditions are a suitablemethod to sustain cell viability also for neuronal-like cells

32 Qualitative Analysis of Phenotype-Specific Markers

321 GL15 Cells To analyse the expression of the GL15cell specific phenotype in the G-aggregates cultured in theRCCS bioreactor for up to 2 weeks immunostaining for glialmarkers was carried out The G-aggregates showed glial-cell-specific protein expression similar to that observed inthe GL15 cells cultured as monolayers under 2D conditions(Figure 3) The G-aggregates and the GL15 cells cultured asmonolayers both showed cytoplasmic localisation of GFAPand S100B as two markers of the glial cytoskeleton (Figures3(a)ndash3(d))

Cell interactions due to gap-junction-mediated intercel-lular communication have been shown to have crucial rolesin the regulation of the glial-cell network and nervous systemfunctions [25] For this reason the expression of Cx43 wasalso investigated as Cx43 is the main gap-junction proteinexpressed by astrocytes As shown in Figure 3(e) the GL15cells grown in two dimensions expressed Cx43 near theplasmalemma at cell-cell contact areas and in the cytoplasmA similar distribution was also seen for Cx43 in the G-aggregates (Figure 3(f))

322 SH-SY5Y Cells We characterised the phenotypeexpressed by the SH-SY5Y cells in the S-aggregatesmaintained in the dynamic 3D culture in the RCCSbioreactor for 2 weeks by determining the expression of theneuronal specific markers N-CAM GAP43 and tyrosinehydroxylase Immunofluorescence analysis revealed thatN-CAM in the S-aggregates was localised towards theplasma membrane and near cell-cell contact areas thusresembling its distribution in the SH-SY5Y cells culturedin 2D monolayers (Figures 4(a) and 4(b)) which showedcell-cell adhesion interactions GAP43 is involved in neuriteoutgrowth and neuronal plasticity [26] and in SH-SY5Ycell monolayers it was localised into neurite-like processes(Figure 4(c)) In the S-aggregates GAP43 was localised in thecytoplasmic compartment (Figure 4(d)) The distributionof tyrosine hydroxylase (TH) which is a rate-limitingenzyme in dopaminenorepinephrine synthesis [27] was inthe cytoplasm under both of these cell-culture conditions(Figures 4(e) and 4(f))


Preinduced Spontaneous

Spontaneous

0

1

2

3

4

Agg

rega

te ar

ea (m

m2 )

GL15-composed aggregates

Preinduced

Figure 1 GL15 cell aggregate morphology Representative images and quantification of sections from preinduced and spontaneously formedGL15 aggregates (as indicated) Data are means plusmn SEM 119899 = 15 for the averaged areas of the aggregate sections calculated using the ImageJsoftware (httpimagejnihgovij)

33 Quantitative Analysis of Phenotype-Specific Markers

331 GL15 Cells The differentiation status of cells is charac-terised not only by marker localisation but also by markerexpression levels To evaluate potential quantitative dif-ferences between the G-aggregate modelled microgravity-exposed cultures and the GL15 cells as 2D static monolayercultures the expression levels of the GFAP S100B and Cx43proteins were determined byWestern blotting (Figure 5)TheG-aggregates showed increased levels of GFAP S100B andCx43 after the first 48 h of culture These levels graduallydecreased over the following 2 weeks when those of S100Band Cx43 were similar to those observed in the GL15 cells as2D static cultures while those of GFAP remained increasedin the G-aggregates (Figure 5)

332 SH-SY5Y Cells Western blotting carried out for the S-aggregates showed that N-CAM-140 and GAP43 expressionlevels were increased during the incubation compared to theSH-SY5Y cells as 2D static monolayer cultures (Figure 6) Inparticular for the S-aggregates N-CAM-140 reached a peakafter 2 weeks while GAP43 peaked after 48 h ConverselyN-CAM-180 and tyrosine hydroxylase did not significantly

change in the S-aggregates compared to SH-SY5Y cells as 2Dstatic cultures (Figure 6)

34 Coculture of GL15 and SH-SY5Y Cells in the RCCSBioreactor The SH-SY5Y cells were also cocultured with theGL15 cells in the RCCS bioreactor with the aim to reestablisha more neural-like microenvironment and thus to be closerto in vivo conditions Initial experiments were carried outto determine if it was possible to establish viable GL15 plusSH-SY5Y cocultures in the RCCS bioreactor henceforthreferred to as GS-aggregates GL15 and SH-SY5Y cells werethus cocultured in the RCCS bioreactor at a 1 1 ratio forup to 2 weeks At the end of this period the sizes of theGS-aggregates were similar to those of the monotypic G-aggregates and S-aggregates (Figure 7) and although the S-aggregates appeared smaller than the others these differencesdid not reach significance Cell viability assays also showedthat the GS-aggregates had low levels of apoptotic andnecrotic cells (data not shown)

To characterise the cell phenotype in theseGS-aggregatesimmunostaining was carried out for N-CAM GFAP andCx43 These coculture conditions induced the establishment


SH-SY5Y

spontaneously

formed

aggregates in

GL15

spontaneously

formed

aggregates in

GL15

preinduced

aggregates in

hanging drop

DAPI PI

RCCS

RCCS

Annexin V

(a)

Total cell number Apoptotic cells ()

Necrotic cells ()

GL15 preinduced aggregates in hanging drop

512 16 19

914 14 007

491 7 1

SH-SY5Y spontaneously

formedaggregates in

GL15 spontaneously

formedaggregates in

RCCS

RCCS

(b)

Figure 2 Cell viability assay (a) Representative images of preinduced and spontaneously formed GL15 aggregates and spontaneously formedSH-SY5Y aggregates (as indicated) Aggregates were stained with DAPI (blue) Annexin V-Alexa 488 (green) and propidium iodide (PIred) DAPI-positiveAnnexin V-Alexa 488-negativePI-negative cells are healthy DAPI-positiveAnnexin V-Alexa 488-positivePI-negativeand PI-positive cells are considered apoptotic (Annexin V arrowheads) DAPI-positiveAnnexin V-Alexa 488-negativePI-positive cellsare necrotic (PI arrowheads) (b) Quantification of apoptotic and necrotic cells in aggregate sections Data derived from 3 independentexperiments


GFAP

2D

(a)

GFAPPI

RCCS

(b)

S100B

(c)

S100BPI

(d)

Cx43

(e)

Cx43PI

(f)

Figure 3 Glial marker localisation in GL15 cells Representative confocal images of GL15 cells cultured as a monolayer (2D (a) (c) and (e))and under the modelled microgravity (RCCS bioreactor (b) (d) and (f)) and immunostained with anti-GFAP ((a) and (b)) anti-S100B ((c)and (d)) and anti-Cx43 ((e) and (f)) antibodies (as indicated)The RCCS G-aggregate sections were also stained with propidium iodide (PI)Insets in (b) (d) and (f) show image magnification Scale bars 25120583m

of GS-aggregates that contained both glial-like and neuronal-like cell phenotypes After the 2 weeks of culture in the RCCSbioreactor these GS-aggregates showed specific fluorescencesignals for astrocyte (GFAP-positive) andneuronal (N-CAM-positive) phenotypes (Figure 8)

The GS-aggregates were double-stained for N-CAM andCx43 These N-CAM-specific and Cx43-specific fluorescentsignals revealed a particular distribution of these proteinswhereby even if colocalisation of the N-CAM and Cx43patterns was not evident possible heterotypic cell-cell inter-actions could not be excluded In particular within the GS-aggregates N-CAM localised to the peripheral areas of thecells while Cx43-specific fluorescent spots appeared to be

sparsely distributed which indicated a low level of cell-cellfunctional interactions (Figure 9) In addition in the sameGS-aggregates there were also evident N-CAM-negativeandor Cx43-negative cells which indicated potential differ-ent cell activities due to different protein expression levels

Western blotting of N-CAM and Cx43 expression levelsrevealed that in the GS-aggregates the monomeric Cx43protein (43 kDa) was downregulated during the RCCS biore-actor incubation Interestingly homogenates from the GS-aggregates showed a Cx43-positive band at 86 kDa whichdemonstrates the presence of a dimeric form of Cx43 whichwas highly expressed in the initial phases of the coculture(over the first 24 h) and which significantly decreased over


2D

N-CAMPI

(a)

N-CAMPI

RCCS

(b)

GAP43PI

(c)

GAP43PI

(d)

THPI

(e)

THPI

(f)

Figure 4 Neuronal marker localisation in SH-SY5Y cells Representative confocal images of SH-SY5Y cells cultured as a monolayer (2D (a)(c) and (e)) and under the modelled microgravity (RCCS bioreactor (b) (d) and (f)) and immunostained with anti-N-CAM ((a) and (b))anti-GAP43 ((c) and (d)) and anti-tyrosine hydroxylase (TH) ((e) and (f)) antibodies (as indicated) All of the cells were also stained withpropidium iodide (PI) Insets show image magnification Scale bars 20120583m


0

1

2

0

1

2

3

4

0

1

2

3

GFAP 50 kDa

S100B 20kDa

Cx43 43kDa

2D 24h 48h 96h 2w

2D 24h 48h 96h 2w2D 24h 48h 96h 2w 2D 24h 48h 96h 2w

GFAP S100B Cx43

Ratio

Ratio

Ratio

RCCS

lowast

lowast


lowast

Figure 5 Expression of glial cell markers RepresentativeWestern blotting and quantification of the levels of GFAP S100B and Cx43 in GL15cells cultured in 2Dmonolayers and in the RCCS bioreactor for 24 h 48 h 96 h and 2 weeks (2w) Data are from densitometric ratio analysesas means plusmn SEM from 3 independent experiments lowast119875 lt 005 versus 2D monolayers

0

1

2

3

4

0

1

2

3

0

1

2

3

N-CAM180 kDa140kDa

TH 70kDa

2D 24h 48h 96h 2w

2D 24h 48h 96h 2w 2D 24h 48h 96h 2w 24h 48h 96h 2w

THN-CAM 140

N-CAM 180

Ratio

Ratio

Ratio

lowastlowastlowast

RCCS

GAP43

C0

GAP43 46kDa

Figure 6 Expression of neuronal cell markers Representative Western blotting and quantification of the levels of N-CAM GAP43 andtyrosine hydroxylase (TH) in SH-SY5Y cells cultured in 2D monolayers and in the RCCS bioreactor for 24 h 48 h 96 h and 2 weeks (2 w)Data are densitometric ratio analyses as means plusmn SEM from 3 independent experiments lowast119875 lt 005 and lowastlowast119875 lt 001 versus 2D monolayers


GL15 SH-SY5Y Coculture0

1

2

3

4

Agg

rega

te ar

ea (m

m2 )

Figure 7 Cell aggregate sizes Quantification of section area ofGL15 SH-SY5Y and cocultured (GL15 plus SH-SY5Y) cell aggre-gates (as indicated) Data aremeans plusmn SEM (119899 = 15) for the averagedareas of the aggregate sections calculated using the ImageJ software(httpimagejnihgovij)

the 2 weeks of the GS-aggregates in the RCCS bioreactor(Figure 10) The N-CAM isoform expression pattern showeda slight but not significant decrease in N-CAM-180 levelsand a significant increase in N-CAM-140 levels (Figure 10)which resembled the N-CAM-140 increase observed in theS-aggregate homogenates

4 Discussion

There are evidences available showing that microgravity canaffect the functioning of the nervous system although thepossible physiological mechanisms of these effects remaindifficult to determine [7 28] Such difficulties in investiga-tions into microgravity effects are mainly due to the poormodels that are available either because of their high costand low availability (eg spaceflight) or because they are littlerepresentative of truemicrogravity conditions (eg hindlimbsuspensiondisusemodel) Among the ground-basedmodelsin vitro culture of cellstissues within clinorotation-basedsystems (eg random positioning machine RCCS bioreac-tor) represents a reasonable alternative to spaceflight TheRCCS bioreactor in particular was initially developed byNASA engineers to maintain cells in culture during spacemissions and to counteract the forces faced during shuttlelaunch and landing The RCCS bioreactor was further usedto maintain cells in dynamic 3D culture on the ground andbecause of its particular properties the RCCS bioreactoralso allows the modelling of microgravity on the groundSetting standardised parameters it is possible to also promotethe colocalisation of cells the establishment of cell-cellcontacts and consequently the spontaneous formation ofmulticellular aggregates [11 13] Moreover the rotation speedcan also be regulated in such away that it is possible to reach avector-averaged gravity that simulates low-gravity conditions[14]

In the present study we designed and investigated a pow-erful human-derived 3D organotypic-like model of nervoussystem tissue The experimental strategy was to study this

3D cell aggregation in terms of the cell phenotypes followingshort-term culture (up to 48 h as a time that allows theformation of multicellular aggregates) and long-term (upto 2 weeks) culture to analyse the effects of this modelledmicrogravity system on cell behaviour However apart fromthe effects related to microgravity the development of areliable neuroglia cell in vitro model is of great interest forbasic and clinical research in the field of the nervous systemThus we developed astrocyte-like and neuron-like in vitromodels here as 3D monotypic (GL15 cells only SH-SY5Ycells only) and heterotypic (cocultures of both GL15 and SH-SY5Y cells) cell cultures in the RCCS bioreactor

The particular dynamic conditions in the RCCS bioreac-tor have been shown to favour the differentiated phenotypeexpression for numerous cell and tissue types [13 24 29ndash31]In our hands over 48 h of culture these optimal dynamicconditions favoured spontaneous formation of healthy mul-ticellular aggregates according to the cell type consideredas demonstrated by the low cell death in these spontaneouscell aggregates The survival of these G-aggregates and S-aggregates and also of the GS-aggregates was assessed for upto 2 weeks in the RCCS bioreactor cultures and the dataconfirm the absence of significant necrosis in their centralcores in contrast to what has been reported in the literaturefor similar static culture conditions [32] This evidence sup-ported our choice to use the spontaneously formed aggregatemethod as this allowed the random distribution of the cellsinside the aggregates which is a feature that is particularlyimportant for the establishment of the heterotypic coculturemodel

Under our 3D cell culture conditions in growth mediumthe GL15 cells showed an astrocyte-like phenotype withthe expression of the glial-specific markers GFAP [33] andS100B Interestingly under these conditions Cx43 expressionwas also evident in these G-aggregates These data confirmthe importance of cell-cell interactions in the regulation ofphenotypic expression The modulation of Cx43 expressionmight be related to the formation of these G-aggregates in the3D culture During the first phase of G-aggregate formationthere was upregulation of Cx43 expression In a previousstudy we showed that these GL15 cells express Cx43 andform junctional channels where the permeability is directlyrelated to the cell proliferation rate as it decreased whentheir differentiated status was reached [16] In the presentstudy this transient upregulation of Cx43 duringG-aggregateformation might support the hypothesis that Cx43 has acrucial role and function in cellular aggregation in additionto its well-known involvement in differentiation processes[34] This hypothesis was also supported by Cotrina andcolleagues who demonstrated a role for Cx43 hemichannelsin cellular adhesion of C6 glioma cells [35]

The optimal dynamic culture conditions provided by theRCCS bioreactor were also demonstrated by the favouredexpression of neuronal-specificmarkers by the SH-SY5Y cellsin the S-aggregates such as tyrosine hydroxylase GAP43and N-CAM The expression levels of tyrosine hydroxylaseappeared similar in both the 2D and the 3D cultures atall of the times tested which demonstrated the adrener-gic phenotype that was expressed by these S-aggregates


N-CAM N-CAMGFAPGFAP

Figure 8 Localisation of glial and neuronal cell markers in the GS-aggregates Representative confocal images of GS-aggregates culturedunder the modelled microgravity for 2 weeks and immunostained with anti-N-CAM and anti-GFAP antibodies (as indicated) Insets showimage magnification Scale bars 20120583m

N-CAM N-CAMCx43PICx43

Figure 9 Localisation of cell-cell interaction markers in the GS-aggregates Representative confocal images of GS-aggregates cultured undermodelled microgravity for 2 weeks and immunostained with anti-N-CAM and anti-Cx43 antibodies (as indicated) The GS-aggregatesections were also stained with propidium iodide (PI right) Insets show image magnification Scale bars 10 120583m

GAP43 expression increased during the cell aggregation(48 h) which confirms active cell-cell interactions with thecytoskeletal modifications shown by GAP43 regulation Thestabilisation of these cell aggregates is supported by theincrease in N-CAM expression

During the long-termexposure tomodelledmicrogravityspecific protein expression was differently regulated in thecell aggregates Even after 2 weeks under culture in themodelled microgravity in the G-aggregates the glial-specificand functional markers (ie GFAP S100B and Cx43) showedlocalisation patterns thatwere similar to those observed in themonolayers under normal gravity conditions Interestinglyunder microgravity S100B and Cx43 expression levels in theG-aggregates were downregulated over two weeks as com-pared to those in the G-aggregate cultures at 48 h exposurewhereas there was a slight transient although not significanteffect on GFAP expression

The exact physiological roles of GFAP in astrocytesremain incompletely understood although they appear to beinvolved in cell-shape maintenance nervous system cytoar-chitecture mechanical stability and synaptic function [36]On the other hand it is well-known that Cx43 modulationis involved in neuronal development and plasticity [37] and

that S100B is expressed and also secreted by astrocytes andcan thus be an extracellular mediator of cell signalling [38]This evidence supports the hypotheses that the microgravitycan affect not only cell shape but also cell function

In the S-aggregates the modelled microgravity condi-tions did not have any significant effects on the localisationof N-CAM and tyrosine hydroxylase but they were shownto induce a switch of the GAP43 protein from the neurite-like processes to the cytoplasmic compartment In additionthe microgravity induced a slight although not significantdecrease in the expression levels of GAP43 and tyrosinehydroxylase while it had no effect on the expression of theN-CAM-180 isoform but significantly increased the expressionof the N-CAM-140 isoform which has been shown to havea key role in neuronal survival and signal transduction [3940] This suggests the involvement of N-CAM-140 duringthis modelled microgravity exposure that could promote asignificant degree of neuronal remodelling and survival

It has been previously reported that in neuroglialcocultures the neurons induce a reduction in astrocyteproliferation [41] In particular this effect was mediated bymembrane-membrane interactions between the neurons andthe astroglia in vitro and raised the possibility that membrane


00

05

10

15

20

N-CAM 180 kDa

N-C

AM

180

N-CAM 140kDa

N-C

AM

140

Cx43 86kDa

(dimer)

Cx43 43kDa

(monomer)

24h 2w

24h2w

GS-aggregates

Ratio

Cx43

(dim

er)

Cx43

(mon

omer

)

lowast

lowast

lowast

Figure 10 Expression levels of functional markers in the GS-aggregates Representative Western blotting and quantification of the levels ofN-CAM and Cx43 in GS-aggregates cultured in the RCCS bioreactor for 24 h and 2 weeks (2 w) Data are densitometric ratio analyses asmeans plusmn SEM from 3 independent experiments lowast119875 lt 005 for 2-week versus respective 24 h cultures

elements involved in glial cell growth regulation includeneuron-glial interaction molecules [41] In our neuron-likeand glial-cell-like coculture we focused our attention on N-CAM and Cx43 expression as these participate in importantintercellular signal interactions The GL15 cells used heredid not express N-CAM isoforms (data not shown) and theimmunofluorescence signals might reveal SH-SY5Y homo-typic interactions even if N-CAM heterotypic interactionscould not be excluded (such as N-CAM-integrins) Howeverthe N-CAM expression pattern in homogenates from GS-aggregates resembled that for the S-aggregates with anincrease in the expression of N-CAM-140 one of the threemain isoforms of N-CAM that is implicated in regenerationand remodelling of the nervous system [42]

Gap-junction-mediated intercellular communicationamong astrocytes has long been thought to contributeto tissue homeostasis in the brain [43] Cx43 has beenused as a marker to investigate neuron-glia interactions[22] Astrocytes express Cx30 and Cx43 which can formhomotypic (Cx43Cx43 andCx30Cx30) but not heterotypicjunctions [44] Interestingly in the homogenates from theGS-aggregates in addition to the classical 43 kDa formof Cx43 a dimeric form of Cx43 (ie 86 kDa) was alsoexpressed This Cx43 dimeric form has been related to

stress conditions In other models oxidative stress status hasbeen related to the appearance of such a higher molecularweight band for Cx43 which suggests that this representsan aggregated form of Cx43 [45] Under our conditions thepresence of this dimeric form of Cx43 might reveal a firstphase of impact between the neuronal-like and the glial-likephenotypes subsequently these Cx43 forms significantlydecreased when the GS-aggregates reached dynamic adaptiveconditions

5 Conclusions

In conclusion the evidence presented here suggests that the3D laminar flow the high mass transfer and the low shear-stress microenvironment generated by the RCCS bioreactorrepresent optimal conditions for the well-being of monotypicneural-like and glial-like cells as well as for heterotypicaggregates and for long-term culture Moreover such modelsystem can reproduce 3D cell-cell interactions that are similarto those under in vivo conditions [46] and can mimic themicrogravity conditions of exposure Our data highlight howsome phenotypic markers of monotypic and heterotypicneuroglia culture models can be influenced by microgravity


The data presented here open a wide range of specificinvestigations in terms of cell transduction pathways cell-cellinteractions and signalling and heterotypic culture biologyand the cell models we have described and analysed hererepresent important tools in the study of in vitro biologicaland pathological processes of the nervous system




Caterina Morabito and Nathalie Steimberg equally con-tributed to this study

Acknowledgments

The authors wish to thank Jennifer Boniotti (UNIBS) fortechnical support This study was supported by ASI 2013 Dec1342013 to MAM and by G drsquoAnnunzio University researchfunds to MAM

References

[1] G Vunjak-Novakovic N Searby J de Luis and L E FreedldquoMicrogravity studies of cells and tissuesrdquo Annals of the NewYork Academy of Sciences vol 974 pp 504ndash517 2002

[2] A-M Duprat D Husson and L Gualandris-Parisot ldquoDoesgravity influence the early stages of the development of thenervous system in an amphibianrdquo Brain Research Reviews vol28 no 1-2 pp 19ndash24 1998


[4] G Pani N Samari R Quintens et al ldquo Morphological andphysiological changes in mature in vitro neuronal networkstowards exposure to short- middle- or long-term simulatedmicrogravityrdquo PLOS ONE vol 8 no 9 Article ID e73857 2013

[5] R Gruener and G Hoeger ldquoVector-averaged gravity altersmyocyte and neuron properties in cell culturerdquo Aviation Spaceand Environmental Medicine vol 62 no 12 pp 1159ndash1165 1991

[6] H Rosner T Wassermann W Moller and W Hanke ldquoEffectsof altered gravity on the actin and microtubule cytoskeleton ofhuman SH-SY5Y neuroblastoma cellsrdquo Protoplasma vol 229no 2ndash4 pp 225ndash234 2006

[7] A Crestini C Zona P Sebastiani et al ldquoEffects of simulatedmicrogravity on the development and maturation of disso-ciated cortical neuronsrdquo In Vitro Cellular amp DevelopmentalBiologyminusAnimal vol 40 no 5-6 pp 159ndash165 2004

[8] L Lossi S Alasia C Salio and A Merighi ldquoCell deathand proliferation in acute slices and organotypic cultures ofmammalian CNSrdquo Progress in Neurobiology vol 88 no 4 pp221ndash245 2009

[9] K Rambani J Vukasinovic A Glezer and S M PotterldquoCulturing thick brain slices an interstitial 3D microperfusionsystem for enhanced viabilityrdquo Journal of NeuroscienceMethodsvol 180 no 2 pp 243ndash254 2009

[10] L E Freed and G Vunjak-Novakovic ldquoSpaceflight bioreactorstudies of cells and tissuesrdquo Advances in Space Biology andMedicine vol 8 pp 177ndash195 2002

[11] D A Wolf and R P Schwarz ldquoRP analysis of gravity-inducedparticle motion and fluid perfusion flow in the NASA-designedrotating zero-heaed-space tissue culture vesselrdquo NASA Techni-cal Paper 3143 1991

[12] C M Begley and S J Kleis ldquoThe fluid dynamic and shearenvironment in the NASAJSC rotating-wall perfused-vesselbioreactorrdquo Biotechnology and Bioengineering vol 70 no 1 pp32ndash40 2000


[14] P S Ayyaswamy and K Mukundakrishnan ldquoOptimal condi-tions for simulating microgravity employing NASA designedrotating wall vesselsrdquo Acta Astronautica vol 60 no 4ndash7 pp397ndash405 2007

[15] V Bocchini T Beccari C Arcuri L Bruyere C Fages and MTardy ldquoGlial fibrillary acidic protein and its encoding mRNAexhibit mosaic expression in a glioblastoma multiform cellline of clonal originrdquo International Journal of DevelopmentalNeuroscience vol 11 no 4 pp 485ndash492 1993

[16] M A Mariggio G Mazzoleni T Pietrangelo et al ldquoCalcium-mediated transductive systems and functionally active gapjunctions in astrocyte-like GL15 cellsrdquo BMC Physiology vol 1no 1 article 4 2001

[17] Y-T CheungW K-W Lau M-S Yu et al ldquoEffects of all-trans-retinoic acid on human SH-SY5Y neuroblastoma as in vitromodel in neurotoxicity researchrdquo NeuroToxicology vol 30 no1 pp 127ndash135 2009

[18] S Guarnieri R Pilla C Morabito et al ldquoExtracellular guano-sine and GTP promote expression of differentiation mark-ers and induce S-phase cell-cycle arrest in human SH-SY5Yneuroblastoma cellsrdquo International Journal of DevelopmentalNeuroscience vol 27 no 2 pp 135ndash147 2009

[19] M Miloso D Villa M Crimi et al ldquoRetinoic acid-inducedneuritogenesis of human neuroblastoma SH-SY5Y cells is ERKindependent and PKC dependentrdquo Journal of NeuroscienceResearch vol 75 no 2 pp 241ndash252 2004

[20] S Pahlman J C Hoehner E Nanberg et al ldquoDifferentiationand survival influences of growth factors in human neuroblas-tomardquo European Journal of Cancer Part A General Topics vol31 no 4 pp 453ndash458 1995

[21] T Fellin ldquoCommunication between neurons and astrocytesrelevance to the modulation of synaptic and network activityrdquoJournal of Neurochemistry vol 108 no 3 pp 533ndash544 2009

[22] A Vernadakis ldquoGlia-neuron intercommunications and synap-tic plasticityrdquo Progress in Neurobiology vol 49 no 3 pp 185ndash214 1996

[23] G Mazzoleni D Di Lorenzo and N Steimberg ldquoModellingtissues in 3D the next future of pharmaco-toxicology and foodresearchrdquo Genes and Nutrition vol 4 no 1 pp 13ndash22 2009

[24] N Steimberg J Boniotti and G Mazzoleni ldquo3D culture ofprimary chondrocytes and bonecartilage tissue explants insimulated microgravityrdquo inMethods in Bioengineering Alterna-tive Technologies to Animal Testing M Yarmush and R LangerEds Artech House 2010

[25] D A Goodenough and D L Paul ldquoGap junctionsrdquo Cold SpringHarbor Perspectives in Biology vol 1 no 1 Article ID a0025762009

[26] M I Mosevitsky ldquoNerve ending ldquosignalrdquo proteins GAP-43MARCKS and BASP1rdquo International Review of Cytology vol245 pp 245ndash325 2005


[27] D A Lewis D S Melchitzky and J W Haycock ldquoFourisoforms of tyrosine hydroxylase are expressed in human brainrdquoNeuroscience vol 54 no 2 pp 477ndash492 1993

[28] I B Krasnov ldquoGravitational neuromorphologyrdquo Advances inSpace Biology and Medicine vol 4 pp 85ndash110 1994

[29] X Chen H Xu C Wan M McCaigue and G Li ldquoBioreactorexpansion of human adult bone marrow-derived mesenchymalstem cellsrdquo Stem Cells vol 24 no 9 pp 2052ndash2059 2006

[30] G Mazzoleni F Boukhechba N Steimberg J Boniotti J MBouler and N Rochet ldquoImpact of dynamic culture in theRCCSTM bioreactor on a three-dimensional model of bonematrix formationrdquo Procedia Engineering vol 10 pp 3670ndash36752011


[32] P Humphreys S Jones and W Hendelman ldquoThree-dimensional cultures of fetal mouse cerebral cortex in acollagen matrixrdquo Journal of Neuroscience Methods vol 66 no1 pp 23ndash33 1996

[33] G Moretto N Brutti V de Angelis C Arcuri and V BocchinildquoA time-dependent increase in glial fibrillary acidic proteinexpression and glutamine synthetase activity in long-termsubculture of the GL15 glioma cell linerdquo Cellular and MolecularNeurobiology vol 17 no 5 pp 509ndash519 1997

[34] T Nakase and C C G Naus ldquoGap junctions and neurologicaldisorders of the central nervous systemrdquo Biochimica et Biophys-ica ActamdashBiomembranes vol 1662 no 1-2 pp 149ndash158 2004

[35] M L Cotrina J H-C LIN and M Nedergaard ldquoAdhesiveproperties of connexin hemichannelsrdquo Glia vol 56 no 16 pp1791ndash1798 2008

[36] VMenetMGimenezYRibottaNChauvet et al ldquoInactivationof the glial fibrillary acidic protein gene but not that ofvimentin improves neuronal survival and neurite growth bymodifying adhesion molecule expressionrdquo Journal of Neuro-science vol 21 no 16 pp 6147ndash6158 2001

[37] R RozentalM Srinivas S Gokhan et al ldquoTemporal expressionof neuronal connexins during hippocampal ontogenyrdquo BrainResearch Reviews vol 32 no 1 pp 57ndash71 2000

[38] R Donato G Sorci F Riuzzi et al ldquoS100Brsquos double lifeintracellular regulator and extracellular signalrdquo Biochimica etBiophysica Acta vol 1793 no 6 pp 1008ndash1022 2009

[39] M A Mariggio C Morabito S Guarnieri A Gentile KKolkova and G Fano ldquoIgIII (270-280)-fragment-like H

2N-

DDSDEEN-COOH peptide modulates N-CAM expression viaCa2+-dependent ERK signaling during ldquoin vitro neurogenesisrdquordquoPeptides vol 29 no 9 pp 1486ndash1497 2008

[40] P S Walmod K Kolkova V Berezin and E Bock ldquoZippersmake signals NCAM-mediated molecular interactions andsignal transductionrdquoNeurochemical Research vol 29 no 11 pp2015ndash2035 2004

[41] M E Hatten ldquoNeuronal inhibition of astroglial cell prolifera-tion is membrane mediatedrdquo Journal of Cell Biology vol 104no 5 pp 1353ndash1360 1987

[42] L C B Roslashnn V Berezin and E Bock ldquoTheneural cell adhesionmolecule in synaptic plasticity and ageingrdquo International Jour-nal of Developmental Neuroscience vol 18 no 2-3 pp 193ndash1992000

[43] M Theis G Sohl J Eiberger and K Willecke ldquoEmergingcomplexities in identity and function of glial connexinsrdquoTrendsin Neurosciences vol 28 no 4 pp 188ndash195 2005

[44] J L Orthmann-Murphy M Freidin E Fischer S S Schererand C K Abrams ldquoTwo distinct heterotypic channels mediategap junction coupling between astrocyte and oligodendrocyteconnexinsrdquo Journal of Neuroscience vol 27 no 51 pp 13949ndash13957 2007

[45] C M L Hutnik C E Pocrnich H Liu D W Laird and QShao ldquoThe protective effect of functional connexin43 channelson a human epithelial cell line exposed to oxidative stressrdquoInvestigative Ophthalmology and Visual Science vol 49 no 2pp 800ndash806 2008

[46] E Fennema N Rivron J Rouwkema C van Blitterswijk andJ de Boer ldquoSpheroid culture as a tool for creating 3D complextissuesrdquo Trends in Biotechnology vol 31 no 2 pp 108ndash115 2013

Review ArticleThe Impact of Microgravity and Hypergravityon Endothelial Cells

Jeanette A M Maier1 Francesca Cialdai2 Monica Monici2 and Lucia Morbidelli3

1 Department of Biomedical and Clinical Sciences ldquoL Saccordquo Universita di Milano Via Gian Battista Grassi 74 20157 Milan Italy2 ASAcampus Joint Laboratory ASA Research Division Department of Experimental and Clinical Biomedical Sciences ldquoM SeriordquoUniversity of Florence Viale Pieraccini 6 50139 Florence Italy

3 Department of Life Sciences University of Siena Via A Moro 2 53100 Siena Italy

Correspondence should be addressed to Lucia Morbidelli morbidelliunisiit

Received 4 July 2014 Revised 20 October 2014 Accepted 4 November 2014


Copyright copy 2015 Jeanette A M Maier et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

The endothelial cells (ECs) which line the inner surface of vessels play a fundamental role in maintaining vascular integrity andtissue homeostasis since they regulate local blood flow and other physiological processes ECs are highly sensitive to mechanicalstress including hypergravity and microgravity Indeed they undergo morphological and functional changes in response toalterations of gravity In particular microgravity leads to changes in the production and expression of vasoactive and inflammatorymediators and adhesion molecules which mainly result from changes in the remodelling of the cytoskeleton and the distributionof caveolae These molecular modifications finely control cell survival proliferation apoptosis migration and angiogenesis Thisreview summarizes the state of the art on how microgravity and hypergravity affect cultured ECs functions and discusses somecontroversial issues reported in the literature

1 The Endothelium

The concept of endothelium as an inert barrier lining theinner side of blood vessels has been overcome by thefinding that the endothelium is a dynamic heterogeneousand disseminated organ which orchestrates blood vessel andcirculatory functions thus exerting a critical role for tissuehomeostasis Indeed the endothelial cells (ECs) possessessential secretory synthetic metabolic and immunologicactivities [1 2]

The endothelium is semipermeable and regulates thetransport of variousmolecules between the blood and under-lying interstitial space by expressing specific carriers ECs alsocontrol vascular permeability especially in microvasculardistricts Moreover ECs importantly contribute to main-taining a nonthrombogenic blood-tissue interface since theyrelease various antithrombotic and fibrinolytic factors as wellas molecules that impact on platelets [1 2]

The endothelium is an immunocompetent organ becauseit exposes histocompatibility and blood group antigens canbe induced to express adhesion molecules for leukocytes

and produce cytokines Finally a functional relation existsbetween endothelial and smooth muscle cells as a conse-quence of the presence of junctions allowing the passageof electric charges and metabolites and the productionand release of vasoactive mediators [1 2] Indeed ECsfinely control vasomotor responses through the productionand metabolism of vasoactive molecules acting on smoothmuscle cells as endothelin-1 (ET-1) nitric oxide (NO) andangiotensin II (AngII) They also tightly control smoothmuscle cells proliferation [1 2] ECs are protagonists inangiogenesis that is the formation of new blood vessels frompreexisting ones Angiogenesis involves the most dynamicfunctions of the endothelium since it requires the migrationof ECs their ability to degrade the extracellular matrixtheir proliferation and differentiation ultimately leading tofunctional capillaries [3] This highly organised process ismodulated by the balance between stimulators and inhibitorsof angiogenesis

Vascular endothelium is structurally and function-ally heterogeneous [4] This heterogeneity is detectable at



different levels that is markers of cell activation gene expres-sion responsiveness to growth factors and antigen com-position and differentiates the behaviour between micro-and macrovascular ECs as well as between cells isolatedfrom different organs and from different vascular districtsof the same organ In fact the arteriolar endothelium isdifferent from the venous one as well as from the micro- andmacrovessel derived ECs The endothelium of the cerebralcirculationmdashwhich is the main component of the blood-brain barrier to protect the brain from toxic substancesmdashdeserves special consideration It is continuous has tightjunctions and differs both from fenestrated endotheliumwhere cells have pores and from discontinuous endotheliumwhere cells have intracellular and transcellular discontinuities[2]

ECs are normally quiescent in vivo with a turnover rateof approximately once every three years [5] Most of ECs inthe adult have a cell cycle variable from months to yearsunless injury to the vessel wall or angiogenesis occurs Onlyendothelium from endometrium and corpus luteum has adoubling time of weeks

ECs act as mechanotransducers whereby the transmis-sion of external forces induces various cytoskeletal changesand activates second messenger cascades which in turnmay act on specific response elements of promoter genesTherefore it is not surprising that ECs are sensitive tovariations of gravity

2 Methods to Simulate Microgravityand Hypergravity on Earth

Gravity is exerted permanently on organisms which are inconstant orientation in the gravity field (static stimulation)and if their orientation is changed with respect to the gravityvector (dynamic stimulation) [6]

The only way to achieve real microgravity is to useparabolic flights rockets space crafts or space labs asavailable on the International Space Station (ISS) Howeverthe possibility to perform experiments in real microgravityis limited because of high costs and the limited numberof missions On the other hand the short duration ofmicrogravity conditions achieved by using parabolic flightsor rockets limits the studies of many complex and prolongedbiological processesTherefore many efforts have been madeto establish methods to simulate microgravity on Earth Allthe devices available however mimic only some aspects ofreal microgravity

21 Clinostat Clinostats are considered reasonably effectiveground-based tools for simulating microgravity [7ndash10] andhave been used to study the effects of microgravity [11ndash19]

The clinostat randomizes motion and theoreticallyreduces the uniform gravity influence In the more widelydiffuse design that is the random positioning machine(RPM) the clinostat consists of an inner chamber containingthe samples which rotate clockwise anticlockwise verticallyand horizontally The horizontal and vertical motions areprovided by an outer chamber All the chambers are operated

by small motors under computer controlThe cells are grownin cell chambers or in flasks filled completely with mediathus diminishing the likelihood of turbulence and shearforces during culture rotation When using the clinostatto simulate microgravity the shear stress and vibrationsgenerated by the clinostat must be taken into account Shearstress can be limited by completely filling the chamberwith the culture medium Parallel controls are necessary toeliminate the effects of vibrations It is also important toconsider the distance of the samples from the centre of theplatform where the maximal reduction of gravity occursAnother important parameter to monitor is the speed ofrotation It has been verified that the effects of clinostat-determined microgravity are similar to those obtained inspace labs [8ndash19]

22 Rotating Wall Vessel Bioreactor This device wasdeveloped at NASArsquos Johnson Space Center to simulatethe effects of microgravity on cells in a ground-basedculture system The bioreactor the rotating wall vessel(RWV)rotating cell culture system (RCCS) from Synthecon(httpwwwsyntheconcom) is a cylindrical vessel thatmaintains cells in suspension by slow rotation around itshorizontal axis with a coaxial tubular silicon membranefor oxygenation Adherent cells need to be cultured onbeads This system represents a new cell culture technologydeveloped for 3D cultures of different cell types andbiotechnological applications The vessel wall and themedium containing cells bound to microcarrier beads or3D cultures rotate at the same speed producing a vector-averaged gravity comparable with that of near-Earth free-fallorbit [20] Most results obtained using the RWV wereconfirmed by experiments in real microgravity [12 21ndash23]

23 Magnetic Levitation This is a relatively novel Earth-based simulation technique used to investigate the biologicalresponse to weightlessness Magnetic levitation takes placewhen the magnetic force counterbalances the gravitationalforce Under this condition a diamagnetic sample is in asimulatedmicrogravity environment However the magneticfield which is generated affects cell behaviour thereforeconfounding the effects of simulated microgravity Mouseosteoblastic MC3T3-E1 cultured in a superconducting mag-net for 2 days showed marked alterations of gene expression[24] Random rotation and magnetic levitation inducedsimilar changes in the actin of A431 cells that were alsodescribed in real microgravity [25] At the moment howeverno studies are available on ECs under magnetic levitationbut they should be fostered as levitation as an alternative tosimulate microgravity might yield novel information or con-firm previous data thereby helping in designing successfulexperiments in real microgravity

24 Models to Generate Hypergravity Variation in gravityexposure is also related to hypergravity as the one to whichthe astronauts are transiently exposed during launch andreturn to Earth Also military pilots and subjects engaged


in certain sports such as motor racing motorcycling bob-sledding and the luge experience hypergravity The com-parison among the conditions of microgravity normogravity(1timesg) and hypergravity may be helpful to understand themechanisms underlying the effects of gravitational alterationson endothelial function and to understand what happenswhen humans quickly pass fromhypergravity tomicrogravityconditions and vice versa

Centrifuges constructed for research under hypergravityconditions are characterized by high precision control of rpmTheir speed and the angle of inclination of the sample can beregulated to obtain the desired hypergravity in a range from 1to many g Centrifuges are also used to perform 1timesg controlexperiments on board of the ISS and spacecraft Studies onendothelial cells in hypergravity are available [12 26 27]

3 The Effects of Microgravity on ECs

Exposure to microgravity during space missions impacts onvarious systems In humansmicrogravity-induced alterationsinclude bone loss muscle atrophy cardiovascular decondi-tioning impairment of pulmonary function and immuneresponse [50 51] The cardiovascular system is affected byspaceflight with changes manifesting as cardiac dysrhyth-mias cardiac atrophy orthostatic intolerance and reducedaerobic capacity [52] These changes can cause adaptationproblems when astronauts return back to Earth especiallyafter long-duration spaceflights [53]

Because ECs are key players in the maintenance ofvascular integrity inflammation and angiogenesis severalstudies have been devoted to the mechanisms by whichmicrogravity affects EC functions

Various reports have indicated that ECs are highly sensi-tive to microgravity and undergo morphological functionaland biochemical changes under these conditions [11 12 2328 29 37 38 46 49 54] These studies have used a varietyof in vitro cell models with divergent results One of thereasons for these discrepancies can be EC heterogeneity orthe isolation from different species Indeed human bovinemurine and porcine endothelial cells have been investigatedunder gravitational unloading With concern to human cellsstudies are available on human ECs from the umbilicalvein (HUVEC) widely considered a model of macrovas-cular endothelial cells as well as on human microvascularECs (HMEC) Moreover studies have been performed onEAhy926 cells a fusion of HUVEC with the lung carcinomacell line A549 [55] Although immortalized cell lines offersignificant logistical advantages over primary cells in in vitrostudies they exhibit important differences when compared totheir primary cell counterparts Indeed microarrays used fora genome-wide comparison between HUVEC and EAhy926in their baseline properties have shown that EAhy926 cellsare useful in studies on genes encodingmolecules involved inregulating thrombohemorrhagic features while they appearto be less suited for studies on the regulation of cell prolifera-tion and apoptosis [56] Moreover immortalized endothelialcell lines show different expression pattern of biomarkerswhen compared to primary cells [57] The controversial

results reported about the response of ECs to microgravitycould be due also to the diverse experimental approachesutilized such as the device simulating microgravity theduration of exposure to simulated microgravity and thedegree of reduction of the gravity that can be reachedoperating these devices differently (see above) Neverthelessaltered EC morphology cell membrane permeability andsenescence are documented by spaceflight experiments oncultured endothelium [21 30 58]

Several aspects of endothelial behaviour have been stud-ied in simulated and real microgravity Table 1 summarizesthe published findings

31 Migration Controversy exists on this topic No signif-icant modulation of cell migration under basal conditionand in response to the angiogenic factor hepatocyte growthfactor (HGF) was observed in HUVEC as well as in HMECcultured in the RPM [12 46] Shi et al [39] demonstratedthat after 24 h of exposure to simulated microgravity ina clinostat HUVEC migration was significantly promotedthrough the eNOS pathway upregulation by means of PI3K-Akt signalling On the contrary the endothelial cell lineEAhy926 in simulated microgravity migrated more thancontrols [31] while in a study on porcine aortic endothelialcells (PAEC) microgravity modelled by a RPM caused amarked impairment of cell migration induced by serum orthe angiogenic factors vascular endothelial growth factor(VEGF) and fibroblast growth factor-2 (FGF-2) [11]

32 Proliferation and Formation of 3D Structures Carlssonand Versari using the RWV and the RPM respectivelyfound that the proliferation rate of HUVECs was reversiblyincreased under simulated microgravity [12 28] Also bovineaortic ECs (BAEC) grew faster in the RWV than controls[35] On the contrary simulated microgravity inhibited thegrowth of HMEC and murine microvascular ECs [23 46]The results obtained using microvascular EC are reinforcedby the in vivo finding showing an impairment of angiogenesisin space Wound healing in which neovascularization is anearly and fundamental step is retarded in space-flown animalmodels [59] and the development of vascular channels in arat fibular osteotomy model is inhibited after flight as shownby an experiment carried out during a shuttle mission [60]

Also in PAECs a marked impairment of EC responsive-ness to angiogenic factors and a reduced ability to proliferatewere reported [11] Using the endothelial cell line EAhy926Grimm et al [43] showed the formation of 3D tubularstructures in clinorotation After two weeks a subtype of 3Daggregates was observed with a central lumen surroundedby one layer of ECs These single-layered tubular structuresresembled the intimas of blood vessels Characterization ofthese tubular structures revealed that they might originatefromdouble-row cell assemblies formedbetween the fifth andseventh day of culture under simulated microgravity [43]

33 Apoptosis Increased apoptosis after culture in the RPMhas been observed in PAEC and the endothelial cell line


Table1Th

eeffectso

frealorsim

ulated

microgravity

ondifferent

endo

thelialcelltypes

Experim

entalm

odel

Experim

entalcon

ditio

nsEff

ects

Authors

Prim

aryhu

man

umbilicalvein

ECs

(HUVEC

)

Rotatin

gwallvessel(RW

V)

Rand

ompo

sitioning

machine

(RPM

)48or

96h

Growth

stim

ulation

uarrNOprod

uctio

nAc

tinremod

ellin

gdarrAc

tin

Versarietal2007

[12]

Spaceflight

(Progress4

0Pmission)

10d

uarrTh

ioredo

xin-interactingprotein

darrhsp-70

and90

uarrsecretionof

IL-1120572andIL-1120573

Ionchannels(TPC

N1KC

NG2KC

NJ14

KCN

G1KC

NT1T

RPM1CL

CN4

CLCA

2)m

itochon

drialoxidativ

epho

spho

rylatio

nandfocaladh

esionwerew

idely

affected

Versarietal2013

[21]

RWV72

hor

96h

uarrPG

I2andNO

Carlssonetal2002

[22]

RPM

24ndash4

8huarrNOuarrCa

v-1p

hospho

rylatio

n(Tyr

14)

Spisn

ietal2006

[26]

RWV4244896144

h

uarruarrhsp70

darrIL-1120572

Remod

ellin

gof

cytoskele

ton

darractin

Carlssonetal2003

[28]

RPM

24h

uarreN

OSCa

v-1and

-2darrof

thelengthandwidth

ofthec

ells

darrICAM-1V

CAM-1E

-sele

ctinand

IL-6

andTN

F-120572

Greno

netal2013

[29]

Spaceflight

12d

Cytoskele

tald

amage

uarrcellmem

branep

ermeability

Inreadaptedcells

persistingcytoskele

talchanges

darrmetabolism

andcellgrow

th

Kapitono

vaetal2012

[30]

2D-C

linostat(developedby

ChinaA

stron

autR

esearchand

Training

Center)30

rpm24h

uarrHUVEC

tube

form

ationandmigratio

ndarrnu

mbero

fcaveolaeinthem

embrane

uarreN

OSactiv

ityby

phosph

orylationof

Akt

andeN

OS

Siam

walae

tal2010

[31]

RWV5m

in30m

in1hand24

h

uarrICAM-1expressio

nDepolym

erizationof

F-actin

andclu

sterin

gof

ICAM-1on

cellmem

brane(short

term

)Ac

tinfib

errearrangem

entand

stablec

luste

ringof

ICAM-1(afte

r24h

)uarrICAM-1andVC

AM-1RN

Aaft

er30

min

Zhangetal2010

[32]

RPM

96h

Alteratio

nof

proteins

regu

latin

gcytoskele

tonassembly

darrIL-1120572IL-8andbF

GF

uarrchem

okines

Rantes

andEo

taxin

involved

inleuk

ocytes

recruitm

ent

Griff

onietal2011

[33]

RPM

24h

uarriN

OSby

amechanism

depend

ento

nsupp

ressionof

AP-1

Wangetal2009

[34]


Table1Con

tinued

Experim

entalm

odel

Experim

entalcon

ditio

nsEff

ects

Authors

Bovine

aorticEC

s(BA

EC)

RWVforu

pto

30d

Growth

stim

ulation

uarrNO

Prod

uctio

nof

NOdepend

ento

ntheR

WVrotatio

nrate73

increase

at8r

pm

262

increase

at15rpmand

500

increase

at20

rpm

Sanfordetal2002

[35]

Porcinea

ortic

ECs(PA

EC)

RPM

72h

uarrproapo

ptoticgenes(p53FA

S-LBA

X)darrantia

poptoticgenes(Bc

l-2)

Diss

olutionof

mito

chon

drialm

embraneintegrity

Impairm

ento

fcellrespo

nsivenesstoexogenou

sstim

uli

Morbidelli

etal2005

[11]

Bovine

coronary

venu

larE

Cs(C

VEC

)RP

M72

h

uarrFibron

ectin

(form

ationof

intricaten

etworkof

FNfib

ers)

uarrLaminin

uarr120573-Actin

(form

ationof

stressfi

bers)

uarr120572120573-Integrin

(form

ationof

cluste

rs)

Mon

icietal2011

[36]

Hum

anEC

lineE

Ahy926

RPM

10days

uarrCa

spase-3Ba

xandBc

l-2uarrcollagentypesI

andIII

Alteratio

nsof

thec

ytoskeletal120572

-and120573-tu

bulin

sand

F-actin

darrbrain-deriv

edneurotroph

icfactorplatele

ttissue

factorV

EGFandET

-1

Infanger

etal2007

[37]

RPM

7days

Mod

ulationof

genese

ncod

ingforsignaltransdu

ctionandangiogenicfactorscell

adhesio

nmem

branetranspo

rtproteinsore

nzym

esinvolved

inserin

ebiosynthesis

Mae

tal2013

[38]

RPM

2h

uarrcellu

larm

igratio

nuarrfilop

odiaandlamellipod

iaAc

tinrearrangem

ents

uarrNO

Shietal2012

[39]

RPM

4122448and

72h

uarrextracellularm

atrix

(ECM

)proteins

Alteratio

nin

cytoskeletalcompo

nents

uarrexpressio

nof

ECM

proteins

(collagentype

Ifib

ronectinoste

opon

tinlam

inin)

andflk

-1protein

Morph

ologicalandbiochemicalsig

nsof

apop

tosis

after

4hfurther

increasin

gaft

er72

h

Infanger

etal2006

[40]

Parabo

licflight(22

smicrogravity18xg

2perio

dsof

20s)

Parabolic

flight

darrCO

L4A5CO

L8A1ITGA6ITGA10and

ITGB3

mRN

Asaft

erP1

(firstp

arabolas)

uarrED

N1and

TNFR

SF12AmRN

Asaft

erP1

darrADAM19C

ARD

8CD

40G

SNP

RKCA

mRN

AsuarrPR

KAA1(AMPK1205721)mRN

Ascytoplasmicrearrangem

ent

uarruarrABL

2aft

erP1

andP3

1

Grossee

tal2012

[41]

Parabo

licflight(22

smicrogravity18xg

2perio

dsof

20s)

Parabolic

flight

Actin

networkrearrangem

ent

uarrCC

NA2CC

ND1CD

C6C

DKN

1AE

ZRM

SNO

PNV

EGFA

CASP

3CA

SP8

ANXA1ANXA2andBIRC

5darrFL

K1uarrEZ

RMSN

OPN

ANXA2andBIRC

5aft

er31P

Wehland

etal2013

[42]


Table1Con

tinued

Experim

entalm

odel

Experim

entalcon

ditio

nsEff

ects

Authors

RPM

71421and28

d

Differentrespo

nsivenesstoVEG

FandbF

GFaddedexogenou

slyAlteredgene

andproteinexpressio

nof

phosph

okinaseA

catalytic

subu

nit

phosph

okinaseC

alph

aandER

K-1and

2darrVEG

FbF

GFsolubleT

NFR

SF5TN

FSF5ICA

M-1T

NFR

2IL-18complem

ent

C3and

vonWillebrand

factor

Grim

metal2010

[43]

RPM

7and28

dDelayed

3Dcellgrow

th

uarrbeta(1)-integrinlam

ininfibron

ectin

120572-tu

bulin

intube-like

structuresa

fter4

weeks

ofcultu

ring

Grim

metal2009

[44]

Hum

anEC

lineE

Ahy926

Bovine

lung

microvascular

ECS

Bovine

pulm

onaryaorticEC

sPo

rcinev

entricular

endo

cardialE

Cs

RPM

2h

Results

indicatethatiN

OSisam

olecular

switchforthe

effectsof

microgravity

ondifferent

kind

sofend

othelialcells

uarrangiogenesisviathe

cyclicg

uano

sinem

onop

hosphate(cGMP-)P

KGdepend

ent

pathway

Siam

walae

tal2010

[45]

Hum

anderm

almicrovascular

cells

(HMEC

)RW

VRP

M48

or96

or168h

uarrTIMP-2

uarrNOdarrproteasomea

ctivity

Mariotti

andMaier2008

[46]

Murinelun

gcapillary

ECs(1G

11cells)

RWV72

h

darrendo

thelialgrowth

uarrp21

darrIL-6

uarreN

OSandNO

Cotrupi

etal2005

[23]

Hum

anpu

lmon

arymicrovascular

ECs

(HPM

ECs)

MG-3

clino

stat(develo

pedby

the

Institu

teof

Biop

hysic

sChinese

Academ

yof

Sciences)

uarrapop

tosis

darrPI3K

Akt

pathway

uarrNF-120581Banddepo

lymerizationof

F-actin

Kang

etal2011

[47]

Hum

anandbo

vine

microvascular

ECs

RWV96

huarrhsp70in

cells

which

maintainedthec

apabilityto

proliferatein

microgravity

Cotrupi

andMaier

2004

[48]

Cocultureso

fend

othelialm

onolayers

human

lymph

ocytesimmun

ecells

and

myeloleucem

ic(K

-560)cells

Spaceflight

(ISS)

uarradhesio

nof

PMA-

activ

ated

lymph

ocytes

Retained

abilityof

immun

ecellsto

contactrecogn

izeanddestroyon

cogenicc

ells

invitro

Buravkovae

tal2005

[49]

Legend



EAhy926 [11 40] In particular following exposure to sim-ulated hypogravity PAEC change their morphology andgene expression pattern triggering proapoptotic signals Thegene expression profile demonstrated the upregulation ofp53 FAS-L and BAX genes and the concomitant down-regulation of the antiapoptotic protein Bcl-2 and prolif-eration marker PCNA The induction of apoptosis wasaccompanied by mitochondrial disassembly thus suggest-ing the activation of the mitochondrial intrinsic pathways[11]

In pulmonary HMEC simulated microgravity-inducedapoptosis by downregulating the PI3KAkt pathways andincreasing the expression of NF120581B [47] On the con-trary no apoptosis was observed in HUVEC and dermalHMEC cultured for various times in the RWV or in theRPM and this has been linked to the rapid inductionof heat shock protein (hsp)-70 [28 46 48] Indeed hsp-70 protects endothelial cells from apoptotic stimuli act-ing downstream of cytochrome c release and upstream ofcaspase 3

34 Alterations of Cytoskeleton and Extracellular Matrix Thecytoskeleton plays a key role in the adaptation to mechanicalstress including alterations of gravity [61 62] Therefore thechanges that cytoskeletal components such as microtubulesundergo inmicrogravity can be a key to explaining the effectsof weightlessness on cells [63 64]

Carlsson et al [28] studied actin microfilaments inHUVEC exposed to microgravity simulated by the RWVIn comparison with controls the cells showed elongatedand extended podia disorganization of actin microfilamentsthat clustered in the perinuclear area and decrease in stressfibers Moreover after 96 h exposure actin RNA levels weredownregulated and total actin amounts were reduced Thecytoskeletal modifications were reversible upon return tonormal growth conditions (1timesg) The authors speculatedthat the reduction in actin amount could be an adaptivemechanism to avoid the accumulation of redundant actinfibers The same results were obtained when the experimentwas replicated by using a RPM to model the micrograv-ity conditions [12] More recently in HUVEC exposed tomechanical unloading by RPM Grenon et al [29] found dis-organization of the actin network with clustering of the fibersaround the nucleus Moreover they observed that caveolin-1was less associated with the plasma membrane and adopteda perinuclear localization Thus the authors advanced thehypothesis that disruption of the actin cytoskeleton orga-nization could impair the translocation of caveolin-1 to thecaveolae

After spaceflight (Soyuz TMA-11) readapted HUVECcells with subsequent passages exhibited persisting changesin the organization of microtubules with prominent bundlesthat occupied the peripheral cytoplasm [30]

In a study carried out by Zhang et al [32] HUVEC acti-vated with TNF-120572 and exposed to microgravity modelled byRWV demonstrated that after 30min depolymerization ofF-actin and clustering of ICAM-1 on cellmembrane occurredMoreover ICAM-1 and VCAM-1 RNA were upregulated

After 24 h actin fiber rearrangement was initiated clusteringof ICAM-1 became stable and the mRNAs of ICAM-1 andVCAM-1 returned to levels comparable with the controlsThe authors speculated that actin cytoskeleton rearrangementand changes in levels and distribution of surface adhesionmolecules could significantly affect transendothelial migra-tion processes

Grosse et al [41] studied the effect of parabolic flighton the cytoskeleton of the endothelial cell line EAhy926Every parabola (P) included two hypergravity (18 g) periodsof 20 s separated by a 22 s microgravity period After P1they observed a rearrangement of120573-tubulin that accumulatedaround the nucleus After P31 120573-tubulin and vimentin weredownregulated Using the EAhy926 cell line exposed toparabolic flight Wehland et al [42] reported that the actinnetwork underwent a drastic rearrangement mostly affectedby vibration

Grimm et al [44] studied the walls of tube-like structuresspontaneously formed by the endothelial cell line EAhy926cultured in a RPM They found that the walls consistedof single-layered endothelial-like cells which had producedsignificantly more 120573

1-integrin laminin (LM) fibronectin

(FN) and 120572-tubulin than controls Microgravity-inducedupregulation of proteins involved in the extracellular matrixbuilding was confirmed in studies carried out by Moniciet al [36] on cultured bovine coronary venular endothelialcells (CVECs) exposed for 72 h to microgravity modelledby a RPM The authors observed an increase in actincontent and impressive production of actin stress fibersaccompanied by the overexpression and clustering of 120573

1-

integrin 40 increase in LM 111 increase in FN contentand formation of a tight and intricate network of FNfibrils

Since FN and LM are strongly involved in the regulationof cell adhesionmigration their upregulation and alterednetworking together with the changes in actin and integrinpatterns induced the authors to hypothesize that the exposureto microgravity causes a dysregulation in cell motility andadhesion to the substrate

In summary all of the studies carried out so far demon-strated that microgravity strongly affects cytoskeleton orga-nization and induces a rearrangement of the actin networkwith clustering of the fibers in the perinuclear area A similarbehaviour has been observed also analysing the microtubulenetwork Moreover clustering of adhesion molecules on theplasma membrane and overexpression of proteins of theextracellular matrix have been reported by some authors

The results are less consistent when considering theexpression of cytoskeleton proteins or their RNA Probablythe discrepancies are due to differences in experimentalmodels (different cell populations) protocols and analyticalprocedures

However it is widely accepted that the microgravity-induced changes in the cytoskeleton can strongly affect thebehaviour of endothelial cells in terms of adhesion migra-tion and production of extracellular matrix and can interferewith other processes such as translocation ofmolecules insidethe cells transendothelial migration and even inflammationand angiogenesis


35 Synthesis of VasoactiveMolecules The levels of vasoactivemolecules such as NO and ET-1 are modified under micro-gravity conditions which also indicates that microgravitymay influence both hemodynamic changes and angiogenesis[33 37] In particular HUVEC and HMEC exposed to simu-lated microgravity using RWV and RPM produce more NOthan controls as the result of increased levels of endothelial-nitric oxide synthase (e-NOS) [12] which correlates withthe increase of caveolins [26 29] In particular Grenonet al suggested that the alterations in NO production aremediated by changes in the cytoskeleton detected in all theendothelial types studied [29] Wang et al [34] explained theincreased amounts of NO in HUVEC after 24 h in simulatedmicrogravity as the results of the upregulation of inducibleNOS through a mechanism dependent on the suppression ofthe activity of the transcription factorAP-1 Also inBAECNOproduction was increased [35]

In the endothelial cell line EAhy926 a reduced releaseof ET-1 and VEGF was reported [37] while the productionof NO was increased via the iNOS-cGMP-PKG pathway [3945] If confirmed in vivo in space these results might in partexplain the hemodynamic changes and the redistribution ofblood flows induced by microgravity

36 Genomic and Proteomic Analysis Microgravity affectsseveral molecular features of ECs markedly modulatinggene expression In HUVEC cultured in the RPM thesecretome was evaluated by a 2D proteomic approach [33]The proangiogenic factor FGF-2 and the proinflammatorycytokines interleukin-1 (IL-1) and IL-8 were decreased insimulatedmicrogravity whereas two chemokines involved inleukocyte recruitment Rantes and Eotaxin were increased[33] The unprecedented gene profile analysis on HUVECcultured on the ISS for 10 days was performed by Versariet al [21] 1023 genes were significantly modulated themajority of which are involved in cell adhesion oxidativephosphorylation stress responses cell cycle and apopto-sis thioredoxin-interacting protein being the most upregu-lated Briefly in cultured HUVEC real microgravity affectsthe same molecular machinery which senses alterations offlow and generates a prooxidative environment that altersendothelial function and promotes senescence [21] Similarconclusions were reached by Kapitonova et al [30 58] whodescribed premature senescence in space-flown HUVEC Byaccelerating some aspects of senescencemicrogravity offers abig challenge to study themechanisms implicated in the onsetof aging

Looking at the endothelial cell line EAhy926 a shortterm lack of gravity (22 s) generated by parabolic flightssignificantly influences the signalling pathways [41] Whenthese cells are cultured for various times from 4 to 72 h onthe RPM a number of proteins of the extracellular matriximplicated in apoptosis are modulated when compared tocontrol cells [40] In theRPMsomeEAhy926 cells form tube-like 3D aggregates while others continue to grow adherently3D aggregates and adherent cells were analyzed by genearray and PCR techniques and compared to controls [38]1625 differentially expressed genes were identified and in

particular the levels of expression of 27 genes changedat least 4-fold in RPM-cultured cells when compared tocontrolsThese genes code for angiogenic factors and proteinsimplicated in signal transduction cell adhesion membranetransport or enzymes Fifteen of them with IL-8 and vonWillebrand factor being the most affected showed linkagesto genes of 20 proteins that are important in the maintenanceof cell structure and in angiogenesis

EAhy926 cell line and human dermal microvascular ECs(HMVECs) were then compared after culture on the RPM for5 and 7 days [54] A total of 1175 types of proteins were foundin EAhy926 cells and 846 in HMVECs 584 of which werecommon and included metabolic enzymes structure-relatedand stress proteins This proteomic study also highlightsthat HMVECs develop tube-like 3D structures faster thanEAhy926 possibly through a transient augmentation ofribosomal proteins during the 3D assembling of ECs

4 The Effects of Hypergravity on ECs

A summary of published data on endothelial cell behaviouris reported in Table 2 HUVECs exposed to hypergravity(3timesg) for 24ndash48 h showed inhibition of cell growth butunaltered apoptosis increased COX-2 eNOS and Cav-1suggesting a possible role of caveolae in mechanotransduc-tion Also an increased synthesis of PGI2 and NO whichare also proangiogenic was observed However surprisinglythe formation of capillary-like structure was inhibited [65]Versari et al [12] studying the same cells exposed to 35timesgfound increased NO production enhanced cell migrationbut no effects on proliferation Moreover altered distributionof actin fibers without modifications of the total amountsof actin was detected [12] In the same conditions HUVECshowed a time-dependent decrease in occludin correlatingwith an increase in paracellular permeability and a decreasein transendothelial electrical resistance indicating a decreasein EC barrier function [66 67] with exactly opposingresults in BAEC cultured under hypogravity in RWV whereincreased barrier properties were detected [35]

Koyama et al [69] reported that after a few minutesof exposure to 3timesg in a centrifuge BAECs showed actinreorganization via Rho activation and FAK phosphorylationincreased cell proliferation andATP release Adaily exposureof 1-2 h repeated for 5 consecutive days promoted cell migra-tion Wehland et al [42] investigated short term (s) effectsof hypergravity (18timesg) on EAhy926 cells and found that thecells were weakly affected by loading in the conditions usedfor the experiment On the contrary short term effects ofmicrogravity were much more evident

In order to evaluate these results two considerations haveto be made

(1) Very different protocols and parameters have beenused for EC exposure to hypergravity continuousversus discontinuous exposure different g force valueand exposure times ranging from minutes to days

(2) ECs both derived from the microvasculature andmacrocirculation are very sensitive to mechanicalstress It should be underscored that in physiological


Table2Th

eeffectso

fhypergravity

cond

ition

sondifferent

endo

thelialcelltypes

Experim

entalm

odel

Experim

entalcon

ditio

nsEff

ects

Authors

Prim

aryhu

man

umbilicalvein

ECs(HUVEC

)

Hypergravity

cond

ition

s(generatedby

aMidiCAR

centrifugea

t35xg)for

24ndash4

8h

uarrmigratio

nuarrNO

Altereddistrib

utionof

actin

fibers

Versarietal2007

[12]

Hypergravity

cond

ition

s(generatedby

acentrifu

geat

3xg)

for2

4ndash48

h

uarrcav-1

uarrdistrib

utionof

caveolae

inthec

ellinterior

uarrCO

X-2NOand

PGI2

prod

uctio

ndarrangiogenesis(th

roug

hap

athw

ayno

tinvolving

apop

tosis)

Spisn

ietal2003

[65]

Lifto

ffsim

ulationby

centrifuge(75

min

simulationof

thep

attern

ofgforces

experie

nced

durin

glift

offof

the

NASA

spaces

huttle)

darrMAPK

phosph

orylation

uarrocclu

dinexpressio

nSumanasekerae

tal2006

[66]

Lifto

ffsim

ulationby

centrifuge(75

min

simulationof

thep

attern

ofgforces

experie

nced

durin

glift

offof

the

NASA

spaces

huttle)

uarrParacellu

larp

ermeability

darrOccludin

darrTransend

othelialelectric

alresistance

darrMAPK

activ

ation

darrEC

barrierfun

ction

Sumanasekerae

tal2007

[67]

Bovine

aorticEC

s(BA

EC)

Hypergravity


aZentrifu

gen

5perio

dsof

10min

expo

sure

to10xg

spaced

with

10min

at1xg)

Mod

ified

integrin

distrib

ution

Reorganizatio

nof

cytoskele

taln

etwork

darrgenesc

ontro

lling

vasoconstrictio

nandinflammation

darrProapo

ptoticsig

nals

Morbidelli

etal2009

[68]

Hypergravity

(3xg)app

liedby

lowspeedcentrifuge

uarrAT

Prelease

uarractin

reorganizatio

nviaR

hoA

activ

ationandFA

Kph

osph

orylation

uarrcellproliferatio

nandmigratio

n

Koyamae

tal2009

[69]

Bovine

coronary

venu

larE

Cs(C

VEC

)

Hypergravity


aZentrifu

gen

5perio

dsof

10min

expo

sure

to10xg

spaced

with

10min

at1xg)

darrproapo

ptoticgenes(FA

DDFasFas-L)

uarrantia

poptoticgene

NF120581

BChangeincytoskele

tonorganizatio

nAlteratio

nof

cellenergy

metabolism

Mon

icietal2006

[27]

Hum

anEC

lineE

Ahy926

Hypergravity

Experim

ents(M

uSICD

LRC

ologneG

ermany

centrifuge18xg)

Vibrationexperim

ents(V

ibraplex

vibrationplatform

frequ

ency

range0

2ndash14k

Hz)

darrCA

RD8NOS3V

ASH

1SE

RPIN

H1(allP

1)C

AV2ADAM19

TNFR

SF12AC

D40

and

ITGA6(P31)m

RNAs

Nosig

nificantchanges

ongene

expressio

nandmorph

olog

yof

the

cells

Grossee

tal2012

[41]

Hypergravity

Experim

ents(M

uSICD

LRC

ologneG

ermany

centrifuge18xg)

Vibrationexperim

ents(V

ibraplex

vibrationplatform

frequ

ency

range0

2ndash14k

Hz)

darrPan-actin

tub

ulinand

Moesin

darrgene

expressio

nof

CCND1MSN

RDX

OPN

BIRC5

and

ACTB

uarrPan-actin

tub

ulinand

ezrin

darr120573-Actin

andMoesin

darrAC

TBC

CND1CD

C6C

DKN

1AV

EGFA

FLK

-1E

ZRITB

G1

OPN

CASP

3CA

SP8ANXA2andBIRC

5

Wehland

etal2013

[42]

Legend



Table 3 Summary of the principal parameters influenced by simulated microgravity and hypergravity in different types of ECs

Microgravity HypergravityEndothelial cellline EAhy926 Dermal HMEC HUVEC PAEC BAEC Endothelial cell

line EAhy926 HUVEC CVEC BAEC

Migration darr = =uarr darr ND ND uarr ND uarr

Proliferation ND darr uarr darr uarr ND (=uarr) = ND uarr

Apoptosis uarr = = uarr = =darr = = =NO synthesis uarr uarr uarr ND uarr ND uarr ND NDCytoskeletal rearrangements +++ +++ +++ +++ +++ +++ ++ +++ +++Legend uarr increased darr decreased = no change ND not determined ++ and +++ highly and strongly upregulated

conditions the quality and intensity of mechanicalstimulation to which the endothelium is exposeddepend on the vascular district

Following the latter consideration we hypothesized thatEC response to gravitational alteration could depend on thedistrict from which the cell population derives and could bedifferent in cells derived from macro- or microcirculationTo verify this hypothesis we studied and compared thebehaviour of coronary venular endothelial cells (CVEC) [27]and BAEC [68] exposed to 10timesg for 5 periods of 10 minuteseach spaced with four recovery periods of the same durationFollowing exposure both the cell types showed similarchanges in cytoskeleton organization and 120572v1205733 integrindistribution The peripheral ring of actin microfilaments wassubstituted by trans-cytoplasmic stress fibers microtubulesand intermediate filaments gathered in the perinuclear areafocal contacts in the protruding lamellipodia disappearedand 120572v1205733 integrin molecules clustered in the central bodyof the cells Both in CVEC and in BAEC the expressionof the cytoskeletal proteins 120573-actin and vimentin increasedIn BAEC the transcripts for the matrix proteins LM andFN decreased In both the cell types exposure to hypergrav-ity decreased the transcription of genes encoding for theproapoptotic factors Fas and FasL Bcl-XL [27 68]

Cell energy metabolism assessed by autofluorescencespectroscopy and imaging did not change significantly inBAECs On the contrary CVECs exposed to hypergravityshowed an increase of the anaerobic metabolism in compar-ison with 1timesg controls [27]

The phenotypic expression of molecules involved ininflammation and angiogenesis such as eNOS FGF-2 andCOX-2 which is not expressed in basal conditions didnot significantly change as assessed by immunofluorescencemicroscopy in CVECs Nevertheless in BAECs the expres-sion of COX-2 and other genes controlling the calibre of thevessels that is renin ET processing enzyme and inflamma-tion such as TNF120572 and its receptor CD40 P and E selectinsCD54 was downregulated Briefly hypergravity does notseem to affect significantly the survival of both macro- andmicrovascular ECs However significant changes have beenobserved in cytoskeleton and integrin distribution in all theECs studied and changes in cell energy metabolism havebeen observed only in CVECs while the downregulation ofsome genes involved in inflammation and vasoconstrictionhas been found only in BAECs Considering the expression of

growthmodulators hypergravity increasedVEGF expressionwhile it decreased a series of interleukins acting as inhibitorsof EC proliferation [27 68] These results are consistentwith the hypothesis that the EC response to gravitationalalterations depends at least in part on the vascular districtfrom which the cells are derived


The effects of simulated gravity changes on endothelialcells described in various papers are rather discordant butall converge in the indication that endothelial behaviouris significantly altered (Table 3) Briefly from studies ondifferent types of ECs exposed to simulated microgravity wecan summarize the following

(i) Impact on cell proliferation and survival all thestudies indicate alterations of cell proliferation OnlyHUVEC and BAEC have been reproducibly foundto proliferate faster in microgravity than controlsMicrovascular EC and other endothelial cells aregrowth inhibited or induced to apoptosis

(ii) Impact on NO synthesis most studies agree on theincreased production of NO through the modulationof NOS isoforms

(iii) Impact on cytoskeleton all the studies describedimportant cytoskeletal remodelling in all the differentEC analyzed

(iv) Impact on gene expression no doubt exists aboutthe profound modifications of gene expression byexposure to simulated or real microgravity

The impact of hypergravity on ECs is less defined Dueto the different experimental approaches adopted on variouscell types the findings are not consistent and deserve furtherconsideration

The effects of gravitational forces on mechanotransduc-tion in ECs responses have been the matter of only a fewinvestigations and remain largely unknown The plausiblemechanosensing targets for gravity changes appear to be thecytoskeletal structure and particularly caveolae [26 29 65]

In conclusion because (i) endothelial cells are crucial forthe integrity of the vessel wall and (ii) vessels are responsiblefor the homeostasis of all the tissues it is pivotal to continuestudies on this topic since the modulation of endothelial


functions can contribute to cardiovascular deconditioningand other disorders observed in space from bone loss tomuscle atrophy However it would be recommended toclearly define the experimental models to use A clear cutdefinition of endothelial cell models to be used and theconditions to model gravity need to be standardized



Acknowledgments

Part of this work has been funded byAgenzia Spaziale Italiana(ASI) and European Spatial Agency (ESA)

References


[2] H F Galley andN RWebster ldquoPhysiology of the endotheliumrdquoBritish Journal of Anaesthesia vol 93 no 1 pp 105ndash113 2004

[3] S P Herbert and D Y R Stainier ldquoMolecular control ofendothelial cell behaviour during blood vessel morphogenesisrdquoNature ReviewsMolecular Cell Biology vol 12 no 9 pp 551ndash5642011

[4] E R Regan and W C Aird ldquoDynamical systems approach toendothelial heterogeneityrdquo Circulation Research vol 111 no 1pp 110ndash130 2012

[5] K E Foreman and J Tang ldquoMolecular mechanisms of replica-tive senescence in endothelial cellsrdquo Experimental Gerontologyvol 38 no 11-12 pp 1251ndash1257 2003

[6] B Buchen M Braun Z Hejnowicz and A Sievers ldquoStatolithspull onmicrofilamentsmdashexperiments undermicrogravityrdquo Pro-toplasma vol 172 no 1 pp 38ndash42 1993

[7] W Briegleb ldquoSome qualitative and quantitative aspects ofthe fast-rotating clinostat as a research toolrdquo ASGSB BulletinPublication of the American Society for Gravitational and SpaceBiology vol 5 no 2 pp 23ndash30 1992

[8] T F B Kraft J JW A van Loon and J Z Kiss ldquoPlastid positionin Arabidopsis columella cells is similar in microgravity and ona random-positioning machinerdquo Planta vol 211 no 3 pp 415ndash422 2000

[9] M A Kacena P Todd L C Gerstenfeld and W J LandisldquoExperiments with osteoblasts cultured under varying orienta-tions with respect to the gravity vectorrdquo Cytotechnology vol 39no 3 pp 147ndash154 2002

