october 2009 • Volume 2 • issue 2 … october 2009 • Volume 2 • issue 2 doi 10.3884/0002.2...

EuropEan JournaL of nanomEdicinEclinamfoundationforclinicalnanomedicine

EuropeanSocietyfornanomedicine

www.clinam.org october 2009 • Volume 2 • issue 2 doi 10.3884/0002.2 iSSn 1662-5986 (print) iSSn 1662-596x (online)

Emerging applications:Nanomedical approaches for Inner

Ear Disease

Advancing nanomedicine:The Importance of the Industrial-

Academic Interface for Innovation in the Pharmaceutical sector

Study nanoscience atthe Swiss NanoscienceInstitute, Basel.Students are invited to apply to study for a BSc in Nanoscience, or for aMasters degree by thesis. PhD positions are also available to youngresearchers who want to work at the frontiers of nanoscience andnanotechnology in the internationally renowned Swiss Nanoscience Institute.

In Basel, we combine basic sciencewith application-orientated research.Research projects can focus onnanoscale structures with the aim ofproviding new approaches to lifesciences, the sustainable use ofresources, and to the informationand communication disciplines.

The University of Basel alsocoordinates the Swiss NCCR networkof universities, federal researchinstitutes, industrial partners andthe Argovia-network (financed by theSwiss Canton of Aargau). Throughthese and other activities, theUniversity continues to reinforceits internationally acknowledgedposition as a centre of excellence innanoscale sciences.

See: www.nanoscience.ch, or contact Dr Katrein Spieler, [email protected] for further information on our nanosciences courses, and how to apply.

nano_now_issue9_aw1:issue 3 14/11/2008 16:01 Page 22

� EuropEanJournaLofnanomEdicinE2009Vol.2issue2

EuropEan JournaL of nanomEdicinEEditorial

Nanomedicine - the challenge of complexitypatrickHunziker,md,EditorinchiefandpresidentoftheEuropeanSocietyfornanomedicine

Thenanosciencesarecharacterizedbytwoapparentlyoppositeparadigms:thetop-downandthebottom-upapproach,eachcharacterizingahistoricrootof the entire field. The top-down perspective was evident in the refinement of microscopy to such a degreethatsinglemoleculesandatomscouldbevisualized and manipulated. Conversely, the bottom-up strategy manifested itself in the development of supramoleculardesigns.

formedicine,bothparadigmsareincreasinglyimportant: Detecting specifically a limited number of, or even single molecules represents the ultimate limitofdiagnosticmedicaltestswithallthepromisesbroughtbysizereduction:Speedup(nanomedicineforpoint-of-carediagnosis),minimizingsamples(nanomedicineasgentlemedicine),minimizingreagents (a cost-effective medicine) and consumables (nanomedicineasgreentechnology),andallowingmultiparametertesting(nanomedicineforcomprehensive assessment).

On the other side, the conventional pharmaceutical approacheswithsmallmoleculardrugsyieldfewernew drugs at strongly rising development costs. Thus, it is increasingly evident that advances in drug therapy will increasingly depend on novel, supramolecular therapeuticconcepts.

Suchsupramoleculartherapeuticdesignsrepresentcomplex systems - a significant step beyond conventional drug therapy, where “simple” drug molecules interacted with “complex systems”, namely thecellandtheentireorganism.

Theinteractionofman-made complex therapeutic systems interacting with the biologic complex systems,cellandorganism,isanewparadigmwithlarge promises and significant hurdles to overcome. An important promise is the capability to individually tailorsuchsystemstopatientsbasedondetailed

insightsintodiseasecharacteristics,leadingtopersonalized medicine, instead of relying on “one size fits all” blockbuster drugs.

The hurdles to overcome, however, are numerous: howtounderstand,rationallydesign,produce,analyse,manufacture,clinicallytest,andregisternewtherapies based on complex supramolecular systems? deeperunderstanding,thefundamentalbasisofallotheraspects,isapre-eminentacademicchallengeneededforsuccessfulimplementationofnanomedicinebyindustryandhealthprofessionalstothe benefit of the patient.

One important effort already being pursued is to develop a systematic classification, e.g. the efforts pioneered by Don Tomalia to create a “periodic table” of nanoobjects, which was stimulated byhisinsightsintothesystematicpropertiesofdendrimermolecules.Tofurtherstrengthentheunderstandingofthefoundationsofnanomedicine,the broadly interdisciplinary field of nanomedicine, alreadypopulatedbyphysicists,chemists,pharmacologists,biologistsandphysiciansneedstobefurtherbroadened:wealsowillneedatheoryofnanomedicine,mathematicalmodels,computersimulations, a “systems nanomedicine” approach.

I expect that such a comprehensive approach combining experimental research, theoretical research,andcomputationalsimulationwillcosttheinitial effort needed in all interdisciplinary endeavours, namely to find together as diverse specialists with a different mentality and talking a different language; itwillneedfundingopportunitiestoperformsolidresearch beyond individual small projects, and it will be highly rewarded, first intellectually, then in the capability to more rationally design novel therapies, and finally and most importantly by a better capability to help our patients in a effective, gentle and individualized manner.

EuropEanJournaLofnanomEdicinE2009Vol.2issue2 �

Editorial

From the 2nd to the 3rd European Conference for Clinical Nanomedicine in May 2010Beat Löffler, MA, CEO of the European Foundation for Clinical Nanomedicine

WelcometothethirdEuropeanJournalofnanomedicine!

The2ndEuropeanconferenceforclinicalnanomedicinecombinedastrongclinicalfocuswithleading-edgetechnology and science and proved to beanewprimaryplatformforproblemandsolution-orienteddiscussionsbetweenphysicians,scientistsandengineers.morethanhalfofthe275participantsofthe2009meetingwereclinicians.participantsfrom��countriesaroundtheworlddiscussedtheapplicationof the nanosciences to the benefit of the patient. With thissecondconferencewebecamethemeetingpointforallthosefocusedonclinicalnanomedicineandtheyear-to-year development of this discipline towards medicalapplications.andwesucceededinstaginga debate conference that offered more than 7 hours of discussion time at a highly interdisciplinary level. We received so much positive feedback that we are encouragedtogoonestepfurther:

atthe�rdEuropeanconferenceforclinicalnanomedicinefrommay9-12,2010therewillbeaFoyer Exhibition and a University Village.

The Foyer Exhibition will highlight the existing tools, instruments and materials in the field of regenerative medicine, diagnostics, targeted drug delivery systems, novel materials for nanoparticles, imaging, biomaterials/biosensors/biomarkers,nanomedicine-relatedmedicaltoolsandmedicalinstrumentation,aswell as existing nano-based clinical medicines.

The University Village is the space for universities and research institutes, providing an opportunity for presenting novel approaches, new research projects and initial outcomes of research and experimental results. Together with the industrial exhibitors, this University Village will make the CLINAM 2010 meeting anoutstandinggathering,wherethestateoftheart in Nanomedicine will be much in evidence. The

University Village is a platform to show not only ideas, butalsopatentedearlysolutions,becauseyoumightencounter other experts from science and industry at CLINAM who are looking for just your kind of patentedideaandproduct.

As in all novel emerging technology fields, the massive investments in scientific and technical researchinnanomedicinearenotyetyieldingthereturnsthattheywilloneday.Theformulaforsuccessful innovation and the creation of a flourishing market with true benefit is triple layered: The Basel Conference discusses the novel possibilities, tools and innovations to solve many of todays’ medical problems to the benefit of the patient. The Exhibition willshowthetoolsandapplicationsthatareemergingfrom Industry and manufacturers. The Universities will showthestateofresearchandsolutionsemergingfrombasicresearch.

There are many reasons to invite you to come to Basel inmay2010:•meetattheestablishedinterdisciplinarymeetingonclinicalnanomedicine•attendthemeetingtofurthertheaccelerationofdevelopments in Nanomedicine • Use Europe’s largest platform for debate between clinical researchers, scientists, engineers and experts frommedicine,physics,chemistry,pharmacy,biology,biochemistryandmaterialsscience.•Seethemilestonesonthepathwayofnanomedicinepresented by companies and universities in the CLINAM Foyer Exhibition and University Village. •discussthepromiseandlimitsandimplicationsofnanomedicaltools,techniques,andmaterialsinthecontext of unsolved medical problems.• Investigate the clinical evidence from established nanodrugsandlearnaboutthelatestandongoingtrials.

findthepreliminaryprogrammeandcallforpapersunder:www.clinam.org

Editorial:

nanomedicine - the challenge of complexitypatrickHunziker,md,Editorinchiefandpresident

oftheEuropeanSocietyfornanomedicine 3

contents

Medical Nanomaterials:

nanoscale characterization of Biological and Mechanical Profile of Carbon Stent nanocoatingsKaragkiozakiV.,LogothetidisS., KassavetisS.,LousinianS.14

Implementing Nanomedicine:The importance of the industrial - academic interface for innovation in the pharmaceutical SectorEatonm. 22

Preclinical Nanomedicine:

internalization of liposome nanoparticles functionalized with TrkB ligand in rat cochlear cell populationsZou J. , Zhang Y., Zhang W., Ranjan S., Sood R. , Mikhailov A., Kinnunenp.,pyykköi. 7

5 EuropEanJournaLofnanomEdicinE2009Vol.2issue2

Editorial:

from the 2the 2nd to the 3rd European conference for clinical nanomedicine in may 2010 Beat Löffler, MA , CEO of the European Foundation for clinicalnanomedicine 4

Contents

Review paper:

Cochlear Implants and Inner Ear Based TherapyCiorba A., Astolfi L., Jolly C., Martini A. 25

EuropEAn JournAL of nAnoMEdiCinE 2009 Vol. 2 issue 2 �

Preclinical Nanomedicine:MRI of the Cochlea with Superparamagnetic Iron Oxide Nanoparticles Compared to Gadolinium Chelate Contrast Agents in a Rat ModelPoe d., Zou J., Zhang W., Qin J., ramadan u., fornara A., Mamoun M., pyykkö i. 29

Nanomethods:The Role of TOF-SIMS in NanomedicineKeller B., Mayerhofer K. 37

Nano Imaging Technologies: Polymer vesicles loaded with precipitated gadolinium nanoparticles: A novel target- specific contrast agent for magnetic reso-nance imagingBroz p., nadav B., Santini f., Marsch S., Scheffler K., Meier W. , Hunziker p. 43

Impressum 49

1department of otolaryngology, univeristy of Tampere, Medical School, Tampere, finland2Helsinki Biophysics and Biomembrane Group, Medical Biochemistry, institute of Biomedicine, university of Helsinki, Helsinki, finland†These authors contributed equally to the work.*Correspondence to: dr. J. Zou, department of otolaryngology, university of Tampere, fM1, 3rd floor, Biokatu �, 33520 Tampere, finland; phone: +358 3 311�4129; fax: +358 3 35517700; Email: [email protected]

internalization of liposome nanoparticles functionalized with TrkB ligand in rat cochlear cell populationsJing Zou1*, Ya Zhang1†, Weikai Zhang1†,Sanjeev ranjan2†, rohit Sood2, Andrey Mikhailov1, paavo Kinnunen2,

ilmari pyykkö1(doi 10.3884�0002.2.3�doi 10.3884�0002.2.3�

AbstractObjectives: To investigate the targetability of TrkB ligand-functionalized liposome nanoparticles for gene delivery to spiral ganglion cells. Materials and methods: A TrkB affinity peptide was synthesized and coupled to liposome nanoparticles carrying the plasmid pGeneClipTM hMGfp encoding shrnA to transiently silence inhibitor of differentiation and dnA binding-2 (id2� along with the reporter gene EGfp. internalization and targetability were analyzed in primary cochlear cell culture, cochlear explants, and live rats. Gene transduction was evaluated in rat cochlear explants. immunofluorescent staining in combination with confocal microscopy was used for observation. Results: Efficient internalization was observed in primary cochlear cell culture for both peptide-functionalized liposome nanoparticles and blank liposome nanoparticles in a concentration-dependant manner. Both particles showed uptake in spiral ganglion cells and adjacent nerve fibers. potential targetability with TrkB affinity peptide-functionalized liposome nanoparticles was observed in the adult rat cochlea. More efficient gene expression was seen for the peptide-functionalized liposome nanoparticles, and the function of the shrnA was demonstrated in cochlear explants and adult rat cochleae. Conclusions: potential targetability of A371-functionalized liposome nanoparticles was observed in the adult rat cochlea. functionalization of liposome nanoparticles with TrkB ligand did not change cellular internalization, but it did enhance gene expression.AbbreviationsAfu, adaptive focused ultrasound; Bdnf: brain derived neurotrophic factor; BSA: bovine serum albumin; dLS, dynamic light scattering; dSpE-pEG-2000, 1, 2-distearoyl-sn-glycero-3-phosphoethanolamine-n-[methoxy(polyethylene glycol�-2000] (ammonium salt�; dSpE-pEG(2000�maleimide, 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-n-[maleimide(polyethylene glycol�2000] (ammonium Salt�; EdTA, ethylenediaminetetraacetic acid; nGf: nerve growth factor; Sph, Sphingosine; EggpC, egg phosphatidylcholine; Hepes, n-2-hydroxyethylpiperazine-n-2-ethanesulfonic acid; pBS: phosphate buffered saline; pBS-T: pBS-tween-20; pdi, polydispersity index; pdnA, plasmid dnA; Tritc-dHpE, n-(�-tetramethylrhodaminethiocarbamoyl�-1,2-dihexadecanoyl-sn-glycero-3-phosphoethanolamine, triethylammonium salt; TrkB: tyrosine kinase receptor B; Trifluoroacetic acid (TfA�; Zav, apparent hydrodynamic particle diameter. Zou, J nanomedicine 2009, 2.2:7-13

Key words: animal, cochlea, gene delivery, gene silence, nanocarrier, peptide, Targeting

7 EuropEAn JournAL of nAnoMEdiCinE 2009 Vol. 2 issue 2

IntroductionTargeted drug delivery is an efficient means of therapy due to improved biodis-tribution and minimized side effects. Lipo-some nanoparticles functionalized with different ligands or antibodies have been broadly applied for targeted treatment of cancer, inflammation, and cardiovascular diseases, among others. (1-7� deafness is a major disease causing disability and lacks a curative treatment using conven-tional drugs. Cochlear implants effective-ly replace the mechanosensory transduc-tion function of lost hair cells and furnish the user with a substantial hearing bene-

fit by exerting direct electrical stimulation on spiral ganglion cells or their dendrites (peripheral processes�. Although there are controversial clinical reports on the contribution of spiral ganglion cells to the efficacy of cochlear implants, at least 10%

survival of spiral ganglion cells is needed for current cochlear implants to succeed in helping patients communicate. in or-der to broaden the application of cochlear implants, strategies aimed at preserving or regenerating deafferented spiral gan-

preclinical nanomedicine

EuropEAn JournAL of nAnoMEdiCinE 2009 Vol. 2 issue 2 8

glion cells in combination with cochlear implants are key for further advances in cochlear implant technology. Targeted delivery of agents to spiral ganglion cells using functionalized liposome nanoparti-cles will be the most efficient way to fulfill this goal.The neurotrophin receptor tropomyosin-related kinase (Trk� receptor tyrosine ki-nase, especially TrkB, is expressed in spi-ral ganglion cells. (8� peptides that bind TrkB have been developed using either phage display or molecular modeling with the aim of mimicking the biological func-tions of brain derived neurotrophic factor (Bdnf� and nerve growth factor (nGf�. (9-11� The sequence CTfVKALTMdG-KQAAWr, denoted in our work as A371, is derived from the peptide hngf_EE, which has been shown to bind the TrkB receptor and is in turn a modification of the natural sequence of amino acids 84-100 of nGf. (11� The sequence was extended with an n-terminal cysteine for coupling to lipo-some nanoparticles. Liposome nanopar-ticles with a payload of plasmid dnA were pegylated (dSpE-pEG-2000� and conjugated to A371 peptide. This study was aimed at evaluating the impact of target peptide functionalization on inter-nalization of nanoparticles in cochlear cell populations. observations were made in primary cochlear cell culture, cochlear ex-plants, and in vivo.

Materials and methodsMaterials for liposome nanoparticle manufacturingSph, EggpC, dSpE-pEG-2000, and dSpE-pEG (2000� maleimide were from Avanti polar lipids (Alabaster, AL�. HpLC-grade trifluoroacetic acid (TfA� was from fluka (Buchs, Switzerland�, and acetonitrile was from rathburn (Walker Burn, Scotland, uK�. The purity of the lipids was checked by thin layer chromatography on silicic acid-coated plates (Merck, darmstadt, Germany� developed with a chloroform�methanol�water mixture (�5:25:4, v�v�v�, with examination of the plates after iodine staining and, when appropriate, with uV illumination revealing no impuri-ties. Lipid concentration was determined gravimetrically with a high precision elec-trobalance (Cahn, Cerritos, CA�. Hepes and EdTA were from Sigma-Aldrich, and Tritc-dHpE was from Molecular probes (Eugene, or�. The concentration of the fluorescent lipid analogue Tritc-dHpE was determined spectrophotometrically using molar absorptivity ε4�5 = 19000 M-1cm-1(in

C2H5oH�. The other chemicals were of an-alytical grade and from standard sources. All experiments were conducted in 5 mM Hepes and 0.1 mM EdTA, pH 7.0, unless otherwise indicated.

Peptide manufacturingpeptides were synthesized by fMoC technology (in Storkbio former inbiolabs, Tallin, Estonia� and purified by HpLC to 90% purity, and the sequences were con-firmed by mass spectrometry. Some of the synthesized peptides were n-termi-nally conjugated to fiTC and used for de-termination of affinity and specificity. for conjugation to nanoparticles, only native peptides were used. in order to test speci-ficity and affinity, two cell lines were se-lected: rAW 2�4 heterologously express-ing the TrkB receptor and the K5�2 line shown to be TrkB-negative but positive for the closely related TrkA receptor. (12, 13� fiTC-conjugated peptides were added to a suspension of rAW 2�4 and K5�2 cells at a concentration of 0.1 µg per 10�cells, incubated for 20 min at room tem-perature, washed four times with pBS, and fixed in 4% paraformaldehyde in pBS. fixed cells were measured using a fAC-SAria (Becton dickenson� flow cytometer. The analysis showed that the sequence CrAniGGTHA had the strongest reactivi-ty, while CTfVKALTMdGKQAAWr (A371� had higher selectivity for TrkB relative to TrkA (Tab. 1�.

Plasmid DNA preparationplasmid pGeneClipTM hMGfp, encoding an shrnA to transiently silence inhibi-tor of differentiation and dnA binding-2 (id2� and the reporter gene EGfp (Super-Array, Bioscience Corp., frederick, Md, uSA�, was propagated in oneShot Top10 Competent Cells (invitrogen, Carlsbad, CA, uSA� and then extracted using the pureLinkTM plasmid dnA Megaprep kit (invitrogen, Carlsbad, CA, uSA� according to the manufacturer’s instructions. The

Sequence Signal from rAW2�4cells (mean ± CV�

Signal from K5�2 cells(mean ± CV�

Signal ratio

CSMAHpYfAr (3� A3�� 34.04 ± 109.55 2187.�1 ± 25�.43 3.03CrAniGGTHA (3� A3�8 8543.85 ± 72.�3 2959.14 ± 93.81 2.89CSpGSiHTLV (3� A370 5708.78 ± 128.13 2302.21 ± 300.�7 2.48CTfVKALTMdGKQAAWr (4� A371 4�18.2 ± 105.09 1417.59 ± 188.35 3.2�

Table 1. flow cytometry showing binding of peptide candidates to TrkB positive (rAW2�4� and TrkB negative (K5�2� cell line.

note: Every peptide binds to both TrkB and TrkA, however A371 had higher selectivity for TrkBote: Every peptide binds to both TrkB and TrkA, however A371 had higher selectivity for TrkBA371 had higher selectivity for TrkB higher selectivity for TrkBhigher selectivity for TrkB selectivity for TrkBselectivity for TrkB relative to TrkA according to signal ratio. according to signal ratio.

purity and concentration of pdnA id2.3 were determined by measuring the ra-tio of absorbance at 2�0 nm�280 nm (ε = ��00 l�mol x cm�.

Coupling of peptide A371 to DSPE-PEG (2000) maleimidedSpE-pEG (2000� maleimide and A371 peptides were incubated at a 1:1.3 molar ratio in a reaction mixture with 100 mM hepes and methanol in a 1:1.11 molar ratio, pH 7.0, for 3 hours at room tem-perature under constant stirring. Lipid-peptide conjugate was purified by HpLC on a reverse phase column (Jupiter 5µm C4 300A ST 4.��150, Amersham Biosci-ences, uppsala, Sweden� and eluted with a linear gradient from 0 to 100% aceto-nitrile in water with 0.1% TfA at a flow rate of 1 mL�min. Samples were moni-tored at a uV absorbance of 20� nm and 280 nm. fractions corresponding to pure lipid-peptide conjugate were collected and lyophilized. The concentration of lipid-peptide conjugate was determined spectrophotometrically by measuring tryptophan fluorescence. Lipid-peptide conjugate was dissolved in methanol pri-or to the experiments.

Dynamic light scattering The apparent hydrodynamic particle di-ameter (Zav� and polydispersity index (pdi� of lipid vesicles were determined by dynamic light scattering at 25°C (Zeta-sizer nano ZS, Malvern instruments Ltd., uK�. The instrument uses photon corre-lation spectroscopy at a scattering angle of 173° to evaluate Zav from the diffusion coefficient (d� using the Stokes-Einstein equation, dZ = kT�3πηd, where k is the Boltzmann constant, T is the absolute temperature, and η is the viscosity of the solvent. The Zav analysis gives two values, a mean value for the size and a width parameter known as the polydispersity index. Zav is also known as the cumulants mean. The cumulants analysis is the fit of


� EuropEAn JournAL of nAnoMEdiCinE 2009 Vol. 2 issue 2

a polynomial to the log of the G1 correla-tion function

Ln[G1] = a + bt + ct2 + dt3 + et4 + ……………

The coefficient of the squared term, c, when scaled as 2c�b2, represents the poly-dispersity index.

Preparation of liposome nanoparticleThe appropriate amounts of the lipid stock solutions in chloroform and the lipid peptide conjugate were mixed in organic solvent to obtain the desired liposome compositions. in brief, lipid film was pre-pared from Sph, eggpC, dSpE-pEG-2000, lipid-peptide conjugate and Tritc-dHpE (0.5:0.44:0.02: 0.01, 0.03 molar ratios�. Solvents were removed under a stream of nitrogen, and the lipid residues were subsequently maintained under reduced pressure for at least 2 h. The dry lipid film was then hydrated at �0°C for one hour in 5 mM Hepes, 0.1 mM EdTA, pH 7.0. The lipid mixture at final concentration of 0.1 mM was subjected to focused ultrasound (Covaris, KBiosciences, uK�. An average particle diameter (Zav� of 182 and 89 nm for particles with and without targeting peptide (A371�, respectively, was deter-mined by dynamic light scattering (Zeta-sizer, nano ZS, Malvern instruments Ltd., uK�.

Preparation of lipoplexesAfter being subjected to focused ultra-sound, liposome nanoparticles were mixed with plasmid dnA at a lipid�dnA charge ratio (+�-� of 1.2:1 to obtain na-noscale particles, also termed as lipo-plexes. After the addition of pdnA, a Zavof 108 and 112 nm was obtained for lipo-plexes with and without targeting pep-tide A371, respectively three batches of liposome nanoparticles with and without A371 peptide conjugation, with and with-out plasmid dnA encapsulation were manufactured.

Primary cochlear cell culturefive p1-p5 pups were decapitated after deep anesthetization and sterilized with 70% ethanol. The cochleae were isolated and cut into small pieces, then dissociat-ed with a pBS-based solution containing elastase (1 mg�ml, Sigma Aldrich, uSA�, collagenase type i (1 mg�ml, Sigma Al-drich, uSA�, and trypsin (0.5 mg�ml, Sig-ma Aldrich, uSA� for 35 min at 37°C, fol-lowed by trituration (every four cochleae were dissociated in 0.� ml of solution�.

The digestion was terminated by adding 1 ml dMEM-f12 (Sigma Aldrich, uSA�, con-taining 10% fetal bovine serum (Sigma Aldrich, uSA�. After centrifugation for 5 min at 250 x g, the cell pellets were resus-pended in 1 ml defined medium (dMEM-f12 with B27 supplement, 1 mM n-acetyl-L-cysteine, penicillin–streptomycin, and 20 ng�ml EGf (Sigma Aldrich, uSA�� and plated on the 4-well Lab-Tek®ii Cham-ber SlideTM System (nalge nunc interna-tional, naperville, uSA� containing 1.0 ml defined medium�well. The cells were cultured at 37ºC in the Co2 incubator overnight, then treated with lipoplexes. A371-functionalized liposome nanopar-ticles (final concentrations: 10 μM, 1.0 μM, and 0.1 μM� and blank liposome nanopar-ticles (final concentrations: 10 μM, 1.0 μM, 0.1 μM, and 0.01 μM� without plas-mid dnA encapsulation were prepared with defined medium, and 1.0 ml of each was applied to the cells for 2 hours and 24 hours under the same incubation condi-tions. After washing with pBS for 3x1 min, cells were fixed with 4% paraformalde-hyde for 30 minutes. The cell cytoplasm was counter-stained with 50 μg�ml fiTC-labeled phalloidin (Sigma Aldrich, uSA� for 30 min, and nuclei were stained with 4’,�-diamidino-2-phenylindole (dApi� (10 ng�ml, Sigma-Aldrich, uSA� for 10 min. Cells were then mounted with fluoro-mount for confocal microscopy (3x5 min pBS washes were applied between each staining step�.

Cochlear explants studyfive p1-p� pups were decapitated after deep anesthetization and sterilized with 70% ethanol. Each cochlea was isolated, cut into 3-4 pieces, and plated on the 4-well Lab-Tek®ii Chamber SlideTM System (nalge nunc international, naperville, uSA� containing 1.0 ml defined medium�well, just as for the primary cochlear cell cultures. The explants were cultured at 37ºC in the Co2 incubator overnight and then treated with liposome nanoparticles carrying plasmid pGeneClipTM hMGfp dnA at concentrations of 1 μM and 0.2 μM for different time points (15 min, 30 min, �0 min, and 120 min, 2 d, and 4 d�. At the end of incubation, explants were washed with pBS 3x3 min and fixed in 4% paraformaldehyde for 30 minutes. After washing with pBS, samples at the 2 d and 4 d time points were counter-stained with dApi (10 ng�ml� for 10 min and mounted with fluoromount for confocal micros-copy. other samples were used for neuro-

filament staining. After washing with pBS, explants were incubated with 0.1% Triton X-100 for 15 min. After washing with pBS, explants were then incubated with pre-in-oculated goat serum (1:20� for 30 min, fol-lowed by rabbit anti-neurofilament 200 antibody (diluted at 1:100 with pBS con-taining 0.1% bovine serum albumin (BSA�, Sigma Aldrich, uSA� overnight. Explants were washed with pBS-tween-20 (pBS-T� for 3x2 min, incubated with fiTC-con-jugated goat anti-rabbit igG (diluted at 1:400 with 0.1% BSA-pBS, Sigma Aldrich, uSA� for �0 min, incubated with dApi (10 ng�ml� for 10 min, washed with pBS-T for 3x2 min, and mounted with fluoromount for confocal microscopy. for a negative control, the primary antibody was re-placed with 0.1% BSA-pBS.

In vivo observationfive male Sprague-dawley rats, 3-10 month old, weighing for 400-750 g, with normal pryer’s reflex (supplied by the experimental animal unit, university of Tampere�, were used in the study in accor-dance with the standards of the local eth-ics committee of the university of Tam-pere (permission no: 985�2003�. All animal experiments were approved by the Ethical Committee of the university of Tampere. Animal care and experimental procedures were conducted in accordance with Euro-pean legislation. for round window ad-ministration of lipoplexes, animals were under general anesthesia with domitor (0.5 mg�kg medetomidine hydrochloride, orion pharma, finland� and Ketalar (75 mg�kg ketamine, pfiZEr AB, finland� giv-en intraperitoneally. The operation was performed under sterile conditions. After local analgesia with lidocaine, a retro-au-ricular incision was used to expose the left bulla. A hole was drilled on the bulla with a 2 mm diameter burr. After visualizing the stapes artery, the round window mem-brane was identified above the artery. A small piece of gelfoam (around 8 mm3� was saturated with liposome nanopar-ticles at a concentration of 100 μM and placed on the round window membrane for 24 hours. Atipamezole hydrochloride (2 mg�kg� was injected i.p. immediately after the operation to accelerate recovery from anesthesia. Saline (2 ml� was admin-istered through subcutaneous injection in the neck. L-polamivet (0.4 ml�kg� was injected b.i.d. to relieve pain.following i.p. injection of pentobarbital (�0 mg�kg�, cochleae were fixed using cardiac perfusion with 4% paraformal-



dehyde, and bulla were removed and further fixed for 1 h. The cochleae were thoroughly washed with tap water for 30 seconds, then opened by breaking the bony wall under a stereomicroscope and washed again with pBS for 2x5 min. The bulla were counter-stained with fiTC-la-beled phalloidin (50 µg�ml, Sigma Aldrich, uSA� for 40 min and then with dApi (10 µg�ml� for 10 min. After washing with pBS for 3x5 min, the round window membrane, lateral wall and modiolus together with basilar membrane were taken under a stereomicroscope, placed on glass slides, and mounted with fluoromount for con-focal microscopy.

Confocal microscopyThe samples were observed under an olympus iX70 microscope with Andor iQ installed. The excitation filters were 488 nm (blue excitation� and 5�8 nm (green excitation�, using an Ar-Kr laser as the excitation source. The corresponding emission filters were 525�50 (fiTC� and �07�45 (TriTC�. dApi was excited with a 340-380 nm filter and detected using a 500 Lp filter. Signal intensity was ana-lyzed with imageJ 1.32j software.

Statistical analysisThe signal intensity was corrected by the intensity of a randomly selected back-ground region. The corrected mean val-ues were analyzed with the SpSS 11.5 program. differences in the levels of sig-nal intensity among concentrations and cell populations were analyzed by Ano-VA (p<0.05 was accepted as an indication of statistical significance�. differences in levels of signal intensity between A371-functionalized liposome nanoparticles and blank liposome nanoparticles were analyzed using a t-test (p<0.05 was ac-cepted as an indication of statistical sig-nificance�.

Resultsinternalization of liposome nanoparticles in primary cochlear cellsTwo batches of liposome nanoparticles without plasmid dnA encapsulation were tested in primary cochlear cell cul-tures. one batch showed aggregation of nps but no cellular internalization. The other batch showed efficient internaliza-tion of both peptide functionalized lipo-some nanoparticles and blank liposome nanoparticles, which were observed in the primary cochlear cells, including spiral ganglion cells, fibrocytes, and intermedi-

ate cells of the stria vascularis. Cytosolic localization, perinuclear localization, and nuclear localization of liposome nanopar-ticles were seen in different cell types. The nanoplex signal intensity was significant-ly dependent on the nanoplex concentra-tion in the medium (p<0.01�. no statisti-cally significant difference was observed, however, between A371-functionalized liposome nanoparticles and blank lipo-some nanoparticles (p>0.5� (fig. 1�.

Figure 1. Confocal microscopy showing the in-ternalization of liposome nanoparticles in pri-mary rat cochlear cell cultures. for A371-func-tionalized liposome nanoparticles without plasmid dnA, efficient uptake was observed in the cytoplasmic and perinuclear regions of type i (A� and type ii (B� spiral ganglion cells and stria intermediate cells (C� at a concen-tration of 1.0 μM. Sparse uptake was also de-tected in fibrocytes at a concentration of 0.1 μM (d�. for the blank liposome nanoparticles without plasmid dnA, efficient internalization occurred in spiral ganglion cells (E� and fibro-cytes (f� at a concentration of 1.0 μM. Sparse internalization was observed in fibrocytes at a concentration of 0.1 μM (G�, but no uptake was detected in spiral ganglion cells at a con-centration of 0.01 μM (H�. Concentration-de-pendant internalization was observed in both A371-functionalized liposome nanoparticles (i� and blank liposome nanoparticles (J�. red: li-posome nanoparticles; green: f-actin stained with fiTC-conjugated phalloidin; blue: nuclei stained with dApi. A371-np: A371-functiona-lized liposome nanoparticles. Scale bar=10 μm.

Nanoplex internalization and possible EGFP expression in cochlear explantsLiposome nanoparticles carrying plasmid pGeneClipTM hMGfp dnA was tested in cochlear explants. dynamic uptake of liposome nanoparticles in both neuro-filaments and spiral ganglion cells was observed, with liposome nanoparticles accumulating in the spiral ganglion satel-lite cells and gradually appearing on the neurofilaments and in the spiral ganglion cells. Abundant distribution of liposome nanoparticles with and without functional peptides was seen in the neurofilaments (fig. 2�.

figure 2. Confocal microscopy showing the internalization of liposome nanoparticles in rat cochlear explants. A371-functionalized liposome nanoparticles carrying plasmid pGeneClipTM hMGfp dnA appeared adjacent to neurofilaments at 1 h post-treatment (A�, abundantly attached to neurofilaments at 2 h post treatment (B�, and distributed in the cytoplasm and nucleus of type i spiral ganglion cells at 2 h post-treatment (C�. Blank liposome nanoparticles carrying plasmid pGeneClipTMhMGfp dnA were detected in spiral ganglion satellite cells adjacent to neurofilaments (d�, attached to neurofilaments (E� at 1 h post treatment, greatly accumulated on neurofilament (f�, and within spiral ganglion cells (G� at 2 h post treatment. red: liposome nanoparticles; green: neurofilaments; blue: nuclei stained with dApi. A371-np: A371-functionalized liposome nanoparticles. Scale bar=10 μm.

Potentially aggregated EGfp expression was seen in the explants on day 2 post-treatment with A371-functionalized lipo-some nanoparticles carrying pGeneClipTMhMGfp plasmid dnA encoding shrnA to transiently silence id2. More EGfp expression was detected on day 4 post-treatment. only sparse EGfp expres-sion was detected in the explants on day



4 post-treatment with blank liposome nanoparticles carrying the same plasmid dnA as the A371-functionalized liposome nanoparticles. in general, EGfp expres-sion was inefficient (fig. 3�.

Figure 3. Confocal microscopy showing EGfp expression in rat cochlear explants. Two days post-gene delivery with A371-functionalized liposome nanoparticles carrying plasmid pGeneClipTM hMGfp dnA, dot-like EGfp expression was detected in the cell (B�, while nanoparticles started to disassemble (A� (C: merged image of A and B�. faint EGfp expression was observed in cells transduced with blank liposome nanoparticles carrying plasmid pGeneClipTM hMGfp dnA (E�, although the nanoparticles showed disassembly (d� (f: merged image of d and E�. four days after gene delivery with A371-functionalized liposome nanoparticles, greater dot-like EGfp expression was seen in cells (H�, while the nanoparticles were disassembling (G� (i: nuclear staining with dApi; J: merged image of G, H, and i�. EGfp expression was still sparse (L�, although abundant blank liposome nanoparticles were internalized by the cells (K, M, and n� at 4 days post-treatment. A371-np: A371-functionalized liposome nanoparticles; Scale bar=10 μm.

Nanoplex distribution and possible EGFP expression in cochlear cell popu-lations after round window membrane permeationin the adult rat cochleae receiving round window membrane permeation with A371-functionalized liposome nanopar-ticles, greater particle distribution was observed in the spiral ganglion region than in the cochleae treated with blank liposome nanoparticles. This difference was not statistically significant, however, possibly due to the small sample size (p>0.05�. dot-like EGfp expression was detected sparsely in the spiral ganglion cells and spiral ganglion satellite cells with nanoplex internalization (figs. 4 and 7�. in the inner hair cell region, there was significantly greater uptake of functional-

ized liposome nanoparticles than blank liposome nanoparticles (p<0.05� (figs. 5 and 7�. nanoplex uptake was also ob-served in the lateral wall, including the spiral ligament and stria vascularis, but was not significantly different between the functionalized and blank liposome nanoparticles (p>0.05� (figs. � and 7�. no uptake was detected in the outer hair cell region in cochleae treated with either A371-functionalized liposome nanopar-ticles or blank liposome nanoparticles (fig. 5�.

Figure 4. Confocal microscopy showing the distribution of liposome nanoparticles in the spiral ganglion region of the rat cochlea at 24 h post-round window membrane permeation. Abundant A371-functionalized liposome nanoparticles carrying plasmid pGeneClipTMhMGfp dnA (A� and blank liposome nanoparticles carrying plasmid pGeneClipTMhMGfp dnA (B� appeared in the spiral ganglion cells. Bright dot-like green fluorescence appeared in SGCs, showing uptake of A371-functionlizaed liposome nanoparticles (arrow in A�, and weaker dot-like green fluorescence appeared in the spiral satellite cells that was suspected to be EGfp expression, showing internalization of blank liposome nanoparticles (arrow in B�. red autofluorescence was also detected in SGCs of untreated cochleae (C�. red: liposome nanoparticles; green: f-actin stained with fiTC-conjugated phalloidin; blue: nuclei stained with dApi. A371-np: A371-functionalized liposome nanoparticles; nC: untreated control; SC: satellite cell; SGC: spiral ganglion cell. Scale bar=10 μm.

Figure 5. Confocal microscopy showing the distribution of liposome nanoparticles in the hair cell region of the rat cochlea at 24 h post-round window membrane permeation. After treatment with A371-functionalized liposome nanoparticles carrying plasmid pGeneClipTM hMGfp dnA, abundant liposome nanoparticles were detected in the inner hair cells and adjacent supporting cells (A, B�, while few liposome nanoparticles were found in the pillar cells (A� and outer hair cells (C�. After treatment with blank liposome nanoparticles carrying plasmid pGeneClipTM hMGfp dnA, fewer liposome nanoparticles appeared in the outer hair cells, pillar cells (d�, and inner hair cells (E�. in the untreated controls, faint red autofluorescence was found in the inner hair cells (f�. red: liposome nanoparticles; green: f-actin stained with fiTC-conjugated phalloidin; blue: nuclei stained with dApi. A371-np: A371-functionalized liposome nanoparticles; BC: border cells; iHC: inner hair cell; oHC: outer hair cell; pC: pillar cell. Scale bar=10 μm.

Figure 6. Confocal microscopy showing the distribution of liposome nanoparticles in the lateral wall of the rat cochlea at 24 h post-round window membrane permeation. After treatment with A371-functionalized liposome nanoparticles carrying plasmid pGeneClipTM hMGfp dnA, aggregated liposome nanopar-ticles were detected in spiral ligament fibro-cytes (A� and stria vascularis (B�. After treat-ment with blank liposome nanoparticles carry-ing plasmid pGeneClipTM hMGfp dnA, small-er dots of liposome nanoparticles appeared in the spiral ligament fibrocytes (C�. red: lipo-some nanoparticles; green: f-actin stained with fiTC-conjugated phalloidin; blue: nuclei stained with dApi. A371-np: A371-function-alized liposome nanoparticles; SL: spiral liga-ment; StrV: stria vascularis. Scale bar=10 μm.




Discussionin primary cochlear cell culture and ex-plants, no cell type specific internalization of A371-functionalized liposome nanopar-ticles was observed. The uptake of blank liposome nanoparticles in the cochlear cells was as efficient as that of function-alized liposome nanoparticles. This indi-cates that the TrkB receptor pathway is not involved in the internalization of lipo-some nanoparticles in spiral ganglion cells, although TrkB internalization occurs upon binding to Bdnf. There are two possible explanations for the different behavior of A371 peptide-functionalized liposome nanoparticles and Bdnf in spiral ganglion cells. first, as a modification of the natural sequence of amino acids 84-100 from the structure of nGf, A371 does not have the full functionality of nGf or Bdnf 11. Sec-ond, coupling of the peptide to liposome nanoparticles significantly increases the size of A371, making it much larger than Bdnf and nGf. The large size of the A371-coupled liposome nanoparticles may prevent the internalization of TrkB upon binding. Since they were equally exposed to the liposome nanoparticles, every cell population displayed the same amount of internalization.

Figure 7.Quantification of the nanoplex dis-tribution in different cell populations of adult rat cochleae at 24 h post-round window mem-brane permeation. There was a significantly greater distribution of A371-functionalized li-posome nanoparticles compared to blank li-posome nanoparticles in the inner hair cells. There was also a tendency toward enhanced distribution of A371-functionalized liposome nanoparticles compared to blank liposome nanoparticles in the spiral ganglion cells. There was no difference in the nanoplex distribution in the spiral ligament of the lateral wall. iHC: inner hair cell; SGC: spiral ganglion cell; SL: spiral ligament.

for the in vivo study, the relatively accu-mulation of A371-functionalized liposome nanoparticles in the spiral ganglion re-gion and inner hair cell region was a result of targeting. our previous study on round window membrane permeation of lipid nanocapsules pegylated with dSpE-pEG-2000, the same coating material used for liposome nanoparticles in the present ex-periment, showed that the nanocapsules mainly appeared in the spiral ganglion region, correlated nerve fibers, inner hair cell region, and spiral ligament of the lat-eral wall. (14� This means that there is a greater chance for the cells and tissues in these regions to be exposed to nanopar-ticles upon round window membrane permeation. in the case of A371-function-alized liposome nanoparticles, the affinity for TrkB on the surface of the spiral gan-glion cells and peripheral processes of the neurons enhanced the distribution of lipo-some nanoparticles in the spiral ganglion and inner hair cell regions. (15� We pro-pose the following mechanism for nano-plex distribution: after permeating the porous modiolar wall of the scala tympani, A371-functionalized liposome nanopar-ticles bind to TrkB on the non-myelinated type ii spiral ganglion cells and peripheral processes, and the amount of liposome nanoparticles along the nerve pathway is enhanced. (14, 1�� This provides more liposome nanoparticles to the inner hair cell region. (14� The movement of lipo-some nanoparticles along nerve fibers was shown in the cochlear explant study, which showed an abundance of liposome nanoparticles attached to neurofilaments (fig. 2�. The access of nanoparticles to the lateral cochlear wall was directly related to the round window membrane and peri-lymph and not limited by the nerve path-way. (14� furthermore, internalization in the spiral ligament fibrocytes and inter-mediate cells of blank-liposome nanopar-ticles was as efficient as that of A371-func-tionalized liposome nanoparticles (fig. 1�.The mechanism of more efficient gene expression mediated by A371-function-alized liposome nanoparticles than by blank liposome nanoparticles is unknown. The A371 peptide might also bind to epi-dermal growth factor protein tyrosine ki-nase (EGfr-pTK� without activation, sim-ilar to the TrkB receptor binding observed in the spiral ganglion cells. This binding blocks EGfr-pTK signaling and improves the efficiency of intracellular traffick-ing and transduction. (17, 18� The unique dot-like expression pattern of EGfp in

the cochlear explants, which possible represents protein aggregation, might be caused by oxidative stress, nitrative insult, or proteasomal impairment. (19, 22� in the cells successfully transfected by liposome nanoparticles, pGeneClipTM hMGfp plasmid dnA encoding shrnA transiently silenced id2 in the host cells. id2 (E47 protein� is reportedly involved in cell survival, cell cycle progression, lipid metabolism, stress response, and lym-phoid maturation. (23�in conclusion, potential targetability of A371-functionalized liposome nanopar-ticles was observed in rat cochleae but not in primary cochlear cell culture or co-chlear explants. functionalization of lipo-some nanoparticles with TrkB ligand did not change cellular internalization, but it enhanced gene expression. in general, the gene transduction efficacy was poor for liposome nanoparticles; this might be resolved by using nuclear localization sig-nal peptides.

AcknowledgementsThe authors thank Mr. Tommi rK Man-ninen (university of Tampere� for plasmid propagation and dnA extraction. This study was supported by the integrated Eu project nanoear (nMp4-CT-200�-02�55��.

References1. Adamo V, Lorusso V, rossello r, et al. pegylated liposomal doxorubicin andpegylated liposomal doxorubicin and gemcitabine in the front-line treatment of recurrent�metastatic breast cancer: a multicentre phase ii study. British journal of cancer 2008;98(12�:191�-1921.2. ElBayoumi TA, Torchilin Vp. Tumor-targeted nanomedicines: enhanced antitumor efficacy in vivo of doxorubicin-loaded, long-circulating liposomes modified with cancer-specific monoclonal antibody. Clin Cancer res 2009;15(��:1973-1980.3. Chang dK, Chiu CY, Kuo SY, et al. Anti-angiogenic targeting liposomes increase therapeutic efficacy of solid tumors. The Journal of biological chemistry 2009.4.Wang M, Lowik dW, Miller Ad, Thanou M. Targeting the urokinase plasminogen activator receptor with synthetic self-assembly nanoparticles. Bioconjugate chemistry 2009;20(1�:32-40.5.Scindia Y, deshmukh u, Thimmalapura pr, Bagavant H. Anti-alpha8 integrin immunoliposomes in glomeruli of lupus-susceptible mice: a novel system for delivery of therapeutic agents to the renal glomerulus in systemic lupus erythematosus. Arthritis and rheumatism 2008;58(12�:3884-3891.�.Khaw BA, daSilva J, Hartner WC. Cytoskeletal-


antigen specific immunoliposome-targeted in vivo preservation of myocardial viability. J Control release 2007;120(1-2�:35-40.7.Arnold AS, Tang YL, Qian K, et al. Specific beta1-adrenergic receptor silencing with small interfering rnA lowers high blood pressure and improves cardiac function in myocardial ischemia. Journal of hypertensionJournal of hypertension 2007;25(1�:197-205.8.Schimmang T, Tan J, Muller M, et al. LackLack of Bdnf and TrkB signalling in the postnatal cochlea leads to a spatial reshaping of innervation along the tonotopic axis and hearing loss. development (Cambridge, England� 2003;130(19�:4741-4750.9. MA Zhongcai WX, CAo Mingmei, pAn Wei, ZHu fenlu, CHEn Jingshan, Qi Zhongtian. Selection of trkB-binding peptides from a phage-displayed random peptide library. SCiEnCE in CHinA (Series C� 2003;4�(1�:77-8�.10.o’Leary pd, Hughes rA. design of potent peptide mimetics of brain-derived neurotrophic factor. The Journal of biological chemistry 2003;278(28�:25738-25744.11.fobian K. roles of nGf-derived peptides in neuritogenesis and neuronal survival. roskilde:roskilde: roskilde university; 2007. 7� p.12.Garcia-Suarez o, Hannestad J, Esteban i, Sainz r, naves fJ, Vega JA. Expression ofExpression of the TrkB neurotrophin receptor by thymic macrophages. immunology 1998;94(2�:235-241.13.Chevalier S, praloran V, Smith C, et al. Expression and functionality of the trkA proto-oncogene product�nGf receptor in undifferentiated hematopoietic cells. Blood 1994;83(��:1479-1485.14.Zou J, Saulnier p, perrier T, et al. distribution of lipid nanocapsules in different cochlear cell populations after round window membrane permeation. Journal of biomedical materials research 2008;87(1�:10-18.15.Tan J, Shepherd rK. Aminoglycoside-induced degeneration of adult spiral ganglion neurons involves differential modulation of tyrosine kinase B and p75 neurotrophin receptor signaling. The American journal of pathology 200�;1�9(2�:528-543.1�.rask-Andersen H, Schrott-fischer A, pfaller K, Glueckert r. perilymph�modiolar communication routes in the human cochlea. Ear and hearing 200�;27(5�:457-4�5.17.Zhong L, Zhao W, Wu J, et al. A dual role of EGfr protein tyrosine kinase signaling in ubiquitination of AAV2 capsids and viral second-strand dnA synthesis. Mol Ther 2007;15(7�:1323-1330.18.Zhong L, Li B, Jayandharan G, et al. Tyrosine-phosphorylation of AAV2 vectors and its consequences on viral intracellular trafficking and transgene expression. Virology 2008;381(2�:194-202.19.norris EH, Giasson Bi, ischiropoulos H, Lee VM. Effects of oxidative and nitrative challenges on alpha-synuclein fibrillogenesis involve distinct mechanisms of protein

modifications. The Journal of biological chemistry 2003;278(29�:27230-27240.20.Stefanis L, Larsen KE, rideout HJ, Sulzer d, Greene LA. Expression of A53T mutant but not wild-type alpha-synuclein in pC12 cells induces alterations of the ubiquitin-dependent degradation system, loss of dopamine release, and autophagic cell death. J neurosci 2001;21(24�:9549-95�0.21.Tanaka Y, Engelender S, igarashi S, et al. inducible expression of mutant alpha-synuclein decreases proteasome activity and increases sensitivity to mitochondria-dependent apoptosis. Human molecular genetics 2001;10(9�:919-92�.22.Seki T, Takahashi H, Adachi n, et al. Aggregate formation of mutant protein kinase C gamma found in spinocerebellar ataxia type 14 impairs ubiquitin-proteasome system and induces endoplasmic reticulum stress. The European journal of neuroscience 2007;2�(11�:312�-3140.23.Schwartz r, Engel i, fallahi-Sichani M, petrie HT, Murre C. Gene expression patterns define novel roles for E47 in cell cycle progression, cytokine-mediated signaling, and T lineage development. proceedings of the national Academy of Sciences of the united States of America 200�;103(2��:997�-9981.


Medical nanomaterials

nanoscale Characterization of Biological and Mechanical profile of Carbon Stent nanocoatingsV.Karagkiozaki 1,2, S.Logothetidis1, S.Kassavetis1, S.Lousinian1(doi 10.3884�0002.2.4�

Abstractnatural tissues are composed of nano-components and cells and these interact directly with nanostructured extra-cellular matrices. Therefore, the surface nanoscale features and properties of biomaterials for vascular applications play a key role in tuning the healing process of the arterial wall after stent implantation. The advantage of nanotechnology comes from the ability to control the properties of the materials to be used in a wide range of biomedical applications. As far as stent coating technology is the challenge is to overcome in-stent restenosis. Thus, a significant amount of research has been recently devoted to the development of nanomaterials that are less thrombogenic but still maintain high mechanical and tribological properties in order toavoid cracking during stent expansion. The scope of this study is to explore the biological and nano-biological and nano- and nano-mechanical properties of amorphous hydrogenated carbon (a-C: H� thin films. This investigation exploits nanoscale sensitive experimental techniques; the comprehension of the properties of the a-C: H films will lead to their improvement so that in the future their use may be applied to stent coatings.Materials and methods: The a-C: H nanocoatings were grown in high vacuum chamber, in Ar�H2 atmosphere using rf Magnetron Sputtering by varying the applied negative bias voltage (Vb� to the substrate. The Vbenables the control of surface nanotopography and the properties of the growing a-C: H nanocoatings via bombardment with the plasma energetic species. The surface morphology and the nanomechanical proper-ties of the a-C: H nanocoatings were investigated by the state-of-the-art non-destructive characterization techniques of Atomic force Microscopy (AfM�, Acoustic Microscopy (AfAM� and depth-sensing nanoinden-tation.The haemocompatibility studies involved protein adsorption mechanisms and platelet adhesion to the case-study materials. Especially, fibrinogen (fib� (5 mg�ml� solution in phosphate buffer saline (pBS, pH 7.4� wasfibrinogen (fib� (5 mg�ml� solution in phosphate buffer saline (pBS, pH 7.4� was used with concentrations similar to those found in the blood of a healthy donors, as this plasma protein takes this plasma protein takes part in blood coagulation, facilitates platelets adhesion, aggregation and determines biomaterial failure. The real-time investigation of its adsorption was made by the real-time Spectroscopic Ellipsometry (SE� inreal-time Spectroscopic Ellipsometry (SE� ineal-time Spectroscopic Ellipsometry (SE� in the Vis-uV energy region which is a non-destructive technique, applied both in air and in liquid environment. ultra fast measurements were performed in-situ, with the use of a flow-cell. Human platelet rich plasma was prepared after centrifugation of whole blood drawn by venopuncture, from healthy donors, and the platelets were fixed with 5 μl glutaraldehyde 1% for one hour. Tapping mode AfM was implemented for their5 μl glutaraldehyde 1% for one hour. Tapping mode AfM was implemented for theirμl glutaraldehyde 1% for one hour. Tapping mode AfM was implemented for their glutaraldehyde 1% for one hour. Tapping mode AfM was implemented for theirglutaraldehyde 1% for one hour. Tapping mode AfM was implemented for their 1% for one hour. Tapping mode AfM was implemented for theirTapping mode AfM was implemented for their study, being a nanoscale surface sensitive technique which is harmless for the biological specimens. being a nanoscale surface sensitive technique which is harmless for the biological specimens.being a nanoscale surface sensitive technique which is harmless for the biological specimens.results: The deposition conditions of the a-C: H nanocoatings were found to affect their surface nanotopog-raphy and nanomechanical properties. in particular, alterations in ion bombardment during deposition and and hydrogen content in plasma result in different bonding configuration and surface properties of the carbon-based films. protein adsorption mechanisms were studied by an appropriate ellipsometric model, involvingadsorption mechanisms were studied by an appropriate ellipsometric model, involving their optical response and the thickness of the protein layers formed on their surfaces. This model supposes that each moment the layer formed on the a-C: H nanocoating consists of fib molecules that have their na-tive form (fib solution� and adsorbed fib molecules (bound form�, and it takes into account the liquid ambi-ent of the experiments. These data were consistent with the AfM images of adherent platelets attached toplatelets attached to the biomaterials and will be discussed in the view of their nanotopography.biomaterials and will be discussed in the view of their nanotopography.The nanomechanical characterization of the a-C:H nanocoatings was accurately made by AfAM and nanoin-dentation and this is a perquisite for their potential use as stent coatings.The proportion of hydrogen in theThe proportion of hydrogen in the a-C: H coatings was found to affect their mechanical behavior whereas AfAM results show that the surfacecoatings was found to affect their mechanical behavior whereas AfAM results show that the surface morphology does not affect the near-surface elastic properties. firthermore, there were no large variations in the elastic properties of the surface, suggesting they formed a homogeneous surface.in conclusion, the combination of tailoring the deposition parameters of the grown a-C: H nanocoatings anda-C: H nanocoatings and nanocoatings and the in-depth examination of their surface properties by a plethora of nanoscale techniques can facilitate the enhancement of their haemocompatibility and mechanical stability, to be used for medical applications.Karagkiozaki V.nanomedicine 2009, 2.2:14-21

Keywords: Atomic force Microscopy, stents, stent nanocoatings, platelets, thrombogenicity, nanotechnology, carbon basedthrombogenicity, nanotechnology, carbon based nanotechnology, carbon based carbon based thin films, nanomechanical properties, fibrinogen adsorption.




1. department of physics, Lab for “Thin films - nanosystems & nanometrolo-gy”, Aristotle university of Thessaloniki (AuTh�, Gr-54124, Greece.2. AHEpA Hospital, 1st CardiologyAHEpA Hospital, 1st Cardiology department, Lab for “CardiovascularLab for “Cardiovascular Engineering & Atherosclerosis”, AuTh, Thessaloniki, Greece, Gr-54124.