[10] Z Barjaktarovic A Nordheim T Lamkemeyer C Fladerer JMadlung and R Hampp ldquoTime-course of changes in amountsof specific proteins upon exposure to hyper-g 2-D clinorota-tion and 3-D random positioning of Arabidopsis cell culturesrdquoJournal of Experimental Botany vol 58 no 15-16 pp 4357ndash43632007

[11] L Morbidelli M Monici N Marziliano et al ldquoSimulatedhypogravity impairs the angiogenic response of endotheliumby up-regulating apoptotic signalsrdquoBiochemical and BiophysicalResearch Communications vol 334 no 2 pp 491ndash499 2005

[12] S Versari A Villa S Bradamante and J A M Maier ldquoAlter-ations of the actin cytoskeleton and increased nitric oxidesynthesis are common features in human primary endothelialcell response to changes in gravityrdquo Biochimica et BiophysicaActa vol 1773 no 11 pp 1645ndash1652 2007


[14] D Grimm P Kossmehl M Shakibaei et al ldquoEffects of sim-ulated microgravity on thyroid carcinoma cellsrdquo Journal ofGravitational Physiology vol 9 no 1 pp P253ndashP256 2002

[15] K Hirasaka T Nikawa L Yuge et al ldquoClinorotation preventsdifferentiation of rat myoblastic L6 cells in association withreduced NF-120581B signalingrdquo Biochimica et Biophysica Acta vol1743 no 1-2 pp 130ndash140 2005

[16] Z Li Y Song M Yuzhong et al ldquoInfluence of simulatedmicrogravity on avian primordial germ cell migration andreproductive capacityrdquo Journal of Experimental Zoology vol292 no 7 pp 672ndash676 2002

[17] D Sarkar T Nagaya K Koga F Kambe Y Nomura andH SeoldquoRotation in clinostat results in apoptosis of osteoblastic ROS1728 cellsrdquo Journal of Gravitational Physiology vol 7 no 2 ppP71ndashP72 2000


[19] C CWoods K E Banks R Gruener andDDeLuca ldquoLoss of Tcell precursors after spaceflight and exposure to vector-averagedgravityrdquoThe FASEB Journal vol 17 no 11 pp 1526ndash1528 2003


[21] S Versari G Longinotti L Barenghi J A M Maier andS Bradamante ldquoThe challenging environment on board theInternational Space Station affects endothelial cell functionby triggering oxidative stress through thioredoxin interactingprotein overexpression the ESA-SPHINX experimentrdquo FASEBJournal vol 27 no 11 pp 4466ndash4475 2013

[22] S I Carlsson M T Bertilaccio I Ascari S Bradamante andJ A Maier ldquoModulation of human endothelial cell behaviourin simulated microgravityrdquo Journal of Gravitational Physiologyvol 9 no 1 pp P273ndashP274 2002

[23] S Cotrupi D Ranzani and J A M Maier ldquoImpact of modeledmicrogravity on microvascular endothelial cellsrdquo Biochimica etBiophysica ActamdashMolecular Cell Research vol 1746 no 2 pp163ndash168 2005


[25] M J A Moes J C Gielen R-J Bleichrodt J J W A VanLoon P C M Christianen and J Boonstra ldquoSimulation ofmicrogravity by magnetic levitation and random positioningeffect on human A431 Cell morphologyrdquo Microgravity Scienceand Technology vol 23 no 2 pp 249ndash261 2011

[26] E Spisni M Toni A Strillacci et al ldquoCaveolae and caveolaeconstituents in mechanosensing effect of modeled micrograv-ity on cultured human endothelial cellsrdquo Cell Biochemistry andBiophysics vol 46 no 2 pp 155ndash164 2006

[27] M Monici N Marziliano V Basile et al ldquoHypergravity affectsmorphology and function in microvascular endothelial cellsrdquo


Microgravity Science and Technology vol 18 no 3-4 pp 234ndash238 2006

[28] S I M Carlsson M T S Bertilaccio E Ballabio and J AMMaier ldquoEndothelial stress by gravitational unloading effectson cell growth and cytoskeletal organizationrdquo Biochimica etBiophysica Acta Molecular Cell Research vol 1642 no 3 pp173ndash179 2003

[29] S M Grenon M Jeanne J Aguado-Zuniga M S Conteand M Hughes-Fulford ldquoEffects of gravitational mechanicalunloading in endothelial cells association between caveolinsinflammation and adhesion moleculesrdquo Scientific reports vol 3p 1494 2013




[33] C Griffoni S di Molfetta L Fantozzi et al ldquoModificationof proteins secreted by endothelial cells during modeled lowgravity exposurerdquo Journal of Cellular Biochemistry vol 112 no1 pp 265ndash272 2011

[34] Y-C Wang S Zhang T-Y Du B Wang and X-Q SunldquoClinorotation upregulates inducible nitric oxide synthase byinhibiting AP-1 activation in human umbilical vein endothelialcellsrdquo Journal of Cellular Biochemistry vol 107 no 2 pp 357ndash363 2009

[35] G L Sanford D Ellerson C Melhado-Gardner A E Sroufeand S Harris-Hooker ldquoThree-dimensional growth of endothe-lial cells in the microgravity-based rotating wall vessel bioreac-torrdquo In Vitro Cellular and Developmental Biology-Animal vol38 no 9 pp 493ndash504 2002

[36] M Monici F Cialdai G Romano et al ldquoAn in vitro studyon tissue repair impact of unloading on cells involved in theremodelling phaserdquo Microgravity Science and Technology vol23 no 4 pp 391ndash401 2011


[38] X Ma M Wehland H Schulz et al ldquoGenomic approach toidentify factors that drive the formation of three-dimensionalstructures by EAhy926 endothelial cellsrdquo PLoS ONE vol 8 no5 Article ID e64402 2013



[41] J Grosse M Wehland J Pietsch et al ldquoShort-term weight-lessness produced by parabolic flight maneuvers altered geneexpression patterns in human endothelial cellsrdquo FASEB Journalvol 26 no 2 pp 639ndash655 2012





[46] M Mariotti and J A M Maier ldquoGravitational unloadinginduces an anti-angiogenic phenotype in humanmicrovascularendothelial cellsrdquo Journal of Cellular Biochemistry vol 104 no1 pp 129ndash135 2008


[48] S Cotrupi and J A M Maier ldquoIs HSP70 upregulation crucialfor cellular proliferative response in simulated microgravityrdquoJournal of Gravitational Physiology vol 11 no 2 pp P173ndash1762004

[49] L Buravkova Y Romanov M Rykova O Grigorieva andN Merzlikina ldquoCell-to-cell interactions in changed gravityground-based and flight experimentsrdquo Acta Astronautica vol57 no 2ndash8 pp 67ndash74 2005



[52] V A Convertino ldquoStatus of cardiovascular issues related tospace flight implications for future research directionsrdquo Respi-ratory Physiology and Neurobiology vol 169 supplement 1 ppS34ndashS37 2009

[53] B J Yates and I A Kerman ldquoPost-spaceflight orthostatic intol-erance possible relationship to microgravity-induced plasticityin the vestibular systemrdquoBrain Research Reviews vol 28 no 1-2pp 73ndash82 1998


[55] C J S Edgell C C McDonald and J B Graham ldquoPermanentcell line expressing human factor VIII-related antigen estab-lished by hybridizationrdquo Proceedings of the National Academyof Sciences of the United States of America vol 80 no 12 pp3734ndash3737 1983

[56] M Boerma G R Burton J Wang L M Fink R E McGeheeJr and M Hauer-Jensen ldquoComparative expression profiling inprimary and immortalized endothelial cells changes in geneexpression in response to hydroxy methylglutaryl-coenzyme Areductase inhibitionrdquo Blood Coagulation and Fibrinolysis vol17 no 3 pp 173ndash180 2006


[57] H F Galley M G Blaylock A M Dubbels and N R WebsterldquoVariability in E-selectin expression mRNA levels and sE-selectin release between endothelial cell lines and primaryendothelial cellsrdquo Cell Biology International vol 24 no 2 pp91ndash99 2000

[58] M Y Kapitonova S L Kuznetsov G R A Froemming etal ldquoEffects of space mission factors on the morphology andfunction of endothelial cellsrdquo Bulletin of Experimental Biologyand Medicine vol 154 no 6 pp 796ndash801 2013

[59] J M Davidson A M Aquino S C Woodward and W WWilfinger ldquoSustained microgravity reduces intrinsic woundhealing and growth factor responses in the ratrdquo FASEB Journalvol 13 no 2 pp 325ndash329 1999

[60] M E Kirchen K M OrsquoConnor H E Gruber et al ldquoEffects ofmicrogravity on bone healing in a rat fibular osteotomymodelrdquoClinicalOrthopaedics andRelatedResearch vol 318 pp 231ndash2421995

[61] D Ingber ldquoHow cells (might) sense microgravityrdquo The FASEBJournal vol 13 no 8 pp S3ndashS15 1999

[62] M Hughes-Fulford and J Boonstra ldquoCell mechanotransduc-tion cytoskeleton and related signalling pathwaysrdquo in CellMechanochemistry Biological Systems and Factors InducingMechanical Stress such as Light Pressure andGravityMMoniciand J W A van Loon Eds pp 75ndash95 Transworld ResearchNetwork Trivandrum India 2010


[64] R G Bacabac D Mizuno and G H Koenderink ldquoMechanicalproperties of living cells on mechanosensing and micrograv-ityrdquo in Cell Mechanochemistry Biological Systems and FactorsInducing Mechanical Stress Such as Light Pressure and GravityM Monici and J W A van Loon Eds pp 23ndash54 TransworldResearch Network Trivandrum India 2010

[65] E Spisni M C Blanco C Griffoni et al ldquoMechanosensingrole of caveolae and caveolar constituents in human endothelialcellsrdquo Journal of Cellular Physiology vol 197 no 2 pp 198ndash2042003

[66] W K Sumanasekera L Zhao M Ivanova et al ldquoEffectof estradiol and dihydrotestosterone on hypergravity-inducedMAPK signaling and occludin expression in human umbilicalvein endothelial cellsrdquo Cell and Tissue Research vol 324 no 2pp 243ndash253 2006

[67] W K Sumanasekera G U Sumanasekera K A Mattingly SM Dougherty R S Keynton and C M Klinge ldquoEstradioland dihydrotestosterone regulate endothelial cell barrier func-tion after hypergravity-induced alterations in MAPK activityrdquoAmerican Journal of Physiology Cell Physiology vol 293 no 2pp C566ndashC573 2007

[68] L Morbidelli N Marziliano V Basile et al ldquoEffect of hyper-gravity on endothelial cell function and gene expressionrdquoMicrogravity Science andTechnology vol 21 no 1-2 pp 135ndash1402009

[69] TKoyamaCKimuraMHayashiMWatanabe Y KarashimaandMOike ldquoHypergravity induces ATP release and actin reor-ganization via tyrosine phosphorylation and RhoA activationin bovine endothelial cellsrdquo Pflugers Archiv European Journal ofPhysiology vol 457 no 4 pp 711ndash719 2009

Research ArticleA Functional Interplay between 5-Lipoxygenase and120583-Calpain Affects Survival and Cytokine Profile of HumanJurkat T Lymphocyte Exposed to Simulated Microgravity

Valeria Gasperi1 Cinzia Rapino23 Natalia Battista45 Monica Bari1

Nicolina Mastrangelo1 Silvia Angeletti6 Enrico Dainese45 and Mauro Maccarrone56

1 Department of Experimental Medicine amp Surgery Tor Vergata University of Rome Via Montpellier 1 00133 Rome Italy2 Faculty of Veterinary Medicine University of Teramo Piazza A Moro 45 64100 Teramo Italy3 StemTeCh Group Via Colle dellrsquoAra 66100 Chieti Italy4 Faculty of Bioscience Technology for Food Agriculture and Environment University of Teramo Piazza A Moro 4564100 Teramo Italy

5 European Center for Brain Research (CERC) IRCCS Santa Lucia Foundation Via del Fosso di Fiorano 64-65 00143 Rome Italy6Center of Integrated Research Campus Bio-Medico University of Rome Via Alvaro del Portillo 21 00128 Rome Italy

Correspondence should be addressed to Valeria Gasperi valeriagasperiuniroma2it Enrico Dainese edaineseuniteitand Mauro Maccarrone mmaccarroneunicampusit

Received 15 May 2014 Accepted 18 August 2014 Published 16 September 2014


Copyright copy 2014 Valeria Gasperi et alThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

A growing body of evidence strongly indicates that both simulated and authentic weightlessness exert a broad range of effects onmammalian tissues and cells including impairment of immune cell function and increased apoptotic deathWe previously reportedthat microgravity-dependent activation of 5-lipoxygenase (5-LOX) might play a central role in the initiation of apoptosis in humanT lymphocytes suggesting that the upregulation of this enzyme might be (at least in part) responsible for immunodepressionobserved in astronauts during space flights Herein we supplement novel information about themolecular mechanisms underlyingmicrogravity-triggered apoptotic cell death and immune system deregulation demonstrating that under simulated microgravityhuman Jurkat T cells increase the content of cytosolic DNA fragments and cytochrome c (typical hallmarks of apoptosis) andhave an upregulated expression and activity of 120583-calpain These events were paralleled by the unbalance of interleukin- (IL-) 2and interferon- (INF-) 120574 anti- and proapoptotic cytokines respectively that seemed to be dependent on the functional interplaybetween 5-LOX and 120583-calpain Indeed we report unprecedented evidence that 5-LOX inhibition reduced apoptotic death restoredthe initial IL-2INF-120574 ratio and more importantly reverted 120583-calpain activation induced by simulated microgravity

1 Introduction

Several studies have shown that authentic space condi-tions markedly alter physiological processes thus leadingto cardiovascular changes [1] loss of bone density [2 3]muscle atrophy [2 4] and immunodepression [5 6] Todate it is well established that cells of the immune systemare severely affected by microgravity conditions [5ndash8] Inparticular alterations observed in astronauts and rodentsflown in space included altered distribution and function ofcirculating leukocytes [9ndash11] lymphocytopenia [12ndash14] and

impaired T cell activation [9 14ndash16] In addition several invivo and in vitro studies reported a weightlessness-dependentalteration of cytokine secretion from T-helper 1 (Th1) and T-helper 2 (Th2) cells that in turn results in a deregulation ofcell-to-cell crosstalk as well as of inflammatory responses [9ndash11 17]

It has been reported that several proinflammatory Th1cytokines including interferon- (INF-) 120574 tumor necro-sis factor- (TNF-) 120573 and interleukin- (IL-) 2 and anti-inflammatory Th2 cytokines like IL-4 and IL-10 as well asleukaemia inhibitory factor (LIF) are related to programmed



cell death (PCD) These glycoproteins indeed are able toinduce or protect cells from apoptosis [18ndash23] so that analternative classification distinguishes them as anti-(LIF IL-2IL-4 IL-10) or proapoptotic (INF-120574) substances A hot topic isthe study of the effect of microgravity (be it real or simulated)on apoptosis of different mammalian cell types includingcerebral vascular smooth muscle [24] thyroid cancer [25]endothelial cells [26] cultured glial cells [27] spermatozoa[28] B lymphocytes [29] and T cells [6 30] In particular5-lipoxygenase (5-LOX) has been proposed as a ldquogravityresponderrdquo which executes the apoptotic events induced bymicrogravity in human lymphocytes [6 30]

Evidence is accumulating that the execution of PCDis finely regulated by a distinct set of signal transductionpathways and catabolic mechanisms (eg mitochondriallysosomal and nuclear alterations lipid modifications andcytosolic calcium accumulation) and recent data providedfirst hints that lipid hydroperoxides impact on PCD [31]Indeed LOX-catalyzed lipid peroxidation has been reportedto be a specific downstream event that triggers apoptosis-inducing factor- (AIF) mediated PCD in primary neuronsin culture and in mice [31] In the same context calpainscleave multiple substrates potentially involved in PCD andincluding cyclin-dependent kinase-5 [32] plasmamembraneCa2+ ATPase isoform-1 [33] and calcineurin [34] Also AIFis a calpain substrate implicated in neuronal death becauseits proteolysis activates PCD through a translocation of AIFitself from the mitochondria to the nucleus [35 36]

Against this background the present study aimed atbetter defining the influence of the space environment onsurvival and cytokine profile of human lymphocytes in orderto identify a possible link between these events In thiscontext we report an unprecedented functional interplaybetween 5-LOX and 120583-calpain in modulating PCD inducedby simulated microgravity


21 Reagents Chemicals were of the purest analytical gradeHuman recombinants IL-2 IL-4 IL-6 IL-10 INF-120574 and LIFcalpain substrate [N-Suc-Leu-Tyr-AMC (7-amido-4-methyl-coumarin)] AA861 (specific inhibitor of 5-LOX) and E64D(specific inhibitor of calpain) were purchased from SigmaChemical Co (St Louis MO USA) Mouse anti-cytochromec antibody was from Cell Signalling Technology Inc (Dan-vers MA USA) mouse anti-calpain-1 was from Calbiochem(Merck Darmstadt Germany) Rabbit anti-LIF anti-IL-2anti-IL-4 anti-IL-6 anti-IL-10 anti-INF-120574 secondary anti-bodies conjugated to horseradish peroxidase (HRP) andenhanced chemiluminescence (ECL) kit were from SantaCruz Biotechnology Inc (Santa Cruz CA USA) Goat anti-rabbit conjugated to alkaline phosphatase (GAR-AP) wasfrom Bio-Rad (Hercules CA USA)

22 Simulated Microgravity Cell Cultures To simulate spaceconditions the rotary cell culture system (RCCS) devel-oped by the National Aeronautics and Space Administration(Washington DC USA) and manufactured by Synthecon

(Houston TX USA) was used Human Jurkat T cells (CloneE6-1) (ATCC Manassas VA USA) were grown in RPMI1640 medium supplemented with 2mM glutamine 25mMsodium pyruvate 100UmL penicillin 100 120583gmL strepto-mycin and 10 heat-inactivated foetal bovine serum Cellswere placed in completely filled 50mL vessels to avoid thepresence of air bubbles that could lead to shear force damageof cells on the RCCS Vessels were rotated at a speed of72 rpm (simulated microgravity and referred to as sim-120583g)as reported [30 37] or cultured at ground gravity (1 g) ascontrols Incubation of 1 g and sim-120583g cells with differentcompounds was performed at 37∘C in an atmosphere of 5CO2 at the indicated concentrations and for the indicated

periods of time

23 Evaluation of PCD PCD was estimated by the cell-deathdetection enzyme-linked immunosorbent assay (ELISA) kit(Boehringer Mannheim Germany) based on evaluationof histone-associated DNA fragments in the cytoplasm aspreviously reported [30]

Cytochrome c release from mitochondria was analyzedas reported [38] Briefly cells were lysed in HB buffer (5mMTris-HCl pH 74 10mM KCl 1mMMgCl

2 and 1mM DTT)

containing protease inhibitor cocktail and centrifuged at1000 timesg for 10min to completely remove nuclei and wholecells The resulting supernatant was centrifuged at 3000 timesgfor 10min then the pellet was saved as membrane-boundorganellar fraction enriched with mitochondria while thesupernatant after centrifugation at 100000 timesg for 40minwas collected as cytosolic fraction These two fractions wereanalyzed for cytochrome c localization by means of ELISAmitochondrial and cytosolic proteins (20 120583gwell) were incu-bated with anti-cytochrome c antibody (diluted 1 500) andafter incubation with a GAR-AP (diluted 1 2000) colourdevelopment of the alkaline phosphatase reaction was mea-sured at 405 nm (A

405 nm) by using 119901-nitrophenyl phosphateas substrate

24 Analysis of 120583-Calpain Activity and Expression Detectionof 120583-calpain mRNA was performed by quantitative reversetranscriptase-polymerase chain reaction (q-RT-PCR) as pre-viously reported [6] Briefly total RNA was extracted fromJurkat cells using the RNeasy extraction kit (Qiagen CrawleyUK) following the manufacturerrsquos instructions RT-PCRreactions were performed using the RT-PCR SuperScript IIIPlatinum Two-Step qRT-PCR Kit (Invitrogen Carlsbad CAUSA) One 120583g total RNA was used to synthesize cDNA with10U120583L SuperScript III reverse transcriptase in the presenceof 2U120583L RNaseOUT 125 120583Moligo (dT) 125 ng120583L randomhexamers 5mMMgCl

2 05mM dNTP mix and DEPC-

treatedwaterThe reactionwas performedusing the followingRT-PCR program 25∘C for 10min 42∘C for 50min 85∘C for5min and then after addition of 01 U120583L of E coli RNase Hthe product was incubated at 37∘C for 20min For expressionstudies target transcripts were amplified in ABI PRISM 7700sequence detector system (Applied Biosystems Foster CityCA USA) Thermal cycling involved 40 cycles of 95∘C for15 sec and 60∘C for 30 sec after initial denaturation for 10min


at 95∘C TaqMan MGB probe was synthesized by AppliedBiosystems (Foster City CA USA) The probe was labelledwith the fluorescent dye 6-carboxyfluorescein at the 51015840 endand a dark quencher at the 31015840 end (Applied Biosystems)Fluorescence was measured after each cycle of PCR andto confirm the quality of isolated RNA and to standardizethe amount of RNA applied glyceraldehyde-3-phosphatedehydrogenase (GAPDH) was used as endogenous controlwith FAMTM dye label and MGB Real-time PCR mixturescontained template cDNA 20x PrimerProbe Mix TaqManMGB Probe with FAMTM dye label no primer limitationMinorGroove Binder andNonfluorescentQuencher Univer-sal PCRMasterMix no AmpErase UNGApplied Biosystems(Foster City CA USA) in a total volume of 25120583L in a 96-wellplate Relative 120583-calpain expression levels were measured byΔΔCT method (PE-Applied Biosystems Sequence DetectorUser Bulletin)

Calpain protein expression was evaluated by Westernblot analysis Briefly cell lysates (20120583gwell) were sub-jected to SDS-PAGE electroblotted onto PVDF membranesincubated with mouse anti-120583-calpain antibody (1 4000)which detects both the full-length (large subunit) and theautoproteolytically cleaved forms of 120583-calpain and detectedwith ECL Calpain quantification was also evaluated throughELISA method by incubating protein lysates (20 120583gwell)with mouse anti-120583-calpain-1 (1 2000) as primary anti-body and HRP-conjugated antibody (1 5000) as secondaryantibody The HRP enzymatic activity was determinedby adding 100 120583Lwell of tetramethylbenzidine containing0002 H

2O2 and the absorbance was read on a microplate

reader (ELISA Ascent Software per Multiskan) at 450 nmAbsorbance values of the samples were within the linearityrange of the ELISA test assessed by calibration curveswith known amounts of 120583-calpain (in the range of 75ndash600 ngwell)

The enzymatic activity of 120583-calpain was measured asreported [39] Briefly cell lysates (40 120583gtest) were incubatedwith 150120583M calpain substrate (N-Suc-Leu-Tyr-AMC) in10mMHepes pH 74 1TritonX-100 and 100 120583MCaCl

2 for

2 hours at 37∘C After incubation hydrolyzed AMC groupswere measured on a fluorimeter LS50B (Perkin-Elmer LifeSciences Inc Boston MA USA) with an excitation filter of380 nm and emission filter of 460 nm

25 5-LOX Activity The activity of 5-LOX (arachido-nateoxygen 5-oxidoreductase EC 1131134) was deter-mined as previously reported [6] Briefly the end prod-uct leukotriene (LT) B

4was extracted from Jurkat cells

(5 times 106 celltest) and quantified at 405 nm by using theLeukotriene B

4EIA Kit (Cayman Chemical Company Ann

Arbor MA USA) and calibration curves drawn according tothe customerrsquos instructions

26 Cytokine Profile Analysis Jurkat cells harvested after 48hours were centrifuged at 200 timesg for 10min to collect cellsand culture medium Cells were lysed in 50mM Tris-HCl(pH 74) containing protease inhibitors and cytokine contentwas quantified by coating proteins (20120583gwell) from whole

lysates overnight in a 96-well ELISA microplate as reported[40] Rabbit anti-LIF anti-IL-2 anti-IL-4 anti-IL-6 anti-IL-10 and anti-INF120574 (diluted 1 500) were used as primaryantibodies GAR-AP (diluted 1 2000) was used as secondaryantibody and absorbance values were read at 405 nm Releaseof LIF and other cytokines from Jurkat cells into the mediumwas quantified through Quantikine Immunoassay kit (RampDSystem Minneapolis MN USA) and a specific MultiproteinProfiling ELISA Kit (SuperArray Bioscience Co Germany)respectively according to the manufacturerrsquos instructions Tothis aim 50120583L of culture medium was used and the contentof each protein was evaluated by comparing A

405 nm values tothose of antigen standard curves (positive controls)

27 Statistical Analysis All values were expressed as meansplusmn SEM of at least three independent experiments Studentrsquosunpaired 119905-test or one-way ANOVA (followed by Bonferronipost hoc analysis) was used to compare quantitative data withnormal distributions and equal varianceThe statistical InStat3 program (GraphPAD Software for Science San DiegoCalifornia) was used and a value of 119875 lt 005 was consideredstatistically significant

3 Results

31 Prolonged Exposure to Simulated Microgravity InducesApoptosis in Human Jurkat T Cells Jurkat T cells wereexposed to simulatedmicrogravity for different times (from 0to 48 hours) and the hallmarks of apoptosis DNA fragmenta-tion and cytochrome c release were analyzed In agreementwith previously reported data [30] RCCS treatment ledto a time-dependent increase of cytosolic DNA fragmentsthat were undetectable after a brief exposure (4 hours) tosimulated microgravity increased after 24 hours (sim2-foldover 1 g cells) and reached a maximum level of sim3-foldover controls 24 hours later (Table 1) Then the subcellularlocalization of cytochrome c upon simulated microgravitywas checked Jurkat cells exposed to weightlessness showed aloss of mitochondrial cytochrome c and a parallel increase inthe cytosolic content with a time-dependence comparable tothat observed for DNA fragmentation (Table 1) ConverselyJurkat cells incubated at 1 g under the same experimental con-ditions did not show significant signs of PCD (Table 1) SinceRCCS treatment for 48 hours yielded a significant increase inPCD we chose to perform all subsequent experiments usingthis time point

32 Prolonged Exposure to Simulated Microgravity Upreg-ulates 120583-Calpain Expression and Activity in Human JurkatT Cells We have previously reported that after 48 hoursof exposure to authentic microgravity human lymphocytesshow increased mRNA levels of 120583-calpain [6] a Ca2+-dependent intracellular cysteine protease that is implicated indifferent physiological functions including cell growth andapoptosis [41] Therefore once established that under ourexperimental conditions Jurkat cells underwent apoptosis wecheckedwhether RCCS treatmentmight engage 120583-calpain Inagreement with our previous data [6] RT-qPCR experiments


Table 1 Time-dependent effect of simulated microgravity on apoptotic markers in Jurkat T cells exposed to simulated microgravity (sim-120583g)or kept at normal gravity (1g)

Parameter 1g 4-hour sim-120583g 24-hour sim-120583g 48-hour sim-120583gDNA fragmentation 100 plusmn 8 107 plusmn 8 270 plusmn18lowast 342 plusmn21lowastlowast

Cytochrome c release (cytosolmitochondria ratio) 100 plusmn 9 111 plusmn 7 347 plusmn41lowast 450 plusmn47lowastlowast

Results are expressed as percentage of 1g cells set to 100 For DNA fragmentation 100 = 0300 plusmn 0030A405 nm for cytochrome c release 100 = 0074 plusmn 0005lowastdenotes 119875 lt 001 versus 1g cells lowastlowastdenotes 119875 lt 0001 versus 1g cells

demonstrated a significant increase of 120583-calpain mRNA insim-120583g Jurkat cells (sim2-fold over 1 g cells) (Figure 1(a))Interestingly upregulation of capn 1 gene which encodes 120583-calpain was paralleled by increased protein content (Figures1(b) and 1(c)) Western blot analysis indeed showed that 48hours of RCCS treatment dramatically increased 120583-calpainprotein levels in particular larger amounts of autocleavedand active fragment of 120583-calpain (sim75 kDa) [42] were foundin sim-120583g whereas no active enzyme was observed in1 g cells (Figure 1(b)) Such a result was further corrobo-rated by ELISA revealing that RCCS almost doubled 120583-calpain protein content after 48 hours (Figure 1(c)) We nextdetermined whether increased mRNA and protein contentmight result in increased enzyme activity By analysing thecleavage of a fluorogenic 120583-calpain substrate we observedan enhanced protease activity in sim-120583g T cells (sim2-foldover 1 g cells) (Figure 1(d)) Specific proteolytic activity ofcalpain was confirmed by the addition of 5 120583M calpastatin(Figure 1(d)) the natural calpain inhibitor [43] Since calpainactivation seemed to be implicated in DNA fragmentation[44 45] we analyzed the effect of E64D a cell permeableand selective inhibitor of the same protease [46] on sim-ulated microgravity-induced PCD As shown in Figure 1(e)inhibition of calpain activity significantly lowered internucle-osomal DNA fragmentation thus preventing weightlessness-induced cell death of T cells

33 Prolonged Exposure to Simulated Microgravity Affects theBalance between Proapoptotic and Antiapoptotic Cytokinesin Jurkat Cells Then we characterized the cytokine profilein Jurkat T cells exposed to simulated microgravity Asdemonstrated by ELISA assay 48 hours of RCCS treatmentsignificantly reduced the synthesis and release of antiapop-totic cytokines like LIF IL-4 and IL-2 while increasingprotein levels of the proapoptotic cytokine INF-120574 (Figure 2)Instead no change in IL-6 and IL-10 content was observedupon simulated microgravity treatment (Figure 2)

Next we went further by investigating whetherRCCS-induced PCD might be related to the unbalancebetween proapoptotic and antiapoptotic cytokines Tothis aim we analyzed apoptosis in Jurkat cells culturedunder simulated microgravity for 48 hours in the presenceof the cytokines that changed upon RCCS exposureNeither LIF nor IL-4 (both at 10 ngmL) reduced cytosolicDNA fragments (Figure 3(a)) and cytochrome c content(Figure 3(b)) on the other hand 10 ngmL IL-2 wasable to protect Jurkat cells from simulated microgravity-triggered cell death since it significantly reduced both DNAfragmentation and cytochrome c release (Figures 3(a) and

Table 2 Time-dependent effect of simulated microgravity on 5-LOX activity in Jurkat cells

Sample LOX activity ( of 1g cells)a

1g cells 100 plusmn 114-hour sim-120583g cells 213 plusmn18lowast

24-hour sim-120583g cells 249 plusmn23lowast

48-hour sim-120583g cells 235 plusmn21lowasta100 of 5-LOX activity = 89 plusmn 7 pg of LTB41 times 10

6 cells lowastdenotes 119875 lt 001versus 1g cells

3(b)) To validate our hypothesis we also analyzed theeffect of INF-120574 (10 ngmL) In agreement with the previousdata (Figure 2) the latter cytokine drastically enhancedRCCS-induced PCD of Jurkat cells (sim 45- and 25-fold over1 g cells and sim-120583g cells resp) (Figures 3(a) and 3(b))

To gain further insights on the evaluation of a possiblerelationship between altered IL-2INF-120574 content and calpainactivation we measured the activity of the latter enzymein the presence of these two cytokines Interestingly IL-2reduced calpain activation due to simulated microgravitywhile INF-120574 did not significantly affect enzyme activity(Figure 3(c))

34 Effect of Inhibition of 120583-Calpain and 5-LOX on Apoptosisand Cytokine Release Since we observed that simulatedmicrogravity triggers apoptosis by altering the content of IL-2and INF-120574 we asked whether such an event might engage 5-LOX which has been proposed as a ldquogravity responderrdquo [30]First we analyzed 5-LOX activity by quantifying the contentof its LTB

4product upon RCCS exposure In agreement

with previous data we found an early increase of 5-LOXactivity (sim2 fold over 1 g cells) with values that remainedunchanged over the whole time period tested (Table 2)Hence we subjected Jurkat cells to simulated microgravityin the presence of 10 120583M AA861 a specific 5-LOX inhibitor[47] As shown inTable 3 we observed that 5-LOX inhibitionreduced DNA fragmentation and cytochrome c release andreverted calpain activation More interestingly it was able torestore the balance between IL-2and INF-120574 that was alteredbyRCCS treatmentThese data seem to suggest that increased5-LOX activity might be (at least in part) responsible foraltered cytokine levels

4 Discussion

The effects of LTs on the secretion of cytokines havebeen reported both in vitro and in vivo [46] Here wedemonstrated that increased LTB

4synthesis upon simulated


0

50

100

150

200

250lowast

1g

120583-c

alpa

in m

RNA

leve

l (

of1

gJu

rkat

cells

)

sim-120583g

(a)

Full length

Cleaved form

1g

80kDa

75kDa

sim-120583g

(b)

0

50

100

150

200

250

lowast

1gng o

f120583-c

alpa

in120583

g pr

otei

n (

of1

gJu

rkat

cells

)

sim-120583g

(c)

Calpastatin0

20

40

60

80

100

120

140

160

180

200

sect

lowast

1g

120583-c

alpa

in ac

tivity

( o

f1g

Jurk

at ce

lls)

mdashsim-120583g

(d)

0

50

100

150

200

250

300

350

400

E64D

sect

lowast

1g mdashDN

A fr

agm

enta

tion

( o

f1g

Jurk

at ce

lls)

sim-120583g

(e)

Figure 1 Effect of simulated microgravity on 120583-calpain expression and activity in Jurkat T cells (a) RT-qPCR analysis of 120583-calpain geneexpression in Jurkat cells exposed to simulated microgravity (sim-120583g) at 37∘C for 48 hours Gene levels were normalized to the housekeepingGAPDH and expressed as percentage of 1 g cells considered as control set to 100 (b)Western blot analysis of protein expression in Jurkat cellstreated as in (a) (c) ELISA analysis of 120583-calpain protein content in Jurkat cells treated as in (a) Results are expressed as percentage of 1 g cellsconsidered as control set to 100 (= 948 plusmn 050 ngper 120583g protein) (d) 120583-calpain activity analysis in Jurkat cells treated as in (a) in absence (minus)or in presence of 5120583Mof calpastatin Results are expressed as percentage of 1 g cells considered as control set to 100 (= 6626 plusmn 365 pmolminper mg protein) (e) DNA fragmentation in Jurkat cells exposed to simulatedmicrogravity for 48 hours in absence (minus) or in presence of 10120583ME64D Values are expressed as percentage of 1 g cells considered as control lowastdenotes 119875 lt 0001 versus 1 g cells sectdenotes 119875 lt 005 versussim-120583g cells


0

50

100

150

200

250

SynthesisRelease

( o

f1g

Jurk

at T

cells

)

LIF IL-2 IL-4 IL-6 IL-10 INF-120574

lowast

lowastlowast lowast

lowast

1g

sim-120583g

Figure 2 Effect of simulated microgravity on cytokine profile ofJurkat T cells Cells were exposed (sim-120583g) or not exposed (1 g) tosimulated microgravity at 37∘C for 48 hours and cytokine synthesis(gray bars) and release (black bars) were analyzed as reported inSection 26 Results are reported as percentage of 1 g cells set to100 For synthesis 100 of IL-2 = 027 plusmn 001 A

405 nm 100 of IL-4 = 034 plusmn 002 A

405 nm 100 of LIF = 022 plusmn 002 A405 nm 100

of IL-6 = 018 plusmn 001 A405 nm 100 of IL-10 = 034 plusmn 002 A

405 nm100 of INF-120574 025 plusmn 002 A

405 nm For release 100 of LIF = 042plusmn 003 Abs

405 nm 100 of IL-2 = 66 plusmn 05 pgmL 100 of IL-4 = 12plusmn 01 pgmL 100 of IL-6 = 20 plusmn 2 pgmL 100 of IL-10 = 25 plusmn03 pgmL 100 INF-120574 = 122 plusmn 01 pgmL lowastdenotes 119875 lt 005 versus1 g cells denotes 119875 lt 001 versus 1 g cells

Table 3 Effect of 5-LOX inhibition on Jurkat T cells exposed for 48hours to simulated microgravity (sim-120583g) or kept at normal gravity(1g)

Parameter 1g sim-120583g sim-120583g +10 120583MAA861

DNA fragmentation 100 plusmn 10 342 plusmn21lowastlowastlowast 250 plusmn11

Cytochrome c release(cytosolmitochondriaratio)

100 plusmn 9 450 plusmn47lowastlowastlowast 230 plusmn24

Calpain activity 100 plusmn 11 177 plusmn9lowastlowastlowast 31 plusmn2

IL-2 protein content 100 plusmn 9 67 plusmn5lowast 93 plusmn8

INF-120574 protein content 100 plusmn 9 179 plusmn15lowastlowast 120 plusmn4

Values are reported as percentage of relative control set to 100 For DNAfragmentation 100 = 030 plusmn 003A405 nm for cytochrome c release 100= 0074 plusmn 0005 for calpain activity 100 = 6626 plusmn 365 pmolmin per mgprotein for IL-2 synthesis 100 = 027 plusmn 001 A405 nm for INF-120574 synthesis100 = 025 plusmn 002 A405 nm

lowastdenotes = 119875 lt 005 versus 1g cells lowastlowastdenotes119875 lt 001 versus 1g cells lowastlowastlowastdenotes 119875 lt 0001 versus 1g cells denotes 119875 lt001 versus sim-120583g cells

microgravity exposure is paralleled by a reduced release ofantiapoptotic cytokines such as LIF IL-4 and IL-2 [19ndash23] as well as by a significant increase of the productionof the proapoptotic cytokine INF-120574 [18 23] These dataare in line with the immunomodulatory role postulatedfor 5-LOX metabolites and especially for LTB

4 Indeed

the latter substance is a powerful chemoattractant forinflammatory cells and induces degranulation superoxide

anion production and adherence of neutrophils to vascularendothelial cells [48] LTB

4has been already demonstrated

to affect the production of several cytokines including IL-1120573 [49 50] IL-2 [51 52] IL-6 [53] INF-120574 [54] IL-4 [55]and IL-10 [56] Moreover LTB

4has been also demonstrated

to modulate the expression of the IL-2 receptor 120573-chain innatural killer cells and in CD8+ lymphocytes [57]

In this context our data add further information onthe mechanism of PCD activation suggesting a crosstalkbetween 5-LOX and 120583-calpain signalling In particular wedemonstrate that exposure of Jurkat T cells to simulatedmicrogravity induced activation of 120583-calpain and 5-LOXOur results suggest that the functional interplay betweenthese two enzymes could be related to the synthesis of aspecific pattern of cytokines In line with this our resultsshow that 5-LOX inhibition (i) reduced DNA fragmentationand cytochrome c release (typical apoptotic markers) (ii)reestablished the initial IL-2INF-120574 ratio and (iii) moreimportantly reverted 120583-calpain activation induced by simu-lated microgravity (Table 3) Furthermore we showed thattreatment of Jurkat T cell with IL-2 whose levels are down-regulated upon simulated microgravity exposure (Figure 2)significantly reduced 120583-calpain activation upon RCCS treat-ment Remarkably the latter result is in agreement with thewell-known antiapoptotic effect of IL-2 [21 23] It shouldbe noted that the lack of any increase in 120583-calpain activityin the presence of 5-LOX inhibitors might be suggestivethat additional and as-yet-unknown 5-LOX products areable to directly activate 120583-calpain Thus in addition to aspecific role of distinct cytokines in modulating the crosstalkbetween 5-LOX and 120583-calpain we can speculate that 5-LOX activation could also induce the formation of specificlipid hydroperoxides that could trigger PCD via 120583-calpainactivation In line with the latter hypothesis hydroperoxidesof cardiolipin and phosphatidylserine have been detectedas byproducts upon PCD [58] Consistently it has beendemonstrated that LOX-induced lipid peroxidation triggersAIF-mediated PCD [31] Indeed although a finely regulatedlipid peroxidationmay have beneficial effects for the cells andthe whole organism leading to different physiological roles ofLOXs (such as eicosanoid synthesis cell maturation and lipidmobilization) when the lipid bilayer of biologicalmembranesis oxidized in an uncontrolled manner (as in the case ofexternal stimuli like microgravity) it may lose its barrierfunction and thus harm the integrity of subcellular organellesor of thewhole cell [59] Consistently an overactivated 5-LOXcan open pore-like structures in mitochondrial membranes[60 61] thus forming the basis for a converging role of thisenzyme in the induction of PCD by unrelated stimuli [59]

Overall our results demonstrate that simulatedmicrogravity-dependent increase in 5-LOX activity regulatessurvival and cytokine release of human T lymphocytes byengaging 120583-calpain

5 Conclusions

Our findings seem to add biochemical support to the immun-odepression observed in astronauts exposed to authentic


0

100

200

300

400

500D

NA

frag

men

t con

tent

( o

f1g

Jurk

at T

cells

)

LIF IL-2 IL-4 INF-120574

lowast

1g mdash

sim-120583g

(a)

0100200300400500600700800900

Cyto

chro

me c

(cyt

osol

mito

chon

dria

ratio

)

lowast

LIF IL-2 IL-4 INF-120574mdash

1gsim-120583g

(b)

0

50

100

150

200

120583-c

alpa

in ac

tivity

( o

f1g

Jurk

at ce

lls) lowast

IL-2 INF-120574mdash

1g

sim-120583g

(c)

Figure 3 Cytokine effects on Jurkat cell apoptosis under simulated microgravity conditions Jurkat T cells were exposed (sim-120583g) or notexposed (1 g) to simulated microgravity in absence (minus) or in presence of the indicated cytokines (10 ngmL) and DNA fragmentation (a)cytochrome c release (b) and 120583-calpain activity (c) were evaluated as reported in Section 2 Results are reported as percentage of 1 g cells setto 100 For DNA fragmentation 100 = 0300 plusmn 0030 A

405 nm for cytochrome c release 100 = 0074 plusmn 0005 for 120583-calpain activity 100 =6626 plusmn 365 pmolmin per mg protein lowastdenotes119875 lt 0001 versus1 g cells denotes 119875 lt 005 versus sim-120583g cells denotes 119875 lt 001 versussim-120583g cells

microgravity for long periods of time (eg InternationalSpace Station crewmembers or astronauts travelling toMars)Taking into account that Jurkat E61 cells are somewhatdifferent from normal human T cells [61] nonetheless theyare considered a valid experimental model especially in thelight of their exaggerated signaling making changes mucheasier to detect Therefore only authentic space conditionswill give a conclusive answer onwhether or not the unbalancebetween proapoptotic and antiapoptotic cytokines due toimpaired 5-LOX and 120583-calpain activities can affect immuneresponse helping to design countermeasures against apopto-sis observed in space

Abbreviations

LOX LipoxygenaseAMC 7-Amido-4-methyl-coumarinAIF Apoptosis-inducing factorRCCS Rotary cell culture system

ECL Enhanced chemiluminescenceELISA Enzyme-linked immunosorbent assayGAPDH Glyceraldehyde-3-phosphate

dehydrogenaseGAR-AP Goat anti-rabbit conjugated to alkaline

phosphatase1 g Ground gravityHRP Horseradish peroxidaseINF-120574 Interferon-120574IL-2 Interleukin-2LIF Leukaemia inhibitory factor(LT)B

4 Leukotriene B

4

sim-120583g Simulated microgravityPCD Programmed cell deathq-RT-PCR Quantitative reverse transcriptase-

polymerase chain reactionTh1 T-helper 1Th2 T-helper 2TNF-120573 Tumor necrosis factor-120573



The authors declare that there is no conflict of interests

Acknowledgment

This investigation was supported under contracts from ldquoDis-turbi del Controllo Motorio e Cardiorespiratoriordquo and ldquoFromMolecules to Manrdquo 2006ndash2009 to Mauro Maccarrone

References


[2] L C Shackelford ldquoMusculoskeletal response to space flightrdquoin Principles of Clinical Medicine for Space Flight M R Barrattand S L Pool Eds pp 293ndash306 Springer Science and BusinessMedia New York NY USA 2008


[4] EWang ldquoAge-dependent atrophy andmicrogravity travel whatdo they have in commonrdquoThe FASEB Journal vol 13 no 8 ppS167ndashS174 1999

[5] D Williams A Kuipers C Mukai and R Thirsk ldquoAcclimationduring space flight effects on human physiologyrdquo CanadianMedical Association Journal vol 180 no 13 pp 1317ndash1323 2009

[6] N Battista M A Meloni M Bari et al ldquo5-Lipoxygenase-dependent apoptosis of human lymphocytes in the Interna-tional Space Station data from the ROALD experimentrdquo TheFASEB Journal vol 26 no 5 pp 1791ndash1798 2012



[9] B E Crucian R P Stowe D L Pierson and C FSams ldquoImmune system dysregulation following short- vslong-duration spaceflightrdquo Aviation Space and EnvironmentalMedicine vol 79 no 9 pp 835ndash843 2008

[10] B Crucian R Stowe S Mehta et al ldquoImmune system dysreg-ulation occurs during short duration spaceflight on board thespace shuttlerdquo Journal of Clinical Immunology vol 33 no 2 pp456ndash465 2013

[11] B S Crucian S R Zwart S Mehta et al ldquoPlasma cytokineconcentrations indicate that in vivo hormonal regulation ofimmunity is altered during long-duration spaceflightrdquo Journalof Interferon amp Cytokine Research 2014

[12] S K Chapes S J Simske A D Forsman T A Bateman andR J Zimmerman ldquoEffects of space flight and IGF-1 on immunefunctionrdquo Advances in Space Research vol 23 no 12 pp 1955ndash1964 1999

[13] S K Chapes S J Simske G Sonnenfeld E S Miller andR J Zimmerman ldquoEffects of spaceflight and PEG-IL-2 on ratphysiological and immunological responsesrdquo Journal of AppliedPhysiology vol 86 no 6 pp 2065ndash2076 1999

[14] A T Ichiki L A Gibson T L Jago et al ldquoEffects of spaceflighton rat peripheral blood leukocytes and bonemarrowprogenitorcellsrdquo Journal of Leukocyte Biology vol 60 no 1 pp 37ndash43 1996

[15] Z Allebban A T Ichiki L A Gibson J B Jones C CCongdon and R D Lange ldquoEffects of spaceflight on thenumber of rat peripheral blood leukocytes and lymphocytesubsetsrdquo Journal of Leukocyte Biology vol 55 no 2 pp 209ndash2131994

[16] D S Gridley JM Slater X Luo-Owen et al ldquoSpaceflight effectson T lymphocyte distribution function and gene expressionrdquoJournal of Applied Physiology vol 106 no 1 pp 194ndash202 2009

[17] K Felix KWise SManna et al ldquoAltered cytokine expression intissues of mice subjected to simulated microgravityrdquoMolecularand Cellular Biochemistry vol 266 no 1-2 pp 79ndash85 2004

[18] S P Tu M Quante G Bhagat et al ldquoIFN-120574 inhibits gastriccarcinogenesis by inducing epithelial cell autophagy and T-cellapoptosisrdquo Cancer Research vol 71 no 12 pp 4247ndash4259 2011

[19] D Duval B Reinhardt C Kedinger and H Boeuf ldquoRole ofsuppressors of cytokine signaling (Socs) in leukemia inhibitoryfactor (LIF) -dependent embryonic stem cell survivalrdquo TheFASEB Journal vol 14 no 11 pp 1577ndash1584 2000

[20] H Slaets D Dumont J Vanderlocht et al ldquoLeukemiainhibitory factor induces an antiapoptotic response in oligoden-drocytes throughAkt-phosphorylation andup-regulation of 14-3-3rdquo Proteomics vol 8 no 6 pp 1237ndash1247 2008

[21] L R Devireddy and M R Green ldquoTranscriptional programof apoptosis induction following interleukin 2 deprivationidentification of RC3 a calciumcalmodulin binding protein asa novel proapoptotic factorrdquoMolecular and Cellular Biology vol23 no 13 pp 4532ndash4541 2003

[22] L M Minter and B A Osborne ldquoNotch and the survival ofregulatory T cells location is everythingrdquo Science Signaling vol5 no 234 article pe31 2012

[23] F C H Pinto G B Menezes S A L Moura G D Cassali MM Teixeira and D C Cara ldquoInduction of apoptosis in tumorcells as a mechanism of tumor growth reduction in allergicmicerdquo Pathology Research and Practice vol 205 no 8 pp 559ndash567 2009

[24] M-J Xie Y-G Ma F Gao et al ldquoActivation of BK119862119886

channel isassociated with increased apoptosis of cerebrovascular smoothmuscle cells in simulated microgravity ratsrdquo American Journalof Physiology Cell Physiology vol 298 no 6 pp C1489ndashC15002010




[28] L H Yan Z Hong M G Ying et al ldquoSimulated microgravityconditions and carbon ion irradiation induce spermatogeniccell apoptosis and sperm DNA damagerdquo Biomedical and Envi-ronmental Sciences vol 26 no 9 pp 726ndash734 2013

[29] B Dang Y Yang E Zhang et al ldquoSimulated microgravityincreases heavy ion radiation-induced apoptosis in human Blymphoblastsrdquo Life Sciences vol 97 no 2 pp 123ndash128 2014



[31] A Seiler M Schneider H Forster et al ldquoGlutathione per-oxidase 4 senses and translates oxidative stress into 1215-lipoxygenase dependent- and AIF-mediated cell deathrdquo CellMetabolism vol 8 no 3 pp 237ndash248 2008

[32] M-S Lee Y T Kwon M Li J Peng R M Friedlander andL-H Tsai ldquoNeurotoxicity induces cleavage of p35 to p25 bycalpainrdquo Nature vol 405 no 6784 pp 360ndash364 2000

[33] D Guerini B Pan and E Carafoli ldquoExpression purificationand characterization of isoform 1 of the plasmamembrane Ca2+pump Focus on calpain sensitivityrdquo The Journal of BiologicalChemistry vol 278 no 40 pp 38141ndash38148 2003

[34] H-Y Wu K Tomizawa Y Oda et al ldquoCritical role of calpain-mediated cleavage of calcineurin in excitotoxic neurodegener-ationrdquo The Journal of Biological Chemistry vol 279 no 6 pp4929ndash4940 2004

[35] B M Polster G Basanez A Etxebarria J M Hardwick andD G Nicholls ldquoCalpain I induces cleavage and release ofapoptosis-inducing factor from isolated mitochondriardquo Journalof Biological Chemistry vol 280 no 8 pp 6447ndash6454 2005

[36] S A Susin H K Lorenzo N Zamzami et al ldquoMolecularcharacterization of mitochodrial apoptosis-inducing factorrdquoNature vol 397 no 6718 pp 441ndash446 1999

[37] R Mitteregger G Vogt E Rossmanith and D FalkenhagenldquoRotary cell culture system (RCCS) a new method for cultivat-ing hepatocytes on microcarriersrdquo The International Journal ofArtificial Organs vol 22 no 12 pp 816ndash822 1999

[38] MV Catani V Gasperi D Evangelista A F Agro L Aviglianoand M MacCarrone ldquoAnandamide extends platelets survivalthrough CB1-dependent Akt signalingrdquo Cellular and MolecularLife Sciences vol 67 no 4 pp 601ndash610 2010

[39] K G Daniel J S Anderson Q Zhong A Kazi P Gupta andQ P Dou ldquoAssociation ofmitochondrial calpain activationwithincreased expression and autolysis of calpain small subunit inan early stage of apoptosisrdquo International Journal of MolecularMedicine vol 12 no 2 pp 247ndash252 2003

[40] V Gasperi F Fezza N Pasquariello et al ldquoEndocannabinoidsin adipocytes during differentiation and their role in glucoseuptakerdquo Cellular and Molecular Life Sciences vol 64 no 2 pp219ndash229 2007

[41] P Łopatniuk and J M Witkowski ldquoConventional calpains andprogrammed cell deathrdquo Acta Biochimica Polonica vol 58 no3 pp 287ndash296 2011

[42] H Sorimachi S Ishiura and K Suzuki ldquoStructure and phys-iological function of calpainsrdquo Biochemical Journal vol 328article 3 pp 721ndash732 1997

[43] T Uemori T Shimojo K Asada et al ldquoCharacterization ofa functional domain of human calpastatinrdquo Biochemical andBiophysical Research Communications vol 166 no 3 pp 1485ndash1493 1990

[44] J Takano M Tomioka S Tsubuki et al ldquoCalpain mediatesexcitotoxic DNA fragmentation via mitochondrial pathwaysin adult brains evidence from calpastatin mutant micerdquo TheJournal of Biological Chemistry vol 280 no 16 pp 16175ndash161842005

[45] A Rami R Agarwal G Botez and JWinckler ldquo120583-Calpain acti-vation DNA fragmentation and synergistic effects of caspase

and calpain inhibitors in protecting hippocampal neurons fromischemic damagerdquo Brain Research vol 866 no 1-2 pp 299ndash3122000

[46] Y Yang Z H Liu C F Ware and J D Ashwell ldquoA cysteineprotease inhibitor prevents activation-induced T-cell apoptosisand death of peripheral blood cells from human immunodefi-ciency virus-infected individuals by inhibiting upregulation ofFas ligandrdquo Blood vol 89 no 2 pp 550ndash557 1997

[47] Y Tanihiro Y Chieko O Kenkichi et al ldquo235-Trimethyl-6-(12-hydroxy-510-dodecadiynyl)-l4-benzoquinone (AA861)a selective inhibitor of the 5-lipoxygenase reaction andthe biosynthesis of slow-reacting substance of anaphylaxisrdquoBiochimica et Biophysica Acta Lipids and Lipid Metabolism vol713 no 2 pp 470ndash473 1982

[48] J C Eun E E Moore A Banerjee et al ldquoLeukotriene B4and its metabolites prime the neutrophil oxidase and induceproinflammatory activation of human pulmonary microvascu-lar endothelial cellsrdquo Shock vol 35 no 3 pp 240ndash244 2011

[49] G Bonizzi J Piette M P Merville and V Bours ldquoDistinctsignal transduction pathwaysmediate nuclear factor- 120581B induc-tion by IL-1beta in epithelial and lymphoid cellsrdquo Journal ofImmunology vol 159 no 11 pp 5264ndash5272 1997

[50] J Marcinkiewicz A Grabowska K Bryniarski and B MChain ldquoEnhancement of CD4+ T-cell-dependent interleukin-2 production in vitro by murine alveolar macrophages the roleof leukotriene B4rdquo Immunology vol 91 no 3 pp 369ndash374 1997

[51] M Los H Schenk K Hexel P A Baeuerle W Droge and KSchulze-Osthoff ldquoIL-2 gene expression and NF-120581B activationthrough CD28 requires reactive oxygen production by 5-lipoxygenaserdquoThe EMBO Journal vol 14 no 15 pp 3731ndash37401995

[52] J Dornand C Sekkat J-C Mani and M Gerber ldquoLipoxy-genase inhibitors suppress IL-2 synthesis relationship withrise of [Ca++]i and the events dependent on protein kinase Cactivationrdquo Immunology Letters vol 16 no 2 pp 101ndash106 1987

[53] M A Brach S de Vos C Arnold H-J Gruszlig R Mertelsmannand F Herrmann ldquoLeukotriene B4 transcriptionally activatesinterleukin-6 expression involving NK-NB and NF-IL6rdquo Euro-pean Journal of Immunology vol 22 no 10 pp 2705ndash2711 1992

[54] H M Johnson and B A Torres ldquoLeukotrienes Positive signalsfor regulation of 120574-interferon productionrdquo Journal of Immunol-ogy vol 132 no 1 pp 413ndash416 1984

[55] NDugas B Dugas J-P Kolb K Yamaoka J F Delfraiss andCDamais ldquoRole of leukotriene B4 in the interleukin-4-inducedhuman mononuclear phagocyte activationrdquo Immunology vol88 no 3 pp 384ndash388 1996

[56] S Jozefowski R Biedron M Bobek and J MarcinkiewiczldquoLeukotrienes modulate cytokine release from dendritic cellsrdquoImmunology vol 116 no 4 pp 418ndash428 2005

[57] J Stankova N Gagnon and M Rola-PleszczynskildquoLeukotriene B4 augments interleukin-2 receptor-beta(IL-2R120573) expression and IL-2R120573-mediated cytotoxic responsein human peripheral blood lymphocytesrdquo Immunology vol 76no 2 pp 258ndash263 1992

[58] V E Kagan V A Tyurin J Jiang et al ldquoCytochrome c actsas a cardiolipin oxygenase required for release of proapoptoticfactorsrdquo Nature Chemical Biology vol 1 no 4 pp 223ndash2322005

[59] M Maccarrone G Melino and A Finazzi-Agro ldquoLipoxyge-nases and their involvement in programmed cell deathrdquo CellDeath and Differentiation vol 8 no 8 pp 776ndash784 2001


[60] K van Leyen R M Duvoisin H Engelhardt and M Wied-mann ldquoA function for lipoxygenase in programmed organelledegradationrdquo Nature vol 395 no 6700 pp 392ndash395 1998

[61] R R Bartelt N Cruz-Orcutt M Collins and J C D HoutmanldquoComparison of T cell receptor-induced proximal signaling anddownstream functions in immortalized and primary T cellsrdquoPLoS ONE vol 4 no 5 Article ID e5430 2009

Research ArticleHow Microgravity Changes Galectin-3 in Thyroid Follicles

Elisabetta Albi1 Francesco Curcio2 Andrea Lazzarini12

Alessandro Floridi1 Samuela Cataldi1 Remo Lazzarini1 Elisabetta Loreti3

Ivana Ferri3 and Francesco Saverio Ambesi-Impiombato2

1 Laboratory of Nuclear Lipid BioPathology CRABiON 06100 Perugia Italy2 Department of Medical and Biological Sciences University of Udine 33100 Udine Italy3 Institute of Pathologic Anatomy and Histology University of Perugia 06100 Perugia Italy

Correspondence should be addressed to Elisabetta Albi elisabettaalbiyahoocom

Received 22 April 2014 Revised 7 August 2014 Accepted 28 August 2014 Published 11 September 2014


Copyright copy 2014 Elisabetta Albi et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

After long-term exposure to real microgravity thyroid gland in vivo undergoes specific changes follicles are made up oflarger thyrocytes that produce more cAMP and express more thyrotropin-receptor caveolin-1 and sphingomyelinase andsphingomyelin-synthase parafollicular spaces loseC cells with consequent reduction of calcitonin productionHerewe studied fourimmunohistochemical tumor markers (HBME-1 MIB-1 CK19 and Galectin-3) in thyroid of mice housed in the Mouse DrawerSystem and maintained for 90 days in the International Space Station Results showed that MIB-1 proliferative index and CK19are negative whereas HBME-1 and Galectin-3 are overexpressed The positivity of Galectin-3 deserves attention not only for itsexpression but also and especially for its localization Our results highlighted that in microgravity conditions Galectin-3 leavesthyrocytes and diffuses in colloid It is possible that the gravity force contributes to the maintenance of the distribution of themolecules in both basal membrane side and apical membrane side and that the microgravity facilitates slippage of Galectin-3 incolloid probably due to membrane remodelling-microgravity induced

1 Introduction

Galectins are endogenous lectins which constitute a galacto-side-binding protein family of 15 members [1] All membersshare close sequence homology in their carbohydrate recog-nition domain but exhibit different affinities for differentsaccharide ligands and can be bi- or multivalent in terms oftheir ligand-binding activity in cell surface [2] Eukaryoticcell surfaces are dominated by the glycocalyx asim100 nmwidemacromolecular structure consisting of glycans attached toproteins and lipids and N-glycans appear to be the majorligand for galectins [3] Each member of the galectin familycontains at least one domain of about 130 amino acids thisdomain binds to saccharides and is designated the carbohy-drate recognition domain (CRD) Based on the number andorganization of domains in the polypeptides the galectinshave been classified into subfamilies (a) the prototype groupcontains one domain the CRD (b) the chimera group

contains a proline- (P-) and glycine- (G-) rich domain (alsoabout 130 amino acids) which fused amino terminal to theCRD and (c) the tandem repeat group contains two CRDs[4]

Galectin-3 (Gal-3) the only representative of the chimeragroup was first discovered as an IgE-binding protein andcharacterized as a 32 kDa antigen on the surface of murinemacrophages [5] It is mainly a cytosolic protein but caneasily traverse the intracellular and plasma membranes totranslocate into the nucleus or mitochondria or get exter-nalized [6] The protein shuttles between the cytoplasm andnucleus on the basis of targeting signals that are recognized byimportins for nuclear localization and exportin-1 for nuclearexport Depending on the cell type specific experimentalconditions in vitro or tissue location Gal-3 has been reportedto be exclusively cytoplasmic predominantly nuclear ordistributed between the two compartments [7] The pres-ence of Gal-3 in the nucleus is dependent on the integrity



of ribonucleoprotein complexes [8] and a Gal-3-U1 smallnuclear ribonucleoprotein (snRNP) complex has been iden-tified which provides a mechanism of incorporation of theGal-3 into the pre-mRNA splicing substrate [9] In additionGal-3 is secreted via nonclassical pathway outside of the cellindependent on the classical secretory pathway through theendoplasmic reticulumGolgi network thus being found onthe cell surface or in the extracellular space [10] Thus Gal-3is a multifunctional protein which regulates pleiotropic bio-logical functions such as cell growth cell adhesion cell-cellinteractions apoptosis angiogenesis and mRNA processingIts unique structure enables interacting with a plethora ofligands in a carbohydrate dependent or independent manner[6]

In thyroid gland Gal-3 plays a role in the pathogenesis ofwell-differentiated carcinoma particularly in papillary carci-noma [11]Therefore it is one of themarkers most commonlyused to assist in distinguishing thyroid lesions together tohuman bone marrow endothelial cell-1 (HBME-1) as a tumormarker of follicular origin and cytokeratin-19 (CK-19) withgeneral intense and diffuse expression in papillary carcinomaand heterogeneous labeling in carcinoma and in follicularadenoma [12] In addition MIB-1 is useful in evaluatingproliferative activity and in predicting the aggressiveness ofthyroid carcinoma [13]

We have previously demonstrated that microgravityinduces changes in the physiology of the thyroid gland Infact in comparison with control animals thyroids of space-flight animals have a more homogenous structure producemore cAMP and overexpress thyrotropin-receptor (TSHR)caveolin-1 [14] and sphingomyelinase and sphingomyelin-synthase [15] and are characterized by a loss of parafollicularcells with reduction of calcitonin production [16]

Data are not available at the time regarding the evaluationof thyroid tumor markers in microgravity We report for thefirst time the effect of long-term exposure to realmicrogravityenvironment on thyroid HBME-1 MIB-1 CK19 and Gal-3


21 Experimental Design and Animal Care All experimentalprocedures were authorized by the Public Veterinary HealthDepartment of the ItalianMinistry ofHealthThe experimentwas also conducted in accordance with the regulations for thecare and use of laboratory animals and with the guidelines ofthe Japanese Physiological Society Furthermore this studywas also approved by the Committee on Animal Care andUse at Graduate School of Medicine Osaka University (no22-071) Finally the protocol utilized in the study has beenauthorized by the Public Veterinary Health Department ofthe Italian Ministry of Health All experiments were carriedout using male C57BL10J mice (8 weeks old)

22 Microgravity Experiment 3 mice were individuallyhoused in the Mouse Drawer System (MDS) a 116 times 98 times84 cmpayload developed byThales-Alenia Space Italy and alltreatments were performed as previously reported [14] Foodand water were supplied ad libitum The MDS was launched

in the Space Shuttle Discovery within the Space TransportSystem (STS)-128 mission on August 28 2009 It was thenhoused in Japanese Experimental Module (Kibou) on the ISSuntil its return to the Earth by Space Shuttle Atlantis (STS-129 mission) on November 27 2009 Only 1 mouse returnedto the Earth alive after 91 days of space flight

Thyroids were sampled bilaterally from eachmouse killedby inhalation of carbon dioxide at the Life Sciences SupportFacility of Kennedy Space Center within 3-4 hours after land-ing and either processed or frozen immediately accordingto the various experimental protocols The procedure wasapproved by the IACUC protocol n∘ FLT-09-070(KSC)

After the spaceflight experiment the on-ground experi-ment was also carried out at the Vivarium of the AdvancedBiotechnology Center in Genoa Italy One group of 3 micewith the same species sex and age was housed in normalvivarium cage as the laboratory control Amount of food andwater supplementation and environmental conditions weresimulated as the flight group After 3 months thyroids weresampled bilaterally and treated for spaceflight mice

23 Thyroid Tissue Treatment The thyroid lobes were fixedin 4 neutral phosphate-buffered formaldehyde solution for24 h as previously reported [14] Thyroids were dropped withessentially randomorientation in paraffinTheparaffinblockswere sectioned into 4-120583m-thick sections All sections weremounted on silane-coated glass slides Each slide contained apair of sections at a distance equal to 140 120583m Between 5 and14 pairs of sections were sampled excluding the first and thelast sections 2 6 and 10 were used for HBME-1 detectionsections 3 7 and 11 for MIB-1 detection sections 4 8 and12 for CK19 detection and sections 5 9 and 13 for Gal-3detection Tissue sectionswere deparaffinized and rehydratedthrough a series of xylene and ethanol washes

24 Immunohistochemical Analysis For immunohistochem-ical analysis Bond Dewax solution was used for removal ofparaffin from tissue sections before rehydration and immun-ostaining on the Bond automated system (Leica BiosystemsNewcastle Ltd UK) as previously reported [17] Immunos-taining detection was performed according to Bancroft andStevens [18] by using HBME-1 and Ki-67 (MIB-1 clone)from Dako (Milano Italy) and CK19 and Gal-3 antibodiesand Bond Polymer Refine Detection from Leica Biosystems(Newcastle Ltd UK) The observations were performed byusing invertedmicroscopy EUROMEXFE 2935 (EDAmhemThe Netherlands) equipped with a CMEX 5000 camerasystem (40x magnification)The analysis of the tissue sectionsize was performed by ImageFocus software

25 Statistical Analysis The experiments have been con-ducted on the thyroid of 1 animal for themicrogravity experi-ment (the only ones that returned alive from themission) and3 control animals for the microgravity experiment (vivarium1) Median and range of sections 2 6 and 10 (HBME-1) ofsections 3 7 and 11 (MIB-1) of sections 4 8 and 12 (CK19)and of sections 5 9 and 13 (Gal-3) were given



Prolonged space flights are known to elicit changes in humancardiovascular musculoskeletal immune and nervous sys-tems whose functions are regulated by the thyroid gland[14] The structure of thyroid shows the presence of folliclescontaining colloid and surrounded by a single layer of thyroidepithelial cells or thyrocytes that produce the metabolicallyactive iodothyronines and parafollicular spaces with thyroidC cells that produce calcitonin [19] We have previouslyreported that thyrocyte cells in culture delay cell growth andenter into a proapoptotic state after long stay on the Inter-national Space Station (ISS) [20] In vivo experiments on theboard of ISS showed that thyroid of spaceflightmice hasmoreordered follicles with thicker thyrocytes containing increasednuclear volume [14] and reduction of interfollicular spacewith loss of C cells [16] in comparison with thyroid glandof ground mice In order to verify whether the structuralchanges of the thyroid gland inmicrogravity conditions couldlead to pathological conditions in this study we investigatedthe immunoexpression of markers known to be related toclinical outcome The limitation of the present paper is thatonly 1 mouse survived to the 91-day spaceflight Howeverthe MDS experiment was a unique opportunity to study themicrogravity long-term exposure effects on several tissuesof an animal model and to collect interesting observationsthat could prepare the field to future experimentsThe resultsshowed that microgravity gives a nonspecific staining inthe colloid during MIB-1 CK19 and Gal-3 immunohisto-chemistry analysis absent in control samples It is reallyhard to pinpoint the reason but it is possible to hypoth-esize an increase of membrane permeability microgravity-dependent on the basis of the observation that at the endof the spaceflight endothelial cells display profound changesindicating cytoskeletal lesions and increased cell membranepermeability [21] MIB-1 and CK19 immunopositivity do notshow changes in thyroid of spaceflight mice in comparisonwith control animals (Figure 1(a)) Differently the immunos-taining is present for HBME-1 and it is very strong for Gal-3(Figure 1(a)) Alshenawy demonstrated that no single markeris completely sensitive and specific for diagnosis of thyroidlesions but only their combination [22] with Gal-3 + HBME-1 was considered the best combination for distinguishingbenign from malignant lesions [23] In thyroid of spaceflightmice the structure of thyroid follicles is more organized thanthat of the control animals [14] and thyrocytes delay theirgrowth [20] and MIB-1 is negative So it is very difficult atthe moment to consider that the expression of HBME-1 andGal-3 markers is linked to tumor transformation Howeverthe possibility that HBME-1 and Gal-3 overexpression mightindicate a premaligne state of thyroid tissue cannot beexcluded by considering that in microgravity follicles aremade up of cells 2 times larger and colloid darker [14] similarto those of papillary carcinoma [24] Our result showedthat HBME-1 is present only in trace in thyroid of controlmice maintained in the vivarium whereas it appears evidentafter space flight with well-defined localization in thyrocytes(Figure 1(a)) Median and range value of immunopositivesurface area is 462 (551ndash457)mm2 and its ratio in relation


HBME-1

MIB-1

CK19

Galectin-3

(a)

ControlMicrogravity

HBME-1 Galectin-30

02

04

06

08

1

Posit

ive s

urfa

ceto

tal s

urfa

ce

(b)

Figure 1 Effect of microgravity on HBME-1 MIB-1 CK19 andGalectin-3 (a) Marker detection in thyroid tissue by immunohis-tochemical staining ldquoControlrdquo mice maintained in vivarium cagesldquomicrogravityrdquo experimental mice in space environment (b) Ratiobetween the immunopositive surface and total surface of thyroidlobe The values are expressed as median and range of two sectionsas reported inMaterial andMethods 40xmagnificationThe arrowsindicate positive areas



Galectin-3

Figure 2 Localization of Galectin-3 in colloid Gal-3 immunohistochemical staining ldquoControlrdquo mice maintained in vivarium cagesldquomicrogravityrdquo experimental mice in space environment 40x magnification The arrows indicate positive areas

to total surface is reported in Figure 1(b) Gal-3 labelling ispresent in some of follicular thyrocytes of control animals andit increases strongly in spaceflight mice (Figure 1(a)) Medianand range value of immunopositive surface area is 172 (199ndash125) mm2 in the control and 794 (859ndash700) in microgravityby increasing 467 times the positive surfacetotal surfaceratio (Figure 1(b)) The presence of Gal-3 in normal thyroidtissue has already been demonstrated [25] Our data show anoverexpression in microgravity We do not have support inthe literature since this is the first study on observation ofthe behavior of thyroid pathological markers in micrograv-ity Nevertheless Grosse et al demonstrated that NF-120581B isoverexpressed and different factors that interact with it aredifferentially regulated under altered gravity conditions [26]In addition spaceflight conditions change gene expressionprofile in thyroid cancer cells [27] Therefore microgravityinfluences gene expression and consequently protein contentHowever the positivity of Gal-3 deserves attention not onlyfor its expression but also and especially for its localizationOur results highlighted that in microgravity conditions Gal-3 leaves thyrocytes and diffuses in colloid (Figure 2) It ispossible that microgravity induces changes of cell membranethat in turn facilitates the escape of Gal-3 accumulated inthyrocytes We have previously demonstrated that thyrocytesin culture (FTRL-5 cell line) release thyrotropin receptorlinked to cholesterol and sphingomyelin in culture mediumduring space missions by indicating a depletion of lipid raftsand consequently cell membrane remodelling [20] Clarke etal told about microgravity-induced decrease in membraneorder [28] and Hsu et al localized Gal-3 in membrane lipidrafts [29] It is possible to suppose that Gal-3 overexpressed inthyrocytes moves into colloid due to the modification of thecell membrane following the variation of gravity force It hasbeen demonstrated that Gal-3 is mainly a cytosolic proteinbut it shuttles to the nucleus or extracellular space the basisof targeting signals [6] Here we do not have specific stainingin these locations but the molecules move in the oppositedirection they do not protrude from the basal membrane of

thyrocytes towards the extracellular space but from the apicalmembrane to the colloid On the other hand Delacour et alsuggested a direct role of Gal-3 in apical sorting as a sortingreceptor [30] It is possible that the gravity force contributes tothe maintenance of the distribution of the molecules in bothbasal membrane side and apical membrane side and that themicrogravity facilitates slippage of Gal-3 in colloid

4 Conclusion

To our knowledge this is the first study correlating thyroidtumor markers with long stay mice in microgravity con-ditions Here we found higher expression of HBME-1 andGal-3 in comparison with ground gravity However MIB-1 proliferative index and CK19 are negative Gal-3 usuallypresent in cytoplasm nuclei and extracellular space leavesthyrocytes and diffuses in colloid probably due to membraneremodelling-microgravity induced



Acknowledgment

This work has been partially supported by grants fromAgenzia Spaziale Italiana (ASI)

References

[1] S H Barondes V Castronovo D N W Cooper et al ldquoGalec-tins a family of animal 120573-galactoside-binding lectinsrdquo Cell vol76 no 4 pp 597ndash598 1994

[2] S Di Lella V Sundblad J P Cerliani et al ldquoWhen galectinsrecognize glycans from biochemistry to physiology and backagainrdquo Biochemistry vol 50 no 37 pp 7842ndash7857 2011


[3] A Grigorian S Torossian and M Demetriou ldquoT-cell growthcell surface organization and the galectin-glycoprotein latticerdquoImmunological Reviews vol 230 no 1 pp 232ndash246 2009

[4] J Hirabayashi and K-I Kasai ldquoThe family of metazoan metal-independent 120573-galactoside-binding lectins structure functionand molecular evolutionrdquo Glycobiology vol 3 no 4 pp 297ndash304 1993

[5] M K Ho and T A Springer ldquoMac-2 a novel 32000 Mr mousemacrophage subpopulation-specific antigen defined by mono-clonal antibodiesrdquo Journal of Immunology vol 14 pp 1221ndash12281982

[6] T Funasaka A Raz and P Nangia-Makker ldquoNuclear transportof galectin-3 and its therapeutic implicationsrdquo Seminars inCancer Biology vol 27C pp 30ndash38 2014

[7] K C Haudek P G Voss L E Locascio J L Wang and R JPatterson ldquoA mechanism for incorporation of galectin-3 intothe spliceosome through its association with U1 snRNPrdquo Bio-chemistry vol 48 no 32 pp 7705ndash7712 2009

[8] J G Laing and J L Wang ldquoIdentification of carbohydratebinding protein 35 in heterogeneous nuclear ribonucleoproteincomplexrdquo Biochemistry vol 27 no 14 pp 5329ndash5334 1988

[9] K C Haudek K J Spronk P G Voss R J Patterson J LWangand E J Arnoys ldquoDynamics of galectin-3 in the nucleus andcytoplasmrdquo Biochimica et Biophysica Acta vol 1800 no 2 pp181ndash189 2010

[10] J Dumic S Dabelic andM Flogel ldquoGalectin-3 an open-endedstoryrdquo Biochimica et Biophysica Acta vol 1760 no 4 pp 616ndash635 2006

[11] T Yoshii H Inohara Y Takenaka et al ldquoGalectin-3 maintainsthe transformed phenotype of thyroid papillary carcinomacellsrdquo International Journal of Oncology vol 18 no 4 pp 787ndash792 2001

[12] L L deMatos A B del Giglio C O Matsubayashi M de LimaFarah andM A da Silva Pinhal ldquoExpression of ck-19 galectin-3 and hbme-1 in the differentiation of thyroid lesions systematicreview and diagnostic meta-analysisrdquoDiagnostic Pathology vol7 no 1 article 97 2012

[13] P Kjellman G Wallin A Hoog G Auer C Larsson and JZedenius ldquoMIB-1 index in thyroid tumors a predictor of theclinical course in papillary thyroid carcinomardquoThyroid vol 13no 4 pp 371ndash380 2003

[14] M A Masini E Albi C Barmo et al ldquoThe impact of long-term exposure to space environment on adult mammalianorganisms a study onmouse thyroid and testisrdquo PLoSONE vol7 no 4 Article ID e35418 2012

[15] E Albi F Curcio R Spelat et al ldquoObserving the mouse thyroidsphingomyelin under space conditions a case study from theMDS mission in comparison with hypergravity conditionsrdquoAstrobiology vol 12 no 11 pp 1035ndash1041 2012

[16] E Albi F Curcio R Spelat et al ldquoLoss of parafollicular cellsduring gravitational changes (microgravity hypergravity ) andthe secret effect of pleiotrophinrdquoPLoSONE vol 7 no 12 ArticleID e48518 2012

[17] E Albi F Curcio R Spelat et al ldquoThe thyroid lobes the differ-ent twinsrdquo Archives of Biochemistry and Biophysics vol 518 no1 pp 16ndash22 2012

[18] J D Bancroft and A Stevens EdsTheory and Practice of Histo-logical Techniques Churchill Livingstone New York NY USA1996

[19] C C Capen and S L Martin ldquoThe effects of xenobiotics onthe structure and function of thyroid follicular and C-cellsrdquoToxicologic Pathology vol 17 no 2 pp 266ndash293 1989

[20] E Albi F S Ambesi-Impiombato M Peverini et al ldquoThyrot-ropin receptor and membrane interactions in FRTL-5 thyroidcell strain in microgravityrdquo Astrobiology vol 11 no 1 pp 57ndash64 2011


[22] H A Alshenawy ldquoUtility of immunohistochemical markers indiagnosis of follicular cell derived thyroid lesionsrdquo Pathology ampOncology Research 2014

[23] H A Saleh J Feng F Tabassum O Al-Zohaili M Husainand T Giorgadze ldquoDifferential expression of galectin-3 CK19HBME1 and Ret oncoprotein in the diagnosis of thyroidneoplasms by fine needle aspiration biopsyrdquo CytoJournal vol6 article 18 2009

[24] R V Lloyd D Buehler and E Khanafshar ldquoPapillary thyroidcarcinoma variantsrdquo Head and Neck Pathology vol 5 no 1 pp51ndash56 2011

[25] J Feilchenfeldt M Totsch S-Y Sheu et al ldquoExpression of ga-lectin-3 in normal and malignant thyroid tissue by quantitativePCR and immunohistochemistryrdquo Modern Pathology vol 16no 11 pp 1117ndash1123 2003



[28] M S Clarke C R Vanderburg and D L Feeback ldquoThe effectof acute microgravity on mechanically-induced membranedamage and membrane-membrane fusion eventsrdquo The Journalof Gravitational Physiology vol 8 no 2 pp 37ndash47 2001

[29] D K Hsu A I Chernyavsky H-Y Chen L Yu S A Grandoand F-T Liu ldquoEndogenous galectin-3 is localized in membranelipid rafts and regulates migration of dendritic cellsrdquo Journal ofInvestigative Dermatology vol 129 no 3 pp 573ndash583 2009

[30] D Delacour C I Cramm-Behrens H Drobecq A Le BivicH Y Naim and R Jacob ldquoRequirement for galectin-3 in apicalprotein sortingrdquo Current Biology vol 16 no 4 pp 408ndash4142006

Research ArticleThe Influence of Simulated Microgravity on PurinergicSignaling Is Different between IndividualCulture and Endothelial and Smooth Muscle Cell Coculture

Yu Zhang12 Patrick Lau3 Andreas Pansky1 Matthias Kassack2

Ruth Hemmersbach3 and Edda Tobiasch1

1 Department of Natural Sciences Bonn-Rhine-Sieg University of Applied Sciences 53359 Rheinbach Germany2 Institute of Pharmacology and Medical Chemistry University of Dusseldorf 40225 Dusseldorf Germany3 Institute of Aerospace Medicine German Aerospace Center 51147 Cologne Germany

Correspondence should be addressed to Edda Tobiasch eddatobiaschh-brsde

Received 25 April 2014 Revised 30 June 2014 Accepted 23 July 2014 Published 28 August 2014


Copyright copy 2014 Yu Zhang et alThis is an open access article distributed under theCreativeCommonsAttribution License whichpermits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Exposure to microgravity conditions causes cardiovascular deconditioning in astronauts during spaceflight Until now no specificdrugs are available for countermeasure since the underlying mechanism is largely unknown Endothelial cells (ECs) and smoothmuscle cells (SMCs) play key roles in various vascular functions many of which are regulated by purinergic 2 (P2) receptorsHowever their function in ECs and SMCs under microgravity conditions is still unclear In this study primary ECs and SMCswere isolated from bovine aorta and verified with specific markers We show for the first time that the P2 receptor expressionpattern is altered in ECs and SMCs after 24 h exposure to simulated microgravity using a clinostat However conditioned mediumcompensates this change in specific P2 receptors for example P2X7 Notably P2 receptors such as P2X7 might be the importantplayers during the paracrine interaction Additionally ECs and SMCs secreted different cytokines under simulated microgravityleading into a pathogenic proliferation and migration In conclusion our data indicate P2 receptors might be important playersresponding to gravity changes in ECs and SMCs Since some artificial P2 receptor ligands are applied as drugs it is reasonable toassume that they might be promising candidates against cardiovascular deconditioning in the future

1 Introduction

Exposure to microgravity conditions during space missionsinduces a variety of health issues in astronauts includ-ing bone loss muscle atrophy decreased immune activityand cardiovascular deconditioning [1ndash3] The cardiovasculardeconditioning is very likely caused by the dysfunction ofthe major vascular cells endothelial cells (ECs) and smoothmuscle cells (SMCs) ECs build up the monolayer coatinginner surface of blood vessels Layers of SMCs arranged infibers support the EC monolayer by providing contractionand relaxation of vessels [4] Importantly the interactionbetween ECs and SMCs has been shown to be a key playerin human cardiovascular physiology [5] ECs are sensitive tomechanical stress and they secret cytokines inhibiting SMCproliferation [6] Purinergic receptors can bind extracellular

nucleotides such as ATP [7 8] and they are crucial playersin regulating a series of physiological and pathological car-diovascular processes such as atherosclerosis hypertensionand vascular pain [9 10] Purinergic receptors are dividedinto P1 receptors and P2 receptors [11] P2 receptors can besubdivided into P2X receptors that are ion channels and P2Yreceptors that are G protein-coupled receptors [12] Untilnow seven P2X (P2X1-7) and eight P2Y (P2Y1 2 4 6 1112 13 and 14) have been characterized However the roleof extracellular nucleotides on vascular cell function undermicrogravity condition is still unknown

Recent publications have shown that cytoskeletonarrangement gene expression of extracellular matrix andcell surface adhesion molecules in ECs were altered after22 seconds and 24 h exposure to microgravity [13ndash16]ECs formed tubes after culturing for longer term (7 days)



(a)

Rotation axis18mm

6mm

(b) (c)

Figure 1 Cell Culture in aClinostat to SimulateMicrogravityConditionsThe clinostat (a) was used to simulate themicrogravity environmentby rotating cells Only cells grown in the 6mm area in the middle of culture slide (b) received the optimal simulated microgravity and werethus harvested using a special cell scraper (c)

under simulated microgravity conditions using a RandomPositioning Machine (RPM) [17] On the other hand SMCsshowed suppressed proliferation and an enhanced rate ofapoptosis after 72 h exposure to simulatedmicrogravity usinga rotating wall vessel (RWV) [18] However these findingswere encountered when single cell type such as endothelialor smooth muscle cells was cultured under real or simulatedmicrogravity Considering ECs already showed to secretedifferent cytokines under simulated microgravity using RPM[19] the important interactions between ECs and SMCsunder microgravity condition should be evaluated and thusrequire investigations

In this study an indirect cell coculture model wasestablished by culturing SMCs with EC-conditionedmediumand vice versa The P2 receptor expression pattern was ana-lyzed and compared under three conditions normal gravitycontrol (1 g) simulated microgravity (MG) and simulatedmicrogravity with conditioned medium For the simulationof microgravity a fast rotating clinostat was used in whichthe cells were quickly rotated around one axis perpendicularto the direction of gravity [20] The influence of condi-tioned medium collected from normal gravity and simulatedmicrogravity on cell proliferation and migration was alsoinvestigated

2 Methods

21 Isolation and Characterization of Bovine Aortic Endothe-lial and Smooth Muscle Cells Bovine aorta was cut longi-tudinal into 5 cm sections and divided again into rectanglesafter removing residual and connective tissues The cut aortainto type I collagenase (10mgmL in PBS) coated cell culturedishes with the inner layer (endothelium) attached to thecollagenase and incubated for 60min at 37∘C The aorticendothelial cells were slightly scraped with a cell scraperand put onto gelatin (in PBS (1 vv)) coated culture plates[21] Medium was added to the freshly scraped cells and theplates were then incubated at 37∘C 5CO

2under humidified

conditions Aortic smooth muscle cells were isolated byobtaining the media layer through removal of the outer layer

and scraping off the endothelial cellsThemedia layer was cutinto 2mm times 2mm sections and put into cell culture dish for2 h without medium to allow these sections adhering tightlyto the surface [22] The medium was added and the pieceswere incubated at 37∘C 5CO

2under humidified conditions

for up to a week to let SMCs migrate and proliferate from thetissue pieces to the surface of culture dish

22 Cell Culture The cells were cultured in DMEMmedium(Merck Millipore Berlin Germany) supplied with 10 FCSand 1 penicillinstreptomycin ECs and SMCs were splitand seeded at a density of 5000 cellscm2 after they reacheda level of 80ndash90 confluence ECs and SMCs with passagenumber 2ndash4 were used The cell line human microvascularendothelial cell-1 (HMEC-1) C2C12 (ATCC number CRL-1772) andMG-63 (ATCC number CRL-1427) and U-87 MG(ATCC number HTB-14) were cultured in DMEM mediumand subsequently used as positive control

23 Clinostat Experiments The fast-rotating 2D clinostatused in this study was originally developed by the Institute ofAerospace Medicine German Aerospace Center (DLR) (seeFigure 1(a)) It has 6 parallel horizontal axes each for fixationfor up to 4 slide flasks ECs and SMCswere seeded at a densityof 10000 cellscm2 onto 9 cm2 cell culture slide flasks (NuncThermo Fisher Scientific Langenselbold Germany) Whenthey reached a confluence level of 60ndash70 the culture flaskswere filled up completely with DMEM medium To avoidshear stress and thus the induction of respective metabolicchanges in signal transduction pathways for example apop-tosis air bubbles were removed carefully The flasks wereinserted on the clinostat and rotated at 60 rpm for 24 h in theCO2incubator at 37∘CControls were also filledwithmedium

and placed simultaneously under normal gravityCells from the whole flask were first used to analyze the

P2 receptor expression pattern that altered subtypes couldbe distinguished from unaffected Later according to theclinostat principle only cells exposed to minimal g-forceswere taken for further analysis This means that only cellsfrom the middle of the flask were taken (see Figure 1(b))


Under a defined constant speed of 60 rpm the maximalresidual acceleration at an area of 6mm provided an optimalquality of simulated microgravity (le00121 g) [23] Thus onlycells within this 6mm area were isolated to evaluate alteredP2 subtypes for both gene and protein expression in detail Tomaintain the cells accurately and consistently in the center aspecial chamber consisting of two cover plates attached to abottom plate [23] was used to allow slide insertion withoutwiping off the cell layer A corresponding cell scraper wasused to scratch the cells from the specific 6mm width areain the center (Figure 1(c))

24 ConditionedMedium To investigate a possible paracrineinfluence on P2 receptor expression ECs and SMCs wereseeded in a density of 2500 cellscm2 Cell growth mediumwas collected when they were 80ndash90 confluent Theconditioned medium (CM) was composed out of cell growthmedium and normal DMEM medium in a ratio of 1 2 onrespective cell type The SMC-conditioned medium was sub-sequently fully added into the culture slide with ECs and setof a 24 h clinorotation as a group of ECMG+CMThe ECs innormal gravity group (EC 1 g) and in clinorotation but filledwith normal DMEM medium (EC MG) were set simultane-ouslyThe similar experimentswere set for SMCswith normalgravity (SMC 1 g) clinorotation (SMC MG) and clinorota-tion filled with EC-conditioned medium (SMCMG + CM)

To evaluate a possible paracrine effect on cell prolif-eration and migration cell growth medium was collectedfrom cells cultured 24 h in normal gravity and cultured24 h under clinorotation respectively The used conditionedmedium (CM) was composed out of cell growth mediumand normal DMEM medium in a ratio of 1 2 on respectivecell type The ECs were subsequently treated with normalDMEM medium SMC-conditioned medium from normalgravity (CM SMC + 1 g) and clinorotation (CM SMC +MG) separately for proliferation or migration assays Similarexperiments were set and performed with SMCs as wellAll experiments were performed with samples from threecows

25 RNA Isolation and Semiquantitative PCR RNAwas extracted after clinorotation using a Ribozol RNAreagent (Amresco OH USA) cDNAs were synthesizedfrom 20 120583g total RNA by using Revert Aid ReverseTranscriptase and oligo-dT primer (Thermo FisherScientific MA USA) Primers for P2 receptors EC andSMC specific markers in the human and bovine system weredesigned and shown in the supplementary data availableat httpdxdoiorg1011552014413708 The RT-PCRconditions such as annealing temperature and magnesiumconcentration are given in the supplementary data as well 1of agarose gels were set up to evaluate the RT-PCR productsAs positive control RNA extracts from the cell lines HMEC-1 MG-63 C2C12 and U-87 MG were used for respective P2receptor subtypes given in the supplementary data

26 Western Blot Analysis The proteins were extracted fromthe cells in a protein lysis buffer (Cell Signaling Technology

MA USA) and subsequently centrifuged at 22000 g for5min at 4∘C to remove cellular debris After boiling for5min the lysate samples were separated by a 12 SDS-PAGEelectrophoresis and electrotransferred to a PVDFmembraneThe membrane was blocked in TBST containing 5 BSAand incubated with anti-P2X7 P2Y1 P2Y2 P2Y11 VEGFR2VE-cadherin PECAM-1 calponin SMA-120572 MYH-11 (1 500)or GAPDH antibodies (1 5000) (Santa Cruz BiotechnologyCA USA) overnight at 4∘C The membranes were washedthree times with TBST and incubated with the secondaryantibodies (1 5000) (CALBIOCHEM CA USA) for 60minat RT After washing with TBST immune-detection wasaccomplished by using the Luminata Forte Western HRPsubstrate (MerckMilliporeMAUSA) and imageswere takenusing Bio-Rad Chemidoc system

27 Immunofluorescence The cells were fixed in 4paraformaldehyde for 15min Cells were incubated withprimary anti-VEGFR2 VE-cadherin PECAM-1 calponinSMA-120572 MHY-11 P2X7 P2Y1 P2Y2 and P2Y11 (SantaCruz Biotechnology CA USA) diluted in a ratio of 1 100in antibody dilution buffer containing 1 BSA and 02Triton-X-100 in PBS at 4∘C overnight After rinsing withPBS 3 times cells were stained with FITC-labeled anti-goator rabbit antibody respectively (1 100) (Southern BiotechAL USA) at RT for 60min Cell nuclei were stained withDAPI (Sigma MO USA) and the cell cytoskeleton waslabeled using rhodamine (1 2000) (Life Technologies CAUSA) After washing with PBS fluorescent signals wereanalyzed with an Axio Observer D1 fluorescence microscope(Carl Zeiss Germany) or a FW300 confocal fluorescentmicroscope (Olympus Japan) respectively

28 DiI-ac-LDL Uptake ECs were incubated with 10 120583gmLDiI-labeled acetylated-low density lipoprotein (DiI-ac-LDL)(Biomedical Technologies Inc MA USA) for 4 hours at 37∘Cand investigated with a fluorescent microscope (Carl ZeissGermany) at a wavelength of 565 nm After staining withLDL cells were fixed with 4 formaldehyde for 15min andsubsequently incubated with DAPI (1 10000 in PBS) andrinsed with PBS Images were taken with an Axio ObserverD1 fluorescent microscopy (Carl Zeiss Germany)

29 Proliferation and Wound Assay For the proliferationassay 20000 ECs were seeded separately in each well of 12-well plates ECs were grown in DMEM medium in a SMC-conditioned medium in normal gravity (SMC CM + 1 g) andin a SMC-conditioned medium in simulated microgravity(SMC CM + MG) (see Section 24 for conditioned mediumdetails) ECs incubated with the respective medium weresubsequently obtained after 24 h and 48 h under normalgravity incubation and numbers were calculated Similarlyexperiments were set for SMCs SMCs were cultured inDMEM medium in an EC-conditioned medium in normalgravity (EC CM + 1 g) and EC-conditioned medium insimulatedmicrogravity (ECCM+MG) for 24 h and 48 h Cellnumber in each well was counted


For the wound assay ECs and SMCs (10000cm2) wereseeded and grown to 80ndash90 confluence A straight scratchinjury was made using a sterile 1mL pipette tip on 6-wellplates The ECs were incubated for 24 h at 37∘C in a CO

2

incubator with normal DMEM medium SMC-conditionedmedium in normal gravity (SMC CM + 1 g) and SMC-conditioned medium in simulated microgravity (SMC CM+ MG) On the other hand SMCs were cultured in DMEMmedium EC-conditionedmedium innormal gravity (ECCM+ 1 g) and EC-conditioned medium in simulated micrograv-ity (ECCM+MG) Hydroxyurea (5mM)was added to inhibitcell proliferation Images were taken using a phase contrastmicroscope (Carl Zeiss Germany)The numbers of migratedcells in three individual areas were calculated and quantifiedusing Image J software (NIH)

210 Statistical Analysis Statistical analysis was applied forthe experiments using the Microsoft Office program Excel2010 and SPSS 120 Data are shown as means plusmn standarddeviation Experiments were repeated at least three times forthree donors which are given as 119899 = number of experimentsThe probability (119875) value was calculated using LSH test toassess differences between two groups Levels of significancewere labeled as follows lowast119875 le 005 lowastlowast119875 le 001 and lowastlowastlowast119875 le0001 Significance was given with the appropriate number ofasterisks or in numbers

3 Results

31 Characterization of Primary ECs and SMCs from BovineAorta The isolated ECs showed positive gene expressionof endothelial cell markers VEGFR2 VE-cadherin andPECAM-1 whereas SMCs positively expressed smooth mus-cle cell markers SMA-120572 calponin andMYH-11 (Figure 2(a))Western blot experiments confirmed the gene expressiondata on the protein level ECs positively expressed VEGFR2VE-cadherin and PECAM-1 while SMCs were positive forcalponin SMA-120572 and MYH-11 (Figure 2(b)) Importantlyboth gene and protein data showed that ECs were negativefor SMC markers except a weak band found in calponinSMCs were negative for the three tested endothelial cellmarkers These results indicate that isolated ECs and SMCswere without major cross contaminations The fluorescentstaining data further confirmed the results from the RT-PCRandWestern blot analysis (Figures 2(c) and 2(d)) In additionisolated ECs also showed the typical endothelial activity byuptaking LDL (Figure 2(e))

32 P2 Receptor Expression in ECs after 24 h Simulated Micro-gravity with and without SMC-Conditioned Medium Allfifteen P2 receptors were analyzed for their gene expressionby RT-PCR In the first experiment RNAwas collected in oneset of clinorotation experiments from the whole culture flaskAll P2 receptors were expressed in ECs with the exceptionof P2X3 and P2Y6 Next to this P2X5 P2Y4 P2Y11 andP2Y14 were upregulated while P2X7 P2Y1 and P2Y4 weredownregulated on the gene expression level in ECs under24 h simulated microgravity condition (MG) induced by

clinorotation if compared to ECs under normal gravity (1 g)(Figure 3(a)) In a further set of clinorotation experimentsthe conditioned medium collected from SMCs grown undernormal gravity condition (see Section 24) was added toECs For this experiment only cells grown in the 6mmdiameter area of the center were taken to isolate RNA andprotein RT-PCR data showed that although the expressionof P2X7 and P2Y1 was decreased after clinorotation P2X7in ECs showed an increase on the gene level when culturedin SMC-conditioned medium P2Y11 protein expression inECs was upregulated and further increased also on theSMC conditioned medium compared to P2X7 (Figure 3(b))Western blot and fluorescence confirmed the change of P2X7on protein level (Figures 3(c) and 3(d))

33 P2 Receptor Expression in SMCs after 24 h SimulatedMicrogravity with and without EC-Conditioned MediumIdentical operational steps were undertaken to investigateSMCs under simulated microgravity In SMCs all P2 recep-tors were expressed except P2X3 P2X7 P2Y6 and P2Y11After 24 h clinorotation RT-PCR showed an increased geneexpression of P2X4 P2X7 and P2Y2 whereas P2X2 P2Y1and P2Y14 were downregulated in SMCs under simulatedmicrogravity condition (MG) if compared to the SMCsunder normal gravity (1 g) (Figure 4(a)) After adding EC-conditionedmedium (see Section 24) within clinostat exper-iment clinorotation induced an upregulation of P2X7 geneexpression in SMCs as revealed by RT-PCR (Figure 4(b))Interestingly P2X7 showed a decreased gene expression afteradding EC-conditioned medium compared to its increasewithout EC-conditioned medium under 24 h clinorotationP2Y1 was upregulated in SMCs under simulated micro-gravity however conditioned medium showed no effect onits expression Gene level alterations of P2X7 and P2Y2were confirmed on the protein level by Western blot orfluorescent staining however P2Y1 showed an increasingprotein expression in simulated microgravity and with EC-conditioned medium (Figures 4(c) and 4(d))

34 Proliferation and Migration of ECs Cultured with SMC-Conditioned Medium Collected under Normal Gravity andSimulated Microgravity Conditioned medium from SMCscollected after 24 h normal gravity and after 24 h simu-lated microgravity was used to culture ECs evaluating theparacrine influence of SMCs on EC proliferation and migra-tion SMC-conditioned medium from normal gravity (SMCCM+ 1 g) did not have a significant influence on ECprolifera-tion after 24 h but caused a decrease of EC numbers after 48 hSMC-conditioned medium collected after simulated micro-gravity (SMC CM + MG) inhibited EC proliferation signif-icantly after both 24 h and 48 h respectively (Figure 5(a))To mimic a wound in the endothelium a straight scratchthrough the cells was set The SMC-conditioned mediumcultured under normal gravity (SMCCM+ 1 g) and simulatedmicrogravity (SMCCM+MG) condition was added to studyEC migration capacity The conditioned medium from MGenhanced EC migration after 24 h and even more signifi-cantly after 48 h in the presence of hydroxyurea (Figures 5(b)


pc EC SMC

PECAM-1

VEGFR2

MYH-11

Calponin

SMA-120572

VE-cadherin

120573-Actin

(a)

pc EC SMC

PECAM-1

VEGFR2

Calponin

SMA-120572

VE-cadherin

120573-Actin

(b)

NegDAPI VEGFR2DAPI

VE-cadherinDAPI PECAM-1DAPI

(c)

NegDAPI CalponinDAPI

MYOCDDAPI MYH11DAPI

(d)

Dil-ac-LDLDAPI

(e)

Figure 2 Characterization of Bovine Endothelial and Smooth Muscle Cells The isolated cells were verified with the EC specific markersVEGFR2 VE-cadherin PECAM-1 and the SMC specific markers calponin SMA-120572 MYH-11 by RT-PCR (a) and Western blot (b) GAPDHserved as internal control The endothelial (c) and smooth muscle cells (d) were also identified with the above-mentioned markers viafluorescent staining The isolated endothelial cells were further examined for the typical endothelial activity of LDL up-take (e) All picturesare representative of one cow sample out of three

and 5(c)) Figure 5 is a representative of example one of threecows The numbers of proliferating and migrating ECs fromthe three individual cows are given in the supplementary data

35 Proliferation andMigration of SMCs Grown in ECDerivedConditioned Medium Collected after 24 h under NormalGravity and SimulatedMicrogravity Experimentswith SMCswere performed in a comparable manner as described forECs The conditioned medium collected from EC grownunder normal gravity (ECCM+ 1 g) reduced the proliferationof SMCs However the conditioned medium collected fromEC grown under simulated microgravity condition (EC CM+ MG) compensated this effect (Figure 6(a)) Conditionedmedium under simulated microgravity induced SMCmigra-tion after 48 h but inhibited it after 24 h (Figures 6(b) and6(c)) Figure 6 is a representative of example one of threeindividual cows The numbers of proliferating and migrating

SMCs from the three individual cows are given in thesupplementary data

4 Discussion

In this study we showed for the first time that several specificP2 receptor expressionswere altered on gene and protein levelafter 24 h under simulated microgravity condition as shownin Figure 7 Culturing ECs with SMC-conditioned mediumunder normal gravity and vice versa can compensate the P2receptor expression change such as P2X7

Similar to the findings of Wang and colleagues [24]our data showed that ECs and SMCs expressed differentP2 receptors on cell membrane P2X4 P2Y1 P2Y2 andP2Y11 were predominantly expressed in ECs while P2X1 andP2Y2 were strongly expressed in SMCs Macrovascular and


P2Y1

P2Y2

P2Y6

P2Y14

P2Y13

P2Y12

P2Y11

P2Y4

GAPDH

pc

P2X5

P2X6

P2X4

P2X1

P2X3

P2X2

P2X7EC

MG

EC1

g

(a)

GAPDH

pc

P2Y13

P2Y11

P2Y1

P2Y2

P2X7

P2X5

EC M

G

ECM

G+

CM

EC1

g

(b)

P2Y1

pc

P2Y2

P2Y11

GAPDH

P2X7

P2X5

EC M

G

ECM

G+

CM

EC1

g

(c)

P2Y1 F-actin Merge P2Y11 F-actin MergeP2X7 F-actin Merge

MG

MG + CM

1g

(d)

Figure 3 P2 Receptor Expression in Endothelial Cells after 24 h under Normal Gravity and SimulatedMicrogravity All cells on the surface offlasks were isolated for RT-PCR P2X5 P2Y4 P2Y11 P2Y13 were up-regulated and P2X7 P2Y1 and P2Y2were down-regulated in the ECs after24 h in the clinostat (a) Cells grown within 6mm of the center had the optimal simulated microgravity condition and were therefore isolatedto confirm the above P2 receptor alteration on the RNA (b) and protein (c) level after 24 h simulated microgravity with and without SMC-conditioned medium collected under normal gravity P2X5 and P2Y11 were up-regulated in ECs but P2X5 up-regulation was not significanton protein level P2X7 P2Y1 P2Y2 were down-regulated on both gene and protein level The SMC-conditioned medium can compensate thedecrease of P2X7 expression but cause no significant effect on the alteration of P2Y1 P2Y2 and P2Y11The fluorescent staining confirmed theprotein change of P2X7 P2Y1 and P2Y11 (d)

microvascular ECs have shown the several functional differ-ences such as matrix metalloproteinase expression [25] andbeta-adrenergic regulation of transendothelial permeability[26]The expression of P2X3 and P2Y4 is low in bovine aorticECs if compared to the control HMEC-1 which suggests thatmacrovascular ECsmight differ in the P2 receptor expressionpattern compared to microvascular ECs

The P2 receptor expression patterns of ECs and SMCshave already been shown to play an important role invarious cardiovascular functions For example in controlling

vascular tone ATP and UTP released from ECs act on P2Y1P2Y2 and P2Y4 leading to the production of NO and sub-sequent vasodilatation Simultaneously ATP released by thesympathetic nerve acts on P2X1 P2X2 and P2X4 resultingin vasoconstriction [9] We found that the expression ofP2X2 and P2X4 in SMCs was significantly increased afterclinorotation indicating to maybe more vasoconstrictionNext to this P2Y1 and P2Y2 expressions were decreasedwhich suggests NO productionmight decrease and cause lessvasodilatation Kang and colleagues found that 72 h exposure


P2X5

P2X6

P2X4

P2X1

P2Y1

P2Y2

P2Y6

P2Y14

P2Y13

P2Y12

GAPDH

P2Y11

P2Y4

P2X7

P2X3

P2X2pc SM

C M

G

SMC1

g

(a)

P2Y2

P2Y1

GAPDH

P2X4pc

P2X7

P2Y14

P2X2

SMC

MG

SMC

MG+

CM

SMC1

g

(b)

pc

P2Y1

GAPDH

P2Y2

P2Y14

P2X7

SMC

MG

SMC

MG+

CM

SMC1

g

(c)

P2X7 F-actin Merge P2Y1 F-actin Merge P2Y2 F-actin Merge

MG

MG + CM

1g

(d)

Figure 4 P2 Receptor Expression in SmoothMuscle Cells after 24 h under Normal Gravity and SimulatedMicrogravityThe experiments forSMCwere performed similarly to those for the endothelial cells above All cells on the surface of flasks were isolated for RT-PCR P2X4 P2X7and P2Y2were up-regulated whereas P2X2 P2Y1 and P2Y14 were down-regulated in the SMCs after 24 h clinorotation (a) Cells grownwithin6mm of the center had the optimal simulated microgravity condition and were thus isolated to confirm the above P2 receptor alteration onboth RNA (b) and protein (c) level after 24 h clinorotation with and without EC-conditioned medium from normal gravity P2X2 and P2X4showed no significantly changed P2X7 and P2Y2 was up-regulated P2Y1 and P2Y14 were down-regulatedThe EC-conditioned medium cancompensate for the increase of P2X7 and P2Y2 expression but no significant effect was observed on P2Y1 and P2Y14 (d)

to clinorotation led to a decreased proliferation but increasedthe rate of apoptotic SMCs Additionally the SMC phenotypewas induced and transferred from the contractive to thesynthetic type [18] Our data showed that the expression ofP2X7 and P2Y2 was altered differentially between ECs andSMCs under simulated microgravity which indicates thatthey could be the key P2 receptor subtypes responding tothe change of gravity To point out P2X7 has an importantrole in cell apoptosis and can activate a series of downstreamsignals due to several protein kinase binding sites on itslong intracellular tail [10] A mechanical force such as shear

stress can induce endothelial cell apoptosis that might beregulated through P2X7 It is of interest that P2X7 which wasdownregulated in ECs was upregulated in SMCsThis changecould be compensated by adding conditioned medium fromthe other cell type Such compensation was not found in theother P2 receptor subtypes which might point to P2X7 as amajor player in the interaction between ECs and SMCs insimulated microgravity with respect to apoptosis regulation

Various proteins of endothelial cells are altered under realor simulated microgravity such as F-actin [27] tubulin [28]cell adhesion molecules [16] integrins [27] eNOS [29] and


0

10000

20000

30000

40000

50000

60000

70000

80000

0

(h)24 48

DMEM SMC CM + MG

lowastlowast

lowastlowast

lowast

lowastlowastlowast

SMC CM + 1g

DMEM SMC CM + MG

200

150

100

50

0

lowastlowast

lowastlowastlowast

SMC CM + 1g

DMEM SMC CM + MGSMC CM + 1g

(a)

(b)

(c)

Figure 5 Effect of SMC-ConditionedMedium on EC Proliferation andMigration To evaluate the EC proliferation normal DMEMmediumand conditioned medium of SMCs after normal gravity (SMC CM + 1 g) and simulated microgravity (SMC CM +MG) was added for a 24 hand 48 h culture period The conditioned medium from SMCs after normal gravity showed a decrease of EC numbers after 48 h SMC-conditioned medium under simulated microgravity inhibited EC proliferation significantly after 24 h and 48 h incubation (a) To evaluateEC migration ECs were scratched and normal DMEM medium SMC-conditioned medium under normal gravity and under simulatedmicrogravity was added for a 24 h culture period (b) Conditioned medium from SMCs under simulated microgravity (SMC CM + MG)enhanced EC migration significantly if compared to the normal DMEMmedium and SMC-conditioned medium under normal gravity (c)

iNOS [30] In line with that ECs also showed a decreasedproliferation rate increased apoptosis [28 31] and migration[29] in simulated microgravity However these data wereobserved based on cultured ECs as single cell type eitherunder real or simulated microgravity An EC and SMCcoculture model was successfully created via EC-conditionedmedium culturing SMCs and vice versa In addition to theparacrine effect on P2 receptor expression such as P2X7our results showed that this effect found under simulatedmicrogravity could influence EC or SMC behavior for cellproliferation and migration SMC proliferation has beendemonstrated to be a crucial process in atherosclerosis sincemigrated and proliferating SMCs form amajor cell type in theplaque [32]

In healthy vessels ECs secrete cytokines that inhibit SMCproliferation and form a monolayer to block small moleculesfrom blood that might cause SMC proliferation We foundthat conditioned medium collected from ECs in normalgravity inhibited SMCproliferation but conditionedmediumcollected from ECs after clinorotation was not able to doso On the other hand EC damage or dysfunction is oneof the first steps during the development of pathologicalchange in atherosclerosis The conditioned medium of SMCgrown in simulated microgravity reduced EC proliferationEnhanced apoptosis was observed when only ECs werecultured in simulatedmicrogravity by Infanger and coauthors[28] They found several caspases such as caspase-3 andcaspase-9 activated after simulated microgravity treatment


0

10000

20000

30000

40000

50000

60000

70000

80000

0 24 48

(h)

DMEM EC CM + MG

lowast

lowastlowast

lowastlowastlowast

lowastlowastlowast

lowastlowastlowast

EC CM + 1g

DMEM

200

150

100

50

0

250

lowastlowast

lowastlowast

lowastlowastlowast

EC CM + MG

EC CM + 1g

DMEM EC CM + MGEC CM + 1g

(a)

(b)

(c)

Figure 6 Effect of EC-Conditioned Medium on SMC Proliferation and Migration To investigate the SMC proliferation normal DMEMmedium and EC-conditioned medium under normal gravity (EC CM + 1 g) and simulated microgravity (EC CM + MG) was added andSMCs were subsequently incubated for 24 h and 48 h Conditionedmedium from EC grown under normal gravity (ECCM+ 1 g) significantlyinhibited SMC proliferation after 24 h and 48 h incubation and conditioned medium from EC grown under simulated microgravity (EC CM+ MG) led to a significant decrease of SMC numbers after 24 h which is not obvious after 48 h (a) To evaluate SMC migration SMCs werescratched and cultured under normal DMEMmedium EC-conditionedmedium under normal gravity and under simulatedmicrogravity for24 h (b) EC-conditioned medium under normal gravity (EC CM + 1 g) inhibited the SMCmigration significantly Whereas EC-conditionedmedium under simulated microgravity (EC CM +MG) enhanced migrated SMC numbers (c)

Apoptosis might be induced by activation of NF-120581B via thePI3KAkt pathway [28 31]These data suggest that astronautsmay be more prone to suffer from cardiovascular diseasessuch as atherosclerosis during space missions and paracrineeffects between ECs and SMCs might be the key factors inthis process On the other hand migration of ECs is thefirst step in angiogenesis and a major factor in metastasisIn addition it also plays an important role in restenosisin the vascular system after application of a stent Thereare evidences that microgravity can promote angiogenesisin both macrovascular and microvascular ECs when onlyECs were cultured under simulated microgravity [29 30]Our data showed an enhanced number of migrated ECswhen cultured with SMC-conditioned medium derived after

clinostat application compared to the DMEM control andSMC-conditioned medium collected after 24 h exposure tonormal gravityThis indicates the effect ofmicrogravitymightenhance the angiogenesis via both autocrine and paracrinesignals

Contradictory observations have been demonstrated inseveral publications for example EC migration increasedin simulated microgravity both in this study and in thestudy of Siamwala and colleagues [30] while Versari andcolleagues found a decreased EC migration under simulatedmicrogravity [33] One explanation for these findings mightbe that different endothelial cells were used such as primaryendothelial cells from an artery or umbilical vein or anendothelial cell line (EAhy926) which might give a different


SMC

EC EC

SMC

Clinorotation(MG)

Paracrine effect

(protective)

Paracrine effect

(pathogenic)

P2X7 P2Y1

P2Y2

P2Y14

P2X5

P2X7

P2Y11P2Y4

P2Y1

P2Y2P2Y13

P2X4

1g

Figure 7 Scheme of P2 Receptor Alteration and the Postulated Paracrine Effect in ECs and SMCs under Simulated Microgravity SeveralP2 receptor expressions were altered in ECs and SMCs under 24 h simulated microgravity condition using a clinostat To point out that theexpression P2X7 and P2Y2 was altered differentially between ECs and SMCs under simulated microgravity Especially the change of P2X7in ECs was compensated under SMC-conditioned medium and vice versa The conditioned medium collected under simulated microgravityshowed the pathogenic influence of EC and SMC proliferation and migration if compared to condition medium from normal gravity

response due to its immortalization and thus prolongedtime in culture Another explanation could be that differentdevices were applied to simulate microgravity conditionssuch as the clinostat the RPM and the RWV Differentequipment might produce different qualities of microgravityas well as a different amount and quality of shear stressduring rotation [20] Clinorotationwas shown to produce thelowest shear forces and the central area used in our studyhas an optimized simulated microgravity environment [23]Furthermore different ECs from different body parts wereused Macrovascular andmicrovascular ECs already revealeda difference in promoting angiogenesis under realsimulatedmicrogravity conditions which is regulated via the iNOS-cGMP-PKG pathway in macrovascular ECs but via eNOS-PI3K-Akt in microvascular ECs [29 30] Taken togetherthe simulated microgravity data independently of ground-based facility we use have to be approved and verified in realmicrogravity for a final statement on the outcome

5 Conclusion

Our data show for the first time that P2 receptor gene andprotein expression in both ECs and SMCs were altered undersimulated microgravity SMC-conditionedmedium collectedunder simulated microgravity influenced some P2 receptorexpressions as well as proliferation and migration of ECs andvice versa Additionally proliferation and migration of ECsand SMCs differed between conditioned medium collectedunder normal gravity and under simulated microgravityThese data suggest that the extracellular environment suchas paracrine signals is an important factor and cannot beignored considering the impact of microgravity on vascularcells Since some P2 receptor artificial ligands are already

applied as drugs for cardiovascular patients specific P2receptor ligandsmight be reasonable candidates to investigat-ing their function for cardiovascular deconditioning undermicrogravity in the future


All authors declare that there are no conflicts of interest andagree with the contents of the paper

Acknowledgments

This work was supported by the Bundesministerium furBildung und Forschung- (BMBF-) FHprofUnt [FKZ03FH012PB2 to ET] NRW FH-Extra [FKZ z1112fh012 toET] DAAD PPP Vigoni [FKZ 314-vigoni-dr and FKZ54669218 to ET] and BMBF-AIF [FKZ 1720X06 to ET]the fellowship of YZ was funded by China ScholarshipCouncil [no 20100602024] and the Helmholtz Space LifeSciences Research School (SpaceLife) SpaceLife is funded inequal parts by the Helmholtz Association and the GermanAerospace Center (DLR)

References


[2] A R Hargens and S Richardson ldquoCardiovascular adaptationsfluid shifts and countermeasures related to space flightrdquo Respi-ratory Physiology and Neurobiology vol 169 pp S30ndashS33 2009

[3] M Coupe J O Fortrat I Larina G Gauquelin-Koch CGharib and M A Custaud ldquoCardiovascular deconditioning


from autonomic nervous system to microvascular dysfunc-tionsrdquo Respiratory Physiology and Neurobiology vol 169 ppS10ndashS12 2009

[4] E L Schiffrin ldquoThe endothelium and control of blood ves-sel function in health and diseaserdquo Clinical and InvestigativeMedicine vol 17 no 6 pp 602ndash620 1994


[6] C A Limbach M Lange M Schulze and E TobiaschldquoRecent patents on biomedical applications for the treatment ofatherosclerosisrdquo Recent Patents on RegenerativeMedicine vol 2no 2 pp 75ndash102 2012

[7] N Zippel C A Limbach N Ratajski et al ldquoPurinergicreceptors influence the differentiation of human mesenchymalstem cellsrdquo Stem Cells and Development vol 21 no 6 pp 884ndash900 2012

[8] Y Zhang D Khan J Delling and E Tobiasch ldquoMechanismsunderlying the osteo- and adipo-differentiation of humanmesenchymal stem cellsrdquoThe ScientificWorld Journal vol 2012Article ID 793823 14 pages 2012

[9] G Burnstock ldquoControl of vascular tone by purines and pyrim-idinesrdquo The British Journal of Pharmacology vol 161 no 3 pp527ndash529 2010

[10] G Burnstock ldquoPurinergic signallingrdquo British Journal of Phar-macology vol 147 no 1 pp S172ndashS181 2006

[11] G Burnstock ldquoPurine and pyrimidine receptorsrdquo Cellular andMolecular Life Sciences vol 64 no 12 pp 1471ndash1483 2007

[12] Y Zhang and E Y Tobiasch ldquoThe role of purinergic receptorsin stem cells in their derived consecutive tissuesrdquo in AdultStem Cell Standardization P di Nardo Ed pp 73ndash98 RiverPublishers 2011






[18] H Y Kang Y B Fan A Q Sun X L Jia and X Y DengldquoSimulated microgravity exposure modulates the phenotype ofcultured vascular smooth muscle cellsrdquo Cell Biochemistry andBiophysics vol 66 no 1 pp 121ndash130 2013

[19] C Griffoni S Di Molfetta L Fantozzi et al ldquoModificationof proteins secreted by endothelial cells during modeled lowgravity exposurerdquo Journal of Cellular Biochemistry vol 112 no1 pp 265ndash272 2011


[21] S M Schwartz ldquoSelection and characterization of bovine aorticendothelial cellsrdquo In Vitro vol 14 no 12 pp 966ndash980 1978

[22] M A Stepp M S Kindy C Franzblau and G E SonensheinldquoComplex regulation of collagen gene expression in culturedbovine aortic smooth muscle cellsrdquo The Journal of BiologicalChemistry vol 261 no 14 pp 6542ndash6547 1986

[23] P Eiermann S Kopp J Hauslage R Hemmersbach R Gerzerand K Ivanova ldquoAdaptation of a 2-D clinostat for simulatedmicrogravity experiments with adherent cellsrdquo MicrogravityScience and Technology vol 25 pp 153ndash159 2013

[24] L Wang L Karlsson S Moses et al ldquoP2 receptor expressionprofiles in human vascular smooth muscle and endothelialcellsrdquo Journal of Cardiovascular Pharmacology vol 40 no 6 pp841ndash853 2002

[25] C J Jackson and M Nguyen ldquoHuman microvascular endothe-lial cells differ from macrovascular endothelial cells in theirexpression of matrix metalloproteinasesrdquo International Journalof Biochemistry and Cell Biology vol 29 no 10 pp 1167ndash11771997

[26] S Zink P Rosen and H Lemoine ldquoMicro- and macrovascularendothelial cells in 120573-adrenergic regulation of transendothelialpermeabilityrdquo The American Journal of PhysiologymdashCell Physi-ology vol 269 no 5 pp C1209ndashC1218 1995



[29] F Shi Y C Wang T Z Zhao et al ldquoEffects of simulated micro-gravity on human umbilical vein endothelial cell angiogenesisand role of the PI3K-Akt-eNOS signal pathwayrdquo PLoS ONE vol7 no 7 Article ID e40365 2012



[32] V J Dzau R C Braun-Dullaeus and D G Sedding ldquoVascularproliferation and atherosclerosis new perspectives and thera-peutic strategiesrdquo Nature Medicine vol 8 no 11 pp 1249ndash12562002

[33] S Versari A Villa S Bradamante and J A M MaierldquoAlterations of the actin cytoskeleton and increased nitric oxidesynthesis are common features in human primary endothelialcell response to changes in gravityrdquo Biochimica et BiophysicaActa vol 1773 no 11 pp 1645ndash1652 2007

Review ArticleHuman Locomotion under Reduced Gravity ConditionsBiomechanical and Neurophysiological Considerations

Francesca Sylos-Labini12 Francesco Lacquaniti123 and Yuri P Ivanenko2

1 Centre of Space Bio-Medicine University of Rome Tor Vergata Via Montpellier 1 00133 Rome Italy2 Laboratory of Neuromotor Physiology IRCCS Santa Lucia Foundation Via Ardeatina 306 00179 Rome Italy3 Department of Systems Medicine University of Rome Tor Vergata Via Montpellier 1 00133 Rome Italy

Correspondence should be addressed to Yuri P Ivanenko yivanenkohsantaluciait

Received 24 April 2014 Accepted 12 June 2014 Published 28 August 2014


Copyright copy 2014 Francesca Sylos-Labini et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Reduced gravity offers unique opportunities to study motor behavior This paper aims at providing a review on current issues ofthe known tools and techniques used for hypogravity simulation and their effects on human locomotion Walking and runningrely on the limb oscillatory mechanics and one way to change its dynamic properties is to modify the level of gravity Gravity has astrong effect on the optimal rate of limb oscillations optimal walking speed andmuscle activity patterns and gait transitions occursmoothly and at slower speeds at lower gravity levels Altered center of mass movements and interplay between stance and swingleg dynamics may challenge new forms of locomotion in a heterogravity environment Furthermore observations in the lack ofgravity effects help to reveal the intrinsic properties of locomotor pattern generators and make evident facilitation of nonvoluntarylimb stepping In view of that space neurosciences research has participated in the development of new technologies that can beused as an effective tool for gait rehabilitation

1 Introduction

Life evolved in the presence of gravity which has two majorimpacts on motor functions specific body orientation inspace and antigravity muscle tone and specific rules ofmotion in the gravity field Gravity plays an essential role interrestrial locomotion The dominant hypothesis regardingtemplates for bipedal walking in the gravity field is thependular mechanism of walking up to intermediate speedsand the bouncing mechanism of running up to the highestspeeds attainable [1]The inverted pendulum-likemechanismof energy exchange taking place during walking would beoptimized at slower speeds in reduced gravity [2 3] Despiteour intuitive appreciation for the influence of gravity wedo not fully understand how gravity interacts with otherforces such as inertia to affect many biological and physicalprocesses and what type of gait andor limb synchronization(trot gallop lateral sequencewalk pace skipping etc) wouldevolve at other gravity levels

Understanding locomotion characteristics is critical forthose working in the area of gait biomechanics and neu-rophysiology f pattern generation networks and of exer-cise countermeasures for astronauts Many researchers haveinvestigated the effects of reducing and eliminating gravityon locomotive kinematics and kinetics [4ndash8] Others havestudied locomotion in actual weightlessness or hypogravity[9 10] The techniques have included supine and erect cablesuspension parabolic aircraft flights water immersion andcentrifugal methods [6] Increased knowledge of locomotionkinematics kinetics muscular activity patterns and sensoryfeedback modulation may help to facilitate more effectiveexercise countermeasures develop innovative technologiesfor gait rehabilitation and provide new insights into ourunderstanding of the physiological effects of gravity In thisreview we will consider the known tools and techniquesused for hypogravity simulation and their effects on humanlocomotion



2 Methods and Apparatuses for ReducedGravity Simulation

Spaceflights are the more direct way to assess the effect ofgravity on locomotion but studying locomotion in actualhypogravity is demanding and expensive [6] The drawbacksto spaceflight experiments include difficulty in using neces-sary data collection hardware and performing an experimentwith adequate sample size Parabolic flight offers a viablealternative but periods of weightlessness are limited to sim20 swhich only allows for acute locomotion investigations [11]

There are several apparatuses that have been used inthe past to simulate reduced gravity locomotion One ofthe more used systems is the vertical body weight support(BWS) (Figures 1(a) and 1(b)) These kinds of simulators areusually obtained supporting the subjects in a harness thatapplies a controlled upward force For example the WARD[12] mechanism consists of a mechanical gear driven by apneumatic cylinder (Figure 1(b)) It is held in a cart that slidesforward and backward over a track Low-friction sliding ofthe mechanism ensures that only vertical forces are appliedto the subject Vertical BWS systems may also make use ofa small increase in air pressure around the userrsquos lower bodyto create a lifting force approximately at the personrsquos centerof mass [13] Other vertical systems [8 14] use a series ofcompliant rubber spring elements that are stretched to createthe upward (to simulate gravity less than 1 g) or downward(to simulate gravity greater than 1 g) force (Figure 1(a))The main limitation of these reduced gravity simulators (inaddition to high local skin pressure via a harness) is that eachsupporting limb experiences a simulated reduction of gravityproportional to the applied force while the swinging limbexperiences 1 g

The tilted BWS systems (Figures 1(c) and 1(d)) are con-structed to simulate more realistic effects of gravity changeson both the stance and swing legs in the sagittal planeThese simulators that have been used in the past by bothRoscosmos (Russian Federal Space Agency) and NASA totrain astronauts before space flights [15ndash17] are based onthe idea of neutralizing the component of the gravity forcenormal to the lying surface [mg sdot cos(120572) where 120572 is theangle of inclination] while the component of the gravityforce acting on the body and swinging limbs in the sagittalplane is reduced in relation to the tilt angle [mg sdot sin(120572)]A similar concept has been used in the reduced gravitysimulator (Figure 1(d)) designed by Ivanenko et al (Italianpatent number Rm2007A000489) the subject lies on the sideon a tilted couch (up to 40∘ from the horizontal position)with both legs suspended in the exoskeleton and steps onthe treadmill which is tilted to the same angle [7 18 19]This simulator included additional mass of the tilted chassis(sim15 kg) and exoskeleton (15 kg for each leg) Thus theentire assembly had a mass of sim18 kg that increased bothgravitational and inertial forces during walking

Another class of gravity-related manipulations is ldquosubjectload devicerdquo (SLD) that applies a gravity replacement forcein the direction down to the surface This type of SLD can beused in the vertical systems to increase the gravity [8] or in thelying position (Figure 1(e)) When an astronaut walks or runs

on a treadmill in weightlessness a subject load device is usedto return him or her back to the treadmill belt and to loadthe limbs The gravity replacement load is transferred via aharness to the pelvis andor the shoulders Gravity simulatorscan simulate active treadmill running in weightlessness andprovide a method of testing proposed improvements in SLDdesign and exercise protocols [20 21] In supine suspensionsystems (Figure 1(e)) subjects are suspended horizontallyattached to latex rubber cords A cloth sleeve and rubber cordare attached each to the upper and lower arms and legs (eighttotal) [20] The limitation of this system is a local pressureon some parts of the body (eg shoulders) andmodificationsin the swing phase dynamics due to nonconstant forcesof rubber cords and gravity acting in the anterioposteriordirection of leg movements (Figure 1(e))

Based on the passive gravity balancing technology Ma etal [22 23] proposed a design concept of a passive reducedgravity simulator to simulate human walking or other activ-ities in a reduced-gravity environment for potential applica-tions of training astronauts and space travelers (Figure 1(f))The system consists of a 3-DOF dual parallelogram mech-anism a 2-DOF torso support assembly and a pair of 3-DOF leg exoskeletons The weight of the body and the legsis compensated by the spring-balanced dual-parallelogrammechanism and torso-support assembly and the weight ofeach leg is compensated by a leg exoskeleton The systemis capable of simulating human walking and jumping in ahypogravity environment [24] Hardware prototyping andexperimental study of the new system are currently under-way

In the following section we discuss the basic principles ofadaptation of locomotion to different gravity values using thetechnologies described here

3 Biomechanical Aspects of Locomotion inReduced Gravity

Despite some differences all reduced gravity simulationapproaches show a reasonable approximation of the reduc-tion in the gravitational force acting on the center of bodymass (COM) and similar results concerning the speed ofgait transitions An important consequence of the pendulum-like behavior of the limbs in the gravity field is the principleof dynamic similarity [29] which states that geometricallysimilar bodies that rely on pendulum-like mechanics ofmovement have similar gait dynamics at the same Froudenumber

Fr sim 1198812

119892119871 (1)

where 119881 is the speed of locomotion 119892 is the accelerationof gravity and 119871 is a characteristic leg length That is alllengths times and forces scale by the same factors Inorder to optimize the recovery of mechanical energy thekinetic energy and the potential energy curves must beequal in amplitude and opposite in phase as in a pendulumAssuming that the change in kinetic energy within each step


PL

R

R

M

B 20 m

(a) Vertical system for altered grav-ity simulation

F

(b) Vertical BWS

(c) Tilted BWS

120572

(d) Tilted BWS

D

E

F

G

CB

A

(e) Supine suspension system

q2

q1

q3

q4

q6

q5

q11 q12

q78

q910

(f) Passive gravity balancing system

Figure 1 Reduced gravity simulators for locomotion (a) Schema of the vertical system used to simulate different gravity values (redrawnfrom [8]) R rubber bands B light metal bars M electric motor to stretch the elastic band system PL pulleys to invert the direction ofthe pull on the subject (dashed lines) (b) Vertical body weight support (BWS) system subject walks on a treadmill with different levels ofBWS while being supported in a harness pulled upwards by a preset unloading force 119865 (c) Tilted BWS system used by Roscosmos (RussianFederal Space Agency) to train astronauts before space flights [15] the subject walks on a truncated cone (60m height 92∘ inclination relativeto the vertical) supported by five ropes sustaining the head trunk and legs (picture portraying Professor Gurfinkel reproduced with his kindpermission) (d) Tilted unloading system for stepping on a treadmill the subject lies on the side on a tilted couch (up to 40∘ from the horizontalposition) with both legs suspended in the exoskeleton and steps on the treadmill which is tilted to the same angle The component of thegravity force acting on the stance and swing limb segments is proportional to the tilting angle 120572 [18] (e) Supine suspension system (adaptedfrom [20] courtesy of Professor Peter Cavanagh) the subject is suspended horizontally attached to latex rubber cords A cloth sleeve andrubber cord are attached each to the upper and lower arms and legs (eight total) The subject is actively pulled toward the treadmill by agravity replacement load through cables attached to a load splitter (f) Passive reduced gravity walking simulator (courtesy of Dr Ou Ma)The system consists of a 3-DOF dual parallelogram mechanism a 2-DOF torso support assembly and a pair of 3-DOF leg exoskeletons Theweight of the body and the legs is compensated by the spring-balanced dual-parallelogram mechanism and torso-support assembly and theweight of each leg is compensated by a leg exoskeleton [22ndash24]

is an increasing function of the walking speed (while thechange in the potential energy is proportional to gravity)the hypothesis was proposed that the inverted pendulum-like mechanism of energy exchange during walking wouldbe optimized at slower speeds in reduced gravity [3 10] An

optimal exchange between potential and kinetic energies ofthe COM occurs at Fr sim 025 [2] (Figure 2(a)) Even thoughspecific limb segment proportions may play an essential rolein the kinematics and energetics of walking [30] animalanatomy and individualized limb segment dimensions are


Margaria and Cavagna [3] Cavagna et al [68]Cavagna et al [10]Griffin et al [67]

Kram et al [4]Ivanenko et al [18]

0 05 10

1

2

15Gravity (g)

Wal

king

spee

d (m

s)

Walk-runtransitionFr = 05

Optimal speed Fr = 025

(a)

0

20

40

60

F

BW

andashpmndashl

0

4

2

0

20

40

60

1 g065 g 05 g

025 g 005 g

005 g

05 g

1 g

(b)

Simulated locomotion on Moon (016 g)

Gravity replacement load mg

Tilted BWS Vertical BWS

Horiz foot excursion Max horiz foot vel

Tilted BWS systemVertical BWS system

Gravity replacement load Vertical BWS system(matched foot excursion)

0

20

40

60

80

(cm

)

0

2

4(km

h)

100 10

6

8

x

1 g 1 g

Fp = 016middotmgFg

120572

Fvert = 084middotmg

(c)Figure 2 Biomechanical features of locomotion in reduced gravity conditions (a) Optimal (blue) and walk-to-run transition (green) speedsas a function of gravity Dynamically similar speeds predicted by Fr = 025 and Fr = 05 are indicated by blue and green dashed curvesrespectively [25] Green circles and stars refer to measurements of optimal walk-to-run transition speeds in simulated low-gravity conditions[5 18]The grey triangle indicates an earlier estimate of optimal walking speed predicted for theMoon gravitational environment byMargariaand Cavagna [3] Blue triangles refer to the optimal speeds (at which most of the mechanical exchange between potential and kinetic energyof the body center of mass occurs) obtained in a simulation study of Griffin et al [26] Blue circles represent measurements of optimal speedobtained during parabolic flight [10 27] (b) Time course of the net vertical component of in-shoe reaction forces plotted as a function ofthe spatial coordinates of the foot at different reduced gravity levels Note change in vertical scale in the 005 g condition The lower rightpanel shows the trajectories of the center of pressure superimposed on a foot outline (adapted from [28]) (c) Maximum longitudinal footvelocity and foot excursion (119909) during walking at 2 kmh at 016 g using three different reduced gravity simulators (represented schematicallyin the upper panels) Horizontal dashed lines indicate values for walking at 1 g The hatched bar (right panel) corresponds to the maximalfoot velocity for the vertical BWS system approximated by matching the foot excursion to that of the tilted BWS system [7] Note significantlylower foot velocities during swing using tilted BWS systems


optimized in such a way that the Froude number can explainoptimal walking velocity

On Earth walking and running gaits are usually adoptedfor different speeds of locomotion with a preferred transitionoccurring at sim2ms for human adults and at slow speeds forchildren (Frsim 05) in accordance with the dynamic similaritytheory [29] Different studies [4 18] demonstrated that atlower levels of gravity the walk-run transition occurred atprogressively slower absolute speeds but at approximately thesame Froude number (Figure 2(a))

Despite similarities in approximating reduced gravitythere are nevertheless essential differences between differentsimulation approaches The variables that showed the great-est differences between vertical and tilted reduced gravitysystems (Figure 1) were maximal longitudinal foot velocityand longitudinal foot excursion (Figure 2(c)) in agreementwith significant influences of gravity on swing leg dynamics[7] Even though the maximal longitudinal foot velocity forthe tilted BWS condition decreased only slightly relative tothe vertical BWS however the actual decrement was muchmore obvious if one takes into account that it was significantlycompensated for ormasked by increments in the stride length[7] A previous modeling study also predicted differentialeffects of gravity during stance and swing phases [31] In factthe changes in the longitudinal foot excursion were basicallyopposite for the vertical and tilted BWS systems (Figure 2(c))For the former system the amplitude of longitudinal footmotion decreased while for the latter system it increasedrelative to the 1 g condition Considering a monotonic (pre-sumably proportional [32]) relationship between the stridelength and the maximal foot velocity at a given gravity level(1 g) the peak foot velocity would be expected to be sim15times higher for the vertical than for tilted BWS conditionif the stride lengths were similar (Figure 2(c)) The previousstudies on parabolic flights investigating the effect of gravityon walkingmechanics demonstrated increments in the swingphase duration (by 29 at 025 g [33] see also [11]) in linewith the substantial contribution of gravity to the swing legOverall the findings demonstrate that gravity acting on bothstance and swing legs plays an important role in shapinglocomotor patterns

4 Nonlinear Reorganization of EMG Patterns

It is known that load plays a crucial role in shaping patternedmotor output during stepping [34ndash36] and humans producea specific heel-to-toe rolling pattern during stance in normalgravity conditions Ground contact forces reflect the netvertical and shear forces acting on the contact surface andresult from the sum of the mass-acceleration products ofall body segments while the foot is in contact with ground[37] Simulating reduced gravity between 005 and 1 g revealsdrastic changes of kinetic parameters but limited changesof the kinematic coordination [28] The reported accuratecontrol of limbfoot kinematics [28]maydependon load- anddisplacement-compensationmechanisms working effectivelythroughout a wide range of ground contact forces from fullbody weight up to lt5 of its value The peak vertical contactforces decrease proportionally to gravity but at 005 g they are

applied at the forefoot only (Figure 2(b)) During lower limbloading a variety of receptors can be activated such as Golgitendon organs cutaneous receptors of the foot and spindlesfrom stretched muscles [36] These sensory signals interactwith central rhythm-generating centers and help in shapingthe motor patterns controlling phase-transitions and rein-forcing ongoing activity [38 39] For instance loading ofthe limb enhances the activity in antigravity muscles duringstance and delays the onset of the next flexion [40] It isimportant to understand the mechanisms of sensorimotoradaptation to the biomechanics of locomotion and footplacementloading in heterogravity especially to longer-termchanges of load

A key feature of adaptation to hypogravity is a remarkablenonlinear scaling of muscle activity patterns contrary tomonotonic changes in foot loading The simplest kind ofchange with simulated reduced gravity [28] was seen inankle extensors the mean amplitude of activity decreasedsystematically with decreasing simulated gravity consistentwith their antigravity function [35 41] By contrast thebehavior of other muscles could not be predicted simply onthe basis of the static load during stance The amplitude andpattern of muscle activity generally depended on speed andcould vary nonmonotonically with body unloading Therewas also a complex reorganization of the pattern of activityof thigh muscles with decreasing simulated gravity as well asnoteworthy individual differences [28] Figure 3(a) illustratesan example of nonlinear reorganization of EMG patterns inone subject walking at 3 kmh With body weight unloadinggluteus maximus and distal leg extensors decreased theiractivity while other muscles demonstrated a ldquoparadoxicalrdquoincrement of activation (eg quadriceps) or considerablechanges in the activation waveforms (hamstring muscles)Note also the absence of the typical burst of RF at thebeginning of the swing phase at low simulated gravity levels(Figure 3(a)) consistent with other studies on the effect ofbody weight unloading [42] and walking speed [43] It isunlikely that these changes are due to the order of trials orthe consequence of learning the hypogravity condition sincepresentation order of speeds andBWSwas randomized acrosssessions and experiments [28] Also the duration of each trialwas sim1min with at least 2 min rest between trials and ashort (sim30 s) training period of walking at different speedswas allowed for each simulated reduced gravity level beforethe actual data collection was begun (the walking patternstypically adapt rapidly to simulated reduced gravity [4 5])This reorganization is presumably related to the multifunc-tional (biarticular) action of these muscles and to the needto repartition the joint torque contributions across differentmuscles as a function of the changes induced by gravity At1 g the main peak of m biceps femoris activity occurringbefore heel-contact serves to decelerate the swinging limb[37] However as gravity is decreased its main activity occursin mid-stance and late stance presumably in relation to theneed to assist vaulting over an inverted pendulum of thestance limb and swing initiation

There might be various factors accounting for the non-linear reorganization of muscle activity patterns with gravityTo start with nonlinear scaling also occurs during walking


005 g

GM VL RF

0 20 40 60 80 100Cycle () Cycle () Cycle ()

0 20 40 60 80 100 0 20 40 60 80 100

025 g

20 120583

V30

120583V

50 120583

V

50 120583

V

30 120583

V

30 120583

V

05 g065 g

1 g

005 g 025 g

05 g065 g

1 g

BF TA LG

(a)

Speed5 kmh 3 kmh

2 kmh 11 kmh

012345

GM

(120583V

)

0 02 04 06 08 111235

0 5

10 15 20

BF

g

Speed (kmh)

(120583V

)

010203040

TA

(120583V

)

05

101520

LG(120583

V)

02468

10

VL

(120583V

)

02468

10

RF

(120583V

)

(b)

Figure 3 Nonlinear reorganization of muscle activity patterns (a) An example of ensemble-averaged electromyographic (EMG) activity oflower limb muscles versus the normalized gait cycle is shown for a subject walking at 3 kmh at different simulated reduced gravity levels[28] (b) Mean EMG activity computed over the gait cycle and averaged across all cycles and subjects (119899 = 8) For each muscle values fortrials performed at each speed are plotted as a function of simulated reduced gravity (adapted from [28]) GM gluteus maximus VL vastuslateralis RF rectus femoris BF biceps femoris TA tibialis anterior LG and lateral gastrocnemius

at different speeds at 1 g For instance VL and RF activity isquite small at low speeds (less than sim3 kmh) but becomesprominent at higher speeds (gt4 kmh) (Figure 3(b)) a speedeffect consistent with that reported in the literature [28 4345 46] Given that it should be stressed that walking at lowergravity levels at the same speed (Figure 3(a)) corresponds towalking at higher speeds if one uses the Froude number as adimensionless parameter (eg walk-run transition at 025 g

occurs at sim4 kmh Figure 2(a)) so that ldquoparadoxicalrdquo incre-ments of VL and RF EMG activity in Figure 3(a) may reflecthigher biomechanical demands on proximal leg muscles athigher dimensionless speeds Nonlinear reorganization ofEMG patterns was also observed when using exoskeletonrobotic devices that provide body weight support [42 47]Changes in the body reference configuration during stance(slightly flexed posture [48 49]) may contribute to a greater


SOL

Walking speed

Time (s)

0

1 g

0 50 100 150 200 250 300

W-R R-W

0 50 100 150

Time (s)

350

0

048

(km

h)

0

100Cycle

()

Cycle ()0 1001 s

Right

Left100 120583V

600 120583VCycle ()

0 1001 s RightLeft

25 120583V

016 g

Δ

(a)

Mars

Spee

d (k

mh

)

0

2

4

6

8

0 02 04 06 08 1 12

10

Fr ~ 05

PlutoMoon

Walking

Abrupt switch

Smooth transitions

Running

Earth

Simulated gravity gEarth

Fr ~ 05

(b)

Figure 4 Smoothnessabruptness of gait transitions at different gravity levels (a) Soleus (SOL) EMG patterns during slow changes intreadmill belt speed (lower panels) in one representative subject at 016 g (left) and 1 g (right)Upper panels examples of SOL EMGwaveforms(left plotted versus time right plotted versus normalized cycle) during 5 consecutive strides of both legs around the transition from walking(black lines) to running (gray lines) Dotted curves denote the (transition) stride of the leg in which the swing phase first exceeded 50gait cycle Bottom horizontal bars denote stance (black) and swing (white) phases Lower panels the color maps represent a sequence ofdiscrete activation waveforms (vertical slices) 119909-axis indicates the number of the gait cycles (corresponding to the appropriate timing of thetrial) 119910-axis indicates normalized gait cycle (from touchdown to another touchdown) and color indicates EMG amplitude The white lineindicates when toe off occurred Vertical dashed lines indicate walk-to-run (W-R) and run-to-walk (R-W) transitions Note abrupt changesin the relative stance duration and muscle activation patterns at gait transitions at 1 g and no obvious distinction in these parameters at thetransition from walking to running at 016 g (b) Schematic representation of the smoothness of gait transitions as a function of gravity Theorange curve symbolizes the dimensionless walk-run transition speed consistent with the theory of dynamic similarity (Fr sim 05) [19 29 44]The blue color range of gravitational levels represents a discontinuous switch from walk to run whereas the white region indicates smoothtransitions

activity of proximal extensors as well Finally there is adifferential effect of speed on quadriceps muscle activity atreduced gravity levels VL and RF activity increases at lowspeeds (lt3 kmh) while it decreases at a high speed (5 kmh)(Figure 3(b)) Potential nonlinear scaling of muscle activityfor most whole body movements in microgravity shouldalso be taken into account for exercise countermeasures forastronauts

5 Different Gaits

Considering complex high-dimensional dynamically cou-pled interactions between an organism and gravitationalenvironment in principle one challenging solution is toadopt different coordination patterns and not only an optimalspeed of locomotion Are different gaits possible on otherplanets

One approach to study locomotor adaptations is to lookat the effect of gravity on gait transitions A gait has beendefined as ldquoa pattern of locomotion characteristic of a limitedrange of speeds described by quantities of which one or morechange discontinuously at transitions to other gaitsrdquo [29]An important aspect of gait transitions is a discontinuousswitch that occurs at some point while varying the speedof progression (although some exceptions may exist [50ndash52]) As already discussed (Figure 2(a)) gravity has a strong

effect on the speed at which gait transitions occur (Fr sim05) Surprisingly however we found [18 19] that at lowerlevels of simulated gravity the transition betweenwalking andrunningwas generally gradual without any noticeable abruptchange in gait parameters or EMG bursts (Figure 4(a)) Thiswas associated with a significant prolongation of the swingphase whose duration became virtually equal to that of stancein the vicinity of the walk-run transition speed and witha gradual shift from inverted-pendulum gait (walking) tobouncing gait (running) A lack of discontinuous changes inthe pattern of speed-dependent locomotor characteristics in ahypogravity environment (Figure 4(b)) is consistent with theidea of a continuous shift in the state of a given set of centralpattern generators rather than the activation of a separate setof central pattern generators for each distinct gait [19]

Interestingly the smoothness of gait transitions is accom-panied by a gradual shift from inverted-pendulum gaitto bouncing gait resulting in a ldquoparadoxicalrdquo inverted-pendulum running in the vicinity of run-walk and walk-runtransitions [18]The swing phasemay havemore influence ongait than it was previously thought For instance relativelyslower swing and longer foot excursions (tilted BWS con-dition Figure 2(c)) may raise questions about optimality orcomfort ofwalking and could account for potentially differentpreferred gaits such as loping on the Moon observed inApollo astronauts (though the Lunar suit limits the range


(a)

25

20

15

10

5

0

10 13 16 19 22 25

10

5

0

Predicted available impulse (N s)

Pred

icte

d av

aila

ble

impu

lse (N

s)Successful subjects

Succ

essfu

l sub

ject

s

MoonEnceladusEuropa Io

Simulated gravity gEarth ()

(b)

Figure 5 Running on water at simulated reduced gravity The blue curve represents the net vertical impulse available to run on water aspredicted by the model used by Minetti et al [53] Bars represent the number of subjects out of 6 capable of avoiding sinking at differentsimulated gravity values Both variables show that 22 of Earth gravity (119892EARTH) is the maximum gravity at which humans can run on waterwhen assisted by a small rigid fin (as illustrated in the left panel)

of motion in the leg joints and may also contribute to theloping gait on the Moon [9]) The resulting changes in theintersegmental and interlimb coordinationmay in turn affectthe COM motion Overall the results support the idea oflooking for new forms of locomotion (both bipedal andquadrupedal) in a heterogravity environment [54] based onthe interplay between stance and swing leg dynamics alteredinterlimb coupling and altered center of mass movements

Other significant influences of gravity on short-term andlong-term gait adaptations may be related to its effects onthe body reference configuration [48 49] and anticipatorymechanisms of limb and body movements [55 56] Forinstance the basis of habitual human posture is postural toneof the skeletal muscles and microgravity elicits substantialchanges in muscle tone and posture [48 49] Based onclinical observations it has been recently argued that anyreflection on the nature and choice of preferred gait (egbipedal versus quadrupedal) should include a considerationof the mechanisms determining the choice of unconscioushabitual posture [57] Also in analogy with the resultsbased on upper-limb movements related to time-to-contact[55] or movement planning [58] anticipatory postural andlocomotor adjustments for lower limb movements (eg forthe control of heel strike or accurate foot placement) shouldtake gravity into consideration Therefore altered gravityconditions may also affect locomotor-related tasks such asthe negotiation of stationary andmoving obstructions duringwalking or gait initiationtermination [56 59 60]

Finally the repertoire of known gaits can be expandedto a variety of animals For instance on Earth only afew legged species such as water strider insects and someaquatic birds and lizards can run on water For most other

species including humans this is precluded by body sizeand proportions lack of appropriate appendages and limitedmuscle power However if gravity is reduced to less thanEarthrsquos gravity running on water should require less musclepower Recently Minetti et al [53] used this hydrodynamicmodel of Glasheen and McMahon [61] to predict the gravitylevels at which humans should be able to run on water andtested the hypothesis in the laboratory using a reduced gravitysimulator (Figure 5)The results showed that a hydrodynamicmodel of Basilisk lizards running on water [61] can also beapplied to humans despite the enormous difference in bodysize andmorphology Particularly 22of Earthrsquos gravity is themaximum at which humans can run on water when assistedby a small rigid fin (Figure 5) [53] It is also worth notingthe limitations for our musculoskeletal system for producingforcepower (endurance) for instance the stride frequencyin humans is limited to about 2Hz whatever the planet isOn Earth the biggest animal that can run on water is likelyWestern Grebes and even these birds can run only for severalseconds since the force production is basically anaerobic(participants in [53] could run at simulated ldquoMoonrdquo gravityonly for sim10 s) In contrast at reduced gravity (Moon) thesebirds could run on water in a charming manner for muchlonger time

6 Clinical Implications

Reduced gravity also offers unique opportunities for adjust-ing the basic patterns to altered locomotor conditions forgait rehabilitation Bodyweight support systems coupledwithrobotic devices or pharmacologic treatments are now oftenused in the rehabilitation practice to assist physical therapy of


ES of peroneal nerve

RF

BF

TA

LG

Hip

Knee

Ankle

Delay

Flex

Ext

5 s

300 120583V

30∘

30∘

15∘

(a)

Hand-walking

Hand

Foot

DELTa

FCU

BIC

ST

BF

5 s

50 cm

30 cm

100 120583V

800 120583V

200 120583V

(b)

Figure 6 Eliciting nonvoluntary limb stepping movements in simulated weightlessness (gravity neutral) conditions (a) An example ofnonvoluntary rhythmic movements of the suspended legs induced by electrical stimulation (ES) of peroneal nerve from the study of Selionovet al [62] Note the absence of ankle joint rotations during evoked air-stepping (b) An example of evoked rhythmic leg movements duringhand walking in one subject from the study of Sylos-Labini et al [63] RF rectus femoris BF biceps femoris TA tibialis anterior LG lateralgastrocnemius FCU flexor carpi ulnaris BIC biceps brachii DELTa anterior deltoid ST and semitendinosusHand and foot denote anterior-posterior displacements of the left hand and foot

individuals with neurological disorders We will not reviewany detailed analysis of clinical outcomes for ambulationwhen using locomotor training with body weight supportsystems and refer to other reviews [64] Nevertheless it isworth emphasizing a facilitatory effect of the lack of gravityon rhythmogenesis and its potential for gait recovery

Novel pharmacological strategies [65] and electromag-netic stimulation techniques [62 66ndash68] are being developedaimed at modulating spinal activity and restoring the loco-motor function The spinal central pattern generator (CPG)circuitry can be easily activated in healthy humans in a gravityneutral position by applying tonic central and peripheralsensory inputs To minimize interference with the ongoingtask of bodyweight and balance control steppingmovementsare elicited during air-stepping in the absence of gravityinfluences and external resistance Figure 6 illustrates exam-ples of nonvoluntary rhythmic movements of the suspendedlegs induced by electrical stimulation of peroneal nerve [62]and during hand walking [63] It has been suggested thatfunctional multisensory stimulations and a functional neuralcoupling between arm and legs can inspect CPG access bysensory and central activations and entrain locomotor neuralnetworks and promote gait recovery Such investigations may

contribute to the clinical development of central patterngenerator-modulating therapies and neuroprosthetic tech-nologies [65 69]


This perspective outlines an interdisciplinary approach toextend our knowledge on adaptation of human locomotionto a hypogravity environment including biomechanical neu-rophysiological and comparative aspects effective exercisecountermeasures for astronauts and even exobiology ofnew forms of locomotion on different planets The toolsand techniques used for hypogravity simulation and theireffects on human locomotion provide new insights into ourunderstanding of the physiological effects of gravity Thebeneficial effect of weightlessness on rhythmogenesis wouldfurther enhance the utility of this approach and developmentsof innovative technologies for gait rehabilitation