IntroductionThe treatment of coronary artery dis-ease using metallic stents has been one of the most revolutionary and effective medical intervention in recent years. despite its positive outcomes, there is a debate concerning the influence of stent design and materials, thickness of stent struts and surface coatings, on the trig-gering of healing of the vessel wall and the formation of neoendothelium onto the stent surface [1]. The problem be-comes even more serious, in the field of drug eluting stents (dES� that elicit antiproliferative drugs, where the late thrombosis of dES has been a lingering consideration within the cardiology com-munity [2]. This type of stent has as ba-sic component, different biodegradable polymers that may offer ideal solutions in terms of providing a carrier to retain and release the required amount of drug, prior to their full degradation [3]. drug compounds mixed with the polymer matrices are gradually released as the polymer degrades within the vessel wall. However, recent studies have shown that the polymers, the drug coatings themselves or their combination may cause platelet activation and inflamma-tory reaction of the vessel wall [4]. The study by Michael Joner comparing au-topsies of 23 people implanted with dES for more than 30 days, with those of 25 matched autopsies implanted with bare metal stents (BMS� showed that dES pa-tients had greater delayed healing, char-acterized by persistent fibrin deposition and poorer endothelialisation compared to BMS stent patients [5]. Although the exact cause of poor endothelialization has not been established, there is a sig-nificant motivation for the development of technologies which can avoid absorb-able or non-absorbable polymers for loading drugs onto the stent surface.As biological systems in nature func-biological systems in nature func-tion fundamentally at the nanoscale, a strong motivation for nanofabrication at the biological level is essential for pro-viding biomaterials for stent coatings that elicit optimal cellular responses.due to the fact that vascular tissue is a layered structure possessing numerous nanostructured features (i.e. because of the presence of collagen and elastin in the vascular extra-cellular matrices�, the nanomaterials used as coatings have shown much promise to improve endo-thelial and smooth muscle cells func-tions, to inhibit thrombosis and severe

inflammation [�]. Choudhary et al. re-ported that vascular cell adhesion and proliferation were greatly improved on nanostructured titanium (Ti� compared to conventional Ti [7]. interestingly, greater competitive endothelial cell ad-hesion, total elastin and collagen syn-thesis were observed on vascular smooth muscle cells grown on nanostructured Ti after 5 days in culture . Since one of the current problems with vascular stents is the overgrowth of smooth muscle cells compared to endothelial cells, these results suggest that endothelial cell functions were enhanced over those of vascular smooth muscle cells, thus, increasing the probability of endothe-lialization on nanostructured stents. it was speculated that the increased nano-roughness and particle boundaries on nanostructured Ti contributed to the observed favorable endothelial cell func-tions [7]. The tunable material properties that nanotechnology can provide were presented in norio Taniguchi’s paper in 1974 where the term ‘‘nanotechnology’’ was first used in a scientific publication [8]. Subsequently, a significant amount of effort has been devoted to the devel-opment of classes of nanomaterials that are less thrombogenic with excellent mechanical and tribological properties in order to avoid cracking during stent expansion. The aim of this study is to ex-ploit the biological and nanomechanicalbiological and nanomechanical and nanomechanical properties of amorphous hydrogenated carbon (a-C: H� thin films which can be used as stent coatings, by nanoscale sen-sitive techniques. in previous studies we developed a methodology using Atomic force Microscopy to study thrombus formation caused by material–platelet interactions, by [9, 10]. in the present study we investigated the biological pro-file of the case study materials, adsorbed plasma proteins and their adsorption mechanisms by real-time Spectroscopicreal-time SpectroscopicSpectroscopic Ellipsometry (SE� in the Vis-uV energySE� in the Vis-uV energy region. The nanomechanical properties of the carbon based studied films were also assessed. Materials and MethodsThe sputtered a-C: H nanocoatings (with 100-120nm thickness� studied in this work were deposited by rf magnetron sputtering on c-Si (100� substrates at room temperature. details about the growth of the films have been described elsewhere [11, 12]. for the growth of a-C: H films, hydrogen (H2� reactive gas was

introduced into the vacuum chamber un-der a partial pressure of 5%. The energy Ei of the ions (mainly Ar+ or H+� bom-barding the growing film surface varied by applying a negative bias voltage to the substrate Vb (-40V� or by applying Vb (floating conditions, +10V�. Consistent with their manufacturing conditions, two types of a-C: H nanocoatings were studied: floating and biased ones.The surface morphology of the a-C: H nanocoatings and their nanomechanical properties were investigated by non-de-structive characterization techniques: AfM and Atomic force Acoustic Micros-copy (AfAM�, respectively [13]. AfM is an ubiquitous imaging tool that resolvesthat resolves resolves features at atomic dimensions in both vertical and lateral dimensions. for the nanomechanical behavior of the carbon-based nanocoatings, the AfAM tech-nique was combined with depth-sensing nanoindentation (ni� [14].The nanoindenter (nano indenter Xp� is equipped with the Continuous Stiffness Measurements, which enables measure-ment of the hardness versus the penetra-tion depth. Several indents (8–10�, with 30 μm spacing and 0.05 mn�s loading rate, were made at every sample for statistical purposes. Thus, the presented hardness (Hf� is the mean value. A Berkovich type diamond tip, with 50 nm nominal radius, was used as the indenter.The near-surface nanomechanical prop-erties of the a-C: H thin films were also investigated by AfAM, which is an experi-mental technique, based on the scanning probe microscopy (SpM� principles. The AfAM configuration includes a piezo-electric transducer, with 2.5 MHz central frequency, which is placed below and in contact with the sample. The transducer emits longitudinal acoustic waves which cause out-of-plane vibrations of the sam-ple surface. By setting the SpM probe (cantilever� in contact with the sample surface, we can acquire the so-called acoustic images, which represent the vi-bration amplitude of the cantilever. The

Fig. 1a. Three-dimensional AFM image of ‘bare’biased a-C: H nanocoating with rRrms

0.3nm.


EuropEAn JournAL of nAnoMEdiCinE 2009 Vol. 2 issue 2 1�

fig. 1b. Three-dimensional AfM image of ‘bare’ floating a-C: H nanocoating (rRrms= 2 nm�.

where A is the molar refractivity (cm3�mol�, M is the molecular weight of the protein, d and nf the thickness and the refractive index of the protein layer. The latter two parameters are determinated by SE. Evaluation of the thrombogenicity of the studied thin films was achieved by determining the relative ratio between the calculated surface concentrations ��for HSA and fib protein layers (��HSA��fibratio�, which is commonly used as a first indication of the thrombogenicity. Given the function of each protein, the higher the value of the ratio the less thrombo-genic the film is considered [17-23].After drying in air, Tapping mode of AfM was implemented for the study of the fib protein lay-ers formed on the a-C: H samples.in parallel with the protein adsorption study, the interactions of platelets with biomaterials at interfaces were inves-tigated. for this, human platelet rich plasma (prp� was prepared by centrifug-ing whole blood drawn by venopuncture from healthy donors, at 800 rpm for 10 minutes, at room temperature. After the appropriate dilution of prp with homol-ogous plasma, the platelets were fixedplatelets were fixed with 5 μl glutaraldehyde 1% for one hour5 μl glutaraldehyde 1% for one hourμl glutaraldehyde 1% for one hour glutaraldehyde 1% for one hourglutaraldehyde 1% for one hour 1% for one hour and then deposited onto the case study coatings. AfM tapping mode was imple-AfM tapping mode was imple-mented for the platelet study, to avoid to avoidto avoid their destruction.

Results and DiscussionNanotopographical features and nanomechanical proper-ties of carbon stent nanocoatings The increased risk of late stent throm-bosis is one of the biggest challenges faced by the current dES technology. it has been well demonstrated that such thrombosis is more frequently associ-ated with poor endothelialization of stent struts caused by polymeric carriers of drugs [24]. nanoapproaches target to modify stent surface by developing nanocoatings with constituent dimen-

sions of approximately 100 nm that can mimic surface properties of the natural tissues. over the last decade, nanomate-rials have been highlighted as promising candidates for improving traditional bio-materials and their surface topography has been shown to be a decisive factor affecting cell morphology, adhesion and motility [25]. Moreover, the ameliora-tion of biocompatibility is important for medical devices such as stents that interact with blood during clinical use. When blood encounters a foreign body the response can be aggressive, result-ing in thrombus formation and activa-tion of the immune system, which will eventually impair the function of the device and compromise the health of the patient. in this context, extensive in-vestigations on the surface modification of biomaterials have been undertaken with the prospect of attaining greater hemocompatibility characteristics [2�].recent studies found pronounced dif-ferences in cell spreading and focal ad-hesion dynamics using surfaces with features between 50 and 500nm; these differences depended not only on fea-ture size, but also on the spacing be-tween cell-recognizable features [27]. The nanomechanical properties of bio-materials have also been demonstrated to produce a profound impact on cell and tissue behavior. for example, the stiff-ness of a cell substrate was optimized to achieve optimal growth of muscle cells. in particular, very soft and stiff gels pro-duced myosin patterns which were dif-ferent from those observed under nor-mal physiological conditions. Endothe-lial cells, fibroblasts all showed increased proliferation on stiffer substrates [28].Tissues of the human body have selec-tive mechanical properties ranging from soft (brain, about 0.5kpa� to moderate stiff (muscles, 10kpa� and stiff (precalci-fied bone >30kpa�.The narrow specifica-

latter vibrates in contact with the sample surface at a fixed frequency, close to the resonance one. The acoustic images of every sample were acquired simultane-ously with the topographic ones, in order to study in terms of surface morphology and nanomechanical properties of the exact same surface areas of the samples. The measurements were performed with a SoLVEr p47H Scanning probe Mi-croscope (nT-MdT, nTi instruments� in ambient environment, at room tempera-ture. Standard silicon cantilevers with nominal spring constant kc=1 n�m, 10 nm nominal radius and typical resonance frequency 20 kHz were used for AfM and AfAM measurements. The quantity used The quantity used for the evaluation of surface roughness of the plain films for ten areas chosen for random, was root-mean-square rough-ness (Rrms� and mean roughness (by x2- statistical test�. The haemocompat- The haemocompat-ibility study involved protein adsorption mechanisms and platelet adhesion to the case-study materials. in order to es-timate their potential thrombogenicity, fibrinogen (fib� solution in pBS (pH 7.4� was prepared at a concentration of 1mg�ml, that is similar to that found in human blood. fibrinogen is considered to be decisive factor for the biocompatibility, failure and long term efficacy of implants as it takes part in blood coagulation, fa-cilitates platelet adhesion and aggrega-tion and it is important in the processes of both haemostasis and thrombosis [15]. The samples were dipped into the protein solutions for a total time of two hours at room temperature., They were then rinsed with deionized water and dried under n2 flow. Ex-situ SE measurements were per-formed in the energy region 1.5–�.5 eV at angle of incidence of 70°, before and after the incubation of the films in the protein solutions. real-time SE mea-surements were performed for the fib adsorption study on a-C: H thin films grown with and without application of Vb and partial H2 pressure 5%. The mea-surements were performed in a solid-liq-uid cell at angle of incidence of �0◦ and in the energy region 1.5–4 eV and the du-ration of the measurements is 500msec per sec, for 5400sec. The relative surface concentration �� (µg�cm2� of the albumin (HAS� and fibrinogen can be derived by the Cuypers formula [1�]:

,2

11.0 2

2

+

−⋅⋅⋅=Γ

f

f

nn

AMd

Fig. 3b.AFAM acoustic image of the a-C: H surface.



-20 0 20 40 60 80 1001

2

3

4

5

6

Har

dnes

s, H

(GPa

)

Negative Bias Voltage, Vb (V)

Fig.2.The hardness (Hf) of the a-C: H nano-coatings versus the negative applied to the substrate voltage (Vb�.

Fig. 3a. 2D AFM image of the a-C: 2D AFM image of the a-C: H surface nanotopography.

tions of elastic moduli create mechani-cally defined microenvironments that effectively support the development of cellular architecture [29]. in order to transduce mechanical stimuli into bio-chemical signals, cells use a palette of localized elements including mechano-sensitive ion channels, forced unfolding of proteins and remodeling of focal ad-hesion sites. Endothelial and other cell types have developed mechanosensory machineries and an increasing number of cellular processes, such as adhesion, cell spreading or cytoskeletal remodel-ing are found to depend on mechanical cues originating from the cellular micro-environment [30]. from a different pro-spective, in the field of stent manufac-turing, the mechanical behavior of the surface coatings needs to be critically evaluated, as the coatings should exhibit superior tribological behavior and resis-tance to fracture during stent expansion.AfM tapping mode of function was em-ployed to obtain information about the surface roughness of the a-C: H nano-coatings. The AfM images (Figs. 1a, 1b� reveal that the carbon thin films grown under the application of bias voltage are atomically smoother (smaller sur-face roughness values: rRrms =0.3nm and Rm=0.2nm, p<0.001� due to the fact that the voltage increases the kinetic energy of the carbon sputtered atoms, resulting in the layer-by-layer growth of the thin film. in contrast, the lack of bias voltage application during deposition (floating conditions� gives the possibility to the carbon atoms to form aggregates and islands, and this results in higher rRrmsroughness values reaching approximate-ly the 2.8nm (rm=2.2 nm�.As far as the characterization of the nanomechanical properties of the stud-ied materials by the nanoindentation technique, in Fig. 2, the Hf values (after subtracting the substrate effect� versus the Vb are presented. The Hf was found to increase from 1.8 to 5.� Gpa, when |Vb|≥−�0 V. A plateau value appears for further increase of |Vb|. These Hf values are characteristic of the a-C: H thin films [31]. in addition, a similar dependence on Vb was also reported for several car-bon based (non-hydrogenated, hydroge-nated, nitrogenated� thin films [31- 32].AfAM characterization of biomateri-als in which the acoustic image, is ac-quired simultaneously with the topo-graphic one during an AfAM scan, can provide information about the varia-

tions of the nanomechanical proper-ties of the surface (Figs. 3a and b). in the acoustic image, the elastic prop-erties of the surface are represented through the amplitude variation (the darker regions in the image correspond to lower contact stiffness�. previous stud-ies showed that the proportion of hydro-gen in the hydrogenated carbon coat-ings affects its mechanical behavior and its increase results in softer samples [33].in the case of the studied a-C: H thin film, the variations in the acoustic image dif-fer from the corresponding topographic ones. Thus, the surface morphology

does not affect the near-surface elastic properties. furhtermore there are no large variations in the elastic properties of the surface, a fact that denotes a ho-mogeneous surface.

Protein adsorption mechanisms and human platelets interactions with car-bon based nanocoatings a) Protein adsorption studyThe use of biomaterials coming into con-tact with blood is challenged by blood coagulation and attack by the immune system. These blood reactions are initi-ated by the characteristics of the surface of biomaterials which induce protein adsorption and blood cell adhesion pat-terns often provoking the activation of the body defense systems [34]. proco-agulant processes still form the major barrier in a variety of demanding appli-cations. under normal conditions, blood contacts an endothelium with anticoag-ulant and antithrombotic properties. The use of a cardiovascular device represents the introduction of a foreign surface in the circulation lacking the properties of the endothelium. Blood–material inter-actions trigger a complex series of events including protein adsorption, platelet and leukocyte activation�adhesion, and the activation of complement and co-agulation; these systems are highly in-terlinked [34, 35] and their interactions become more significant in the cases of vascular stents, where there is a direct damage of the endothelium during stent implantation. The intact endothelial cells constitute an anticoagulant lining of the vessel wall with essential synthetic and metabolic functions relating to the hae-mostatic system [3�]. After injury, the damaged blood vessels are exposed to tissue factors produced constitutively by cells beneath the endothelium. They then bind to factor Vii or Viia and this un-derlies the formation of thrombin [37].The amount of adherent platelets and the activation on the film surfaces is in-versely proportional to the ratio of albu-min to fibrinogen adhesion on the sur-faces [18, 38-39]. While the adhesion of albumin (a water-soluble protein in the blood plasma� on the film can prevent the adhesion of platelets, fibrinogen is a protein that can enhance the adhesion and activation of platelets and hence the thrombus when it is converted into fibrin by the action of the enzyme thrombin [39]. in a first approximation, the inves-tigation and evaluation of the implant

a�

b�

Fig. 6.Fib layer morphology after 90 min of incubation on a) the floating and b) Biased a-C:H samples


10 20 30 40 50 60 70 80 900,0

0,2

0,4

0,6

0,8

1,0

1,2

MS a-C:H CVD a-C:H MS a-C ta-C

Ã HSA / Ã

Fib

% sp3

Fig. 4. Protein adsorption ratio (ΓHSA /ΓFib) indicative of the haemocompatibility of the a-C and a-C: H samples grown by different methods [40]. (*ΜS: Magnetron Sputtering, CVd: Chemical vapor deposition�

Figure 5. a� Measured psi & delta ellipsomet-ricangles for a-C:H and for fibrinogen�a-C:H asfunction of time. The insets show two ran-dom experimental and fitted data for 50s and 540s duringfib adsorption on a-C:H floating thin film, b� Evolution of fib thickness and c� Volume fractions of adsorbed fib, on float-ing & Biased a-C: H samples.

a�

b�

c�

blood compatibility can be carried out by probing the interaction of its surface with the proteins albumin and fibrinogen. The enhancement of albumin adhesion com-pared to that of fibrinogen is highly de-sirable for the success of haemocompat-ible coatings.in preliminary studies, ex situ SE measurements were performed in the energy region 1.5–�.5 eV, to study human albumin and fibrinogen adsorp-tion onto amorphous carbon (a-C� and a-C: H samples. in Figure 4, the relation between the protein adsorption ratio and the content of sp3 bonds of various amorphous carbon thin films developed by different techniques is demonstrated [40]. The relative surface concentration �� (µg�cm2� of the HSA and fib can be derived by the Cuypers formula as de-scribed in the methods section, whereas, whereaswhereas a higher ratio of �� HSA ��fib is indicative of a better haemocompatibility. As shown in this figure, the a-C: H films with a spthe a-C: H films with a sp3content of 40–45% have a higher higher ��HSA ��fib ratio, an index of a higher hae-higher hae-mocompatibility [41- 42].in the following section, we provide a thorough SE analysis of fibrinogen ad-sorption mechanisms on the two types

of a-C: H thin films, with deposition conditions as mentioned above. We sup-We sup-posed that each moment there is a layer formed on the a-C: H film surface that is consisted of fib molecules that are in na-tive state (as in fib solution� and in the bound state (adsorbed fib� and that the volume fraction of the adsorbed fib in-creases during the evolution of the pro-tein adsorption, reaching its maximum (~100%� when the equilibrium of the phe-nomenon takes place. Thus, the protein

layer formed on a-C: H thin film is com-posed of adsorbed fib and molecules of fib in the solution, with the volume frac-tions of the two phases varying through time. The volume fraction of fib in ad-sorbed and liquid state was estimated by using Bruggeman Effective Medium Approximation (BEMA� [43] to describe the fib adsorption phenomenon. The ef-fect of the liquid ambient was taken into account in the suggested ellipsometric model. The measurements of psi & deltameasurements of psi & delta ellipsometric angles for a-C: H and for

fibrinogen�a-C: H as a function of time are shown in Fig. 5a. The evolution of fibThe evolution of fib thickness and the volume fraction of ad-sorbed fib with time during fib adsorp-tion on the floating and biased a-C: H samples are presented in Fig. 5b and c,respectively. The equilibrium of the phe-nomenon is reached at about 2�00s for the floating sample, while on the biased sample the equilibrium is reached earlier, at about 1200s. it is observed that the fib thickness as well as the volume fraction of adsorbed fib is larger on the floating a-C: H thin film, especially at the initial stage of the protein adsorption. This could be attrib-uted to the surface topography of the a-C: H thin films as the rrms of the floating sample (~2.8nm� is one order of magni-tude larger than that of the biased one (~0.3nm�. Thus, the topography of the floating a-C: H thin film offers a larger area for the fib molecules to bind and to transform to the adsorbed form of fib. Another feature that differs between the two studied samples is the variation of the thickness and the volume fraction of adsorbed fib during time (Fig.5c�. on the floating sample, a continuous


1� EuropEAn JournAL of nAnoMEdiCinE 2009 Vol. 2 issue 2

a�

b�

Fig. 7. AfM topography image of a platelet on a-C: H film a� at an early stage of activation , after 15min of incubation [47] b� highly acti-vated with pseudopodia after one hour of in-cubation and c� highly activated with pseudo-podia after 2 hours of incubation.

Fig. 8.AFM topography image of platelets on a-C: H film after 1 hour - incubation a) onto floating nanocoating, at different stages of activation. The platelet denoted with the red circle has a pseudonucleus and the other within the white circle, is more activated and elaborates pseudopodia and b) onto biased nanocoating, they make aggregations, as presented with blue circle.

increase of both thickness and volume fraction of adsorbed fib takes place, while on the biased one, variations of the values of the above parameters occur, revealing the conformational changes of the fib molecules during the adsorption.After the real-time study, the samples were left to dry in the air and they were studied by tapping mode AfM. in Figures 6a and 6b the topography images of 1x1μm of adsorbed fib on a-C: H floating and biased thin films are presented, respectively. The rrms value is 4.8nm for the fib layer formed on the floating a-C: H and 1.�nm for the fib layer formed on the biased a-C: H sample.The aggregates are larger on the floating a-C: H sample, due to the larger roughness value of the bare sam-ple, which gives available area for the fib molecules to bind on the surface. on the other hand, the trinodular structure of fib molecules is obvious in Fig. 6b.b) Platelet adhesion studyinformation obtained by the platelet adhesion study is essential for elucidat-ing the cellular interactions at interfaces and the factors that are associated with the thrombogenic potential of biomate-rials. The thrombotic complications as-sociated with cardiovascular devices are clearly linked to their ability to activate platelets [44]. platelets respond to min-imal stimulation and become activated when they come into contact with any thrombogenic surface such as injured endothelium, subendothelium and arti-ficial surfaces. platelet activation is ini-tiated by the interaction of an extracel-lular stimulus with the platelet surface. This interaction involves the coupling of the agonist to specific receptors on the platelet plasma membrane. plasma pro-teins such as thrombin and fibrinogen, vascular wall products such as collagen and molecules derived from inflamma-tory cells (i.e., leukocytes� or platelets, such as platelet activating factor or ca-thepsin G, are all potent platelet activa-tors [45].inactive platelets are disc shaped with a size of 0.5-3μm. When they adhere onto biomaterial surface they become acti-vated, their shape becomes spherical, their granules gather into the cell center, they develop a pseudonucleus (egg-like type cells� and pseudopodia are extend-ed. These events involve the remodeling of the internal cytoskeleton of platelets

which is mainly composed of actin and tubulin, whereby new actin filaments as-semble and their granulae redistribute. The extension of pseudopodia which are finger like projections also called filo-pods is an essential step for their spread-ing onto surfaces [4�]. All these structural changes occurring in activated platelets were taken into ac-count for the analysis of AfM images. in Fig. 7a, a three-dimensional AfM image

of a single activated adherent platelet on carbon substrate is presented [47] after 15min of incubation; as shown it looses its discoid shape and becomes spherical. Fig. 7b shows an image of a platelet ,one hour after activation; as shown the platelet develops pseudopodia and the concentration of the granula and the ac-tin and tubulin polymers at the centre of the cell cause the formation of a pseudo-nucleus. After two hours of platelet ad-hesion on the same substrate (Fig. 7c� the platelets interact with each other by means of their pseudopodia, assembling a network.for the case study materials, the AfM observations of the human platelets ad-herent onto their surfaces are presented in Figs. 8a and b. As shown, after one hour, the platelets adherent onto float-ing a-C: H thin film activate and a few of them (with red color� forming the egg-like type structure (denoted with blue cir-cle� whereas other develop pseudopodia (denoted with white circle�. in contrast, on biased a-C:H thin film the platelets make aggregations (blue circle� which is a warning of thrombus formation.As far as the influence of surface roughness of the carbon nanolayers on platelet aggre-gation, the AfM measurements establishthe AfM measurements establish


a�

b�

c�


a more direct correlation between the morphological differentiation of plate-lets with the surface nanostructure. it can be clearly seen by comparing the AfM images of the case study materials, that on the biased carbon films, platelets are highly activated and aggregate whereas on the floating ones they exhibit a struc-tural conformation indicative of differ-ent stages of activation (a few of them having egg-like structure and others with filopods� but without any aggrega-tions. Karagkiozaki et al 2007 shown that the higher content of hydrogen in the plasma during growth of the examined materials is associated with the AfM ob-servations that the platelets tend to ag-gregate and make higher clusters onto their surfaces [47]. Additionally, prelimi-nary studies by our Lab reveal that the Human albumin� fib ratio of a-C films increases with % sp3 bonds of the carbon biomaterials. The a-C: H films with a sp3content of 40–45% have shown a smaller possibility for thrombus formation when deposited under floating conditions and with small H2 partial pressure. Especially, floating a-C: H coatings, present a higher Rrms roughness and a higher HSA � fib ra-tio (an index of better haemocompatibil-ity and lower platelets aggregation than the biased a-C: H ones [40-42, 48-49]. As the cellular interactions at interfaces are influenced by a variety of physical prop-erties of biomaterials, besides nanoto-pography, the electrical fied distribution onto their surfaces, the wettability and surface free energy were also previously evaluated [50].

ConclusionsA thorough, non-destructive analysis of protein and platelet interactions with potential stent nanocoatings for haemo-compatibility is made by AfM & SE tech-niques. The analysis of the SE measure-ments, with the development of a new possible ellipsometric model that de-scribes in detail the adsorption phenome-non, provide valuable information on the adsorption mechanisms of fib formed on the a-C:H samples and the protein tran-sition rate from liquid to adsorbed state. This model supposes that each moment the layer formed on the a-C: H thin film consists of fib molecules that have their native form (fib solution� and adsorbed fib molecules (bound form�, and it takes into account the liquid ambient of the experiments. The surface characteristics of the a-C: H thin films that are control-

led by tuning the deposition conditions have a significant effect on the adsorp-tion mechanisms of fib. in particular, the rougher floating thin film allows fib molecules to bind to a larger area and at the same time it compels them to pack and to form aggregates. This could be the main reason why the fib thickness and the volume fraction of adsorbed fib on the floating thin film are larger than those on the biased thin film. The finding that the examined biased thin film is more thrombogenic in terms of fib adsorption compared to floating one was verified by the AfM platelets study. The platelets were found to be highly activat-ed, forming clusters at biased a-C: H na-nocoating whereas on floating one, these cells were found at various stages of acti-vation but withoutaggregate formation. The nanomechanical characterization of the a-C: H thin films was made accu-rately by AfAM and nanoindentation and this is a prerequisite for their poten-tial usage as stent coatings. The latter The latter technique revealed that the measured Hf was affected by bias voltage (Vb� and was found to increase from 1.8 to 5.� Gpa with increasing |Vb|. AfAM results showed that the surface morphology does not affect the near-surface elastic properties. furthermore there were no large variations in the elastic properties of the surface, a fact that is indicative of a homogeneous surface. The enhanced mechanical stability and haemocompat-ibility of a-C: H nanocoatings for medi-a-C: H nanocoatings for medi- nanocoatings for medi-cal applications are facilitated by tuning their deposition parameters and by the thorough examination of their surface properties by a wide variety of nanoscale techniques.