Acknowledgments

This work was supported by the Italian Health MinistryItalian Ministry of University and Research (PRIN project)and Italian Space Agency (DCMC CRUSOE and COREAgrants)

References

[1] R J Full and D E Koditschek ldquoTemplates and anchorsneuromechanical hypotheses of legged locomotion on landrdquoThe Journal of Experimental Biology vol 202 no 23 pp 3325ndash3332 1999

[2] A E Minetti ldquoInvariant aspects of human locomotion indifferent gravitational environmentsrdquo Acta Astronautica vol49 no 3ndash10 pp 191ndash198 2001

[3] R Margaria and G A Cavagna ldquoHuman locomotion insubgravityrdquo Aerospace Medicine vol 35 pp 1140ndash1146 1964

[4] R Kram A Domingo and D P Ferris ldquoEffect of reducedgravity on the preferred walk-run transition speedrdquoThe Journalof Experimental Biology vol 200 no 4 pp 821ndash826 1997

[5] J M Donelan and R Kram ldquoThe effect of reduced gravity onthe kinematics of human walking a test of the dynamic simi-larity hypothesis for locomotionrdquo The Journal of ExperimentalBiology vol 200 no 24 pp 3193ndash3201 1997

[6] B L Davis and P R Cavanagh ldquoSimulating reduced gravity areview of biomechanical issues pertaining to human locomo-tionrdquo Aviation Space and Environmental Medicine vol 64 no6 pp 557ndash566 1993

[7] F Sylos-Labini Y P Ivanenko G Cappellini A Portone MJ Maclellan and F Lacquaniti ldquoChanges of gait kinematicsin different simulators of reduced gravityrdquo Journal of MotorBehavior vol 45 no 6 pp 495ndash505 2013

[8] G A Cavagna A Zamboni T Faraggiana and R MargarialdquoJumping on the moon power output at different gravityvaluesrdquo Aerospace Medicine vol 43 no 4 pp 408ndash414 1972

[9] C E Carr and J McGee ldquoThe apollo number space suits self-support and the walk-run transitionrdquo PLoS ONE vol 4 no 8Article ID e6614 2009

[10] G A Cavagna P A Willems and N C Heglund ldquoThe role ofgravity in human walking pendular energy exchange externalwork and optimal speedrdquo Journal of Physiology vol 528 part 3pp 657ndash668 2000

[11] J K De Witt G P Perusek B E Lewandowski et al ldquoLoco-motion in simulated and real microgravity horizontal Suspen-sion vs parabolic flightrdquo Aviation Space and EnvironmentalMedicine vol 81 no 12 pp 1092ndash1099 2010

[12] F Gazzani A Fadda M Torre and V Macellari ldquoWARD apneumatic system for body weight relief in gait rehabilitationrdquoIEEE Transactions on Rehabilitation Engineering vol 8 no 4pp 506ndash513 2000

[13] A M Grabowski and R Kram ldquoEffects of velocity and weightsupport on ground reaction forces and metabolic power duringrunningrdquo Journal of Applied Biomechanics vol 24 no 3 pp288ndash297 2008

[14] J P He R Kram and T A McMahon ldquoMechanics of runningunder simulated low gravityrdquo Journal of Applied Physiology vol71 no 3 pp 863ndash870 1991

[15] V A Bogdanov V S Gurfinkel and V E Panfilov ldquoHumanmotion under lunar gravity conditions (Human performancein various locomotive tasks under simulated lunar reduced

gravity conditions classifying test stands and equipment)rdquoKosmicheskaya Biologiya i Meditsina vol 5 pp 3ndash13 1971

[16] J R Hansen Spaceflight Revolution NASA Langley ResearchCenter from Sputnik to Apollo National Aeronautics and SpaceAdministration Washington DC USA 1995

[17] D E Hewes ldquoReduced-gravity simulators for studies of manrsquosmobility in space and on the moonrdquoHuman Factors vol 11 no5 pp 419ndash431 1969

[18] Y P Ivanenko F Sylos-Labini G Cappellini V MacellariJ McIntyre and F Lacquaniti ldquoGait transitions in simulatedreduced gravityrdquo Journal of Applied Physiology vol 110 no 3pp 781ndash788 2011

[19] F Sylos-Labini Y P Ivanenko G Cappellini S Gravanoand F Lacquaniti ldquoSmooth changes in the EMG patternsduring gait transitions under body weight unloadingrdquo Journalof Neurophysiology vol 106 no 3 pp 1525ndash1536 2011

[20] K O Genc V E Mandes and P R Cavanagh ldquoGravity replace-ment during running in simulatedmicrogravityrdquoAviation Spaceand Environmental Medicine vol 77 no 11 pp 1117ndash1124 2006

[21] J L McCrory H A Baron S Balkin and P R CavanaghldquoLocomotion in simulated microgravity gravity replacementloadsrdquo Aviation Space and Environmental Medicine vol 73 no7 pp 625ndash631 2002

[22] Q Lu C Ortega and O Ma ldquoPassive gravity compensationmechanisms technologies and applicationsrdquo Recent Patents onEngineering vol 5 no 1 pp 32ndash44 2011

[23] OMa and JWang ldquoApparatus andmethod for reduced-gravitysimulationrdquo 2012

[24] Q Lu J McAvoy and O Ma ldquoA simulation study of a reduced-gravity simulator for simulating human jumping and walkingin a reduced-gravity environmentrdquo in Proceedings of the ASMEDynamic Systems and Control Conference 2009

[25] A E Minetti ldquoWalking on other planetsrdquo Nature vol 409 no6819 pp 467ndash469 2001

[26] T M Griffin N A Tolani and R Kram ldquoWalking in simulatedreduced gravity mechanical energy fluctuations and exchangerdquoJournal of Applied Physiology vol 86 no 1 pp 383ndash390 1999

[27] G A Cavagna P A Willams and N C Heglund ldquoWalking onmarsrdquo Nature vol 393 no 6686 article 636 1998

[28] Y P Ivanenko R Grasso V Macellari and F LacquanitildquoControl of foot trajectory in human locomotion role ofground contact forces in simulated reduced gravityrdquo Journal ofNeurophysiology vol 87 no 6 pp 3070ndash3089 2002

[29] R Alexander McN ldquoOptimization and gaits in the locomotionof vertebratesrdquo Physiological Reviews vol 69 no 4 pp 1199ndash1227 1989

[30] F Leurs Y P Ivanenko A Bengoetxea et al ldquoOptimal walkingspeed following changes in limb geometryrdquo The Journal ofExperimental Biology vol 214 part 13 pp 2276ndash2282 2011

[31] D A Raichlen ldquoThe effects of gravity on human walking anew test of the dynamic similarity hypothesis using a predictivemodelrdquoThe Journal of Experimental Biology vol 211 no 17 pp2767ndash2772 2008

[32] Y Osaki M Kunin B Cohen and T Raphan ldquoThree-dimensional kinematics and dynamics of the foot duringwalking a model of central control mechanismsrdquo ExperimentalBrain Research vol 176 no 3 pp 476ndash496 2007

[33] J F Roberts ldquoWalking responses under lunar and low gravityconditionsrdquo AMRL-TR 6570th Aerospace Medical ResearchLaboratory 1963


[34] S H Scott and D A Winter ldquoBiomechanical model of thehuman foot kinematics and kinetics during the stance phase ofwalkingrdquo Journal of Biomechanics vol 26 no 9 pp 1091ndash11041993

[35] S J Harkema S L Hurley U K Patel P S Requejo B HDobkin and V R Edgerton ldquoHuman lumbosacral spinal cordinterprets loading during steppingrdquo Journal of Neurophysiologyvol 77 no 2 pp 797ndash811 1997

[36] J Duysens F Clarac and H Cruse ldquoLoad-regulating mecha-nisms in gait and posture comparative aspectsrdquo PhysiologicalReviews vol 80 no 1 pp 83ndash133 2000

[37] D A Winter The Biomechanics and Motor Control of HumanGait Normal Elderly and Pathological University of WaterlooPress Waterloo Canada 1991

[38] K G Pearson ldquoCommon principles of motor control invertebrates and invertebratesrdquo Annual Review of Neurosciencevol 16 pp 265ndash297 1993

[39] K G Pearson ldquoProprioceptive regulation of locomotionrdquoCurrent Opinion inNeurobiology vol 5 no 6 pp 786ndash791 1995

[40] J Duysens B M H van Wezel H W A A van de CrommertM Faist and J G M Kooloos ldquoThe role of afferent feedback inthe control of hamstrings activity during human gaitrdquo EuropeanJournal of Morphology vol 36 no 4-5 pp 293ndash299 1998

[41] L Finch H Barbeau and B Arsenault ldquoInfluence of bodyweight support on normal human gait development of a gaitretraining strategyrdquoPhysicalTherapy vol 71 no 11 pp 842ndash8551991

[42] J C Moreno F Barroso D Farina et al ldquoEffects of roboticguidance on the coordination of locomotionrdquo Journal of Neu-roEngineering and Rehabilitation vol 10 no 1 article 79 2013

[43] A R den Otter A C H Geurts T Mulder and J DuysensldquoSpeed related changes in muscle activity from normal to veryslow walking speedsrdquo Gait and Posture vol 19 no 3 pp 270ndash278 2004

[44] F Saibene and A E Minetti ldquoBiomechanical and physiologicalaspects of legged locomotion in humansrdquo European Journal ofApplied Physiology vol 88 no 4-5 pp 297ndash316 2003

[45] Y P Ivanenko R E Poppele and F Lacquaniti ldquoSpinalcord maps of spatiotemporal alpha-motoneuron activation inhumanswalking at different speedsrdquo Journal ofNeurophysiologyvol 95 no 2 pp 602ndash618 2006

[46] A Pepin K E Norman and H Barbeau ldquoTreadmill walkingin incomplete spinal-cord-injured subjects 1 Adaptation tochanges in speedrdquo Spinal Cord vol 41 no 5 pp 257ndash270 2003

[47] F Sylos-Labini V La Scaleia I Pisotta et al ldquoEMG patternsduring assisted walking in the exoskeletonrdquo Frontiers in HumanNeuroscience vol 8 article 423 2014

[48] J Massion K Popov J-C Fabre P Rage and V Gurfinkel ldquoIsthe erect posture in microgravity based on the control of trunkorientation or center of mass positionrdquo Experimental BrainResearch vol 114 no 2 pp 384ndash389 1997

[49] G Andreoni C Rigotti G Baroni G Ferrigno N A Colfordand A Pedotti ldquoQuantitative analysis of neutral body posturein prolonged microgravityrdquo Gait amp Posture vol 12 no 3 pp235ndash242 2000

[50] S M Gatesy and A A Biewener ldquoBipedal locomotion effectsof speed size and limb posture in birds and humansrdquo Journal ofZoology vol 224 no 1 pp 127ndash147 1991

[51] J Rubenson D B Heliams D G Lloyd and P A FournierldquoGait selection in the ostrich mechanical and metabolic char-acteristics of walking and running with and without an aerial

phaserdquo Proceedings of the Royal Society B Biological Sciencesvol 271 no 1543 pp 1091ndash1099 2004

[52] L Ren and J R Hutchinson ldquoThe three-dimensional locomotordynamics of African (Loxodonta africana) and Asian (Elephasmaximus) elephants reveal a smooth gait transition at moderatespeedrdquo Journal of the Royal Society Interface vol 5 no 19 pp195ndash211 2008

[53] A E Minetti Y P Ivanenko G Cappellini N Dominici andF Lacquaniti ldquoHumans running in place on water at simulatedreduced gravityrdquoPLoSONE vol 7 no 7 Article ID e37300 2012

[54] M Srinivasan and A Ruina ldquoComputer optimization of aminimal biped model discovers walking and runningrdquo Naturevol 439 no 7072 pp 72ndash75 2006


[56] G Clement V S Gurfinkel F Lestienne M I Lipshits andK E Popov ldquoAdaptation of postural control to weightlessnessrdquoExperimental Brain Research vol 57 no 1 pp 61ndash72 1984

[57] Y P IvanenkoW GWright R J St George andV S GurfinkelldquoTrunk orientation stability and quadrupedalismrdquo Frontiers inNeurology vol 4 article 20 2013

[58] C Papaxanthis T Pozzo K E Popov and J McIntyre ldquoHandtrajectories of vertical arm movements in one-G and zero-G environments Evidence for a central representation ofgravitational forcerdquo Experimental Brain Research vol 120 no4 pp 496ndash502 1998

[59] P Crenna D M Cuong and Y Breniere ldquoMotor programmesfor the termination of gait in humans organisation and velocity-dependent adaptationrdquo Journal of Physiology vol 537 no 3 pp1059ndash1072 2001

[60] B J McFadyen and H Carnahan ldquoAnticipatory locomotoradjustments for accommodating versus avoiding level changesin humansrdquo Experimental Brain Research vol 114 no 3 pp500ndash506 1997

[61] J W Glasheen and T A McMahon ldquoSize-dependence of water-running ability in basilisk lizards (Basiliscus basiliscus)rdquo TheJournal of Experimental Biology vol 199 no 12 pp 2611ndash26181996

[62] VA Selionov Y P Ivanenko I A Solopova andV S GurfinkelldquoTonic central and sensory stimuli facilitate involuntary air-stepping in humansrdquo Journal of Neurophysiology vol 101 no 6pp 2847ndash2858 2009

[63] F Sylos-Labini Y P Ivanenko M J Maclellan G Cappellini RE Poppele and F Lacquaniti ldquoLocomotor-like leg movementsevoked by rhythmic armmovements in humansrdquoPloSONE vol9 no 3 Article ID e90775 2014

[64] P Sale M Franceschini A Waldner and S Hesse ldquoUse ofthe robot assisted gait therapy in rehabilitation of patients withstroke and spinal cord injuryrdquo European Journal of Physical andRehabilitation Medicine vol 48 no 1 pp 111ndash121 2012

[65] P A Guertin ldquoPreclinical evidence supporting the clinicaldevelopment of central pattern generator-modulating therapiesfor chronic spinal cord-injured patientsrdquo 2014

[66] YGerasimenko PMusienko I Bogacheva et al ldquoPropriospinalbypass of the serotonergic system that can facilitate steppingrdquoJournal of Neuroscience vol 29 no 17 pp 5681ndash5689 2009

[67] V A Selionov I A Solopova D S Zhvansky et al ldquoLackof non-voluntary stepping responses in Parkinsonrsquos diseaserdquoNeuroscience vol 235 pp 96ndash108 2013


[68] C A Angeli V R Edgerton Y P Gerasimenko and S JHarkema ldquoAltering spinal cord excitability enables voluntarymovements after chronic complete paralysis in humansrdquo Brain2014

[69] D Borton M Bonizzato J Beauparlant et al ldquoCorticospinalneuroprostheses to restore locomotion after spinal cord injuryrdquoNeuroscience Research vol 78 pp 21ndash29 2014

Research ArticleConditioned Media from Microvascular Endothelial CellsCultured in Simulated Microgravity Inhibit Osteoblast Activity

Alessandra Cazzaniga Sara Castiglioni and Jeanette A M Maier

Dipartimento di Scienze Biomediche e Cliniche Luigi Sacco Universita di Milano Via GB Grassi 74 Milano Italy

Correspondence should be addressed to Jeanette A M Maier jeanettemaierunimiit

Received 23 April 2014 Revised 9 July 2014 Accepted 9 July 2014 Published 19 August 2014


Copyright copy 2014 Alessandra Cazzaniga et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Background and Aims Gravity contributes to the maintenance of bone integrity Accordingly weightlessness conditions duringspace flight accelerate bone loss and experimental models in real and simulated microgravity show decreased osteoblastic andincreased osteoclastic activities It is well known that the endotheliumandbone cells cross-talk and this intercellular communicationis vital to regulate bone homeostasis Because microgravity promotes microvascular endothelial dysfunction we anticipated thatthe molecular cross-talk between endothelial cells exposed to simulated microgravity and osteoblasts might be altered ResultsWe cultured human microvascular endothelial cells in simulated microgravity using the rotating wall vessel device developed byNASA Endothelial cells in microgravity show growth inhibition and release higher amounts of matrix metalloproteases type 2 andinterleukin-6 than controls Conditionedmedia collected frommicrovascular endothelial cells in simulatedmicrogravity were usedto culture human osteoblasts and were shown to retard osteoblast proliferation and inhibit their activityDiscussion Microvascularendothelial cells inmicrogravity are growth retarded and release high amounts of matrix metalloproteases type 2 and interleukin-6which might play a role in retarding the growth of osteoblasts and impairing their osteogenic activityConclusions We demonstratethat since simulated microgravity modulates microvascular endothelial cell function it indirectly impairs osteoblastic function

1 Introduction

Bone development and remodeling depend mainly uponcomplex interactions between osteoblasts and osteoclastsIndeed an intimate communication exists between osteo-blasts and osteoclasts since osteoclasts control osteoblasticgrowth and function while osteoblasts regulate the dif-ferentiation and the activity of osteoclasts [1] Recentlyother cells of the bone microenvironment are emerging asimplicated in bone health Among others endothelial cellsare players of the communication network in the bone [2]In embryonic skeletal tissue osteogenesis and angiogenesisare temporally related [3] and in the adults osteoblasts arealways located adjacent to endothelial cells in blood ves-sels at sites of new bone formation [4] The fact thatolder subjects with osteoporosis have decreased bloodvessels in their skeletal tissue accompanied by a paral-lel decrease in osteoblasts further highlights this closerelation [5] Several lines of evidence indicate that a

mutual communication system exists between the endothe-lium and the osteoblasts At the cellular and molec-ular levels vascular endothelial cells have been shown toregulate bone remodelling via cell signalling networks ofligand-receptor complexes and osteoblasts release growthfactors that influence endothelial cells [3]

In long duration space missions astronauts experienceconsiderable bone loss about 1-2 of bone mass per monthin the weight-bearing regions of the leg and the spinemainly because of an uncoupling between osteoblasts andosteoclasts [6ndash8] We anticipate that endothelial-osteoblastcommunication might be impaired in space and contributesto bone loss Indeed dysfunctions in human endothelial cellscultured in simulated microgravity have been described [9ndash15] and alterations in the capillaries of the epiphyses andmetaphyses of femoral bones of rats flown aboard the USlaboratory SLS-2 were detected [16]

Cross-talk between endothelial cells and osteoblasts insimulated microgravity has not been deciphered yet As



a first approach to investigate this issue we exposedosteoblasts to conditioned media (CM) from microvascularendothelial cells (HMEC) cultured in the rotating wall ves-sel (RWV) which simulates some aspects of microgravityStudies utilizing CM are considered a successful strategy forthe identification of soluble factors interconnecting differentcell types and candidate biomarkers for further validation inclinical samples [17] Indeed CM reveal the cell secretomethat is the collection of proteins that are released throughthe classical and nonclassical secretion pathways and alsoproteins shed from the cell surface These secreted proteinsinclude enzymes growth factors cytokines and other solublemediators and are important contributors to cell survivalgrowth and differentiation [17] We here show that CMfrom HMEC grown in simulated microgravity impair theproliferation and activity of cultured primary osteoblasts andosteoblast-like Saos-2 cells


21 Cell Culture HMEC were obtained from CDC (AtlantaUSA) and grown in MCDB131 containing epidermal growthfactor (10 ngmL) and 10 fetal bovine serum (FBS) on 2gelatin-coated dishes Normal human osteoblasts (NHOst)were maintained in osteoblast growth media (OGM) asindicated by the manufacturer (Lonza Basel Switzerland)at 37∘C in a humidified atmosphere containing 5 CO

2

[18] Saos-2 cells (American Type Culture Collection) werecultured inDMEMcontaining 10FBS Before beginning theexperiments with CM from HMEC NHOst and Saos-2 cellswere gradually adapted to be cultured in 1 1 HMEC growthmedium and OGM or DMEM respectively To simulatemicrogravity we utilized the RWV (Synthecom Inc HoustonTX USA) HMEC were seeded on beads (Cytodex 3 SigmaAldrich St Louis MO USA) as controls (CTR) HMECgrown on beads were cultured in the vessels not undergoingrotation [11] In the RWV the vessel rotates around a horizon-tal axis (28 rpm) and allows diffusion of oxygen and carbondioxide across a semipermeable membrane The vessel walland the medium containing cells bound to microcarrierbeads rotate at the same speed producing a vector-averagedgravity comparable with that of near-earth free-fall orbit [19]The beads do not form aggregates in the RWV and tend tobe evenly distributed throughout the vessel Such a rotationreduces gravity to approximately 3 times 10minus2 g [10] After 72 h inthe RWV or in the vessels without rotation the media fromHMEC were collected centrifuged filtered through 02 120583mfilter diluted 1 1 with fresh culture medium to replenishnutrients and used to culture NHOst and Saos-2 cells Inthese experiments the medium was changed every 48 h

22 DNA Fragmentation HMEC cell death was evalu-ated using the cell death detection ELISA (Roche) whichdetermines cytoplasmic histone-associated DNA fragmentsBriefly after 48 and 72 h in the RWV or under controlconditions the cells were lyzed and centrifuged and thesupernatant was analyzed according to the manufacturerrsquosinstruction As a positive control we used HMEC exposed

for 30min to H2O2(10 120583M) and cultured for additional 48 h

in their growth medium

23 Cell Proliferation For MTT assay NHOst and Saos-2at 50 confluence were cultured in 96-well plates for 24 hbefore being exposed for different times to the media col-lected from HMEC MTT measures the reduction of yellowtetrazolium salt MTT to dark purple formazan by succinatedehydrogenase mainly in mitochondria and it is now widelyaccepted as a reliable way to examine cell viability and prolif-eration [20] Briefly at the end of the experiment the mediawere replaced with medium containing 3-(45-Dimethyl-2-thiazolyl)-25-diphenyl-2H-tetrazolium bromide (MTT05mgmL) (Sigma Aldrich St Louis MO USA) Formazancrystals generated by the cellular reduction activity weredissolved in DMSO Absorbance was measured at 550 nm

Neutral red uptake assay was also used to estimateNHOstviability Briefly 24 h after seeding in 96-well dishes the cellswere exposed to CM from HMEC After 3 days neutralred was added to the medium to a final concentration of50 120583gmL 2 h later the wells were washedwith PBS and fixedAbsorbance was measured at 550 nm [21]

HMECand Saos-2 cells were trypsinized and stainedwithtrypan blue solution (04) and the viable cells were countedusing a Burker chamber

24 Osteoblast Activity NHOst and Saos-2 cells at 80confluence were cultured in 24-well plates with conditionedmedia from HMEC added with 100 nM dexamethasone50 120583M L-ascorbate-2-phosphate and 10mM glycerophos-phate at 37∘C in a 5CO

2for 7 and 14 days Osteoblast activ-

ity was evaluated quantifying alkaline phosphatase (ALP)enzymatic activity in the medium by a colorimetric assaybased on the hydrolysis of P-nitrophenyl phosphate Theabsorbance was measured at 405 nm [18] To analyze calciumdeposition the cellswere rinsedwith PBS fixed (70ethanol1 h) and stained for 10min with 2 Alizarin Red S (pH 42)Cultures were photographed with a digital camera AlizarinRed was then released from the cell matrix by incubationfor 15min in 10 cetylpyridinium chloride in 10mM sodiumphosphate (pH 70) The absorbance was measured at 562 nm[18]

25 Measurements of TIMP-2 and IL-6 by ELISA Condi-tioned media were centrifuged and filtered The amountsof tissue inhibitor of matrix metalloprotease (TIMP)-2 andinterleukin (IL)-6 were measured using a double-antibodysandwich ELISA (GE Healthcare) according to the manufac-turerrsquos instructions The concentrations of TIMP-2 and IL-6were determined by interpolation from a standard curve

26 Western Blot HMEC cells were lysed separated onSDS-PAGE and transferred to nitrocellulose sheets Westernanalysis was performed using antibodies against p21 p53 andGAPDH (Tebu Bio-Santa Cruz) Secondary antibodies werelabelled with horseradish peroxidase (Amersham PharmaciaBiotech) The SuperSignal chemiluminescence kit (Pierce)was used to detect immunoreactive proteins


0

1

2

3

4

5

6

7

8

24 48 72 96

HMEC-CHMEC-RWV

(h)

Tota

l cel

l num

ber (times105

)

(a)

C RWV

p21

GAPDH

p53

(b)

0100

200

300

400

500

600

700

48 72

HMEC-CHMEC-RWV

Abso

rban

ce 405

nm

(h)

lowastlowastlowast

HMEC + H2O2

(c)

Figure 1 Simulated microgravity inhibits HMEC growth (a) HMEC were cultured for different times in the RWV (HMEC-RWV) andtrypsinized and viable cells were counted HMEC-C control (b) Cell extracts (50120583glane) were loaded on a 15 SDS-PAGE blotted intonitrocellulose filter incubated with anti-p21 and anti-p53 antibodies and visualized by chemiluminescence as described After stripping theblot was incubated with an anti-GAPDH antibody to show that comparable amounts of proteins were loaded per lane (c) Apoptosis wasevaluated by ELISA on HMEC lysates after 48 and 72 h in the RWV or under control conditions Our positive control is represented byHMEC exposed to H

2O2for 30min and then cultured for additional 48 h

27 Statistical Analysis All experiments were repeated atleast three times in triplicate Data are presented as means plusmnstandard deviation Statistical differences were determinedusing the unpaired two-tailed Studentrsquos 119905 test Consider lowast119875 lt005 lowastlowast119875 lt 001

3 Results

31 Simulated Microgravity Alters HMEC BehaviourFigure 1(a) shows that culture in the RWV retarded HMECproliferation Accordingly growth inhibition correlatedwith the upregulation of p21 (WAF1) an inhibitor of cyclin-dependent kinases as detected by western blot and thisevent seems to be p53-independent since no modulationof p53 was observed in HMEC (Figure 1(b)) We also showthat no cell death is detectable after 48 and 72 h culture inthe RWV (Figure 1(c)) It is noteworthy that similar resultswere obtained when microgravity was simulated using therandom positioning machine (RPM) (data not shown) Onthe basis of results obtained by protein array on 40 proteinsinvolved in inflammation we validated the increase of IL-6and TIMP-2 in the CM from HMEC cultured for 48 and72 h in the RWV and relative controls by ELISA Figure 2(a)shows that TIMP-2 is significantly increased in the mediacollected from HMEC after 48 and 72 h in the RWV whilesecreted IL-6 was increased after 72 h culture in simulatedmicrogravity (Figure 2(b)) On these bases we decided touse 72 h conditioned media fromHMEC for the experimentson bone cells

32 HMEC Secreted Factors Impact on NHOst Cell Prolif-eration and Osteogenic Activity We evaluated the effects ofCM from HMEC on NHOst cell proliferation MTT assayrevealed a significant reduction of NHOst cell proliferationcultured in the presence of CM from HMEC in simulatedmicrogravity (Figure 3(a)) These results were confirmedby neutral red assay which estimates the number of viablecells in a culture on the basis of their ability to incorporateand bind the supravital dye neutral red in the lysosomes(Figure 3(b)) We did not detect any significant difference incell death in NHOst exposed to the conditioned media fromHMEC cultured for 72 h in the RWV and relative controls(not shown)

To evaluate osteoblastic activity NHOst cells were cul-tured for 7 and 14 days in a 24-well plate with CM fromHMEC added with an osteogenic cocktail containing 100 nMdexamethasone 50 120583ML-ascorbate-2-phosphate and 10mMglycerophosphate Two parameters were evaluated that isALP activity which has long been recognised as a reliableindicator of osteoblastic activity and calcium deposition byAlizarin Red Staining

ALP enzymatic activity was measured after 7 and 14 daysby a colorimetric assay Figure 4(a) shows that media fromHMEC in simulated microgravity inhibited ALP activityTo analyze calcium deposition we used the Alizarin Red SStaining Figure 4(b) shows that CM fromHMEC exposed tosimulated microgravity markedly inhibited the deposition ofmineral matrix


(h)

CM-CCM-RWV

0

10

20

30

40

50

60

48 72

lowast

lowastlowast

TIM

P-2

(ng106

cell)

(a)

(h)

CM-CCM-RWV

48 720

05

1

15

2

25

3

35lowast

IL-6

(ng106

cell)

(b)

Figure 2 Simulated microgravity induces TIMP-2 and IL-6 secretion by HMEC TIMP-2 (a) and IL-6 (b) were measured by ELISA in mediacollected after different times of culture in the RWV (CM-RWV) or from relative controls (CM-C)

0

01

02

03

04

05

06

07

08

24 48 72(h)

CM-CCM-RWV

Abso

rban

ce550

nm

(a)

0

02

04

06

08

1

12

14

0 24 48 72(h)

CM-CCM-RWV

Abso

rban

ce570

nm

(b)

Figure 3 CM from HMEC in simulated microgravity inhibit NHOst proliferation NHOst were cultured for different times with CM fromHMEC in simulated microgravity (CM-RWV) or by HMEC controls (CM-C) Viable cells were evaluated by MTT assay (a) and neutralred (b) and the absolute absorbance values are shown Data are expressed as the mean plusmn standard deviation of three different experimentsperformed in triplicate

33 HMEC Secreted Factors Impact on Saos-2 Cell Prolif-eration and Osteogenic Activity Many factors such as agegender and site of isolation influence the behavior of primaryosteoblasts [22]We therefore performed experiments also onan immortalized cell line to reproduce the results obtainedin NHOst and we chose Saos-2 cells because they closelyresemble primary osteoblasts [22] Indeed Saos-2 cells are

used as representative of primary osteoblasts when standardtests are evaluated [23]

Saos-2 cells were exposed toCM fromHMEC in theRWVand relative controls for different times MTT assay showsthat media fromHMEC in the RWV impair cell proliferation(Figure 5(a)) These results were confirmed when the cellswere counted (Figure 5(b))


0

01

02

03

04

05

06

07lowast

Abso

rban

ce405

nm

7 14

Days

(a)

0

02

04

06

08

1

12

14

16

18 lowast

lowast

Abso

rban

ce562

nm

7 14

Days

CM-CCM-RWV

7 days 14 days

CM-C

CM-R

WV

(b)

Figure 4 CM from HMEC in simulated microgravity inhibit NHOst activity NHOst were cultured for 7 and 14 days with mediumconditioned by HMEC in simulated microgravity (CM-RWV) or by HMEC controls (CM-C) both added with osteogenic stimuli (a) ALPenzymatic activitywas quantified by spectrophotometric analysis as described Absorbancewasmeasured at 405 nm (b)Alizarin Red Stainingwas performed Photographs were taken before acid extraction Absorbance was measured at 562 nm

Confluent Saos-2 cells were then cultured in CM fromHMEC in simulated microgravity or HMEC controls bothadded with the osteogenic cocktail and were stained withAlizarin Red to evaluate the formation of calcium phos-phate in culture [18] We found that 14-day culture in theconditioned media from HMEC in the RWV inhibited ALPactivity (Figure 6(a))The inhibition of Saos2 cell activity wasconfirmed by demonstrating lower amounts of deposition ofmineral matrix in cell cultured with the CM from HMEC inthe RWV (Figure 6(b))

4 Discussion

Bone loss in space has been reported in humans and in severalexperimental models [8] All the in vivo results obtained inspace point to major alterations of bone cells Bone cells

have been extensively studied in vitro both in space and onground using different devices to simulate microgravity toconclude that microgravity alters the morphology of thesecells [24] impairs the differentiation of osteoblasts [25] andincreases the activity of osteoclasts [8] All these results arenot surprising since gravitational forces contribute to themaintenance of bone integrity and affect bone remodeling toadjust to mechanical demands

Bone vasculature is important for skeletal developmentduring the embryonic stage postnatal growth and boneremodeling It supplies oxygen nutrients hormones cytok-ines and bone precursor cells Moreover the communicationbetween bone endothelium and bone cells is vital to regulateand modulate bone homeostasis The endothelium con-tributes to bone health by releasing osteogenic factors [26]and bone cells produce angiogenic factors that are crucial


0

05

1

15

2

25

3

35

0 h 2 days 4 days 6 days 8 days

CM-CCM-RWV

Abso

rban

ce570

nm

(a)

0

05

1

15

2

25

T0 2 days 5 days

CM-CCM-RWV

Tota

l cel

l num

ber (times105)

(b)

Figure 5 CM from HMEC in simulated microgravity inhibit Saos-2 proliferation Saos-2 were cultured for different times with CM fromHMEC in simulated microgravity (CM-RWV) or by HMEC controls (CM-C) Viable cells were evaluated by MTT assay (a) and the absoluteabsorbance values are shown After trypsinization viable cells were stained with trypan blue and counted (b)

for endothelial viability and survival under physiologicalconditions and that drive angiogenesis when needed [3]

We have shown that human endothelial cells from theumbilical vein widely used as a model of macrovascularendothelial cells are deeply influenced by simulated micro-gravity [10 11 27]These results were confirmed by our recentstudy performed on the International Space Station (ISS)[28] Other experiments have been performed on differenttypes of macrovascular endothelial cells with discordantresults which can be ascribed to poor definition of theendothelial cells used [14 15] the different culture conditionsthe use of different microgravity simulators and also theinadequate descriptions of how they were operated Less isknown about microvascular endothelial cells which coveran area 50 times greater than that of all large vesselscombined [29] In an animal model of wound healing andin a rat fibular osteotomy model microgravity retards neo-vascularization [30 31] thus indicating the occurrence ofmicrovascular endothelial dysfunction Moreover bed restwhichmimics some aspects of spaceflight causes impairmentof endothelium-dependent functions in the microcirculation[32] We have previously demonstrated that RWV-simulatedmicrogravity induces an antiangiogenic phenotype inHMEC[11] In the present study we confirm and broaden theseresults by showing that culture in the RWV retards HMECcell growth without inducing apoptosis This correlates withthe upregulation of p21 an inhibitor of the cyclinCDK2complexes necessary for the transition from the G1 to the Sphases through a p53-independent mechanism Our resultsare in disagreement with a recent report showing that culturein a clinostat induces apoptosis in pulmonary microvascularendothelial cells [12] As mentioned above these contrastingresults might be due to differences in the cells used in the cellculture conditions and in themicrogravity simulator utilized

The aim of this work was to understand whether simu-lated microgravity impairs endothelial-osteoblast communi-cation To this purpose we evaluated the effects producedon osteoblasts by CM from HMEC cultured in simulatedmicrogravity

We show that HMEC release factors that retard thegrowth of osteoblasts and severely impair their osteogenicactivity It is noteworthy that we found increased amounts ofsecreted TIMP-2 and IL-6 known to affect both endothelialcells and osteoblasts Interestingly TIMP-2 inhibits endothe-lial cell proliferation by a matrix metalloproteases (MMP)independent mechanism [33] and might therefore play arole in HMEC growth retardation in simulated microgravityTIMP-2 also impairs osteoblast activity Indeed TIMP-2nearly abolishes ALP expression [34] by inhibiting MT1-MMP (membrane type 1-metalloprotease) [34] a proteasewhich is implicated in multiple steps of osteogenic differ-entiation and is mainly involved in ALP upregulation [35]Interestingly TIMP-2 inhibits cell survival of osteoblastsforced to transdifferentiate into osteocytes [36] This resultmight offer a molecular explanation at least in part to thelysis of osteocytes in spaceflight described byBlaber et al [37]In media from HMEC cultured in the RWV we also foundincreased amounts of IL-6 a pleiotropic cytokine implicatedin acute phase response and inflammation IL-6 not onlypromotes endothelial dysfunction [38] but also affects humanosteoblast differentiation [39] thus contributing to osteope-nia

We therefore propose that microgravity impacts bothdirectly and indirectly on osteoblasts Microgravity has beenshown to directly inhibit osteoblasts In addition by modu-lating microvascular endothelial cell function microgravityindirectly exerts inhibitory effects on osteoblasts


0

01

02

03

04

05

06

07

08

7 14

Days

lowast

Abso

rban

ce405

nm

(a)

0

02

04

06

08

1

12

14

7 14

Days

CM-CCM-RWV

lowast

lowast

Abso

rban

ce562

nm

7 days 14 days

CM-C

CM-R

WV

(b)

Figure 6 CM fromHMEC in simulated microgravity inhibit Saos-2 activity Saos-2 were cultured for 7 and 14 days with CM fromHMEC insimulated microgravity (CM-RWV) or by HMEC controls (CM-C) both added with osteogenic stimuli (a) ALP enzymatic activity and (b)Alizarin Red Staining were performed as above

The current space programs onboard the ISS and thefuture human exploration of Mars require long durationmissions However several biomedical issues still need to beclarified before these missions can take place without causinghealth problems to the astronauts Our results suggest thatendothelial dysfunction might represent a common denomi-nator for cardiovascular deconditioning and for bone loss andoffer a new light to interpret the behaviour of mammalianskeleton in microgravity Eventually these results mightfoster studies to develop countermeasures that target theendothelium to improve both bone homeostasis and vascularfunction

Conflict of InterestsThe authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This work was supported by a grant from the European SpaceAgency to Jeanette A M Maier

References

[1] T C Phan J Xu andM H Zheng ldquoInteraction between osteo-blast and osteoclast impact in bone diseaserdquoHistology and His-topathology vol 19 no 4 pp 1325ndash1344 2004

[2] B Guillotin C Bourget M Remy-Zolgadri et al ldquoHuman pri-mary endothelial cells stimulate human osteoprogenitor celldifferentiationrdquo Cellular Physiology and Biochemistry vol 14no 4ndash6 pp 325ndash332 2004

[3] J Kular J Tickner S M Chim and J Xu ldquoAn overview of theregulation of bone remodelling at the cellular levelrdquo ClinicalBiochemistry vol 45 no 12 pp 863ndash873 2012


[4] B Decker H Bartels and S Decker ldquoRelationships betweenendothelial cells pericytes and osteoblasts during bone for-mation in the sheep femur following implantation of trical-ciumphosphate-ceramicrdquoAnatomical Record vol 242 no 3 pp310ndash320 1995

[5] R D Prisby M W Ramsey B J Behnke et al ldquoAging reducesskeletal blood flow endothelium-dependent vasodilation andno bioavailability in ratsrdquo Journal of Bone andMineral Researchvol 22 no 8 pp 1280ndash1288 2007

[6] L Vico P Collet A Guignandon et al ldquoEffects of long-termmicrogravity exposure on cancellous and cortical weight-bear-ing bones of cosmonautsrdquo The Lancet vol 355 no 9215 pp1607ndash1611 2000

[7] AD LeBlanc E R SpectorH J Evans and J D Sibonga ldquoSkel-etal responses to space flight and the bed rest analog a reviewrdquoJournal of Musculoskeletal and Neuronal Interactions vol 7 no1 pp 33ndash47 2007

[8] M P Nagaraja and D Risin ldquoThe current state of bone lossresearch data from spaceflight and microgravity simulatorsrdquoJournal of Cellular Biochemistry vol 114 no 5 pp 1001ndash10082013

[9] S I M Carlsson M T S Bertilaccio E Ballabio and J A MMaier ldquoEndothelial stress by gravitational unloading effectson cell growth and cytoskeletal organizationrdquo Biochimica etBiophysica Acta vol 1642 no 3 pp 173ndash179 2003

[10] S Versari A Villa S Bradamante and J A M Maier ldquoAlter-ations of the actin cytoskeleton and increased nitric oxide syn-thesis are common features in human primary endothelial cellresponse to changes in gravityrdquo Biochimica et Biophysica ActamdashMolecular Cell Research vol 1773 no 11 pp 1645ndash1652 2007

[11] M Mariotti and J A M Maier ldquoGravitational unloading indu-ces an anti-angiogenic phenotype in human microvascularendothelial cellsrdquo Journal of Cellular Biochemistry vol 104 no1 pp 129ndash135 2008


[13] S M Grenon M Jeanne J Aguado-Zuniga M S Conteand M Hughes-Fulford ldquoEffects of gravitational mechanicalunloading in endothelial cells association between caveolinsinflammation and adhesion moleculesrdquo Scientific Reports vol3 article 1494 2013

[14] LMorbidelli MMonici NMarziliano et al ldquoSimulated hypo-gravity impairs the angiogenic response of endothelium byup-regulating apoptotic signalsrdquo Biochemical and BiophysicalResearch Communications vol 334 no 2 pp 491ndash499 2005

[15] M Infanger P Kossmehl M Shakibaei et al ldquoInduction ofthree-dimensional assembly and increase in apoptosis ofhuman endothelial cells by simulated microgravity impact ofvascular endothelial growth factorrdquo Apoptosis vol 11 no 5 pp749ndash764 2006

[16] N V Rodionova and V S Oganov ldquoChanges of cell-vascularcomplex in zones of adaptive remodeling of the bone tissueunder microgravity conditionsrdquo Advances in Space Researchvol 32 no 8 pp 1477ndash1481 2003

[17] PDowling andMClynes ldquoConditionedmedia fromcell lines acomplementarymodel to clinical specimens for the discovery ofdisease-specific biomarkersrdquo Proteomics vol 11 no 4 pp 794ndash804 2011

[18] M Leidi F DelleraMMariotti and J AMMaier ldquoHighmag-nesium inhibits human osteoblast differentiation in vitrordquoMag-nesium Research vol 24 no 1 pp 1ndash6 2011


[20] S Castiglioni S Casati R Ottria P Ciuffreda and J A MMaier ldquoN6-isopentenyladenosine and its analogue N6-ben-zyladenosine induce cell cycle arrest and apoptosis in bladdercarcinoma T24 cellsrdquo Anti-Cancer Agents in Medicinal Chem-istry vol 13 no 4 pp 672ndash678 2013

[21] S Casati R Ottria E Baldoli E Lopez J A Maier and PCiuffreda ldquoEffects of cytokinins cytokinin ribosides and theiranalogs on the viability of normal and neoplastic human cellsrdquoAnticancer Research vol 31 no 3 pp 3401ndash3406 2011

[22] E M Czekanska M J Stoddart R G Richards and J S HayesldquoIn search of an osteoblast cell model for in vitro researchrdquoEuropean Cells and Materials vol 24 pp 1ndash17 2012

[23] L Saldana F Bensiamar A Bore and N Vilaboa ldquoIn searchof representative models of human bone-forming cells forcytocompatibility studiesrdquo Acta Biomaterialia vol 7 no 12 pp4210ndash4221 2011


[25] G Carmeliet G Nys and R Bouillon ldquoMicrogravity reducesthe differentiation of human osteoblastic MG-63 cellsrdquo Journalof Bone and Mineral Research vol 12 no 5 pp 786ndash794 1997

[26] S M Chim J Tickner S T Chow et al ldquoAngiogenic factorsin bone local environmentrdquo Cytokine amp Growth Factor Reviewsvol 24 no 3 pp 297ndash310 2013

[27] M Mariotti and J A M Maier ldquoHumanMicro- and macrovas-cular endothelial cells exposed to simulated microgravityupregulate hsp70rdquoMicrogravity Science and Technology vol 21no 1-2 pp 141ndash144 2009

[28] S Versari G Longinotti L Barenghi J A Maier and S Brad-amante ldquoThe challenging environment on board the Interna-tional Space Station affects endothelial cell function by trig-gering oxidative stress through thioredoxin interacting proteinoverexpression the ESA-SPHINX experimentrdquo FASEB Journalvol 27 pp 4466ndash4475 2013

[29] S Danese E Dejana and C Fiocchi ldquoImmune regulation bymicrovascular endothelial cells directing innate and adaptiveimmunity coagulation and inflammationrdquo The Journal ofImmunology vol 178 no 10 pp 6017ndash6022 2007

[30] J M Davidson A M Aquino S C Woodward and W WWilfinger ldquoSustained microgravity reduces intrinsic woundhealing and growth factor responses in the ratrdquo The FASEBJournal vol 13 no 2 pp 325ndash329 1999

[31] M E Kirchen K M OrsquoConnor H E Gruber et al ldquoEffects ofmicrogravity on bone healing in a rat fibular osteotomymodelrdquoClinical Orthopaedics and Related Research no 318 pp 231ndash2421995

[32] M Coupe J O Fortrat I Larina G Gauquelin-Koch CGharib and M A Custaud ldquoCardiovascular deconditioningfrom autonomic nervous system to microvascular dysfunc-tionsrdquo Respiratory Physiology amp Neurobiology vol 169 supple-ment 1 pp S10ndashS12 2009

[33] W G Stetler-Stevenson and D Seo ldquoTIMP-2 an endogenousinhibitor of angiogenesisrdquo Trends in Molecular Medicine vol 11no 3 pp 97ndash103 2005

[34] S Barthelemi J Robinet R Garnotel et al ldquoMechanical forces-induced human osteoblasts differentiation involves MMP-2MMP-13MT1-MMP proteolytic cascaderdquo Journal of CellularBiochemistry vol 113 no 3 pp 760ndash772 2012


[35] P Manduca A Castagnino D Lombardini et al ldquoRole of MT1-MMP in the osteogenic differentiationrdquo Bone vol 44 no 2 pp251ndash265 2009

[36] M A Karsdal T A Andersen L Bonewald and C Chris-tiansen ldquoMatrix metalloproteinases (MMPs) safeguard osteo-blasts from apoptosis during transdifferentiation into osteo-cytes MT1-MMP maintains osteocyte viabilityrdquo DNA and CellBiology vol 23 no 3 pp 155ndash165 2004

[37] E A Blaber N Dvorochkin C Lee et al ldquoMicrogravity inducespelvic bone loss through osteoclastic activity osteocytic oste-olysis and osteoblastic cell cycle inhibi tion by CDKN1ap21rdquoPLoS ONE vol 8 no 4 Article ID e61372 2013

[38] S Wassmann M Stumpf K Strehlow et al ldquoInterleukin-6induces oxidative stress and endothelial dysfunction by over-expression of t he angio tensin II type 1 receptorrdquo CirculationResearch vol 94 no 4 pp 534ndash541 2004

[39] B Peruzzi A Cappariello A del Fattore N Rucci F deBenedetti and A Teti ldquoC-Src and IL-6 inhibit osteoblastdifferentiation and integrate IGFBP5 signallingrdquo Nature Com-munications vol 3 article 630 2012

Research ArticlePhenotypic Switch Induced by Simulated Microgravity onMDA-MB-231 Breast Cancer Cells

Maria Grazia Masiello12 Alessandra Cucina2 Sara Proietti2 Alessandro Palombo3

Pierpaolo Coluccia2 Fabrizio DrsquoAnselmi2 Simona Dinicola2 Alessia Pasqualato2

Veronica Morini2 and Mariano Bizzarri3

1 Department of Clinical and Molecular Medicine ldquoSapienzardquo University of Rome Piazza Sassari 3 00161 Rome Italy2 Department of Surgery ldquoPietroValdonirdquo ldquoSapienzardquo University of Rome Via A Scarpa 14 00161 Rome Italy3 Department of Experimental Medicine ldquoSapienzardquo University of Rome Systems Biology Group Viale Regina Elena 324Via A Scarpa 14 00161 Rome Italy




Copyright copy 2014 Maria Grazia Masiello et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Microgravity exerts dramatic effects on cell morphology and functions by disrupting cytoskeleton and adhesion structures as wellas by interfering with biochemical pathways and gene expression Impairment of cells behavior has both practical and theoreticalsignificance given that investigations of mechanisms involved in microgravity-mediated effects may shed light on how biophysicalconstraints cooperate in shaping complex living systems By exposing breast cancer MDA-MB-231 cells to simulated microgravity(sim0001 g) we observed the emergence of twomorphological phenotypes characterized by distinctmembrane fractal values surfacearea and roundness Moreover the two phenotypes display different aggregation profiles and adherent behavior on the substrateThese morphological differences are mirrored by the concomitant dramatic functional changes in cell processes (proliferation andapoptosis) and signaling pathways (ERK AKT and Survivin) Furthermore cytoskeleton undergoes a dramatic reorganizationeventually leading to a very different configuration between the two populations These findings could be considered adaptive andreversible features given that by culturing microgravity-exposed cells into a normal gravity field cells are enabled to recover theiroriginal phenotype Overall these data outline the fundamental role gravity plays in shaping form and function in living systems

1 Introduction

Space flights induce relevant changes in human physiologysuch as bone loss muscle atrophy deregulation of immunefunction hematological anomalies and cardiovascular func-tion impairment Microgravity effects may be ascribed tosystemic interferences with body fluids distribution disap-pearance of fluid shear perturbation of the circadian clockaltered endothelial function and reduced loading on skeletalstructures [1] Yet a direct effect on cell and signalingpathways inside the cell has been documented despite thefact that microgravity has been previously thought to be tooweak for contrasting the intermolecular forces [2]Thereby itis likely that spaceflight could exert its detrimental effects onastronauts via changes in cellular structure andor functions

Several studies performed both in simulated and actualmicrogravity have shown that normal as well as neoplas-tic cells undergo dramatic changes after exposition to amicrogravity field Cell morphology as well as features ofsubcellular organelles and cytoskeleton structure has beenreported to be dramatically influenced by gravity [3 4] Sim-ilarly relevant modifications in tissue organization have beenrecorded in microgravity-exposed organs andor animals [56] Shape changes are likely to be mediated by concomitantstructural rearrangement of cytoskeleton (CSK) which isseverely disorganized undermicrogravity [7 8] CSK conveysmechanical signals into the cells and by that way it influencesboth biochemical pathways [9 10] and gene expression [1112] As a consequence many metabolic proliferative anddifferentiating processes end up to be deeply perturbed [13]



Microgravity effects may be ascribed to both indirect anddirect effects [14] Meanwhile specialized cells and structuresin the plant realm have been found to be sensitive to evensubtle change in gravity vector [15] no components in themammalian cells have been so far identified as having asufficiently large mass density difference in respect to the sur-roundingmedium thus the force exerted by the gravitationalfield is nowhere higher than the energy of random thermalmotion and cannot significantly modify the behaviour of anysingle subcellular structure Instead mammalian cells maybe able to sense some environmental changes due to gravityaffecting a wide range of biophysical parameters buoyancyshear forces viscosity diffusion process and many othersYet a lot of gravity-related phenomena at the cellular levelinvolving shape rearrangement cytoskeleton disruption andeven modified gene expression would hardly be explainedby only considering changes in ldquoexternalrdquo environmentalbiophysical parameters Indeed gravity may likely affectsome general properties of the systems acting ldquodirectlyrdquo asan organizing field parameter We have previously reportedthat by ldquoremovingrdquo the gravitational constraint according tothe nonequilibrium theory [16] murine osteoblasts under-went a transition after a bifurcation point thus recoveringdegrees of freedom enabling the system in accessing newattractor states that is new phenotypic configurations [17]Indeed microgravity induces the emergence of two distinctphenotypes characterized by different morphologies Hereinwe investigate if a similar pattern could be retrieved inbreast cancer cells and how such features are associated withdifferences in their biochemical pathways Indeed conflictingdata have been reported by investigations carried out oncancer cells exposed to microgravity some authors haverecorded an overall inhibitory effect on cancer cell prolif-eration motility and survival [18 19] whereas others haveobserved the opposite [20ndash22] We hypothesize that suchresults may be likely explained by the emergence of distinctcell phenotypes characterized by different functional andreproductive features


21 RPM (Random Positioning Machine) Microgravity con-ditions were simulated by a Desktop RPM a particular kindof 3D clinostat [23] manufactured by Dutch Space (LeidenThe Netherlands) The degree of microgravity simulationdepends on angular speed and on the inclination of the diskThese tools do not actually eliminate the gravity but allowyou to apply a stimulus rather than a unidirectional omni-directional 1 g Effects generated by the RPM are comparableto those of the real microgravity provided that the directionchanges are faster than the response time of the system togravity field The desktop RPM was located in a standardincubator (to maintain temperature CO

2 and humidity


22 Cell Culture MDA-MB-231 human breast cancer cellline was purchased from European Collection of Cell Cul-tures (ECACC Sigma-Aldrich St Louis MO USA) Cells

were seeded into Nunc OptiCell Cell Culture Systems gas-permeable cell culture disks (Thermo Scientific RochesterUSA) and cultured in Dulbeccorsquos modified Eaglersquos medium(DMEM Euroclone Ltd Cramlington UK) supplementedwith 10 Fetal Bovine Serum (FBS HyClone LaboratoriesLogan UT USA) 200mM L-glutamine 100 IUmL Peni-cillin and 100 120583gmL Streptomycin (all from Euroclone LtdCramlington UK) Then OptiCells containing subconfluentmonolayers were fixed onto the RPM as close as possibleto the center of the platform which was rotated at aspeed of 60∘s using the random mode of the machineOn ground control (1 g static cultures) and RPM cultureswere kept in the same humidified incubator at 37∘C in anatmosphere of 5 CO

2in air Experiments were performed

for 24 and 72 hours After 24 and 72 hours of microgravityexposure cell clumps swimming in culture supernatants werefound in addition to adherent cells and separately collectedThe three cell populations (on ground control cells RPMadherent cells and RPM cell clumps) were characterizedseparately

23 Optical Microscopy Cell clumps were collected washedin PBS and deposited onto a clearly defined area of a glassslide using a Shandon CytoSpin 4 Cytocentrifuge ThermoScientific while maintaining cellular integrity Cell clumpsand adherent and on ground control cells were fixed in 4paraformaldehyde for 10 minutes at 4∘C and photographedwith Nikon Coolpix 995 digital camera coupled with ZeissAxiovert optical microscope The images were obtained witha 320x magnification saved as TIFF files and used for imageanalysis

24 Image Analysis Image analysis was performed on 10images for each group of MDA-MB-231 cells As the analysiswas performed blindly the image groups were classified asfollows A (on ground cells 24 h) B (RPM adherent cells24 h) C (RPM cell clumps 24 h) D (on ground cells 72 h)E (RPM adherent cells 72 h) and F (RPM cell clumps 72 h)In each image single randomly chosen cells (50 for eachgroup) were contoured with a fine black marker by differentresearchers simply scanned and cataloged according to thetime of study 24 and 72 hours This method was chosenbecause pathologists are used to correlate the shape the cellsacquire with their malignancy by means of morphologicalqualitative and subjective observations Thus we decided toperform a semiautomatic analysis coupling the expertise ofresearchers with a computerized parameterization methodAll the images were processed by Adobe Photoshop CS4All the pictures (ie all the sheets of the groups for eachtime point) were resized at 2560 times 1920 pixels according tooriginal scale of image acquisition For each black contouredcell edges were refined Then cells were black filled andthreshold was adjusted in order to exclude from the imageother cells and background For each time point a single sheetof all the cells considered was created To obtain single cellshape parameters (area 119860 roundness solidity and fractaldimension FD) ImageJ v147h software was used Then


the software analyzed single cells by the function ldquoshapedescriptorrdquo In addition to area 119860 were calculated

Roundness = 4119860120587radicma

Solidity = 119860CA

(1)

where 119860 is the area of the cell ma is the major axis and CAis the convex area namely the area of the convex hull of theregion The convex hull of a region is the smallest region thatsatisfies two conditions (a) it is convex and (b) it contains theoriginal region

As for FD it was obtained by means of box countingmethod using FracLac plugin

FD = lim120576rarr0

[1 minuslog [119871

120576(119862)]

log 120576] (2)

where 119862 is the considered curve 119871 is the length of the curve119862 and 120576 is the length of the segment used as unit to calculate119871

Single graphs about roundness solidity and FD wereobtained for each set of images

25 Fluorescence Microscopy MDA-MB-231 cells were fixedin 4 paraformaldehyde for 10 minutes at 4∘C and incu-bated over night at 4∘C with PBS (CMF Calcium andMagnesium Free) 15 goat serum plus the following specificantibodies anti-120572-tubulin (T5168 Sigma-Aldrich) and anti-vimentin (sc-6260 Santa Cruz biotechnology) For F-actinvisualization rhodamine-phalloidin (Invitrogen MolecularProbes Eugene) was used Cells were washed three timeswith PBS (1 BSA 02 Triton X 100) and incubated withrhodamine-phalloidin the anti-mouse IgG-FITC PN IM1619secondary antibody (Beckman-Coulter Inc Fullerton CAUSA) and HOECHST 33342 (Sigma-Aldrich St Louis MOUSA) to stain the DNA Finally cells were washed mountedin buffered glycerol (01M pH 95) and analyzed usinga Zeiss Fluorescent Microscope The images were scannedunder 40x objective

26 Cell Cycle Analysis Cell clumps were collected andcentrifuged and pellets were trypsinized and washed twicewith PBS (Phosphate Buffered Saline Sigma-Aldrich StLouis MO USA) Adherent and ground control cells weretrypsinized and washed twice with PBS Cells were fixed with70 ethanol at 4∘C for 24 h and stainedwithDNAPREP Stain(Beckman Coulter Fullerton USA) at 4∘C overnight Stainedcells weremeasured by flow cytometry Cell cycle analysis wasperformed three times

27 Annexin V7-AAD Staining Cell clumps were collectedand centrifuged and pellets were trypsinized and washedtwice with PBS Adherent cells and ground control cells weretrypsinized andwashed twicewith PBSThe cells were stainedwith FITC labeled annexin V7-AAD (7-aminoactinomycin-D) according to the manufacturerrsquos instructions (annexin

V7-AAD kit Beckman Coulter Marseille France) Brieflya washed cell pellet (5 times 104 cellsmL) was resuspended in500120583L binding buffer 10 120583L of annexinV together with 20120583L7-AAD was added to 470 120583L cell suspension The cells wereincubated for 15min on ice in the dark The samples wereanalyzed by flow cytometry Apoptosis assay was performedthree times

28 Flow Cytometry Flow cytometry was performed usingan EPICS Coulter XL (Beckman-Coulter Inc) The flu-orescence of 20000 events was measured An excitationwavelength of 488 nmwas used in combinationwith standardfilters to discriminate between the FL1 (forward scatter) andFL3 (side scatter) channels Data were analyzed byModFit LTSoftware (Verity Software Inc USA)

29 Western Blot Cell clumps were washed twice withice-cold PBS and resuspended in RIPA lysis buffer (SigmaChemical Co) Adherent and ground control cells werewashed twice with ice-cold PBS and scraped in RIPAlysis buffer (Sigma Chemical Co) A mix of proteaseinhibitors (Complete-Mini Protease Inhibitor CocktailTablets Roche Mannheim Germany) and phosphataseinhibitors (PhosStop Roche Mannheim Germany) wasadded just before use Cellular extracts were then centrifugedat 8000timesg for 10min The protein content of supernatantswas determined using the Bradford assay For westernblot analysis cellular extracts were separated on SDS-polyacrylamide gels with a concentration of acrylamidespecific for the proteins studied Proteins were blottedonto nitrocellulose membranes (BIO-RAD Bio-RadLaboratories Hercules CA USA) and probed with thefollowing antibodies anti-Cyclin D1 (AB-90009) fromImmunological Sciences anti-survivin (number 2808)anti-phospho-AKT (ser473) (number 9271S) anti-AKT(number 9272S) anti-phospho-ERK12 (number 9106)anti-cleaved PARP (number 9541) anti-GAPDH (number2118) all from Cell Signaling Technology anti-Bax (sc-493)anti-Bcl-2 (sc-492) anti-ERK1 (sc-94) all from Santa CruzBiotechnology Antigens were detected with enhancedchemiluminescence kit (Amersham Biosciences LittleChalfont Buckinghamshire England) according to themanufacturerrsquos instructions All Western blot images wereacquired and analyzed through Imaging Fluor S densitometer(Biorad-Hercules)

210 Statistical Analysis Data were expressed as mean plusmnstandard deviation (SD) or mean plusmn standard error (SE)Data were statistically analyzed with the analysis of variance(ANOVA) followed by the Bonferroni post-test Differenceswere considered significant at the level of 119875 lt 0 05 Statisticalanalysis was performed by using GraphPad Instat software(GraphPad Software Inc San Diego CA USA)

3 Results

31 Effect of Microgravity on MDA-MB-231 MorphologyMDA-MB-231 cell line grew as a monolayer when culturedunder static 1 g condition (on ground control Figures 1(a)


100120583m

(a)

100120583m

(b)

100120583m

(c)

100120583m

(d)

100120583m

(e)

100120583m

(f)

Figure 1 Microphotographs of MDA-MB-231 by optical microscopy MDA-MB-231 cell line in on ground control condition at 24 (a) and 72hours (c)MDA-MB-231 cells exposed tomicrogravity for 24 (b) and 72 hours (d) RPMcell clumps after reseeding into a normal gravitationalfield after 6 (e) and 24 hours (f) Magnification times320 (a) (b) (c) (d) times100 (e) (f)

and 1(c)) After 24 and 72 hours of simulated microgravityexposure cells were distributed into two populations thefirst adhering to the substrate represented by flat spindlecells the second represented by rounded cells aggregatedin cell clumps floating in the culture medium (Figures 1(b)and 1(d)) This distribution does not represent a transitorystate given that the percentage of cells at both 24 and 72hours still remains constant However beside the fact thatsuch changes are likely to involve several modifications onshape and biological function the observed nonadherentphenotype is still wholly reversible after 72 hours Indeedafter reseeding into a normal gravitational field cell clumpswere de novo able to adhere to the culture plate alreadyafter 6 hours (Figure 1(e)) and to fully recover their native

morphological traits and topological distribution after 24hours (Figure 1(f))

32 Effect of Microgravity on Quantitative MorphologicalParameters Quantitative image analysis was performed byquantifying roundness solidity and fractal dimension (FD)Significant differences were recorded among the three cellpopulations (Table 1) Roundness no statistically significantdifferences between on ground cells and RPM adherent cellshave been observed at both 24 and 72 hours Instead RPMcell clumps showed a significant strong increase in roundnesscompared to control and RPM adherent cells at 24 and72 hours Solidity at 24 hours no statistically significantdifferences between on ground cells and RPM adherent cells


Table 1

Roundness plusmnSE Solidity plusmnSE FD plusmnSE24 hours

On ground cells 04563 00301 06690 00225 17482 00091RPM adherent cells 03991 00275 06499 00200 17406 00063RPM cell clumps 07894 00219 lowastlowast 08966 00263 lowastlowast 14625 00015 lowastlowast

72 hoursOn ground cells 03227 00263 04687 00136 16677 00036RPM adherent cells 04081 00311 06115 00226 lowast 17245 00067RPM cell clumps 07961 00208 lowastlowast 08573 00469 lowastlowast 14990 00015 lowastlowast

Roundness solidity and fractal dimension (FD) mean values plusmn SE in on ground control cells RPM adherent cells and RPM cell clumps lowastP lt 001 versus onground control cells lowastlowastP lt 0001 versus on ground control and RPM adherent cells by ANOVA followed by Bonferroni post-test

were recorded meanwhile in RPM cell clumps the solidityindex was significantly higher with respect to on groundcells and RPM adherent cells At 72 hours the solidity indexsignificantly increased in both RPM cell populations reach-ing its highest level in RPM cell clumps Fractal dimensionno statistically significant differences between on groundcells and RPM adherent cells were recorded both at 24 and72 hours Instead RPM cell clumps showed a statisticallysignificant decrease in FD compared to control and RPMadherent cells at 24 and 72 hours These results are coherentwith the qualitativemorphological assessment and confirmedthat microgravity exposure leads to the emergence of twomorphologically distinct cell populations

33 Effect of Microgravity on MDA-MB-231 CytoskeletonArchitecture After 24 hours of microgravity exposure bothMDA-MB-231 RPM adherent cells and RPM cell clumpsshowed a large rearrangement of F-actin 120572-tubulin andvimentin compared to on ground control cells In onground control cells the network of cytosolic F-actin bundlesappeared well organized and the microtubules appearedorderly radiating from the perinuclear area throughout thecytoplasm toward the cell periphery (Figure 2(a)) In RPMadherent cells the actin filaments showed a disappearanceof the complex cytosolic network which appeared mostlylocalized on the cell border microtubules were disorganizedwith a more evident thickening in perinuclear position(Figure 2(b)) In floating cell clumps the actin meshworkappeared completely disrupted and the filaments weremainly localized behind the cell border Tubulin mesh-work was also completely disrupted and a slight diffusefluorescence was observed spreading throughout the entirecytoplasm (Figure 2(c)) In the on ground cells vimentinfilaments were well organized all over the cytoplasm (Fig-ure 3(a)) In both RPM adherent cells and cell clumps thevimentin network was disrupted appearing in the form ofdense aggregates closely associated with the nucleus (Figures3(b) and 3(c)) Cytoskeleton rearrangements were almoststable given that no significant changes have been observedeven after 72 hours in microgravity-exposed cells (data notshown)

34 Microgravity Modifies MDA-MB-231 Cell Cycle Distribu-tion MDA-MB-231 cells subjected tomicrogravity displayed

relevant modification in their cell cycle (Figures 4(a) and4(b)) Nonadherent RPM-treated MDA-MB-231 cells weredistributed in a significantly different manner when com-pared to both control and RPM-adherent cells indeed after24 hours floating cell clumps in G0G1 and in S phasewere significantly decreased whereas cells in G2M phaseincrease up to 6-fold On the contrary adherent RPM-treated cells displayed only a slight increase in the S phasedistribution when compared to controls After 72 hours ofmicrogravity exposure MDA-MB-231 RPM cell clumps stillshowed a relevant decrease in the S phase thus demonstratinga persistent inhibition of cell growth cells number in G2Mphase was significantly higher meanwhile no significantchange in G0G1 was observed (Figure 4(b)) Again besidesminor differences control and adherent RPM-treated cellsdisplayed an overlapping distribution in theG0G1 andG2Mphase whereas the percentage of cells in the S phase was stillhigher than that recorded in floating RPM cell clumpsThesedata are exemplarily mirrored by Cyclin D1 data Cyclin D1is one of the main factors that regulate the activation of thecell cycle and its increase is required to foster cell growth Asexpected a statistically significant decrease of CyclinD1 levelsin RPMcell clumpswas recordedmeanwhile CyclinD1 levelsare higher in adherent proliferating RPM cells as well as incontrol samples These effects were observed at both 24 and72 hours (Figure 4(c))

35 Microgravity Induces Apoptosis in MDA-MB-231 CellClumps Data obtained by cytofluorimetric assays reporteda statistically significant increase in the apoptotic rate after 24and 72 hours of microgravity exposure in cell clumps withrespect to both adherent cells and on ground control cells(Figure 5(a)) Western blot analysis revealed a statisticallysignificant increase of BaxBcl-2 ratio at 72 hours in RPMcell clumps compared to RPM adherent cells and on groundcontrol cells (Figure 5(b)) Similarly cleaved-PARP levels adirect marker of caspase-3 activation [24] were significantlyincreased at 24 and 72 hours in RPM cell clumps compared toRPM adherent cells and on ground control cells (Figure 5(c))Overall these data suggest that nonadherent cells were signif-icantly constrained in their viability given that microgravityinhibits cell growth and at the same time enhances theapoptotic process Adherent cells in microgravity on the



(a)


(b)


(c)

Figure 2 Immunofluorescence images of F-actin and 120572-tubulin inMDA-MB-231 Rhodamine-phalloidin staining of MDA-MB-231 showingF-actin distribution patterns (red color) and immunostaining of 120572-tubulin (green color) and HOECHST 33342 to stain nuclei (blue color)after 24 hours in on ground control cells (a) RPM adherent cells (b) and RPM cell clumps (c) Magnification times400

contrary display only minor changes in both apoptosis andproliferation rate

36 Microgravity Modifies MDA-MB-231 Survival PathwaysMicrogravity exposure is associated with a statistically sig-nificant decrease in the phosphorylation of AKT in adherentcells and cell clumps with respect to on ground control cellsafter 24 h Instead after 72 hours of microgravity expositionRPM adherent cells showed a statistical increase of p-AKTexpression with respect to on ground control cells andRPM cell clumps (Figure 6(a)) such biphasic effect on AKTactivation may help explain the biphasic trend observedin apoptosis rate in adherent RPM-exposed cells apoptosisincreases indeed at 24 hours when p-AKT values are lowthe opposite is observed when p-AKT levels increase at72 hours A similar behavior may be described for thetwo other prosurvival factors Survivin and phosphorylated-ERKMicrogravity exposure induced a statistically significantdecrease in Survivin levels in both adherent and nonadherent

RPM-treated cells at 24 hours However at 72 hours Survivinlevels significantly increased in RPM adherent cells anddecreased in nonadherent RPM-treated cells (Figure 6(b))Likewise ERK phosphorylation was severely inhibited inRPM cell clumps after 24 and 72 hours in respect tovalues observed in both RPM adherent and control cells(Figure 6(c)) Taken as a whole prosurvival factors increasedin adherent RPM-treated cells meanwhile they decreasedin nonadherent RPM-exposed cells mirroring so far theobserved mentioned trend in apoptosis

4 Discussion

Breast cancer cells exposed to microgravity acquire twodistinct phenotypes already after the first 24 hours Such out-standing result has been previously observed in osteoblastscultured in microgravity [17] and can be interpreted inthe light of the nonequilibrium theory Briefly a dissipa-tive nonlinear system sufficiently far from the equilibrium


Vimentin Merge

(a)

Vimentin Merge

(b)

Vimentin Merge

(c)

Figure 3 Immunofluorescence images of vimentin in MDA-MB-231 Immunostaining of vimentin (green color) and HOECHST 33342 tostain nuclei (blue color) after 24 hours in on ground control cells (a) RPM adherent cells (b) and RPM cell clumps (c) Magnification times400

can form spatial stationary patterns after experiencing aphase transition leading to new asymmetric configurations[25] These states are equally accessible that is to say thatthere exists a complete symmetry between the emergingconfigurations as it is reflected in the symmetry of thebifurcation diagram However the superimposition of anexternal field even if a weak one like gravity may breakthe systemrsquos symmetry bestowing a preferential directionalityaccording to which the system will preferentially evolve intoone of the states and not the others Indeed bifurcations far

from equilibrium endow a system with a very pronouncedsensitivity allowing it to capture the slightest environmentalasymmetry and select a preferred polarity or chirality Inother words the ldquoweakrdquo force dramatically influences thesystem in selecting one out ofmany other configurations [26]On the contrary by removing the gravitational constraints thesystem can freely access different attractor states recoveringhenceforth new configuration states (ldquophenotypesrdquo) Suchmodel has been experimentally confirmed showing that sev-eral cell components characterized by a nonlinear dynamics


0

10

20

30

40

50

60

70

G0G1 S G2M

Cell

s in

each

pha

se (

)


24h

lowastlowastlowast

lowastlowast

lowast

(a)

Cell

s in

each

pha

se (

)

G0G1 S G2M


0102030405060708090

72h

lowastlowastlowast

(b)

Cyclin D1

005

115

225

335

4

Relat

ive e

xpre

ssio

n

GAPDH


cells) clumps)


cells) clumps)

24h

24h

72h

72h

lowastlowast

lowast


(c)

Figure 4 Cell cycle analysis in MDA-MB-231 Cells distribution along the different phases of the cell cycle at 24 (a) and 72 hours (b) in onground control cells RPM adherent cells and RPM cell clumps (c) Immunoblot bar chart showing the expression of Cyclin D1 in MDA-MB-231 in on ground control cells RPM adherent cells and RPM cell clumps at 24 and 72 hours Columns and bars represent densitometricquantification of optical density (OD) of specific protein signal normalized with the OD values of the GAPDH served as loading control Eachcolumn represents the mean value plusmn SD of three independent experiments lowast119875 lt 0 05 lowastlowast119875 lt 0 01 lowastlowastlowast119875 lt 0 001 versus on ground control119875 lt 0 01 119875 lt 0 001 versus RPM adherent cells by ANOVA followed by Bonferroni post-test

when exposed to microgravity may experience bifurcationtransitions leading to the appearance of new self-organizedstates from an initially homogeneous conformation [27 28]It is tempting to speculate that such transitions may arise inthe cell when self-organization processes (cytoskeleton com-ponents assembly andmitosis) take place In our experimentthe annihilation of gravity enables the system to recovermoredegree of freedom through subsequent symmetry breakingswith the appearance of new morphological and functionalphenotypes

IndeedMDA-MB-231 cells exposed tomicrogravity werealmost equally split into two distinct populations char-acterized by very different morphologies The first cluster

is represented by cells adherent to the substrate roughlypreserving their native spindle profile The second one isrepresented by rounded smallest cells grouped and linkedto each other forming aggregates floating in the supernatant

Fractal analysis provides a quantitative assessment ofthose qualitative differences [17 29] Adherent cells inmicrogravity showed fractal values significantly higher thansuspended cells coherently roundness values were greaterin suspended than in adherent cells Additionally solidityestimation evidences how different these populations are interms of ldquopotentialrdquo deformability Solidity is a good descrip-tor of cell deformability indeed as it describes in geometricalterms the stiffness and deformability of an object Thus the


0020406080

100120140160

Apop

totic

cells

()


24h 72h

lowastlowastlowast

(a)

0

05

1

15

2

25

3

Relat

ive e

xpre

ssio

n

Bcl-2Bax


24h 72h


cells) clumps)


cells) clumps)

24h 72h

lowastlowast

(b)

cl-PARPGAPDH

05

0

1

15

2

25

3

Relat

ive e

xpre

ssio

n


24h 72h


cells) clumps)


cells) clumps)

24h 72h

lowastlowast

lowastlowast

(c)

Figure 5 Apoptosis analysis in MDA-MB-231 Apoptotic rate in RPM cultured MDA-MB-231 and on ground cells was determined by a dualparameter flow cytometric assay (a) Histograms show the percentage of apoptotic cells (Annexin V+7-AAD-) each column represents themean value plusmn SD of three independent experiments Immunoblot bar chart showing the expression of BaxBcl-2 ratio (b) and cleaved PARP(c) in on ground control cells RPM adherent cells and RPM cell clumps at 24 and 72 hours Columns and bars represent densitometricquantification of optical density (OD) of specific protein signal normalized with the OD values of the GAPDH served as loading controlEach column represents the mean value plusmn SD of three independent experiments lowast119875 lt 0 05 lowastlowast119875 lt 0 01 versus on ground control 119875 lt 0 05119875 lt 0 01 versus RPM adherent cells by ANOVA followed by Bonferroni post-test

higher the solidity is the lower the cell deformability isNonadherent cells growing in microgravity are grouped indiscrete clusters and they establish tight cell-to-cell contactsAs expected their solidity value is higher than that recordedin isolated adherent cells growing in RPM given thatthe multiple cell-to-cell adhesion is thought to ldquostabilizerdquothe cells shape by mutually reinforcing their stiffness Thecombined estimation of these parameters suggests that thetwo emerging populations significantly exhibit differences intheir respective morphological features

Aggregates of floating cells retain their viability potentialand after reseeding into a normal gravitational field they

are able to fully recover their native morphological traitsalready after 24 hours This is not really an unexpectedevent since it has been previously reported that microgravityexposed cellsmay recover their original profilewhen replacedin a normal gravity environment [30] Thereby gravity-related phenotypic variability may be considered an adaptivereversible phenomenon

Changes in cell shape are likely mediated by associatedmodification in cytoskeleton architecture which also conveysmechanical stress to the cell biochemicalgenetic machin-ery Therefore different cytoskeleton arrangements will endup in activating different gene sequences leading hence


00

05

10

15

20

25

30

35

Relat

ive e

xpre

ssio

n

AKTp-AKT


cells) clumps)