References1. Hara H, nakamura M, palmaz JC, Schwartz rS, Adv drug deliv rev 58 (200�� 377.2. John MC, Wessely r, Kastrati A, Schömig A, Joner M, et al. JACC: Cardiovascular interven-tions, 1 (2008� 535.3. Smith EJ, Jain AK, rothman MT. J int Cardio � (200�� 493.4. Stone GW, Moses JW, Ellis SG. n Engl J Medn Engl J Med 35� (2007� 998.5. Joner M, finn AV, farb A, et al. JACC 48 (200�� 193.�. Zhang L, Webster T. nano Today 4 (2009� ��.7. Choudhary S, Haberstroh K, Webster T. Tis-sue Eng. part A 13 (2007� 1421. 8. Taniguchi n. proc. intl. Conf. prod. London, part ii, British Society of precision Engineer-

ing, 1974.9. Karagkiozaki V, Logothetidis S, Laskarakis A, Giannoglou G, Lousinian S. Mater Sci Engin B 152 (2008� 1�.10. Karagkiozaki V, Logothetidis S, Kalfagian-nis n, Lousinian S, Giannoglou G. nanomed: nanotech Biol Med 5 (2009� �4.11. Gioti M, Logothetidis S. diam relat Mater 12 (2003� 957.12. Logothetidis S. Handbook of thin films materials, in: H.S. nalwa (Ed.�, Characteriza-tion and Spectroscopy of Thin films, vol. 2, Academic press, 2002, p. 277.13. Bhushan B. 2004. Handbook of nanotech-nology, Berlin, Springer, u. rabe, K. Janser, W. Arnold, rev. Sci. instrum. �7 (199�� 3281.14. oliver WC, pharr GM. J Mat research 19 (2004� 3.15. Sugio S, Kashima A, Mochizuki S, noda M, Kobayashi K. prot Eng 12 (1999� 439.1�. Cuypers p, Corsel J, Janssen, M, et al. J Biol Chem 258 (1983� 242�.17. dion i, Baquey C, Candelon B, Monties J int J Artif organs 15 (1992� �17.organs 15 (1992� �17.18. Jones M, McColl i, Grant d, parker K, par-ker T. diam relat Mater 8 (1999� 457.19. Yuu L, Wang X, Liu X. Surf Coat Technol 128-129 (2000� 484.20. Jones M, McColl r, Grant d, parker K, par-ker T. J Biomed Mater res 52�2 (2000� 413.21. Cui f, Li d. Surf Coat Technol 131 (2000� 481.22. Hauert r. diam rel Mat 12 (2003� 583.23. Lin CH, Jao WC, et al. Col Surf B: Biointer-Col Surf B: Biointer-faces 70 (2009� 132.24. Steigerwald K, Sabine Merl S, Adnan Kas-trati A, et al. Biomaterials 30 (2009� �32.25. peppas nA, Langer r. Science 2�3 (2004� 487.2�. Zhanga S, dua H, et al. Thin Solid films 515 (200�� .27. Chen CS, Mrksich M, Huang S, et al. Scien-Scien-ce 27� (1997� 1425.28. Griffin MA Sen, Sweeney HL et al. J Cell Sci 117 (2004� 5855.29. Engler A, richert L, et al.Surf Sci. 570 (2004� 142.30. Bershadsky A et al. Eur J Cell Biol. 85 (200�� 1�5.31. robertson J. Mater Sci. Eng. r rep. 37 (2002� 129.32. Laskarakis A, Logothetidis S, Charitidis C, Gioti M, panayiotatos Y, et al. relat Mat 10 (2001�1179.33. Kassavetis S, Logothetidis S, Matenoglou G. Surf Coat Tech 200 (200�� 400.34. Maud B, Gorbet M, et al. Biomaterials 25Biomaterials 25 (2004� 5�81.35. Sperling C, Salchert K, Streller u, Werner C. Biomaterials 25 (2004� 5101.3�. Jaffe EA. Biochemistry, immunology and



cell Biology of Endothelium. in:colman rW,HirshJ,marderVJ,SalzmanEWEditors,Bloodvessels in hemostasis and thrombosis, Lip�pincott company, philadelphia (1994), pp.718–744.37.diqueloua,dupouyd,etal.�hromb Hae��hrombHae�most74(1995)778.38.cuiZ, Li d.Surf coat�echnol 131 (2000)481.39.ongSE,ZhangS,duH,etal.Biomaterials28(2007)4033.40. LogothetidisS. diam rel mat 16 (2007)1847.41.LogothetidisS,Giotim,LousinianS,fotia�douS.�hinSolidfilms482(2005)126.42.LousinianS,LogothetidisS.microeleEng84(2007)479.43.aspnesd.�hinSolidfilms89(1982)249.44.GrunkemeierJ,�saiW,Horbett�.JBiomedmaterres41(1998)657.45. Blockmans d, deckmyn H,VermylenJ.JBloodrev9(1995)143.46.fritzm,radmacherm,GaubHE.BiophysJ66(1994)1328.47.KaragkiozakiV,LogothetidisS,GiannoglouG.EurJnanomedicine1(2008)24.48.mitsakakisK,LousinianS, Logothetidis S.S,LogothetidisS.BiomolEng24(2007)119.49.LousinianS,LogothetidisS,etal.BiomolEng24(2007)107.50. KaragkiozakiV, LogothetidisS, LousinianS, Giannoglou. int J nanomedicine, issue onLine�Early,2008.

medicalnanomaterials

implementingnanomedicine

The Importance of the Industrial -academicinterfaceforinnovationinthepharmaceuticalSectormikeEaton,ucB,uK(doi10.3884/0002.2.5)

The pharmaceutical sector has in recent years suffered from a perceived lack of successful new drugs, espe�cially blockbusters. There is much debate as to the causes of this, but the current regulatory and financial en�vironment requires that the sector tackles new and innovative therapeutic targets that challenge its current science base. These new targets will continue the shift away from chemical entities (NCEs), to biologicals andinthelongertermtoexploreall“drugspace”.“drugspace”thatisnotpartofthecurrentpharmareper�toire includes higher molecular weight drugs e.g. nanomedicines, nucleic acid-based drugs1,and“non�Lipin�ski”inhibitorsofprotein�proteininteractions(figure1).asaresultthetherapeuticsectorisactivelylookingoutside its walls, taking lessons from the biotech revolution, for new thinking from academics and SMEs. Eaton,mnanomedicine2009,2.2:22�24Keywords: nEc,drugspacenanomedicines,biotechrevolution

Examples from the pastHistorically, industrial�academic contactinthencEareahasbeen,ingeneral,lim�ited to the education of new employees. This interface changed somewhat with the emergence of the biotech industry,largely based on the new and unexpected molecular and cell biology tools devel�opedbytheacademicsector,especiallyinthe US. These opportunities were seized by visionary entrepreneurs and despitetheconservatismandscepticismoflargecompanies, new businesses, such as Ge�nentech and Amgen were formed and were successful. Large companies can be slow to take on transforming technolo�gies, having evolved processes which sup�portincrementaltechnicalimprovements,with a view that they can later buy in ad�ditionalrequiredtechnologies.Someen�terprisingcompaniesarecurrentlytryingto avoid idea stagnation by encouragingclustersofstart�upsbusinesses�researchincubators where less conventional ideas can be evaluated, or through creatingsmaller, devolved research groups with greaterindependenceandautonomy.

The whole pharma sector is being squeezed by generic companies, who are predictedtohave~90%oftheglobalmar�ket of $820bn in 2009. many companiesno longer find “me toos” commercially viableandasaresultarefocusingondif�

ficult drug targets with high unmet medi�cal need, outside the traditional pharmacomfort zone. Within the non�genericmarket, biologicals and antibodies arecompeting with NCEs for market share with the expectation that within ten years they will share the revenues equally2.Weare now seeing the emergence of new therapeutics, occupying the last remain�ing untouched theoretical drug space (fig 2). Drug space can be defined by molecu�lar size going from aspirin (picometre)to stem cells and regenerative medicine(micrometre or larger). Early examplesof larger drugs are nanoparticulates butnucleic acid�based therapeutics and re�generative medicine are just aroundthecorner. inparallel, therearestepstobroaden the market for the major drugclasses, such as ncEs inhibiting protein�

Figure 1 Diversification of “Drug Space” after the Biotech Revolution

protein interactions and antibodies thattackleintra�cellulartargets,currentlytheexclusiveprovinceofncEs.�herapeuticsare set to become much more eclecticwith the emergence of these new modali�ties. These are difficult challenges and thepharmaandSmEsectorisonceagainlookingfor inspirationfromacademia. ithashappenedbefore�notso longago�and gave rise to the biotech sector.canhistoryrepeatitself?

The Communications Dilemmaaproblemforbothacademiaandindustryisthelackofacommonunderstandingofthetechnicalissuesfacingdrugdiscoveryinthe21stcentury.doesacademia,espe�ciallyinEurope,haveanunderstandingofwhat is required in the present pharma en�vironment?�hesideshavedriftedapart,

EuropEanJournaLofnanomEdicinE2009Vol.2issue2 22

Drug Delivery Systems

nanomedicines

Vaccines

Biologicals

antibodies

ncEs

Figure 2 New Molecules Entitiesapprovedin2008byclass

perhaps due to industrial arrogance, butthere is also a perception that academicfreedom may be adversely impacted byworking on industrial applied research. certainly “Blue Sky” research should beencouraged � national research councilsand especially the peer review system probablydonotdoenoughtosupportthistype of activity. However there is within Europe a considerable body of fundedresearch that purports to be therapeuticand hence is applied research by defini�tion.appliedresearch,academicfreedomand“BlueSky”aretotallycompatibleandthese seemingly different and antagonis�ticapproachescanandmustbebroughttogether to produce radical high valuenew products.

How is this publicly-funded applied re�search strategically guided? in general itis down to the individual and the reviewer to judge whether the project objective fits into an industrial wish list. European �echnologyplatformshavebeenbroughtinto being by the Ec to advise, from anindustrial perspective, how tax payers’ research monies should be best spent.The Commission’s objective in the thera�peutics sector is both to fund academicresearch into new treatments and diag�nostics for diseases and to support Eu�ropean pharmaceutical industries efforts to be more competitive. The author’s ex�perience with the ETP on Nanomedicine3has shown that there is a significant need for more information flow from industry to academia. Without this information,much of Europe’s funded research will benon�translatabletoindustry,theclinicandultimatelythepatient;consequentlymuch needed new ideas will not reach the pharmasector.

The Value of Communication Whilst academics are familiar with their

departmentsbeinggivenastarratingbypeer review, most will be unaware that industry often reviews departments and individuals independently. Industry will not only choose to work with the best sci�entists but also those most informed onwhat is needed to translate an idea to a product.Breakthrough research requires a lessconservative mentality on the part ofgrant reviewers. Traditionally, chemistry reviewers have strongly favoured extrap�olations from the published literature.Equallyitrequiresinformedinvestigatorsto recognise where discoveries or seren�dipity can produce step changes for sci�enceandindustry.So we need to:

•Selectfertileresearchareas(forindustry)•Exploregoodideasorhypotheses•Leave room for serendipity and most

importantlyitsrecognition.•createmoreacademic�basedkeyopin�

ion leaders with industrial knowledge

What is in it for Industry?providinginformationforacademicscien�tistsisatime�consumingactivitythatcon�flicts with mounting pressures on many Business development departments andtheirindustrialr&dcolleagues.academ�ics need to know whether their idea is ap�plicable, its manufacturability and why it has not been done before.�hey need tounderstand the industrial requirementsand their project’s potential position in the industrial r & d process. regretta�bly,muchofthisexperiencehasnotbeenpublished, neither is it academically ac�cessible, and the commercial framework is changing all the time. fortunately thetraditional secrecy barriers are slowly be�ing pulled down, as the need to search glo�bally for ideas takes root in forward look�ing companies. For academia the benefits

are fiscal, highly rated publications being in the driving seat for new industrial sec�tors and the identification of new health�care areas which can help patients.

What is in it for Academics? Whilst some individuals are perfectlyhappytojustpublishtheirresearch,somedepartments are increasingly trying tobuild bridges with industry. Their aim is to increase potential funding � but alsoto solve real healthcare problems.�hereare perhaps two approaches to applied research:

•Laissez faire(defaultstrategy)•amoreinformed(butnot managed)strategy.Given that academics by and large arenot familiar with the patent literature or current industrial priorities the formerapproach is somewhat wasteful, albeit it could produce a “Black Swan”4, a seren�dipitous discovery that would produce a paradigm shift. industrial experience isnot well documented, but knowledge of it could reduce time spent on research with nopossiblecommercial translatability. itisimportantthatcommercialinformationand experience is available to academ�ics,butthatnoattemptshouldbemadeto impose industrial management proc�esses. It is better to know and understand the perceived industrial roadblocks fromthe outset, rather than to travel hope�fully.

The known obstacles to the marketVeryoftentranslatabilityofacademicre�searchispredictablefromtheoutset,oriftherouteisnotsoclearthenatleasttheroad-blocks can be identified and strate�gies developed to overcome or bypassthem.�heauthorhascomeacrossmanysuch predictable and manageable road�blocks5, the following are purely illustra�tive!

Drug Pharmacokinetics�heimportanceofin vivodmpKstudiesisoften not appreciated by academics, es�peciallythoseofinorganicmaterialsthathave not commonly been into man, forexamplesilicaandgoldparticulates.Verysignificant work has been done on gold imagingandtherapy,butunlessthereareappropriate clearance studies these will remain in academic laboratories and will not be translated to the clinic. anotherarearequiringmuchmore in vivo work is polymertherapeutics;polymerclearance



being very difficult to monitor in vivoorinurine or faeces. Synthetic gene therapyvectorshavealsostagnatedbecauseofareluctancetocarryoutin vivostudies.

Drug StabilityIn the drug delivery field much effort has been expended on drug release systemsexploiting the lysosome’s lower pH, fol�lowing endocytosis of a delivery sys�tem.SuchsystemsoftenuseastabilisedSchiff’s base, consequently they are in�herently unstable to low pH. To produce a marketedproductfromsuchaconjugaterequires a simple reliable process and low exposure to water even at pH 7, to ensure thatbatchesareanalyticallyreproducibleto regulatory requirements. The field of antibody drug conjugates has wrestled with this problem for a decade and has now adopted a manufacturable stable linkage approach � this industrial lessonhasnottransferredintorelatedacademicresearch, often with disastrous results on scale�up.

Polymer Therapeutics�heuseofpolymersintherapeuticsispartof a multi�billion dollar sector.aggrega�tionofsomepolymerscouldleadtoana�phylacticshockorimmuneresponses,butpolymerself�associationstudiesareoftenleft until the pre�clinical stage.althoughnaturehasevolvednaturalpolymersnotto aggregate at low concentrations, com�mercial products are formulated at highconcentrations.

Industrial ContextOften the answer as to why a concept has notbeen implementedbefore is there isno appreciation of what the advantages must be. an example here is molecularimprinting of polymers to a therapeutictarget. �his technology requires a sub�strateataone�to�oneratiotoimprintanexpensive polymer. Such systems oftencompete with antibodies, which have much higher affinities and are cheaper andeasiertoscaleup.understandingtheindustrialcontextofyourresearch is im�portantifyouaretotakemakethemostof your efforts.

Cost DNA cages are an interesting way to en�trap a drug, which potentially could be re�leasedbyaparticularmrnarelevanttoaspecific disease. Nucleic acids are, howev�er,veryexpensivedrugdeliverymaterialswhen accessed by chemical synthesis. It is

likelythatthefutureofnucleicacidthera�peutics lies in other directions where they are the agent rather than the formulantordrugdeliverysystem.

Analytical ChallengesThe new types of drugs being discussed throw up many analytical challenges. Ana�lyticalchemistryisoftenregardedasrou�tine but with the advent of such diverse drugtypesitisachallengeforGmpandin vivo studies. It may even be a significant obstacletoprogressingtodevelopment;a point that is often not appreciated bythosenotdirectlyinvolved.

What can we do to improve industrial-academic communication?This is a challenge which has no easy solu�tion but it must involve more effort from the industrial sector. The following are ideasbuteachresearcherordepartmentshouldhavealocalstrategy.

• Better industrial peer review of applied researchproposals

• More effective industrial contacts with universities

• More industrialists connected with ma�jorresearchdepartments

•more sabbaticals in industry and vice versa

• A change in academic culture which en�courages and rewards real innovation andentrepreneurshipinEurope

•Eachacademicdepartmentdevelopinganindustrialliaisonpolicy

•an available source of information onindustrialpriorities

•industry should share its specialisedtechnologiesandexpertise

• Create “reverse symposia” on what in�dustry needs or what they do not know

SummaryforthepharmaceuticalsectortosurviveinEuropeitneedsthesupportofacademia.�hereisaneedformoreindustrialexperi�encetobeavailabletotheacademicsec�tor, which should be encouraged to find original solutions to serious healthcareproblems.Equally for academia to flourish in Europe it needs the continuation of its sciencebased industries and a dialogue with in�dustry.Another question which needs answering is whether the US is better than Europe in producingtranslatableresearch?perhapsthe streak of entrepreneurship is not soevident in European culture, for cultural

or fiscal reasons. This aspect will not be changedbyspendingyetmoremoneyonresearch,anditvariesquiteconsiderablybetween European countries. R & D is now a global activity within the industrial sector�nolongerrestrictedtoitsnationalorcontinentalboundaries.

References 1.www.archemix.com/website/_popup_press_release.php?release=542.Witty,a.Scrip20083.www.etp-nanomedicine.eu/public4. The Black Swan, Nassim N Taleb, Penguin 20085.Eaton,m.naturematerials6,251�3,2007



Review paper

Cochlear Implants and Inner Ear Based Therapy

ciorbaa.1, Astolfi L.1,Jollyc.2,martinia.1(doi10.3884/0002.2.6)

AbstractHearinglossresultingfromhaircelldamageisirreversiblebecausethehumanearisincapableofregen�erating or repairing sensory elements following severe injury. Moreover, hair cell loss leads to secondary degenerationofspiralganglionneurons.cochlearimplantationisarecognizedtreatmentforadultsandchildren with severe to profound cochlear hearing loss. Improvements in nanotechnologies and progress ininnerearregenerationresearchmayfacilitatethedevelopmentofdevicessuchascochlearimplantscombined with local drug/gene devices or stem cell therapy. New devices could improve patient auditory performancebyinducingcellular/neuronalprotectionand/orregeneration/resproutingandbypromotingcochlearregeneration. martini,aEurJnanomed2009,2.2:25�28

Key Words: bio�interactivecochlearimplants,cochlearimplants,innerearregeneration,innereartherapy,nanoparticles.

Introductionauditoryhaircell(Hc)andspiralganglionneuron (SGn) loss cause permanenthearing impairment in humans becausethesensoryelementsdonotregenerate.Sofar,adultstemcellsinthemammalianinner ear have only been found within themouseutricularmaculae,andseveralattempts have performed in order toachievesuccessfulinnerearregenerationstrategies(1,2).it is estimated that 900 million peoplein the developed world will be affected by hearing loss by 2050 (3). presentlyhearinglossisconsideredthethirdmostcommon chronic condition amongst theolder population after arthritis and highblood pressure (http://www.pamf.org/health/healthinfo/reutershome_top.cfm?fx=article&id=44363). Furthermore, the number of US citizens with hearing loss has doubledduring the past 30 years (4,5,6).independentresearchreportedthat28.6 million americans had an auditorydisorder in 2000 (7).as manyas738,000individuals in the u.S. have severe toprofoundhearingloss,andalmost8%ofthem are under the age of 18 years (8).according to recent estimates, cochlearimplants (cis) have been implantedin more than 110,000 deaf adults andchildren worldwide (4), providing deaf patients important auditory cuesnecessary for auditory awareness and speech perception via direct stimulationoftheauditorynerve.

�hereisconsiderablesocialandeconomicdemand for new therapeutic approaches for hearing impairment. it has been

reported that, so far, investment inhearingresearchhasbeenmodestrelativeto measured social and economic costs(3). at the same time, expectations forthe development of new therapeutic tools haveincreasedbecauseresearchhasbeenconductedoninnereargene and stem cellgeneandstemcelltherapy.�hedevelopmentofhearinglosstreatments at the cellular and molecularlevel is just beginning. in the future,improvements in nanotechnologies, aswell as progress in inner ear regeneration, may yield new devices such as CIs in combination with local drug/gene devices orstemcelltherapy.

Implant electrode damageBefore aci may be considered for innereartherapy,preventionandreductionofelectrode insertion trauma and resultinghearing loss is necessary. Two primary typesofcochleardamageareassociatedwith electrode insertion: early and late in�nereardamage.Early inner ear damage.Histologicaleval�uation of the cochlea after ci electrodeinsertionintothescalatympaniofcadav�er temporal bones indicates that therecanbeimmediatedamagetoseveralco�chlear structures including the basilarmembrane, spiral ligament, osseous spi�rallamina,andstriavascularis(7,8,9,10,11,12).recentresearchsuggeststhatthedegreeofdamagetothelateralcochlearwall during surgery may greatly influence the amount of new tissue formation post�operatively(13).inaddition,intracochleartrauma increases with deep electrode in�sertions(14).mechanicalinsertiontraumaitselfcanalsoinducetheactivationofin�

flammatory and cell death pathways (7). Even when a very soft surgical technique is used for electrode insertion, hearingloss progression following initial insertion traumamaybemeasured(8).�hese observations may lead to the hy�pothesisthatelectrodeinsertionitselfin�ducestrauma/changesonacellular levelthat still need to be clarified (8). If mac�roscopic mechanical damage to the co�chlea could be prevented or reduced byimprovements of both, electrode designand surgical technique, further studiesareneededinordertoclarifythecochlearcellularandmoleculardamagesaftertheinsertion.Late inner ear damage. intra�cochlear“scarring”isapossiblesequelaofcochle�arimplantation(10).althoughthedegreeof tissue growth around the electrode variesamongpatients,scarring isgener�allymoresubstantialinthebaseandlesswell developed in the apical region (10). itisspeculatedthatthismaybebecausethe electrode is thinner towards its tip and doesnotreachthecochlearapex(15).Fibrous scar tissue within scala tympani

1audiologydepartmentandHearingprotectionplasticityandregenerationLaboratory,universityofferrara2mEdELHearingimplants,innsbruck,austria

correspondingauthor: prof.alessandromartiniu.o.diaudiologia,universityofferraracorsoGiovecca203,44100ferrara,italyEmail:[email protected]�el:+39�0532�237451fax:+39�0532�236887


Review paper

aftercochlearimplantationcouldchangethepassivemechanicalpropertiesofthebasilar membrane and HCs, as well nerve fibre integrity, thus affecting preservation of residual hearing (16, 17). Even in cas�esofmilddamage,intracochleartraumamay result in reduced numbers of func�tional peripheral dendrites or spiral gan�glion cells. This could influence stimula�tion efficiency along the length of the im�planted array.�he implications of inser�tion trauma may be even more seriousfor patients with residual hearing than for profoundly deaf patients with little or no residual hearing. resolving, reversing orcontrollingtheseproblemsevenpartiallymakes the consideration of cochlear im�plantation as a first step in the cochlear “regenerative”processpossible.

A “biointeractive cochlear implant” (bci)The development of new CI devices in�cluding the coexistence of an electrodeand a drug delivery system have beenreportedrecently inthe literature.�hereare two aims of the research: (i) to reduce cochlear insertion damage and (ii) to di�rectlysupportandincreasethesurvivalorevenregenerationof innerearcells/neu�rons (18,19,20,21). interest indevelop�ing a “bioactive” or “biointeractive” co�chlear implant (BCI) is growing because a BCI device could offer new possibilities forcochlearcellularandmolecularthera�py(fig1).

Fig 1. Cochlear section (10 x, mouse). Exogenousneurotrophicfactors/genes/stemcells,introducedintotheinnerearthroughabio�interactivecochlearim�plant,canreachhaircellsorspiralgan�glionneurons.

Ideally a BCI would minimize electrode insertion trauma and preserve residu�al hearing by reducing cochlear traumaduring electrode insertion and reducing/guiding cochlear tissue responses to theelectrode, particularly avoiding scar orfibrous tissue formation. Furthermore, a Bci should stimulate cochlear neuronaltissueinordertoimproveciperformanceandinducecellular/neuronalconservationand/orregeneration.in our opinion, main characteristics of aBci should be: (i) avoidance and reduc�tionoflocaltoxicity,allergicreactions,orinfections, particularly biofilm formation; (ii) drug release regulation; (iii) possibili�tyofexplantingthelocaldeliverysystemwithout electrode avulsion; (iv) refilling. delivery system development requirescareful preclinical evaluation to assesssafety,tolerabilityandperformance.par�ticularly, it is necessary to determine inanimal models whether the delivery sys�tem conjugated to the cochlear implantallows local administration of drugs as well as genes or cells. furthermore,toimprovetheperformanceofaBci,ofparticularinterestisthedevel�opmentofnanoparticlesaspropercarri�ersfortheintracochlearadministrationoftherapeuticsupplies.oncereleasedintheinnerear,nanoparticlesshouldactasspe�cific vectors; they should be able to target a peculiar cell population, as well as pos�sess several characteristics such as be�ingbiodegradable,invivotraceable,andequipped with controlled release activity (22). Several european partner laborato�ries are already working on this matter inajointprojectfundedbytheEuropeancommission (nanoear consortium �for�mer EuBioEar program�, www.nanoear.orgbothcoordinatedbyilmaripyykkö).in the future then, especially if all theserequirements will be satisfied, BCIs could be used as a resource for cochlear andneuronalstructureregeneration.

Possible tools for bci inner ear therapyDrugs.�hepurposeofdrugsistopreserveandregeneratetheinnerear.directinfu�sion of exogenous neurotrophic factorsintotheinnerearresultsintheprotectionofSGnsandHcsinexperimentalanimals.Severalresearchershavereportedthatacombinationofelectricalstimulationandneurotrophic factor administration has astronger influence on SGN survival than electrical stimulation alone (23, 24, 25,26,27,28).administrationofneurotroph�ins has been shown to enhance the re�

sproutingofauditoryperipheralprocess�es. in ototoxically deafened guinea pigsSGN peripheral processes were observed and monitored following treatment with Brain�derivedneurotrophicfactor(Bdnf)plusneurotrophin3(n��3),andincreasedresprouting was described in the bas�alturnofthecochleaclosetothesiteofneurotrophinapplication(29,30).

Genes. �he purpose of gene�therapy istopreserveandregeneratetheinnerear.preservation and regeneration of SGnsareimportantresearchareasinthedevel�opment of effective gene-based strate�gies for inner ear diseases.Several stud�ies have examinedthe potential of genetherapy;although,forthemostpart,theyhave focused on vectors and delivery is�suesintotheinnerear(31,32,33,34,35,36).astheunderstandingofgenethera�py improves, investigators will focus on targeted,single�genereplacementtreat�ment of inner ear disorders and will ap�plygenetherapytosensorycell(HcsandSGns)replacementortoensuregenein�duced growth of neurites toward the elec�trodes. However, there also are concerns regarding gene therapy use, particular�ly regarding the safety of virus vectors.careful evaluation of laboratory animalwork and more basic research are neces�sarytofurtherunderstandtheapplicabil�ityofgenetherapytocis.

Stem Cells (SCs). �hepurposeofScsistoregenerate the inner ear. SCs provide new hope for treating sensorineural hearingloss and inner ear disease. In the past five years, a few research groups have con�centrated their efforts on SCs. Research on mammals to date has found residentstem cells only in the utricle (37) with pre�liminary studies showing that stem cells can replace lost SGNs as well as HCs. Sev�eralsourcesofScshavebeenusedtore�generate cochlear cells and SGns; em�bryonic SCs are one of the most powerful stem cell types tested (38, 39). more re�cently, umbilical cord and bone marrow stem cells (40, 41) have been reportedtobecapableofgeneratingSGns.inad�dition to understanding which subtype of Scc is most suitable for cochlear andspiralganglionregeneration(particularlydemonstrating their functionality), it isalso necessary to understand if and how newly formed neural peripheral processes or neurons might interact within the co�chlea, and, most importantly, couple with aci.


Review paper

Conclusionsminimizing cochlear electrode damage,stimulating the spiral ganglion neuralpopulationandachievingcochlear/neuro�nalregenerativeprocessesareimportantgoalsforthefuturedevelopmentofcis.inparticular, the development of an effec�tiveandadaptableelectrodearraycouldoptimizeinnereartherapeuticstrategiesalthoughcarefullaboratoryandpreclini�calevaluationisrequiredtobringdeliverysystemstofruition.In our opinion, the development of new CI systemscouldrepresentareliabletooltointroduce into clinical practice new tech�nologies/advancesininnerearresearch,including regenerative strategies, in anearfuture.