24hOn RPM RPM


72h

24h 72h



(a)

0

05

1

15

2

25

Rela

tive e

xpre

ssio

n

SurvivinGAPDH


cells) clumps)

24hOn RPM RPM


72h

24h 72h


lowastlowast

lowastlowast

lowast

lowast

(b)

005

115

225

335

4

Relat

ive e

xpre

ssio

n

ERKp-ERK


cells) clumps)

24hOn RPM RPM


72h

24h 72h


lowast

lowast

(c)

Figure 6 Survival pathways analysis in MDA-MB-231 Immunoblot bar chart showing the expression of p-AKTAKT ratio (a) Survivin(b) and p-ERKERK ratio (c) in on ground control cells RPM adherent cells and RPM cell clumps at 24 and 72 hours Columns and barsrepresent densitometric quantification of optical density (OD) of specific protein signal normalized with the OD values of the GAPDH servedas loading control Each column represents the mean value plusmn SD of three independent experiments lowast119875 lt 0 05 lowastlowast119875 lt 0 01 versus on groundcontrol 119875 lt 0 05 119875 lt 0 001 versus RPM adherent cells by ANOVA followed by Bonferroni post-test

to triggering different biochemical pathways The balancebetween tensional forces and the cytoskeleton architecturemodulates thereupon several complex cell functions likeapoptosis differentiation proliferation ECM remodellingand so forth [31] Compelling data demonstrated that bothsimulated and real space-based microgravity can severelyaffect cytoskeleton structure and function [8 32] The mostimpressive modifications were observed in nonadherentRPM-exposed cells in which stress fibers disappear and actinarchitecture is severely compromised thus jeopardizing thechances of cell anchoring to the substrate In the same cells

tubulin microfilaments are almost completely disorganizedThis finding may help in explaining the cell cycle inhibitionobserved in floating cell clumps given that a correct arrange-ment of the tubulin meshwork is required to ensure a properfunctioning of the mitotic process microtubules performindeed a special task during mitosis and meiosis by formingthe spindle assembly to align and separate the chromosomes[33] Yet it is worth of noting that cytoskeleton changes greatlydiffer between the two RPM-cultured populations outliningtherefore that microgravity enacted the emergence of twovery different cytoskeleton phenotypes


That architectural diversity is associated with significantdifferences in cell cycle and apoptosis Adherent breast cancercells growing in RPM are trying to counteract microgravityeffects by increasing the number of cells in the S phase and bystabilizing the apoptotic rate On the contrary suspended cellaggregates display a very different behavior characterized byreduced proliferative capability and enhanced apoptosis

However most of the cells in the floating clumps resultedto be viable in fact these cells readhered and grew up whenonce they were reseeded in normal gravity environmentHence cell population blocked in G2M underwent apopto-sis meanwhile cell population blocked in G0G1 recoveredthe original profile when they were reseeded

It is worth noting that such results have been obtainedby treating highly malignant growing cancer cells In thisregard both cell phenotypes cultured in RPM greatly differfrom their counterpart growing in 1 g gravity field Suchprocesses are remarkably mirrored by the concomitantcoherent changes in several biochemical pathways mech-anistically linked to both proliferation and programmedcell death Cyclin D1 a key regulatory factor for cellcycle G1S transition is significantly increased in adherentMDA-MB-231 cells meanwhile in suspended cell aggregatesCyclin D1 release is almost completely abolished Likewiseproapoptotic effectors (BAX PARP) dramatically increasein suspended RPM-cultured cells meanwhile prosurvivalfactors (Bcl-2 Survivin) significantly decrease Survivin awell-known critical factor triggering a plethora of survivalsignaling cascades was indeed dramatically downregulatedand resulted to be undetectable after 72 hours of exposi-tion Opposite findings were observed in adherent breastcancer cells exposed to microgravity the Bcl-2 inhibitor ofcaspase activation increases whereas proapoptotic effectorsconcomitantly decline

Regulation of apoptotic processes relies on the mod-ulation of an intricate interplay between several upstreammolecular pathways involving mainly activation of p-ERKand p-AKT expression As expected p-AKT and p-ERK weresignificantly reduced in suspended cell aggregates mean-while they increase in adherent apoptosis-resistant cellsOverall these results become evident already at early timesthat is after 24 hours of exposition

5 Conclusions

Our results confirm previous findings demonstrating thatmicrogravity enacted the emergence of distinct pheno-types characterized by significant recognizable differences inshape configuration biochemical pathways architecture andbehavioral processes [17] Additionally it should be remarkedthat the coexistence of two different cell populations maycontribute to explain some contradictory results providedby earlier studies [34 35] indeed increase or reduction incell proliferation as well as enhanced or reduced apoptosiscould well be both found during microgravity experimentsgiven that such opposite behaviors must be ascribed to verydifferent cell clusters

Spontaneous emergence of different phenotypes inmicrogravity after the system has experienced a symmetry

breaking is a finding worth of interest and may have relevantconsequences for human space flights Phenotypic switchleading to divergent morphological and biochemical config-uration is triggered by nonlinear processes taking place nearthe transition point Such transition enables the system torecover new degree of freedom and as such it may be viewedas a spontaneous process allowed by the nonequilibriumthermodynamics That finding highlights the relevance ofbiophysical constraints in shaping the form biological dis-sipative systems acquire [36] and may help understand howcells and tissues behave during development pathologicalevents or in extreme environmental fields



Acknowledgment

Thisworkwas partially supported by a grant fromASI (ItalianSpace Agency) COREA Program

References

[1] J C Buckey Space Physiology Oxford University Press 2006[2] G Albrecht-Buehler ldquoPossible mechanisms of indirect gravity

sensing by cellsrdquo ASGSB Bulletin vol 4 no 2 pp 25ndash34 1991[3] MHughes-Fulford andM L Lewis ldquoEffects ofmicrogravity on

osteoblast growth activationrdquo Experimental Cell Research vol224 no 1 pp 103ndash109 1996

[4] G Carmeliet and R Bouillon ldquoThe effect of microgravity onmorphology and gene expression of osteoblasts in vitrordquo TheFASEB Journal vol 13 no 8 pp S129ndashS134 1999


[6] D A Riley J L W Bain J L Thompson et al ldquoThin filamentdiversity and physiological properties of fast and slow fibertypes in astronaut leg musclesrdquo Journal of Applied Physiologyvol 92 no 2 pp 817ndash825 2002

[7] M L Lewis L A Cubano B Zhao et al ldquocDNA microarrayreveals altered cytoskeletal gene expression in space-flownleukemic T lymphocytes (Jurkat)rdquo The FASEB journal vol 15no 10 pp 1783ndash1785 2001

[8] D Vorselen W H Roos F C MacKintosh G J L Wuite andJ J W A van Loon ldquoThe role of the cytoskeleton in sensingchanges in gravity by nonspecialized cellsrdquoThe FASEB Journalvol 28 no 2 pp 536ndash547 2014

[9] A Cogoli ldquoSignal transduction in T lymphocytes inmicrograv-ityrdquo Gravitational and Space Biology Bulletin vol 10 no 2 pp5ndash16 1997

[10] J P Hatton F Gaubert M L Lewis et al ldquoThe kinetics oftranslocation and cellular quantity of protein kinaseC in humanleukocytes aremodified during spaceflightrdquoTheFASEB Journalvol 13 no 8 pp S23ndashS33 1999

[11] T GHammond E Benes K C OrsquoReilly et al ldquoMechanical cul-ture conditions effect gene expression gravity-induced changes


on the space shuttlerdquo Physiological Genomics vol 2000 no 3pp 163ndash173 2000

[12] J Renn D Seibt R Goerlich M Schartl and C WinklerldquoSimulated microgravity upregulates gene expression of theskeletal regulator Core binding Factor 1205721Runx2 inMedaka fishlarvae in vivordquo Advances in Space Research vol 38 no 6 pp1025ndash1031 2006

[13] J Boonstra ldquoGrowth factor-induced signal transduction inadherent mammalian cells is sensitive to gravityrdquo The FASEBJournal vol 13 no 8 pp S35ndashS42 1999

[14] M Bizzarri A Cucina A Palombo and M G MasielloldquoGravity sensing by cells mechanisms and theoretical groundsrdquoRendiconti Lincei Scienze Fisiche e Naturali vol 25 no 1 ppS29ndashS38 2014

[15] M Braun ldquoPrimary responses of gravity sensing in plantsrdquoin Biology in Space and Life on Earth Effects of Spaceflighton Biological Systems E Brinckmann Ed chapter 2 WileyWeinheim Germany 2007

[16] D K Kondepudi and P B Storm ldquoGravity detection throughbifurcationrdquo Advances in Space Research vol 12 no 1 pp 7ndash141992

[17] F Testa A Palombo and S Dinicola ldquoFractal analysis ofshape changes in murine osteoblasts cultured under simulatedmicrogravityrdquo Rendiconti Lincei Scienze Fisiche e Naturali vol25 no 1 pp S39ndashS47 2014

[18] J Vassy S Portet M Beil et al ldquoWeightlessness acts on humanbreast cancer cell line MCF-7rdquo Advances in Space Research vol32 no 8 pp 1595ndash1603 2003

[19] AQianW Zhang L Xie et al ldquoSimulatedweightlessness altersbiological characteristics of human breast cancer cell lineMCF-7rdquo Acta Astronautica vol 63 no 7ndash10 pp 947ndash958 2008

[20] J M Jessup M Frantz E Sonmez-Alpan et al ldquoMicrogravityculture reduces apoptosis and increases the differentiation ofa human colorectal carcinoma cell linerdquo In Vitro Cellular ampDevelopmental BiologymdashAnimal vol 36 no 6 pp 367ndash3732000

[21] E N Grigoryan H J Anton and V I Mitashov ldquoReal andsimulated microgravity can activate signals stimulating cellsto enter the S phase during lens regeneration in urodeleanamphibiansrdquoAdvances in Space Research vol 38 no 6 pp 1071ndash1078 2006

[22] K Nakamura H Kuga T Morisaki et al ldquoSimulated micro-gravity culture system for a 3-D carcinoma tissue modelrdquoBioTechniques vol 33 no 5 pp 1068ndash1076 2002


[24] A H Boulares A G Yakovlev V Ivanova et al ldquoRole ofpoly(ADP-ribose) polymerase (PARP) cleavage in apoptosisCaspase 3-resistant PARPmutant increases rates of apoptosis intransfected cellsrdquo The Journal of Biological Chemistry vol 274no 33 pp 22932ndash22940 1999

[25] G Nicolis and I Prigogine Exploring Complexity FreemanNew York NY USA 1989

[26] G Nicolis and I Prigogine ldquoSymmetry breaking and patternselection in far-from-equilibrium systemsrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 78 no 2 part 1 pp 659ndash663 1981

[27] J Tabony N Glade J Demongeot and C Papaseit ldquoBiologicalself-organization by way of microtubule reaction-diffusionprocessesrdquo Langmuir vol 18 no 19 pp 7196ndash7207 2002

[28] S J Moorman and A Z Shorr ldquoThe primary cilium as agravitational force transducer and a regulator of transcriptionalnoiserdquo Developmental Dynamics vol 237 no 8 pp 1955ndash19592008

[29] A R Qian D Li J Han et al ldquoFractal dimension as a measureof altered actin cytoskeleton inMC3T3-E1 cells under simulatedmicrogravity using 3-D2-D clinostatsrdquo IEEE Transactions onBiomedical Engineering vol 59 no 5 pp 1374ndash1380 2012

[30] R Coinu A Chiaviello G Galleri F Franconi E Crescenziand G Palumbo ldquoExposure to modeled microgravity inducesmetabolic idleness in malignant human MCF-7 and normalmurine VSMC cellsrdquo FEBS Letters vol 580 no 10 pp 2465ndash2470 2006

[31] M E Chicurel C S Chen and D E Ingber ldquoCellular controllies in the balance of forcesrdquo Current Opinion in Cell Biologyvol 10 no 2 pp 232ndash239 1998


[33] J L Carminati and T Stearns ldquoMicrotubules orient the mitoticspindle in yeast through dyne independent interactionswith thecell cortexrdquoThe Journal of Cell Biology vol 138 no 3 pp 629ndash641 1997

[34] L Vico ldquoWhat do we know about alteration in the osteoblastphenotype with microgravityrdquo The Journal of MusculoskeletalNeuronal Interactions vol 6 no 4 pp 317ndash318 2006

[35] M Hughes-Fulford ldquoPhysiological effects of microgravity onosteoblast morphology and cell biologyrdquo Advances in SpaceBiology and Medicine vol 8 pp 129ndash157 2002


Research ArticleOxidative Stress and NO Signalling in the Root Apex asan Early Response to Changes in Gravity Conditions

Sergio Mugnai12 Camilla Pandolfi1 Elisa Masi1 Elisa Azzarello1 Emanuela Monetti1

Diego Comparini1 Boris Voigt3 Dieter Volkmann3 and Stefano Mancuso1

1 DISPAA University of Florence Viale delle Idee 30 50019 Sesto Fiorentino Italy2HSO-USB ESTEC European Space Agency Keplerlaan 1 2200 AG Noordwijk The Netherlands3 IZMB University of Bonn Kirschallee 1 53115 Bonn Germany

Correspondence should be addressed to Sergio Mugnai sergiomugnaiunifiit



Copyright copy 2014 Sergio Mugnai et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Oxygen influx showed an asymmetry in the transition zone of the root apex when roots were placed horizontally on ground Theinflux increased only in the upper side while no changes were detected in the division and in the elongation zone Nitric oxide(NO) was also monitored after gravistimulation revealing a sudden burst only in the transition zone In order to confirm theseresults in real microgravity conditions experiments have been set up by using parabolic flights and drop tower The production ofreactive oxygen species (ROS) was also monitored Oxygen NO and ROS were continuously monitored during normal and hyper-and microgravity conditions in roots of maize seedlings A distinct signal in oxygen and NO fluxes was clearly detected only inthe apex zone during microgravity with no significant changes in normal and in hypergravity conditions The same results wereobtained by ROS measurement The detrimental effect of Drsquoorenone disrupting the polarised auxin transport on the onset of theoxygen peaks during the microgravity period was also evaluated Results indicates an active role of NO and ROS as messengersduring the gravitropic response with probable implications in the auxin redistribution

1 Introduction

During evolution plants have developed elaborate sensoryand signaling systems to cope with and adjust to rapidenvironmental changes Among them gravity remains aconstant stimulus playing a central role in driving theevolution of plants on Earth [1] Gravitropism involves a fineand reliable coordination of the activity of different cells andtissues deputed to gravity sensing with a growth responseoccurring in spatially distinct regions In roots for examplethe centrally located columella cells in the root cap are theproposed site of gravity sensing but the growth response(root curvature) occurs in the elongation zone (EZ)Themostcommon and accepted explanation for gravity sensing inplants is the starch-statolith hypothesis (which is the physicalsedimentation of starch-filled organelles called amyloplasts(statoliths) in gravity-sensing cells (statocytes) located at theroot tip) which triggers biochemical and physiological signals

[2] After the first event (sensing the change in the gravityvectorlevel) a signal is transduced and then transmitted tothe EZ stimulating the differential cellular growthmentionedabove This response is mainly driven by auxin [3] whichaccumulates to higher levels along the lower side of the rootthus provoking the inhibition of growth (Cholodny-Wenttheory) Recently other investigators highlighted the role ofcytoskeleton in the gravitropic response [4 5] Its central roleinmodulating cell polarity organellemovement intracellulartransport and cell expansion leads cytoskeleton to be astrong candidate in mediating the gravity signal transductioncascade (tensegrity model [6])

In the last years experimental evidences depicted thestructure of the root apex as divided into three different zonesa transition zone (TZ) located between two other regionsthe apical division zone (DZ) and the elongation zone (EZ)[7] The cells belonging to TZ have a specific cytoarchi-tecture with centralized postmitotic nuclei surrounded by



perinuclear microtubules radiating toward the cell peripheryIn contrast to the mitotically active DZ cells which arecontinually assembling and disassembling mitotic spindlesthe TZ cells are not deputed to perform these activities but tohave more specific sensory activities [3] Experimental datasuggest that the TZ is more a sort of sensory and informationprocessing zone devoted to a continuous monitoring ofthe environmental parameters and triggering appropriateresponses rather than being implicated in the division andgrowth processTheTZ cells are very sensitive to a wide rangeof stress sources such as touch [8] water and salt stress [9]aluminium [10] and hypoxia [11 12] However little is knownabout the role of TZ cells in the graviresponse especiallyrelated to the transductiontransmission phases after gravitysensing by the cells of the root tip

Among key signalling molecules in plants nitric oxide(NO) has recently emerged as an essential compound [13]Among its tasks NO regulates the actin cytoskeleton [14]endocytosis vesicle trafficking and the polarity of growingtip cells [15] root formation [16] and stomatal regulation [17]In addition NO is widely implicated in the plant response toenvironmental stress [18] but its exact role in the response ofplants to change in gravity levels is still unclear and not wellinvestigated

Reactive oxygen species (ROS) such asH2O2are generally

considered to be toxic by-products of respiration Howeverrecent experiments suggest that the production of ROSshould have an important and active role as components ofintracellular and extracellular signalling [19] In particularH2O2is starting to be accepted as a second messenger for

signals generated by means of ROS because of its relativelylong life and high permeability across membranes [20] Therole of ROS in the gravitropic response is still under debateas this topic has been rarely investigated [21]

Elucidating the mechanisms behind the signal transmis-sion from the site of gravity sensing to the site of gravibendingis therefore the main objective of this paper with particularinterest towards the role of the root apex (and in particular theTZ) in the transduction process and the importance of sens-ing molecules such as NO and H

2O2 in the gravity response

process After preliminary experiments on ground using themethod of gravistimulation via a horizontal displacement ofthe root the response of maize seedlings to gravity changeshas been studied for the first time ever in a real situationof microgravity thanks to a set of experiments performedduring three ESAparabolic flight campaigns and a drop towercampaign


21 Gravistimulation on Ground Oxygen fluxmeasurementswere performed using the vibrating probe technique [22]Briefly healthy Z mays L root apices (5-6mm long) werecut carefully washed with deionized water and placed indi-vidually at the bottom of a measuring chamber containingan electrophysiological solution (10mM CaCl

2 pH 65)

The flux measurement was performed at 24 plusmn 025∘C bypositioning a custom-built oxygen-selective microelectrode

(tip diameter of 1120583m[23 24]) near the root surface To ensurethe flux detection on the bottompart of a gravistimulated root(gravistimulation was performed by rotating the seedlingand the measurement system by 90 degrees) an electrodewith a hooked tip was also built During the recording themicroelectrode oscillated in a square wave parallel to theelectrode axis over a distance of 10 120583m (01Hz frequency)moving along the entire root length The calculation of thedifference between the voltage of each electrode position andthat of the previous one at the other extreme position as wellas the evaluation of amoving average of these differences overany desired time period producing the potential differencewere computer generated The O

2influxes were calculated

using Fickrsquos first law of diffusion assuming a cylindricaldiffusion geometry The flux measurements were performedon at least 10 different root apices per treatment (119899 ge 10)

To localize the production of NO in the different regionsof the root apex with a spatial resolution of a few microm-eters a NO-selective microelectrode of carbon fibers withdiameters as small as 5 120583m was constructed [12] using thesame system described above for the measurement Thedimensions and the response time (lt05 s) allowed the useof this electrode in a self-referencing mode [23 25] with aresolution as small as 50 fmol cmminus2 sminus1

22 Parabolic Flight Experiments All parabolic flight exper-iments were conducted aboard the Airbus A300 ZERO-Gwhich is operated by Novespace and is based in BordeauxFrance Every parabolic flight which lasts sim3 h includ-ing takeoff and landing encompasses 31 parabolas Everyparabola started from a steady normal horizontal flightand typically included 2 hypergravity (18 g) periods of 20 sseparated on average by a 22-s microgravity period (lt005 g)The first test parabola was followed by 6 series of 5 parabolasseparated by breaks of 4 and 8min respectively The datapresented emerged from the 41st and 45th Parabolic FlightCampaigns (PFC) and the ldquoFly Your Thesis 2012rdquo (studentcampaign) of the European Space Agency (ESA) represent-ing a total of 8 parabolic flights or 248 parabolas

During the 41st ESA Parabolic Flight Campaign (PFC)the measurements of oxygen influxefflux from the seedlingroots were conducted For each parabolic flight of the PFCa set of three 3-day-old seedlings of Zea mays L with ahomogeneous length of 5 plusmn 05 cm have been installed intoan Eppendorf vial (1 seedling = 1 vial) filled with an electro-physiological solution (10mMCaCl

2 pH 65)The fourth vial

was left empty and used as a control without seedlings Acouple ofO

2needlemicrosensors (OX50UnisenseDK) have

been horizontally inserted at two different levels for each vialcorresponding to the root apex and to the mature zone of theseedling root The tip of the electrode was placed close to theroot tissues (distance lt 1mm) The vials with the seedlingsand the electrodes were placed inside a thermostated cham-ber (temperature = 24-25∘C) Each electrode was connectedto a picoammeter (PA2000 Unisense DK) a four-channellaboratory amplifier that enables the measurement of mul-tiple parameters The output of the picoammeter was thenconnected to a datalogger A dedicated LabView software on


a laptop recorded the oxygenmeasurement Concurrently anaccelerometer provided gravity measurement

During the 45th ESA PFC the respiration rate of detachedroot apices (119899 = 6 with known weight) from 3-day-oldZea mays L seedlings was measured The root apices wereplaced inside an oximeter chamber (Oxytherm HansatechInstruments) with controlled temperature (25∘C) A smallmagnetic stirrer provided a continuous stirring of the solu-tion Measurements have been performed with the apicesin distilled water or in a solution containing Drsquoorenone(C18

ketone (5E7E)-6-methyl-8-(266-trimethylcyclohex-1-enyl) octa-57-dien-2-one 10 120583gmL) The oximeter chamberwas connected to a laptop with a dedicated software fordatalogging

During the ldquoFly your thesis 2012rdquo PFC campaign theproduction of H

2O2was assessed The Amplex Red reagent

was used after a preliminary evaluation in our lab due to itshigh sensitivity and successful use for the measure of H

2O2

production in plant root as previously reported by [26] Weused the Amplex Red Hydrogen PeroxidePeroxidase AssayKit (Invitrogen A22188) in combination with horseradishperoxidase (HRP Invitrogen) to detect H

2O2released from

biological samples In the presence of HRP Amplex redreagent reacts with H

2O2in a 1 1 stoichiometry to produce

the red-fluorescent oxidation product resorufin As first stepthe tips of maize roots were cut and immediately washedtwice for 15 minutes in PBS to eliminate ROS derived fromthe cut Tips were divided in samples constituted by 10mgof fresh tissue Then 50 120583L of working solution (containingdye and HRP) was added to 10mg of plant tissue (root)and the samples were incubated in a 96-well microplate atroom temperature for 30 minutes in darkness During theflight fluorescence was measured using excitation at 530 plusmn125 nm and fluorescence detection at 590plusmn175 nm by usinga microplate reader (Tecan Infinite 200 PRO) A dataloggerconnected to the Tecan and a laptop provided data storagewhich were then normalised to plant biomass

23 Drop Tower Campaign The Drop Tower in Bremen(Germany) is one of only a few facilities worldwide providinggravitational forces as small as 10minus5 g even if only for ashort time of 47 seconds The cylindrical falling capsule ofa diameter of 80 cm a height of 28m and a mass of 500 kg isdropped from 110m height of the tower whose inner tube isevacuated within 2 h to an air pressure of less than 10 mPaOn the bottom of the tower the capsule (reaching 170 kmhminus1) is decelerated within 130ms by a huge basket of app 25times 8m filled with styropor grains There the motion energy(6 times 105Nm) is converted into heat Gravitational sensorswere provided by ZARM Deceleration of the capsule leads togravitational values of about 30 g An oximeter (see ParabolicFlights section) was used and adapted for the measurementof nitric oxide by using selective microelectrodes (amiNO-30 Innovative Instruments Inc USA) connected to a NOelectrochemical detector with automatic temperature com-pensation (in NO-T-II Innovative Instruments Inc USA)The output of the detector was connected to a laptop via

200

175

150

125

100

75

50

25

0

175

150

125

100

75

50

25

0

Oxy

gen

influ

x (p

mol

cmminus

2sminus

1 )

00 05 10 15 20 25 30 35 40 45 50

Distance from root apex (mm)

DZ TZ EZ

Upper side

Bottom side

GravistimulatedNongravistimulated

Figure 1 Oxygen influx measured on the two sides of a gravistim-ulated and a vertical root Upper graph refers to the upper side of agravistimulated root whereas the second graph refers to the bottomside of a gravistimulated root

200

175

150

125

100

75

Oxy

gen

influ

x (p

mol

cmminus

2sminus

1 )

minus60 0 60 120 180 240 300

Time (s)

Upper side

Bottom side

Gra

visti

mul

atio

n

Figure 2 Timeline of the oxygen influx change at TZ level aftergravistimulation (time = 0)


14

12

10

8

6

4

2

14

12

10

8

6

4

2

14

12

10

8

6

4

2

0

1 3 5 7 9 11 13 15 17 19

Time (s)

Division zone

Transition zone

Elongation zone

NO

(nM

)N

O (n

M)

NO

(nM

)

minus3 minus1

Figure 3 Timeline of the NO production in a maize root apex inthe three different constituent zones after gravistimulation (time =0)

USB port running the inNO-T-II specific software for dataacquisition

3 Results

31 Gravistimulation on Ground In normal (vertical) con-ditions strong differences between the constituent zones ofthe maize root (DZ TZ and EZ) were clearly evident TheTZ appeared to be the most active zone in the uptake of

Table 1 Average time of burst appearance calculated for differentgroups of parabolas Data were analyzed by ANOVA using Tukeyrsquostest (119875 lt 005)

Group ofparabolas

Average time(secs)

Standarddeviation Significance

1ndash10 251 108 ns11ndash20 265 122 ns21ndash30 207 099 ns1ndash30 24 11 mdash

oxygen from the surrounding solution (Figure 1) The spatialpatterns of the oxygen influxes in the entire root apex showedamarked peak in theTZ (110 pmol cmminus2 sminus1) at 1ndash15mm fromthemaize root tip Aminor oxygen influx peak (75 pmol cmminus2sminus1) was also evident in the DZ Importantly the TZ was theonly root apex region significantly affected with regards togravistimulation in fact the marked peak of oxygen influxwas greatly enhanced in the upper part of the horizontal(gravistimulated) root whereas the DZ maintained a similarpattern On the contrary the bottom side of the horizontalroot showed a normal behaviour The increase of oxygeninflux at TZ level appeared as a very quick response followinggravistimulation as it was clearly evident after less than 30seconds in the upper side of the root (Figure 2) while thebottom part remained unaffected

Gravistimulation also promoted a very fast NO produc-tion from the root apex (Figure 3) A burst of NO was sud-denly produced after only 2-3 seconds from gravistimulationreaching a peak of 10 nM and lasting approximately 8 s beforereturning to the steady-state values in the TZ Only a smalland negligible efflux of NO was detectable in the DZ andimportantly NO bursts were not detected in the EZ region

32 Parabolic Flight and Drop Tower Campaigns A typicalparabola is shown in Figure 4 with a period of microgravity(lt005 g 20ndash22 secs) inserted between two periods of hyper-gravity (18 g 20 secs) Oxygen concentrations in the solutionmeasured at two different root levels (root apex and maturezone) are also reported for three different parabolas Burstsand peaks of oxygen concentration are clearly evident at rootapex level during the microgravity periods with significantdetection during the entire flight Interestingly no peakswere detected in the mature zone and during hypergravityperiods in the root apex Control without seedlings showedno activity demonstrating that the results previously shownwere not related to a background noise of themicroelectrodesduring microgravity

For each parabola the average detection time (1198791) of the

first oxygen peaks from the start of the microgravity period(119892 lt 1 119879

0 see Figure 5) has been calculated The average

time for all the parabolas is 24 secs In order to evaluatea difference in the appearance of the first peak during theflight the parabolas have been separated in three differentgroups (Table 1) Each group was composed of 10 parabolasThe objective was to investigate if a sort of ldquomemory effectrdquoof the stimulus during the repeated parabolas could cause a


400

375

350

325

300

275

250

225

200

Oxy

gen

conc

entr

atio

n (p

A)

0 10 20 30 40 50 60 70 80 90 100

Gra

vity

0

1Apex

Mature zone

Time (s)

(a)

Apex

Mature zone

400

375

350

325

300

275

250

225

200

Oxy

gen

conc

entr

atio

n (p

A)

0 10 20 30 40 50 60 70 80 90 100

Gra

vity

0

1

2

Time (s)

(b)

Apex

Mature zone

400

375

350

325

300

275

250

225

200

Oxy

gen

conc

entr

atio

n (p

A)

0 10 20 30 40 50 60 70 80 90

Time (s)

Gra

vity

0

1

2

Gravity

(c)

Apex-no root

Mature zone-no root

400

350

300

250

200

150

100

Oxy

gen

conc

entr

atio

n (p

A)

0 10 20 30 40 50 60 70 80 90 100

Gra

vity

0

1

2

Gravity

Time (s)

(d)

Figure 4 Oxygen concentration in the solution measured at two different levels close to (distance lt 1mm) the root apex and to the rootmature zone during three different parabolas one parabola at the beginning (a) one in the middle (b) and one at the end of the experiment(c) Control experiment with no roots inside the Eppendorf vial is shown in (d)

180

1

005

011

010

009

008

Oxy

gen

conc

entr

atio

n (p

A)

50 100

Time (s)

T0 T1

Gra

vity

Figure 5 Timeline of burst appearance (1198791) after the onset of

microgravity (1198790) during a parabola

different response during the time of the flight (higherloweranticipatedretarded)The results showed a reduction but notstatistically significant in the onset of the first peak during thelast set of parabolas

Interesting results have been also obtained from themeasurement of the respiration rate by oximeters The res-piration rate inside a single parabola has been divided intofive segments each segment being related to a differentgravity level 1 g 2 g 0 g 2 g (after microgravity) and 1 g(after microgravity) Negative values indicate oxygen influxRespiration by root apices led to an unavoidable reductionof the oxygen content in the solution due to the plantmetabolism thus the parabolas have been divided intodifferent groups by taking into account the real oxygenconcentration because the respiration rate is directly relatedto the amount of oxygen present in the solution Fourgroups related to different [O

2] in the solution have been

therefore created gt1500 nM 1000ndash1500 nM 500ndash1000 nMand lt500 nM The values relative to the control (Figure 6)show no significant differences among the different gravitylevels in each parabola for every oxygen concentration groupexcept for the last group ([O

2] lt 500 nM) with an increased

respiration rate during the second period of hypergravity Onthe contrary the presence ofDrsquoorenone in the solution did notlead to any variation neither in the respiration rate among thedifferent gravity levels nor compared to the control (Figure 7)


Control

ns ns

ns

2

0

minus2

minus4

minus6


1G 2G 0G 2G after 1G after1G 2G 0G 2G after 1G after

0

minus1

minus2

minus3

minus4

aa a

ab

b

minus03

minus06

minus09

minus12

minus15

minus18

minus21

minus24

025

000

minus025

minus050

minus075

minus100

O2

(nM

sminus1 )

O2

(nM

sminus1 )

O2

(nM

sminus1 )

O2

(nM

sminus1 )

gt1500nM O2 1500ndash1000nM O2


Figure 6 Respiration rate in control roots for the different groups related to the oxygen concentration in the solution

Drsquoorenone has been utilized in this experiment because itincreases PIN2 protein abundance without affecting PIN2transcripts with the consequence that the PIN2 expressiondomain enlarges and shifts basipetally resulting in moreactive auxin transport To deeply analyse the previous resultsthe behaviour of the respiration rate during a single parabolahas been evaluated It has been noted that when [O

2] was

lt 700 nM a sudden burst of oxygen was produced only inthe control a few seconds after the onset of microgravity(Figure 8)This large amount of oxygenwas quickly absorbedby the roots for respiration thus explaining the increasedrespiration rate during the second hypergravity period Thisphenomenon was clearly evident during each parabola with[O2] lt 700 nM The fact that the bursts were evident only

when [O2] lt 700 nM was probably due to the electrode

sensitivity which was not able to discriminate very lowdifferences in the respiration rate (around 25 nM)with higheroxygen concentrations in the solution These oxygen burstshave been characterised by calculating the area inside thecurve (Figure 9) The values of area response time after theonset of microgravity peak duration and peak amplitudeare reported in Figure 10 with a discrimination based onthe parabolasrsquo groups No significant differences among thegroups were noted in the peak area with an average value

of 27473 nM representing the moles of oxygen producedduring the microgravity period and then consumed in theresponse time after the onset of microgravity (average valueof 079 seconds) and the timing of the maximum peak (1103seconds) On the contrary significant differences amonggroups were registered in the peak duration The first 20parabolas had an average peak duration of 20-21 secondswhile the last 10 parabolas had a longer duration (averagevalue around 30 seconds) Finally peak amplitude showedno significant differences As expected when no roots werepresent in the oximeter a stable signal was registered (datanot shown)

Production ofH2O2measuredwithAmplex Red is shown

in Figure 11 Data were grouped according to the class ofgravity level Data recorded during microgravity (0 g) werestatistically compared with data recorded during normograv-ity (1 g) and hypergravity (2 g) conditions Transition from 1 gto 2 g and 2 g to 1 g had no significant effect on H

2O2 On the

contrary transition from 2 g to 0 g results stimulated a higherH2O2production from the root samples Interestingly we did

not observe any difference in H2O2production between 2 g

condition in comparison to 1 g Control experiment withoutany root tip inside the microplate showed no significantdifference between all the gravity levels (data not shown)


ns ns

nsns

minus10

minus15

minus20

minus25

minus30

minus35


minus050

minus075

minus100

minus125

minus150

minus175

minus200

minus225

minus025

minus050

minus075

minus100

minus125

minus150

minus175

minus200

025

000

minus025

minus050

Drsquoorenone



O2

(nM

sminus1 )

O2

(nM

sminus1 )

O2

(nM

sminus1 )

O2

(nM

sminus1 )

gt1500nM O21500ndash1000nM O2


Figure 7 Respiration rate in roots incubated in Drsquoorenone for the different groups related to the oxygen concentration in the solution

ControlDrsquoorenone

1200

1000

800

600

400

200

0

350 450 550 650 750 850 950

Time (s)

2

1

0

Gra

vity

O2

(nM

)

Figure 8 Respiration rate during single parabolas when [O2] in the

solution was lt700 nM Both control roots and roots incubated inDrsquoorenone are reported

Finally the production of NOwas detected and evaluatedduring an ESA Drop Tower campaign (Figure 12) Interest-ingly a burst of NO was clearly evident after 2 secondsfrom the start of the microgravity period which then started

O2

(nM

)

350

325

300

275

250

Time (s)T0 T1 Tmax Tend

Peak

Baseline T1

18

1

005

Gra

vity

Figure 9 Characterisation of an oxygen burst measured with anoximeter 119879

0is the time of the onset of microgravity 119879

1is the time

when the burst of oxygen begins Its value in the 119884-axis is taken asthe baseline for the calculation of the area amplitude and durationbetween 119879

1and 119879end

to decline resembling the behaviour of the gravistimulatedroots on ground Oximeter chambers without roots showed


O2

(nM

)

O2

(nM

)

500

400

300

200

100

0


Peak area Response time

Resp

onse

tim

e (s)

20

18

16

14

12

10

08

06

04

02

00

nsns

ns50

40

30

20

10

0

Peak amplitudePeak duration50

40

30

20

10

0

Peak

dur

atio

n (s

)

a

bb




Figure 10 Peak area response time peak duration and peak amplitude measured for different groups of parabolas by using the methoddescribed in Figure 9

no bursts or signal detected by the microelectrode (data notshown)

4 Discussion

Although both sensing (Cholodny-Went theory and tenseg-rity model) and signal transduction (role of auxin in the rootbending) in the gravitropic response are well and comprehen-sively described in the literature little is known about the greyarea of signal transmission the series of events comprisedbetween sensing and bending Hu et al [27] reported thatgravistimulation induced the asymmetric accumulation ofnitric oxide (NO) on the lower side of the apical region ofgravistimulated (horizontal) soybean seedling roots leadingto a subsequent auxin accumulation in the upper part Ourresults confirmed this hypothesis with a massive productionof NO in a very short time (2-3 seconds) Moreover we alsointegrated these results with the interesting information thatthe NO is mainly produced at TZ level thus confirming therole of TZ as a sensing zone of the root directly and activelyimplicated in the response to gravity changes

Gravistimulation also induced a sudden burst of oxygenin the upper part of TZ level 20ndash30 seconds after gravis-timulation Our results suggest the hypothesis that after the

displacement of statolythes under gravistimulation the chainof events is related to a sudden emission of NO which leadsto an improved plant metabolism which needs more oxygenfor respiration especially at TZ level to produce ATP tobe used as a source of energy Rapid changes in cytosolicCa2+ and pH have been proposed as components of thegravisignaling machinery [28] therefore it is plausible thatthe control of Ca2+ and H+ channels would require moreATP (ie more oxygen consumption) after gravistimulationand during microgravity The fact that oxygen burst at TZlevel following gravistimulation can be inhibited by BFA [3]and that TZ shows significant higher auxin secretion via theendocytic vesicle recycling [29] might correlate the oxygenbursts observed during gravistimulationunder microgravityand the auxin metabolism thus provoking a differentialgrowth response

In plants the simultaneous generation of O2and NO has

a synergistic function in defense responses [30] as well as inplants exposed to abiotic stress [31] NO is also generated atthe same time as ROS such as hydrogen peroxide for exam-ple during abiotic stress [32] Root gravitropism appears to beanother example of a physiological process in which bothNOand ROS play key roles in a simultaneous way [33] as ROSwere recently associated with auxin-mediated gravitropic


5

4

3

2

1

0

H2O

2(120583

mol

mgminus

1 )

1 I 2 I 0 2

(g)1 II

bbbb

a

Figure 11 H2O2production measured with Amplex Red during

different gravity conditionsThe letters I and II refer to the 1 g and 2 gphases prior (I) and after (II) the 0 g condition Data are presentedas average among each parabola of two parabolic flight days (62parabolas in total are considered) Data were analyzed by ANOVAData statistically different are indicated with different letters (119875 lt005) Error bars are also indicated

100

095

090

NO

(nM

)

minus5 0 5

Time (s)

Gravity

Gra

vity

(g)

NO

0

1

Figure 12 Typical nitric oxide curve measured in the oximeterchamber during a drop in the ESA Drop Tower campaign

responses in maize [21] and in the graviresponsive pulvinusof maize [34] In gravistimulated roots ROS accumulatedasymmetrically to the lower cortex within 30min of reori-entation becoming symmetrical upon longer stimulation[21] Interestingly Long et al [35] have shown that auxinasymmetries are detectable only after 2 h of gravistimulationof the pulvinus making the ROS changes reported muchfaster than the generation of gradients in auxin and so incontrast to the conclusions from the gravitropically respond-ing root possibly placing them upstream of the action ofthis hormone Our results support this hypothesis with thegeneration of an oxygen burst after a few seconds after theonset of microgravity which could be directly linked to theproduction of ROS as a stress messenger The fact that therespiration rate in root apices increased during microgravitycould also be related to the necessity of activating defensiveand scavenging mechanisms for ROS molecules In fact

the production of ROS during real microgravity has beenconfirmed during a parabolic flight campaign

Drsquoorenone rapidly and significantly activates theDR5 pro-moter [36] and also targets processes that are related to PIN2degradation [16] causing slower turnover and increasedprotein levels of this auxin efflux transporter thus suggestingthat this apocarotenoid interacts with auxin signaling at theroot apex Our results indicate that Drsquoorenone has also aninhibitor activity on the respiration rate and on the oxygenproduction thus giving indirect clue to a link between thesudden increase of oxygen during microgravity and auxinredistribution via PIN2 activity which is one of the majorresponses to changes in the gravity vectorlevels

5 Conclusions

For the first time ever a systemic and comprehensive seriesof experiments concerning the role of oxygen and stressmessengers (NO and ROS) during a real microgravity envi-ronment has been conductedThe timeline and the cascade ofevents detected during these experiments suggest an activerole of NO and ROS during the transmission step of thegravity response with probable implications in the auxinredistribution



Acknowledgments

The authors would like to thank ESA (European SpaceAgency) Novespace and ZARM for their kind supportduring the parabolic flight campaigns and the drop towercampaigns

References

[1] D Volkmann and F Baluska ldquoGravity one of the driving forcesfor evolutionrdquo Protoplasma vol 229 no 2ndash4 pp 143ndash148 2006

[2] E B Blancaflor ldquoRegulation of plant gravity sensing andsignaling by the actin cytoskeletonrdquoAmerican Journal of Botanyvol 100 no 1 pp 143ndash152 2013

[3] F Baluska and S Mancuso ldquoRoot apex transition zone asoscillatory zonerdquo Frontiers in Plant Science vol 4 article 3542013

[4] H Tatsumi T Furuichi M Nakano et al ldquoMechanosensitivechannels are activated by stress in the actin stress fibres andcould be involved in gravity sensing in plantsrdquo Plant Biologyvol 16 pp 18ndash22 2014

[5] F Baluka and D Volkmann ldquoMechanical aspects of gravity-controlled growth development and morphogenesisrdquo inMechanical Integration of Plant Cells and Plants pp 195ndash223Springer Berlin Germany 2011

[6] Y Chebli and A Geitmann ldquoGravity research on plants use ofsingle-cell experimental modelsrdquo Frontiers in Plant Science vol2 no 56 pp 1ndash10 2011


[7] F Baluska S Mancuso D Volkmann and P W Barlow ldquoRootapex transition zone a signalling-response nexus in the rootrdquoTrends in Plant Science vol 15 no 7 pp 402ndash408 2010

[8] H Ishikawa and M L Evans ldquoInduction of curvature inmaize roots by calcium or by thigmostimulation role of thepostmitotic isodiametric growth zonerdquo Plant Physiology vol100 no 2 pp 762ndash768 1992

[9] E S Ober and R E Sharp ldquoElectrophysiological responses ofmaize roots to low water potentials relationship to growth andABA accumulationrdquo Journal of Experimental Botany vol 54 no383 pp 813ndash824 2003

[10] M Amenos I Corrales C Poschenrieder P Illes F Baluskaand J Barcelo ldquoDifferent effects of aluminum on the actincytoskeleton and brefeldin A-sensitive vesicle recycling in rootapex cells of twomaize varieties differing in root elongation rateand aluminum tolerancerdquo Plant and Cell Physiology vol 50 no3 Article ID pcp013 pp 528ndash540 2009

[11] S Mugnai A M Marras and S Mancuso ldquoEffect of hypoxicacclimation on anoxia tolerance in vitis roots response ofmetabolic activity andK+ Fluxesrdquo Plant and Cell Physiology vol52 no 6 pp 1107ndash1116 2011

[12] S Mugnai E Azzarello F Baluska and S Mancuso ldquoLocal rootapex hypoxia induces no-mediated hypoxic acclimation of theentire rootrdquo Plant andCell Physiology vol 53 no 5 pp 912ndash9202012

[13] M Simontacchi C Garcıa-Mata C G Bartoli G E Santa-Marıa and L Lamattina ldquoNitric oxide as a key component inhormone-regulated processesrdquo Plant Cell Reports vol 32 no 6pp 853ndash866 2013

[14] K A Wilkins J Bancroft M Bosch J Ings N Smirnoff andV E Franklin-Tong ldquoReactive oxygen species and nitric oxidemediate actin reorganization and programmed cell death in theself-incompatibility response of papaverrdquo Plant Physiology vol156 no 1 pp 404ndash416 2011

[15] M C Lombardo and L Lamattina ldquoNitric oxide is essentialfor vesicle formation and trafficking in Arabidopsis root hairgrowthrdquo Journal of Experimental Botany vol 63 no 13 pp4875ndash4885 2012

[16] M Schlicht J Ludwig-Muller C Burbach D Volkmann and FBaluska ldquoIndole-3-butyric acid induces lateral root formationvia peroxisome-derived indole-3-acetic acid and nitric oxiderdquoNew Phytologist vol 200 no 2 pp 473ndash482 2013

[17] A M Distefano D Scuffi C Garcıa-Mata L Lamattina andA M Laxalt ldquoPhospholipase D120575 is involved in nitric oxide-induced stomatal closurerdquo Planta vol 236 no 6 pp 1899ndash19072012

[18] W Qiao C Li and L M Fan ldquoCross-talk between nitric oxideand hydrogen peroxide in plant responses to abiotic stressesrdquoEnvironmental and Experimental Botany vol 100 pp 84ndash932014

[19] S S Gill and N Tuteja ldquoReactive oxygen species and antioxi-dant machinery in abiotic stress tolerance in crop plantsrdquo PlantPhysiology and Biochemistry vol 48 no 12 pp 909ndash930 2010

[20] L J Quan B Zhang W W Shi and H Y Li ldquoHydrogenperoxide in plants a versatile molecule of the reactive oxygenspecies networkrdquo Journal of Integrative Plant Biology vol 50 no1 pp 2ndash18 2008

[21] J H Joo Y S Bae and J S Lee ldquoRole of auxin-induced reactiveoxygen species in root gravitropismrdquo Plant Physiology vol 126no 3 pp 1055ndash1060 2001

[22] C Pandolfi S Pollastri E Azzarello E Masi S Mugnai andS Mancuso ldquoThe vibrating probe technique in root studiesrdquo inMeasuring Roots pp 67ndash81 Springer Berlin Germany 2011

[23] S Mancuso G Papeschi and A M Marras ldquoA polarographicoxygen-selective vibrating-microelectrode system for the spa-tial and temporal characterisation of transmembrane oxygenfluxes in plantsrdquo Planta vol 211 no 3 pp 384ndash389 2000

[24] S Mancuso and A M Marras ldquoDifferent pathways of theoxygen supply in the sapwood of young Olea europaea treesrdquoPlanta vol 216 no 6 pp 1028ndash1033 2003

[25] S Mancuso and M Boselli ldquoCharacterisation of the oxygenfluxes in the division elongation and mature zones of Vitisroots influence of oxygen availabilityrdquo Planta vol 214 no 5pp 767ndash774 2002

[26] R Shin R H Berg and D P Schachtman ldquoReactive oxygenspecies and root hairs in arabidopsis root response to nitrogenphosphorus and potassium deficiencyrdquo Plant and Cell Physiol-ogy vol 46 no 8 pp 1350ndash1357 2005

[27] X Hu S J Neill Z Tang and W Cai ldquoNitric oxide mediatesgravitropic bending in soybean rootsrdquo Plant Physiology vol 137no 2 pp 663ndash670 2005

[28] G B Monshausen N D Miller A S Murphy and S GilroyldquoDynamics of auxin-dependent Ca2+ and pH signaling in rootgrowth revealed by integrating high-resolution imaging withautomated computer vision-based analysisrdquo The Plant Journalvol 65 no 2 pp 309ndash318 2011

[29] S Mancuso A M Marras V Magnus and F Baluska ldquoNonin-vasive and continuous recordings of auxin fluxes in intact rootapex with a carbon nanotube-modified and self-referencingmicroelectroderdquo Analytical Biochemistry vol 341 no 2 pp344ndash351 2005

[30] S Asai K Mase and H Yoshioka ldquoA key enzyme for flavinsynthesis is required for nitric oxide and reactive oxygen speciesproduction in disease resistancerdquo Plant Journal vol 62 no 6pp 911ndash924 2010

[31] G Tanou A Molassiotis and G Diamantidis ldquoHydrogenperoxide- andnitric oxide-induced systemic antioxidant prime-like activity under NaCl-stress and stress-free conditions incitrus plantsrdquo Journal of Plant Physiology vol 166 no 17 pp1904ndash1913 2009

[32] S J Neill R Desikan and J T Hancock ldquoNitric oxide signallingin plantsrdquo New Phytologist vol 159 no 1 pp 11ndash35 2003

[33] S Swanson and S Gilroy ldquoROS in plant developmentrdquo Physi-ologia Plantarum vol 138 no 4 pp 384ndash392 2010

[34] A M Clore S M Doore and S M N Tinnirello ldquoIncreasedlevels of reactive oxygen species and expression of a cytoplasmicaconitaseiron regulatory protein 1 homolog during the earlyresponse of maize pulvini to gravistimulationrdquo Plant Cell andEnvironment vol 31 no 1 pp 144ndash158 2008

[35] J C Long W Zhao A M Rashotte G K Muday and S CHuber ldquoGravity-stimulated changes in auxin and invertase geneexpression inmaize pulvinal cellsrdquo Plant Physiology vol 128 no2 pp 591ndash602 2002

[36] M Schlicht O Samajova D Schachtschabel et al ldquoDorenoneblocks polarized tip growth of root hairs by interfering with thePIN2-mediated auxin transport network in the root apexrdquo ThePlant Journal vol 55 no 4 pp 709ndash717 2008

Research ArticleCytoskeleton Modifications and Autophagy Induction inTCam-2 Seminoma Cells Exposed to Simulated Microgravity

Francesca Ferranti12 Maria Caruso2 Marcella Cammarota3

Maria Grazia Masiello45 Katia Corano Scheri2 Cinzia Fabrizi2 Lorenzo Fumagalli2

Chiara Schiraldi3 Alessandra Cucina56 Angela Catizone2 and Giulia Ricci3

1 Italian Space Agency (ASI) Via del Politecnico snc 00133 Rome Italy2 Department of Anatomy Histology Forensic Medicine and Orthopedics Sapienza University of RomeViale Regina Elena 336 00161 Rome Italy

3 Department of Experimental Medicine Second University of Naples Via Santa Maria di Costantinopoli 16 80138 Naples Italy4Department of Clinical and Molecular Medicine Sapienza University of Rome Viale Regina Elena 291 00161 Rome Italy5 Systems Biology Group Sapienza University of Rome Via A Scarpa 16 00161 Rome Italy6Department of Surgery ldquoPietro Valdonirdquo Sapienza University of Rome Viale del Policlinico 155 00161 Rome Italy

Correspondence should be addressed to Giulia Ricci giuliaricciunina2it

Received 12 May 2014 Revised 4 July 2014 Accepted 4 July 2014 Published 17 July 2014


Copyright copy 2014 Francesca Ferranti et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

The study of how mechanical forces may influence cell behavior via cytoskeleton remodeling is a relevant challenge of nowadaysthat may allow us to define the relationship between mechanics and biochemistry and to address the larger problem of biologicalcomplexity An increasing amount of literature data reported thatmicrogravity condition alters cell architecture as a consequence ofcytoskeleton structuremodifications Herein we are reporting themorphological cytoskeletal and behavioral modifications due tothe exposition of a seminoma cell line (TCam-2) to simulatedmicrogravity Even if no differences in cell proliferation and apoptosiswere observed after 24 hours of exposure to simulatedmicrogravity scanning electronmicroscopy (SEM) analysis revealed that thechange of gravity vector significantly affects TCam-2 cell surface morphological appearance Consistent with this observation wefound that microtubule orientation is altered by microgravity Moreover the confocal analysis of actin microfilaments revealed anincrease in the cell width induced by the low gravitational force Microtubules and microfilaments have been related to autophagymodulation and interestingly we found a significant autophagic induction in TCam-2 cells exposed to simulated microgravityThis observation is of relevant interest because it shows for the first time TCam-2 cell autophagy as a biological response inducedby a mechanical stimulus instead of a biochemical one

1 Introduction

An increasing number of experimental observations havedemonstrated that tissue homeostasis could be stronglyinfluenced and regulated by physical forces such as themodulation of gravity vector In the recent years many effortshave been made to elucidate the effect of microgravity oncell behavior and accumulating data show that micrograv-ity alters permanently or transiently important biological

processes such as mitosis differentiation survival cell mor-phology and gene expression profiles [1ndash7] However howcells sense these signals and convert them into a biochemicalresponse remains an important question that needs to beaddressed Recent studies have focused on the cytoskeletonas initial gravity sensor [1 8] Cytoskeleton plays importantroles in cell physiology being responsible for chromosomalsegregation during mitosis providing a mechanical supportto dividing cells contributing to maintain cell shape and



spatially organizing cell proteins and organelles in cell cyto-plasm Moreover cytoskeleton is involved in cell motilitymembrane trafficking signal transduction and cell adhesionIn addition cytoskeletal proteins can transduce and amplifymembrane receptor-captured signals transmitting the infor-mation to the nucleus and finally regulating gene expression[2 9 10] Considering all these observations it appearseasy to understand why cytoskeleton disorganization couldcompromise a lot of cell functions leading in some cases tocell death It is well known that microgravity exposure couldstrongly influence cytoskeleton organization [10ndash17] and itis commonly accepted that cellular tensegrity alteration inmicrogravity exposed cells could explain at least in part theconversion of a mechanical cue into a biological responseIn this regard recent studies have revealed the importanceof cytoskeletal integrity such as F-actin and microtubulesin the physiological specific aspects of autophagy and somepapers described the capability of microgravity to induceautophagy in living cells [18ndash22] Autophagy is an importanthousekeeping physiological process that is involved in cellu-lar remodeling during development elimination of aberrantorganelles or misfolded proteins and in the recycling ofunnecessary cellular components to compensate for thelimitation of nutrients during starvation It is of interestingnotice that this biological process is highly conserved fromyeast to mammals Despite several studies suggested a tumorsuppressive role for autophagy other reports support thehypothesis that this process is instead exploited by cancercells to prime their proliferation and promote their survival[23ndash27]

Microgravity condition is a stressful change in the physi-cal microenvironment for living cells however they seem tobe able to adapt to this change of gravitational force sincein the major part of studies available in the literature thebehavioral cellular modifications induced by microgravityare transient This observation has led to the intriguinghypothesis that cells in response to gravity changes reacttriggering adaptive biological processes and autophagy couldbe one of them

Testicular cells appear to be sensitive to microgravity ithas been demonstrated in fact that testicular function isimpaired by microgravity exposure [28ndash34] Moreover somein vitro observations revealed that microgravity influencescell proliferation apoptosis and testosterone secretion oftesticular organ cultures [35 36] In addition microgravitycondition has differentiating effect in cultured spermatocytesand influences germ cell survival [37 38] This effect onmale germ cell lineage has triggered the hypothesis that alsotesticular cancer germ cells could be altered by microgravitycondition For this reason we decided to study the effect ofmicrogravity on TCam-2 cells that are the only accreditedseminoma cell line [39ndash42] These cells have been recentlycharacterized at molecular and biochemical level [43ndash51]and thus represent a good tool to investigate male germcell behavior modification in response to a mechanical forcemodification In this paper we report for the first timecytoskeletal modifications and the activation of autophagicprocess induced by acute exposure to microgravity in TCam-2 cell line


21 Random Positioning Machine The random positioningmachine (RPM desktop RPM Dutch Space Leiden theNetherlands) we used in the investigation is a particularkind of 3D clinostat It consists of two independently rotatingframes One frame is positioned inside the other givinga very complex net change of orientation to a biologicalsample mounted in the middle The degree of microgravitysimulation depends on angular speed and on the inclinationof the disk These tools do not actually eliminate the gravitybut it is a microweight simulator based on the principleof ldquogravity-vector averagingrdquo it allows you to apply a 1 gstimulus omnidirectionally rather than unidirectionally andthe sum of the gravitational force vectors tends to zeroEffects generated by the RPM are comparable to those ofthe real microgravity provided that the direction changes arefaster than the response time of the system to gravity fieldThe desktop RPM we used has been positioned within anincubator (for maintaining temperature CO

2 and humidity


22 TCam-2 Cell Cultures The TCam-2 human cell line wasderived in 1993 from a primary testicular tumor sample ofpure classical seminoma [42] TCam-2 cells were culturedin RPMI 1640 (Lonza) supplemented with 10 fetal bovineserum (FBS Lonza) and penicillinstreptomycin (Invitrogen)at 37∘C in a humidified atmosphere with 5 carbon dioxide[41] The time 0 plating cell density is 3 times 104cm2 Asdescribed in the paragraph above microgravity conditionwas simulated using the randompositioningmachine (RPM)Experiments were performed on cells cultured for 24 and 48hours at 1 g or in RPM after additional 24 hours of preplatingon glass slides or IBIDImicroscopy chambers (IBIDI 80826)Glass slides were silicone fixed to the culture dishes at least 48hours before plating Cell culture dishes in both 1 g and RPMculture conditions were completely filled with fresh culturemedium in order to avoid air bubbles and to minimize liquidflow thus making negligible the effects of both buoyancy andshear stress during rotation

23 Proliferation Apoptosis and Autophagy QuantificationCells cultured at 1 g or under microgravity conditions (asdescribed above) were fixed in 4 paraformaldehyde (PFA)in phosphate buffered saline (PBS) 1X for 10 minutes at 4∘Cand permeabilized with 1 bovine serum albumin (BSA)01-Triton X-100 in PBS 1X for 1 hour at room temperature(RT) Nonspecific antibody binding was blocked with glycine1M pH 88 and with 1 BSA 01 Triton X-100 and 5donkey serum (Jackson ImmunoResearch Laboratories) inPBS 1X Cells were incubated overnight (ON) in PBS 1Xadded with 1 BSA01 Triton X-100 at 4∘C with thefollowing primary antibodies anticleaved Caspase-3 (CellSignaling rabbit polyclonal 9661 1 200 dilution) anti-p-histone H3 (Santa Cruz Biotechnology mouse monoclonalsc-374669 1 50 dilution) or anti-LC3 (Sigma-Aldrich L75431 120 dilution) After rinsing samples were incubated with


the opportune secondary antibody (FITC-conjugated donkeyanti-rabbit 711-095-152 or donkey anti-mouse 715-095-150IgG Jackson ImmunoResearch Laboratories 1 200 dilution)in PBS 1X for 90min at RT In negative controls primaryantibody was omitted After secondary antibody incubationsamples were washed and mounted in buffered glycerol(01M pH 95) All experiments were performed at least intriplicate

For proliferation and apoptosis analyses samples werephotographed with a Zeiss fluorescence microscope (Axio-scope) and positive cells were counted For LC3 immunolo-calization a Leica confocal microscope (Laser Scanning TCSSP2) equipped with ArArKr and HeNe lasers was usedImages were acquired utilizing the Leica confocal softwareThe laser line was at 488 nm for FITC excitation Theimages were scanned under a 20x objective or 40x oilimmersion objective In order to get a quantitative analysis offluorescence optical spatial series each composed of 2325optical sections with a step size of 2120583m were performed inareas in which cells reached confluence both in nonrotatedand in RPM cultured samples The fluorescence intensitywas determined by the Leica confocal software using thefollowing parameters the maximum amplitude of fluores-cence (MAX Amplitude) the sum of intensity (SUM (I))and the mean amplitude of fluorescence intensity (MEAN(A)) of LC3 positive areas The MAX Amplitude representsthe maximum intensity of fluorescence of each series TheSUM (I) represents the total amount of fluorescence intensityrecovered within the entire 119911-axis of each series The MEAN(A) represents the arithmeticalmean of fluorescence intensityrecovered within the entire 119911-axis of each series We analyzed12 equivalent sized regions (regions of interest (ROI)) for eachexperiment both in 1 g and in RPM culture conditions (36total ROI for each experimental condition)

24 Western Blotting of LC3 Autophagy Marker Cells cul-tured at 1 g and in RPM condition for 24 and 48 hourswere lysed in RIPA buffer (Sigma-Aldrich) Samples werethen clarified by centrifugation at 10000 rpm for 10minEquivalent amount of protein (10 120583g) from each samplewas electrophoretically resolved on 125 precast SDS-polyacrylamide gels (ExcelGel GE Healthcare Biosciences)using horizontal apparatus (Pharmacia Biotech UppsalaSweden) Then separated proteins were electrotransferredonto nitrocellulose membranes (Schleicher amp Schuell) by asemidry system (Novablot Pharmacia Biotech) Membraneswere blocked with 3 nonfat milk in PBS and then wereincubated (ON at 4∘C) with the LC3B monoclonal antibody(1 2000 Sigma) After extensive washing with PBS contain-ing 01 tween-20 (TBST) blots were incubated with 1 2000dilution of HRP-conjugated secondary antibody (AmershamBiosciences) for 1 hour at RT Immunopositive bands weredetected with a chemiluminescencersquos detection system (GEHealthcare Biosciences) To check for equal loading of thegel membranes were stripped and reprobedwithmouse anti-120573-actin antibody (1 20000 Sigma) and with anti-GAPDHantibody (1 1000 Cell Signalling Technology) Densitomet-ric analysis was performed with the Quantity One software(BioRad Laboratories)

25 F-Actin and Tubulin Distribution Pattern For F-actinvisualization Rhodamine Phalloidin (Invitrogen MolecularProbes Eugene 1 40 dilution) was used Cells were fixed in4 paraformaldehyde (PFA) in PBS for 10 minutes at 4∘Cand then permeabilized with cold ethanol Acetone 1 1 for10 minutes at 4∘C After rinsing cells were incubated withRhodamine Phalloidin for 25min in the dark Cells were thenwashed in PBS and mounted in buffered glycerol (01M pH95)

Cell height analysis (119911-axis) was performed using theconfocal microscope already described (Leica IRE SP2 LaserScanningTCS SP2) equippedwithArArKr andHeNe lasersImages of the optical sections were acquired using the Leicaconfocal software The Laser Line was at 543 nm for TRITCexcitation Images were scanned under a 40x oil objective Inorder to evaluate cell height three different experiments wereperformedusing cells cultured 1 g and inRPMconditions Foreach experiment 45 optical spatial series with a step size of2 120583mwere recovered and a total of at least 80 optical sectionswere analyzed for each experimental condition Cell heightof the examined samples was calculated using Leica confocalsoftware

For microtubules localization immunofluorescenceexperiments using anti-120572-tubulin (Biomeda mousemonoclonal V10178 1 75 dilution) as primary antibodywere performed The protocol used is the same alreadydescribed in the paragraph above Donkey anti-mouse(715-095-150 IgG Jackson ImmunoResearch Laboratories1 200 dilution) as secondary antibody was used Sampleswere then observed using both fluorescence microscope(Axioscope Zeiss) and confocal microscope (Leica)

26 Scanning Electron Microscopy Samples were fixed inGlutaraldehyde 25 in cacodylate buffer 01M pH 73 ONand then postfixed with 1 osmium tetroxide in cacodylatebuffer 1M dehydrated with increasing ethanol percentage(30ndash90 in water for 5min twice 100 for 15min) treatedin Critical Point Dryer (EMITECH K850) sputter coatedwith platinum-palladium (Denton Vacuum DESKV) andobserved with Supra 40 FESEM (Zeiss)

27 Statistical Analysis All experiments were performed atleast in triplicate All quantitative data are presented as themean value plusmn standard error (SEM) Studentrsquos 119905-test andANOVA test for multigroup comparison were carried outwhen appropriate to evaluate the significance of differencesThe significance level was fixed at a 119875 value lt 005


31 Microgravity Does Not Affect TCam-2 Cell Proliferationand Apoptosis Microgravity exposure is known to influencecell proliferation and apoptosis in normal and cancer cells[52] In order to asses proliferation rate of TCam-2 seminomacells maintained at 1 g or in RPM culture conditions for 24and 48 hours we performed immunofluorescence analysesof the M-phase marker p-histone H3 We observed thatactually this acute microgravity exposure does not affect


the number of mitotic cells at all the culture times considered(Figure 1) Literature data have demonstrated that TCam-2cells do not have a high proliferation rate (58 hours doublingtime) when compared with JKT1 (27 hours doubling time)that is another germ cell tumor cell line [40] Since thepercentage of proliferating cells we expect in the time frameof 24 and 48 hours is not high we can hypothesize that thisaltered gravitational stimulus is not long enough to determinea modification of cell proliferation in this particular cell lineInterestingly after 48 hours of culture the number of mitoticcells decreases significantly in a similar amount both in 1 gand in RPM cultured samples (Figure 1) indicating that cellproliferation in this particular cell line starts to be inhibitedby cell-to-cell contact even if these cells are cancer cells It hasto be noticed that we chose to plate cells at high density inorder to let them attach each other before the RPM exposureand react thanks to their tensional forces to the changes ofgravitational field Due to the high density of plating at theend of the longer culture time we analyzed cell culture dishesare crowded of cells so it appears not possible to prolongmore the culture without detaching and replate cells To thisregard it is fair to say that we cannot exclude that TCam-2 cellproliferationmight be altered by RPM exposure if they wouldhave been cultured at a different density

To test whether microgravity would be able to modifyTCam-2 cell apoptosis we performed immunofluorescencesfor the active fragment of the apoptosis marker Caspase-3We found that the change of gravity vector does not affect thenumber of apoptotic cells after 24 hours of culture (Figure 2)However it has to be noticed that after 48 hours of culturethe number of apoptotic cells increases significantly in theRPM cultured samples even if the large majority of cellsappear to tolerate this mechanical stress (Figure 2) and tosurvive The latter observation indicates that a small part ofTCam-2 cells appears more sensible to the change of gravityvector when the mechanical stimulus is prolonged a bitbut this sensibility does not seem related to mechanical cellstability because due to the high density of plating all cellsare stably attached to each other and to the substrate Inaddition apoptotic cells are observable uniformly dispersedin the culture dish On the basis of this observation wehypothesized that TCam-2 cells need to trigger rescue pro-cesses that let them survive after a prolonged change of gravityvector Possibly rescue processes are not correctly inducedor exploited by the whole population of TCam-2 cells andthis hypothesis may explain why a small percentage of themappears not able to survive to the change of gravity Thechange of physical forces is sensed by the cells through theircytoskeleton components and one of the first features thatreveal a cytoskeletal modification is the change in the plasmamembrane morphology We studied first membrane surfaceand cytoskeletal modifications due to RPM exposure to besure the TCam-2 cells are able to sense and modify theirshape in response to this mechanical stress Then we evalu-ated in the same culture conditions the autophagic processmodulation in response to RPM exposure since autophagyis the most known biological rescue mechanism that let cellto change rapidly and survive to suddenmicroenvironmentalchanges

32 Microgravity Strongly Influences TCam-2 Cell MembraneSurface To study if the alteration of the mechanical forcesacting on TCam-2 cells during microgravity simulation maymodify cell membrane surface morphology samples wereanalyzed by scanning electron microscopy We observed thepresence of two morphologically distinguishable cell popula-tions in the 1 g cultured samples one has smooth membranesurface and the other one is characterized by the presence ofmembrane expansions morphologically similar to microvilli(Figure 3) Noteworthy we found that microgravity stronglyaffects membrane surface appearance after 24 hours ofculture microvilli appeared collapsed and the differencesbetween the two cell populations are less evident (Figure 3)It is of interesting notice that cell microvilli are considered tobe an important site of mechanotransduction both in sensoryspecialized cells and not-sensory cells [53] After 48 hours ofculture the membrane surface differences appear recoveredand microvilli-like structures appear reconstituted in RPMcultured samples (Figure 3) On the basis of these obser-vations we hypothesized that cell mechanosensor-systemwas transiently altered by RPM exposure and this stronglysuggested the occurrence of cytoskeleton remodeling due toan acute exposure to gravitational vector change

33 Microgravity Induces TCam-2 Cytoskeleton RemodelingA huge amount of literature data demonstrated that micro-gravity is able to influence cell cytoskeletal architecture pro-moting cell morphofunctional alterations [54] In the light ofthese observations and on the basis of our scanning electronmicroscopy data we decided to evaluate the possible effectsof simulated microgravity on TCam-2 microfilament andmicrotubule organization Herein we report microfilamentdistribution pattern analyzed by F-actin staining of TCam-2 cells cultured at 1 g or in RPM culture conditions Even ifno apparent significant alterations in the actin cytoskeletonorganization were found both in 24 (Figure 4(a)) and 48hours of culture (not shown) a more detailed analysis byconfocal microscopy using Leica confocal software allowedus to evaluate cell height (cell 119911-axis) (Figures 4(b) 4(c)and 4(d)) in all the considered experimental conditions Weobserved that simulated microgravity significantly increasesTCam-2 cell height after 24 hours of RPM exposure withrespect to 1 g cultured cells (1562 plusmn 110 120583m versus 110 plusmn066 120583m 119875 lt 0001) indicating that RPM culture conditionwas able to modify TCam-2 cell shape Noteworthy after48 hours of culture the differences in cell height in 1 gand RPM cultured cells are no more statistically significant(Figure 4(d)) indicating that TCam-2 cells are able to recoverrapidly after the exposure to this mechanical stressThe latterobservation appears consistent with the reported recovery ofsurface membrane microvilli-like structures after 48 hours ofRPM exposure (Figure 3)

Microtubule distribution pattern was studied by anti-120572-tubulin immunofluorescence staining After 24 hours ofculture we observed that microtubule distribution is alteredinTCam-2 cells exposed toRPMculture condition centriolarpolarization is much less visible in these samples and micro-tubules appear to be distributed in an apparently random


p-H

iston

e H3

posit

ive c

ells

()

0

1

2

3

4

5

6

7

1GRPM

aa

bb

(h)24 48

(a)

(I) (II)

(b)

Figure 1 RPM exposure does not influence TCam-2 cell proliferation (a) Graphical representation of the percentage of proliferating cells(p-histone H3 positive cells) at 24 and 48 hours of culture No differences were observed between TCam-2 cells cultured at 1 g or in RPMculture conditions Data are expressed as the mean plusmn SEM Same letters indicate no statistical difference Different letters indicate 119875 lt 005(b) Representative images of TCam-2 cells cultured for 24 hours at 1 g (I) and in RPM condition (II) after p-histone H3 immunofluorescenceBar 50 120583m

manner within the cells (Figure 5) Microtubules are keyregulators of membrane trafficking organelle distributioninside the cells and together with actin microfilaments seemsto regulate autophagosome formation [55ndash57] In additionit is of interesting notice that LC3 the marker protein ofthe autophagic process is a microtubule associated protein(MAP) As well as actin filaments after 48 hours of culturethe microtubule distribution pattern appears recovered inRPM exposed samples since it is not possible to observesignificant differences between 1 g and RPM cultured cellsThese observations again clearly indicate the capability ofTCam-2 cell to sense the change of physical forces in theirmicroenvironment and also to recover rapidly from thisphysical stress These data strongly suggest the trigger ofrescue mechanisms due to TCam-2 RPM exposure

It is worth mentioning that the reported microtubulealteration does not appear to significantly alter the properformation of the mitotic spindle (Figure 5(g) white box)

This observation is consistent with the results reported inFigure 1 in which we observed that TCam-2 cell proliferationdoes not appear to be affected by RPM exposure

34 Microgravity Induces TCam-2 Cell Autophagy Somepapers in the literature reported that in other cellular sys-tems microgravity is involved in autophagy induction [18ndash20] and as previously stated cytoskeleton plays importantroles in autophagy regulation [22] In particular in mam-mals microtubules appear to be involved in the fusion ofautophagosome with late endosome and to bind and trans-port autophagosomes once terminally completedThe role ofactin filaments on mammalian autophagy process regulationis still a matter of debate but it is worth mentioning thatmicrofilaments depolymerization agents are able to blockautophagosome formation

TCam-2 cells cultured at 1 g and in RPM conditionswere immunostained to detect the autophagic marker LC3


123456789

10

Clea

ved

casp

ase-

3 po

sitiv

e cel

ls (

)

aa a

b

0

1GRPM

(h)24 48

(a)

(II)

(IV)

(I)

(III)

(b)

Figure 2 RPM exposure and TCam-2 cell apoptosis (a) Graphical representation of the percentage of apoptotic cell number (anticleavedCaspase-3 positive cells) No differences were observed between TCam-2 cells cultured for 24 hours at 1 g or in RPM culture conditions Onthe contrary a slight increase in apoptotic cell percentage is observed after 48 hours of culture Data are expressed as the mean plusmn SEM Sameletters indicate no statistical difference Different letters indicate 119875 lt 001 (b) Representative images of 1 g (I III) and RPM (II IV) exposedTCam-2 cells in 24 (I II) and 48 (III IV) hours of culture after cleaved Caspase-3 immunofluorescence Bar 50120583m (I and II) 35 120583m (III andIV)


RPM 24h1G 24h

RPM 48h

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

1G 48h

(i)

(l)

(m)

(n)

lowast

lowast

lowast

Figure 3 Microgravity effect on TCam-2 cell membrane surface Scanning electron microscopy pictures with increasing magnificationshowing cell membrane surfacemorphology of TCam-2 cells cultured for 24 (a b c and d) and 48 hours (i l) at 1 g or for 24 (e f g and h) and48 hours (m n) inRPMculture conditions In (a)white asterisks indicate TCam-2 cells with smoothmembrane surfacewhile the other TCam-2 cells of the same image are characterized by the presence of microvilli-like structures In (b) the boundary between one smooth membraneand one microvilli membrane presenting cells is reported (c) and (d) represent higher magnifications of the microvilli-like structures ofTCam-2 cells maintained at 1 g In (e) (f) (g) and (h) it is well evident that in RPM cultured cells membrane surface is more similar in allthe cells and it is difficult to clearly identify the two cell populations In particular in (h) it is possible to observe that microvilli-like structuresappeared collapsed in RPM exposed TCam-2 cells The morphological appearance of cell surface (i m) and microvilli-like structures (l n)appeared indistinguishable in 1 g (i l) and RPM exposed cells (m n) after 48 hours of culture


(I)

100200

300375

(I) (II)

RPM

0

5

10

15

20

1G RPM

48h culture

Cell

hei

ght (120583

m)

z-ax

is (c

ell w

idth

)z-

axis

(cel

l wid

th)

(I)(I)(I)(I) (II((II((IIII((( ))))))

1G

24h culture

24h culture

0

2

4

6

8

10

12

14

16

18 24h culture

1G RPM

Cell

hei

ght (120583

m)

(a)

(b)

(c)

(d)

lowast

100200

300375

x (120583m)x

(120583m)

100200

300375

100200

300375

(II)

x (120583m)x

(120583m)

Figure 4 Simulated microgravity influences TCam-2 cell height (a) Rhodamine-phalloidin staining of TCam-2 cells showing F-actindistribution pattern after 24 hours of culture at 1 g (I) or under RPM (II) conditions Bar 20120583m (b) Representative images of cell heightobtained using the Leica confocal software of samples cultured for 24 hours at 1 g (I) or in RPM (II) conditions (c) Graphical representationof cell height obtained by confocal microscopy analysis on 1 g and RPM exposed cells after 24 hours of culture (lowast1 562plusmn1 10120583m versus 110 plusmn066 120583m 119875 lt 0001) Data are expressed as the mean plusmn SEM (d) Graphical representation of cell height obtained by confocal microscopyanalysis on 1 g and RPM exposed cells after 48 hours of culture (1302 plusmn 132120583m versus 1602 plusmn 249 120583m resp) Data are expressed as themean plusmn SEMThe values are not statistically significant


1G 24h

(a)

RPM 48h1G 48h

(b)

(c)

(g)

(h)

RPM 24h

(d)

(e)

(f)

(i)

(l)

Figure 5 Microtubule distribution pattern in TCam-2 cells exposed to simulated microgravity Immunodetection of 120572-tubulin in TCam-2cells cultured for 24 hours (a b c d e and f) and 48 hours (g h i and l) at 1 g (a b c g and h) or under RPM conditions (d e f i and l) Inimages (g) and (i) in the white box representative images of mitotic spindles are also shown Bar 20120583m