References1.LiH,roblinG,LiuH,HellerS.Genera�tion of hair cells by stepwise differentia�tion of embryonic stem cells. proc natlacadSci2003;100:13495–500.2.LiH,LiuH,HellerS.pluripotentstemcellsfromtheadultmouseinnerear.na�ture2003;9:1293–9.3.Vio mm, Holme rH. Hearing loss andtinnitus: 250 million people and a uS 10billion potential market. drug discovery�oday,2005,10:1263�1265.4.fallonJB,irvinedr,ShepherdrK.co�chlearimplantsandbrainplasticity.Hearres.2008apr238:(1�2):110�7.5.ries,p.W.(1994).prevalenceandchar�acteristics of persons with hearing trou�ble: united States, 1990�91. nationalcenter for HealthStatistics.Vital HealthStat10(188).6.Benson,V.,&marano,m.a.(1995).cur�rent estimates from the national HealthInterview Survey, 1993. National Cen�terforHealthStatistics.VitalHealthStat10(190).7.Kochkin,S.(2001,december).marke��rakVi:�heVaanddirectmailsalessparkgrowth in hearing aid market. The Hear�ing Review, 8(12): 16-24, 63-65.8. Blanchfield, B.B., et. al. (2001). The se�verely to profoundly hearing�impairedpopulation in the united States: preva�lenceestimatesanddemographics.Jour�naloftheamericanacademyofaudiolo�gy,12,183�189.9. Eshraghi aa. prevention of cochle�ar implant electrode damage. curr opinotolaryngol Head neck Surg. 2006 oct;14(5):323�8.10.Eshraghiaa,VandeWater�r.coch�lear implantation trauma and noise�in�ducedhearingloss:apoptosisandthera�

peuticstrategies.anatrecadiscovmolcellEvolBiol.2006apr;288(4):473�81.11.WellingdB,Hinojosar,GantzBJ,LeeJ�.1993.insertionaltraumaofmultichan�nel cochlear implants. Laryngoscope103:995–1001.12.nadolJr.,J.B.,Shiao,J.Y.,Burgess,B.J.,Ketten,d.r.,Eddington,d.K.,Gantz,B.J.,Kos,i.,montandon,p.,coker,n.J.,rolandJr.,J.�.,Shallop,J.K.,2001.Histopatholo�gy of cochlear implants in humans.ann.otol.rhinol.Laryngol.110,883–891.13. Clark GM, Shepherd RK, Dowell RC (1988) Histopathology following cochlear implantationinapatient.acta otolaryn�actaotolaryn�golSuppl(Stockh)106:44814.Balkany�J,HodgesaV,Gomez�marino,Birdpa,etal(1999)cochlearreimplan�tation.Laryngoscope109:351–35515. Li pm, Somdas ma, Eddington dK,nadol JB Jr. analysis of intracochlearanalysis of intracochlearnew bone and fibrous tissue formation in human subjects with cochlear im�plants. ann otol rhinol Laryngol. 2007oct;116(10):731�8.16. adunka o, Kiefer J. impact of elec�trode insertion depth on intracochleartrauma. otolaryngol Head neck Surg.2006Sep;135(3):374�82.17. Skinner, m.W., Ketten, d.r., Holden,L.K., Harding, G.W., Smith, p.G., Gates,G.a., neely, J.G., Kletzker, G.r., Bruns�den,B.,Blocker,B.,2002.c��derivedesti�mationofcochlearmorphologyandelec�trode array position in relation to word recognition in nucleus�22 recipients. J.assoc.res.otolaryngol.3,332–350.18.choi,c.H.,Spector,a.a.,oghalai,J.S.,2004.acochlearmodeldesignedtotestthe effect of modulating outer hair cell biophysical properties on basilar mem�brane mechanics, Twenty-Seventh Annu�almidWinterresearchmeetingoftheas�sociationofresearch inotolaryngology,vol.27.florida,pp.346.19. choi cH, oghalai JS. predicting theeffect of post-implant cochlear fibrosis on residualhearing.Hearres.2005Jul;205(1�2):193�200.20.Hochmair i, nopp p, Jolly c, SchmidtHochmair i,noppp,Jollyc,Schmidtm,Schösser H,Garnhamc,anderson i...mEd�ELcochlear implants: State of theartandaGlimpseintothefuture.�rendsin Amplification Volume 10 Number 4 De�cember2006201�220.21. rebscher SJ, Hetheringtonam, Sny�der rL, Leake pa, Bonham BH. designand fabrication of multichannel cochlearimplants for animal research Journal ofneurosciencemethods166(2007)1–12.22. Volckaerts B, corless ar, mercan�

zini a, Silmon am, Bertsch a,Van Him�beeck C, Wasikiewicz J, Vanden Bulcke M, Vadgamap,renaudp.�echnologydevel�opments to initiate a next generation ofcochlear implants. conf proc iEEE Engconf proc iEEE EngmedBiolSoc.2007;1:515�8.23.Jollyc.,Garnhamc.,mirzadehH,�ruyE.,martinia.,KieferJ..Electrodefeaturesfor hearing preservation and drug deliv�ery strategies.advances inorL, Karger,inpress.24. Richardson T, Noushi F, O’Learly S. In�nereartherapyforneuronalpreservation.audiolneurotol,2006,11,343�56.25.EvansaJ,�hompsonBc,WallaceGG,Millard R, O’Leary SJ, Clark GM, Shepherd rK, richardson r�. promoting neuriteoutgrowth from spiral ganglion neuron explants using polypyrrole/Bdnf�coatedelectrodes. J Biomed mater resa. 2008Sep23.26.pettingillLn,richardsonr�,WiseaK,O’Leary SJ, Shepherd RK. Neurotroph�ic factors and neural prostheses: poten�tial clinical applications based upon find�ings in the auditory system. iEEE �ransBiomed Eng. 2007 Jun; 54 (6 pt 1):1138�48. Review.27. Shepherd rK, coco a, Epp SB. neu�rotrophins and electrical stimulation forprotection and repair of spiral ganglionneurons following sensorineural hearing loss.Hearres.2008aug;242(1�2):100�9.28.richardsonr�,�hompsonB,moultonS, Newbold C, Lum MG, Cameron A, Wal�lace G, Kapsa R, Clark G, O’Leary S. The effect of polypyrrole with incorporated neurotrophin�3onthepromotionofneu�rite outgrowth from auditory neurons. Biomaterials.2007Jan;28(3):513�23.29. mcGuinness SL, Shepherd rK. Exo�genousBdnfrescuesratspiralganglionneuronsinvivo.otolneurotol.2005Sep;26(5):1064�72.30. pettingill Ln, richardson�,WiseaK,O’Learly S, Shepherd RK. Neurotrophic factors and neural prostheses: potentialclinical applications based upon findings intheauditorysystem.iEEE�ransBiomedEng.2007June;54(6):1138�1148.31.rejalid,LeeVa,abrashkinKa,Huma�yun N, Swiderski DL, Raphael Y. Cochlear implantsandexvivoBdnfgenetherapyprotectspiralganglionneurons.Hear res.Hearres.2007Jun;228(1�2):180�7.32. Kawamoto K, Ishimoto S, Minoda R, BroughdE,raphaelY.math1 gene trans�math1genetrans�fer generates new cochlear hair cells in mature guinea pigs in vivo. J neurosci.2003Jun1;23(11):4395�400.


Review paper


33. praetorius m, Baker K, Weich cm,plinkert pK, Staecker H. Hearing preser�vation after inner ear gene therapy: theeffect of vector and surgical approach. orLJotorhinolaryngolrelatSpec.2003Jul�aug;65(4):211�4.34.Liduanm,Bordet�,mezzinam,Kahna, ulfendahl m. adenoviral and adeno�adenoviral and adeno�associated viral vector mediated genetransfer in the guinea pig cochlea. neu�roreport.2002Jul19;13(10):1295�9.35. Kawamoto K, Oh SH, Kanzaki S, Brown n,raphaelY.�hefunctionalandstructur�aloutcomeofinnereargenetransferviathe vestibular and cochlear fluids in mice. mol�her.2001dec;4(6):575�85.36. Wenzel Gi, Xia a, funk E, Evans mB,palmer dJ, ng p, pereira fa, oghalai JS.Helper�dependent adenovirus�mediatedgenetransferintotheadultmousecochlea.otolneurotol.2007dec;28(8):1100�8.37. praetorius m, Baker K, Brough dE,plinkertp,StaeckerH.pharmacodynam�ics of adenovector distribution within the innereartissuesofthemouse.Hearres.2007may;227(1�2):53�8.38. oshima K, Grimm cm, corrales cE,Sennp,martinezmonederor,GéléocGS,Edge A, Holt JR, Heller S. Differential dis�tributionofstemcellsintheauditoryandvestibularorgansoftheinnerear.Jassocresotolaryngol.2007mar;8(1):18�31.39.JeonSJ,oshimaK,HellerS,EdgeaS.Bone marrow mesenchymal stem cells are progenitorsinvitroforinnerearhaircells.molcellneurosci.2007Jan;34(1):59�68.40.revoltellarp,papiniS,rosellinia,mi�chelinim,franceschiniV,ciorbaa,Berto�lasoL,magossoS,HatzopoulosS,LoritoG,Giordanop,SimoniE,ognioE,cillim,Saccardi R, Urbani S, Jeffery R, Poulsom r, martini a. cochlear repair by trans�plantation of human cord blood cd133+cells to nod-scid mice made deaf with ka�namycinandnoise.cell�ransplant.2008;17(6):665�78.


MRI of the Cochlea with Superpara-magneticironoxidenanoparticlescomparedtoGadoliniumchelatecontrastagentsinaratmodeldennispoe1,JingZou1��,WeikaiZhang1,JianQin2,usamaaboramadan3,andreafornara2,mamounmuhammed2,ilmaripyykkö1(doi10.3884/0002.2.7)

AbstractObjectives:Superparamagneticironoxidenanoparticles(Spions)havebeenproducedintheanticipationthat they could be traced using high resolution MRI while also serving as a delivery vehicle for clinical therapeutic agents. These particles were evaluated in vivointheratmodeltostudytheirabilitytoenhancemriimagingofthecochleaandtracethemthroughtheirdistribution.Methods: SPIONs were manufactured with high temperature decomposition methods and coated with pluronic® F127 copolymer (POA@SPIONs). The relaxation times were measured in phantom tests in a 4.7 Tesla MRI scanner. POA@SPIONs or Gadolinium DOTA were delivered to rats by I.V injection, round window membrane permeation by intratympanic application, and POA@SPIONS were also given by intracochlear injection through a catheter. Baseline pretreatment MRI scans or contralateral untreated ears were used for controls. MRI scans of the cochleae using T2-weighted sequences; rapid acquisition with relaxation enhancement (RARE) sequences and T2*-weighted sequences; fast low angle shot (FLASH) sequences that were obtained repeatedly up to 4 hours.Results: Gadolinium-DOTA given IV or IT produced strong enhancement within the perilymph space of the cochlea and yielded high resolution images that differentiated the scala tympani and scala vestibuli from the dark scala media in T1 weighted images. POA@SPIONs enhanced 1/T1 and 1/T2 relaxation rates and were effective in reducing T2 relaxation times. As a result, POA@SPIONs quenched T2 weighted signal intensities in the cochlear fluids that created voids wherever the particles distributed or aggregated. Decreases in intracochlear signal intensities were observed using the intracochlear catheter injection. Intratympanic applicationdidnotconsistentlydemonstratelossofsignalintheperilymphatic space. intravenous injectionatic space. intravenous injectionspace.intravenousinjectionresulted in signal decreases over the cochlea and other surrounding tissues without any specific uptake in thecochlearperilymphorendolymphcompartments.Conclusions: POA@SPIONs were demonstrated to image the cochlea and offer the possibility for future differential imaging between the perilymphatic and endolymphatic compartments. The ability to distinguish these scalae could translate into developing a critical new understanding of clinical inner ear diseases and provide a potential means for delivering treatments with a labeled carrier vehicle while tracing their course throughtheinnerear. poe,dnanomedicine2009,2.2:29�36

Keywords: animal, contrast agent, inner ear, fluids, MRI, nanotechnology, neuronanimal, contrast agent, inner ear, fluids, MRI, nanotechnology, neuroncontrast agent, inner ear, fluids, MRI, nanotechnology, neuron

IntroductionGreat advances in magnetic resonanceimaging (mri) of the cochlea have beenrealised with the recent introduction of intratympanic (i�) and high dose intra�venous(iV)administrationofGadoliniumchelates(Gd)inanimalmodelsandinhu�mans.(1)i�applicationsareaccomplishedbydirectlydeliveringGdtothemiddleear

space through the tympanic membrane.�hereisconsiderableinterestinimaginginner ear disorders, such as sensorineu�ral hearing loss and Menière’s disease and treating these conditions with drugs car�riedbynanoparticlesthataretargetedforspecific sites within the cochlea. Nanopar�ticles are being developed that incorpo�ratemricontrastenhancerstotracether�

apeuticagentsthroughtheircourseinthebody and to visualise their effects with conventionalormolecular(ie.cellularandsubcellular)scaleimaging.(2)�hecochleacontainsthehearingsensoryportionoftheinnerearandisasmallspi�ralsnail�shapedorganmeasuring9mmindiameteratthebaseand5mminheightinhumans.(3) It is contained within the otic

preclinicalnanomedicine


capsule of the temporal bone, the dens�estboneinthebody.itsanatomyconsistspredominantly of fluid-filled spaces that contain endolymph and perilymph. En�dolymphisseparatedfromperilymphbywatertight membranes in order to main�tain the electrolyte homeostasis that isessential for the normal function of theorgan of Corti, which is the site of the sen�sory neuroepithelium (4,5). changes intheintegrityofthesemembranes,inthepermeability of the stria vascularis, or intheabsorptivefunctionsoftheendolym�phaticsacmayproduceanumberofdis�orders, such as endolymphatic hydrops,thatcanproducepathologicalchangesinthe sensory endorgans. �reatments forthe cochlea are being developed in which nanoparticles will be targeted for specific areas or cells within the inner ear. mr imaging of the ultrastructure of thissmall organ housed in dense bone andsubdivided into different fluid-filled com�partments, remains a challenge that isbeing addressed with the use of higher field strength magnets. Increasing the magnetic field strength improves spectral resolution, signal�to�noise and contrast�to�noise ratios, reduce scan acquisitiontimes, and can be adapted to molecularimaging.(6,7)clinical mr imaging uses the principlesof proton nuclear magnetic resonance(nmr)andmeasurestherelaxationtimesto return to baseline energy states oftissues and fluids after the protons are aligned in a strong magnetic field and perturbed by radio frequency emissions.contrast agents improve the sensitivityof the measurements and enhance thedifferences between tissues to highlight regions of interest. �he most common�lyusedclinical contrast agentsarepara�magnetic Gd chelates, which have longi�tudinalrelaxivity(r1)valuesrangingfrom10 to 20 s−1 mm−1 (8,9). contrast agentsinteract with adjacent protons to influ�ence their signal characteristics and theeffects are measured as the longitudinal relaxivity (r1) and transverse relaxivity(r2). Higher r1 values result in brighten�ingorenhancementoftissueandhigherr2 values result in darkening. in�1 (lon�gitudinal relaxation time) weighted se�quences,thehighr1valuesofGdproducedesirable bright positive contrast effects but the r1 decreases rapidly in higherfield strengths. (10) Another disadvan�tage of Gd chelates is that they requiremicromolar concentrations for visualisa�tion whereas molecular imaging requires

sensitivitiesinthenanomolarrange.forthese reasons, efforts are being made to synthesize nanoparticulate contrastagents with significantly higher relaxivi�ties(2,11).Someexamplesofthesenovelnanoparticlesareliposomes(12,13,14)ormicelles(15,16)thatcontainparamagnet�icGdchelate,nanoparticlescreatedfromGdoxide(6),andsuperparamagneticironoxidenanoparticles(Spions).(17,18)in order to gain access to their intendedtargets within the various compartments of the inner ear, particles must passthrough barriers such as the round win�dow membrane for nanoparticles applied i�,theblood�perilymphbarrierforiVde�livery.additionally,thereisthebarrierofthe plasma membrane. Gd chelate hasbeen recently shown in animal models andinhumanstopassthroughtheroundwindow membrane and blood-perilymph barriers to significantly accumulate with�in the perilymph space. �he perilymphcompartments (scala tympani and scalavestibuli) appear with high signal on T1 weighted images compared to the persis�tentlydarkscalamediacontainingtheen�dolymph.allthreecompartmentsappearbright and cannot be differentiated on T2 weighted images. (1)Differential demonstration of the endo�lymph and perilymph compartments with MRI was accomplished in 1999 using high dose (1.5mmol/kg) gadolinium�diethyl�enetriaminepentaacetate�bis�methylam�ide (Gd-DTPA-BMA), which has a small molecular weight of less than 2500 Da, administered iV to guinea pigs and visu�alised with a 4.7 Tesla (T) Bruker scanner (19). using 3d imaging techniques theywere able to demonstrate even further details of the normal fine structure of the cochleaandvestibularorganguineapigs.(20) Wesubsequentlydemonstratedthedynamics of the vascular origin of peri�lymph in guinea pigs using iVGd�d�pa�BMA (1.5mmol/kg) enhanced MRI with a 4.7 T Bruker scanner. Following injec�

tion, gadolinium enhancement began toappear in the perilymph within the sca�la tympani and by one hour it had filled the perilymphatic spaces. �hese resultsserved to confirm theories that perilym�phatic fluid is derived from the cochlear blood supply via the cochlear glomeruliwithin the modiolus, which are in close proximitytothescalatympani(21).peri�lymphisalsobelievedtoderivefromcap�illaries within the spiral ligament, but that was not conclusively demonstrated in the study.Imaging with IT Gd has been investigated asanalternativetohighdoseiVdelivery.inhumans,systemicallyadministeredGdhaspossiblerenaltoxicityrisksthatmaybe dose related and nephrogenic Sys�temicfibrosisandnephrogenicfibrosingdermopathyhavebeenreportedincaseswith pre-existing renal failure. (22) Medi�cationsintroducedthroughthetympanicmembraneintothemiddleearappeartoentertheinnerear,probablythroughtheround window membrane. In guinea pigs, we showed obvious uptake of Gd-DTPA-Bmainbothcochlearandvestibularperi�lymph at 10 min after i� administration.Contrast within the perilymph appeared to progress from the scala tympani tothemodiolous,andthentothescalaves�tibule beginning in the basal turn andslowly progressing distally. These results suggestthatthecommunicationofperi�lymph between scala tympani and scala vestibuli is mainly through the vascula�tureregionofmodiolus.(23)In humans, we have reported imaging of the inner ear after i� administration ofGd-DTPA-BMA in two cases of Menière’s disease and two cases of sudden idiopath�icsensorineuralhearingloss.mrimageswere obtained at 2 hours, 12 hours, or 24 hours after injection with a 1.5T magnet using conventional T2 weighted images fortheearlierscansandtherecentscansemployed a Siemans 3� machine usingfast spin�echo sequences. results at 2

1departmentofotolaryngology,universityof�ampere,medicalSchool,fm1,3rdFloor, Biokatu 6, 33520 Tampere, Finland. Email: [email protected],royalinstituteof�echnology(K�H),Electrum229,Isafjordsgatan 22, SE-164 40 Kista, Stockholm, Sweden3 Experimental mri Laboratory, department of neurology, Helsinki universitycentralHospital,fin�00029HuS,Helsinki,finland

correspondence to: dr. J. Zou, department of otolaryngology, university of�ampere, fm1, 3rd floor, Biokatu 6, 33520 �ampere, finland; phone: +358 331164129;fax:+358335517700;Email:[email protected]

Eu research ip: nmp�2004�3.4.1.5�1 nanotechnology�vectors for targeted drug nmp�2004�3.4.1.5�1 nanotechnology�vectors for targeted drugandgenedelivery



hours showed bright enhancement of the perilymph within the vestibulum, lesser intensity in the scala tympani and scalavestibuliinthebasalturnofthecochlea,and no uptake in the scala media. In two cases, the enhancement was uniformly reduced inallpreviouslybrightsitesandthe apex was not filled. In two other cases there was increased signal change in the whole cochlea at both 12 and 24 hours, suggesting that there may be some se�questration of the contrast over timewithin the cochlea. In the cases with Me�nière’s disease, dilation of the scala media was observed, consistent with endolym�phatichydropsandinoneofthesecases,there was a distinct absence of uptake in thevestibulethatcouldbeduetohydrop�icblockageoftheperilymphpassageintothat compartment. (1,23) Naganawa et al. confirmed these findings reporting de�flection of the basilar membrane toward thescalatympani,presumablyduetoen�dolymphatic hydrops, in 11 patients with Menière’s disease imaged with a 3 T MRI scanner using 3d�fLair (fluid attenua�tion inversion recovery) sequences af�ter intratympanic administration of Gd(24).imaging of the cochlear compartmentshas been achieved but for dynamic im�aging of nanoparticles and their effects within the cochlea, there is a need for still higher resolution of the cochlear ultra�structural with molecular imaging that will require high magnetic fields, high relaxivity contrast agents, and appropri�atesequences.Giventhelimitationsthathave been reviewed with Gd chelate, we have done initial investigations with SPI�onstoimagethecochlea.SomeSpionsincurrentclinicalusehaver2valuesrangingfrom50to>600s−1mm−1(25,26) and the values are stable with in�creasing magnetic field strengths. Similar to Gd agents, the r1 values decrease with increasing field. (10) Thus, in high mag�netic fields of 3.0 T or more, SPIONs will retain their strong darkening effect on T2 (transverse relaxation time) weighted se�quences, which would be seen in contrast tothebrightsignalordinarilydetectedinfluid spaces. SPIONs targeted for specif�ic cell types may be useful for differential imagingsomeoftheultrastructuralcom�partments within the cochlea.WehaverecentlycreatedanoveltypeofSPION manufactured with high temper�ature decomposition methods and con�structed from iron oxide nanoparticleswith a hierarchical coating consisting of a

surfacelayerofpluronic®f127copolymer(pf127)thatoverliesa layerofoleicacidonthesurfaceoftheironoxidenanopar�ticles.(27,28) pf127/oleic acid@super�paramagneticironoxidenanoparticlesarehereafter abbreviated as poa@Spion.�hepf127coatingisanaBa�typetriblockcopolymer consisting of poly(propyleneoxide) (ppo) and poly(ethylene oxide)(pEo) that serves to make the nanopar�ticles water soluble, which is an invalu�able characteristic for medical applica�tions.�hediameteroftheironoxidecoreis12.4nmandhydrodynamicdiameterofPOA@SPION is 92 nm with polydispersi�tyindex0.073.Weundertookapreliminarystudytoas�sessthefeasibilityofusingnon�targetedpoa@Spionsforimagingofthecochleaand compared the results to with gado�liniumchelateimagingoftheinnerearinrats.

Materials and MethodsTwenty-five healthy male Sprague-Daw�ley rats weighing from 218 – 470 g were providedbytheExperimentalmriLabo�ratory,departmentofneurology,Helsin�kiuniversitycentralHospital,fin�00029HUS, Helsinki, Finland. Rats were studied usingi�,iV,andintracochleardeliverybyasiliconcatheterconnectedtoareservoir(tubing OD=0.64 mm, MedEl, Innsbruck, austria).asummaryofthestudydesignis shown in Fig 1.inmostcasesofi�andintracochlearap�plications the treatment was delivered totheleftearandtherightearservedasa control. Animals were maintained un�der general anaesthesia with Domitor (0.5mg/kgmedetomidinehydrochloride,orionpharma,finland)andKetalar(75mg/kg ketamine, pfiZEr aB, finland)givenintraperitoneally.animalsthenre�

ceived either POA@SPION (4.5 μmol Fe/ml) by i�, iV, or intracochlear routes orGadolinium do�a (Gd�do�a)(279.3mg/ml Gadoterate meglumine (dotarem®),Guerbet, paris) by i� or iV routes. Se�rial MRI scanning was subsequently per�formed. Animals were maintained on a warming blanket during the administra�tion procedures and were covered with a warming blanket with circulating water during MRI scanning. Eyes were protect�ed with saline-based ointment. All animal experiments were approved by the Ethi�calcommittee of theuniversity of�am�pere (permission: S�H527a ESLH�2006�07528/Ym23). animal care and experi�mental procedures were conducted in ac�cordance with the European legislation.

Intratympanic administration techniqueUnder the operating microscope and with a transcanal approach, a posterior myr�ingotomy was made and gelfoam pled�getssoakedinGd�do�aorpoa@Spionwere placed into the round window niche and to fill the posterior middle ear cavity. inmostofthepoa@Spioncasesasimi�lar procedure was done on the contralat�eral side and gelfoam pledgets soakedin saline were applied as a control. One animal had diluted POA@SPION 1:5 with water. One animal additionally received IV Gd 1.5mmol/kg two hours after the IT application to investigate whether there would be any decrease in T1 signal on the treated side. one animal received a 1:9dilutionofGd�do�a:normalsalineappli�cation.�heremainderGd�do�aapplica�tions were undiluted.mri scanning commenced immediatelyafter the IT administration except in two SPION treated animals that were scanned 24hourslater.

Figure 1. Summary of the study’s design



Intravenous administration techniqueThe lateral tail vein was dilated by soaking the tail in warm water followed by topi�cal application of 70% alcohol. �he veinwas cannulated with a 30 gauge needle connected to catheter tubing and tapedsecurely intoposition. Salinecontaining5% heparin was initially instilled to main�tainpatency.Baselinepre�treatmentmriscans were obtained as a control. POA@SPION at a dosage of 10 μmol/kg or Gd-do�a 1.5 mmol/kg in one rat and 0.75mmol/kg in two was slowly injected intra�venously. mri scanning commenced im�mediatelyaftertheiVadministration

Intracochlear administration techniqueA post-auricular approach was used to expose and open the left bulla. underthe operating microscope a 0.5 mm dia�mond burr was used to create an open�ing into the cochlea (cochleostomy) in�ferior to the round window to access the scalatympaniinthebasalturn.aproto�type silastic catheter designed for intra�cochlearinsertion(medElinc.,innsbruck,austria)(fig2)orcustommadepolyure�thane catheter (O.D.=0.25 mm, I.D.=0.12 mm, AgnTho’s AB, Sweden) was slowly advanced atraumatically approximately1.0 mm into the scala tympani.�he co�chleostomy, with the catheter in place, was sealed with Histoacryl (enbucrilate) glue (aesculap, �uttlingen, Germany) inthe first animal and in the remaining, it was sealed with a muscle plug that was subsequently covered with the glue. After the glue was dry 10 μl-20μl POA@SPION was slowly instilled into the scala tympani in the first three animals with the MedEL catheter and 5μl was used for the remain�ing two with the custom made catheter. In one animal, POA@SPION was diluted 1:50 with saline. The wound was sutured closed.mriscanningcommencedimme�diatelyaftertheintracochlearadministra�tion.

Phantom preparationPOA@SPION stock solution was seri�

ally diluted by 100, 200, 400, and 800times with water to make concentrations 45, 22.5, 11.25, and 5.6 μmol/L respec�tively. Plastic phantom tubes (AgnTho’s AB, Sweden) were filled by either 400 µl poa@Spion solutions or the sameamount of water as controls. The tubes were grouped in concentric circles with the SPIONs centrally and the water ar�rangedperipherallyandallhousedina50ml syringe. Relaxivities were measured in themriscanner.

MRI measurementsMRI studies were performed with a 4.7 T scanner (pharmaScan, Bruker BioSpin,Germany) with bore diameter of 155 mm using a 90 mm shielded gradient that iscapable of producing maximum gradi�ent amplitude of 300 mT/m with 80-µs rise time. A linear birdcage RF coil with an inner diameter of 38 mm was used. The body temperatures of animals were maintained by circulating warm water and their respirations were recorded with physio�ool�1.0.b.2program(BrukerBio�Spin, Germany). Rats were placed in the magnet with the ears positioned at the isocenter.2D MRI measurements were performed immediatelyaftertreatmentinallgroupswith serial scans taken approximately hourly up to 4 hours. Two animals were reimaged after 3 days and two animals re�imagedafter7days.for 2d imaging, the inner ear geometrywas established by taking three 2D im�ages at oblique orientations using thestandard Bruker technique of fast spinecho sequence; Rapid acquisition with re�laxation enhancement (rarE) sequence[repetition time (TR) = 2500 ms, effective echotime(�Eeff) = 40 ms, number of av�erages (NA) = 3, matrix size = 256 x 256, field-of-view (FOV) = 50 x 50 mm, 3 slices with slice thickness = 0.5 mm]. Then T2-weighted image with RARE sequence (TR = 2500 ms, TEeff = 40 ms, NA = 10, matrix size = 256 x 256, FOV = 2.5 x 2.5 mm, 3 slices with slice thickness = 0.5 mm)T2*-weighted images were acquired with a FLASH (Fast Low Angle Shot) sequence (TR=350 ms, TE= 10 ms, flip angle = 40, matrix size = 256x128, FOV = 25x25 mm, NA = 17, 3 slices, slice thickness= 0.5 mm). 3D T1-weighted images were acquired with turboRARE sequence, FOV= 50 x 50 mm, matrix size = 64x64x64, TR 500 ms, �Eeff = 10 ms, flip angle 180 degrees.for the one animal that had additionalIV Gd, both 2D and 3D sequences were

used. 2D T1-weighted images were ac�quired with RARE sequence (TR/TEeff = 500/10 ms, matrix size=256x192, FOV = 25x25 mm, NA = 10, rare factor = 4). 3D T1-weighted images were acquired with turboRARE sequence, FOV= 50 x 50 mm, matrix size = 64x64x64, TR 500 ms, TEeff = 10 ms, flip angle 180 degrees.intratympanic and intracochlear deliv�eries were done prior to imaging. In the intravenous administration groups, ini�tial T2-weighted images were taken after satisfactory geometry was obtained. The Gd-DOTA or POA@SPION was subse�quently followed by similar serial MRI im�aging with T2-weighted images. At the completion of each animal’s stud�ies, they were sacrificed with an overdose of barbiturate intraperitoneally adminis�tered. inmostofthepoa@Spioncas�es, intracardiac and intracochlear perfu�sion were performed and the cochleae harvestedforfuturehistologicalstudy.

ResultsSPION phantomT2-weighted images of the SPION phan�toms showed high r2 values such that sig�nificant darkening of signal was seen at a 100-fold dilution and the effect was nearly equal to water at an 800-fold dilution. (Fig 3)

Figure 2. catheterforintracochleardelivery..

Figure ��.�2 map of poa@Spion phantom�2 map of poa@Spion phantomwith vials containing dilutions of 1:100, 200, 400, 800 in the center compared to theperipheral ring containing water.

Intratympanic administration techniqueFour rats were treated with IT Gd-DOTA. at1hourafteradministration,rarE�1�weighted images showed no changes on thecontrolearbutthetreatedeardem�onstrated mild enhancement within the scala tympani and scala vestibuli filling that was most visibly prominant in the first 1.5 – 2.5 turnsof the cochlea and with the basal turn showing the strongest up�take. At 3 hours, the entire cochleae were maximally enhanced and the controlsides remained unchanged. decreased



Figure �.�. RARE T1-weighted image of rat cochleae one hour after intratympanicadministration of Gd�do�a to the leftmiddleear.Gadoliniumenhancementismostprominentinthebasalturn.

Figure �.�. RARE T1-weighted image of the sameratcochleaeasinfig4threehoursafterintratympanicadministrationofGd�do�atotheleftmiddleear.Gadoliniumenhancementis prominent in the entire cochlea with clear distinction between the scalae.

Figure �.�.3dmrireconstructionofratcochleaimaged with 3D T2-weighted sequence.

signal was noted in one animal scanned at five hours. The scala media remained unenhancedthroughoutthestudiesmak�ing for a strong distinction between the bright perilymph and dark endolymphcompartments. (fig4,5)3dvolumereconstructionsfacilitatedthevisualisation of the distinctions between thescalae.(fig6,7)

Ten rats underwent IT administration of POA@SPION. rarE �2 images consistently demon�strated absence of signal in the middle

three showed decreased signal on the control side, and one showed no chang�es on either side. one of the three thatshowed decreased signal on the POA@SPION side was rescanned after three days and showed decreased signal on the controlsideatthattime,thenonrepeatscanning after 6 days, showed decreased signalagainonthepoa@Spionside.

Intravenous administration techniqueThree animals underwent IV administra�tionofGd�do�a(1.5mmol/kginoneratand 0.75 mmol/kg in two). at1hourafterinjectionofGd�do�a,theentirecochleawas mildly enhancing and the signal be�camestrongerin3hours.(fig8,9,10)Three animals underwent IV delivery of poa@Spion. after one hour, all threedemonstrated mild diffuse decreases in signalinalltissues,includingthecochle�ae, cerebrospinal fluid (CSF), brain, and muscles. �he reduction in signal frombaselinescanningpersistedforthedura�tionoftheexamination.

Intracochlear administration techniquefive animals had intracochlear deliveryof POA@SPION to the left ear with the opposite ear left untreated as a control.In RARE T2-weighted MR images, the first animal showed darkening of signal only in the basal turn of the cochlea ad�jacent to the cochleostomy and the de�crease persisted throughout the exami�nation. (fig 11) At autopsy, it was found that the glue had developed a leak atthe cochleostomy. �he second animalshowed immediate decrease in cochlear signal on the poa@Spion treated sidewith darkening seen diffusely, but espe�ciallyprominantly inthemiddleandapi�cal turns. at one hour, most of the leftcochlea was darkened and not visualised exceptforsomefaintstreaksnotedpos�sibly inthelocationofthescalamedia. .(figs 12)at three hours, the cochlea be�gan to reappear with some mild return of signalintheperilymph.�hethirdanimalreceived POA@SPION diluted 1:50 with saline. There was no observable change incochlearsignalduringtheexamination.�he4thand5thanimalsdemonstratedim�mediate prominent darkening of signalthroughoutthescalatympaniandvestib�uli with bright signal persisting in the sca�la media. The signal darkening was per�sistentinoneoftheseanimalsat4hours.The signal darkening was visibly less at 4 hours intheotheranimalandat6hourshadevenmorevisiblereturnofsignal.

Figure �.�. 3d mri reconstruction of samerat cochlea as in Fig 6 imaged with 3D T2-weighted sequence and rotated 90 degrees.


ear space containing gelfoam soaked inpoa@Spion and bright signal in themiddle ear region containing gelfoamsoaked saline on the control side. allcochleae on the control side remainedbrightonallrarE�2images.intheco�chleae treated with POA@SPION, there was no decrease in signal seen in any an�imal immediatelyafterapplication. oneanimal (10%) demonstrated mild de�creased signal within the perilymph in themiddleandapicalturnsoftheactivelytreatedcochleaafteronehourandtheef�fect was slightly more pronounced at two hours. One additional animal showed no change in cochlear signal on the poa@Spiontreatedsideatonehour,butdem�onstrated decreased signal in the peri�lymph space at three hours and slightlydecreasedsignaladditionallyintheendo�lymph space at five hours. Eight animals (80%) showed no change in signal from the cochlea at any time.Two of these eight were only imaged at 24 hours after the treatment. Two oth�ers of the eight were imaged initially and re�imagedafterthreedays.oneofthesetwo was further re-imaged after 6 days. oneoftheseeighthadnosignalchangesby two hours after treatment and had IV Gd instilled. There was bright symmetric enhancementofbothcochleaeafteronehour that persisted for the subsequentthree hours imaged. �he middle ear re�maineddarkonthepoa@Spiontreatedside throughout the examination. noneof these eight animals showed any signifi�cant signalchanges in [email protected]*-weighted sequences were performed in seven animals. images demonstratedincreasednoiseartefactandreducedres�olutionof thecochleaecomparedto�2�weighted sequences. Three of the seven showed decreased signalonthepoa@Spiontreatedside,


CommentsThis study is the first to employ intratym�panic or intracatheter administration ofcontrastagentsintoratsformrimaging.intravenous and intratympanic deliveryof Gd�do�a produced bright enhance�ment of the cochleae in T1-weighted im�ages and the effect was even detectable in T2-weighted images. The mechanism forGdchelateuptakeintotheperilymphspaceremainsuncertain.intratympanicGdmostlikelygainedaccesstotheperi�lymphscalatympanibypassagethroughthe round window, either via active trans�port or passive diffusion. It then very rap�idly passed into the scala vestibuli suchthat there was no differential imaging between the scala tympani and vestibuli seeninanyanimal.�hepassageintothecochlea by intratympanic delivery was slower than by IV route and progressed from the basal turn toward the apex

whereas the intravenous delivery pro�duced diffuse enhancement of the peri�lymph space beginning within one hour. iV delivery demonstrated that the con�trast agent effectively passed through the blood�perilymph barrier. intraco�chlear detail was excellent with both ad�ministration routes and T1-weighted im�ages revealed a sharp contrast between thebrightperilymphspacesandthedarkendolymph compartment. These find�ingscloselysupportedthoseofpreviouswork with Gd chelate in guinea pigs. (21) �he results provided an excellent basisfor comparison with the SPIONs, which cause darkening of signal on T2-weight�edimages.poa@Spions demonstrated a strongreduction of signal on T2-weighted se�quencesinthisstudy.�hischaracteristicmakes them useful as an alternative toGd chelates, which enhance signal on T1-weighted images. As a negative contrast agent,Spions could be used to darkenportions of the cochlea that would be surrounded by otherwise bright endo�lymph or perilymph on T2 weighted im�ages. �argeted Spions could be visu�alised as they coursed to their specific locations or signals might be activated

by interactions with the targeted cells. �hepreservationofhighr2valuesinhighmagnetic fields is a particularly desirable featureofSpionsforpossiblefutureusein this type of molecular scale imaging.magnetic particles can be directed to�ward targets by manipulation of mag�netic fields. Magnetic nanoparticles can be manipulated to “switch on and off” their functions in magnetic fields and of�fer interesting possibilities for control�lingthetherapyatmolecularlevel.T2* -weighted sequences should be able toincreasethesensitivityfortracingSpi�ONs in tissues but we found that they yielded significant artefacts when trying to image the cochlea. using methodsto reducethe artefacts could make�2�amoreusefulimagingtechniqueforfu�tureinvestigations.�he present study found that non�tar�geted POA@SPIONs were not effective in demonstrating the cochlea when ad�ministeredtothemiddleear.mostlike�ly, these nanoparticles were unable to pass through the round window mem�brane in quantities sufficient for MRI imaging. �hereasonsforfailuretoper�meate the membrane are unclear andcouldberelatedtoissuesofsize,charge,coating properties, and other variables.It is possible that with reduction of par�ticle size and other modifications, SPI�onsmaymorereadilypassthroughtheround window membrane. The addition oftargetbindingligandstothenanopar�ticle may also augment the quantity ofcontrastagentsintheregionofinterest.furtherstudyofthesealterationsisindi�cated.�heamountofnanoparticlesthatmayhavegainedentranceintotheinnerear will be assessed in the future histo�logicalexaminationofthecochleaefromthisstudy.Nanoparticles that were delivered by IV routecreatedmildareductionofsignaldiffusely in all of the vascular tissues in the head and CSF, without any specif�ic predilection for the cochlea. in com�parison,GdchelateinhighiVdosesap�pearedtobesequesteredintotheinnerear over the first three hours and yielded positivelycontrastedimagesoftheperi�lymph compartment on T1-weighted imaging that were comparable with the excellent intracochlear resolution seenwith IT delivery. intracochlear delivery of poa@Spionsdemonstratedastrongnegativecontrasteffect within the cochlea on T2-weighted images. The first animal showed darken�

Figure �.�. RARE T1 weighted image of rat cochleapriortoiVGd�do�aadministration

Figure �.�. RARE T2 weighted image of same rat cochlea as in fig 8 prior to iVGd�do�aadministration

Figure 10.10. RARE T1 weighted image of same ratcochleaasinfig81.5hoursafteriVGd�do�a administration. �he scalae tympaniand vestibule are maximally contrast�enhancing along the full length of bothcochleae.�hescalamediaremainsdark.

Figure 11.11. rarE�2image of rat cochleae 1image of rat cochleae 1ofratcochleae1hourafterleftintracochlearcatheterinjectionof POA@SPION that was noted at autopsy to have likely leaked into the middle ear.darkeningofsignal isseeninthebasalturnadjacent to the cochleostomy site (Arrow)

Figure 12.12. rarE�2 image of rat cochleaeimage of rat cochleae of rat cochleae1 hour after left intracochlear catheterinjection of poa@Spion. most of cochleaisdarkenedcomparedtothebright�2signalfromtherightcontrolside.



ing of signal only around the cochleos�tomysiteandmiddleearduetoleakageof the agent out of the cochlea. in thesecondanimal,thelargerdoseofSpionquenched the perilymph signal stronglythroughout the whole cochlea, but there was some faintly bright signal demon�strating the endolymph compartment.�he perilymph�endolymph barrier is ex�tremely tight and would not be expected to allow ready passage of nanoparticles early after administration into the peri�lymph. Most likely, the quenching effect of signal in the perilymph was sufficiently greatastospilloverandsuppresstheen�dolymph compartment as well, without actualpassageofparticlesintotheendo�lymph. In the last two animals, in which a smaller dose was employed, there was an immediate strong suppression of sig�nal throughout the perilymph compart�ment that strongly contrasted with the bright endolymph signal. �hese resultsare being further investigated and will be reportedindetailinafuturereport.ad�ditional study with various doses, delivery methods, and quantified intensity mea�surements will be necessary to elucidate theintracochlearbehaviouroftheseSpi�ons.duetotherisksofhighdosesystemicGdadministration, i� delivery for cochlearimagingappearsmoreattractivethanbyiV route in humans. �here are possiblerisks with the IT route as well. IT Gd che�late had some possible ototoxic effects in a study that showed electrophysiological and histological changes when adminis�tered at full strength, but when diluted 1:8 no evidence of injury was shown. (29) it remains a challenge to develop ameans to facilitate the passage of Spi�ONs through the round window mem�brane. currently, Gd chelate continuesto be the most useful contrast agent forinner ear imaging. As a powerful positive contrastagentthatisrapidlysequesteredinto the perilymph, it is now being used in clinicaltrials.i�administrationofGdche�late with T1 weighted sequences creates bright enhancement of the perilymphcompartment in contrast to the persis�tently dark endolymph within the scala media. However, the ability to visualise molecular interactions within the cochlea maybebeyondthelimitationsofGdche�late and further investigation ofSpionsand Gd nanoparticles will be important. It ispossiblethatSpionstargetedforspe�cialcochlearcellpopulationsmaybeuse�ful if they can be shown to sequester spe�

cifically within the inner ear, either by IV ori�administration.

ConclusionsGadolinium chelate, delivered i� and iV,remains the currently preferred positivecontrast agent for T1 weighted imaging of the cochlea. T2 weighted imaging with poa@Spions as a negative contrastagenthasbeendemonstratedtobefeasi�bleinthisstudyusingintracochleardeliv�ery. �he limitsof intracochlear imaging,resolution,distribution,andpharmacoki�netics all remain to be studied with these nanoparticles.Spions need to be modi�fied in order to gain access to the cochlea through the round window membrane or byvascularapproach.onceenteringthecochlea, Spions may be promising fordifferential imaging of intracochlear com�partmentsandforfuturemolecularinter�action imaging. further investigation ofthese contrast agent nanoparticles is war�ranted.

References1.ZouJ,poed,BjelkeB,pyykköi.Visu�alization of inner ear disorders with MRI in vivo: from animal models to humanapplication. acta oto�Laryngol 2009;129:22�31.2. Sosnovik d, Weissleder r. Emergingconcepts in molecular mrcurrentopin�ioninBiotechnology2007,18:4–103.poppera,fayr.�hecochlea,Spring�erHandbookofauditoryresearched.bypeterdallos,(1996)4. Leonova EV, raphaelY. organizationof cell junctions and cytoskeleton in thereticular lamina in normal and ototoxi�callydamagedorganofcorti.Hearingre�search1997;113:14�28.5. Hibino H, Kurachi Y. molecular andphysiological bases of the K+ circulationin the mammalian inner ear. physiology(Bethesda,md)2006;21:336�45.6.mcdonaldm,WatkinK.investigationsinto the physicochemical properties ofdextran Small particulate Gadoliniumoxide nanoparticles. acad radiol 2006;13:421–4277. Jacobs rE, cherry Sr. complemen�tary emerging techniques: high resolu�tion pE� and mri. curr opin neurobiol2001;11:621.8. frias Jc, ma Y, Williams KJ, fayadZa, fisher Ea. properties of a versatilenanoparticle platform contrast agent toimage and characterize atheroscleroticplaquesbymagneticresonanceimaging.nanoLett2006;6:2220–2224.

9. Morawski AM, Winter PM, Crowder Kc,caruthersSd,fuhrhoprW,ScottmJ,robertson Jd, abendschein dr, LanzaGm,WicklineSa.�argetednanoparticlesforquantitativeimagingofsparsemolec�ular epitopes with MRI. Magn Reson Med 2004;51:480–486.10.Sosnovikd,nahrendorfm,Weissled�err.magneticnanoparticlesformrim�aging:agents,techniquesandcardiovas�cularapplicationsBasicrescardiol.2008march;103(2):122–13011. Sosnovik d,Weissleder r: magneticresonance and fluorescence based mo�lecular imaging technologies. prog drugres2005,62:83�115.12. Gløgård c, Stensrud G, Hovland r,fossheim S, Klaveness J. Liposomes ascarriers of amphiphilic gadolinium che�lates: the effect of membrane composi�tion on incorporation efficacy and in vitro relaxivity. international Journal of phar�maceutics233(2002)131–14013.LanzaGm,abendscheindr,YuX,Win�terpm,KarukstisKK,ScottmJ,fuhrhoprW,ScherrerdE,WicklineSa:molecularimaging and targeted drug delivery with a novel, ligand�directed paramagneticnanoparticle technology. acad radiol2002,9(Suppl2):S330�S331.14.Winterpm,caruthersSd,YuX,SongSK, chen J, miller B, Bulte JW, robert�son JD, Gaffney PJ, Wickline SA et al.: Im�provedmolecularimagingcontrastagentfordetectionofhumanthrombus.magnresonmed2003,50:411�416.15.friasJc,WilliamsKJ,fisherEa,fay�adZa:recombinantHdL�likenanoparti�cles: a specific contrast agent for MRI of atherosclerotic plaques. Jamchem Soc2004,126:16316�16317.16. Lipinski mJ, amirbekian V, frias Jc,aguinaldo JG, mani V, Briley�Saebo Kc,fuster V, fallon J�, fisher Ea, fayad a:MRI to detect atherosclerosis with gado�linium�containing immunomicelles tar�getingthemacrophagescavengerrecep�tor.magnresonmed2006,56:601�610.17.Shen�,Weisslederr,papisovm,Bog�danovaJr,Brady�J:monocrystallineironoxide nanocompounds (mion): physico�chemical properties. magn reson med1993,29:599�604.18. Wunderbaldinger p, Josephson L,Weissleder r: crosslinked iron oxides(CLIO): a new platform for the develop�ment of targeted mr contrast agents.acadradiol2002,9(Suppl2):S304�S306.19.counterSa,BjelkeB,Klason�,chenZ, Borg E. magnetic resonance imagingof the cochlea, spiral ganglia and eighth



nerve of the guinea pig. neuroreport1999;10:473�9.20.counterSa,ZouJ,BjelkeBKlason�.3d mri of the in vivo vestibulo�cochlealabyrinth during Gd�d�pa�Bma uptake.neuroreport2003;14:1707�12.21. Zou J, pyykko i, counter Saet al. invivo observation of dynamic perilymphformation using 4.7 T MRI with gadolin�ium as a tracer. acta oto�laryngologica2003;123:910�5.22. Thomsen H. Nephrogenic systemic fi�brosis: a serious late adverse reaction togadodiamide Eurradiol.2006december;16(12):2619–2621.23.Zou J, pyykko i, Bjelke B dastidar p,Toppila E. Communication between the perilymphatic scalae and spiral ligamentvisualized by in vivo mri. audiology &neuro�otology2005;10:145�52.24. Naganawa S, Satake H, Iwano S, Fu�katsu H, Sone m, nakashima�. imagingendolymphatic hydrops at 3 tesla using3D-FLAIR with intratympanic Gd-DTPA administration. magn reson med Sci2008;7:85�91.25. Moffat BA, Reddy GR, McConville P, Hall dE, chenevert �L, Kopelman rr,philbert m,Weissleder r, rehemtulla a,ross Bd. a novel polyacrylamide mag�neticnanoparticlecontrastagentformo�lecular imaging using mri. mol imaging2003;2:324–332.26. Wunderbaldinger p, Josephson L,Weissleder r. crosslinked iron oxides(CLIO): a new platform for the develop�ment of targeted mr contrast agents.acadradiol2002;9(Suppl2):S304–306.27.QinJ,Laurent,JoYS,rocha,mikhay�lova M, Bhujwalla ZM, Muller RN, and muhammed m. a High�performancemagneticresonanceimaging�2contrastagent.adv.mater.2007,19,1874–187828.QinJ,asempahi,LaurentS,fornaraa,mullerr,muhammedm.injectableSu�perparamagneticferrogelsforcontrolledrelease of Hydrophobic drugsadv. ma�ter.2009,21,1354–135729.Kakigia,nishimuram,�akeda�etal.Effects of gadolinium injected into the middle ear on the stria vascularis. actaoto�laryngologica2008;128:841�5.



The Role of TOF-SIMS in NanomedicineBeata.Keller,Karlmayerhofer1(doi 10.3884/0002.2.8)doi10.3884/0002.2.8)

Abstract

The toolbox of analytical techniques available today for materials scientists offers a wide selection of instruments for nearly every problem and application. Here we present a brief introduction into tissue imaging by time-of-fight secondary ion mass spectrometry (TOF-SIMS) for the non-expert. Possibilities and limitationsof�of�SimSfor identifyingchemicalcomponents inbiologicalmatterarediscussed.�hepotential of the technique is shown in a case study of 2D imaging of co-polymer based nanosized drug

deliverycontainersembeddedinmouseliver.Keller,Bnanomedicine2009,2.2:37�43

Keywords: �of�SimS,cluster primary ion Beam Sources, imaging in cell and �issue researchclusterprimaryionBeamSources,imagingincelland�issueresearch

Swiss Federal Laboratories for

materials�estingandresearch

ueberlandstrasse129,

CH-8600 Dübendorf, Switzerland

analyticalnanomethods

IntroductionTOF-SIMSTime-of-flight secondary ion mass spec�trometry (�of�SimS) is a very versatilesurface sensitive analytical techniquethat yields elemental as well as molecular informationfromsolidsubstratesrangingfrombiologicalmaterialstosemiconduc�torapplications.afterthebombardmentof a surface with energetic charged pro�jectiles, the emitted secondary speciesoriginatingfromthetopmostlayersofthematerial are collected and analyzed with highmassresolution.�heparalleldetec�tion scheme not only provides spectralinformation with high accuracy, but also allows the operation of the instrument in a microprobe mode to obtain chemi�callyresolvedimagesofthesurface.�heprimary ionbeamisscannedoverapre�selectedsurfacearea,andcompletemassspectra are recorded from a pixel map.3d distributions of chemical species areproduced by alternating cycles of mate�rialablationandimagerecording. inthisdualbeammode,asputtergunisremov�ing a well defined layer of the material surface under accurately controlled low-energy conditions. Afterwards a second ionbeamisscannedoverthecraterareato specify the chemical composition ofthe freshly exposed surface. it is impor�tant to be aware that TOF-SIMS is gen�erally a destructive analytical technique,andthequalityofdataisverysensitivetothechemicalnatureofsputterions.Highdensitydatastorageofmassandlocationinformation offers the possibility of retro�spective analysis of samples containingpreviously unknown chemical species. In a reverse mode, the software can recon�structthedistributionofachemicalcom�

pound in a user-defined area.�of�SimS applications of biomaterialsurfaces have become a traditional field of the technique. due to the importantroleofmedicaldevicesformodernmedi�caltreatment,andtheincreasingnumberof applications in all fields of trauma and reconstructionchirurgythedevelopmentandqualitycontrolofsurfacesinmedicalsystems has become vital for the prog�ress of materials design. It is not within the scope of this paper to discuss �of�SimS analysis of biomaterials. for a de�taileddiscussion,thereaderisreferredtorecent reviews and articles of the field [1 – 4], and the literature cited therein.

Limitations of TOF-SIMS�he pivotal idea of surface analysis is toanswer basic questions about what is on the surface, how much, where. While �of�SimS provides detailed informationabout the nature and location of chemi�calcompounds,itislesssuitabletoobtainquantitativedata.�hereasonforthislimi�tationisastrongdependenceofsecond�ary particle emission from the chemicalenvironment, i.e. the matrix. �he con�ductivity of the material influences the ability to neutralize the surface chargeinducedbytheimpactofchargedprimaryprojectiles.althoughitispossibletocom�pensate this effect to a certain degree, surface charging tends to degrade massandspectralresolutionduetoinstabilitiesoftheacceptanceangleofthemassana�lyzer. Even under stable emission condi�tions,thetruechemicalnatureofsurfacescan be masked by contaminations fromsample preparation or improper han�dling.�hepreparationofcomplexmateri�als as encountered in biological samples

therefore requires careful preparationprotocols to avoid the build�up of arti�facts. if molecules are present in serumlevel concentrations (i.e. metabolites),secondaryparticlescanonlybedetectedusingstate�of�the�artclusterprimaryionbeam sources. Molecular fragments with a mass exceeding m/z = 1500 are often formed in extremely low abundance and therefore not detected. Sophisticatedprincipalcomponentanalysis(pca)isap�pliedtothedatasetstoidentifyunequiv�ocally such signals [5, 6].

TOF-SIMS with Cluster Primary Ion Beam Sources�he desorption of secondary ions afterthe impact of energetic ions with typically several thousand electron volts was first used to by Benninghoven et al. [7] in 1976 to specify surface located organic mol�ecules. However, it became soon evident thatabovetheso�calledstaticSimSlimit(iondoseabout1012ions/cm2) where more thanabout1%ofsurface locatedmole�culesareconsumedduringmeasurement,an increasing amount of unspecific low-mass carbon builds�up through surfacedamage,causinginitiallycompound�spe�cific secondary signals to disappear. This later effect restricts the concentration of surface species that can be detected instaticSimSmodeusingmonoatomicpri�maryions.Sincetheusefullateralresolu�tion and the efficiency are connected, a

EuropEan JournaL of nanomEdicinE 2009 Vol. 2 issue 2 38

lower ion yield also affects the area from which an analytically useful information can be obtained. Therefore, it became ap-parent that the overall secondary ion yield must be increased in order to be able to detect low abundance chemical species on a surface. Experiments undertaken in the late 1980’s using SF6

+ and csxiy+ as

primary ion sources [8, 9] showed im-proved signal intensities in polymers and organic thin films [10]. Other molecular ion species proved to be less effective and it was more difficult to maintain steady and uniform beam conditions. However, SF6

+ sources, still difficult to operate and providing only moderate beam current, could be commercialized by several man-ufacturers. The later development of Aun

+, Bin

+ and c60+ guns overcame these limita-

tions [11, 12] and are state-of-the-art today. Recent calculations indicate that the damage accumulation rate is in many cases lowest for fullerene primary ions, probably due to the low kinetic energy of individual carbon atoms as compared to the atoms in a small polyatomic cluster [13]. Cluster primary ion sources produce an up to 1000 times higher amount of secondary ions and due to the less vig-orously fragmentation also allow for the detection of very specific high-mass mo-lecular fragments, typically encountered in biological systems. For a recent review on cluster primary ion sources, the reader is referred to [14].The most striking advantage of TOF-SIMS using cluster ion sources, however, is the fact that the high intensity of signal aris-ing from individual pixels makes it pos-sible to detect small amounts of a species from a small area, i.e. obtaining a smaller useful lateral resolution as required in many biological applications. As the pixel dimension (typically 256 x 256 pixel per area) approaches sub-micrometer size, the number of analytically useful surface molecules located within this area is in the micromole range.