As shown in Figures 6(a)(II) and 6(a)(IV) LC3 is detectableboth in 1 g and in RPM cultured samples and it ismainly localized in cytoplasmic vesicles Interestingly thenumber of these LC3 positive vesicles appears stronglyincreased in TCam-2 cells exposed to microgravity con-ditions (Figure 6(a)(IV)) with respect to 1 g cultured cells(Figure 6(a)(II)) after 24 hours of cultureMoreover a quanti-tative analysis carried out using the Leica confocal softwareallows us to quantify the fluorescence intensity increase ofLC3 stained cells exposed to simulated microgravity (Figures6(b) and 6(c)) In particular Figure 6(b) shows a stackprofile of 12 regions of interest (ROI) of a representativeexperiment both in 1 g (I) and in RPM cultured samples (II)The two groups of peaks reported in this figure representthe Max amplitude of fluorescence detected by the confocalmicroscope from the beginning to the end of the sample (total119911-axis) It is well evident thatMax amplitude of fluorescence isincreased in simulated RPM exposed samples We evaluatedalso the SUM (I) and the MEAN (A) of fluorescenceConsistent with the data reported in Figure 6(b) we observedalso an increase of both the SUM (I) and the MEAN (A) inRPM cultured cells after 24 hours of culture (Figure 6(c))According to the described confocal quantitative analyseswestern blots performed with the anti-LC3 antibody showedthat besides the increase of LC3-I protein amount LC3-II (the LC3 active isoform) protein content is increased inRPM with respect to 1 g cultured samples (Figure 7) Sameresults were obtained normalizing the LC3 bands versus 120573-actin (Figure 7) and versus GAPDH signal (not shown)Autophagy induction is a naturally transient process thisphenomenon is called autophagic flux [58] since when itworks autophagy protein machinery has to be degraded vialysosomes or proteasome together with the portion of the cellthat needs to be eliminated On the contrary when autophagyis blocked the autophagy protein machinery is not degradedand is maintained at high level in the cytoplasm In oursamples after 48 hours of culture autophagy active proteinLC3-II together with LC3-I appears quantitatively similar in1 g and RPM cultured cells demonstrating that autophagy isrestored at the same level with respect to 1 g culture conditionSame results were obtained normalizing the LC3 bands versus120573-actin (Figure 7) and versus GAPDH signal (not shown)Consistent with this observation the LC3 cytoplasmic fluo-rescence is lowered in the RPM exposed cells demonstratingthat autophagy was not blocked by this mechanical stress(Figure 6(a)(VI)) It has to be mentioned that LC3-II proteinis present at basal level at 24 and 48 hours of culture aswell as cytoplasmic LC3 dots even in cells cultured at 1 gindicating that autophagy is a housekeeping process thatworks inTCam-2 cells even in control samples and suggestingthat this cancer cell line may exploit autophagy as a survivalmechanism

There is a common agreement indicating that thereis a relationship between autophagy and apoptosis whenautophagy is not able to rescue cell frommicroenvironmentalchanges apoptotic process is triggered On the light of thistheory we might interpret the small increase in the apoptoticindex at 48 hours of culture in RPM cultured samples(Figure 2) as the autophagy efficiency threshold or the limit

of autophagy efficiency in the rescue of cell survival aftermechanical stress exposure

All together these qualitative and quantitative analysesallow us to conclude that microgravity is able to positivelymodulate the autophagic process in TCam-2 seminomacell line Autophagy induced in TCam-2 cells by Estrogenexposure through ER120573 activation was recently reported [59]Herein we reported for the first time autophagy induced inTCam-2 cells by a mechanical cue (or more precisely by aremoval of a mechanical stimulus) instead of a biochemicalone The analysis of the autophagy related pathways inducedby RPM exposure and the direct role of microtubules andmicrofilaments in this process as well as the other possiblebiological meanings of RPM induced TCam-2 autophagydeserves further investigations

4 Conclusions

Gravitational biology could be considered part ofmechanobi-ology the science that investigates the impact of forces onliving organisms At cellular level cytoskeleton elements arelikely candidates for force sensing and transduction pro-cesses These biomechanical properties of cell cytoskeletonexplain the capability to propagate a mechanical stimulusover long distances in living tissues and represent the basisof the intriguing hypothesis that many if not all reportedchanges in ion fluxes protein phosphorylation membranepotential changes and so forth are indeed provoked by amechanical modification somewhere within the cell or onits membrane [60 61] This paper is in line with this theoryand adds experimental data supporting the importance ofmechanotransduction and cell behavior In this paper in factwe reported the effects of the exposure to changes of gravityvector on TCam-2 seminoma cells In this experimentalmodel simulated microgravity is able to induce TCam-2 cellsurfacemodifications andmicrovilli-like structure alterationMoreover microtubules and microfilaments organizationresult to be influenced by microgravity (a) TCam-2 cellsshow actin cytoskeleton remodeling and cell height increase(b) centriolar polarization becomes much less visible in thesesamples and microtubules appear to be distributed in anapparent random manner within the cells All these modi-fications appear to be transient indicating that cells modifytheir cytoskeletal components in response to gravitationalforce change but that are also able to recover their shapewhen the gravitational change is prolonged InterestinglyRPM exposure is able to induce TCam-2 cell autophagyThe latter observation allows us to hypothesize that TCam-2 cells are able to rapidly respond to acute exposure tomicrogravity inducing adaptive biological processes suchas autophagy that probably allow them to survive in thechanging physical microenvironment Since autophagy isconsidered a biological survival mechanism the apoptosisinduction in a small percentage of TCam-2 cells after 48 hoursof culture might be speculated as the limit in the efficiencyof this survival process All together these data provideevidences of TCam-2 sensitivity to changes of gravitationalforce direction and lay the groundwork to further studies onTCam-2 cell autophagy and its biological meaning


(I) (III)

(IV) (VI)(II)

(V)

(a)

(120583m)

90

80

70

60

50

40

30

20

10

5 10 15 20 25 30 35 40

(I)Max amplitude

(120583m)

90

80

70

60

50

40

30

20

10

5 10 15 20 25 30 35 40

(II)Max amplitude

(b)

16

14

12

10

8

6

4

2

0

1G RPM

SUM

(I) (

au)

lowast

35

30

25

20

15

10

5

0

1G RPM

Mea

n(A

) (au

)

lowast

(c)

Figure 6 Autophagy induction in TCam-2 cells exposed to microgravity (a) Immunodetection of LC3 in TCam-2 cells cultured for 24 hoursat 1 g (II) or under RPM (IV) conditions In VI LC3 immunodetection of TCam-2 cells cultured in RPM condition for 48 hours is reported InI III and V the respective bright fields are shown (b) Stack profile of 24 hours of culture representative experiment showing the maximumamplitude (MAXAmplitude) of fluorescence in 12 regions of interest (ROI) randomly drawn in an area in which the cells reached confluencein nonrotated (I) and RPM cultured samples (II) It is evident an increase of maximum amplitude of fluorescence in microgravity exposedsamples (II) with respect to the 1 g-cultured cells (I) (c) MEAN (A) (lowast2 762plusmn104 versus 1434 plusmn 059 119875 lt 0001) and SUM (I) (lowast1 292plusmn085versus 695 plusmn 152 119875 lt 005) confirm an increase of LC3 positivity in RPM exposed sample with respect to 1 g cells after 24 hours of cultureData are expressed as the mean plusmn SEM


LC3-I

LC3-II

1G RPM

120573-Actin

24hours 48hours1G RPM

(a)

20

40

60

80

100

120

140

160

0

LC3-ILC3-II

1G RPM

Relat

ive i

nten

sity

(LC3

120573-a

ctin

)

24hours

lowastP le 005

lowast

(b)

20

40

60

80

100

120

140

0

LC3-ILC3-II

1G RPM

Relat

ive i

nten

sity

(LC3

120573-a

ctin

)

48hours

(c)

Figure 7 Western blot analysis of LC3 autophagy marker (a) Representative images of the bands revealed by anti-LC3 western blot analysison 24 and 48 hours cultured samples As expected anti-LC3 antibody detected both the LC3 isoforms (LC3-I cytosolic isoform LC3-IIautophagosomal membrane-conjugated isoform) (b) Graphical representation summarizing the densitometric analysis of the LC3-I andLC3-II bands normalized versus 120573-actin in 24 hours cultured samples Data are expressed as the mean plusmn DS lowastversus 1 g 119875 lt 005 (c)Graphical representation summarizing the densitometric analysis of the LC3-I and LC3-II bands normalized versus 120573-actin in 48 hourscultured samples Data are expressed as the mean plusmn DS The values are not statistically significant




Angela Catizone and Giulia Ricci are equal senior authors

Acknowledgments

The authors wish to thank Professor Mariano Bizzarri forthe valuable suggestions and the Italian Space Agency for thesupport received to carry out this research project

References

[1] D Vorselen W H Roos F C MacKintosh G J Wuite and JJ van Loon ldquoThe role of the cytoskeleton in sensing changes ingravity by nonspecialized cellsrdquoThe FASEB Journal vol 28 no2 pp 536ndash547 2014

[2] G Aleshcheva J Sahana X Ma et al ldquoChanges inmorphologygene expression and protein content in chondrocytes culturedon a random positioning machinerdquo PLoS ONE vol 8 no 11Article ID e79057 2013

[3] S Li Z Ma Z Niu et al ldquoNASA-approved rotary bioreactorenhances proliferation and osteogenesis of human periodontalligament stem cellsrdquo Stem Cells and Development vol 18 no 9pp 1273ndash1282 2009


[5] S J PardoM J PatelMC Sykes et al ldquoSimulatedmicrogravityusing the Random Positioning Machine inhibits differentiationand alters gene expression profiles of 2T3 preosteoblastsrdquoAmerican Journal of Physiology-Cell Physiology vol 288 no 6pp C1211ndashC1221 2005

[6] A Guignandon M H Lafage-Proust Y Usson et al ldquoCellcycling determines integrin-mediated adhesion in osteoblasticROS 1728 cells exposed to space-related conditionsrdquo TheFASEB Journal vol 15 no 11 pp 2036ndash2038 2001




[9] T D Ross B G Coon S Yun et al ldquoIntegrins inmechanotrans-ductionrdquo Current Opinion in Cell Biology vol 25 no 5 pp 613ndash618 2013

[10] B Geiger A Bershadsky R Pankov and K M YamadaldquoTransmembrane extracellular matrix-cytoskeleton crosstalkrdquoNature Reviews Molecular Cell Biology vol 2 no 11 pp 793ndash805 2001

[11] M Y Kapitonova N Salim and S Othman ldquoAlteration of cellcytoskeleton and functions of cell recovery of normal humanosteoblast cells caused by factors associated with real spaceflightrdquo Malaysian Journal of Pathology vol 35 no 2 pp 153ndash163 2013

[12] G Pani N Samari R Quintens et al ldquoMorphological andphysiological changes in mature in vitro neuronal networkstowards exposure to short- middle- or long-term simulatedmicrogravityrdquo PLoS ONE vol 8 no 9 Article ID e73857 2013

[13] J Nakashima F Liao J A Sparks Y Tang and E B BlancaflorldquoThe actin cytoskeleton is a suppressor of the endogenous skew-ing behaviour of Arabidopsis primary roots in microgravityrdquoPlant Biology vol 16 supplement 1 pp 142ndash150 2013

[14] C Nouri J A Tuszynski M W Wiebe and R Gordon ldquoSimu-lation of the effects of microtubules in the cortical rotation ofamphibian embryos in normal and zero gravityrdquo BioSystemsvol 109 no 3 pp 444ndash449 2012

[15] J Li S Zhang J Chen T Du Y Wang and ZWang ldquoModeledmicrogravity causes changes in the cytoskeleton and focaladhesions and decreases in migration in malignant humanMCF-7 cellsrdquo Protoplasma vol 238 no 1ndash4 pp 23ndash33 2009



[18] H W Ryu S H Choi S Namkoong et al ldquoSimulated micro-gravity contributes to autophagy induction by regulating AMP-activated protein kinaserdquo DNA and Cell Biology vol 33 no 3pp 128ndash135 2014

[19] Y SambandamM T Townsend J J Pierce et al ldquoMicrogravitycontrol of autophagy modulates osteoclastogenesisrdquo Bone vol61 pp 125ndash131 2014

[20] Y C Wang D Y Lu F Shi et al ldquoClinorotation enhancesautophagy in vascular endothelial cellsrdquo Biochemistry and CellBiology vol 91 no 5 pp 309ndash314 2013

[21] D Sandona J Desaphy G M Camerino et al ldquoAdaptation ofmouse skeletal muscle to long-term microgravity in the MDSmissionrdquo PLoS ONE vol 7 no 3 Article ID e33232 2012

[22] I Monastyrska E Rieter D J Klionsky and F ReggiorildquoMultiple roles of the cytoskeleton in autophagyrdquo BiologicalReviews vol 84 no 3 pp 431ndash448 2009

[23] J H Choi Y S Cho Y H Ko S U Hong J H Park andM A Lee ldquoAbsence of autophagy-related proteins expressionis associated with poor prognosis in patients with colorectaladenocarcinomardquo Gastroenterology Research and Practice vol2014 Article ID 179586 10 pages 2014

[24] N Orfali S L McKenna M R Cahill L J Gudas and N PMongan ldquoRetinoid receptor signaling and autophagy in acutepromyelocytic leukemiardquo Experimental Cell Research vol 324no 1 pp 1ndash12 2014

[25] C Cerella M H Teiten F Radogna M Dicato and MDiederich ldquoFrom nature to bedside Pro-survival and celldeath mechanisms as therapeutic targets in cancer treatmentrdquoBiotechnology Advances 2014

[26] C Fabrizi V S De F Somma et al ldquoLithium improves survivalof PC12 pheochromocytoma cells in high-density cultures andafter exposure to toxic compoundsrdquo International Journal of CellBiology vol 2014 Article ID 135908 7 pages 2014

[27] L Yu L Strandberg and M J Lenardo ldquoThe selectivity ofautophagy and its role in cell death and survivalrdquo Autophagyvol 4 no 5 pp 567ndash573 2008

[28] F Strollo G Riondino B Harris et al ldquoThe effect of micro-gravity on testicular androgen secretionrdquo Aviation Space andEnvironmental Medicine vol 69 no 2 pp 133ndash136 1998

[29] F Strollo M A Masini M Pastorino et al ldquoMicrogravity-induced alterations in cultured testicular cellsrdquo Journal ofGravitational Physiology vol 11 no 2 pp P187ndash188 2004

[30] Y Ding J Tang J Zou et al ldquoThe effect of microgravity ontissue structure and function of rat testisrdquo Brazilian Journal ofMedical and Biological Research vol 44 no 12 pp 1243ndash12502011

[31] T Kaneko S Sasaki Y Umemoto Y Kojima T Ikeuchiand K Kohri ldquoSimulated conditions of microgravity increasesprogesterone production in I-10 cells of Leydig tumor cell linerdquoInternational Journal ofUrology vol 15 no 3 pp 245ndash250 2008

[32] MAHMotabagani ldquoMorphological andmorphometric studyon the effect of simulated microgravity on rat testisrdquo ChineseJournal of Physiology vol 50 no 4 pp 199ndash209 2007

[33] F Strollo G Strollo M More et al ldquoChanges in humanadrenal and gonadal function onboard Spacelabrdquo Journal ofGravitational Physiology vol 4 no 2 pp 103ndash104 1997

[34] U Engelmann F Krassnigg and W- Schill ldquoSperm motilityunder conditions of weightlessnessrdquo Journal of Andrology vol13 no 5 pp 433ndash436 1992

[35] G Ricci A Catizone R Esposito and M Galdieri ldquoMicro-gravity effect on testicular functionsrdquo Journal of gravitationalphysiology vol 11 no 2 pp 61ndash62 2004

[36] G Ricci R Esposito A Catizone and M Galdieri ldquoDirecteffects of microgravity on testicular function analysis of hys-tological molecular and physiologic parametersrdquo Journal ofEndocrinological Investigation vol 31 no 3 pp 229ndash237 2008

[37] S di Agostino F Botti A di Carlo C Sette and R GeremialdquoMeiotic progression of isolated mouse spermatocytes undersimulatedmicrogravityrdquo Reproduction vol 128 no 1 pp 25ndash322004

[38] M Pellegrini S di SienaGClaps et al ldquoMicrogravity promotesdifferentiation andmeiotic entry of postnatal mouse male germcellsrdquo PLoS ONE vol 5 no 2 Article ID e9064 2010

[39] J de Jong H Stoop A J M Gillis et al ldquoFurther char-acterization of the first seminoma cell line TCam-2rdquo GenesChromosomes and Cancer vol 47 no 3 pp 185ndash196 2008

[40] D Eckert D Nettersheim L C Heukamp S Kitazawa K Bier-mann and H Schorle ldquoTCam-2 but not JKT-1 cells resembleseminoma in cell culturerdquo Cell and Tissue Research vol 331 no2 pp 529ndash538 2008

[41] N C Goddard A McIntyre B Summersgill D Gilbert SKitazawa and J Shipley ldquoKIT and RAS signalling pathways


in testicular germ cell tumours new data and a review of theliteraturerdquo International Journal of Andrology vol 30 no 4 pp337ndash348 2007

[42] YMizuno A Gotoh S Kamidono and S Kitazawa ldquoEstablish-ment and characterization of a new human testicular germ celltumor cell line (TCam-2)rdquoNihon Hinyokika Gakkai Zasshi vol84 no 7 pp 1211ndash1218 1993

[43] D Nettersheim L C Heukamp F Fronhoffs et al ldquoAnalysisof TET expressionactivity and 5mC oxidation during normaland malignant germ cell developmentrdquo PLoS ONE vol 8 no12 Article ID e82881 2013

[44] F Ferranti B Muciaccia G Ricci et al ldquoGlial cell line-derivedneurotrophic factor promotes invasive behaviour in testicularseminoma cellsrdquo International Journal of Andrology vol 35 no5 pp 758ndash768 2012

[45] F Ferranti F DAnselmi M Caruso et al ldquoCorrection TCam-2 seminoma cells exposed to egg-derived microenvironmentmodify their shape adhesive pattern and migratory behavioura molecular and morphometric analysisrdquo PLoS ONE vol 8 no10 2013

[46] S M Russell M G Lechner A Mokashi et al ldquoEstablishmentand characterization of a new human extragonadal germ cellline SEM-1 and its comparison with TCam-2 and JKT-1rdquoUrology vol 81 no 2 pp 464ndashe9 2013

[47] R Franco F Boscia V Gigantino et al ldquoGPR30 is overex-pressed in post-puberal testicular germ cell tumorsrdquo CancerBiology ampTherapy vol 11 no 6 pp 609ndash613 2011

[48] F Esposito F Boscia V Gigantino et al ldquoThe high-mobilitygroup A1-estrogen receptor 120573 nuclear interaction is impairedin human testicular seminomasrdquo Journal of Cellular Physiologyvol 227 no 12 pp 3749ndash3755 2012

[49] DNettersheimA JMGillis LH J Looijenga andH SchorleldquoTGF-1205731 EGF and FGF4 synergistically induce differentiationof the seminoma cell line TCam-2 into a cell type resemblingmixed non-seminomardquo International Journal of Andrology vol34 no 4 part 2 pp e189ndashe203 2011

[50] D Nettersheim A Gillis K Biermann L H J Looijengaand H Schorle ldquoThe seminoma cell line TCam-2 is sensitiveto HDAC inhibitor depsipeptide but tolerates various otherchemotherapeutic drugs and loss of NANOG expressionrdquoGenes Chromosomes and Cancer vol 50 no 12 pp 1033ndash10422011

[51] U Eppelmann F Gottardo J Wistuba et al ldquoRamanmicrospectroscopic discrimination of TCam-2 culturesreveals the presence of two sub-populations of cellsrdquo Cell andTissue Research vol 354 no 2 pp 623ndash632 2013


[53] O P Hamill and B Martinac ldquoMolecular basis of mechan-otransduction in living cellsrdquo Physiological Reviews vol 81 no2 pp 685ndash740 2001


[55] A Aplin T Jasionowski D L Tuttle S E Lenk and W ADunn Jr ldquoCytoskeletal elements are required for the formationand maturation of autophagic vacuolesrdquo Journal of CellularPhysiology vol 152 no 3 pp 458ndash466 1992

[56] E Fass E Shvets I Degani K Hirschberg and Z ElazarldquoMicrotubules support production of starvation-induced

autophagosomes but not their targeting and fusion withlysosomesrdquo Journal of Biological Chemistry vol 281 no 47 pp36303ndash36316 2006

[57] R Kochl X W Hu E Y W Chan and S A ToozeldquoMicrotubules facilitate autophagosome formation and fusionof autophagosomes with endosomesrdquo Traffic vol 7 no 2 pp129ndash145 2006

[58] N Mizushima T Yoshimori and B Levine ldquoMethods inMammalian Autophagy Researchrdquo Cell vol 140 no 3 pp 313ndash326 2010

[59] CGuido S PanzaM Santoro et al ldquoEstrogen receptor120573 (ER120573)produces autophagy and necroptosis in human seminoma cellline through the binding of the Sp1 on the phosphatase andtensin homolog deleted fromchromosome 10 (PTEN) promotergenerdquo Cell Cycle vol 11 no 15 pp 2911ndash2921 2012

[60] J J W A van Loon ldquoMechanomics and physicomics ingravisensingrdquo Microgravity Science and Technology vol 21 no1-2 pp 159ndash167 2009

[61] M Bizzarri A Cucina A Palombo and M G MasielloldquoGravity sensing by cells mechanisms and theoretical groundsrdquoRendiconti Lincei vol 25 pp 29ndash38 2014

Research ArticleGravity Affects the Closure of the Traps in Dionaea muscipula

Camilla Pandolfi1 Elisa Masi1 Boris Voigt2 Sergio Mugnai1

Dieter Volkmann2 and Stefano Mancuso1

1 DISPAA University of Florence Viale delle idee 30 50019 Sesto Fiorentino Italy2 IZMB University of Bonn Kirschallee 1 53115 Bonn Germany

Correspondence should be addressed to Camilla Pandolfi camillapandolfiunifiit

Received 12 May 2014 Accepted 27 June 2014 Published 15 July 2014


Copyright copy 2014 Camilla Pandolfi et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Venus flytrap (Dionaea muscipula Ellis) is a carnivorous plant known for its ability to capture insects thanks to the fast snapping ofits traps This fast movement has been long studied and it is triggered by the mechanical stimulation of hairs located in the middleof the leaves Here we present detailed experiments on the effect of microgravity on trap closure recorded for the first time during aparabolic flight campaign Our results suggest that gravity has an impact on trap responsiveness and on the kinetics of trap closureThe possible role of the alterations of membrane permeability induced by microgravity on trap movement is discussed Finally weshow how the Venus flytrap could be an easy and effective model plant to perform studies on ion channels and aquaporin activitiesas well as on electrical activity in vivo on board of parabolic flights and large diameter centrifuges

1 Introduction

The response of Venus flytrap (Dionaea muscipula) tomechanical stimulation has long been known and it is oneof the most rapid movements in the plant kingdom [1 2]

The plant produces a rosette of leaves each divided intotwo parts a lower part called the lamina and the upper partcalled the trap The trap catches prey thanks to a very rapidmovement of its bilobed halves that shut when the triggerhairs are stimulated At room temperature two touches acti-vate the trap which snaps shut in a fraction of second [3]At higher temperature only one stimulus is required for trapclosure [4] The stimulation of the trigger hairs activatesmechanosensitive ion channels and generates receptor poten-tials inducing the action potentials (APs) that initiate theclosure [5] electrical signals are the immediate cause of thetrapmovements irrespective of the way in which the signal istriggered (mechanical stimulation or electrostimulation) [5]Once the insect is caught the lobes seal tightly allowing diges-tion to take place [6 7] The APs in Dionaea muscipula havebeen extensively studied (eg [5 8 9]) Trigger hair-inducedgeneration of action potentials is not exclusively associatedwith the trap closure The struggling of the entrapped preyin the closed trap results in the generation of further action

potentials which cease to occur just when the prey stopsmovingTheseAPsmay induce inhibition of the dark reactionof photosynthesis [10] showing that chlorophyll-A fluores-cence is under electrochemical control [11] Although thisspectacular example of plant movement has long fascinatedscientists the mechanism by which the trap works remainspoorly understood [12] Some explanations proposed involvean irreversible cell wall loosening induced by the acidifica-tion of the cells [6] or a rapid loss of turgor pressure similarlyto what happens in stomata [13] However the validity ofboth mechanisms has been questioned because they cannotexplain the speed at which the movement happens Morerecently other two models have been proposed the elasticdeformation that results from a snap-buckling instability [14]and a hydroelastic curvature mechanism based on the fastopening ofwater channels [9] Bothmodelsmay convincinglyaccount for the speed of the movement

Thepossibility to study the effect ofmicrogravity on livingorganisms is a unique opportunity to observe the alterationof phenomena in the absence of the otherwise omnipresentgravity force Although the understanding of the effect ofgravity on animal and plant bodies is crucial in view of thepossible future space travels the research is moving slowly if



compared with other research fields due to the accessibilityto microgravity conditions and the challenging experimentalcondition

In the present study we report the effect of microgravityon trap closure conducted during a parabolic flight campaignThis gave us the opportunity to test Dionaea muscipula aspossible candidate to study the effect of gravity on the elec-trical activity of organisms bymonitoring the variation in theexcitability of the traps and the alteration in the kinetics oftheir closure Furthermore the changes in the kinetics of trapclosure gave us important hints on themechanism at the baseof the fast trap snapping


Parabolic flight experiments were performed in an A300 air-plane (Novespace France) during the 9th DLR parabolicflight campaign A typical parabolic flight manoeuvre pro-vides alternating acceleration levels of regular gravity (1 g)microgravity (0 g) for 22 s and two periods (20 s) of hyper-gravity (up to 18 g for 20 s) before and after each period ofmicrogravity Twenty Dionaea muscipula J Ellis plants weregrown in a growth chamber with 1410 h lightdark period inwell-drained peat moss in plastic pots irrigated with distilledwater All experiments were performed on healthy adult spec-imens The pots were sealed with parafilm to avoid the float-ing of the substrate during the zero gravity periods and plantswere secured inside a plexiglass growth chamber Digital HDvideo recorder Sony HDR-SR11E was used to film the Venusflytraps at 25 fps Each day experiments were done on 4different plants All the traps were mechanically stimulatedduring 0 1 and 2 g using a wooden stick by gently touchingthe trigger hairsThewooden stick was immediately removedafter stimulation The collected movies were analysed frameby frame and the trap closure was quantified by measuringthe change of distance between two laminas with ImageJ soft-ware [15] Space and time constraints limited the number ofplants that could be carried on board and consequently lim-ited the number of traps available each day More emphasiswas given to the zero gravity condition therefore 25 trapswere tested in microgravity 20 were tested in 1 g during theflight and 8 traps were devoted to 2 g The distance 119910(119905)between the edges of the trap leaf wasmeasured in the closingprocess In the open state the distance between the edges ofthe trap leaf is 119910max As individual plants have different open-ing distances data were normalized 119909 = 119910119910max The speedof trap closure was calculated as V = 119889119909119889119905 and it has thedimension of sminus1

3 Results

The trap closure was studied at different gravity conditions(Figure 1)

Results showed a low responsiveness of traps in micro-gravity 36 of the traps did not close at all and 48 mani-fested a slower closing motion The traps stimulated in nor-mal gravity demonstrated a normal closure in 80 of thecases and finally the traps stimulated in hypergravity reactedpromptly to the stimulation with 50of the traps being fasterthan controls (Table 1)

Table 1 Trap behaviour recorded under different gravity conditionsthe number of traps tested 119899 is reported in the table

119899No

responseNormalclosure

Slowclosure

Fastclosure

0G 25 36 16 48 01G 20 10 80 10 02G 8 0 50 0 50

In the graph (Figure 2) three representative exampleshave been reported

Trap closure is strongly affected by gravity in micrograv-ity the kinetics of snapping is slower (Figure 2(a)) and theacceleration is low if compared with 1 and 2 g where the speedof closure increases sharply after the trigger (Figure 2(b))

Because of the constraints involved in performing theexperiments on a plane we were unable to measure thereaction time between the trigger and the start of the closure(our time resolution was 40ms and the traps were stimulatedmanually) However visual observation revealed a delayedresponse in zero gravity and an anticipated response inhyper-g

4 Discussion

Volkov et al described the trap closure as consisting of threedifferent phases [9] (1) a mechanically silent period with noobservable movement immediately after stimulation (2) theperiod when the movement starts accelerating (3) the fastmovement of the trap when the leaves quickly relax to thenew equilibrium state In our results it appears that micro-gravity acts inDionaea at two different levels (i) by impairingthe signal transduction as suggested from the high per-centage of inactive traps and from the lower responsiveness(phase 1) and (ii) by altering the trap kinetics by signifi-cantly reducing the trap closing time (phases 2-3) eventuallysuggesting that the mechanisms leading to trap closure aregravity-related The electrical properties of excitable cells areextremely important in higher organisms Changes of theirparameters under microgravity can impair the functionalityof the neural systems and have significant consequences forhuman especially in view of long space travels The fewreports available on animal cells suggest that action potentialsare affected by gravity [16] in particular the propagationvelocity and their intensity seem to be gravity-dependentthat is they increase under hypergravity and decrease undermicrogravity compared to 1 g [17] Very little is known forhigher plants Masi et al monitored for the first time theelectrical activity of root cells during a parabolic flight andobserved alterations of the frequency of APs [18] Alteredparameters have been reported also under hypergravity (Masiet al unpublished) suggesting that the excitability of bothplant and animal cells is heavily affected by altered gravityconditions

InDionaeamuscipula it is well known that the stimulationof trigger hairs generates the twoAPs required for trap closure[9 19 20]Thehigh number of inactive traps (ie no responseto the trigger) and the apparent slower response time of


0g

0g

0g

0g

0g

0g

1g

1g

1g

1g

1g

1g

2g

2g

2g

2g

2g

2g

0ms

180ms

360ms

540ms

720ms

900ms

Figure 1 Closing of the trap in micro- (0 g) normal (1 g) and hypergravity (2 g)

the trap closure suggest an alteration in the generation orpropagation of the APs in microgravity In plants as well asin animals action potentials are induced by the fast openingand closing of ion channels whose functionality has beenlittle studied and understood in microgravity so far In fact

ion channels are integral membrane proteins and they couldbe affected either directly or indirectly by gravity Gravitycould directly affect the protein integrity whereas changesin the thermodynamical properties of the membrane couldhave an indirect effect on the ion channel functionality [21]


0 200 400 60000

05

10

Time (ms)

0g1g

2g

Trap

clos

ure (

yminusy

max

)

(a)

0 200 400 600 80000

01

02

03

04

05

Time (ms)

Spee

d (1

sminus1)

0g1g

2g

(b)

Figure 2 Effect of gravity on trap closure (a) Kinetics of trap closure under different gravity conditions 119910 is the distance between the edgesof the lobes (b) Dependency of the speed of trap closure on time after stimulation

In 2001 Goldermann and Hanke showed for the first timethat gravity influences the integral open state probabilityof ion channels providing a first explanation of the effectsof gravity on electrical signalling [22] Those findings werefurther confirmed by patch-clamp analysis [21]

Nothing similar has been done for plants The first silentstage of the trap closing involves transduction of electrical sig-nal and hence it is related to ion channel gating Interestinglyresults similar to the ones obtained here under microgravitywere observed when applying channel blockers to the traps[9]The use of BaCl2 ZnCl2 andTEACl significantly delayedtrap closure and altered its speed [9]

Of course the results presented here are just preliminaryFurther studies will be necessary to consolidate the resultsand to investigate in deeper detail the possible effect of gravityon the generation and propagation of action potentialsParticularly interesting would be to stimulate electrically thetraps allowing measuring and quantifying the delay in trapclosure under altered gravity conditions

To conclude our results demonstrate the role of micro-gravity on the events leading to trap closure The possiblealterations of ion and water channel permeability that couldbe at the base of the lower responsiveness and slow closureobserved in microgravity are a possibility worthy to be inves-tigated In fact if properly demonstrated it would strengthenthe validity of the hydroelastic curvature model suggested byVolkov et al [9] Finally we want to stress the fact that Venusflytrap could be an easy and effective model plant to performstudies on ion channels and aquaporin activities as well ason electrical activity in vivo on board of parabolic flights andlarge diameter centrifuges



References

[1] J Burdon-Sanderson ldquoOn the electrom otive properties of theleaf of dionaea in the excited and unexcited statesrdquoPhilosophicalTransactions of the Royal Society of London vol 173 pp 1ndash551882

[2] C Darwin and F Darwin Insectivorous Plants John MurrayLondon UK 1888

[3] B B E Juniper R J Robins and D M Joel The CarnivorousPlants Academic Press London UK 1989

[4] WH Brown andLW Sharp ldquoThe closing response in dionaeardquoBotanical Gazette vol 49 no 4 pp 290ndash302 1910

[5] A G Volkov T Adesina and E Jovanov ldquoClosing of venusflytrap by electrical stimulation of motor cellsrdquo Plant Signalingamp Behavior vol 2 no 3 pp 139ndash145 2007

[6] S E Williams and A B Bennett ldquoLeaf closure in the venusflytrap an acid growth responserdquo Science vol 218 no 4577 pp1120ndash1122 1982

[7] J Scala K Iott D Schwab and F Semersky ldquoDigestive secretionof Dionaea muscipula (venusrsquos flytrap)rdquo Plant Physiology vol44 no 3 pp 367ndash371 1969

[8] K Trebacz and A Sievers ldquoAction potentials evoked by light intraps ofDionaea muscipula ellisrdquo Plant and Cell Physiology vol39 no 4 pp 369ndash372 1998

[9] A G Volkov T Adesina V SMarkin and E Jovanov ldquoKineticsand mechanism of dionaea muscipula trap closingrdquo PlantPhysiology vol 146 no 2 pp 694ndash702 2008

[10] A Pavlovic L Slovakova C Pandolfi and S Mancuso ldquoOn themechanism underlying photosynthetic limitation upon triggerhair irritation in the carnivorous plant Venus flytrap (Dionaeamuscipula Ellis)rdquo Journal of Experimental Botany vol 62 no 6pp 1991ndash2000 2011

[11] A Pavlovic and SMancuso ldquoElectrical signaling and photosyn-thesis can they co-exist togetherrdquo Plant Signaling and Behaviorvol 6 no 6 pp 840ndash842 2011

[12] D Hodick and A Sievers ldquoOn themechanism of trap closure ofVenus flytrap (Dionaea muscipula Ellis)rdquo Planta vol 179 no 1pp 32ndash42 1989


[13] B S Hill and G P Findlay ldquoThe power of movement in plantsthe role of osmotic machinesrdquo Quarterly Reviews of Biophysicsvol 14 no 2 pp 173ndash222 1981

[14] Y Forterre JM Skotheim J Dumals and LMahadevan ldquoHowthe Venus flytrap snapsrdquoNature vol 433 no 7024 pp 421ndash4252005

[15] C A Schneider W S Rasband and K W Eliceiri ldquoNIH imageto imageJ 25 years of image analysisrdquo Nature Methods vol 9no 7 pp 671ndash675 2012

[16] M Wiedemann F P Kohn H Roesner and W R HankeldquoBehavior of action potentials under variable gravity condi-tionsrdquo in Self-Organization and Pattern-Formation in NeuronalSystems under Conditions of Variable Gravity pp 95ndash109 2011

[17] K Meissner and W Hanke ldquoAction potential properties aregravity dependentrdquoMicrogravity Science and Technology vol 17no 2 pp 38ndash43 2005

[18] E Masi M Ciszak S Mugnai et al ldquoElectrical network activityin plant roots under gravity-changing conditionsrdquo Journal ofGravitational Physiology pp 167ndash168 2008

[19] D Hodick and A Sievers ldquoThe action potential of Dionaeamuscipula Ellisrdquo Planta vol 174 no 1 pp 8ndash18 1988

[20] W H Brown ldquoThe mechanism of movement and the durationof the effect of stimulation in the leaves of dionaeardquo TheAmerican Journal of Botany vol 3 no 2 pp 68ndash90 1916

[21] M Wiedemann F P Kohn H Roesner and W R HankeldquoInteraction of gravity with molecules and membranesrdquo in Self-Organization and Pattern-Formation in Neuronal Systems underConditions of Variable Gravity pp 57ndash93 2011

[22] M Goldermann and W Hanke ldquoIon channel are sensitive togravity changesrdquo Microgravity Science and Technology vol 13no 1 pp 35ndash38 2001

Review ArticleThe Impact of Simulated and Real Microgravity onBone Cells and Mesenchymal Stem Cells

Claudia Ulbrich1 Markus Wehland2 Jessica Pietsch2 Ganna Aleshcheva2 Petra Wise3

Jack van Loon456 Nils Magnusson7 Manfred Infanger2 Jirka Grosse8 Christoph Eilles8

Alamelu Sundaresan9 and Daniela Grimm10

1 Department of Physiology Membrane Physiology University of Hohenheim 70593 Stuttgart Germany2 Clinic for Plastic Aesthetic and Hand Surgery Otto-von-Guericke University 39120 Magdeburg Germany3HematologyOncology Childrenrsquos Hospital Los Angeles University of Southern California Los Angeles CA 90027 USA4Department of Oral and Maxillofacial SurgeryOral Pathology VU University Medical Center Amsterdam1007 MB Amsterdam The Netherlands

5 Department of Oral Cell Biology Academic Centre for Dentistry Amsterdam (ACTA) University of Amsterdam andVU University Amsterdam 1081 LA Amsterdam The Netherlands

6 European Space Agency Technology Center Gravity Lab (ESA-ESTEC-TEC-MMG) 2201 AZ Noordwijk The Netherlands7Medical Research Laboratory Institute of Clinical Medicine Aarhus University 8000 Aarhus C Denmark8Department of Nuclear Medicine University of Regensburg 93052 Regensburg Germany9Department of Biology Texas Southern University 3100 Cleburne Houston TX 77004 USA10Institute of Biomedicine Pharmacology Aarhus University Wilhelm Meyers Alle 4 8000 Aarhus C Denmark

Correspondence should be addressed to Daniela Grimm danielagrimmfarmaudk

Received 4 April 2014 Revised 6 June 2014 Accepted 6 June 2014 Published 10 July 2014


Copyright copy 2014 Claudia Ulbrich et alThis is an open access article distributed under theCreative CommonsAttribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Howmicrogravity affects the biology of human cells and the formation of 3D cell cultures in real and simulatedmicrogravity (r- ands-120583119892) is currently a hot topic in biomedicine In r- and s-120583119892 various cell types were found to form 3D structures This review willfocus on the current knowledge of tissue engineering in space and on Earth using systems such as the random positioning machine(RPM) the 2D-clinostat or the NASA-developed rotating wall vessel bioreactor (RWV) to create tissue from bone tumor andmesenchymal stem cells To understand the development of 3D structures in vitro experiments using s-120583119892 devices can providevaluable information about modulations in signal-transduction cell adhesion or extracellular matrix induced by altered gravityconditions These systems also facilitate the analysis of the impact of growth factors hormones or drugs on these tissue-likeconstructs Progress has beenmade in bone tissue engineering using the RWV andmulticellular tumor spheroids (MCTS) formedin both r- and s-120583119892 have been reported and were analyzed in depth Currently these MCTS are available for drug testing andproteomic investigations This review provides an overview of the influence of 120583119892 on the aforementioned cells and an outlook forfuture perspectives in tissue engineering

1 Introduction

It is well known that microgravity influences different bio-logical systems like bone and muscle as well as the heart andbrain and it enhances cancer risk [1] During their stay atthe MIR astronauts and cosmonauts did show a distinct lossof bone mineral density in the lumbar spine the pelvis and

the proximal femur [2] and the extent of bone loss varied upto 20 [3]

As it is not feasible to gather enough material fromastronauts to do in-depth investigations another device hasbeen developed for the International Space Station (ISS) themice drawer system (MDS) as a facility to study long-timeinfluence of radiation on the biology and behavior of mice



Tavella et al for example report an altered bone turnoverin different strains of mice which were kept on the ISS for91 days This resulted in bone loss due to increased boneresorption and a decreased bone deposition [4]

While the past biological physiological and medicalresearch nearly exclusively focused on investigating the bio-chemical processes of living cells and organisms more andmore attention was paid to the biomechanical properties andmechanical environment of cells and tissues during the lastdecades When culturing cells on Earth they usually settleon the bottom of the culture flask forming two-dimensional(2D) monolayers A three-dimensional (3D) growth moreresembling the tissue environment found in living organismsis prevented by the presence of the gravitational field Fora scaffold-free 3D tissue growth it is therefore necessary tocircumvent this problem by effectively eliminating the influ-ence of the gravitational pull during cultivation One of thebyproducts of various space flight endeavors is the possibilityto perform long-term near-weightlessness or microgravity(120583119892) experiments [5 6] In a 120583119892 environment cells will notsettle like on Earth This provides an increased opportunityfor freely floating cells to interact with each other and develop3D structures [7]

2 Space Flights forCell-Biological Experiments

Long-term orbital space flight experiments are however nottrivial Flight opportunities are very scarce and the costs ofhardware development are high Furthermore science is notalways a priority in space flight activities Such preconditionsare delaying the advancement of research in areas such as cellbiology and tissue engineering disciplines which could profittremendously from more frequent research options in a realmicrogravity (r-120583119892) environment

Some researchers recently pointed out that osteoblastsundergo a disintegration of their cytoskeleton which mayexplain dramatic changes in size and shape of the cells andtheir surface specializations [47] Also other studies havebeen performed using the ISS or space shuttle flights to learnmore about the behavior of bone cells in space [48] but flightopportunities are sparse and therefore other platforms hadto be elucidated

It is due to the aforementioned limitations that over theyears various devices have been developed in an attemptto reduce the impact of gravity and simulate a near-weightlessness environment (s-120583119892) on Earth From a physicalpoint of view gravity is a force exhibiting bothmagnitude anddirection Therefore the influence of gravity can be reducedby either manipulating magnitude or direction An orbitalspace flight as on the ISS is physically identical to a free-fall Here the gravitation acts in a perpendicular manneron the spacecraftrsquos velocity vector effectively changing itsdirection constantly but not affecting its magnitude Free-fallis also found when using sounding rockets which provider-120583119892 during a time span of up to 15 minutes On Earthr-120583119892 can also be attained although only for periods inthe range of seconds in drop towers and during parabolic

flights missions [49 50] Although time periods of secondsor minutes limit their use for tissue engineering studiessuch periods can be useful to explore various intra- andintercellular processes responsible for gene expression andprotein content changes which can be observed after only afew hours of culturing cells in 120583119892 [49ndash51]

3 Devices Simulating Microgravity on Earth

In this respect we should mention an instrument that wasintroduced by the European Space Agency (ESA) in the earlynineties called the free fall machine (FFM) [52] This instru-ment was specifically developed for biological experimentsand could generate a free fall for a period of about 800mswith an intermediate ldquobouncerdquo ofsim20 g for around 50msTheparadigm of the FFM is that cells might not be sensitive tothe relatively short period of 50ms of hypergravity while theyexperience the relatively longer period of free-fall Long-termexperiments (hours days) which might be useful for tissueengineering studies could be performed on this platformHowever thus far only two studies were published usingthe FFM one investigatingChlamydomonas [53] and anotherone researching T-lymphocytes [54] The Chlamydomonasstudy showed similar results to what was found in realspace flight while the T-lymphocytes experiments did notConsidering the very limited number of studies performedonthis ground-based device the FFM still might deserve somemore exploration

Levitating magnets are also used to produce s-120583119892 onEarth Such systems compensate themagnitude of the gravityvector by preventing sedimentation of relatively heavy struc-tures like cells by the application of a high gradientmagneticfield This principle was first described for biological systemsby Berry and Geim in 1997 [55] who demonstrated that atoad could be levitated and survivewhile exposed to a 16 Teslamagnetic field Various experiments in cell biology havemadeuse of such systems [56ndash58] The magnetic field acts on indi-vidual molecules and atomswithin a cell based on theirmag-netic susceptibility preventing them from sedimentationHowever the magnetic field as such confounds possible s-120583119892effects The direction of the field might force (bio-)polymersinto a certain orientation Different polymers within a cellor on the cell membrane have different susceptibilitiespossibly producing artifacts by forcing polymers into specificarrangements which may not reflect the actual physiologicalsituation [59ndash61] Superconducting high gradient magnetsare especially capable of performing long-term experimentsand might be useful in the area of tissue engineering [62ndash64] In this context another promising technique should bementioned This method is the use of magnetic particles for3D cell cultures It is not based on a high-gradient magneticfield but on ferromagnetic particles attached to cells whichcan subsequently be levitated by a conventional magnetfacilitating the formation of 3D structures [65 66]

Another option is to manipulate the direction of thegravity vector with respect to the sample The reduction ofthe gravitational impact on biological systems by constantlychanging its orientation was shown first in experiments by


the German botanist von Sachs in 1879 growing Lepidiumsativum and Linumusit [67] He constructed a slowly rotatingsystem and named it a clinostat in which for example a plantcan be placed horizontally and rotated around its longitudinalaxis In doing so the gravity vector stimulus is constantlychanging its impact angle on the sample As a result aplant grows straight without the characteristic gravitropiccurvature seen when the plant is placed horizontally and notrotating Based on these initial studies other rotating systemslike the fast rotating clinostat have been developed

The initial clinostats were rotating relatively slowly in arange fromone rotation per couple of hours up to amaximumof about 10 rpmThis is adequate for relatively ldquosolid samplesrdquosuch as plants but too slow for cell culture systems thatinvolve a large liquid phase In a biphasic system that isa liquid with particles (cells) both of different density theheavy particles tend to settle Rotating such a system around ahorizontal axis keeps the heavy particles in suspension Thisphenomenon depends mainly on the relative density of theliquid and the particles the viscosity of the liquid the rotationspeed and the diameter of the rotated container When a cellis in a static vessel and the vessel is rotated by 90∘ the cell willsettle in the direction of the gravity vector One can repeatthis for a full 360∘ and upon an increase in the frequency ofrotation the traveling distance of the cell decreases If thisrotation is performed constantly with increased speed wefinally end up rotating a cell around its own axis Such acontrolled rotation not only applies to the cells but also itssurrounding boundary liquid phase [68]

Another well-known device to simulate 120583119892 is the so-called random positioning machine (RPM) a 3D clinostat[69] consisting of two frames each driven by a dedicatedmotor This allows a randomized movement of both framesindependent of each other [69ndash74] One of the advantages ofthe RPM is its size as cell culture flasks can easily bemountedon it so it is possible to work with quite large liquid volumesThis ranges from regular T25 flasks [75 76] to multi-wellplates [77] flasks on slides [78] or more dedicated devices[79] As cells move freely within the liquid they usuallyinteract with each other and form multicellular spheroids

The best simulation of 120583119892 is achieved in the rotationcenter of the two axes which limits the preferred volume sizeof the samples Depending on the speed of rotation and thedistance from the center an acceptable residual gravity canbe obtained in the order of 10minus4 g by a maximum angularvelocity of 60∘ sminus1 at a radial distance of 10 cm [70] EarlierRPM models had no possibility to add constituents duringthe experiment but newer models have been developed toenable fluid management during rotation [73 74] RPMsare commercially available by Mitsubishi Heavy Industries(Kobe Japan) and Dutch Space (Leiden The Netherlands)while various academic groups developed similar systems ontheir own [80ndash84] (Figure 1)

The rotating wall vessel (RWV) prevents cells from set-tling via a constant rotation It has been developed by NASA[85] and is now commercially available through SyntheconInc (Houston TX USA) Basically RWVs consist of aslow rotating relatively large liquid filled container (vessel)

The rotation speed has to be adapted to the specific weightof the cells the fluid density and viscosity The cells andtissues in the RWV are constantly falling within the fluidThesettling velocity and direction combined with the rotation ofthe fluid create spiral trajectories within the vessel [86] Thismotion of the sample relative to the fluid generates fluid shearforces on a particle surface ranging from 180 to 320mPa (18ndash32 dynecm2) for 50 120583m beads [87] sim500mPa (5 dynecm2)with 3D aggregates of BHK-21 cells [88] to 520ndash780mPa (52ndash78 dynescm2) for a 200 or 300120583mspherical object [89] Overthe years variousmodels based on the initial RWVhave beendeveloped differing in vessel geometry aspect ratio and gassupply such as the slow turning lateral vessel (STLV) [90]the high aspect ratio vessel (HARV) [91] or the rotating-wallperfused vessel (RWPV) [92]

Hence it can be concluded that annulling the gravityforces which pull the cells constantly towards the Earthdeliver the ultimate trigger to eukaryotic cells to leave a cellmonolayer and assemble in 3D aggregates [5]

It is still unknown which cellular and biochemical mech-anisms are involved in the altered signal transduction and inthe change of the cellular growth behavior

4 Transition from Two- toThree-Dimensional Cell Growth

A few publications appeared in the literature in recent yearsproviding some clues for understanding the weightlessness-induced transition from two- (2D) to three-dimensional (3D)cell growth

Several signaling pathways are affected by annullinggravity forces in the cell interior [93] However it is unknownwhich of these signaling pathways contribute to the formationof three-dimensional aggregatesWhen endothelial cells formtubes the nitric oxide signaling pathway appears to beaffected [94] Siamwala et al reported that iNOS (induciblenitric oxide synthase) acts as a molecular switch whichcontrolswhether the effects of120583119892on vascular endothelial cellsinduce angiogenesis via the cyclic guanosinemonophosphate(cGMP)-PKG-dependent pathway [94] iNOS is upregulatedinHUVECby amechanismdependent on suppression ofAP-1 after clinorotation of the cells [95] In addition the endothe-lial nitric oxide synthase is phosphorylated by phosphoinosi-tide 3-kinase under weightlessness simultaneously with Akt[96] The organoid formation by PC12 pheochromocytomacells in a RWV bioreactor is accompanied by prolongedactivation of the ERK p38 and jnk signaling pathways [97]

3D cell culture techniques have attracted much attentionnot only among biologists but also clinicians interested intissue engineering [98 99] of artificial vessels [100ndash104] orcartilage [105ndash108] Moreover osteoarthritis and cartilagetrauma occur in patients with a high incidence but currenttreatmentmethods are still limited [109] Even aminor injuryto articular cartilage may lead to progressive damage anddegeneration [110]


(a) (b) (c)

Figure 1 (a) Two 2D clinostat devices in an incubator constructed by the German Aerospace Center (DLR) Institute of AerospaceMedicineBiomedical Science Support Center Gravitational Biology Cologne Germany (b c) Random Positioning Machine simulating microgravityIt was developed by T Hoson in Japan and manufactured by Dutch Space (former Fokker Space)The basic principle consists of an inner andan outer frame rotating independently from each other in random directionThe samples in the center of the machine experience low gravityas the gravity vector is averaged to zero over time The redesign of the classical RPM with a CO

2-Incubator with temperature and CO

2-level

control was realized by Professor Jorg Sekler Fachhochschule Nordwestschweiz (FHNW) Institut fur Automation Switzerland and tested byPD Dr Marcel Egli Hochschule LuzernmdashTechnik amp Architektur CC Aerospace Biomedical Science amp Technology Hergiswil Switzerland

5 Tissue Engineering of Bone

Bone loss has been documented for many years in 120583119892 (1-2 a month) Increased bone loss and risk of fractures isan identified risk in the bioastronautics critical roadmap forlong-term cosmic missions to the moon and mars In vitrodrug screening both in 1 g 120583119892 and in artificial gravity isessential to adequately address countermeasures for boneloss Bone loss in 120583119892 is the second most important risk tospace missions [5 6]

Exposure to the 120583119892 environment of space causes astro-nauts to lose calcium from bones [5 6] This loss occursbecause the absence of Earthrsquos gravity disrupts the processof bone maintenance in its major function of supportingbody weight Exposure to the 120583119892 environment of space causesmen and women of all ages to lose up to 1 of their bonemass per month due to disuse atrophy a condition similar toosteoporosis It is not yet clear whether loss in bone mass willcontinue as long as a person remains in the 120583119892 environmentor level off in time

There are indeed four major bone cell types and eachof them seems to be influenced by 120583119892 Bone mesenchymalstem cells (MSC) are able to differentiate into adipocytesosteoblasts and osteoclasts Proliferation and differentiationare very sensitive to 120583119892 as the lack of gravity in spacecan reduce mechanical stress leading to a decreased rate ofosteogenesis and an increased adipogenesis rate [111] As thesignaling pathways involved in MSC differentiation form acomplicated network it has been found that the reduction inthe osteogenesis ofMSCs in the presence of 120583119892 is mediated bya decrease in the integrinmitogen-activated protein kinase(MAPK) signaling pathway [112] as well as RhoA andcytoskeletal disruption [113]

Osteoblasts are derived from MSCs but in 120583119892 the dif-ferentiation does not function properly and the resultingbone loss has been attributed to osteoblasts due to their (1)reduced proliferation and activity (2) reduced differentiationand (3) decreased responsiveness to bone-related factors inthe microenvironment [114] Observations have also beenmade regarding the cytoskeleton of osteoblasts there isgrowing evidence that the cytoskeleton is closely connected tonuclear morphology and function [115] The enlarged nucleiobserved in flight osteoblasts could be a result of cytoskeletaldisruption [116]

Osteocytes regulate bone resorption and formationand are considered the terminal differentiation stage ofosteoblasts The osteocytes in cortical bone and periosteumdegenerated after a 125-day flight in space on the CosmosBiosatellite [117] Osteocyte apoptosis has been observedafter a 2-week flight increasing the number of functionallyactive osteoclasts [118] Apoptotic osteocytes are essential forthe initiation of bone remodeling but it is the neighboringnonapoptotic osteocytes that produce proosteoclastogenicsignaling [119] Osteocytes seem to be the key effectors of 120583119892induced bone loss [120]

Osteoclasts are bone-resorbing cells and their differenti-ation seems to be enhanced in 120583119892 [121]This could be anotherexplanation of bone-loss in space

Themystery for the moment is what signals permit bonetissue to adapt to a weightless or an Earth (1 g) environmentResearchers do not yet know whether the biomechanicalstimuli that are changed by 120583119892 directly affect osteoblastand osteoclast function or if other physiological factorssuch as hormone levels or poor nutrition contribute tobone loss NASA investigators are studying gravity-sensingsystems in individual bone cells by flying cultures of these


cells on the space shuttle and observing how they functionDiscoveries made in the course of space biomedical researchon bone are already contributing to a better understandingof osteoporosis and the treatment of bone mass loss on Earthas well as in space The single most important contributionthat NASA research has made to the understanding of bonedeterioration in osteoporosis is heightened awareness of theimportance of gravity activity and biomechanicsmdashthat is themechanical basis of biological activitymdash in bone remodeling

Mechanical forcesmdashthe action of energy on mattermdashappear to coordinate bone shaping processes The standardtheory of bone remodeling states the body translatesmechan-ical force into biochemical signals that drive the basic pro-cesses of bone formation and resorption Aging especially inpostmenopausal women and exposure to 120583119892 uncouple boneresorption and formation When this uncoupling occursformation lags behind resorption and the result is bone loss

Researchers are not yet certain whether bone resorptionspeeds up or the bone formation slows down though recentexperimentation in space indicates that 120583119892 might somehowaffect both processes Progress in developing methods of pre-venting or treating disuse atrophy and osteoporosis dependson better understanding of the mechanisms that cause theproblem Determining how the body translates mechanicalloading (physical stress or force) into the signals that controlbone structure may reveal how aging inactivity and spaceflight uncouple bone formation and resorption Only in theabsence of gravity can we determine the influence of weightand stress on bone dynamics

By studying whatmechanisms translatemechanical stresson bones into biochemical signals that stimulate bone for-mation and resorption space life scientists may be able todetermine how tomaintain bonemass Researchers donot yetknow exactly what type and amount of exercise hormonesor drugs might prevent bone loss or promote bone forma-tion However some combination of sex hormones growthhormones and exercise seems to be the key to preventingbone mass loss associated with chronological aging andpostmenopausal hormone changes on Earth

Bone is made up of several different cell populationsOsteoclasts are responsible for the breakdown of mineralizedbone in preparation for bone remodeling In contrast theosteoblasts synthesize mineralized bone in the remodelingprocess The goal of this project is to develop an ldquoinvitrordquo three-dimensional cellular model of osteoclasts andosteoblasts (human and rodent) cultured together in 120583119892analog culture conditions to identify the underlying biomark-ers related to bone loss in 120583119892 and the cellular mechanismsinvolved in bone resorption The NASA rotating-wall vessel(RWV) permits the growth of mixed cell cultures for muchlonger periods than traditional culture methods This wouldset the stage for development of countermeasure strategies forbone loss in space as well as in osteoporosis and rheumatoidarthritis which are increased health risks on Earth ProfessorSundaresan and collaborators [122ndash124] have developed a 3Dcell culture bone tissue model using a specialized rotating-wall vessel culture system to address a more physiologicallyrelevant model to the human body The use of the cells by

themselves also eliminates confounding variables such asneuroendocrine stress found in vivo (Figure 2(a))

The human body needs a framework to withstand gravityThis framework is given by the skeletal system Duringlong-term space missions bone loss has been reported inastronauts at a rate that is both substantial and progressivewith time spent in 120583119892 [125ndash128] But what is the reason forthismassive bone loss Some studies suggested that this effectmight be attributed to increased resorption in load-bearingregions of the skeleton [129ndash131] and evidence of a decreasein bone formation had also been described For example theloss of bone in 120583119892 is about 10 times greater than the bonemineral density loss per month of postmenopausal womenon Earth who are not on estrogen therapy [132ndash135]The lossof bone mineral density in a six-month mission appeared tobe reversible in 1000 days after return to Earth [136 137] butchanges in the bone structure are irreversible and seem tomimic changes in the elderly [137]

Until now there are still knowledge gaps on the mecha-nism of bone loss especially on the molecular and cellularmechanisms also the question of fracture repair arisesMoreover more information is needed on the influence ofradiation hormones and fluid shifts

Investigations in humans and animals are quite difficultdue to the lack of long-term flight opportunities the absenceof animal housing facilities in space and the problem ofmaterial collection from returning astronauts Thus otherpossibilities have to be sought in order to investigate bone Sofar most commonly used are bone cell culture experimentswhich are a viable opportunity for investigating cells in 3Dacting as tissue like samples while they are cultivated underconditions of weightlessness However 3D embryonic bonetissue cultures have been used in the past and show a cleardecrease in matrix mineralization in mineralizing cartilageand by osteoblasts combined with an increased mineralresorption by osteoclasts [138]

Besides this tissue engineering is a very up-to-date topicThe ultimate goal is to generate functional 3D constructswhich can be used as replacement organs or structures withnormal function or serve for in vitro studies [5 139] Bonereplacement especially is quite difficult as large bone defectsusually require reconstructive surgery to restore function[140] Up to date the treatment includes autograft or allografttransplantation and the use of syntheticmaterials [141]Whileautograft transplantation is the preferred treatment it suffersfrom limited supply and donor site morbidity [142] Asthe autogenous origin of cells prevents potential immunerejection the amount of bone marrow suitable for trans-plantation is limited New techniques have been developedallowing selection of bone marrow osteoprogenitor cells andexpanding them in culture so that a large amount of trans-plantable cells can be generated after only one biopsy [143ndash145]

In principle culturing bone cells is not that easy A combi-nation of osteoconductive matrices bone-forming cells andosteogenic growth factors is needed for the engineering ofbone tissue [146] The first important factor is the cell typeOsteoblasts are in a close to mature stage showing a lowproliferative potential Mesenchymal stromal cells represent


(a)

TC 4d RPM MCTS

(b)

Figure 2 (a) Production of large numbers of small (200 120583m diameter) immature (7-day-old) osteospheres with labeled osteoclast cells (red)viewed by confocal imaging in living constructs-USPTO 80736136 and (b) follicular thyroid cancer cells (TC) cultured on the RPM Severalmulticellular tumor spheroids are visible after 4 days

a proliferating and undifferentiated cell source but theiravailability is limited [147 148] An option to increase theirlifespan in vitro is the overexpression of human telomerasereverse transcriptase (hTERT) The second factor is an idealscaffold which possesses mechanical properties comparableto bone It should support cell adhesion and should bebiodegradable to facilitate natural bone remodeling [146]As of now different studies have shown the advantagesand disadvantages of several types of scaffolds like chitingelatin poly(lactic acid) poly(glycolic acid) poly(lactic acid-co-glycolic acid) polycaprolactone hydroxyapatite coraland so forth Several in vitro studies revealed an ideal scaffoldpore size for osteoblasts from 200 to 400 120583m [149 150] It isimportant to recognize that the scaffold architecture influ-ences the distribution of shear stress the range of mechanicalstimuli as well as the proliferation and differentiation ofosteoprogenitor cells [151 152]

To simulate an ideal in vivo situation for in vitro cellsspecific cytokines and growth factors are necessary For bonemorphogenesis the bone morphogenetic proteins (BMP)which belong to the transforming growth factor beta (TGF-120573) superfamily are essential [153] Currently only BMP-2and -7 are commercially available so alternatives to stimulateosteoprogenitor cells by growth factors are required It hasbeen reported that autologous platelet-rich plasma is aneffective bioactive supplement as it contains osteogenic andangiogenic growth factors [154]

Several different bioreactor systems are already availablefor bone tissue engineering Awell-known and simple systemis the spinner flask bioreactor Convective forces are providedby a stirrer and the medium flows around the cells Theemerging shear stress is not applied homogenously as thereappears to form a gradient in the flask [146] This factorcertainly needs to be considered when conducting studieswith the spinner flask system

Other suitable instruments are rotating bioreactor sys-tems for example the RWV It has been used with different

kind of bone cells which are often grown with the help ofmicrocarriers [8 155] or scaffolds [8ndash11 15 155] The highaspect ratio vessel (HARV) [91] was used by Lv et al [12]to engineer tissue on poly(lactic acid glycolic acid)nano-hydroxyapatite composite microsphere-based scaffolds

Some researchers used bone marrow mesenchymal stemcells for their investigations Jin et al [16] were able to trans-plant RWV-grown bone constructs in cranial bone defectsof Sprague-Dawley rats and found them to be more effectivein repairing the defects than the 1 g controls after 24 weeksMoreover a 3D environment as in a rotary cell culture systemenhanced osteoblast cell aggregation andmineralization [13]Preosteoblasts cultured in a RWV could be engineered intoosseous-like tissue [14]

6 Mesenchymal Stem Cells and Microgravity

Mesenchymal stem cells (MSCs) are cells capable of long-term proliferation and differentiation into various stromaltissue cell types The state of MSCs rests on the cellularmicroenvironment and several soluble factors In additiongravity can influence MSC features Disuse as encounteredduring long-term bed-rest or space travel and the accom-panying absence of mechanical stimuli lead to an inhibitionof osteogenesis and simultaneously to an induction of adi-pogenesis in MSCs Hence it is crucial to provide a propermechanical stimulation for cellular viability and osteogenesisparticularly under unusual conditions

In 2004 Merzlikina et al [27] studied the effects of pro-longed clinorotation on cultured human MSC morphologyproliferation rate and expression of specific cellular markersAfter exposure of the cells to clinorotation for time framesfrom 1 h to 10 days it was shown that the proliferativerate decreased in the experimental cultures as compared tocells growing under normal conditions Clinorotated MSCsseemed more flattened and reached confluence at a lower


cell density which advocates that cultured hMSCs sense thechanges in the gravity vector and respond to s-120583119892 by alteredfunctional activity The group around Myoui [28] examinedwhether gravity-induced stress is linked to osteoblast dif-ferentiation and function Rat marrow mesenchymal cells(MMCs) were cultured in pores of interconnected porouscalcium hydroxyapatite (IP-CHA) for 2 weeks on a 3D clino-stat In MMCs subjected to s-120583119892 the marker of osteoblasticdifferentiation alkaline phosphatase activity was decreased by40 compared to the control group Also the clinostat groupexhibited less extensive extracellular matrix formation thanthe control group The implantation of the IP-CHAMMCcomposites in syngeneic rats showed that bone formationwas significantly lower for the clinostat group than for thecontrol group Yuge et al [29] also used a 3D clinostat fortheir experiments on the proliferation behavior of hMSCsThe proliferation rate of the cells of the clinostat group waselevated almost 3-fold in comparison to the control groupand the number of hMSCs double-positive for CD44CD29or CD90CD29 in the clinostat group after 7 days in cultureincreased 6-fold The hMSCs cultured in a 3D-clinostat werestill able to differentiate into hyaline cartilage after trans-plantation into cartilage defective mice and displayed thestrong proliferative characteristic of stem cells thus showingthat s-120583119892 may be used to expand stem cell populations invitro In contrast to these findings Dai et al [24] reportedin 2007 that 120583119892 simulated by a clinostat inhibited populationgrowth of bone marrow mesenchymal stem cells (rBMSCs)and their differentiation towards osteoblasts The cells grownon the clinostat were arrested in the G(0)G(1) phase ofcell cycle and growth factors such as insulin-like growthfactor-I epidermal growth factor and basic fibroblast growthfactor had only a slight stimulatory effect compared to thestatic control group Gershovich and Buravkovarsquos [17] worksupports this hypothesis After 20 days of clinostat-exposurethe proliferative activity of hBMCs was reduced whereas itincreased the number of large flat cells in the culture andstimulated migration activity of cells In 2009 Gershovichand Buravkova [30] demonstrated the effects of s-120583119892 byclinostat and RPMon the interleukin production by hBMSCsand MSC osteogenous derivatives 20-day exposure on aclinostat increased the interleukin-8 (IL-8) content 14 to 32times in the culturemedium while the average increase of IL-production on the RPM amounted to 15ndash6 times (10 days)and 16ndash21 times (20 days) respectively This suggests thatresults of s-120583119892 vary by the use of different modeling systemsrMSCs grown in a clinostat demonstrate that s-120583119892 can boostthe differentiation of MSCs into neurons as demonstratedby Chen et al [156] In s-120583119892 neuronal cells derived fromrMSCs were found to express higher microtubule-associatedprotein-2 (MAP-2) tyrosine hydroxylase (TH) and cholineacetyltransferase (CHAT) Furthermore the excretion ofneurotrophins such as nerve growth factor (NGF) brainderived neurotrophic factor (BDNF) or ciliary neurotrophicfactor (CNTF) was increased In comparison to 1 g controlsneuronal cells from the s-120583119892 group generated more matureaction potentials and displayed repetitive action potentialsThis might benefit the search for new strategies for thetreatment of central nervous system diseases

Zayzafoon et al [18] demonstrated that s-120583119892 inhibitsthe osteoblastic differentiation of hMSC and induces thedevelopment of an adipocytic phenotype In the effort ofunderstanding space flight-induced bone loss the group usedthe rotary cell culture system (RCCS) to model 120583119892 and deter-mine its effects on osteoblastogenesis Human MSCs werecultured and osteogenic differentiation was induced beforethe initiation of s-120583119892 As a result the important mediator ofadipocyte differentiation peroxisome proliferator-activatedreceptor gamma (PPARgamma2) and adipsin leptin andglucose transporter-4 was highly expressed These changeswere not adjusted after 35 days of readaptation to normalgravity Moreover 120583119892 decreased ERK- and increased p38-phosphorylation pathways known to regulate the activityof runt-related transcription factor 2 and PPARgamma2These results were supported by Saxena et al [19] in 2007who demonstrated that s-120583119892 inhibited osteoblastogenesisand increased adipocyte differentiation in hMSCs incubatedunder osteogenic conditions using the RCCS They couldshow that a reduced RhoA activity and cofilin phosphoryla-tion disruption of F-actin stress fibers and decreased inte-grin signaling through focal adhesion kinase were involvedin this process Meyers et al [20] also investigated the effectsof s-120583119892 on integrin expression and function in hMSCssince a reduced osteoblastic differentiation might be causedby impaired type I collagen (Col I)-integrin interactionsor a reduction of integrin signaling Culturing of hMSCsfor 7 days in s-120583119892 lead to reduced expression of Col Iwhile Col I-specific alpha2 and beta1 integrin protein expres-sion increased However autophosphorylation of adhesion-dependent kinases focal adhesion kinase (FAK) and proline-rich tyrosine kinase 2 (PYK2) was significantly reducedThese findings indicate that a reduction in osteoblastogenesisin s-120583119892 is at least in part caused by a reduced integrinMAPKsignaling The group around Duan [16] studied the relation-ships between the composition and mechanical properties ofengineered bone constructs BMSCs were grown for 15 dayson ceramic bovine bone scaffolds in different environmentsnamely static flasks and the RWV DNA content and alkalinephosphatase (ALP) were higher for cells grown on the RWVAfter transplantation into Sprague-Dawley rats with cranialbone defects the bone constructs engineered on the RWVrepaired the defects better and showed histologically betterbone connection

Sheyn et al [21] evaluated the effect of s-120583119892 on all genesexpressed in hMSCs with the hypothesis that many impor-tant pathways are affected during culture on a rotating wallvessel systemThe analysis of gene expression by use of wholegenome microarray and clustering showed that 882 geneswere downregulated and 505 genes were upregulated afterexposure to s-v A multitude of genes belonging to cell com-partment biological process and signaling pathway clusterswere modulated as identified by gene ontology clusteringSignificant reductions in osteogenic and chondrogenic geneexpression and an increase in adipogenic gene expressionwere shown and could be validated by a parallel adipogenicdifferentiation assay In order to investigate the effects ofs-120583119892 on chondrogenic differentiation of human adipose-derived MSCs (ADSCs) Yu et al [22] cultured cells on


a RCCS in pellets with or without the chondrogenic growthfactor TGF-120573

1 Analysis of real-time PCR and histological

results demonstrated that s-120583119892 has a synergistic effect onchondrogenesis with TGF-120573

1 The p38 MAPK pathway was

activated by TGF-1205731alone and was additionally stimulated

by s-120583119892 Inhibition of p38 activity with SB203580 suppressedchondrocyte-specific gene expression andmatrix productionThis indicates that the p38 MAPK signal mediates s-120583119892-induced chondrogenesis of ADSCs InMSCs cultured duringchondrogenic induction in a rotating culture combined withpolyglycolic acid (PGA)mRNAand proteins of collagen typeII and aggrecan were significantly more expressed in the s-120583119892 rotating culture group than the static culture group asreported by Wu et al [25] Zhang et al [26] described thatMSCs spread out in a spindle shape when cultured in normalgravity while they become unspread and round under s-120583119892 Also under s-120583119892 their cytoskeleton fibers are beingreorganized The function of MSCs was affected by thesemorphological changes transmitted through the activity ofRhoA To test the hypothesis that s-120583119892 has the capacity tooffer a novel choice in the stimulation of neovascularizationMSCs were cultured under s-120583119892 stimulation followed byVEGFdifferentiationThe responses revealed thatMSCsweredifferentiated into endothelial-like cells after 72 h incubationand were able to form a capillary network Their endothelialdifferentiation potential improved compared with the staticcontrol group

Another approach of modeling 120583119892 in hMSCs is the use ofa large gradient high magnetic field (LGHMF) produced bya superconducting magnet Shi et al [64] analyzed the effectsof LGHMF-120583119892 on survival cytoskeleton and osteogenicpotential of hMSCs Results showed that the LGHMF-120583119892treatment disrupted the cytoskeleton of hMSCs a LGHMF-120583119892 treatment for 24 h led to cell death LGHMF-120583119892 treat-ments in early stages of osteogenic induction resulted in sup-pression of osteogenesis of hMSCsThe suppression intensitywas reduced gradually as the treatment stage of LGHMF-MG was postponed A LGHMF-120583119892 treatment during theending-stage of osteogenic induction had no visual effecton osteogenesis of hMSCs which indicates that LGHMF-120583119892affects the initiation of osteogenesis

Furthermore a study of Uddin and Qin [23] examinedthe effects of low intensity pulsed ultrasound (LIPUS) onthe osteogenic differentiation of adipose-derived human stemcells (Ad-hMSC) under s-120583119892 conditions Microgravity wassimulated in a 1D clinostat and treated with LIPUS at30mWcmminus2 for 20min dayminus1 Hypothetically the applica-tion of LIPUS to s-120583119892 cultures would restore osteogenesisin Ad-hMSCs The results showed significant increases inALP OSX RANKL and RUNX2 and decreases in OPGgene expression in LIPUS treated SMG cultures of Ad-MSCcompared to nontreated cultures LIPUS also restored OSXRUNX2 and RANKL gene expression in osteoblast cells s-120583119892 significantly reduced ALP positive cells by 70 (119875 lt 001)and ALP activity by 22 (119875 lt 001) while LIPUS treatmentrestored ALP positive cell number and activity to equivalencewith normal gravity controls Extracellular matrix collagenand mineralization was assessed by Sirius red and Alizarinred staining respectively s-120583119892 cultures showed little or no

Table 1 Comparative methods of 3D cell culture systems usingsimulated 120583119892

Device Working principle

Free fall machine FFM Free fall for 800ms ldquobouncerdquoof 20 g for 50ms

Levitating magnets LM A high gradient magnetic fieldprevents sedimentation

2D-clinostat Rotation along one axisRandom positioningmachine RPM Two frames with randomized

movement

Rotating wall vessel RWV Constant rotation preventscells from settling

collagen or mineralization but LIPUS treatment restoredcollagen content to 50 (119875 lt 0001) and mineralization by45 (119875 lt 0001) relative to s-120583119892mdashonly cultures

7 Multicellular Tumor Spheroids

3D growth of tumor cells creating MCTS in vitro has beenobserved in various tumor cell lines including thyroid andcolorectal cancer [31] MCTS mimic the growth of solidtumors and represent a simple model approaching someof the characteristics found in vivo including physiologicalcharacteristics such as multicellular architecture and naturalbarriers of mass transportation Therefore the use of MCTSas an in vitro tool for testing anticancer drugs has gainedsignificant interest as MCTS potentially provide a morereliable model for drug testing compared to single layeradherent cell cultures During the approval process of drugsbefore clinical testing in trials the mechanisms of deliveryand the effectiveness of the drugs must be determined Thefirst steps of preclinical drug testing are typically carriedout using adherent cell formats growing in two dimensions[157] However the outcome of such investigations in two-dimensional cell systems is often very different from what isobserved in a whole-body situation This makes it difficult todraw clear conclusions of the drug properties anticipated invivo In terms of drug delivery a spheroid test platform hasinherent advantages providing a natural barrier resemblingthe natural tumor environment Spheroids of a particularsize exhibit certain gradients of oxygen and nutrition [35ndash37 158 159] Spheroids larger than 400ndash500120583m in diametershow characteristics of layered structures with a hypoxiccore consisting of necrotic cells surrounded by quiescentcells and an outer layer of proliferating cells [38 160 161]Hence 3D tumor cell systems are a valuable tool for studyingdrug delivery and the response and metabolism of hypoxictumor cells to cancer therapy Fang et al reported thatmulticellular spheroids of primary human colon cancercells were resistant to chemotherapy-induced apoptosis andretained the expression of colon cancer marker CD133mimicking colorectal cancer [162] Were these cells grownunder normal conditions they did not retain these char-acteristics Size control of MCTS is a major challenge inobtaining uniform and reliable high throughput test systems