Applications of TOF-SIMS Imaging in Cell and Tissue ResearchSample preparation and handling are sig-nificant for the quality of TOF-SIMS data obtained from biological matter. In a re-cent paper, Kurczy et al. [15] concluded that both, the nature of the projectile and sample preparation protocols are equally important to acquiring successful infor-mation from biological samples.Preparation of Biological Matter for TOF-SIMS measurements

Sample preparation for imaging TOF-SIMS experiments include several impor-tant issues that each influences the qual-ity of the acquired data. Most notable these are:

A. specimen collectionB. specimen handling (freezing, embed-ding, chemical fixation, etc).C. sectioning and mounting manipula-tionsD. measurement temperature

The collecting of biological matter, which is either harvested from an organ (liver, kidney, spleen etc.), originating from a biopsy, or taken from a cell culture is the very first step of a successful bio imaging experiment. Several instruments like bi-opsy needles or collecting plates are used for this purpose. Before transferring the tissue sample to a suitable substrate hold-er, it is imperative to freeze the specimen as fast as possible either in liquid pentane, isopentane or another appropriate water free cooling medium. As is well documented, freezing gener-ally damages cell membranes, and there-fore reduces the histological information from biological samples. This effect is most prominent after slow freezing and in large volume samples. The former pro-cess promotes ice crystal formation, and the latter reduces the cooling rate due to the Leidenfrost effect. Large thermal gradients inside the specimen would be the consequence. Since the rate of freez-ing is more important than the actual end-temperature, it is important to avoid such behavior. Shock-freezing in hydro-carbon solvents kept at liquid nitrogen temperature has been established as working protocol to prepare biological samples at freezing temperature. How-ever, even if a specimen is quickly enough frozen down to a vitreous (amorphous) state, it may not stay that way for a long time because vitreous ice forms an un-stable state. Above -212 °C, amorphous ice begins to gradually undergo a phase restructuring process to cubic ice, associ-ated with volume expansion. As a conse-quence the water stretches and starts to penetrate cell membranes. While this is a slow process, it still sets rather tight time limits to the measurement, and particu-larly at the time of sample introduction. Modern TOF-SIMS equipment is outfitted with liquid nitrogen cold traps to assist cold temperature measurements, but the experiment still requires experience and

skills. The correct way to prepare samples of biological origin for TOF-SIMS imag-ing investigations has stimulated a long debate among specialists. For a recent re-view of tissue preparation for TOF-SIMS measurements, the reader is referred to [16] and the literature cited therein.For some time, chemical fixation proto-cols, using glutaraldehyde or a water/acetic acid mixture have been used to prepare samples. Glutaraldehyde reacts through its two aldehyde groups with one primary amine group of a protein. Thus, glutaraldehyde acts as protein linker, and preserves cell structure, although in a completely water-free state. On the other hand, treatment with a water/acetic acid mixture dissolves the cell membrane. The remaining tissue sample mainly consist of the nucleus (or parts of it), and the cyto-skeleton of cells. Certainly, one does not expect to find the original native state of tissue after such harsh chemical treat-ment. Nevertheless, chemical fixation protocols have been frequently used to prepare biological samples [17, 18]. Since a freshly cut sample is unlikely to be completely flat, one must also con-sider the roughness of the specimen on the molecular level. In order to reflect the real distribution of molecular species within a sample, it is important to con-sider this fact. In particular if a 3D image is reconstructed from a depth profiling experiment, e.g. using C60

+ sputter ions. It is immediately evident that for example a spherical object can be distorted due to impact angle dependent shadowing ef-fects during etching. A better way to han-dle this problem would be the construc-tion of a slicing instrument that produces a series of freshly prepared surfaces from which 2D molecular images are measured in the static SIMS mode and subsequently stored. The 3D map is then reconstructed from slices rather than from a damage plagued sputter profile.In recent years, SIMS bioimaging has been applied to a range of tissue specimen us-ing static TOF-SIMS. Among the most promising studies were the mapping of fatty acids, lipids, and vitamin E, among others [12, 19] in mouse and rat brain sec-tions. Distribution maps showed the loca-tion of these rather large molecules in the intracellular medium by pooling relevant fragments. Molecular details of low mo-lecular weight compounds located at the single cell level are obviously much more difficult to extract. First of all, the amount of available molecules is generally very

Analytical Nanomethods

39 EuropEan JournaL of nanomEdicinE 2009 Vol. 2 issue 2

small, and secondly, the focus require-ments of the primary ion beam are very demanding. However, several studies have been performed, which reveal some details about the molecular organization of single cell samples [20].

2D Mapping of artificial Polymers in Tis-sue – a TOF-SIMS case StudyThe advantage of targeting specific cells and components of biological interest in tissue samples has been outlined by Broz et al. [21]. In particular the possi-bility to transport drugs and molecules with diagnostic functionality intact, i.e. by avoiding the harsh digestion process, to certain parts of a vertebrate body has been proposed as therapeutic concept with great potential. Conventional ways to deliver drugs for certain therapies are not very specific, and generally drug ac-tion can not be restricted to well defined parts of a body. Therefore the amount of drug or diagnostic material often exceeds the amount which is minimally necessary to perform the desired action. The con-sequences for the patient are oftentimes very uncomfortable and sometimes ac-companied by sickness and the danger of allergic reactions. Although the concept of drug delivering by means of small con-tainers which are specifically targeting the membrane proteins of a certain cell type is well documented, the interplay between recognition and controlled drug release by subsequent trigger action is not easy to combine. The experimental proof of this principle has yet to be shown.A first important step towards this goal is the 2D mapping of drug delivery contain-ers of nanometer dimensions. Here we present the application of time-of-flight secondary ion mass spectrometry (TOF-SIMS) to produce chemically resolved sur-face images of thin mouse liver slices con-taining 50 nm sized drug delivery contain-ers. Characteristic molecular fragments of PDMS (part B of an ABA triblock copo-lymer) can be identified in the TOFSIMS image and reflect the two-dimensional distribution of the containers on a scale of 500 x 500 μm2. The chemical composition of the polymer is shown in figure 1.As shown in figure 1, the synthetic nano-containers consist of a hydrophilic (PMOXA = poly(2-methyloxazoline) and a hydro-phobic (PDMS = poly(dimethylsiloxane) building block, giving the container wall an amphiphilc character [21]. The ABA copolymer forms stable vesicles in aque-ous media with a diameter of 50 – 100

nm (in some cases also up to 250 nm). The PMOXA-PDMS-PMOXA copolymer structure is biocompatible and shows very little susceptibility to non-specific plasma protein binding (e.g. serum al-bumin). Therefore the material has been chosen by Broz et al. [21, 22] as model for cell targeting using fluorescent markers for position location While the use of confocal microscopy is a key technology in nanomedicine due to its sensitivity and ability to perform time-resolved measurements that allow the experimental investigation of interac-tion processes, the method requires tag-ging of molecules with fluorophores [23]. This principle is not always accessible in a straightforward way, and therefore appli-cations are sometimes prone to synthetic restrictions or artifacts. Time-of-flight secondary ion ,mass spectrometry offers an alternative, label-free way to record spatial distributions of chemical species within samples of biological origin.

Experimental SectionTOF-SIMS spectra and 2D fragment dis-tribution maps were acquired using a TOFSIMS V instrument (IonTof GmbH, Münster, Germany) equipped with a Bi-liquid metal primary ion source angled at 45° to the sample plane. The Bi-cluster ion gun (25 keV, 10 kHz) produces maxi-mum pulsed target currents of 1 pA for Bi+, 0.45 pA for Bi3

+ and 0.3 pA for Bi32+,

respectively, in a high current bunched mode. However, the operation of the instrument in a mode for simultaneous high mass resolution in combination with high spatial resolution, as required for the detection of small molecules within a tis-sue sample, will require some alterations in the source settings. The trade-off for working with this option would be a lower target current and therefore also a longer measurement time for sufficient signal-to-noise ratio. Alternatively, the instru-ment can be operated in a so-called burst alignment mode, whith a reasonable

good lateral resolution and beam bright-ness, but only with unit mass resolution, which is only acceptable for high mass im-aging with minimal mass interferences. An extraction lens biased at ± 2000 V col-lected positive and negative secondary ions which were subsequently analyzed in a reflectron type mass filter. A maximum mass resolution of m/∆m ≥ 10’000 was achieved with this experimental set-up. A pre-biased microchannel plate detector operating in a single ion counting mode was used to obtain statistical numbers of secondary ion counts for each fragment. The sample itself was mounted onto a Ln2-cooled standard stage obtained as accessory from IonTOF GmbH (Münster, Germany). A maximum temperature of -120 °C was maintained through all mea-surements. TOF-SIMS images were recorded by ras-tering the primary ion beam across a pre-selected area of interest on the sample. As standard, a pixel resolution of 128 x 128 or 256 x 256 was chosen. A full data set consisted of one complete mass spec-trum for each pixel. However, very small features are difficult to analyze because in this case the size of the biological units approaches the minimum size of the pix-el. For features smaller than this dimen-sion, it becomes necessary to compare neighboring pixel intensities rather than averaging over the signal from a large number of pixels. The underlying concept of useful (or best) lateral resolution (∆L), defined as the side length of the mini-mum square area from which a number of N secondary ions of a certain mass are de-sorbed and detected has been introduced by Benninghoven and Kötter [24]. Several authors defined a number N = 4 as ade-quate for most samples. One might, how-ever speculate if this value is sufficiently describing the case of a small number of pixels. The influence of counting statistics in TOF-SIMS imaging experiments is still not fully explored.Samples of rat kidney, liver and spleen

Figure 1: Chemical structure of drug delivery container wall polymer according to ref. [14]


EuropEan JournaL of nanomEdicinE 2009 Vol. 2 issue 2 40

were prepared by cutting thin slices us-ing a standard laboratory microtome, transferring the tissue sections onto gold coated steel plates and shock freezing the specimen by immersing the plates into liquid nitrogen. After this procedure, the samples were stored in a LN2-storage container and kept in a frozen state until measurement. The transfer of the tissue slices onto the TOF-SIMS cold stage was performed in a portable glove box tent kept under inert and water free conditions to avoid excessive condensation of water and hydrocarbons onto the substrate. In order to avoid secondary contaminations caused by laboratory equipment, all sol-vents, cutting blades, tweezers, beakers and Ln2-Dewar containers were carefully cleaned before use. However, it was not possible to completely wipe-out all com-ponents of the microtome cutting instru-ment.

TOF-SIMS Images of Mouse Liver SlicesThe TOF-SIMS spectrum of positively charged secondary ions in the so-called fingerprint region between masses m/z = 1 - 100 is shown on the left-hand side of figure 2. Within one mass spectrum, there are a few hundred signals which in the-ory can each be assigned to a multitude of chemical compounds. Therefore, the identification of specific signals from the drug delivery containers can be difficult and time consuming. This rather complex situation is shown in figure 3.Therefore, the identification of small fea-tures within TOF-SIMS images requires a two-fold approach:

1. The possibility to perform TOF-SIMS imaging with high mass resolution and high lateral resolution at the same time. 2. A standardized approach to verify ob-served features using either mathemati-cal tools or an alternative analytical tech-nique that acquires images in another form than ion-matter interaction, e.g. fluorescence imaging [23, 25]. The development of primary ion beam technology meets the first requirement by operation the source in a so-called burst mode, where the ion pulse is sepa-rated into 4 sub-pulses (patented technol-ogy of IonTof GmbH, Münster, Germany). In this mode of operation, it is possible to achieve an average mass resolution of about m/∆m= 3000 for fragment signals in the range of m/z = 1- 300. Even with these restricted figures, the acquisition of secondary ion images is possible for spe-cific masses. However, in some cases the unit resolution of other existing operation modes is sufficient to separate a specific fragment signal from interfering masses. Note that the relatively low mass resolu-tion in our experiments is due to consid-

erable charging of the sample rather than a limited instrumental mass resolution. In theory, fragments of water are not sepa-rable from PDMS fragments at mass 73 (see table 1). However, under realistic ex-perimental conditions, the mass of PDMS fragments tends to shift for about +200 ppm, thus becoming distinguishable from ice fragments. The definition of a measurement tech-nique for data validation is in no way as easy as it seems. Only methods with com-parable interaction and pixel definition as found in secondary ion mass spectrom-etry will yield data sets that allow for the confirmation of observed distribution of metabolites or other molecules of bio-logical interest.Positive-ion images of rat liver sections are shown in images A to D of figure 4. A Bi+ primary ion beam, producing a pulsed ion current of 2 pA on the target was used in the burst mode operation mode. The distribution maps in figure 4 show 500 x 500 µm2 areas from left to right: A = total ion image, B = distribution of frag-ment mass m/z = 73, C = sum of specific

Figure 2: TOF-SIMS images of positively charged secondary ions obtained from mouse liver slices with A: total ion image, B: total ion image of rectangle in A, C: and D: distribution of mass m/z = 73 (characteristic fragment of drug delivery container. Arrows indicate the area distribution of characteristic fragments found in the survey mass spectrum on the left side.

ion Origin mass m/∆mc3H5o2

+ tissue 73.029 4058Si(CH3)3

+ container 73.047 -H9o4

+ water 73.050 24349c3H7no+ Tissue (???) 73.053 12174c4H11n

+ tissue 73.089 1739

Table 1: fragments around nominal mass m/z =73 which contribute to the signal interference at low mass resolution



Figure 3: High resolution spectra and TOF-SIMS images (500 x 500 µm2) of mouse liver slices, showing mass interferences at specific fragment mass (m/z = 73) of drug delivery containers. The situation clearly shows the need for high mass resolution spectroscopy combined with high lateral resolution imaging.

mass fragments of lipids and D = overlay image. The position of drug delivery con-tainers are marked as red dots fin figure 4D. Domains of an unknown compound, characterized by a fragment with a specif-ic mass of m/z = 269.27 are shown as well as silicon contaminations arising from the underlying substrate which suggest the presence of small fractures and holes in the sample. Despite the fact that all mea-surements were performed at a tempera-ture less than -125 °C, tissue samples were gradually loosing water, probably sublim-ing to the cold fingers of the LN2 cooling equipment. Although the integrity of the specimen may be altered somewhat by this effect, the positions of the containers are clearly visible as individual red points. The overlay image further indicates that no drug delivery containers were found within the lipid rich area. Whether this

finding has some therapeutic value is not clear and the answer to it will need further investigations.A minimum raster size of 60 x 60 µm2 was still yielding statistically useful data from our sample. However, as discussed be-fore, the significance of low signal levels found in some of the pixels are not yet ful-ly explored. Here, additional experiments providing refined statistical values have to be undertaken in the future.

ConclusionsAlthough ion-induced sputtering of or-ganic material is a highly destructive pro-cess in general, a TOF-SIMS instrument can be operated at a minimal destructive mode while at the same time maintaining well controlled low-temperature (-140°C) conditions over the measurement time. With this configuration, TOF-SIMS pro-

vides an analytical technique for 2D-map-ping of chemical species in tissue and cell samples. Here we present chemically re-solved images of thin slices of mouse liver tissue, containing small (50 nm) drug de-livery containers. The container wall con-sists of a biocompatible co-polymer. The distribution of characteristic molecular fragments of PDMS (part B of an ABA tri-block copolymer) can be identified in the TOF-SIMS image and the location of the containers on a scale of 500 x 500 μm2 can be specified. Presently the analysis area maintaining minimal material degrada-tion is limited to 60 x 60 μm2. In the future we plan to downscale this value in order to obtain 2D-images of low-abundance metabolites.

Figure 4: Positive-ion TOF-SIMS images (500 x 500 µm2) of mouse liver with A: total ion image, B: distribution of fragment m/z = 73, C: distribution of specific mass for lipids and D: overlay image showing drug delivery containers as red dots and the location of lipid rich areas as grey-green domains. Note that no containers are detected within the large lipid domain. Dark blue areas (2) indicate a yet unknown metabolite (m/z = 269.27and the light blue (1) spot is a hole in the slice with a silicon contamination from the substrate shining through. All measurements were performed at T < -125 °C.


EuropEan JournaL of nanomEdicinE 2009 Vol. 2 Issue 2 42

AcknowledgementsThe authors with to thank Prof. P. Hun-ziker and Mrs. R. Baenziger from Univer-sity Hospital Basel for stimulating discus-sions, providing mouse liver, kidney and spleen samples and help to prepare the tissue slices.References:1. B. D. Ratner in R. L. Reis and S. Weiner, eds, NATO Science Series II: Mathematics, Physics and Chemistry, Kluver Academic Publishers, Netherlands, 2004, “Surface Analysis of Biomaterials and Biomineral-ization”2.J. Grams, New Trends and Potentialities of ToF-SIMS in Surface Studies, “Bioma-terials, Biomolecules, Biologic, Biomate-rials, Biomolecules, Biological Systems and Applications in Medicine”, Nova Sci-ence Publishers, Inc, Hauppauge, USA, 2007, 21 – 593. Chilkoti, A. “Biomolecules on Surfaces”, in TOF-SIMS: Surface Analysis by Mass Spectrometry , Vickerman, J. C. , Briggs, D., eds, IM Publishers, Chichester, U.K., 2001, p.627 - 6504. Wagner, M. S., Castner, D. G., Applied Surface Science, 2004, 231 - 232, 366 – 376, “Analysis of Adsorbed Protein Films by Static TOF-SIMS”5. Wagner, M. S., Pasche, S., Castner, D. G., M. Textor, M., Analytical Chemistry, 2004, 76, 1483 -1492, “Characterization of Poly(L-Lysine)-graft-Poly(Ethylene Gly-col) Assembled Monolayers on Niobium Pentoxide Substrates using ToF-SIMS and Multivariate Analysis”6. Graham, D. J., Wagner, M. S., Castner, D. G., Applied Surface Science, 2006, 252, 6860 – 6868 “Information from complex-ity: Challenges of TOF-SIMS data inter-pretation”7. Benninghoven, A., Jaspers, D., Sichter-man, W. Appl. Phys. 1976, 11, 35 - 39, “Sec-ondary-Ion Emission of Amino Acids”8.Blain, M. G., Della-Negra, S., Joret, H., Le Bayec, J., Schweikert, E.A. Phys Rev. Lett. 1989, 63, 1625 -1628, “Secondary-ion yields from surfaces bombarded with keV molecular and cluster ions”9. Appelhans, A. D., Delmore, J. E. Anal. Chem. 1989, 61, 1087 – 1093, “Compari-son of Polyatomic and Atomic Primary Beams for Secondary Ion Mass Spectrom-etry of Organics”10. Gillen, G., Roberson, S. Rapid com-munications in Mass Spectrometry 1998, 12, 1303 – 1312, “Preliminary evaluation of an SF5

+ polyatomic primary ion beam for analysis of organic thin films by second-ary ion mass spectrometry”

11. Touboul, D., Halgand, F., Brunelle, A., Kersting, R. et al. Anal. Chem. 2004, 76, 1550 – 1559, “Tissue molecular ion imag-ing by gold cluster ion bombardment”12. Touboul, D., Kollmer, F., Niehuis, E., Brunelle, A., et al. J. Am. Soc. Mass Spec-trom. 2005, 16, 1608 – 1618, “Improve-ment of biological time-of-flight second-ary ion mass spectrometry imaging with a bismuth cluster ion source”13.Postawa Z., Czerwinski, B., Szewczyk, M., Edward J. Smiley, E. J., Winograd, N., Garrison, B., J. Phys. Chem. B 2004, 108, 7831 – 7838, “Microscopic Insights into the Sputtering of Ag{111} Induced by C60 and Ga Bombardment”14.Winograd, N. Analytical Chemistry 2005, 77, 143A – 149A, “The Magic of Clus-ter SIMS”15.Kurczy, M. E., Piehowski, P.D., Parry, S. A., Jiang, M., Chen, G., Ewing, A. G., Winograd, N. Appl. Surf. Sci. 208, 255, 1298 – 1304, “Which is more important in bioimaging SIMS experiments – The sample preparation or the nature of the projectile?”16. McDonnell, L. A. and Heeren, R. M. Mass Spectrometry Reviews 2007, 26, 606 – 643, “Imaging mass spectrometry”17. Levi-Setti, R., Gavrilov, K. L., Strissel, P. L., Strick, R. Appl. Surf. Sci 2004, 231, 479 – 484, “Ion microprobe imaging of 44Ca-labeled mammalian chromosomes”18. Levi-Setti, R., Gavrilov, K. L., Neilly, M. E., Appl. Surf. Sci. 2005, 252, 6765 – 6769, “Cations in mammalian cells and chromo-somes: Sample preparation protocols af-fect elemental abundances by SIMS”19. Sjovall, P., Lausmaa, J., Johannson, B. Analytical Chemistry 2004, 76, 4271 – 4278 “Mass Spectrometric Imaging of Lipids in Brain Tissue”20. Roddy, T.P., Cannon, D.M., Ostrowski, S.G., Winograd, N., Ewing, A.G. Anal. Chem 2002, 74, 4020 – 4026, “Identifica-tion of Cellular Sections with Imaging Mass Spectrometry Following Freeze Fracture”21. Broz, P., Benito, S. M., Saw, CL, Burg-er, P., Heiderd, H., Matthias Pfisterer, M., Marsch, S., Meier, W. Hunziker, P. Jour-nal of Controlled Release 2005, 102, 475 – 488, “Cell targeting by a generic recep-tor-targeted polymer nanocontainer plat-form” 22. Broz, P., Driamov, S. , Ziegler, J., Ben-Haim, N., Marsch, S., Meier, W., Hunziker, P. Nano Letters 2006, 6, 2349-2353, „To-ward Intelligent Nanosize Bioreactors: A pH-Switchable, Channel-Equipped, Func-tional Polymer Nanocontainer”

23. Pickup, J., Khan, F., Zhi, Z.-L., Saxl, T. Europ. J. Nanomed. 2009, 2 , 16 – 21, “Fluorescence in Nanometrology: its Potential Role in Diabetes Research and Management”24. Kötter, F., Benninghoven, A. Appl. Surf. Sci 1998, 133, 47 – 57, “Second-ary ion emission from polymer surfaces under Ar+, Xe+ and SF5+ ion bombard-ment”25. Valeur, B. “Molecular Fluorescence – Principle and Applications”, Wiley-VCH (Weinheim), 2002


Polymer vesicles loaded with precipitated gadolinium nanoparticles: A novel target-specific contrast agent for magnetic resonance imagingPavel Broz1*, Nadav Ben-Haim1, Francesco Santini2, Stephan Marsch1, Klaus Scheffler2, Wolfgang Meier3

and Patrick Hunziker1* (DOI 10.3884/0002.2.9)

AbstractGadolinium-based contrast agents for imaging of cellular or subcellular compartments by clinical magnetic resonance imaging (MRI) currently lack a satisfying signal-to-noise ratio and acceptable target specificity. Water-soluble enzyme nanoreactors based on amphiphilic triblock copolymer membranes functionalized with size-selective pore proteins were used to precipitate highly insoluble gadolinium phosphate nanopar-ticles inside the vesicles. Surface modification with the macrophage scavenger receptor A1 (SRA-1) ligand polyguanylic acid and fluorescence-labeled streptavidin resulted in a novel type of bimodal target-specific contrast agent for MR and fluorescence imaging. Physicochemical analyses showed vesicles with a mean di-ameter of 46 +/- 9 nm and distinguishable particulate content with high electron density. In MRI, the nano-carrier displayed a T1 relaxivity of 3.5e10 L x s(-1) x mmol(-1) vesicles. Uptake to target cells and intracellular localization was proven both with confocal microscopy and MRI. In conclusion, we introduce a novel recep-tor-targetable gadolinium nanoparticle-based vesicular contrast agent for bimodal imaging in clinical medi-cine and show its use in first in vitro and cell culture experiments. Broz, P Poe, D Nanomedicine 2009, 2.2:43-48Keywords: clinical magnetic resonance imaging, confocal microscopy, nanoparticles, intracellular localization

1Medical Intensive Care Unit, University Hospital Basel, Petersgraben 4, 4031 Basel, Switzerland2MR-Physics, Department of Radiology, University Hospital Basel, Petersgraben 4, 4031 Basel, Switzerland3Department of Chemistry, University of Basel, Klingelbergstrasse 80, 4056 Basel, Switzerland* To whom correspondence and requests for materials should be addressed. E-mail: [email protected] and [email protected]

Nano Imaging Technologies

IntroductionSupramolecular nanometer-sized struc-tures such as particles, micelles, or vesicles built from synthetic copolymer materials have aroused enormous interest in recent years and promise to be useful for novel or improved biomedical applications(1, 2). The common goal of these projects is the development of well-defined, self-assem-bled, highly organized, multifunctional, biocompatible, non-immunogenic, and target-specific tools. Nanometer-sized polymer vesicles self-assembled in aque-ous solution from the amphiphilic triblock copolymer (poly(2-methyloxazoline)-b-poly(dimethylsiloxane)-b-poly(2-methy-loxazoline) (PMOXA-PDMS-PMOXA)(3, 4) are one example of such a supramo-lecular structure and have demonstrated to fulfill the above stated conditions(5, 6). The copolymer has shown to build highly stable and homogenous unilamellar ves-icles with a predictable and controlled diameter (mostly defined by the chain lengths and the proportions of the indi-

vidual polymer blocks) when introduced into an aqueous environment(3), thereby enclosing a defined proportion of the so-lution into the aqueous core of the vesi-cle(7). The end group of the PMOXA poly-mer building block allows chemical modi-fications of the vesicle surface, e.g. with biotin-avidin-biotin coupled oligonucle-otide ligands for specific cell receptor tar-geting(5). Furthermore, the fully synthet-ic vesicles can be equipped with biological transmembrane proteins such as OmpF(6, 8), LamB(9), Tsx(10), and FhuA(11) to en-able a size-, charge-, or molecule-specific transmembrane transport. Encapsulated enzymes of bacterial or plant origin finally create an organized nanometer-sized bio-reactor that can be used for activation of prodrugs(10) or pH-controllable hydroly-sis of phosphatase substrates(6).Here, we present the use of this versatile polymer vesicle system for the creation of a novel receptor-targeted contrast agent for flu-orescence microscopy and magnetic res-onance imaging (MRI)(12). Based on the

relaxation properties of excited hydro-gen nuclei in body water and fat, MRI is being used to obtain anatomical images of physiological and pathological tissues and functional information on the patho-genesis of inflammation, cancer, or isch-emic diseases (stroke, myocardial infarc-tion). To enhance the gained information, contrast agents have been introduced to clinical practice and help to detect and vi-


Figure 1: Schematic visualization of the nanoparticle precipitation concept. The ABA triblock copolymer vesicles are functionalized with transmembrane OmpF pore proteins and the track-ing ligand polyguanylic acid. Fluorescence-labeled streptavidin is being used to couple the lig-and to the vesicle surface via biotin interaction. The pH-triggerable enzyme acid phosphatase is encapsulated inside the vesicle to create a nanometer-sized bioreactor (A). In a second step, glucose-1-phosphate and gadolinium chloride are added and processed to insoluble gadolinium phosphate precipitates by enzymatic action (B). The final targetable polymer vesicle based con-trast agent is packed with multiple gadolinium phosphate nanoparticles (C).

sualize pathological processes. Tradition-al agents for contrast-enhanced MRI are based on the paramagnetic properties of the gadolinium (III) ion, a rare earth met-al with 7 unpaired f electrons. Gadolinium decreases the T1 relaxation time of the surrounding hydrogen nuclei, resulting in an increase of signal strength. To prevent toxic effects on liver, spleen, and bone marrow upon injection, the gadolinium ions are typically administered as high stability complexes with chelators such as diethylentriaminepentaacetic acid (DTPA) or tetraazacyclododecanetetraacetic acid (DOTA)(13). Due to the low amount of gadolinium atoms compared to the total mass of the chelated construct, the sig-nal-to-noise ratio is consequently insuf-ficient for cellular or subcellular imaging with these contrast agents(14). The use of contrast agents based on chelated gado-linium (III) is therefore limited to imaging of organ function, organ perfusion(15), inflammation, and other water accumu-lations.To circumvent these problems and to al-low molecular magnetic resonance imag-ing(16), new contrast agents have been developed in the last years that might allow an imaging of defined populations of cancer cells, stem cells, or white blood cells and even of subcellular constituents. The most prominent example are poly-mer (mostly dextran) – coated superpara-magnetic iron oxide (SPIO) nanoparticles with diameters ranging roughly from 5 nm to 200 nm(17). They have shown to be beneficial for detection of liver tumors in clinical practice(18) and are currently be-ing modified for improved targeting and detection applications, e.g. for cancer or atherosclerosis(19). However, their pre-dominant effect on T2/T2* relaxation times and consequently a negative con-trast enhancement in standard MR imag-ing sequences impede a wide application of this contrast agent for standard clinical molecular imaging. In contrast to SPIO-based contrast agents, gadolinium-based agents show a decrease of T1 relaxation times, result-ing in a positive contrast enhancement in standard MR imaging sequences. Even so, most ongoing research in the field of target-specific contrast agents with pre-dominant T1 effect relies on the above mentioned chelated gadolinium ions. The use of crystalline gadolinium complexes promises to be an important step forward in the field of cellular or subcellular diag-nostics with MRI, but problems with tox-

icity, water solubility, and stability need to be solved to fulfill the promises(20). As a model to demonstrate the first use of our novel polymer vesicle based agent, we have chosen cell cultures of highly ac-tive macrophages(5). Macrophages are phagocytic monocyte-derived cells of the innate immune system and play a key role in major diseases such as atherosclerosis, cancer, and autoimmune disorders. In the case of atherosclerosis, macrophages are main constituents of type V vulnerable plaques(21) that are in risk of shear-in-duced rupture that in the case of coronary arteries leads to myocardial infarction. The macrophages represent up to 20% of total cell amount of the fibrous cap of vul-nerable plaques(22) and weaken the me-chanical resistance of the cap mainly by producing potent matrix metalloprotein-ases and attracting lymphocytes(23). A high percentage and activity of the plaque macrophages correlates significantly with the risk of plaque rupture and therefore vessel thrombosis and myocardial infarc-tion (24). For a successful prevention of this disease, it is vital to detect high risk plaques and treat them specifically. Most

concepts for vulnerable plaque detection such as intracoronary thermography(25), radionuclide imaging(26), and targeted SPIOs(27) represent basically the macro-phage distribution and their activity state, but have several limitations until now(28). Experimental proceduresExperimentally, acid phosphatase con-taining, macrophage targeted ligand functionalized nanoreactors self-as-sembled from triblock copolymer mem-branes and size-selective, bacterial OmpF pores(6) were incubated with gadolinium chloride (GdCl3) and a phosphatase sub-strate that are able to reach the interior of the vesicles through the pores (Fig. 1). Inside the polymer vesicles, the hydroly-sis of the phosphatase substrate by the encapsulated enzyme results in the pro-duction of free phosphate groups (phos-phos-phoric acid, respectively), followed by a, followed by a rapid precipitation of the highly water-in-soluble hydrated gadolinium phosphate (GdPOGdPO4·xH2O) as described by Lessing et) as described by Lessing et al.(29). Synthesis of gadolinium phos-phate containing polymer vesicles



Figure 2: A) Cryogenic transmission electron microscopy of the final targetable contrast agent shows a homogenous population of vesicles with a mean diameter of slightly less than 50 nm and polyparticulate content with high density. B) A control population of polymer vesicles filled only with acid phosphatase shows similar mean diameter, but a homogenous content with no dense particles. The images prove the presence of electron-dense particles inside the polymer vesicles.