Table 2 Overview of various cell types used for tissue engineering and 120583119892-conditions involved

Cell type Engineered tissue Method ReferencesOsteoblasts Bone RWV bioreactor (s-120583119892) [8ndash12]Osteoprogenitor cells Bone RWV bioreactor (s-120583119892) [13 14]Mesenchymal stem cells Bone RWV bioreactor (s-120583119892) [12 15ndash23]Mesenchymal stem cells Divers RWV bioreactor (s-120583119892) [18 21 24ndash26]Mesenchymal stem cells Divers RPM (s-120583119892) [27ndash30]Several cell types MCTS RWV bioreactor (s-120583119892) [31ndash34]Several cell types MCTS Spinner flask (s-120583119892) [35ndash38]Several cell types MCTS RPM (s-120583119892) [39ndash44]Several cell types MCTS Space (r-120583119892) [45]Hepatocytes liver RWV bioreactor (s-120583119892) [34 46]

various techniques such as forced aggregation techniquesmicro textured surfaces and porous 3D scaffolds are beingemployed to solve these issues [163ndash166] There are severalconventional methods for generating 3D aggregates of cancercells including NASA rotary cell culture systems hangingdrops and culturing of cells using nonadherent surfaces[166ndash168] Conditions of s-120583119892 using the RPM (Figure 2(b)) orthe HARV have been shown to induce the growth of MCTSwithout the use of scaffolds in several types of human cancercells [31ndash33 39 169] The molecular switches initiating s-120583119892-induced spheroid formation are still unknown Severalchanges in morphology and gene expression profiles havebeen observed in follicular thyroid cancer cells grown unders-120583119892 conditions with the main features involving changes inthe ECM and early induction of apoptosis [39 40] Signalingbetween exogenous ECM and tumor surface receptors haslong been thought to be an essential component in reg-ulating the tumorigenic phenotype in 3D cultures Thesephenomena have been demonstrated in studies showing thatblocking specific ECM-integrin signaling can cause a shiftin the malignant potential of tumor cells leading to a morebenign phenotype [40 170] In an attempt to identify gravitysensitive genes responsible for MCTS formation Grosseet al [41] identified 487 transcripts which were differentlyregulated after 24 h of s-120583119892 in comparison to the groundcontrol Comparing adherent cells with MCTS under s-120583119892conditions revealed significant differences in terms of growthpatterns and signaling Interestingly the rate of apoptosis wasincreased in adherent cells compared to MCTS indicatingthat the early phase induction of apoptosis may be concomi-tant with the transition of cells shifting from2D to 3D growthSeveral NF-120581B-driven genes involved in the regulation oftumor invasion were upregulated by s-120583119892 in adherent cellshighlighting that s-120583119892 initiates distinct adaptive mechanismsin the cells

8 Summary

The development of tools like the RPM or RWV gave newimpacts in the field of tissue engineering (Table 1) Growingcells in simulated or real weightlessness for example on theRPMmight be a highly promising new technique to generate

tissue constructs in a scaffold-free manner Cultivation ofchondrocytes might lead to small cartilage particles whichcould be used to replace injured or outworn cartilageRestoring normal osteogenic differentiation of MSCs from s-120583119892 exposure by daily short-term stimulation could be helpfulso that tissue products may become commercially availablelike it has already happened for some bone tissues (BioSeed-Oral Bone codon osteotransplant and Osteocel)

At the moment studies analyzing the molecular mech-anisms behind spheroid formation of for example thyroidcells chondrocyte and others have increased the knowledgeof the complex regulation of 3D growth in 120583119892 [42ndash45 171]

To be able to use this new technique more efficientlyfurther studies are necessary to better understand the exactcellular changes specific to these conditions Tissue whichwas produced under s- or r-120583119892 conditions might be helpfulto better understand cell signaling intercellular contactand tissue growth as well as being sufficient for medicaltransplantationMCTS can be used as an alternative to animalexperiments

Although the 120583119892 environment is not a common fieldfor biologists and medical researchers recent studies haveclearly shown that the loss of gravity impacts the cells and itdramatically changes the genome proteome and secretomeof these cells [43 45] Therefore it is important to systemati-cally explore the advantages of this new research opportunityDifferent space flights have already demonstrated a 3D cellgrowth (Table 2) and similar results have been detectedwith the help of devices simulating 120583119892 in ground-basedlaboratories [34 44 46 172ndash174]



Acknowledgments

This paper was supported by the German Space Agency DLR(DanielaGrimm) (BMWiGrants 50WB0824 and 50WB1124)the European Space Agency (ESA CORA-GBF-2013-001


CORA-GBF-2013-004) and Aarhus University Denmark(Daniela Grimm) The authors would like to thank Mr PeterLindborg Los Angeles CA USA for reviewing the paper forlanguage and grammar errors

References


[2] A I Grigoriev V S Oganov A V Bakulin et al ldquoClinicaland physiological evaluation of bone changes among astronautsafter long-term space flightsrdquo Aviakosmicheskaia i Ekologich-eskaia Meditsina vol 32 no 1 pp 21ndash25 1998

[3] L Vico P Collet A Guignandon et al ldquoEffects of long-term microgravity exposure on cancellous and cortical weight-bearing bones of cosmonautsrdquoTheLancet vol 355 no 9215 pp1607ndash1611 2000

[4] S Tavella A Ruggiu A Giuliani et al ldquoBone turnover in wildtype and pleiotrophin-transgenicmice housed for threemonthsin the International Space Station (ISS)rdquo PLoS ONE vol 7 no3 Article ID e33179 2012



[7] D Grimm M Wehland J Pietsch J et al ldquoGrowing tissuesin real and simulated microgravity new methods for tissueengineeringrdquo Tissue Engineering Part B Reviews 2014

[8] C Granet N Laroche L Vico C Alexandre and M HLafage-Proust ldquoRotating-wall vessels promising bioreactors forosteoblastic cell culture comparison with other 3D conditionsrdquoMedical and Biological Engineering and Computing vol 36 no4 pp 513ndash519 1998

[9] K Song Z Yang T Liu et al ldquoFabrication and detection oftissue-engineered bones with bio-derived scaffolds in a rotatingbioreactorrdquo Biotechnology and Applied Biochemistry vol 45 no2 pp 65ndash74 2006

[10] K Song T Liu Z Cui X Li and X Ma ldquoThree-dimensionalfabrication of engineered bone with human bio-derived bonescaffolds in a rotating wall vessel bioreactorrdquo Journal of Biomed-ical Materials Research A vol 86 no 2 pp 323ndash332 2008

[11] K Kyriakidou G Lucarini A Zizzi et al ldquoDynamic co-seedingof osteoblast and endothelial cells on 3D polycaprolactonescaffolds for enhanced bone tissue engineeringrdquo Journal ofBioactive and Compatible Polymers vol 23 no 3 pp 227ndash2432008

[12] Q Lv L Nair and C T Laurencin ldquoFabrication charac-terization and in vitro evaluation of poly(lactic acid gly-colic acid)nano-hydroxyapatite composite microsphere-basedscaffolds for bone tissue engineering in rotating bioreactorsrdquoJournal of Biomedical Materials Research A vol 91 no 3 pp679ndash691 2009

[13] S R Facer R S Zaharias M E Andracki J Lafoon S KHunter and G B Schneider ldquoRotary culture enhances pre-osteoblast aggregation and mineralizationrdquo Journal of DentalResearch vol 84 no 6 pp 542ndash547 2005

[14] G B Schneider J K Boehrs J V Hoopes and D A SeaboldldquoUse of 3-dimensional environments to engineer osseous-liketissuerdquo Journal of Developmental Biology andTissue Engineeringvol 3 no 4 pp 42ndash47 2011

[15] D Turhani E Watzinger M Weiszligenbock et al ldquoAnalysis ofcell-seeded 3-dimensional bone constructs manufactured invitro with hydroxyapatite granules obtained from red algaerdquoJournal of Oral andMaxillofacial Surgery vol 63 no 5 pp 673ndash681 2005

[16] F Jin Y Zhang K Xuan et al ldquoEstablishment of three-dimen-sional tissue-engineered bone constructs under microgravity-simulated conditionsrdquo Artificial Organs vol 34 no 2 pp 118ndash125 2010

[17] J G Gershovich and L B Buravkova ldquoMorphofunctionalstatus and osteogenic differentiation potential of human mes-enchymal stromal precursor cells during in vitro modelingof microgravity effectsrdquo Bulletin of Experimental Biology andMedicine vol 144 no 4 pp 608ndash613 2007

[18] M Zayzafoon W E Gathings and J M McDonald ldquoModeledmicrogravity inhibits osteogenic differentiation of human mes-enchymal stem cells and increases adipogenesisrdquo Endocrinol-ogy vol 145 no 5 pp 2421ndash2432 2004

[19] R Saxena G Pan and J M McDonald ldquoOsteoblast andosteoclast differentiation in modeled microgravityrdquo Annals ofthe New York Academy of Sciences vol 1116 pp 494ndash498 2007


[21] D Sheyn G Pelled D Netanely E Domany and D GazitldquoThe effect of simulated microgravity on human mesenchymalstem cells cultured in an osteogenic differentiation system abioinformatics studyrdquo Tissue Engineering A vol 16 no 11 pp3403ndash3412 2010

[22] B Yu D Yu L Cao et al ldquoSimulated microgravity using arotary cell culture system promotes chondrogenesis of humanadipose-derived mesenchymal stem cells via the p38 MAPKpathwayrdquo Biochemical and Biophysical Research Communica-tions vol 414 no 2 pp 412ndash418 2011

[23] S M Uddin and Y X Qin ldquoEnhancement of osteogenicdifferentiation and proliferation in human mesenchymal stemcells by a modified low intensity ultrasound stimulation undersimulated microgravityrdquo PLoS ONE vol 8 no 9 Article IDe73914 2013


[25] X Wu S-H Li L-M Lou and Z-R Chen ldquoThe effect ofthe microgravity rotating culture system on the chondrogenicdifferentiation of bonemarrowmesenchymal stem cellsrdquoMolec-ular Biotechnology vol 54 no 2 pp 331ndash336 2013


[27] N VMerzlikina L B Buravkova and Y A Romanov ldquoThe pri-mary effects of clinorotation on cultured human mesenchymalstem cellsrdquo Journal of Gravitational Physiology vol 11 no 2 pp193ndash194 2004

[28] M Nishikawa H Ohgushi N Tamai et al ldquoThe effect ofsimulated microgravity by three-dimensional clinostat on bonetissue engineeringrdquoCell Transplantation vol 14 no 10 pp 829ndash835 2005


[29] L Yuge T Kajiume H Tahara et al ldquoMicrogravity potentiatesstem cell proliferation while sustaining the capability of differ-entiationrdquo Stem Cells and Development vol 15 no 6 pp 921ndash929 2006

[30] I G Gershovich and L B Buravkova ldquoInterleukine productionin culture of mesenchymal stromal cells of humans duringsimulation of the microgravity effectsrdquo Aviakosmicheskaia IEkologicheskaia Meditsina vol 43 no 3 pp 44ndash50 2009

[31] M IngramG B Techy R Saroufeem et al ldquoThree-dimensionalgrowth patterns of various human tumor cell lines in simulatedmicrogravity of a nasa bioreactorrdquo In Vitro Cellular and Devel-opmental Biology Animal vol 33 no 6 pp 459ndash466 1997

[32] B Marrero J L Messina and R Heller ldquoGeneration of atumor spheroid in a microgravity environment as a 3D modelof melanomardquo In Vitro Cellular and Developmental BiologymdashAnimal vol 45 no 9 pp 523ndash534 2009

[33] S M Konstantinov M M Mindova P T Gospodinov andP I Genova ldquoThree-dimensional bioreactor cultures a usefuldynamic model for the study of cellular interactionsrdquo Annals ofthe New York Academy of Sciences vol 1030 pp 103ndash115 2004

[34] T T Chang and M Hughes-Fulford ldquoMonolayer and spheroidculture of human liver hepatocellular carcinoma cell line cellsdemonstrate distinct global gene expression patterns and func-tional phenotypesrdquo Tissue Engineering A vol 15 no 3 pp 559ndash567 2009

[35] A Bredel-Geissler U Karbach S Walenta L Vollrath and WMueller-Klieser ldquoProliferation-associated oxygen consumptionand morphology of tumor cells in monolayer and spheroidculturerdquo Journal of Cellular Physiology vol 153 no 1 pp 44ndash52 1992

[36] L A Kunz-Schughart C Groebe and W Mueller-KlieserldquoThree-dimensional cell culture induces novel proliferative andmetabolic alterations associated with oncogenic transforma-tionrdquo International Journal of Cancer vol 66 no 4 pp 578ndash5861996

[37] M Wartenberg F Donmez F C Ling H Acker J Hes-cheler and H Sauer ldquoTumor-induced angiogenesis studied inconfrontation cultures of multicellular tumor spheroids andembryoid bodies grown from pluripotent embryonic stemcellsrdquoThe FASEB Journal vol 15 no 6 pp 995ndash1005 2001

[38] R-Z Lin and H-Y Chang ldquoRecent advances in three-dimensional multicellular spheroid culture for biomedicalresearchrdquo Biotechnology Journal vol 3 no 9-10 pp 1172ndash11842008






[44] J Pietsch A Sickmann G Weber et al ldquoA proteomic approachto analysing spheroid formation of two human thyroid cell linescultured on a random positioning machinerdquo Proteomics vol 11no 10 pp 2095ndash2104 2011


[46] T T Chang and M Hughes-Fulford ldquoMolecular mechanismsunderlying the enhanced functions of three-dimensional hep-atocyte aggregatesrdquo Biomaterials vol 35 no 7 pp 2162ndash21712014

[47] M Y Kapitonova N Salim S Othman et al ldquoAlteration of cellcytoskeleton and functions of cell recovery of normal humanosteoblast cells caused by factors associated with real spaceflightrdquo The Malaysian Journal of Pathology vol 35 no 2 pp153ndash163 2013

[48] E A Blaber N Dvorochkin C Lee et al ldquoMicrogravityinducespelvic bone loss throughosteoclastic activity osteocytic osteoly-sis and osteoblastic cell cycle inhibition by CDKN1120572p21rdquo PLoSONE vol 8 no 4 Article ID 61372 2013



[51] R Hemmersbach M Krause R Braucker and K IvanovaldquoGraviperception in ciliates steps in the transduction chainrdquoAdvances in Space Research vol 35 no 2 pp 296ndash299 2005

[52] D A M Mesland ldquoNovel ground-based facilities for researchin the effects of weightrdquo ESA Microgravity News vol 9 1996

[53] D A M Meslana A H Anton H Willemsen and H vanden Ende ldquoThe free fall machinemdasha ground-based facility formicrogravity research in life sciencesrdquoMicrogravity Science andTechnology vol 9 no 1 pp 10ndash14 1996

[54] M Schwarzenberg P Pippia M A Meloni G Cossu MCogoli-Greuter and A Cogoli ldquoSignal transduction in Tlymphocytesmdasha comparison of the data from space the freefall machine and the random positioning machinerdquo Advancesin Space Research vol 24 no 6 pp 793ndash800 1999

[55] M V Berry and A K Geim ldquoOf flying frogs and levitronsrdquoEuropean Journal of Physics vol 18 no 4 pp 307ndash313 1997

[56] J M Valles Jr K Lin J M Denegre and K L Mowry ldquoStablemagnetic field gradient levitation ofXenopus laevis toward low-gravity simulationrdquo Biophysical Journal vol 73 no 2 pp 1130ndash1133 1997

[57] M J A Moes J C Gielen R-J Bleichrodt J J W A van LoonP C M Christianen and J Boonstra ldquoSimulation of micro-gravity by magnetic levitation and random positioning effecton human A431 Cell morphologyrdquo Microgravity Science andTechnology vol 23 no 2 pp 249ndash261 2011



[59] T Higashi A Yamagishi T Takeuchi et al ldquoOrientation oferythrocytes in a strong static magnetic fieldrdquo Blood vol 82 no4 pp 1328ndash1334 1993

[60] K A Mirica F Ilievski A K Ellerbee S S Shevkoplyasand G M Whitesides ldquoUsing magnetic levitation for threedimensional self-assemblyrdquo Advanced Materials vol 23 no 36pp 4134ndash4140 2011

[61] J S Brooks J A Reavis R A Medwood et al ldquoNew opportu-nities in science materials and biological systems in the low-gravity (magnetic levitation) environment (invited)rdquo Journal ofApplied Physics vol 87 no 9 pp 6194ndash6199 2000


[63] C E Dijkstra O J Larkin P Anthony et al ldquoDiamagneticlevitation enhances growth of liquid bacterial cultures byincreasing oxygen availabilityrdquo Journal of the Royal SocietyInterface vol 8 no 56 pp 334ndash344 2011

[64] D Shi R Meng W Deng et al ldquoEffects of microgravitymodeled by large gradient highmagnetic field on the osteogenicinitiation of humanmesenchymal stem cellsrdquo Stem Cell Reviewsand Reports vol 6 no 4 pp 567ndash578 2010


[66] F Castro-Chavez K C Vickers J S Lee C Tung and J DMorrisett ldquoEffect of lyso-phosphatidylcholine and Schnurri-3on osteogenic transdifferentiation of vascular smooth musclecells to calcifying vascular cells in 3D culturerdquo Biochimica etBiophysica Acta vol 1830 no 6 pp 3828ndash3834 2013

[67] F G J R von Sachs ldquoUber Ausschliessung der geotropischenund heliotropischen Krummungen wahrend des WachstumsrdquoWurzburger Arbeiten vol 2 pp 209ndash225 1879


[69] S Murakami and M Yamada ldquoArchitecture of statocytes andchloroplasts under the microgravity environmentrdquo BiologicalSciences in Space vol 2 no 4 p 301 1988

[70] J J W A van Loon ldquoEffects of spaceflight on biologicalsystemsrdquo in Biology in Space and Life on Earth E BrinckmannEd vol 17 Wiley-VCH 2007


[72] T Hoson S Kamisaka Y Masuda M Yamashita and BBuchen ldquoEvaluation of the three-dimensional clinostat as asimulator of weightlessnessrdquo Planta vol 203 pp S187ndashS1971997


[74] C A D Leguy R Delfos and M J B M Pourquie ldquoFluidmotion for microgravity simulations in a random positioningmachinerdquoGravitational and Space Biology vol 25 no 1 pp 36ndash39 2011

[75] S Wakayama Y Kawahara C Li K Yamagata L Yuge andT Wakayama ldquoDetrimental effects of microgravity on mousepreimplantation development in vitrordquo PLoS ONE vol 4 no 8Article ID e6753 2009

[76] M Z Luo R Meng S S Li et al ldquoWeightlessness simulatedwith random positioning machine influences the cytoskeletonand migration of MC3T3-E1 cellsrdquo Journal of the Japan Societyof Microgravity Application vol 28 no 2 pp S41ndashS45 2011

[77] W A Loesberg X F Walboomers E M Bronkhorst J JW A Van Loon and J A Jansen ldquoThe effect of combinedsimulated microgravity and microgrooved surface topographyon fibroblastsrdquo Cell Motility and the Cytoskeleton vol 64 no 3pp 174ndash185 2007

[78] G Aleshcheva J Sahana X Ma et al ldquoChanges inmorphologygene expression and protein content in chondrocytes culturedon a Random Positioning Machinerdquo PLoS ONE vol 8 no 11Article ID e79057 2013

[79] L Prodanov J J W A van Loon J te Riet J A Jansenand X F Walboomers ldquoNanostructured substrate conforma-tion can decrease osteoblast-like cell dysfunction in simulatedmicrogravity conditionsrdquo Journal of Tissue Engineering andRegenerative Medicine 2012

[80] J Ichigi and M Asashima ldquoDome formation and tubulemorphogenesis by Xenopus kidney A6 cell cultures exposed tomicrogravity simulated with a 3D-clinostat and to hypergrav-ityrdquo In Vitro Cellular Developmental Biology Animal vol 37 no1 pp 31ndash44 2001

[81] Y D Jiang W N Li L F Wang Z Y Zhang B M Zhang andH J Wu ldquoSeveral new types of clinostatsrdquo Space Medicine ampMedical Engineering vol 21 no 4 pp 368ndash371 2008


[83] L Xiang F Qi D Dai C Li and Y Jiang ldquoSimulatedmicrogravity affects growth of Escherichia coli and recombi-nant 120573-D-glucuronidase productionrdquoApplied Biochemistry andBiotechnology vol 162 no 3 pp 654ndash661 2010

[84] R Hemmersbach S M Strauch D Seibt and M SchuberldquoComparative studies on gravisensitive protists on ground (2Dand 3D clinostats) and in microgravityrdquo Microgravity Scienceand Technology vol 18 no 3-4 pp 257ndash259 2006

[85] T J Goodwin J M Jessup and D A Wolf ldquoMorphologic dif-ferentiation of colon carcinoma cell lines HT-29 andHT-29KMin Rotating-Wall Vesselsrdquo In Vitro Cellular and DevelopmentalBiology Animal vol 28 no 1 pp 47ndash60 1992


[87] T Liu X Li X Sun X Ma and Z Cui ldquoAnalysis on forcesand movement of cultivated particles in a rotating wall vesselbioreactorrdquo Biochemical Engineering Journal vol 18 no 2 pp97ndash104 2004

[88] T J Goodwin T L Prewett D A Wolf and G F SpauldingldquoReduced shear stress a major component in the ability ofmammalian tissues to form three-dimensional assemblies insimulated microgravityrdquo Journal of Cellular Biochemistry vol51 no 3 pp 301ndash311 1993

[89] E A Nauman C M Ott E Sander et al ldquoNovel quantitativebiosystem formodeling physiological fluid shear stress on cellsrdquoApplied and EnvironmentalMicrobiology vol 73 no 3 pp 699ndash705 2007

[90] XGuo Y ZhaoHChang et al ldquoCreation of engineered cardiactissue in vitro from mouse embryonic stem cellsrdquo Circulationvol 113 no 18 pp 2229ndash2237 2006


[91] D L Tucker C M Ott S Huff et al ldquoCharacterization ofEscherichia coli MG1655 grown in a low-shear modeled microgravity environmentrdquo BMCMicrobiology vol 7 article 15 2007

[92] M N Cinbiz R S Tıgli I G Beskardes M Gumusdereliogluand U Colak ldquoComputational fluid dynamics modeling ofmomentum transport in rotating wall perfused bioreactor forcartilage tissue engineeringrdquo Journal of Biotechnology vol 150no 3 pp 389ndash395 2010

[93] A Puca G Russo and A Giordano ldquoProperties of mechano-transduction via simulated microgravity and its effects onintracellular trafficking of VEGFRrsquosrdquo Oncotarget vol 3 no 4pp 426ndash434 2012


[95] Y C Wang S Zhang T Y Du B Wang and X Q SunldquoClinorotation upregulates inducible nitric oxide synthase byinhibiting AP-1 activation in human umbilical vein endothelialcellsrdquo Journal of Cellular Biochemistry vol 107 no 2 pp 357ndash363 2009


[97] P I Lelkes N Akhtar E Lelkes et al ldquoNeuroendocrine tissueengineering in rotating wall vessel bioreactors under simulatedmicrogravity conditionsrdquo in Proceedings of the 23rd AnnualInternational Conference of the IEEE Engineering in Medicineand Biology Society vol 3 pp 2987ndash2990 October 2001

[98] A Abbott ldquoCell culture biologyrsquos new dimensionsrdquoNature vol424 no 6951 pp 870ndash872 2003






[104] X Ma M Wehland H Schulz et al ldquoGenomic Approach toidentify factors that drive the formation of three-dimensionalstructures by EAhy926 endothelial cellsrdquo PLoS ONE vol 8 no5 Article ID e64402 2013

[105] C Ulbrich K Westphal J Pietsch et al ldquoCharacterizationof human chondrocytes exposed to simulated microgravityrdquoCellular Physiology and Biochemistry vol 25 no 4-5 pp 551ndash560 2010

[106] L E Freed and G Vunjak-Novakovic ldquoCultivation of cell-polymer tissue constructs in simulated microgravityrdquo Biotech-nology and Bioengineering vol 46 no 4 pp 306ndash313 1995


[108] V Stamenkovic G Keller D Nesic A Cogoli and S P GroganldquoNeocartilage formation in 1 g simulated and microgravityenvironments implications for tissue engineeringrdquo Tissue Engi-neering A vol 16 no 5 pp 1729ndash1736 2010

[109] L Kock C C van Donkelaar and K Ito ldquoTissue engineering offunctional articular cartilage the current statusrdquoCell and TissueResearch vol 347 no 3 pp 613ndash627 2012

[110] C H Chang F H Lin T F Kuo and H C Liu ldquoCartilagetissue engineeringrdquo Biomedical Engineering Applications Basisand Communications vol 17 no 2 pp 61ndash71 2005

[111] E Ozcivici Y K Luu B Adler et al ldquoMechanical signals asanabolic agents in bonerdquo Nature Reviews Rheumatology vol 6no 1 pp 50ndash59 2010

[112] R Yang W Lin Y Chen et al ldquoRegulation by ultrasoundtreatment on the integrin expression and differentiation ofosteoblastsrdquo Bone vol 36 no 2 pp 276ndash283 2005

[113] H C Hsu Y C Fong C S Chang et al ldquoUltrasound inducescyclooxygenase-2 expression through integrin integrin-linkedkinase Akt NF-120581B and p300 pathway in human chondrocytesrdquoCellular Signalling vol 19 no 11 pp 2317ndash2328 2007

[114] Y I Arfat W Z Xiao S Iftikhar et al ldquoPhysiological effects ofmicrogravity on bone cellsrdquo Calcified Tissue International vol94 no 6 pp 569ndash579 2014

[115] D A Starr ldquoCommunication between the cytoskeleton and thenuclear envelope to position the nucleusrdquoMolecular BioSystemsvol 3 no 9 pp 583ndash589 2007


[117] S B Doty E R Morey-Holton G N Durnova and A SKaplansky ldquoCosmos 1887 morphology histochemistry andvasculature of the growing rat tibiardquo FASEB Journal vol 4 no1 pp 16ndash23 1990

[118] N V Rodionova O V Polkovenko and V S Oganov ldquoInter-actions of cells in zones of bone resorption under microgravityand hypokinesiardquo Journal of Gravitational Physiology vol 11no 2 pp P147ndash151 2004

[119] O D Kennedy B C Herman D M Laudier R J MajeskaH B Sun and M B Schaffler ldquoActivation of resorption infatigue-loaded bone involves both apoptosis and active pro-osteoclastogenic signaling by distinct osteocyte populationsrdquoBone vol 50 no 5 pp 1115ndash1122 2012

[120] C Lin X Jiang Z Dai et al ldquoSclerostinmediates bone responseto mechanical unloading through antagonizing Wnt120573-cateninsignalingrdquo Journal of Bone and Mineral Research vol 24 no 10pp 1651ndash1661 2009


[122] A Sundaresan S F Clarke and N R Pellis Production ofBoneMorphogenic Proteins (BMPS)Using aNovel Tissue CulturePlatform 2011 US Patent 8076136


[123] A Sundaresan S F Clarke and N R Pellis ldquoConstructionsosseuses mineralisees en trois dimensionsrdquo European PatentNumber EP2013730 2011

[124] A Sundaresan M S F Clarke andM Brinker Development ofa human colloidal bone graft material US 8506982 B2 2013

[125] J R Milstead S J Simske and T A Bateman ldquoSpaceflightand hindlimb suspension disuse models in micerdquo BiomedicalSciences Instrumentation vol 40 pp 105ndash110 2004

[126] H M Frost and W S Jee ldquoOn the rat model of humanosteopenias and osteoporosesrdquo Bone and Mineral vol 18 no3 pp 227ndash236 1992


[128] S M Smith M EWastney K O OrsquoBrien et al ldquoBonemarkerscalcium metabolism and calcium kinetics during extended-duration space flight on the Mir Space Stationrdquo Journal of Boneand Mineral Research vol 20 no 2 pp 208ndash218 2005

[129] JMVogel ldquoBonemineralmeasurement Skylab experimentM-078rdquo Acta Astronautica vol 2 no 1-2 pp 129ndash139 1975

[130] S M Smith M E Wastney B V Morukov et al ldquoCalciummetabolism before during and after a 3-mo spaceflight kineticand biochemical changesrdquo American Journal of Physiology vol277 no 1 part 2 pp R1ndashR10 1999

[131] J E Zerwekh L A Ruml F Gottschalk and C Y C PakldquoThe effects of twelve weeks of bed rest on bone histology bio-chemicalmarkers of bone turnover and calciumhomeostasis ineleven normal subjectsrdquo Journal of Bone and Mineral Researchvol 13 no 10 pp 1594ndash1601 1998

[132] T Lang A LeBlanc H Evans Y Lu H Genant and A YuldquoCortical and trabecular bone mineral loss from the spine andhip in long-duration spaceflightrdquo Journal of Bone and MineralResearch vol 19 no 6 pp 1006ndash1012 2004

[133] A LeBlanc C Lin L Shackelford et al ldquoMuscle volume MRIrelaxation times (T2) and body composition after spaceflightrdquoJournal of Applied Physiology vol 89 no 6 pp 2158ndash2164 2000

[134] M Iki E Kajita Y Dohi et al ldquoAge menopause bone turnovermarkers and lumbar bone loss in healthy Japanese womenrdquoMaturitas vol 25 no 1 pp 59ndash67 1996

[135] J Sirola H Kroger R Honkanen et al ldquoFactors affecting boneloss around menopause in women without HRT a prospectivestudyrdquoMaturitas vol 45 no 3 pp 159ndash167 2003


[137] T F Lang A D Leblanc H J Evans and Y Lu ldquoAdaptationof the proximal femur to skeletal reloading after long-durationspaceflightrdquo Journal of Bone andMineral Research vol 21 no 8pp 1224ndash1230 2006

[138] J J W A Van Loon D-J Bervoets E H Burger et alldquoDecreased mineralization and increased calcium release inisolated fetal mouse long bones under near weightlessnessrdquoJournal of Bone andMineral Research vol 10 no 4 pp 550ndash5571995

[139] R Langer ldquoTissue engineering a new field and its challengesrdquoPharmaceutical Research vol 14 no 7 pp 840ndash841 1997

[140] G M Crane S L Ishaug and A G Mikos ldquoBone tissueengineeringrdquoNatureMedicine vol 1 no 12 pp 1322ndash1324 1995

[141] M J Yaszemski R G Payne W C Hayes R Langer and A GMikos ldquoEvolution of bone transplantation molecular cellular

and tissue strategies to engineer human bonerdquoBiomaterials vol17 no 2 pp 175ndash185 1996

[142] E M Younger and M W Chapman ldquoMorbidity at bone graftdonor sitesrdquo Journal of Orthopaedic Trauma vol 3 no 3 pp192ndash195 1989

[143] S Gronthos and P J Simmons ldquoThe biology and applicationof human bone marrow stromal cell precursorsrdquo Journal ofHematotherapy and Stem Cell Research vol 5 no 1 pp 15ndash231996

[144] N Jaiswal S E Haynesworth A I Caplan and S P BruderldquoOsteogenic differentiation of purified culture-expandedhuman mesenchymal stem cells in vitrordquo Journal of CellularBiochemistry vol 64 no 2 pp 295ndash312 1997

[145] J E Aubin ldquoOsteoprogenitor cell frequency in rat bonemarrowstromal populations role for heterotypic cell-cell interactions inosteoblast differentiationrdquo Journal of Cellular Biochemistry vol72 no 3 pp 396ndash410 1999

[146] J Rauh F Milan K Gunther and M Stiehler ldquoBioreactorsystems for bone tissue engineeringrdquo Tissue Engineering BReviews vol 17 no 4 pp 263ndash280 2011

[147] S Kern H Eichler J Stoeve H Kluter and K BiebackldquoComparative analysis of mesenchymal stem cells from bonemarrow umbilical cord blood or adipose tissuerdquo StemCells vol24 no 5 pp 1294ndash1301 2006

[148] M Kassem and BM Abdallah ldquoHuman bone-marrow-derivedmesenchymal stem cells biological characteristics and poten-tial role in therapy of degenerative diseasesrdquo Cell and TissueResearch vol 331 no 1 pp 157ndash163 2008

[149] J E Dennis S E Haynesworth R G Young and A I CaplanldquoOsteogenesis in marrow-derived mesenchymal cell porousceramic composites transplanted subcutaneously effect offibronectin and laminin on cell retention and rate of osteogenicexpressionrdquo Cell Transplantation vol 1 no 1 pp 23ndash32 1992

[150] B D Boyan T W Hummert D D Dean and Z SchwartzldquoRole of material surfaces in regulating bone and cartilage cellresponserdquo Biomaterials vol 17 no 2 pp 137ndash146 1996

[151] A L Olivares E Marsal J A Planell and D Lacroix ldquoFiniteelement study of scaffold architecture design and culture con-ditions for tissue engineeringrdquo Biomaterials vol 30 no 30 pp6142ndash6149 2009

[152] S B Vangordon R S Voronov T B Blue R L ShambaughD V Papavassiliou and V I Sikavitsas ldquoEffects of scaffoldarchitecture on preosteoblastic cultures under continuous fluidshearrdquo Industrial amp Engineering Chemistry Research vol 50 no2 pp 620ndash629 2011

[153] E Canalis A N Economides and E Gazzerro ldquoBonemorpho-genetic proteins their antagonists and the skeletonrdquo EndocrineReviews vol 24 no 2 pp 218ndash235 2003

[154] B L Eppley J E Woodell and J Higgins ldquoPlatelet quantifi-cation and growth factor analysis from platelet-rich plasmaimplications for wound healingrdquo Plastic and ReconstructiveSurgery vol 114 no 6 pp 1502ndash1508 2004

[155] E A Botchwey S R Pollack E M Levine and C T LaurencinldquoBone tissue engineering in a rotating bioreactor using amicrocarrier matrix systemrdquo Journal of Biomedical MaterialsResearch vol 55 no 2 pp 242ndash253 2001

[156] J Chen R Liu Y Yang et al ldquoThe simulated microgravityenhances the differentiation of mesenchymal stem cells intoneuronsrdquo Neuroscience Letters vol 505 no 2 pp 171ndash175 2011

[157] D N Karunaratne P S Silverstein V Vasadani et al ldquoCellculture models for drug transport studiesrdquo in Drug Delivery


Principles and Applications B Wang T Siahaan and R SolteroEds pp 103ndash124 JohnWiley amp Sons Hoboken NJ USA 2005

[158] M Gassmann J Fandrey S Bichet et al ldquoOxygen supply andoxygen-dependent gene expression in differentiating embry-onic stem cellsrdquo Proceedings of the National Academy of Sciencesof the United States of America vol 93 no 7 pp 2867ndash28721996

[159] R M Sutherland ldquoCell and environment interactions in tumormicroregions the multicell spheroid modelrdquo Science vol 240no 4849 pp 177ndash184 1988

[160] L A Kunz-Schughart J P Freyer F Hofstaedter and R EbnerldquoThe use of 3-D cultures for high-throughput screening themulticellular spheroid modelrdquo Journal of Biomolecular Screen-ing vol 9 no 4 pp 273ndash285 2004

[161] F Hirschhaeuser H Menne C Dittfeld J West W Mueller-Klieser and L A Kunz-Schughart ldquoMulticellular tumorspheroids an underestimated tool is catching up againrdquo Journalof Biotechnology vol 148 no 1 pp 3ndash15 2010

[162] D D Fang Y J Kim C N Lee et al ldquoExpansion of CD133+colon cancer cultures retaining stem cell properties to enablecancer stem cell target discoveryrdquo British Journal of Cancer vol102 no 8 pp 1265ndash1275 2010

[163] S M Ong C Zhang Y Toh et al ldquoA gel-free 3D microfluidiccell culture systemrdquo Biomaterials vol 29 no 22 pp 3237ndash32442008

[164] Y-S Torisawa B-H Chueh D Huh et al ldquoEfficient formationof uniform-sized embryoid bodies using a compartmentalizedmicrochannel devicerdquo Lab on a Chip vol 7 no 6 pp 770ndash7762007

[165] L Y Wu D Di Carlo and L P Lee ldquoMicrofluidic self-assemblyof tumor spheroids for anticancer drug discoveryrdquo BiomedicalMicrodevices vol 10 no 2 pp 197ndash202 2008

[166] Y S Torisawa A Takagi Y Nashimoto T Yasukawa HShiku and T Matsue ldquoA multicellular spheroid array to realizespheroid formation culture and viability assay on a chiprdquoBiomaterials vol 28 no 3 pp 559ndash566 2007

[167] J M Kelm and M Fussenegger ldquoMicroscale tissue engineeringusing gravity-enforced cell assemblyrdquo Trends in Biotechnologyvol 22 no 4 pp 195ndash202 2004

[168] J Friedrich R Ebner and L A Kunz-Schughart ldquoExperimentalanti-tumor therapy in 3-D spheroidsmdashold hat or new chal-lengerdquo International Journal of Radiation Biology vol 83 no11-12 pp 849ndash871 2007

[169] J M Kelm and M Fussenegger ldquoScaffold-free cell delivery foruse in regenerative medicinerdquoAdvanced Drug Delivery Reviewsvol 62 no 7-8 pp 753ndash764 2010

[170] V M Weaver O W Petersen F Wang et al ldquoReversionof the malignant phenotype of human breast cells in three-dimensional culture and in vivo by integrin blocking antibod-iesrdquoThe Journal of Cell Biology vol 137 no 1 pp 231ndash245 1997


[172] Y Naito T Shinoka D Duncan et al ldquoVascular tissue engi-neering towards the next generation vascular graftsrdquo AdvancedDrug Delivery Reviews vol 63 no 4-5 pp 312ndash323 2011


[174] S M Grenon M Jeanne J Aguado-Zuniga M S Conteand M Hughes-Fulford ldquoEffects of gravitational mechanical

unloading in endothelial cells association between caveolinsinflammation and adhesion moleculesrdquo Scientific Reports vol3 article 1494 2013

Review ArticleMultisensory Integration and Internal Models forSensing Gravity Effects in Primates

Francesco Lacquaniti123 Gianfranco Bosco123 Silvio Gravano13 Iole Indovina13

Barbara La Scaleia3 Vincenzo Maffei3 and Myrka Zago3

1 Centre of Space Bio-Medicine University of Rome Tor Vergata Via Montpellier 1 00133 Rome Italy2 Department of Systems Medicine University of Rome Tor Vergata Via Montpellier 1 00133 Rome Italy3 Laboratory of Neuromotor Physiology IRCCS Santa Lucia Foundation Via Ardeatina 306 00179 Rome Italy

Correspondence should be addressed to Francesco Lacquaniti lacquanitimeduniroma2it

Received 2 May 2014 Accepted 26 May 2014 Published 1 July 2014


Copyright copy 2014 Francesco Lacquaniti et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Gravity is crucial for spatial perception postural equilibrium and movement generation The vestibular apparatus is the mainsensory system involved in monitoring gravity Hair cells in the vestibular maculae respond to gravitoinertial forces but theycannot distinguish between linear accelerations and changes of head orientation relative to gravityThe brain deals with this sensoryambiguity (which can cause some lethal airplane accidents) by combining several cues with the otolith signals angular velocitysignals provided by the semicircular canals proprioceptive signals from muscles and tendons visceral signals related to gravityand visual signals In particular vision provides both static and dynamic signals about body orientation relative to the vertical butit poorly discriminates arbitrary accelerations of moving objects However we are able to visually detect the specific accelerationof gravity since early infancy This ability depends on the fact that gravity effects are stored in brain regions which integrate visualvestibular and neck proprioceptive signals and combine this information with an internal model of gravity effects

1 Introduction

Intuitively sensing gravity effects should be a trivial problemfor a complex nervous system such as our own On the onehand direction and magnitude of gravity are quasi-constanton Earth Thus gravitational acceleration varies by lt1 bychanging latitude or altitude while the vertical deflection islt005∘ On the other hand our nervous system is compu-tationally high-powered being endowed with asymp1011 neuronsinterconnected via asymp1015 synapses All axons pieced togetherwould cover the distance between the Earth and the Moon(about 400000 km) One would assume that we are able tomonitor gravity directly by means of our sensory systemsbut this is not the case As we shall review in this papergravity effects are only extrapolated indirectly by the brain bycombining multisensory information with internal modelsthat is with neural processes which mimic a physical event

Sensing and coping with gravity is crucial for spaceperception control of upright posture and generation of

movements Indeed gravity provides a unique reference axisto which we can anchor body orientation and monitor orien-tation changes Gravity effects on limb and body movementsare two-sided insofar as gravity acts both as a perturbingforce that must be counteracted to avoid falling down and asa facilitating force which allows walking and running via theground contact forces

2 Vestibular Information

The vestibular receptors lie inside the labyrinth of thetemporal bone Somewhat similar sensors evolved first ininvertebrates and then in vertebrates about 500 Myrs ago[1] The vestibular apparatus acts as an inertial navigationsystem including in each ear three semicircular canalsoriented roughly orthogonal to each other and two otolithicorgans the sacculus and utriculus with sensory epitheliaoriented roughly vertically and horizontally respectively [2]



The vestibular sensors function as accelerometers the semi-circular canals transducing angular accelerations (roll yawand pitch) and the otoliths transducing linear accelerationsHead acceleration bends the cilia of the hair cells in thesensory organs resulting in a change of the membranepotential and synaptic transmission of the neurons of thevestibular ganglion innervating the receptors The signalsfrom the vestibular neurons carry information about headvelocity and acceleration to the vestibular nuclei in the brainstem In turn signals from these nuclei are relayed andprocessed in several regions of the brain and spinal cordgiving rise to sensations and movements [3]

Thewidely distributed polarities of response of the otolithreceptors in the maculae allow monitoring acceleration vec-tors in any arbitrary direction (Figure 1(a)) These receptorsare extremely sensitive being able to detect displacementsof the cilia as small as 03 nm (typical atomic diameter) andcorrespondingly small accelerations In fact the receptorsin the maculae respond to the projection of an appliedforce (or acceleration) Thus the component of gravitationalacceleration projected on the saccular macula is 119892 cos120572where 120572 is the angle of tilt of the head relative to thegravity direction whereas the component of gravitationalacceleration projected on the utriclemacula is119892 sin120572 Gravityaccelerates the body downwards and is opposed by theground contact forcesThese contact forces are transmitted toall body segments and to the head where they are monitoredby the otolith receptors These receptors respond to a tiltof the head relative to gravity but in general they cannotprovide a unique measurement of gravity effects As anyaccelerometer also the sacculus and utricle respond to netgravitoinertial accelerations and they cannot distinguishbetween the gravitational and the inertial component Forinstance otolith afferents cannot distinguish whether weare accelerating backward (Figure 1(b)) or tilting the headforward (Figure 1(c)) This is because the effect of gravity islocally indistinguishable from the effect of a linear accelera-tion of the reference system [8] In fact the otolith afferentssignal the net gravitoinertial acceleration (a) resulting fromthe vector difference between the gravity vector (g) and thelinear acceleration vector (119891)

119886 = 119892 minus 119891 (1)

All vectors are time-varying referred to head-fixed coordi-nates of the vestibular sensors

This intrinsic ambiguity can give rise to perceptualillusions which become extremely dangerous under somespecific conditions For instance during takeoff an airplanepilot may sense an erroneously high value of upward pitchbecause the resultant of the vector sum of gravity andbackward inertial acceleration is misperceived as the actualorientation relative to the vertical Under conditions of lowvisibility and without the aid of instruments the pilot maythen attempt to correct the aircraft attitude by pitchingdownward with the risk of impacting the ground Spatialdisorientations originating from sensory ambiguities of thiskind are often involved in severe aviation accidents [9]

Under usual conditions however we have no difficultyin sensing the orientation of the head relative to the vertical

even with the eyes closed and in the presence of appreciableaccelerations provided the latter have short duration (asthose of a car or train) This is because the ambiguity can besolved by the brain using a variety of ldquotricksrdquo First the brainfilters the otolith signals so that the low-frequency (longerlasting) signals are automatically interpreted as a change inthe tilt angle of the head relative to gravity [10] Converselyhigh-frequency (shorter lasting) signals are interpreted asrelated to a linear acceleration Also the perceptual illusionof the aircraft pilot mentioned above is consistent withthe frequency segregation hypothesis A prolonged linearacceleration (such as that at airplane takeoff) is a very rareevent when it occurs it is interpreted erroneously as a tiltrelative to gravity Notice however that unlike the outputof a simple low-pass filter the phase of perceived tilt hasbeen shown to be relatively constant across a broad frequencyrange [3 11]

A second ldquotrickrdquo used to disambiguate gravitoinertialacceleration consists in combining the otolith signals withthose of the semicircular canals [11ndash13] just as the man-made inertial systems which combine accelerometers andgyroscopesWhen we turn our heads the semicircular canalsintegrate the angular acceleration and signal the correspond-ing angular velocity for frequencies above about 005Hz [14]Information about angular head velocity can then be usedto keep track of changes in orientation of the gravity vectorrelative to the head [15 16] Formally

119892 = 119892 times 120596 (2)

where 119892 denotes the time derivative of the gravity vec-tor 120596 denotes the angular velocity and times denotes vectorcross-product An internal three-dimensional estimate of thegravity vector in head coordinates can then be obtained byintegrating (2) if the initial conditions for 119892 are known

119892 = int119892 times 120596119889119905 (3)

Notice that the angular velocity that needs to be integratedin (3) is represented by the component parallel to the Earthhorizon because this component changes the orientation ofthe head relative to gravity Given the estimate of 119892 providedby (3) gravitoinertial accelerations can be disambiguated bysolving (1)

A potential problemwith themodel outlined above is thatthe semicircular canals do not provide a reliable estimate ofangular velocity at steady-state [14] Errors in the estimate of120596 would determine an error in the estimate of tilt relativeto gravity provided by (3) A solution consists in correctingthe errors by means of the so-called somatogravic feedback(Figure 2) which tends to align the estimate of the gravita-tional acceleration with the gravitoinertial acceleration [4 517] In other words the time-average of the gravitoinertialacceleration over several seconds yields an estimate of gravityorientation at low frequencies The somatogravic effect canbe incorporated in the model of (2) by including a low-pass


Endolymph fluid

Otolith membrane

Gel layer

Hair cells

(a)

Translation

120572

a = f

(b)

Tilt

120572

a = g

(c)

Figure 1 (a)Otolith organs Left utricle Center saccule Arrows indicate the local on-directions of the hair cells thick black lines indicate thestriola Right cross-section through the otolith membrane showing the different layers Licensed under the Creative Commons Attribution-Share (authorsThomasHaslwanter andRudi Jaeger) (b)-(c) Tilt-translation ambiguity of otolith receptors (b)Theupright head is acceleratedbackward (c) The head is tilted forward These two gravitoinertial accelerations cannot be discriminated by the otolith sensory neurons

filtered term to the tilt estimate thereby canceling any drift[4] The resulting equation is

119892 = 119892 times 120596 minus119892 minus 119886

120591 (4)

The time constant 120591 controls the gain and phase of the 119892estimate when the otolith organs alone are activated for

example by pure translation Alternatively the somatogravicfeedback effects can be substituted by a Bayesian prior atzero translational acceleration [4 17] This prior is also com-patible with the aviation illusion mentioned above Indeedwhile the correction due to the feedback or the prior isbeneficial under normal conditions it can result in the so-called somatogravic illusion [4 9] During translation the tilt


Somatogravicfeedback

IGIA-GAccelerationA

G

A

= G-GIA

GIA

Retinal flow

Canals

Gra

vity

Gravityestimator

+ Ω

k

ko

go

kFmiddotGIAxG

minus1Tvs

intdt

intGx Ω

(a) Complete vestibular information processing model

AccelerationA = G-GIA

GIA

Gra

vity

Zero translation prior

Zero velocity prior

+

minus

Ωintdt

intGx Ω

(b) Bayesian model

Figure 2 Model of visuovestibular processing proposed by Laurens and Angelaki [4] (a) Blue lines vestibular pathways grey lines visualpathways green lines inertial pathways (b) Schematicmodel of Bayesian inference for vestibular processing Black lines deterministicmodellightning bolts sources of noise question marks points of error accumulation blue lines influence of the zero velocity prior green linesinfluence of the zero translation prior grey lines incorporation of visual information (reproduced with permission from [4])

estimate increases over time as the estimated gravitymoves towards the gravitoinertial acceleration This causes adecrease in the translation estimate and in an aftereffect at theend of the translational acceleration

Neural correlates of the operations described above havebeen discovered in the monkey by Angelaki and colleagueswho found that the neural computation of translation (119891)occurs in the so-called Vestibular-only neurons of thevestibular nuclei in the rostral portion of cerebellar fastigiumand nodulus [18] Neurons in these regions combine tem-porally processed signals from the canals and otoliths aspredicted by the internalmodel hypothesis Recently neuronsextracting gravity have been discovered in the cerebellum [5]Laurens et al identified a group of Purkinje cells in the caudalcerebellar vermis with responses that reflect an estimate ofhead tilt (Figure 3) These tilt-selective cells are complemen-tary to the translation-selective Purkinje cells mentionedabove such that their population activities sum to the netgravitoinertial acceleration encoded by the otolith organs

3 Multisensory Integration

As we remarked in the previous section vestibular sensationsresult from composite signals because the otolith signalsare centrally combined with those of the semicircular canalsalready at the level of second-order sensory neurons in the

vestibular nuclei of the brainstem As far as gravity transduc-tion is concerned the vestibular signals are centrally com-bined with other sensory information such as proprioceptivesignals from muscle and tendon receptors visceral signals(from the kidneys vena cava etc) and visual signals Visionin particular provides both static and dynamic (eg opticflow) signals about the orientation of the body relative to thevertical Finally also the so-called efference copy ofmotor sig-nals (ie a copy of the motor commands sent by higher braincenters) and internal estimates of the body axis orientation[19] contribute to an estimate of body orientation All thesesignals are centrally combined yielding accuratemultisensoryestimates about gravity direction Indeed in darkness an erectperson makes errors lt2∘ when aligning an initially tiltedluminous bar with the expected direction of gravity [20]

Under normal light conditions of daily life there areseveral visual cues which point to the direction of gravity[21] Thus trees are rooted downwards and grow verticallyupwards and the walls of the houses are also vertical asare the chandeliers hanging from the ceiling The visualreference to gravity is so strong that there exist touristattraction places (so-called mystery spots) where someanomaly of the environment is exploited to provide theillusion that the gravity law is violated For instance in someplaces there is a strong slope of the terrain and trees growslanted Similar effects can be obtained with tilted walls in


Translation Roll

100

0

100

04 s

OVAR beginning OVAR steady-state

Tran

slatio

n ce

llTi

lt ce

llFi

ring

rate

(spi

kes

s)

180∘s

62∘s

04

G

GA GIA

(a) (b)

(c)(l) (m) (n) (o)

(h) (i) (j) (k)

(d) (e) (f) (g)

TAFB axis

Roll Pitch

LR axisYaw

OVAR tiltTranslation Roll tilt

Yaw signal

Roll signalGIA

(OTO)

Figure 3 (a) Equivalence principle the otolith organs are sensitive to the gravitoinertial acceleration (GIA) equal to the difference betweenthe gravity vector (GA) and the translational acceleration (TA) (b) Naming conventions of the headrsquos translation and rotation axes FBforward-backward LR leftward-rightward (c) Representation of the motion protocols used by Laurens et al [9] GIA along the LR axisrepresented by a swinging pendulum (bottom) is identical in the 3 protocols (translation tilt and off-vertical axis rotation [OVAR]) ((d)ndash(o))Responses from a translation-selective cell (red) and a tilt-selective cell (green) during left-right (LR) translation ((d) and (h)) roll tilt ((e)and (i)) and constant velocity OVAR ((f) (g) (j) and (k)) (l) (m) (n) and (o) show the corresponding yaw velocity (detected by horizontalcanals blue) roll velocity (detected by vertical canals cyan) and GIA along the LR axis (detected by otolith organs [OTO] black) Graycurves fit to the LR translation response (shown in (d) translation cell) or the roll tilt response (shown in (i) tilt cell) (reproduced withpermission from [5])

houses built for the purpose of creating such illusions Theslope angle distorts the perspective of the observer and mayeven create the perceptual illusion that a ball can roll upwardsby itself In Italy such illusions can be felt inside the leaninghouse designed by Vicino Orsini at Bomarzo (see httpenwikipediaorgwikiFileBomarzo parco mostri casa pen-dentejpg)

Except when some cue is so strong as to drive spaceperception by itself (a winner-take all situation) neuralestimates of gravity direction normally are computed by thecentral nervous system as a weighted average of multicueinformation including vestibular visual neck and truncalsignals plus a prior distribution about head and bodyorientation based on experience [20 22ndash24] In Bayesianterms the posterior estimate is obtained by combining noisysensory measurements with a prior each term being weighedinversely to its variance (noise [23])

4 Visual Perception ofGravitational Acceleration

So far we considered the problemofmonitoring the directionof gravity A different problem concerns monitoring its

magnitudeHowdowe estimate the gravitational accelerationof an object in a visual scene This situation occurs quitefrequently as when we experience the vision of objects infree-fall projectile or pendulum motion In addition toobject motion also self-motion may involve visual stimuli(optic flow) accelerated by gravity as when we fall or jumpfrom a height When confronted with gravity effects thevisual system faces a unique challenge In contrast withbody graviceptors (such as those of the vestibular systemmuscle and tendon proprioceptors and visceral organs) thevisual system does not deal with physical gravity directlybut only with the acceleration of the retinal image Whereasgravitational acceleration is constant at a given location thecorresponding retinal acceleration varies inversely with theviewing distance (distance between the observer and thescene) Therefore the visual estimate of gravity effects ona target motion requires accurate estimates of both imageacceleration and viewing distance Both types of estimatesare potentially problematic Indeed while the visual systemis very accurate in velocity estimates it is rather poor inacceleration estimates In fact the visual discrimination ofacceleration is about 5 times worse than that of velocity[25] Also viewing distance may be difficult to assess Eyevergence accommodation and stereo-disparity contribute


Figure 4 Test with curved tubes in preschoolers (reproduced withpermission of Prof Bruce Hood University of Bristol)

to estimating viewing distance of target motion in three-dimensional space but these cues are ineffective when thetarget is far (because of trigonometry) or when it moves on atwo-dimensional video display (as in a videogame) Pictorialinformationmdashsuch as that provided by the presence of objectsof known size (people trees houses etc) in the visual scenemdashalso aids recovering an environmental reference and scale andthus allows the calibration of the retinal image [26]

An internal model of gravity effects represents a criticalcomponent of the visual estimates in addition to raw sensorysignals Indeed gravity represents a special case of visualacceleration to which we are exposed since birth Thereforeit is very likely that it has been internalized in the brain Infact it has been shown that gravity effects on a visual objectare detected early in life [27] Between 5 and 7 months of ageinfants expect that an object moving down an inclined planeaccelerates and an upwardly moving object decelerates andare surprised to see the effects of an artificial reversed gravity(ie objects decelerating while moving downwards andaccelerating while moving upwards) Implicit expectationof gravity effects can generate striking judgment errors inpreschoolers Children around 2 years of age believe that adescending object always falls vertically downwards Thuswhen they are asked to find a ball that is dropped along acurved tube they search directly under the point of fall ratherthan at the exit of the tube [28] However if the ball motion isartificially reversed so that the ball seems to rise upwards allchildren solve the task perfectly (Figure 4) Notice that falsebeliefs about free-fall can still persist in adulthood (so-callednaıve physics) For instance several people without formalscientific background believe that heavier objects fall fasterthan lighter objects of the same size [6]

Strikingly however the motor system has an implicitknowledge of physics which ismuch better than that availableto the cognitive system [6] Daily life offers several examplesof behavior demonstrating the implicit knowledge of physicsand the anticipation of the effects of gravitational and cen-trifugal forces for instance when we try to keep equilibriumwhile riding a bike Also the automatic motor responses

Elbow angle

Wrist angle

Biceps EMG

02 kg

06 kg

130ms

Figure 5 Catching balls of different weight In different trials asubject caught a 02 kg ball (upper panel) and a 06 kg ball (lowerpanel) dropped from a 12m height In each panel traces from topto bottom correspond to elbowflexion angle wrist flexion angle andrectified electrical activity (EMG) of biceps muscle The right-mostvertical line denotes the time of impact of the ball on the hand Theleft-most vertical line denotes the time of onset of the anticipatoryEMG activity (reproduced with permission from [6])

evoked by seeing a falling object are programmed by the brainby taking into account the law of free-fall first formulatedby Galileo Galilei In a laboratory experiment (Figure 5)subjects were asked to catch with the hand a ball that wasdropped vertically from 12m height relative to the hand [6]The ball could weigh 200 g or 600 g in different trials andfell in about 05 s Subjects prestiffened their arm muscles toabsorb the impact at about 130ms irrespective of the specificmass of the ball Instead the amplitude of muscle activationscaled in proportion to the ball mass because a stronger forceis required to counteract a stronger ball momentum [29]It has also been shown that the time of muscle contractionalways leads the impact time by the same amount irrespectiveof the height of fall (Figure 6) [29] Therefore the motorsystem is accurately tuned to the effects of Earth gravity

This tuning persists at the beginning of orbital flightdespite the sensory and cognitive evidence of weightlessnessand despite the motor responses being inappropriate tothe new conditions [30] These results are compatible witha Bayesian interpretation of the estimate of gravitationalacceleration if one assumes that the variance in the prior of1 g acceleration is very small compared with the variance inthe sensory likelihood If so the 1 g prior would bias stronglythe estimate until it is updated with prolonged exposure toweightlessness [31]

Gravity effects are taken into account not only by themotor system but also perceptually as when people judge theduration of motion of a falling target [32 33] or the periodof oscillation of a pendulum [34] Thus in experiments inwhich a pendulum oscillates faster or slower than normalthe observers rate the oscillations violating the physicallength-period relation as less natural than the oscillationscomplying with physics [34] The implicit bias toward grav-itational motion when viewing an oscillating pendulum is


Time-course of anticipatory biceps activity

Fall duration (s) Height (m)

16

12

08

06

04

057

049

040

029

04 02 00

Time-to-contact (s)

Figure 6 Time course of the EMG anticipatory responses ofbiceps Traces correspond to the results obtained for catches of ballsdropped from the heights indicated on the right (fall durations areindicated on the left) EMG traces have been scaled in amplitudeto their maximum and aligned relative to collision time Time axisindicates the time remaining prior to collision (reproduced withpermission from [6])

also revealed by the observation that harmonic motion isperceived as uniform [35] Also the perceptual judgment ofpassive egomotion along the vertical directionmdashsimulated bymeans of immersive visual stimulimdashis based on the internalmodel of gravity [36]

Just as in the case of the estimates of the directionof gravity also those of visual gravitational accelerationgenerally depend on a combination of multiple cues andsuch combination may obey Bayesrsquo rules The internal modelprovides the prior while various sensory cues provide thelikelihood of the estimate with a reliability that depends onthe context In one study observers were asked to judgethe duration of motion of a target accelerating in one offour different directions downwards upwards leftwards andrightwards relative to a visual scene [33] Downward motioncomplied with the gravity constraint whereas motion in theother directions violated this constraint Observers watchedeither a pictorial or an empty scene while being upright or

tilted by 45∘ relative to the monitor and Earthrsquos gravity Inanother condition observers were upright and the scene wastilted by 45∘ Discrimination precisionwas significantly betterfor downwardmotion than for the other directions Howeverthe difference in precision was not constant across conditionsbut was highest when both the observer and the pictorialscenewere upright and lowestwhen the target direction in theempty scene was tilted by 45∘ relative to an upright observerThus the behaviour observed in the study was consistentwith the combination of pictorial cues orientation of theobserver relative to the physical vertical and orientation oftarget motion relative to the physical vertical

The interaction of the visual signals with vestibularsignals about subject orientation relative to physical gravitywas shown in a study performed during a parabolic flightcampaign [37] During each parabola a 20 s weightless (0 g)phase was preceded and followed by 20 s of hypergravity(15ndash18 g) Strikingly the timing of interception of a visualtarget moving along the visual vertical reversed sign duringthe weightless phases compared with the responses at normalgravity [37] This reversal depends on the reversal of theotolith responses during the transition from hypergravity tohypogravity which was sensed as a negative gravity that isas a gravitational pull in the upward direction (comparableto when we are suspended upside-down)

5 Neural Substrates of the Internal Model ofGravity Effects on Visual Motion

The hypothesis that the effects of gravity on a target motionare taken into account by combining multisensory informa-tion including visual and vestibular cues is supported byhuman neuroimaging studies In a series of fMRI studies[7 38ndash40] visual gravitational acceleration (involving eitherobject motion or simulated egomotion) engaged a network ofbrain regions located within and around the Sylvian fissureclose to the temporoparietal junction (TPJ) posterior insularcortex retroinsula parietal operculum supramarginal gyrustemporal operculum and superior and middle temporalgyri In addition gravitational motion engaged primarysomatosensory and motor cortex ventral premotor cortexSMA cingulate cortex visual cortex including the lingualgyrus and several subcortical structures posterior thalamusputamen cerebellum and vestibular nuclei A causal linkbetween TPJ activity and the processing of visual gravita-tional motion has been demonstrated by transiently disrupt-ing the activity of TPJ by means of transcranial magneticstimulation (TMS) [41]

As reviewed earlier electrophysiological studies in themonkey showed that a population of Purkinje cells in thecaudal cerebellar vermis encodes head tilt thus reflectingan estimate of gravity direction based on vestibular infor-mation [5] Interestingly human posterior cerebellar vermis(a homologue region of that studied by Laurens et al [5]in monkeys) and vestibular nuclei appear to be involved incombining pictorial information with the internal model ofgravity to extract gravitational motion from visual scenes(Figure 7) [7]


(a) (b)

(c)

6

4

2

0

minus2

minus4a b c

T-v

alue

PictorialNonpictorial

(gminusrg)

(d)

Figure 7 Functionalmagnetic resonance imaging of the cerebellum and brainstem in a task of interception of a ballmoving along the verticalBrain areas showing preferential activation for natural gravity motion specifically for the pictorial visual context (a) (b) (c) Activations inthe midline cerebellum ((a) axial section (c) medial sagittal section) and vestibular nuclei ((b) axial section) Roman numerals in (c) denoteLarsell lobules White circles are centered on maximal statistical activation peaks (a) Lobules IXX (b) Left vestibular nuclei (c) LobulesVIIVIII (d) Bar-graphs of the difference (plusmn between-subjects sem) t-values for natural gravity (119892) and artificial reversed gravity (119903119892) trialsin pictorial (white) and nonpictorial (black) context for the activity peaks circled in (a) (b) and (c) (reproduced with permission from [7])

In sum the neuroimaging studies reviewed above indi-cate that the effects of gravity on visual motion are encodedin a highly distributed cortical-subcortical network Sev-eral regions of this network colocalize with the regionsindependently activated by vestibular caloric stimuli [38]These regions then presumably belong to the multimodalvestibular network which also responds to visual and neckproprioceptive stimuli [42 43] Lesions of vestibular cortexcan lead to a tilt of the perceived visual vertical and rotationalvertigounsteadiness [44] while focal electrical stimulationor epileptic discharges can elicit sensations of self-motion oraltered gravity [45 46]

6 Conclusions

We argued that an apparently simple problem such as that ofmonitoring gravity effects on our body and on the external

environment is in fact computationally very demandingeven for a high-powered brain such as that of primatesMeasurements derived from individual sensory organs areoften ambiguous (due to the intrinsic constraints of physicallaws) and noisy (due to biological limitations) Howeverthe combination of multisensory signals (visual vestibularproprioceptive and visceral) and the reliance on internalmodels of physics yield estimates which are very accurateunder normal conditions but which can fail badly underanomalous conditions (such as the early phases of spaceflight) Central processing of multisensory information andinternal models occurs in a widely distributed network ofcortical and subcortical regions The extensive integrationof sensory and motor information in this network makesgravity-related information available to many vital functionsof the organism



The research reviewed in this paper was conducted in theabsence of any commercial or financial relationships thatcould be construed as a potential conflict of interests

Acknowledgments

The authorsrsquo work was supported by the Italian Ministryof University and Research (PRIN grant) and Italian SpaceAgency (CRUSOE COREA SLINK andARIANNA grants)

References

[1] K W Beisel Y Wang-Lundberg A Maklad and B FritzschldquoDevelopment and evolution of the vestibular sensory appa-ratus of the mammalian earrdquo Journal of Vestibular ResearchEquilibrium and Orientation vol 15 no 5-6 pp 225ndash241 2005

[2] M E Goldberg M F Walker and A J Hudspeth ldquoThevestibular systemrdquo inPrinciples ofNeural Science E KKandel JH Schwartz T M Jessell S A Siegelbaum and A J HudspethEds pp 917ndash934 McGraw-Hill New York NY USA 2013

[3] D E Angelaki and K E Cullen ldquoVestibular system the manyfacets of a multimodal senserdquo Annual Review of Neurosciencevol 31 pp 125ndash150 2008

[4] J Laurens and D E Angelaki ldquoThe functional significance ofvelocity storage and its dependence on gravityrdquo ExperimentalBrain Research vol 210 no 3-4 pp 407ndash422 2011

[5] J Laurens HMeng andD E Angelaki ldquoNeural representationof orientation relative to gravity in the macaque cerebellumrdquoNeuron vol 80 no 6 pp 1508ndash1518 2013

[6] M Zago and F Lacquaniti ldquoCognitive perceptual and action-oriented representations of falling objectsrdquo Neuropsychologiavol 43 no 2 pp 178ndash188 2005

[7] W L Miller V Maffei G Bosco et al ldquoVestibular nuclei andcerebellum put visual gravitational motion in contextrdquo Journalof Neurophysiology vol 99 no 4 pp 1969ndash1982 2008

[8] C Fernandez and J M Goldberg ldquoPhysiology of peripheralneurons innervating otolith organs of the squirrel monkey IResponse to static tilts and to long duration centrifugal forcerdquoJournal of Neurophysiology vol 39 no 5 pp 970ndash984 1976

[9] R Gibb B Ercoline and L Scharff ldquoSpatial disorientationdecades of pilot fatalitiesrdquo Aviation Space and EnvironmentalMedicine vol 82 no 7 pp 717ndash724 2011

[10] R Mayne ldquoA systems concept of the vestibular organsrdquo inVestibular System Psychophysics Applied Aspects and GeneralInterpretations H Kornhuber Ed vol 6 part 2 pp 493ndash580Springer Berlin Germany 1974

[11] S Glasauer ldquoInteraction of semicircular canals and otoliths inthe processing structure of the subjective zenithrdquo Annals of theNew York Academy of Sciences vol 656 pp 847ndash849 1992

[12] D E Angelaki M Q McHenry J D Dickman S D Newlandsand B J M Hess ldquoComputation of inertial motion neuralstrategies to resolve ambiguous otolith informationrdquo Journal ofNeuroscience vol 19 no 1 pp 316ndash327 1999

[13] D M Merfeld L Zupan and R J Peterka ldquoHumans useinternal models to estimate gravity and linear accelerationrdquoNature vol 398 no 6728 pp 615ndash618 1999

[14] J M Goldberg and C Fernandez ldquoVestibular mechanismsrdquoAnnual Review of Physiology vol 37 pp 129ndash162 1975

[15] D M Merfeld and L H Zupan ldquoNeural processing of grav-itoinertial cues in humans III Modeling tilt and translationresponsesrdquo Journal of Neurophysiology vol 87 no 2 pp 819ndash833 2002

[16] AM Green andD E Angelaki ldquoAn integrative neural networkfor detecting inertial motion and head orientationrdquo Journal ofNeurophysiology vol 92 no 2 pp 905ndash925 2004

[17] J Laurens and J Droulez ldquoBayesian processing of vestibularinformationrdquo Biological Cybernetics vol 96 no 4 pp 389ndash4042007

[18] D E Angelaki A G Shaikh A M Green and J D DickmanldquoNeurons compute internal models of the physical laws ofmotionrdquo Nature vol 430 no 6999 pp 560ndash564 2004

[19] HMittelstaedt ldquoA new solution to the problem of the subjectiveverticalrdquo Naturwissenschaften vol 70 no 6 pp 272ndash281 1983

[20] R A A Vingerhoets M de Vrijer J A M van Gisbergen andW PMedendorp ldquoFusion of visual and vestibular tilt cues in theperception of visual verticalrdquo Journal of Neurophysiology vol101 no 3 pp 1321ndash1333 2009

[21] I P Howard Human Visual Orientation John Wiley amp SonsNew York NY USA 1982

[22] L H Zupan D M Merfeld and C Darlot ldquoUsing sensoryweighting to model the influence of canal otolith and visualcues on spatial orientation and eye movementsrdquo BiologicalCybernetics vol 86 no 3 pp 209ndash230 2002

[23] P R MacNeilage M S Banks D R Berger and H H BulthoffldquoABayesianmodel of the disambiguation of gravitoinertial forceby visual cuesrdquo Experimental Brain Research vol 179 no 2 pp263ndash290 2007

[24] D E Angelaki E M Klier and L H Snyder ldquoA vestibular sen-sation probabilistic approaches to spatial perceptionrdquo Neuronvol 64 no 4 pp 448ndash461 2009

[25] P Werkhoven H P Snippe and A Toet ldquoVisual processing ofoptic accelerationrdquo Vision Research vol 32 no 12 pp 2313ndash2329 1992

[26] S E Palmer Vision Science Photons to Phenomenology MITPress Cambridge UK 1999

[27] I K Kim and E S Spelke ldquoInfantsrsquo sensitivity to effectsof gravity on visible object motionrdquo Journal of ExperimentalPsychology Human Perception and Performance vol 18 no 2pp 385ndash393 1992

[28] B M Hood ldquoGravity does rule for falling eventsrdquo Developmen-tal Science vol 1 no 1 pp 59ndash63 1998

[29] F Lacquaniti and C Maioli ldquoThe role of preparation in tuninganticipatory and reflex responses during catchingrdquoThe Journalof Neuroscience vol 9 no 1 pp 134ndash148 1989



[32] M A Grealy C M Craig C Bourdin and S G ColemanldquoJudging time intervals using a model of perceptuo-motorcontrolrdquo Journal of Cognitive Neuroscience vol 16 no 7 pp1185ndash1195 2004

[33] A Moscatelli and F Lacquaniti ldquoThe weight of time gravita-tional force enhances discrimination of visualmotion durationrdquoJournal of Vision vol 11 no 4 pp 1ndash17 2011

[34] J B Pittenger ldquoDetection of violations of the law of pendulummotion observersrsquo sensitivity to the relation between periodand lengthrdquo Ecological Psychology vol 2 no 1 pp 55ndash81 1990


[35] B La Scaleia M Zago A Moscatelli F Lacquaniti and PViviani ldquoImplied dynamics biases the visual perception ofvelocityrdquo PloS ONE vol 9 no 3 Article ID e93020 2014

[36] I IndovinaVMaffei andF Lacquaniti ldquoAnticipating the effectsof visual gravity during simulated self-motion estimates oftime-to-passage along vertical and horizontal pathsrdquo Experi-mental Brain Research vol 229 no 4 pp 579ndash586 2013

[37] P Senot M Zago A le Seacrsquoh et al ldquoWhen up is down in 0ghow gravity sensing affects the timing of interceptive actionsrdquoJournal of Neuroscience vol 32 no 6 pp 1969ndash1973 2012

[38] I Indovina V Maffei G Bosco M Zago E Macaluso and FLacquaniti ldquoRepresentation of visual gravitational motion inthe human vestibular cortexrdquo Science vol 308 no 5720 pp416ndash419 2005

[39] VMaffei EMacaluso I Indovina G Orban and F LacquanitildquoProcessing of targets in smooth or apparent motion alongthe vertical in the human brain an fMRI studyrdquo Journal ofNeurophysiology vol 103 no 1 pp 360ndash370 2010

[40] I Indovina V Maffei K Pauwels E Macaluso G A Orbanand F Lacquaniti ldquoSimulated self-motion in a visual gravityfield sensitivity to vertical and horizontal heading in the humanbrainrdquo NeuroImage vol 71 pp 114ndash124 2013

[41] G Bosco M Carrozzo and F Lacquaniti ldquoContributions of thehuman temporoparietal junction and MTV5+ to the timingof interception revealed by transcranial magnetic stimulationrdquoThe Journal of Neuroscience vol 28 no 46 pp 12071ndash120842008

[42] S Bense T Stephan T A Yousry T Brandt and M DieterichldquoMultisensory cortical signal increases and decreases duringvestibular galvanic stimulation (fMRI)rdquo Journal of Neurophys-iology vol 85 no 2 pp 886ndash899 2001

[43] G Bottini H O Karnath G Vallar et al ldquoCerebral representa-tions for egocentric space functional-anatomical evidence fromcaloric vestibular stimulation and neck vibrationrdquo Brain vol124 no 6 pp 1182ndash1196 2001

[44] T Brandt andMDieterich ldquoThe vestibular cortex its locationsfunctions and disordersrdquo Annals of the New York Academy ofSciences vol 871 pp 293ndash312 1999

[45] O Blanke S Ortigue T Landis and M Seeck ldquoStimulatingillusory own-body perceptionsrdquo Nature vol 419 no 6904 pp269ndash270 2002

[46] D K Nguyen D B Nguyen R Malak et al ldquoRevisiting the roleof the insula in refractory partial epilepsyrdquo Epilepsia vol 50 no3 pp 510ndash520 2009

Research ArticleIntegration Analysis of MicroRNA andmRNA Expression Profiles in Human Peripheral BloodLymphocytes Cultured in Modeled Microgravity

C Girardi1 C De Pittagrave1 S Casara1 E Calura1 C Romualdi1 L Celotti12 and M Mognato1

1 Dipartimento di Biologia Universita degli Studi di Padova Via U Bassi 58B 35131 Padova Italy2 Laboratori Nazionali di Legnaro INFN Viale dellrsquoUniversita 2 Legnaro 35020 Padova Italy

Correspondence should be addressed to L Celotti luciacelottiunipdit and M Mognato maddalenamognatounipdit

Received 16 April 2014 Revised 22 May 2014 Accepted 22 May 2014 Published 23 June 2014


Copyright copy 2014 C Girardi et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

We analyzed miRNA and mRNA expression profiles in human peripheral blood lymphocytes (PBLs) incubated in microgravitycondition simulated by a ground-based rotating wall vessel (RWV) bioreactor Our results show that 42miRNAswere differentiallyexpressed in MMG-incubated PBLs compared with 1 g incubated ones Among these miR-9-5p miR-9-3p miR-155-5p miR-150-3p and miR-378-3p were the most dysregulated To improve the detection of functional miRNA-mRNA pairs we performedgene expression profiles on the same samples assayed for miRNA profiling and we integrated miRNA and mRNA expressiondata The functional classification of miRNA-correlated genes evidenced significant enrichment in the biological processes ofimmuneinflammatory response signal transduction regulation of response to stress regulation of programmed cell death andregulation of cell proliferation We identified the correlation of miR-9-3p miR-155-5p miR-150-3p and miR-378-3p expressionwith that of genes involved in immuneinflammatory response (eg IFNG and IL17F) apoptosis (eg PDCD4 and PTEN) andcell proliferation (eg NKX3-1 and GADD45A) Experimental assays of cell viability and apoptosis induction validated the resultsobtained by bioinformatics analyses demonstrating that in human PBLs the exposure to reduced gravitational force increases thefrequency of apoptosis and decreases cell proliferation

1 Introduction

Exposure to spaceflight environment is known to cause inhumansmany adverse physiological changes including skele-tal muscle atrophy [1ndash3] cardiovascular and microvasculardisorders [4ndash6] bone deterioration [7 8] and impairedimmune system function [9 10] Immune systemdysfunctiondue to exposure to microgravity has been documented aswell in terms of reduced activationproliferation alteredcytokine production and altered signal transduction [11 12]Alterations in global gene expression patterns have beenalso observed in space-flown human cells involving mainlygenes of immune system activation [13 14] cytoskeleton [15]and cell cycle [16 17] However due to the difficulty andlimitations of performing experiments in the real micro-gravity of space many investigations have been conductedunder simulated microgravity conditions in which cells are

cultured in ground-based machines such as clinostats androtating wall vessel bioreactors that generate a residual 10minus3ndash10minus6 g force that approximates microgravity [5 18ndash24] Theresults indicate that similar to space microgravity simulatedmicrogravity affects both cell structure and function as wellas gene expression inmammalian cells [14 19 25] in bacteria[26 27] or in other living organisms [28ndash30]

Since molecular changes at the gene level may compro-mise cell function it is important to understand the cellularresponse to reduced gravity at the molecular level For thispurpose a class of noncoding RNAs called microRNAs(miRNAs) plays a key role miRNAs are a large family ofsmall RNAs of 18ndash24 nucleotides that are involved in post-transcriptional regulation of gene expression by interactingwith 31015840-untranslated regions (UTR) of target genes The reg-ulatory process is complex and occurs posttranscriptionallythrough miRNA interaction with a target site in the mRNA



that has partial or complete complementarity to the miRNAThe binding of miRNAs to complementary sequence of theirtarget mRNAs may repress translation or induce degradationofmRNAs [31] Recently the destabilization of targetmRNAsinstead of translational repression has been shown to be thepredominant mechanism for reduced protein output [32]Less often dsRNA formed by miRNA target complexes cantarget gene promoters and actually enhance transcription oftarget genes sometimes termed RNAa (RNA activation) [33]A single miRNA may have broad effects on gene expressionnetworks such as regulating cell lineage specificity cellularfunctions or stress response Besides a physiological roleof miRNAs in a variety of important biological processesincluding differentiation apoptosis [34] and fat metabolism[35] the miRNA-mediated gene regulation operates alsoduring viral infection [36] stress response pathway [37] andpathological processes such as tumorigenesis [38ndash42]

The present study is addressed to identify alterationsin miRNA profiles of human peripheral blood lympho-cytes (PBLs) incubated in modeled microgravity (MMG)with respect to those incubated in gravity 1 g To simulatemicrogravity we used a specialized bioreactor developed atthe NASA-Johnson Space Center (Houston TX USA) therotating wall vessel which represents a valid ground modelto simulate as far as possible a condition of reduced gravityTo identify miRNA-correlated genes whose expression levelwas significantly altered as a function ofMMGwe performedgene expression profiling on the same PBL samples assayedfor miRNA profiling and we integrated microRNAome andtranscriptome by using MAGIA2 [43] a web tool for theintegrative analysis of miRNA and genes expression dataincorporating transcriptional regulation A group ofmiRNA-mRNA pairs related to immunity cell proliferation andapoptosis was identified in PBLs incubated in MMG Thedifferences between MMG and 1 g on correlated miRNA-mRNA pairs involved in cell proliferation and apoptosiswere investigated by in vitro assays of clonogenic ability andapoptosis induction in PBLs incubated in MMGwith respectto those incubated in 1 g conditions


21 Lymphocytes Isolation and Microgravity SimulationHuman peripheral blood lymphocytes (PBLs) were obtainedfrom freshly collected ldquobuffy coatsrdquo from blood samples oftwelve healthy anonymous donors at the Blood Centre ofthe City Hospital of Padova (Italy) This study obtainedthe Ethics Approval from the Transfusion Medicine (TM)Ethics Committee of Blood Centre of the City Hospital ofPadova PBLs were isolated by separation on Biocoll densitygradient (BIOCHROM Berlin Germany) After isolationPBLs were preincubated overnight at a concentration of 3times 106mL in basal medium RPMI 1640 containing Gluta-MAX I (Invitrogen Life Technologies Carlsbad CA USA)124UmL penicillin 63120583gmL streptomycin sulfate and 10fetal bovine serum (FBS BIOCHROM Berlin Germany)After the overnight incubation PBLs consisting of peripheralmononuclear cells depleted of monocytes were suspended

at 1 times 106mL in basal medium and subjected to modeledmicrogravity simulated by the rotating wall vessel (RWV)bioreactor (Synthecon Cellon) placed inside a humidifiedincubator vertically rotating at 23 rpm [44] In the rotatingsystem the gravity is balanced by equal and oppositemechan-ical forces (centrifugal Coriolis and shear components)and the gravitational vector is reduced to about 10minus3 g Inthese conditions single cells are nearly always in suspensionrotating in quasi-stationary manner with the fluid in a low-shear culture environment [19 45] Ground-based (1 g) PBLswere kept at the same cell density in 75 cm2 flasks (FALCON)in the same medium After 24 48 or 72 h of incubationin MMG and 1 g PBLs were activated to enter cell cycle tomeasure cell proliferation by incubation in culture medium(CM) containing phytohaemagglutinin (PHA BIOCHROMBerlin Germany) and interleukin 2 (IL2 Chiron Siena Italy)as stimulating factors [46]

22 Total RNA Isolation Total RNA was isolated from107 PBLs at the end of 24 h incubation in MMG and 1 g byusing Trizol Reagent (Invitrogen Life Technologies CarlsbadCA USA) according to the manufacturerrsquos protocol TotalRNA was quantified using the ND-1000 spectrophotometer(Nanodrop Wilmington DE USA) and RNA integrity andthe content of miRNAs were assessed by capillary elec-trophoresis using the Agilent Bioanalyzer 2100 as previouslydescribed [47] Only total RNA samples with RNA integritynumber (RIN) values ge6 and with miRNA lt20 were usedfor microarray analysis

23 miRNA and Gene Expression Profiling MicroRNAs pro-filing was carried out in PBL samples incubated in MMGversus 1 g Analyses were performed by using the ldquoHumanmiRNA Microarray kit (V2)rdquo (Agilent Technologies) thatallows the detection of 723 known human (miRBase v101)and 76 human viral miRNAs Total RNA (200 ng) waslabeled with pCp Cy3 according to the Agilent protocoland unincorporated dyes were removed with MicroBioSpin6columns (BioRad) [48] Probes hybridization and slideswashing were performed as previously reported [47] AgilentFeature Extraction software version 10511 was used forimage analysis

Gene expression profiling was carried out in MMG-incubated PBLs versus 1 g incubated PBLs on total RNAextracted from the same PBL samples assayed for miRNAprofiling We used the ldquoWhole Human Genome Oligo Mi-croarrayrdquo (Agilent) consisting of sim41000 (60-mer) oligonu-cleotide probes which span conserved exons across thetranscripts of the targeted full-length genes 800 ng of totalRNAwas labeled with ldquoAgilent One-Color Microarray-BasedGene Expression protocolrdquo according to the manufacturerrsquosinstructions The method uses T7 RNA polymerase whichsimultaneously amplifies target material and incorporatescyanine 3-labeled CTP The Cy3-labeled cRNAs were puri-fied using Qiagenrsquos RNeasy mini spin columns (Qiagen)and quantified using the ND-1000 spectrophotometer (Nan-odrop Wilmington DE USA) Probes hybridization andslides washing were performed as previously reported [47]


Slides were scanned on an Agilent Microarray Scanner Sys-tem (model G2565CA) and Agilent Feature Extraction soft-ware version 10511 was used for image analysis Raw data areavailable on the Gene Expression Omnibus (GEO) website(httpwwwncbinlmnihgovgeo) using SuperSeries acces-sion number GSE57418 that groups microRNA (GSE57400)and mRNA expression profiles (GSE57408)

24 Statistical Analysis of miRNA and Gene Expression DataInterarray normalization of expression levels was performedwith cyclic Lowess formiRNA experiments and with quantilefor gene expression profiling [49] to correct possible exper-imental distortions Normalization function was applied toexpression data of all experiments and then values of spotreplicates within arrays were averagedThemodalities of spotquality measures and hybridization are reported previously[47] The identification of differentially expressed genes andmiRNAs was performedwith one- and two-class SignificanceAnalysis of Microarray (SAM) program [50] with defaultsettingsThe expression level of eachmiRNA andmRNAwascalculated as the log2 (MMG1 g) PBLs of the same donorPathway analysis on differentially expressed genes has beenperformed using Graphite web [51] hypergeometric test onReactome Pathways considering significant those categorieswith FDR lt 01

25 Identification of miRNA Target Genes and CorrelationAnalysis of miRNA and mRNA Expression Data To predictmiRNA targets we performed a computational analysis inte-grating mRNA and miRNA expression measurements fromthe same donor usingMAGIA2 web tool [43] freely availableat httpgencompbiounipditmagia2We used httpwwwmicrornaorg predictions and Pearson correlation (119903 gt 04)to estimate the degree of correlation between any putativepairs of miRNA and mRNA [52 53] To identify the bio-logical processes most involved in target prediction we haveperformed an enrichment analysis on Gene Ontology (GO)usingDAVIDweb tool v67 [54] considering significant thosecategories with FDR lt 02

Intraclass analyses have been performed consideringMMGand 1 g samples separately In order to have comparableresults the intraclass analysis has been performed usingMAGIA2 software with the same parameters predictors andcutoff described for the previous analysis Specific inter-actions for MMG and 1 g networks have been identifiedand a GO enrichment analysis (FDR lt 02) was performedseparately on the nodes belonging to the specific MMG and1 g networks using DAVID web tool

26 Quantitative Real-Time PCR (qRT-PCR) Assay In orderto verify the expression data generated by miRNA andmRNA microarrays we performed qRT-PCR experimentsfor miRNAs and genes which showed significant expressionchanges in MMG The following miRNAs were subjectedto the RT-qPCR validation miR-9-5p miR-378a miR-155-5p and miR-150-3p Reverse transcription of 10 ng of totalRNA with primers corresponding to each miRNA and toU48 small nuclear RNA (RNU48) as endogenous control

was performed as directed by the protocol of the two-stepTaqMan MicroRNA Assay kit (Applied Biosystems FosterCity CA USA) that incorporates a target-specific stem-loopreverse transcription primer to provide specificity for themature miRNA target For the PCR reaction 1 120583L of the RTreaction was combined with 05 120583L of TaqMan MicroRNAAssay 20x and 5 120583L of TaqMan Universal PCR Master Mixin a 10 120583L final volume The reactions were incubated ina Mastercycler EP gradient 119878 (Eppendorf) in 02mLPCRtubes for 30min at 16∘C and 30min at 42∘C followed by5min at 85∘C and then held at 4∘CThe resulting cDNA wasquantitatively amplified in 40 cycles on an ABI 7500 Real-Time PCR System using TaqMan Universal PCRMaster Mixand TaqMan MicroRNA Assays

For mRNA detection 1 120583g of total RNA was retro-transcribed with ImProm-II Reverse Transcription System(Promega) qRT-PCR was performed with the GoTaq qPCRMaster Mix (Promega) and gene-specific primers for IFNGIL17F TLR4 HLA-DRB1 and BCL6 genes and for GAPDHas reference qRT-PCR reactions were performed in quad-ruplicates in PBL samples from 6 to 8 donors Real-timePCR was performed using an Applied Biosystems 7500 FastReal-Time PCR System with cycling conditions of 95∘C for10min followed by 95∘C for 15 sec and 60∘C for 60 sec 45cycles in total The relative expression levels of miRNAsand mRNAs between samples were calculated using thecomparative delta CT (threshold cycle number) method(2minusΔΔCT) implemented in the 7500 Real-Time PCR Systemsoftware [55] Primersrsquo pairs used are as follows GAPDH(glyceraldehyde-3-phosphate dehydrogenase) fw 51015840-TCC-TCTGACTTCAACAGCGA-31015840 rev 51015840-GGGTCTTACTCC-TTGGAGGC-31015840 IFNG (interferon gamma) fw 51015840- GGC-ATTTTGAAGAATTGGAAAG-31015840 rev 51015840-TTTGGATGC-TCTGGTCATCTT-31015840 IL17F (interleukin 17) fw 51015840-GGC-ATCATCAATGAAAACCA-31015840 rev 51015840- TGGGGTCCCAAG-TGACAG-31015840 TLR4 (Toll-like receptor 4)fw 51015840-CCTGCG-TGAGACCAGAAAG-31015840 rev 51015840-TTCAGCTCCATGCAT-TGATAA-31015840 HLA-DRB1 (major histocompatibility com-plex class II DR beta 1) fw 51015840-ACAACTACGGGGTTG-TGGAG-31015840 rev 51015840-GCTGCCTGGATAGAAACCAC-31015840BCL6 (B-cell CLLlymphoma 6) fw 51015840-CGAATCCACACA-GGAGAGAAA-31015840 rev 51015840-ACGCGGTATTGCACCTTG-31015840

27 Cell Proliferation and Apoptosis Induction Cell viabilitywas determined by the T-cell cloning assay [44 46] at theend of 24 h 48 h and 72 h incubation in MMG and 1 gBriefly four 96-well U-bottomed microtiter plates with twoviable lymphocyteswell were seeded in medium CM in thepresence of 1 times 104 feeder cellswell (TK6 lymphoblastoidcells lethally irradiatedwith 40Gy of 120574-rays) Twoweeks laterthe plates were scored for growing colonies to calculate thecloning efficiency (CE) from the proportion of negative wellsassuming a Poisson distribution (CE = minusIn P

0N where P

0is

the fraction of wells without cells andN is the number of cellsseeded into wells) [56]

Apoptotic index was determined in PBLs incubated for24 and 48 h in MMG and in parallel in 1 g For detectionof apoptotic morphology PBLs were fixed and stained with




hsa-let-7e-5phsa-miR-376c-3phsa-miR-99b-5p

hsa-miR-29b-1-5phsa-miR-132-3p

hsa-let-7i-3phsa-miR-376a-3phsa-miR-193b-3p

hsa-let-7i-5phsa-miR-146b-5p

hsa-miR-221-5phsa-miR-505-5phsa-miR-10a-5p


hsa-miR-200a-3phsa-miR-342-5phsa-miR-192-5phsa-miR-7-1-3phsa-miR-185-5p





hsa-miR-663ahsa-miR-940


hsa-miR-34b-5phsa-miR-150-3p

hsa-miR-378a-3p

Expression valueminus4 minus2 0 2 4 6

(a)

hsa-miR-376a-3phsa-miR-376c-3phsa-miR-192-5phsa-miR-7-1-3phsa-miR-625-5phsa-miR-342-5phsa-miR-10a-5phsa-miR-505-5phsa-miR-221-5phsa-miR-7i-5phsa-miR-7-5phsa-miR-629-3p5phsa-miR-146b-5phsa-miR-200a-3phsa-miR-145-5phsa-miR-193b-3phsa-miR-7i-5phsa-miR-29b-1-5phsa-miR-1323phsa-miR-99b-5phsa-miR-125a-5phsa-miR-7e-5phsa-miR-155-5phsa-miR-9-3phsa-miR-9-5phsa-miR-150-3phsa-miR-378a-3p5phsa-miR-378a-5phsa-miR-181a-3phsa-miR-34a-5phsa-miR-34b-5phsa-miR-362-5phsa-miR-532-5phsa-miR-423-3phsa-miR-103a-3phsa-miR-107hsa-miR-185-5phsa-miR-223-3phsa-miR-940

G D E F I C L A P B H Mminus371 00 307

(b)

Figure 1 Differentially expressed miRNAs in human PBLs incubated in MMG (a) The expression level of each miRNA indicated as foldchange is the mean of the expression values obtained from the transformed log2 ratio (MMG1 g) (b) Dendrogram of miRNAs differentiallyexpressed in MMG The range of expression value is from minus37 (green downregulation) to 307 (red upregulation) Grey boxes correspondto not available (NA) fluorescent signal from the microarray platform

2 120583gmL 46-diamino-2-phenylindol (DAPI Roche) in anantifade solution (Vectashield Vector Lab) as previouslydescribed [57] At least 2000 cells were scored for each time-point by fluorescence microscopy (1000x magnification)The activation of caspase-3 was measured by the caspase-3fluorescent assay kit (Clontech BD Biosciences) at the endof incubation in 1 g or MMG as previously described [58]The fluorescent emission at 505 nm (excitation at 400 nm)of cleaved 7-amino-4-trifluoromethyl coumarin (AFC) wasmeasured with a PerkinElmer LS-50 B spectrofluorimeter

3 Results

31 Identification of miRNAs Affected by MicrogravitymiRNA expression profiling was performed on total RNAextracted from PBLs of twelve healthy donors at the end of

24 h incubation time in MMG and in 1 g conditions By com-paring the expression profile of MMG-incubated versus 1 gincubated PBLs of the same donor we found 42 differentiallyexpressed miRNAs 25 upregulated and 17 downregulatedfor which raw data and means of miRNA expression valuesare available at Supplementary Table S1 (see SupplementaryTable S1 in the Supplementary Material available online athttpdxdoiorg1011552014296747) miR-9-5p miR-9-3pand miR-155-5p were the most upregulated (46- 35- and24-fold resp) whereas miR-378a and miR-150-3p were themost downregulated (sim2-fold) (Figure 1)

32 Effect of Microgravity on Gene Expression Profile Geneexpression analysis was performed in PBLs incubated for 24 hin MMG and in 1 g By comparing the expression profilesof MMG-incubated and 1 g incubated PBLs we identified1581 differentially expressed genes inMMG versus 1 g among


0

200

400

600

800

1000

Num

ber o

f diff

eren

tially

expr

esse

d ge

nes

1200

Upregulated Downregulated

(a)

57

1919

19 19

19

ImmunityHemostasisLipid metabolic processSignal transductionMetabolismCell migrationdevelopment

Cell surface interactionsDiseaseNeuronal systemTransmembrane transportof small moleculesMuscle contraction

283

188151

132

94

(b)

Figure 2 Results of gene expression analysis Differentially expressed genes (a) and pie chart of biological process () containing pathwayssignificantly enriched in PBLs incubated in MMG (b)

which 465 (29) genes were upregulated whereas 1116 (71)genes were downregulated (Figure 2(a) and SupplementaryTable S2) By selecting a 2-fold cut-off threshold we iden-tified 312 (197) genes in MMG 157 genes (10) showedalterations in expression level with a fold change greater than40 and among these 20 genes showed a fold change ge160(Supplementary Table S3) To identify sets of genes withexpression changes in MMG condition we used Graphite[51] a novel web tool for topological-based pathway analysesbased on high-throughput gene expression data analysesPathway analysis on differentially expressed genes has beenperformed by using hypergeometric test on Reactome Path-ways as implemented inGraphiteweb considering significantthose categories with FDR lt 01 We evidenced biologicalpathways significantly enriched in MMG 15 (283) wererelated to immunity 10 (188) to hemostasis 7 (132) tolipid metabolic process and signal transduction 5 (94) tometabolism 3 (57) to cell migrationdevelopment and 1(19) tocell surface interactions disease neuronal systemtransmembrane transport of small molecules and musclecontraction (Figure 2(b)) The list of pathways is reported inSupplementary Table S4

Among the immune pathways significantly enriched inMMG those of ldquoMHC class II antigen presentationrdquo ldquoTollReceptor Cascadesrdquo and ldquoDAP12 signalingrdquo showed the high-est number of differentially expressed genes (30 20 and 20genes resp) Also the ldquointerferon gamma signalingrdquo pathwaywas significantly enriched in MMG with 13 differentiallyexpressed genes Ten genes codifying for proteins of MHCclass II (HLA-DRA HLA-DRB1 HLA-DRB3 HLA-DRB4

HLA-DRB5 HLA-DQA1 HLA-DQA2 HLA-DPA1 HLA-DPB1 and HLA-DQB1) were common to nine pathways(Figure 3) CD4 codifying for a membrane glycoprotein of Tlymphocytes that interacts with the major histocompatibilitycomplex class II antigens was common to eight immunepathways RELA PTEN and PAK1 were included in threepathways whereas HLA-DOA andHLA-DMBwere includedin two pathways

33 Target Prediction and Integration Analysis of miRNAand mRNA Expression Profiles We examined the regulatoryeffects of miRNAs on global gene expression under modeledmicrogravity (MMG) condition in comparison with groundgravity (1 g) To predict the target genes of differentiallyexpressed miRNAs in MMG we performed a computationalanalysis using TargetScan tool which predicts biologicaltargets of miRNAs by searching for the presence of conserved8mer and 7mer sites that match the seed region of eachmiRNA [59] However all available software for target pre-diction is characterized by a large fraction of false positivethus the integration of target predictions with miRNA andgene target expression profiles has been proposed to refinemiRNA-mRNA interactions The correlation analyses on thedifferentially expressed miRNAs and mRNAs were carriedout with MAGIA2 software [43] by microRNA Pearson pre-diction analysis which allowed the identification of miRNA-mRNA interactions (Supplementary Table S5) To discoverfunctional relationships between miRNAs and the transcrip-tome and uncover the gene pathways that are regulated


Gene symbol description Fold change Toll

rece

ptor

casc

ade

HLA-DPA1 Major histocompatibility complex class II DP alpha 1 HLA-DPB1 Major histocompatibility complex class II DP beta 1 HLA-DQA1 Major histocompatibility complex class II DQ alpha 1HLA-DQA2 Major histocompatibility complex class II DQ alpha 2HLA-DQB1 Major histocompatibility complex class II DQ beta 1 HLA-DRA Major histocompatibility complex class II DR alphaHLA-DRB1 Major histocompatibility complex class II DR beta 1 HLA-DRB3 Major histocompatibility complex class II DR beta 3 HLA-DRB4 Major histocompatibility complex class II DR beta 4 HLA-DRB5 Major histocompatibility complex class II DR beta 5 HLA-DOA Major histocompatibility complex class II DO alphaHLA-DMB Major histocompatibility complex class II DM betaCD4 CD4 moleculeRELA v-rel reticuloendotheliosis viral oncogene homolog APTEN Phosphatase and tensin homologPAK1 p21 protein (Cdc42Rac)-activated kinase 1

minus178minus183minus151minus164minus14minus155minus161minus185minus174minus159minus095minus147minus078051

minus026minus132

DA

P12

signa

ling

Cos

timul

atio

n by

the C

D28

fam

ily

TCR

signa

ling

Dow

nstre

am T

CR si

gnal

ing

MH

C cla

ss II

antig

en p

rese

ntat

ion

Gen

erat

ion

of se

cond

mes

seng

er m

olec

ules

Tran

sloca

tion

of Z

AP-70

to im

mun

olog

ical

syna

pse

Phos

phor

ylat

ion

of C

D3

and

TCR

zeta

chai

ns

Inte

rfero

n ga

mm

a sig

nalin

g

PD-1

signa

ling

Figure 3 Differentially expressed genes common to immune-related pathways identified by Reactome database in PBLs incubated inMMGThe expression value of each gene indicated as fold change is the mean of expression levels calculated as the log2 ratio (MMG1 g) on PBLsamples (see Supplementary Table S2)

by miRNAs in MMG we performed Gene Ontology (GO)analysis using DAVID [54]

In our analysis we used high classification stringency andconsidered only GO terms that have 119875 lt 01 after permu-tation corrections (Benjamini) (Table 1) Several GO termsbelonged to immune system function (ie ldquoinnate immuneresponserdquo ldquoinflammatory responserdquo ldquoregulation of cytokineproductionrdquo ldquopositive regulation of immune system processrdquoand ldquoresponse to bacteriumrdquo) in accordance with the resultsof pathway analysis on transcriptome (see SupplementaryTable S4) GO terms of ldquocell developmentrdquo ldquoregulation ofcell differentiationrdquo ldquoregulation of cell communicationrdquo ldquocellmotilityrdquo and ldquocell migrationrdquo were significantly enriched inMMG together with the category ldquoorgan developmentrdquo Inaddition the biological categories of ldquoregulation of signaltransductionrdquo ldquoregulation of response to stressrdquo ldquoregulationof cell deathrdquo and ldquoregulation of cell proliferationrdquo wereenriched in PBLs incubated in MMG

To determine whether different miRNAs within a GOcategory interact with the same target genes we performednetwork analysis using MAGIA2 [43] a software platformfor the visualization of complex miRNA-mRNA interactionsWe focused on miRNAs that correlated both positively andnegatively with the GO categories of immuneinflammatory

response regulation of programmed cell death and reg-ulation of cell proliferation As shown in Figure 4 mosttranscripts are associated with more than one miRNA as inthe case of TLR4 transcript correlated with eight differentmiRNAs (miR-10a-5p miR-7-5p miR-135a-3p miR-103a-3pmiR-7-1-3p miR-107 miR-629-5p and miR-362-5p)

By using Cytoscape [60] we visualized the func-tional interactions between miRNAs whose expression levelschanged the most in PBLs incubated in MMG such as miR-9-5p miR-9-3p miR-155-5p miR-150-3p and miR-378a-3pand correlated target genes involved in GO categories ofimmuneinflammatory response regulation of programmedcell death and regulation of cell proliferation (Figure 5)

Our results show that miR-155-5p correlates with IFNGIL17F BCL6 and RELA involved in immuneinflammatoryresponse with PTEN BNIP3L APAF1 and PDCD4 involvedin regulation of programmed cell death and with NKX3-1 involved in regulation of cell proliferation miR-150-3pcorrelates with immune-related genes (IFNG IL1A andHLA-DRB1) and with proapoptotic gene PDCD4 miR-9-3p correlates with genes regulating cell proliferation (NKX3-1 GADD45A and TP53BP1) apoptosis (APAF1 BNIP3L)and immunity (CCL7 CXCL5 and BCL6) Among genesenriched within the three functional categories miR-9-5p


Table 1 Selected GO terms of biological processes significantly affected by microgravity The complete list of GO terms can be found inSupplementary Table S6

GOID Term Count 119875 value Fold enrichment FDRGO0045087 Innate immune response 25 242 times 10minus5 2566978193 0040412626GO0009966 Regulation of signal transduction 88 167 times 10minus4 1471577879 027973741GO0048468 Cell development 62 171 times 10minus4 1610338849 0285253218GO0045595 Regulation of cell differentiation 54 268 times 10minus4 1648 0448057857GO0006954 Inflammatory response 41 323 times 10minus4 1787513228 0539044451GO0080134 Regulation of response to stress 37 543 times 10minus4 1806696296 090478009GO0010646 Regulation of cell communication 95 822 times 10minus4 1380599647 1366092587GO0048513 Organ development 139 822 times 10minus4 1290910116 1366713032GO0042060 Wound healing 26 838 times 10minus4 2025910165 1393334889GO0007165 Signal transduction 198 0001398603 1211224944 2313943814GO0001817 Regulation of cytokine production 27 0001406003 1926233766 2326052387GO0048514 Blood vessel morphogenesis 26 0001709187 193009009 2820918727GO0001817 Regulation of cytokine production 15 0001406003 1926233766 2326052387GO0007399 Nervous system development 88 0002097026 1359812471 3450524099GO0022603 Regulation of anatomical structure morphogenesis 28 0002408369 1831111111 3953172097GO0048870 Cell motility 33 0002838826 1710188679 464407324GO0048522 Positive regulation of cellular process 159 0003551259 1221594406 5777298149GO0002684 Positive regulation of immune system process 31 0003855111 1711490787 6256755641GO0050865 Regulation of cell activation 26 0003880839 1819447983 6297247375GO0051174 Regulation of phosphorus metabolic process 51 0003964604 1486259947 6428964883GO0001944 Vasculature development 28 0003981783 1767969349 6455956993GO0043066 Negative regulation of apoptosis 41 0004040803 156952381 6548633436GO0010941 Regulation of cell death 83 0004115909 1341019608 6666445839GO0009617 Response to bacterium 22 0004991824 19032021 8030143455GO0043067 Regulation of programmed cell death 82 000552374 1328770895 8849108723GO0008285 Negative regulation of cell proliferation 40 000576007 154741784 9210770389GO0010557 Positive regulation of macromolecule biosynthetic process 62 0006543793 139014966 104004889GO0016477 Cell migration 30 0006706505 1664646465 1064564555GO0042127 Regulation of cell proliferation 71 0009848433 1331149033 152576907GO0048523 Negative regulation of cellular process 139 0012514364 119870225 189945349

correlates with BCL6 miR-378a-3p is correlated with HLA-DRB1 GPNMB and NKX3-1 the same transcripts togetherwith IL17F are correlated also with miR-378a-5p

The microarray data from miRNA and gene expressionprofiling were validated by real-time qPCR experiments forfour miRNAs (miR-9-5p miR-155-5p miR-378a and miR-150-3p) and five mRNAs (IFNG IL17F BCL6 HLA-DRB1and TLR4) whose expression level was significantly alteredby MMG incubation (Figure 6) miR-9-5p and miR-155-5p together with IFNG IL17F and BCL6 transcripts wereupregulated in MMG whereas miR-378a and miR-150-3ptogether with HLA-DRB1 and TLR4 transcripts were down-regulated in MMG

34 Intraclass Integrated Analysis Recently Censi and col-leagues [61] observed a significant increase in the numberand strength of genes correlation under stress conditionssuch as disease and environmental or physiological changesTo evaluate whether the stress induced by MMG increasesthe amount of correlation of the system with respect to

1 g control condition we integrated mRNAs and miRNAsdata separately for MMG and 1 g using MAGIA2 By com-paring the two regulatory networks we observed a similarnumber of interactions between MMG (190 interactions)and 1 g (218 interactions) (Figure S1) indicating that therewas no significant connectivity enrichment under modeledmicrogravity By contrast GeneOntology analysis performedon MMG- and 1 g-specific interactions reported 50 GOcategories significantly enriched in only MMG condition(119875 lt 02 after Benjamini corrections Supplementary TableS7) 10 out of these were previously described in Table 1 Inparticular the GO categories ldquoregulation of cellular processrdquoand ldquocell differentiationrdquo were significantly affected byMMGWith intraclass analysis the GO categories of immunity andcell death were not enriched in MMG probably because inour study the number of PBL samples available for suchanalysis was relatively small On the whole the intraclassanalysis shows that modeled microgravity does not increasethe general connectivity of miRNA-gene interaction networkbut rather increases the transcriptome plasticity with respect


hsa-let-7i-3p

hsa-miR-200a-3p

TP73

CXCL10

hsa-miR-34a-5p

IRAK2

hsa-miR-663aIL18RAP

CCL19

hsa-miR-150-3p

BMP6

IL6

CXCL11CD180

CCR7

hsa-miR-378a-5p

FOS

FCN2

hsa-miR-125a-5p

hsa-miR-132-3pPDPN

hsa-miR-505-5phsa-miR-34b-5p

hsa-miR-155-5p

hsa-miR-9-3p PRDX2

ATRN

FN1

SPP1HMOX1

RELAhsa-miR-7-5p

hsa-let-7i-5p

LYZ

CLEC7AVSIG4

hsa-miR-10a-5p

hsa-miR-940

hsa-miR-362-5p

hsa-miR-532-5pCD55

hsa-miR-9-5p

hsa-miR-146b-5p

IL1RAPhsa-miR-7-1-3p

hsa-miR-103a-3p

CXCL1

CIITA

TLR4

TLR7

hsa-miR-629-5p

hsa-miR-107

MEFV

hsa-miR-376a-3p

hsa-miR-376c-3p

IL17F

IL1A

hsa-let-7e-5phsa-miR-221-5p

hsa-miR-185-5p

CCL7

hsa-miR-135a-3p

SIGLEC1 TLR5

LIPA AIF1

CFP


hsa-miR-1225-5p

C1QCF2R

C5

C1QB

hsa-miR-378a-3p

(a) Immuneinflammatory response

GIMAP5

DDAH2 HMOX1

SOX4PRDX1

hsa-miR-505-5p

hsa-miR-223-3p

SNCA hsa-miR-221-5p

hsa-miR-192-5p

UNC13B

PTPRFRNF7

TXNIPAPAF1

RUNX3

MGMTPTENNOTCH2IFNG

MPOETS1

NRG1

PRDX2

ANXA4

TNFSF13

NEFL GPX1NAIP

BNIP3L hsa-miR-185-5p

STK17ATLR4


hsa-miR-135a-3p

hsa-miR-99b-5p

IL12A

ITGA1CHD8

RAG1

UACASH3RF1

IL1A

HGF

KCNMA1

HTATIP2

SMOCAMK1D

IL6NOL3

AVEN

APOE

hsa-miR-663a

hsa-miR-200a-3p

hsa-miR-378a-5phsa-let-7i-5phsa-miR-532-5p

hsa-miR-376c-3phas-miR-362-5p

TAF9B

LTB

RASGRF2

LGALS1

TRIO

MEF2CBCL6

NQO1

ERBB3

AIFM3

TRAF1

FURIN

NR4A2

SERPINB2

FGD2

PIM2

NR4A1

TNFSF14

TIMP3

CEBPG

FGD4CARD6

RAB27A

ARHGEF3NF1

PPT1

F2RRELA

VAV2

SMAD6

ESR1

BMF ITSN1

MUC2

FOXO3

hsa-miR-132-3p

hsa-miR-423-3p

hsa-miR-9-5p

hsa-miR-146b-5p


hsa-miR-378a-3p

hsa-miR-629-5p

SLC25A4

hsa-let-7e-5p

MAP3K5

hsa-miR-181a-3p

TP73hsa-let-7i-3p

PPP1R13Bhsa-miR-9-3p

hsa-miR-940

hsa-miR-7-1-3p

hsa-miR-625-5p


hsa-miR-342-5p

hsa-miR-7-5p

hsa-miR-155-5p

hsa-miR-29b-1-5p

hsa-miR-10a-5p

(b) Regulation of programmed cell deathLTB

VSIG4

hsa-a-423-3p

SMO

hsa-let-7i-5p

hsa-miR-629-3p

hsa-let-7e-5p

VEGFB

NCK2

hsa-miR-505-5pHMOX1

IL12RB2 IL6CEBPA

CD86

CYP27B1

hsa-miR-7-5p

RARRES1

hsa-miR-940

hsa-miR-629-5p

hsa-miR-146b-5p

KLF11

RAC2

APOEPDCD1LG2

IFI30

F2R

hsa-miR-342-5p

hsa-miR-625-5pIL1A

SYKKLF4

HHEX

BCL6

AIF1

PDGFRB

IL12A

IFNGCTTNBP2

GPNMB

EREG

HES1

ETS1

hsa-miR-378a-5p

hsa-miR-155-5p

hsa-miR-378a-3p

NKX3-1

PPARG

CD33

NOTCH4

hsa-miR-221-5p

hsa-miR-135a-3p

ERBB3

CXCL10

COMTTNFRS13B

hsa-miR-34a-5p

hsa-miR-376c-3p

TNFSF13

hsa-miR-1225-5p

GPX1

hsa-miR-132-3p

hsa-miR-185-5p CXCL5CXCL1

KLF5FIGF

SOX2

KIFAP3 DLG3

hsa-miR-532-5p

hsa-miR-192-5p

hsa-miR-193b-3pPLAU

PTGS1PTEN

PDGFBIRS1

hsa-miR-10a-5p

SOX4NDN

hsa-miR-223-3p

hsa-miR-29b-1-5p

NOTCH2

hsa-miR-7-1-3pTXNIP

hsa-miR-107

CD9

NF1

NRG1

SLAMF1

RUNX3

PTPRF

MUC2

hsa-miR-125a-5p

hsa-miR-103a-3p

CHRNA10

RELA

PDGFCIL13RA1

hsa-miR-9-3p

hsa-miR-34b-5p hsa-miR-663a

GRN

hsa-miR-9-5p

hsa-miR-362-5pTIMP2

(c) Regulation of cell proliferation

Figure 4 Network analysis on correlated miRNA-mRNA pairs in PBLs incubated in MMG Network analyses were performed by MAGIA2software using miRNAs correlating both positively and negatively with transcripts involved in immuneinflammatory response (a) inregulation of programmed cell death (b) and in regulation of cell proliferation (c) Circles represent transcripts and triangles representmiRNAs

to 1 g gravity condition as evidenced by the enriched GOcategories

35 In Vitro Validation of GO Analysis Effects of MMG onCell Proliferation and Apoptosis Experimental assays wereperformed to validate the results obtained by bioinformaticsanalyses on the GO categories ldquoregulation of cell prolifera-tionrdquo and ldquoregulation of programmed cell deathrdquo associatedwith variations in miRNA expression under MMG To mea-sure cell proliferation quiescent (119866

0) PBLs from the same

donorwere incubated for different times (24 h 48 h and 72 h)in 1 g and inMMGAt the end of incubation times the colony

forming ability has been determined by the T-cell cloningassay [44] in which cells were incubated in medium contain-ing mitogen factors (ie PHA and IL2) to trigger their cellcycle entry Our results showed that cloning efficiency (CE)decreased with time in both gravity conditions howeverMMG incubation affected the ability of PBLs to form colonies(119875 lt 005 at 24 h Figure 7(a)) To investigate whether MMGincubation increased the frequency of apoptotic cells PBLswere scored for the presence of apoptotic bodies Apoptoticindex was very similar at 24 and 48 h and significantlyhigher in PBLs incubated in MMG than in 1 g (Figure 7(b)119875 lt 005) In the same PBL samples caspase-3 activationassayed by the cleavage of the peptide substrate DEVD-AFC


Expression value

Regu

latio

n of

cell

prol

ifera

tion

Regu

latio

n of

pro

gram

med

cell

deat

hIm

mun

ein

flam

mat

ory

resp

onse

hsa-miR-155-5pIFNG

hsa-miR-125a-5p hsa-miR-940

hsa-miR-150-3p

hsa-miR-150-3p

hsa-miR-9-3p

CCL7hsa-let-7e-5p hsa-miR-7-5p

hsa-miR-135a-3p

hsa-miR-378a-5p

hsa-miR-135a-3p

IL17F

hsa-miR-155-5p

hsa-miR-9-3p

CXCL5

IL1A

hsa-miR-7-5pRELA


hsa-miR-378a-5p hsa-miR-150-3p

HLA-DRB1hsa-miR-378a-3p hsa-miR-629-5p

hsa-miR-9-3p hsa-miR-9-5pBCL6

hsa-miR-629-5p

hsa-miR-34a-5p

hsa-miR-185-5p

hsa-miR-223-3p

hsa-miR-107

hsa-miR-185-5pPTEN

hsa-miR-10a-5p hsa-miR-7-1-3p

hsa-miR-29b-1-5p


hsa-miR-7-5p

hsa-miR-9-3p

hsa-miR-7-1-3p

hsa-miR-29b-1-5p

hsa-let-7e-5p

hsa-miR-505-5pAPAF1


hsa-miR-221-5p

hsa-miR-376a-3p

hsa-miR-9-3pTP53BP1

hsa-miR-9-3p

hsa-miR-29b-1-5p

GADD45A


hsa-miR-9-3p

hsa-miR-378a-3p

hsa-miR-378a-3p

hsa-miR-155-5p

hsa-miR-1225-5p

hsa-miR-185-5phsa-miR-185-5pNKX3-1

hsa-miR-107GPNMB


PDCD4

hsa-miR-200a-3p

hsa-miR-135a-3p

hsa-miR-532-5p

hsa-miR-629-5p

hsa-miR-362-5p

minus51 0 75

hsa-miR-378a-5p

BNIP3Lhsa-miR-155-5p

hsa-let-7i-5p

Figure 5 Cytoscape visualization of miRNA-mRNA correlations in PBLs incubated 24 h in MMG Relationships between miRNAs andcorrelated target genes involved in ldquoimmuneinflammatory responserdquo (IFNG CCL7 IL17F HLA-DRB1 IL1A CXCL5 RELA and BCL6)ldquoregulation of programmed cell deathrdquo (PTEN BNIP3L APAF1 and PDCD4) and ldquoregulation of cell proliferationrdquo (GPNMB NKX3-1TP53BP1 and GADD45A) Circles represent transcripts and triangles represent miRNAs the expression levels of each feature are representedas color scale


0

05

1

15

2

25

3

35

4

45

miR-9-5p miR-155-5p

MMG

Relat

ive e

xpre

ssio

n

lowastlowastlowast

lowastlowast

1g

0

02

04

06

08

1

12

miR-378a miR-150-3p

Relat

ive e

xpre

ssio

n

lowastlowastlowast lowastlowastlowast

MMG1g

(a)

0

5

10

15

20

25

IFNG IL17F BCL6

80

120

140

Relat

ive e

xpre

ssio

n

lowast

lowastlowastlowast

MMG1g

0

02

04

06

08

1

12

TLR4 HLA-DRB1

Relat

ive e

xpre

ssio

nlowastlowastlowast lowastlowastlowast

MMG1g

(b)

Figure 6 Microarray data validation by quantitative real-time PCR (qRT-PCR) Validation of microarray data by qRT-PCR in MMG-incubated versus 1 g incubated PBLs The results are consistent with the cumulative microarray data of miRNAs (a) and mRNAs (b) Values(fold change dark grey bars) are means plusmn SE of expression levels calculated as the log2 (MMG1 g) on PBL samples from 4 to 6 differentdonors The value ldquo1rdquo of control 1 g PBLs (light grey bars) is arbitrarily given when no change is observed ( lowastlowastlowast119875 lt 0001 lowastlowast119875 lt 001 andlowast

119875 lt 005 t-test)

increased significantly in PBLs incubated 48 h in MMG withrespect to those in 1 g (Figure 7(c) 119875 lt 005)

4 Discussion

In the present study we evaluated the effects of modeledmicrogravity (MMG) on human PBLs by analyzing miRNAand gene expression profiles in comparison with PBLs cul-tured in Earth gravity condition (1 g) Our results reported 42differentially expressed miRNAs in PBLs cultured for 24 h inMMGwith respect to 1 g of which 14 (miR-34a-5p miR-34b-5p miR-663a miR-135a-3p miR-1225-5p miR-940 miR-221-5p miR-29b-1-5p miR-10a-5p let-7i-3p miR-200a-3p miR-7-5p miR-7-1-3p and miR-505-5p) were found altered also

by 120574-irradiation as assessed in our previous study [47] Themost dysregulatedmiRNAs identified in the present work arethe upregulated miR-9-5p miR-9-3p and miR-155-5p andathe downregulated ones are miR-150-3p and miR-378a-3pSuch miRNAs have been found altered in human tumors inparticular miR-9 is an oncogenic miRNA overexpressed inmixed lineage leukemia- (MLL-) rearranged acute myeloidleukemia [62] in muscle-invasive bladder cancer [63] andin osteosarcoma cell lines [64] miR-155 is commonly upreg-ulated in hematological malignancies [65 66] and has beenlinked to the development of breast lung and stomachtumors [67ndash70] miR-150 is significantly downregulated inmost cases of acute myeloid leukemia [71] and colorectalcancer [72] in addition miR-150 has an important role in


25

20

15

10

5

0

lowast

24 48 72


Clon

ing e

fficie

ncy (

)

1gMMG

(a)

)

5

4

3

2

1

024 48


Apop

totic

inde

x (

1gMMG

lowast lowast

(b)

60

50

40

30

20

10

0

24 48

70


Casp

ase-3

activ

ation

(au

)

lowast

1gMMG

(c)

Figure 7 Cell proliferation and apoptosis induction in human PBLs incubated in MMG and in 1 g (a) T-cell cloning assay performed atthe end of 24 h 48 h and 72 h of incubation in the two gravity conditions Data are means plusmn SE from thirteen independent experiments(b) Apoptotic index at the end of incubation for 24 h and 48 h in MMG and 1 g determined by nuclear chromatin condensation with DAPIstaining (c) Caspase-3 activation at the end of 24 h and 48 h incubation in MMG and 1 g assessed by fluorimetric assay (au arbitrary units)Data in (b) and (c) are means plusmn SE from 3-4 independent experiments (lowast119875 lt 005 t-test)

normal hematopoiesis and its aberrant downregulation is asensitivemarker indicative of lymphocyte depletion and bonemarrow damage [73] miR-378 (actually annotated as miR-378a) is significantly downregulated in colorectal cancer [74]in cutaneous squamous cell carcinoma [75] and in renalcell carcinoma [76] The effects of microgravity on miRNAexpression profile are currently reported in only one studycarried out in human lymphoblastoid TK6 cells incubatedunder simulated microgravity for 72 h [77] Among thedysregulated miRNAs only two were common to our datamiR-150 and miR-34a although the direction and intensityof fold change were different demonstrating the cell typespecific signature of miRNA profile

miRNAs modulate gene expression by interacting withthe 31015840UTR of target genes and since a single miRNA couldhave hundreds to thousands of predicted target genes [25]it is difficult to determine the true target regulated by themiRNA which affects a biological function Moreover bind-ing of multiple miRNAs to one target could further increasethe complexity of target prediction The identification ofmiRNA target genes is usually performed by bioinformaticprediction algorithms based on (i) sequence similarity searchpossibly considering target site evolutionary conservationand (ii) thermodynamic stability However it is known thatthe results of target prediction algorithms are characterizedby very low specificity [78] For this purpose the integrationof target predictions with miRNA and gene expression pro-files has been recently proposed to improve the detection offunctional miRNA target relationships [79 80] Therefore toidentify the most likely target genes of miRNAs differentiallyexpressed in MMG we defined gene expression signatureon the same samples of PBLs assayed for miRNA profilingthen we integrated expression profiles from both miRNAsand mRNAs with in silico target predictions to reducethe number of false positives and increase the number of

biologically relevant targets [81ndash83] Our results of geneexpression profiling reported the downregulation of multiplegenes in MMG (71) in accordance with previous findingsin activated human T lymphocytes incubated for 24 h insimulated microgravity [21] Moreover we found that about20 of genes responded to MMG by more than 2-foldchange in expression level and twenty genes showed a ge16-fold change in expression Most of these top dysregulatedgenes were immune-related such as those codifying forinflammatory cytokines (CCL1 CCL7 CXCL5 CXCL11 andIL1A) and for proteins with a role in immunoregulatoryfunctions (IFNG TNIP3 TREM1 APOC1 FCN1 FCN2and CPVL) (Supplementary Table S3) Biological pathwaysenriched in PBLs exposed to MMG were mainly involved inimmunity (Figure 2(b)) including adaptive immune systemresponse (ie PD-1 signaling phosphorylation of CD3 andTCR zeta chains translocation of ZAP-70 to immunologicalsynapse MHC class II antigen presentation TCR signalingand costimulation by the CD28 family) innate immunesystem response (ie Toll Receptor Cascades) and cytokinesignaling in immune system (ie interferon gamma signal-ing) (Supplementary Table S4) All these pathways includedten downregulated genes codifying for MHCmolecules classII (HLA-DPA1 HLA-DPB1 HLA-DQA1 HLA-DQA2 HLA-DQB1 HLA-DRA HLA-DRB1 HLA-DRB3 HLA-DRB4and HLA-DRB5) which are expressed in antigen presentingcells (APC) and play a central role in the immune system bypresenting peptides derived from extracellular proteins [84]Therefore the downregulation of these genes suggests that thedisplay of antigens at the cell surface of APCmay be disturbedby gravity reduction affecting the efficiency of immuneresponse as observed in astronauts during spaceflight andimmediately afterwards [85ndash87] Moreover our data are inaccordance with the inhibition of immediate early genesin T-cell activation observed in space microgravity [14]


and with alterations of gene expression in human activatedT-cells incubated in modeled microgravity including thedownregulation of HLA-DRA gene [21]

By integrating the transcriptome and microRNAomewe detected significant miRNA-mRNA relationships underMMG Since miRNAs act prevalently through target degra-dation expression profiles of miRNAs and target genesare generally expected to be inversely correlated Neverthe-less since miRNA activity is part of complex regulatorynetworks and gene expression profiles are the result ofdifferent levels of regulation also positive correlation (ieupregulated miRNAupregulated mRNA or downregulatedmiRNAdownregulated mRNA) is expected Indeed in an invivo mouse model the activated expression of miRNAs hasbeen shown to correlate with activated expression of mRNAsrather thanwithmRNAdownregulation [88] GeneOntology(GO) analysis conducted on the significantly correlatedmiRNA-mRNA pairs evidenced the biological categoriessignificantly overrepresented in MMG (Table 1) Many GOterms of immune response were enriched such as ldquoinnateimmune responserdquo ldquoinflammatory responserdquo ldquoregulation ofcytokine productionrdquo ldquopositive regulation of immune systemprocessrdquo and ldquoresponse to bacteriumrdquo Notably the mostdysregulated miRNAs detected in the present study miR-378a-3p miR-150-3p miR-155-5p miR-9-3p and miR-9-5pare significantly correlated with immune-related genes Inparticular miR-378a-3p is positively correlated with tran-scripts of MHC molecules class II such as HLA-DRB1(Figure 5) and HLA-DOA HLA-DRB5 and HLA-DQA2(not shown) miR-150-3p is negatively correlated with HLA-DRB1 IFNG and IL1Atranscripts whereas miR-155-5p ispositively correlated with IFNG and IL17F and negativelycorrelated with RELA and BCL6 (Figure 5) IL1A IL17Fand IFN-120574 are proinflammatory cytokines acting during theimmune response IFN-120574 is a soluble cytokine having broaderroles in activation of immune responses in part throughupregulating transcription of genes involved in antigen pro-cessingpresentation in cell cycle regulation and apoptosisand its correlation with miR-155 has been recently validated[89] BCL6 which encodes a nuclear transcriptional repres-sor has a role not only in regulation of lymphocyte functionbut also in cell survival and differentiation Similarly thepleiotropic transcription factor RELA has a role in immunebiological process and it is also involved in cell growth andapoptosis Besides miR-155-5p BCL6 is correlated with fourmiRNAs including miR-9 (3p and 5p) in addition miR-9-3p is positively correlated with CCL7 and CXCL5 (Figure 5)Recent evidences show that miR-9 is highly involved inimmunity and inflammatory diseases [90ndash92] by enhancingIFN-120574 production in activated human CD4(+) T-cells [91]Moreover Gao et al [90] have shown that miR-9 disturbsthe display of antigens at the cell surface by suppressing theexpression of MHC class I gene transcription

Together with the categories of immune response GOanalysis reported that also the categories of regulation of cellproliferation and regulation of programmed cell death weresignificantly enriched in MMG as previously reported in 120574-irradiated PBLs [47] Interestingly such categories were notenriched from pathway analysis conducted on transcriptome

and the reason could be that integrated analysis of miRNAsand mRNAs expression profiles evidences the posttran-scriptional effect mediated by miRNAs on gene expressionAmong genes involved in cell proliferation the transcriptionfactor NKX3-1 (24-fold upregulated) which mediates non-cell autonomous regulation of gene expression and inhibitscell proliferation is correlated with miR-9-3p miR-155-5pmiR-378a-3p and miR-378-5p (Figure 5) TP53BP1 (14-foldupregulated) encoding for a chromatin-associated factorinvolved in cell cycle checkpoint and growth is correlatedwith miR-9-3p GADD45A (15-fold upregulated) regulatingcell cycle arrest DNA repair cell survival senescence andapoptosis is also correlated with miR-9-3p GPNMB (32-fold downregulated) expressed in a wide array of normaltissues such as bone hematopoietic system and skin whereit influences cell proliferation adhesion differentiation andsynthesis of extracellular matrix proteins [93] is targetedby five miRNAs including miR-378a-3p Among miRNA-correlated genes involved in apoptosis we identified PDCD4(proapoptotic 14-fold upregulated) and found out that it iscorrelated with seven miRNAs including miR-155-5p andmiR-150-3p BNIP3L (proapoptotic 15-fold downregulated)also correlated with seven miRNAs including miR-155-5pandmiR-9-3p APAF1 (proapoptotic 13-fold downregulated)correlated with four miRNAs including miR-155-5p andmiR-9-3p and PTEN (proapoptotic) correlated with sevenmiRNAs including miR-155-5p Notably many transcripts(ie BCL6 PTEN BNIP3L PDCD4 NKX3-1 and GPNMB)are targeted by multiple miRNAs indicating a pleiotropiceffect in gene regulation by coexpressed endogenousmiRNAsin MMG Moreover our results suggest that under MMGcondition a small group of miRNAs regulates transcriptomeby modulating the same transcripts within one pathway

To validate the results of Gene Ontology analysis weevaluated whether the GO categories ldquoregulation of cell pro-liferationrdquo and ldquoregulation of programmed cell deathrdquo wereaffected byMMG incubation by performing biological assaysof T-cell cloning and apoptosis induction Our results showthat cloning ability of PBLs was lower after 24 h incubationin MMG than in 1 g in accordance with the suppression ofproliferative response of human lymphocytes to mitogenicstimulation in microgravity [94 95] The ability to originateclones in PBLs incubated for 48h and 72 h decreased in bothgravity conditions probably because the longer119866

0-phase con-

dition experienced by PBLs affected their responsiveness toenter into cell cycleThe antiproliferative effect of micrograv-ity has been recently reported also in human thyroid cancercells [96] and in human lung adenocarcinoma cells [97]Our results of apoptosis induction demonstrated that boththe formation of apoptotic bodies activation and caspase-3activation increased significantly in PBLs incubated inMMGthan in 1 g The activation of apoptotic process seems relatedto the overexpression of PDCD4 and RELA rather than tothe underexpression of BNIP3L and APAF1 indicating theexistence of complex regulatory networks between miRNAsand mRNAs that occur at different levels of regulation IFN-120574 besides having an important role in activating innateand adaptive immune responses plays important roles in


inhibiting cell proliferation and inducing apoptosis Its over-expression in MMG mediated by miR-9-3p and miR-155-5p could thus mediate the antiproliferative effect and theapoptosis induction In addition the correlation betweenmiR-9-3p and TP53BP1 could explain the clonogenicitydecrease and apoptosis increase in PBLs incubated in MMGIndeed overexpression of TP53BP1 has been demonstrated todecrease the clonogenicity and induce apoptosis in ovariancancer cells [98]

5 Conclusions

Our results show that MMG leads to changes in expressionlevel of a considerable fraction of microRNAome and tran-scriptome in human PBLs miRNAs differentially expressedin MMG are correlated with immuneinflammatory-relatedgenes as IFNG accordingly with the important role ofmiRNAs in immune function regulation Since inflammationworks as a tumor-promoting agent the abnormal expressionof such miRNAs under microgravity condition could influ-ence the carcinogenic process by affecting cancer cell immuneescape Moreover miRNAs mostly dysregulated in MMGsuch as miR-9 miR-155 and miR-150 are oncogenic sug-gesting that their abnormal expression can influence the car-cinogenic process The results of miRNA-mRNA integrationanalysis demonstrate that MMG increases the transcriptomeplasticity compared with 1 g condition and that categories ofregulation of cell proliferation and programmed cell deathare affected by MMG as confirmed by in vitro experimentalvalidation Taken togetherour results of high-throughputexpression analysis and miRNA-mRNA integration analysisgive new insight into the complex genetic mechanisms of cellresponse to stress environment under reduced gravity




M Mognato and C De Pitta conceived and designed theexperiments C Girardi S Casara and M Mognato per-formed the experiments C De Pitta C Romualdi E CaluraL Celotti and M Mognato analyzed the data M Mognatoand L Celotti wrote the paper Cristina Girardi and CristianoDe Pitta contributed equally to this work

Acknowledgments

The authors gratefully acknowledge M De Bernard forcritical discussion and R Mazzaro for graphical supportThis work was done with the support of the Italian SpaceAgency (ASI XMAB fromMolecules to Man 1014060) toL Celotti and of the University of Padova (CPDA061783) toMMognatoThe authors also apologize to the authors whosework could not be cited due to space limitations

References

[1] R H Fitts D R Riley and J J Widrick ldquoFunctional and struc-tural adaptations of skeletal muscle to microgravityrdquo Journal ofExperimental Biology vol 204 no 18 pp 3201ndash3208 2001

[2] M Narici B Kayser P Barattini and P Cerretelli ldquoEffects of 17-day spaceflight on electrically evoked torque and cross-sectionalarea of the human triceps suraerdquo European Journal of AppliedPhysiology vol 90 no 3-4 pp 275ndash282 2003

[3] S Trappe D Costill P Gallagher et al ldquoExercise in spacehuman skeletal muscle after 6 months aboard the InternationalSpace Stationrdquo Journal of Applied Physiology vol 106 no 4 pp1159ndash1168 2009

[4] S I M Carlsson M T S Bertilaccio E Ballabio and J AMMaier ldquoEndothelial stress by gravitational unloading effectson cell growth and cytoskeletal organizationrdquo Biochimica etBiophysica Acta vol 1642 no 3 pp 173ndash179 2003






[10] B E Crucian R P Stowe D L Pierson and C F SamsldquoImmune system dysregulation following short- vs long-duration spaceflightrdquo Aviation Space and Environmental Medi-cine vol 79 no 9 pp 835ndash843 2008

[11] G Sonnenfeld J S Butel and W T Shearer ldquoEffects of thespace flight environment on the immune systemrdquo Reviews onEnvironmental Health vol 18 no 1 pp 1ndash17 2003

[12] G Sonnenfeld ldquoEditorial space flight modifies T cellactivationmdashrole of microgravityrdquo Journal of Leukocyte Biologyvol 92 no 6 pp 1125ndash1126 2012

[13] A Semov N Semova C Lacelle et al ldquoAlterations in TNF-and IL-related gene expression in space-flown WI38 humanfibroblastsrdquoTheFASEB Journal vol 16 no 8 pp 899ndash901 2002

[14] T T Chang I Walther C-F Li et al ldquoThe RelNF-120581B pathwayand transcription of immediate early genes in T cell activationare inhibited bymicrogravityrdquo Journal of Leukocyte Biology vol92 no 6 pp 1133ndash1145 2012


[16] S J PardoM J PatelMC Sykes et al ldquoSimulatedmicrogravityusing the Random Positioning Machine inhibits differentiationand alters gene expression profiles of 2T3 preosteoblastsrdquoAmerican Journal of Physiology vol 288 no 6 pp C1211ndashC12212005


[17] M Monticone Y Liu N Pujic and R Cancedda ldquoActivationof nervous system development genes in bone marrow derivedmesenchymal stem cells following spaceflight exposurerdquo Jour-nal of Cellular Biochemistry vol 111 no 2 pp 442ndash452 2010




[21] N E Ward N R Pellis S A Risin and D Risin ldquoGeneexpression alterations in activated human T-cells induced bymodeledmicrogravityrdquo Journal of Cellular Biochemistry vol 99no 4 pp 1187ndash1202 2006

[22] J Q Clement S M Lacy and B L Wilson ldquoGene expres-sion profiling of human epidermal keratinocytes in simulatedmicrogravity and recovery culturesrdquo Genomics Proteomics andBioinformatics vol 6 no 1 pp 8ndash28 2008

[23] R Kumari K P Singh and J W DuMond Jr ldquoSimulatedmicrogravity decreases DNA repair capacity and induces DNAdamage in human lymphocytesrdquo Journal of Cellular Biochem-istry vol 107 no 4 pp 723ndash731 2009


[25] J Krutzfeldt M N Poy andM Stoffel ldquoStrategies to determinethe biological function of microRNAsrdquoNature Genetics vol 38no 1 pp S14ndashS19 2006

[26] C A Nickerson C M Ott J W Wilson et al ldquoLow-shearmodeled microgravity a global environmental regulatory sig-nal affecting bacterial gene expression physiology and patho-genesisrdquo Journal of Microbiological Methods vol 54 no 1 pp1ndash11 2003

[27] K Arunasri M Adil K Venu Charan C Suvro S HimabinduReddy and S Shivaji ldquoEffect of simulated microgravity on Ecoli K12 MG1655 growth and gene expressionrdquo PLoS ONE vol8 no 3 Article ID e57860 2013

[28] O Marcu M P Lera M E Sanchez et al ldquoInnate immuneresponses of Drosophila melanogaster are altered by space-flightrdquo PLoS ONE vol 6 no 1 Article ID e15361 2011

[29] Y Honda A Higashibata Y Matsunaga et al ldquoGenes down-regulated in spaceflight are involved in the control of longevityin Caenorhabditis elegansrdquo Scientific Reports vol 2 article 4872012


[31] D P Bartel ldquoMicroRNAs target recognition and regulatoryfunctionsrdquo Cell vol 136 no 2 pp 215ndash233 2009

[32] H Guo N T Ingolia J S Weissman and D P BartelldquoMammalian microRNAs predominantly act to decrease targetmRNA levelsrdquo Nature vol 466 no 7308 pp 835ndash840 2010

[33] V Huang Y Qin J Wang et al ldquoRNAa is conserved inmammalian cellsrdquo PLoS ONE vol 5 no 1 Article ID e88482010

[34] L B Frankel N R Christoffersen A Jacobsen M Lindow AKrogh and A H Lund ldquoProgrammed cell death 4 (PDCD4)is an important functional target of the microRNA miR-21 inbreast cancer cellsrdquoThe Journal of Biological Chemistry vol 283no 2 pp 1026ndash1033 2008

[35] M N Poy M Spranger and M Stoffel ldquomicroRNAs and theregulation of glucose and lipid metabolismrdquo Diabetes Obesityand Metabolism vol 9 no 2 pp 67ndash73 2007

[36] N Stern-Ginossar N Elefant A Zimmermann et al ldquoHostimmune system gene targeting by a viral miRNArdquo Science vol317 no 5836 pp 376ndash381 2007

[37] C J Marsit K Eddy and K T Kelsey ldquoMicroRNA responsesto cellular stressrdquo Cancer Research vol 66 no 22 pp 10843ndash10848 2006

[38] P M Voorhoeve C le Sage M Schrier et al ldquoA geneticscreen implicates miRNA-372 and miRNA-373 as oncogenes intesticular germ cell tumorsrdquo Cell vol 124 no 6 pp 1169ndash11812006

[39] E A C Wiemer ldquoThe role of microRNAs in cancer no smallmatterrdquo European Journal of Cancer vol 43 no 10 pp 1529ndash1544 2007

[40] W C S Cho ldquoOncomiRs the discovery and progress ofmicroRNAs in cancersrdquo Molecular Cancer vol 6 article 602007

[41] S Mi J Lu M Sun et al ldquoMicroRNA expression signaturesaccurately discriminate acute lymphoblastic leukemia fromacute myeloid leukemiardquo Proceedings of the National Academyof Sciences of the United States of America vol 104 no 50 pp19971ndash19976 2007

[42] S M Hammond ldquoMicroRNAs as tumor suppressorsrdquo NatureGenetics vol 39 no 5 pp 582ndash583 2007

[43] A Bisognin G Sales A Coppe S Bortoluzzi andC RomualdildquoMAGIA2 from miRNA and genes expression data integrativeanalysis to microRNA-transcription factor mixed regulatorycircuits (2012 update)rdquoNucleic Acids Research vol 40 no 1 ppW13ndashW21 2012

[44] M Mognato and L Celotti ldquoModeled microgravity affects cellsurvival and HPRT mutant frequency but not the expressionof DNA repair genes in human lymphocytes irradiated withionising radiationrdquoMutation Research vol 578 no 1-2 pp 417ndash429 2005


[46] S-M Hou F J Van Dam F De Zwart et al ldquoValidation ofthe human T-lymphocyte cloning assaymdashring test report fromthe EU concerted action on HPRT mutation (EUCAHM)rdquoMutation Research vol 431 no 2 pp 211ndash221 1999

[47] C Girardi C de Pitta S Casara et al ldquoAnalysis of miRNAandmRNA expression profiles highlights alterations in ionizingradiation response of human lymphocytes under modeledmicrogravityrdquo PLoS ONE vol 7 no 2 Article ID e31293 2012

[48] H Wang R A Ach and B O Curry ldquoDirect and sensitivemiRNA profiling from low-input total RNArdquo RNA vol 13 no1 pp 151ndash159 2007



[50] V G Tusher R Tibshirani and G Chu ldquoDiagnosis of multiplecancer types by shrunken centroids of gene expressionrdquo Pro-ceedings of the National Academy of Sciences of the United Statesof America vol 98 pp 5116ndash5121 2001

[51] G Sales E Calura P Martini and C Romualdi ldquoGraphite webweb tool for gene set analysis exploiting pathway topologyrdquoNucleic Acids Research vol 41 pp 89ndash97 2013

[52] F Xin M Li C Balch et al ldquoComputational analysis ofmicroRNA profiles and their target genes suggests significantinvolvement in breast cancer antiestrogen resistancerdquo Bioinfor-matics vol 25 no 4 pp 430ndash434 2009

[53] HWang andW-H Li ldquoIncreasingMicroRNA target predictionconfidence by the relative R2 methodrdquo Journal of TheoreticalBiology vol 259 no 4 pp 793ndash798 2009

[54] D W Huang B T Sherman and R A Lempicki ldquoSystematicand integrative analysis of large gene lists using DAVID bioin-formatics resourcesrdquo Nature Protocols vol 4 no 1 pp 44ndash572009

[55] K J Livak and T D Schmittgen ldquoAnalysis of relative geneexpression data using real-time quantitative PCR and the2minusΔΔ119862T methodrdquoMethods vol 25 no 4 pp 402ndash408 2001

[56] R J Albertini K L Castle and W R Borcherding ldquoT-cellcloning to detect the mutant 6-thioguanine-resistant lympho-cytes present in human peripheral bloodrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 79 no 21 I pp 6617ndash6621 1982

[57] MMognato CGirardi S Fabris and L Celotti ldquoDNA repair inmodeledmicrogravity double strand break rejoining activity inhuman lymphocytes irradiated with 120574-raysrdquoMutation Researchvol 663 no 1-2 pp 32ndash39 2009

[58] S Canova F Fiorasi M Mognato et al ldquoldquoModeled micrograv-ityrdquo affects cell response to ionizing radiation and increasesgenomic damagerdquo Radiation Research vol 163 no 2 pp 191ndash199 2005

[59] B P Lewis C B Burge and D P Bartel ldquoConserved seedpairing often flanked by adenosines indicates that thousandsof human genes are microRNA targetsrdquo Cell vol 120 no 1 pp15ndash20 2005

[60] M S Cline M Smoot E Cerami et al ldquoIntegration ofbiological networks and gene expression data using CytoscaperdquoNature Protocols vol 2 no 10 pp 2366ndash2382 2007

[61] F Censi A Giuliani P Bartolini and G Calcagnini ldquoA multi-scale graph theoretical approach to gene regulation networks acase study in atrial fibrillationrdquo IEEETransactions onBiomedicalEngineering vol 58 no 10 pp 2943ndash2946 2011

[62] P Chen C Price Z Li et al ldquomiR-9 is an essential onco-genic microRNA specifically overexpressed in mixed lineageleukemia-rearranged leukemiardquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 110 no28 pp 11511ndash11516 2013

[63] G Pignot G Cizeron-Clairac S Vacher et al ldquoMicroRNAexpression profile in a large series of bladder tumors identifi-cation of a 3-miRNA signature associated with aggressivenessof muscle-invasive bladder cancerrdquo International Journal ofCancer vol 132 no 11 pp 2479ndash2491 2013

[64] H M Namloslashs L A Meza-Zepeda T Baroslashy et al ldquoModulationof the osteosarcoma expression phenotype by microRNAsrdquoPLoS ONE vol 7 no 10 Article ID e48086 2012

[65] P S Eis W Tam L Sun et al ldquoAccumulation of miR-155and BIC RNA in human B cell lymphomasrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 102 no 10 pp 3627ndash3632 2005

[66] Y Pan M Meng G Zhang H Han and Q Zhou ldquoOncogenicmicroRNAs in the genesis of leukemia and lymphomardquoCurrentPharmaceutical Design 2014

[67] M V Iorio M Ferracin C-G Liu et al ldquoMicroRNA geneexpression deregulation in human breast cancerrdquo CancerResearch vol 65 no 16 pp 7065ndash7070 2005

[68] Z Lu Y Ye D Jiao J Qiao S Cui and Z Liu ldquoMiR-155 andmiR-31 are differentially expressed in breast cancer patientsand are correlated with the estrogen receptor and progesteronereceptor statusrdquo Oncology Letters vol 4 no 5 pp 1027ndash10322012

[69] F Gao J Chang H Wang and G Zhang ldquoPotential diagnosticvalue ofmiR-155 in serum from lung adenocarcinoma patientsrdquoOncology Reports vol 31 no 1 pp 351ndash357 2014

[70] G Higgs and F Slack ldquoThe multiple roles of microRNA-155 inoncogenesisrdquo Journal of Clinical Bioinformatics vol 3 no 1 p17 2013

[71] H Fayyad-Kazan N BitarM Najar et al ldquoCirculatingmiR-150andmiR-342 in plasma are novel potential biomarkers for acutemyeloid leukemiardquo Journal of TranslationalMedicine vol 11 no1 article 31 2013

[72] M Yanlei P Zhang F Wang et al ldquomiR-150 as a potentialbiomarker associated with prognosis and therapeutic outcomein colorectal cancerrdquo Gut vol 61 no 10 pp 1447ndash1453 2012

[73] N K Jacob J V Cooley T N Yee et al ldquoIdentification of Sensi-tive SerummicroRNABiomarkers for Radiation BiodosimetryrdquoPLoS ONE vol 8 no 2 Article ID e57603 2013

[74] G J Zhang H Zhou H X Xiao Y Li and T Zhou ldquoMiR-378 isan independent prognostic factor and inhibits cell growth andinvasion in colorectal cancerrdquo BMC Cancer vol 14 no 1 p 1092014

[75] M Sand M Skrygan D Georgas et al ldquoMicroarray analysis ofmicroRNA expression in cutaneous squamous cell carcinomardquoJournal of Dermatological Science vol 68 no 3 pp 119ndash1262012

[76] S Hauser L M Wulfken S Holdenrieder et al ldquoAnalysisof serum microRNAs (miR-26a-2lowast miR-191 miR-337-3p andmiR-378) as potential biomarkers in renal cell carcinomardquoCancer Epidemiology vol 36 no 4 pp 391ndash394 2012

[77] L S Mangala Y Zhang Z He et al ldquoEffects of simulatedmicrogravity on expression profile of microRNA in humanlymphoblastoid cellsrdquo The Journal of Biological Chemistry vol286 no 37 pp 32483ndash32490 2011

[78] P Alexiou MMaragkakis G L Papadopoulos M Reczko andA G Hatzigeorgiou ldquoLost in translation an assessment andperspective for computational microrna target identificationrdquoBioinformatics vol 25 no 23 pp 3049ndash3055 2009

[79] J Nunez-Iglesias C-C Liu T E Morgan C E Finch and XJ Zhou ldquoJoint genome-wide profiling of miRNA and mRNAexpression in Alzheimers disease cortex reveals alteredmiRNAregulationrdquo PLoS ONE vol 5 no 2 Article ID e8898 2010

[80] LMa YHuangWZhu et al ldquoAn integrated analysis ofmiRNAand mRNA expressions in non-small cell lung cancersrdquo PLoSONE vol 6 no 10 Article ID e26502 2011

[81] J C Engelmann and R Spang ldquoA least angle regression modelfor the prediction of canonical and non-canonical miRNA-mRNA interactionsrdquo PLoS ONE vol 7 no 7 Article ID e406342012

[82] N Bossel Ben-Moshe R Avraham M Kedmi et al ldquoContext-specific microRNA analysis identification of functionalmicroRNAs and their mRNA targetsrdquo Nucleic Acids Researchvol 40 no 21 pp 10614ndash10627 2012


[83] S Artmann K Jung A Bleckmann and T Beiszligbarth ldquoDetec-tion of simultaneous group effects inmicroRNA expression andrelated target gene setsrdquo PLoS ONE vol 7 no 6 Article IDe38365 2012

[84] R N Germain ldquoMHC-dependent antigen processing andpeptide presentation providing ligands for T lymphocyte acti-vationrdquo Cell vol 76 no 2 pp 287ndash299 1994

[85] I V Konstantinova E N Antropova V I Legenkov and VD Zazhirey ldquoStudy of the reactivity of blood lymphoid cells increw members of Soyuz 6 7 and 8 before and after space flightrdquoKosmicheskaia Biologiia iMeditsina vol 7 no 6 pp 35ndash40 1973(Russian)

[86] A Cogoli and A Tschopp ldquoLymphocyte reactivity duringspaceflightrdquo Immunology Today vol 6 no 1 pp 1ndash4 1985

[87] N Gueguinou C Huin-Schohn M Bascove et al ldquoCouldspaceflight-associated immune system weakening preclude theexpansion of human presence beyond Earths orbitrdquo Journal ofLeukocyte Biology vol 86 no 5 pp 1027ndash1038 2009

[88] Y O Nunez J M Truitt G Gorini et al ldquoPositively correlatedmiRNA-mRNA regulatory networks in mouse frontal cortexduring early stages of alcohol dependencerdquo BMCGenomics vol14 p 725 2013

[89] R P Sullivan L A Fogel J W Leong et al ldquoMicroRNA-155 tunes both the threshold and extent of NK cell activationvia targeting of multiple signaling pathwaysrdquo The Journal ofImmunology vol 191 no 12 pp 5904ndash5913 2013

[90] F Gao Z-L Zhao W-T Zhao et al ldquoMiR-9 modulates theexpression of interferon-regulated genes and MHC class Imolecules in humannasopharyngeal carcinoma cellsrdquoBiochem-ical and Biophysical Research Communications vol 431 no 3pp 610ndash616 2013

[91] F Bazzoni M Rossato M Fabbri et al ldquoInduction andregulatory function of miR-9 in human monocytes and neu-trophils exposed to proinflammatory signalsrdquo Proceedings of theNational Academy of Sciences of theUnited States of America vol106 no 13 pp 5282ndash5287 2009

[92] S Thiele J Wittmann H-M Jack and A Pahl ldquomiR-9enhances IL-2 production in activated human CD4+ T cells byrepressing Blimp-1rdquo European Journal of Immunology vol 42no 8 pp 2100ndash2108 2012

[93] M Singh F Del carpio-Cano J Y Belcher et al ldquoFunctionalroles of osteoactivin in normal and disease processesrdquo CriticalReviews in Eukaryotic Gene Expression vol 20 no 4 pp 341ndash357 2010

[94] M Cogoli-Greuter M A Meloni L Sciola et al ldquoMovementsand interactions of leukocytes in microgravityrdquo Journal ofBiotechnology vol 47 no 2-3 pp 279ndash287 1996


[96] X Ma J Pietsch M Wehland et al ldquoDifferential gene expres-sion profile and altered cytokine secretion of thyroid cancer cellsin spacerdquoThe FASEB Journal vol 28 no 2 pp 813ndash835 2014

[97] D Chang H Xu Y Guo et al ldquoSimulated microgravity altersthe metastatic potential of a human lung adenocarcinoma celllinerdquo In Vitro Cellular and Developmental BiologymdashAnimal vol49 no 3 pp 170ndash177 2013

[98] S Hong X Li Y ZhaoQ Yang and B Kong ldquo53BP1 suppressestumor growth and promotes susceptibility to apoptosis ofovarian cancer cells through modulation of the Akt pathwayrdquoOncology Reports vol 27 no 4 pp 1251ndash1257 2012

Date post:	11-Sep-2021
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

How Microgravity Affects the Biology of Living Systems

Documents