Nanoreactors based on polymer vesi-cles from (poly(2-methyloxazoline)-b-poly(dimethylsiloxane)-b-poly(2-methy-loxazoline) triblock copolymer building blocks (JW05, total Mn 7090 g/mol, 10% with biotin-modified end-groups) were produced as described previously(6). The main features are encapsulated acid phosphatase (total molecular weight 55 kDa; Sigma-Aldrich, Buchs, CH) and OmpF pore proteins (total molecular weight 110 kDa) that integrate into the polymer membrane. In the next step, 33 mmol/L glucose-1-phosphate (Sigma-Al-drich) and 100 mmol/L gadolinium chlo-ride (GdCl3) (Sigma-Aldrich) were add-ed and the solution was adjusted to pH 5 with 1 mol/L hydrochloric acid. Both constituents reached the interior of the polymer vesicles through the pore pro-teins driven by passive diffusion, where the glucose-1-phosphate was hydrolyzed by the acid phosphatase. The resulting free phosphate groups (or phosphoric acid, respectively) and the gadolinium (III) ions now precipitated inside the vesicles as highly insoluble hydrated gadolinium phosphate (GdPOGdPO4·xH2O)(29). After 24 h,(29). After 24 h, the surface of the complete vesicle was further functionalized by attaching the SRA-1 ligand polyguanylic acid (Micro-synth, Balgach, CH; 23 G, 3` modified with biotin) as described previously(5). Strep-tavidin-Alexa Fluor® 610 (Invitrogen, Ba-sel, CH) was used as a coupling molecule to label the vesicles fluorescently. The fi-nal vesicles were then purified from non-encapsulated and non-bound molecules by gel chromatography (Sepharose® 4B, Sigma-Aldrich, Buchs, CH in a 37 cm col-umn with 1 cm inner diameter, Bio-Rad, Reinach, CH).

Characterization of polymer vesicles For characterization, the vesicles were imaged by cryogenic transmission elec-tron microscopy (LEO 910, Carl Zeiss AG, Feldbach, CH) and atomic force micros-copy (Agilent 5100, Molecular Imaging, Tempe, USA). Cryo-TEM was performed according to standard methods. For AFM, the vesicle solution was dropped on a mica layer and dried in a desiccator for 2 hours. The samples were analyzed in tap-ping mode with a silicon cantilever (fr = 300 kHz). Magnetic resonance experi-ments were performed on a 1.5T human imaging scanner (Magnetom Avanto, Sie-mens Medical Solutions, Erlangen, Ger-many) on 1 ml vials filled with a solution in water of the sample at different con-

centrations (5e12, 3.75e12, 2.5e12, 5e11 polymer vesicles/ml) submerged in wa-ter to provide enough signal for gradient shimming. Standard Spin-Echo sequenc-es with different echo times (TE = 10, 30, 50 ms, Repetition Time TR = 4 s) and Gra-dient-Echo with Inversion Recovery mag-netization preparation (Inversion Times IR = 150, 300, 500, 750, 1000, 1500 ms, Echo Time TE = 3.61 ms, TR = 4 s) were used as a source of data for the exponential fitting needed to calculate T1 and T2 values of the solutions. The same sequence proto-col was also used to compare the magnet-ic properties of the samples with different concentrations (1 mmol/L, 250 µmol/L, 50 µmol/L) of a commercially available gado-butrol contrast agent (Gadovist®, Scher-ing, Berlin, Germany) and to analyze the suspended cell samples.

Cell culture experiments

THP-1 cells (ECACC, Salisbury, UK) were cultured as recommended by the manu-facturer. Differentiation into functional adherent macrophages was initiated 72 hours prior to experiment with 100 nmol/L phorbol 12-myristate 13-acetate (PMA) (Sigma-Aldrich) in 24-well multiwell plates (Becton-Dickinson, Basel, Switzer-land) with glass cover slips (12 mm diam-eter), resulting in a mean cell density of approximately 1.25e5 cells/cm2. For ves-icle uptake experiments, macrophages were incubated with 4 nmol/L vesicles for 1 h. After washing with PBS, the cells were fixed in 4% paraformaldehyde for 20 min at room temperature. Fixed cells

were mounted with an anti-bleaching glycerol-based mounting medium (Sig-ma-Aldrich) on glass slides. Fluorescence microscopy was done by laser confocal microscopy Zeiss ConfoCor 2, HAL 100, HBO 100, Software LSM 510 Zeiss) as de-scribed previously(5). For MRI scans, the cells were detached from the glass cover slips with 0.5% trypsin (Sigma-Aldrich) af-ter the incubation period and resuspend-ed in cell culture medium. Cell numbers were determined using standard micro-scope cell counting chambers. The treat-ed cell population was compared to un-treated cells and free water. To assess the toxicity of the gadolinium phosphate con-taining polymeric vesicles, we performed a colorimetric LDH release assay (Sigma-Aldrich) in 96-well multiwell plates. Dead cells release intracellular LDH into the serum-free medium, where it is detect-ed with a standard method based on the NADH-mediated conversion of a tetrazo-lium dye. The dye was detected with an automated microplate absorption reader (MPP 4001, Mikrotek, Overath, Germany) at a wavelength of 490 nm. Ratio of living cells in the sample compared to the cell number in an untreated control was taken as cell survival parameter.

Results Physicochemical characterization of the vesicles was done by cryogenic transmis-sion electron microscopy (cryo-TEM) and atomic force microscopy (AFM). Locali-zation and toxicology experiments were performed in cell cultures of human THP-



Figure 3: Same samples as in Fig. 2 were studied with atomic force microscopy in tapping mode after drying on a mica layer. Final polymer vesicles with precipitated gadolinium phosphate na-noparticles (A-D) show a continuous layer of retracted polymer material and islands with a de-fined hexagonal pattern (detail in C and D) of a hard material in phase imaging. Control polymer vesicles filled only with acid phosphatase (E and F) show similar layers of polymer material, but no hexagonal pattern in-between. The images prove the presence of hard matter, most likely gadolinium phosphate nanoparticles that are building an organized pattern upon drying of the vesicle solution.

1 macrophages and analyzed by confocal laser microscopy and a light absorption microassay. MR experiments were per-formed on a 1.5T human imaging scan-ner as described in the methods section. The resulting vesicles had a mean diame-ter of 46 nm +/- 9 nm and distinguishable polyparticulate, high density content in cryo-TEM. Control vesicles that contained only acid phosphatase but did not under-go the nanoparticle producing reaction

had a diameter in the same range, but homogenous low density content (Fig. 2). The images indicate the presence of multiple particles with high electron den-sity and a size in the range of 5 nm inside the polymer vesicles that did undergo the gadolinium phosphate precipitating reac-tion.The same polymer vesicles, gadolinium phosphate containing and the control, were then dried on a mica layer for AFM. Measurements in tapping mode in air with

a silicon cantilever revealed characteristic aggregations of soft, most probably poly-meric material as seen in Fig. 3 in phase imaging and islands with a regular, hex-agonal pattern of hard material between the soft aggregations. The control vesicles did not produce the hexagonal pattern af-ter drying on the mica layer. The regular pattern of hard material is most likely a deposition of the water-insoluble gado-linium phosphate nanoparticles, while the water-soluble polymer is retracted into the aggregates during the drying process. In cell cultures of active macrophages, the polymer vesicle contrast agent was tested for fluorescence imaging. By us-ing a fluorescence labeled streptavidin for ligand coupling, the vesicles can be easily tagged with a variety of stable and bright dyes. The targeting moiety polyguanylic acid guarantees the specific delivery of the polymer vesicles to active, scavenger receptor expressing macrophages(5). Af-ter 1 h incubation time, the ligand func-tionalized fluorescent polymeric vesicles could be found in high numbers inside macrophages (Fig. 4A). Macrophages treated with control vesicles without the fluorescence labeled streptavidin showed no significant intracellular fluorescence (Fig. 4B). A confocal microscopy z-stack analysis of macrophages that were treated with the polymer vesicles revealed several hundreds to thousands of distinguishable signals in vesicular shape inside the cells. The results demonstrate the potential of the polymeric vesicles as high signal im-aging contrast agents for fluorescence detection. In MRI, an aqueous solution with 8 nmol/L gadolinium phosphate con-taining vesicles showed a T1 relaxation time of 949 ms and a T2 relaxation time of 367 ms. Similar values were achieved with 50 µmol/L [Gd] gadobutrol, whereas free water resulted in a T1 relaxation time of 1879 ms and T2 relaxation time of 414 ms. Following control vesicles did not gen-erate a significant change in relaxation times compared to free water: 1) no pore proteins in the membrane; 2) no encapsu-lated acid phosphatase; 3) no GdCl3 during the synthesis; 4) no glucose-1-phosphate during the synthesis. T1 relaxivity was de-termined by a linear fitting of the calculat-ed T1 values for different concentrations and a value of 9.1e-17 ml/(msec x number of vesicles) was found (Fig. 5). Based on mol [vesicles], this corresponds to 3.5e10 L x s(-1) x mmol(-1). Samples with 50`000 polymer vesicle-treated macrophages (same as in Fig. 4) in a suspension of 1



Figure 4: Confocal microscopy imaging of THP-1 macrophages treated with fluorescence-la-beled polymer vesicles (A) and unlabeled vesicles (B) shows the intracellular localization of the fluorescence labeled gadolinium phosphate containing polymer vesicles after an incubation time of 1 hour. Owing to the targeting ligand polyguanylic acid, the vesicles are taken up by scavenger receptor expressing, active macrophages in high numbers. Cells without scavenger receptor expression do not take up the polymer vesicles. The results demonstrate the applica-bility of the contrast agent for bimodal imaging of macrophages with fluorescence detection methods (Online data supplement shows z-stack of image A).

Figure 5: MRI measurements. A) Gradient echo with inversion recovery (IR=1500ms, TR=4000ms). Below the samples, the respec-tive concentration of polymer vesicles/ml is shown. B) Linear fitting of 1/T1 values for re-laxivity measurement. The results demon-strate the linear positive signal enhancement when increasing the vesicle concentration in the solution. The polymer vesicle contrast agent shows a gadolinium typical predomi-nant T1 effect.

Figure 6: Toxicity of gadolinium phosphate containing polymer vesicles on THP-1 mac-rophages was assessed with a standard cell death assay (see methods section for details). Macrophages were incubated with increas-ing concentrations of polymer vesicles for 24 h and tested for membrane integrity with an automated microplate based assay. No signifi-cant increase of cell death could be detected during the observation period, indicating that gadolinium phosphate loaded polymer vesi-cles are not cytotoxic (error bars represent the standard deviation over a series of experi-ments).

ml phosphate buffered saline showed a detectable increase of signal in MRI com-pared to untreated cells and free water. The relative grey value intensity of the treated cells was 56% higher than the sig-

nal of the untreated cells and 47% higher than the signal of free water.Gadolinium ions are known to exhibit strong toxic effects on cells(30), therefore we performed a standard cytotoxicity testing using the LDH release assay (see methods for details). Macrophages were treated with increasing doses of gado-linium phosphate containing polymer vesicles for 24 hours and tested for mem-brane integrity compared to populations of untreated cells. Up to a concentration of 7 nmol/L vesicles in the cell culture me-dium, there was no significant increase of cell death in the treated macrophage cultures (Fig. 6). The results indicate that the vesicle based bimodal contrast agent does not exhibit acute toxicity in cell cul-tures.

DiscussionIn conclusion, we introduce a novel re-ceptor-targetable polymer vesicle-based contrast agent for MR and fluorescence imaging in clinical medicine and show its use in in vitro and cell culture experi-ments. The environment-reactive enzyme nanoreactor system based on polymeric biomimetic membranes can be used to control the precipitation of gadolinium minerals locally and temporally, to target a specific cell line, and to protect the tar-get cells from toxic effects of gadolinium. The gadolinium phosphate nanoparticles are protected inside the polymer vesicles and can be transported to a defined cellu-

lar receptor or surface structure. The pol-ymer vesicles exhibit a gadolinium typical decrease of T1 relaxation time and are taken up by active macrophages in high numbers. The T1 relaxivity of 3.5e10 L x s(-1) x mmol(-1) [vesicle] is significantly higher than the T1 relaxivity of commer-cially available MRI contrast agents (e.g. 3.6 L x s(-1) x mmol(-1) for gadobutrol in water(31)). This extremely strong T1 re-laxivity per single vesicle opens the door for cellular and molecular imaging with standard MRI scanners. A small number of 50`000 macrophages can be detect-ed with MRI when using the targetable vesicles. Unwanted acute toxic effects of gadolinium on the targeted cells are ab-sent and should not endanger the use of the novel polymer vesicles based contrast agent for molecular imaging of macro-phages, e.g. in inflamed atherosclerotic plaques. Future experiments will investi-gate the intracellular fate of the polymer vesicle constituents. The use of stable, biocompatible and protein-repellent tri-block copolymer vesicles with a biological targeting function solves the stability and solubility problems of gadolinium miner-als in an elegant way.Compared to superparamagnetic iron oxide nanoparticles, the polymer vesi-cle system is based on the predominant T1 effect of gadolinium and produces a positive increase of signal when using a standard MRI scanner for humans. No additional imaging hardware or soft-ware is necessary for the detection of injected gadolinium phosphate contain-



ing polymer vesicles, a clear advantage compared to other nanotechnological contrast agent approaches. Furthermore, by exchanging the tracking ligand on the vesicles surface, the whole system could be used for bimodal MR and fluorescence imaging of other diseases such as cancer or infections.

AcknowledgementsThis work was financed by the Swiss Na-tional Science Foundation and the Na-tional Centre of Competence in Research Nanoscale Science. The authors thank Joerg Ziegler, University Hospital Basel for his aid with cryo-TEM and Susana Moreno-Flores, University of Basel for her aid with atomic force microscopy.Competing interests statement: The au-thors declare no competing interests.

References(1) Putnam, D. (2006) Polymers for gene deliv-ery across length scales. Nat Mater 5, 439-51. (2) Uchegbu, I. F. (2006) Pharmaceutical nano-technology:polymeric vesicles for drug and gene delivery. Expert Opin Drug Deliv 3, 629-40. (3) Nardin, C., Hirt, T., Leukel, J., and Meier, W. (2000) Polymerized ABA triblock copo-lymer vesicles. Langmuir 16, 1035-1041. (4) Nardin, C., and Meier, W. (2002) Hybrid ma-terials from amphiphilic block copolymers and membrane proteins. J Biotechnol 90, 17-26. (5) Broz, P., Benito, S. M., Saw, C., Burger, P., Heider, H., Pfisterer, M., Marsch, S., Meier, W., and Hunziker, P. (2005) Cell targeting by a ge-neric receptor-targeted polymer nanocontain-er platform. J Control Release 102, 475-488. (6) Broz, P., Driamov, S., Ziegler, J., Ben-Haim, N., Marsch, S., Meier, W., and Hunziker, P. (2006) Toward intelligent nanosize bioreactors: A pH-switchable, channel-equipped, functional poly-mer nanocontainer. Nano Lett 6, 2349-2353. (7) Rigler, P., and Meier, W. (2006) Encap-sulation of fluorescent molecules by func-tionalized polymeric nanocontainers: in-vestigation by confocal fluorescence im-aging and fluorescence correlation spec-troscopy. J Am Chem Soc 128, 367-73. (8) Meier, W., Nardin, C., and Winterhalter, M. (2000) Reconstitution of channel proteins in (polymerized) ABA triblock copolymer mem-branes. Angew Chem Int Ed Engl 39, 4599-4602.(9) Graff, A., Sauer, M., Van Gelder, P., and Meier, W. (2002) Virus-assisted loading of polymer nanocontainer. Proc Natl Acad Sci U S A 99, 5064-5068.(10) Ranquin, A., Versees, W., Meier, W., Stey-aert, J., and VanGelder, P. (2005) Therapeutic nanoreactors: Combining chemistry and biol-ogy in a novel triblock copolymer drug deliv-erysystem. Nano Lett 5, 2220-2224.(11) Nallani, M., Benito, S., Onaca, O., Graff, A., Lindemann, M., Winterhalter, M., Meier,

W., and Schwaneberg, U. (2006) A nanocom-partment system (synthosome) designed for biotechnological applications. J Biotechnol 123, 50-59.(12) Pykett, I., Newhouse, J., Buonanno, F., Brady, T., Goldman, M., Kistler, J., and Pohost, G. (1982) Principles of nuclear magnetic reso-nance imaging. Radiology 143, 157-168.(13) Wedeking, P., Kumar, K., and Tweedle, M. F. (1992) Dissociation of gadolinium chelates in mice: relationship to chemical characteris-tics. Magn Reson Imaging 10, 641-8.(14) Wilensky, R. L., Song, H. K., and Ferrari, V. A. (2006) Role of magnetic resonance and in-travascular magnetic resonance in the detec-tion of vulnerable plaques. J Am Coll Cardiol 47, C48-C56.(15) Constantine, G., Shan, K., Flamm, S. D., and Sivananthan, M. U. (2004) Role of MRI in clinical cardiology. Lancet 363, 2162-2171.(16) Sosnovik, D. E., Nahrendorf, M., and Weissleder, R. (2007) Molecular magnetic res-onance imaging in cardiovascular medicine. Circulation 115, 2076-2086.(17) Matuszewski, L., Persigehl, T., Wall, A., Schwindt, W., Tombach, B., Fobker, M., Poremba, C., Ebert, W., Heindel, W., and Bremer, C. (2005) Cell tagging with clinically approved iron oxides: Feasibility and effect of lipofection, particle size, and surface coat-ing on labeling efficiency. Radiology 235, 155-161.(18) Nasu, K., Kuroki, Y., Nawano, S., Kuroki, S., Tsukamoto, T., Yamamoto, S., Motoori, K., and Ueda, T. (2006) Hepatic metastases: Dif-fusion-weighted sensitivity-encoding versus SPIO-enhanced MR imaging. Radiology 239, 122-130.(19) Nahrendorf, M., Jaffer, F. A., Kelly, K. A., Sosnovik, D. E., Aikawa, E., Libby, P., and Weissleder, R. (2006) Noninvasive vascular cell adhesion molecule-1 imaging identifies inflammatory activation of cells in atheroscle-rosis. Circulation 114, 1504-1511.(20) McDonald, M. A., and Watkin, K. L. (2006) Investigations into the physicochemical prop-erties of dextran small particulate gadolinium oxide nanoparticles. Acad Radiol 13, 421-427.(21) Stary, H. C., Chandler, A. B., Dinsmore, R. E., Fuster, V., Glagov, S., Insull, W., Jr, Rosen-feld, M. E., Schwartz, C. J., Wagner, W. D., and Wissler, R. W. (1995) A definition of advanced types of atherosclerotic lesions and a histo-logical classification of atherosclerosis : A re-port from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Arterioscler Thromb Vasc Biol 15, 1512-1531.(22) Tearney, G. J., Yabushita, H., Houser, S. L., Aretz, H. T., Jang, I.-K., Schlendorf, K. H., Kauffman, C. R., Shishkov, M., Halpern, E. F., and Bouma, B. E. (2003) Quantification of macrophage content in atherosclerotic plaques by optical coherence tomography. Circulation 107, 113-119.(23) Moreno, P., Falk, E., Palacios, I., Newell, J., Fuster, V., and Fallon, J. (1994) Macrophage

infiltration in acute coronary syndromes. Im-plications for plaque rupture. Circulation 90, 775-778.(24) Hansson GK. (2005) Inflammation, ath-erosclerosis, and coronary artery disease. N Engl J Med. 352 (16), 1685-95.(25) Madjid, M., Willerson, J. T., and Casscells, S. W. (2006) Intracoronary thermography for detection of high-risk vulnerable plaques. J Am Coll Cardiol 47, C80-C85.(26) Davies, J. R., Rudd, J. H. F., Weissberg, P. L., and Narula, J. (2006) Radionuclide imaging for the detection of inflammation in vulnera-ble plaques. J Am Coll Cardiol 47, C57-C68.(27) Jaffer, F. A., Libby, P., and Weissleder, R. (2006) Molecular and cellular imaging of ath-erosclerosis: Emerging applications. J Am Coll Cardiol 47, 1328-1338.(28) MacNeill, B. D., Lowe, H. C., Takano, M., Fuster, V., and Jang, I.-K. (2003) Intravascular modalities for detection of vulnerable plaque: Current status. Arterioscler Thromb Vasc Biol 23, 1333-1342.(29) Lessing, P. A., and Erickson, A. W. (2003) Synthesis and characterization of gadolinium phosphate neutron absorber. J Eur Ceram Soc 23, 3049-3057.(30) Shellock, F. G., and Kanal, E. (1999) Safe-ty of magnetic resonance imaging contrast agents. J Magn Reson Imaging 10, 477-484.(31) Goyen, M., Herborn, C. U., Vogt, F. M., Kröger, K., Verhagen, R., Yang, F., Bosk, S., Debatin, J. F., and Ruehm, S. G. (2003) Using a 1 M Gd-chelate (gadobutrol) for total-body three-dimensional MR angiography: Prelimi-nary experience. J Magn Reson Imaging 17, 565-571.



Primary contact: Beat Löffler, MA, CEO of the European Foundation for Clinical Nanomedicine, Alemannengasse 12, P.O. Box, CH-4016 Basel, Switzerland, +41 61 695 93 95 (phone), +41 61 695 93 90 (fax), +41 78 654 37 07 (mobile), [email protected] (E-mail), www.clinam.org

Editorial team: Patrick Hunziker, MD, Prof. (Chief Editor), Beat Löffler, MA, Wolfgang Meier, Prof., Giacinto Sco-les, Prof.

Section Editors: Martin Erdmann, PhD, Patrick Hunziker, MD, Beat Löffler, MA, Felix Fluri, MD

Layout Editor: Sigrid Gombert, Dipl.Ing., Dipl.Bio.

Editing Office: European Foundation for Clinical Nanomedicine, Alemannengasse 12 PO Box, CH 4016 Basel, Switzerland

Focus and scope: The European Journal of Nanomedicine is a scientific journal focusing on the clinical application of nanoscience tools, methods and materials and on the exploration of the implications of nano-medicine. Its readers consist of physicians who will apply such technology to their patients, of scientists and engineers who develop such technologies in view of clinical applications, and of the informed public and policy makers who are interested in the most recent developments in the field.Original research articles are peer reviewed, typically by 2 reviewers, if the article, as judged by the editorial office, covers the scope of the journal, is considered of sufficient priority for the readers and has no evident significant flaws. Reviewers are invited to judge papers according their scientific visibility in the field. An all-electronic submission-reviewing-revision-acceptance system allows to reduce the time delays from submission to publication to a minimum.

Author guidelines: Submission text and each submission figure are to be uploaded as separate files. Additional sup-portive material like data-sets, research materials etc is also welcome, for electronic access. As part of the submission process, authors are required to check off their submission‘s compliance with all of the following items, and submissions may be returned to authors that do not adhere to these guidelines 1. The submission has not been previously published, nor is it before another journal for consideration (or an explanation has been provided in Comments to the Editor). 2. The submission file is in OpenOffice, Microsoft Word or RTF document file format. For all figures, tables etc, a placeholder of the form FIGURE_1, TABLE_3 or similar is placed within the text as a separate paragraph. Figure captions and table captions are included at the end of the text files. 3. Where available, URLs for the references have been provided. 4. Tables are produced with using OpenOffice, Word or RTF to be included with the text; alternatively, they are delivered as sepa-rate Excel file. Figures are in standard graphics formats (JPG,TIF,BMP,PNG,SVG, other formats on request) and are uploaded as supplementary material using file names like FIGURE_1.jpg or TABLE_3.xls . Further supplementary material carries the label SUPPLEMENT_1 etc. Each up-loaded file is not larger than 5 megabytes. Large image files may be ZIP compressed. 5. The text adheres to the stylistic and bibliographic requirements of the APA style (http://en.wikipedia.org/wiki/APA_style). 6. If submitting to a peer-reviewed section of the journal, the instructions in En-suring a Blind Review) have been followed.

Copyright notice: The submitting author acts on behalf of all authors and transfers to the CLINAM Foundation the right for publication in print and electronically of the submitted article in this journal and in the Foundation‘s derivative publications. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as this can lead to productive exchanges, as well as to earlier and greater citation of published work, as long as there is explicit reference to this journal as submitted to the „European Journal of Nanomedicine“ in printed in the “European Journal Nanomedicine“ or, after publica-tion, full biobliographic reference to this article. Otherwise, all rights of the work remains with the authors. Each article mirrors the opinion of the author and not necessarily of the journal. Publica-tion does not imply a recommendation by the journal for clinical use of tools, methods or drugs mentioned in the journal.

Privacy statement: The names and email addresses entered in this journal site will be used exclusively by, and for the purpose of, the European Foundation of Clinical Nanomedicine and will not be made available for any other purpose or to any other party. Journal Sponsorship Publisher: European Foundation for Clinical Nanomedicine

printed by Linsenmann Gissler AG, Binningerstrasse 95, Postfach 944, 4123 Allschwil 1Fon +41 61 567 22 94,Fax +41 61 567 22 33 Mobil +41 79 774 99 81, www.linsenmann-gissler.ch, Price 24€, ISNN Number ISSN 1662-5986 (Print) ISSN 1662-596x (Online), DOI (of journal)

Coverpage:

Endosomal uptake of fluorescent polymer nanocarrier (Hunziker, P.)

DOI 10.3884/0002.2 ISSN 1662-5986 (Print)

ISSN 1662-596x (Online)




The CLINAM Foundation knows why they co-operate with Linsenmann Gissler AG. They know that all services from Layout, Printing, Consulting until Publishing includingAcquisition of Advertisement comes from one professional partner that is outmost skilled in all these fi elds.

Is this what you are looking for too?

Linsenmann Gissler AG Binningerstr. 95 | CH-4123 Allschwil

Fon +41 61 567 22 22 | Fax +41 61 567 22 33

www.linsenmann-gissler.ch

Date post:	18-Jun-2020
Category:	Documents
Upload:	others
View:	7 times
Download:	0 times

october 2009 • Volume 2 • issue 2 … october 2009 • Volume 2 • issue 2 doi 10.3884/0002.2...

Documents