Gap junctions and the propagation of cell survival and cell death signals

Promotors:

Prof. Dr. K. D’Herde Prof. Dr. P. Vandenabeele

Thesis submitted in partial fulfillment of the requirements for the degree of

“DOCTOR IN MEDICAL SCIENCES”

Cell death: initiation, propagation and clearance

Dmitri V. Krysko

Promotors:Prof. Dr. K. D’Herde

Prof. Dr. P. Vandenabeele

Thesis submitted in partial fulfilment of the requirements for the degree of

“DOCTOR IN MEDICAL SCIENCES”

Cell death: initiation,propagation and clearance

Dmitri V. Krysko

2006

Department of Anatomy, Embryology,Histology, Medical Physics and

Department for Molecular Biomedical Research,

Ghent University

Cover illustration: Scanning Electron Micrograph

Clearance of apoptotic L929 cells by a macrophage

To my wife and my parents

Academic year 2005-2006

Promotors:

Prof. Dr. Katharina D’Herde (Department of Anatomy, Embryology, Histology and Medical Physics, Faculty of Medicine and Health Sciences, Ghent University)

Prof. Dr. Peter Vandenabeele (Department for Molecular Biomedical Research, Faculty of Sciences, Ghent University)

Examination committee:

Chairman: Prof. J. Van de Voorde1

Reading committee: Prof. Z. Zakeri2

Prof. C. Cuvelier 3

Prof. W. Van den Broeck4

Dr. W. Declercq5

Other members: Prof. H. Thierens6

Prof. L. Leybaert1

Prof. R. Beyaert5

1. Department of Physiology and Pathophysiology, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium

2. Department of Cell Biology, Queens College and Graduate Center of the City University of New York, New York, USA

3. Department of Pathology, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium

4. Department of Morphology, Faculty of Veterinary Medicine, Ghent University, Ghent, Belgium

5. Department for Molecular Biomedical Research, Faculty of Sciences, Ghent University, Ghent, Belgium

6. Department of Anatomy, Embryology, Histology and Medical Physics, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium

The research described in this thesis was performed:

At the Department of Anatomy, Embryology, Histology and Medical Physics, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium

At the Department for Molecular Biomedical Research, Faculty of Sciences, Ghent University and Flemish Institute for Biotechnology, Ghent, Belgium

i

Table of contents

Acknowledgements ..............................................................................................iv

List of Abbreviations ...........................................................................................vii

Summary..................................................................................................................1

Samenvatting ..........................................................................................................3

Résumé ....................................................................................................................5

I. Introduction.......................................................................................................7

1. Types of cell death at a glance ......................................................................7

1.1 Historical overview of cell death concept ......................................................7

1.2 Apoptotic cell death ......................................................................................7

1.2.1 Extrinsic pathway of caspase activation .............................................8

1.2.2 Intrinsic pathway of caspase activation...............................................9

1.2.3 Other organelles than mitochondria in initiation of apoptosis ...........10

1.3 Autophagy as a strategy for cell survival or for cell death?........................... 12

1.4 Necrotic cell death ......................................................................................15

1.5 Cell death as a consequence of aberrant mitosis ........................................17

1.6 Cell senescence ..........................................................................................18

2. Gap junctions and the propagation of cell survival and cell

death signals ...................................................................................................27

2.1 Abstract ......................................................................................................27

2.2 Introduction ................................................................................................27

2.3 Gap junction channels a conduit for spreading cell fate

modulating signals .............................................................................................. 28

2.4 Hemichannels in cell survival and cell death ...............................................33

2.5 Connexins outside gap junction channels and hemichannels also

affect cell survival and cell death ................................................................33

ii Table of contents

2.6 Conclusions and perspectives ....................................................................34

3. Clearance of apoptotic and necrotic cells and its

immunological consequence .......................................................................38

3.1 Abstract ......................................................................................................38

3.2 Introduction ................................................................................................38

3.3 Phagocytosis of dying cells: a journey through evolution ...........................39

3.4 Molecules involved in recognition and clearance of apoptotic

versus necrotic cells....................................................................................41

3.5 Phosphatidylserine receptor: unresolved controversies .............................44

3.6 Interactions of macrophages with apoptotic and necrotic cells:

functional consequences for immunity ........................................................46

3.6.1 Interaction with apoptotic cells and secondary necrotic cells ...........46

3.6.2 Interaction with necrotic cells ...........................................................48

3.6.3 Clearance of apoptotic and necrotic cells promotes cell life,

cell growth and cell death .................................................................50

3.6.4 Phagocytosis in anti-tumour defence ................................................52

3.7 Impairment of phagocytosis of dead cells and its role in the

development of diseases .............................................................................54

II. Aim of the study............................................................................................67

III. Results

1 Intracellular level

1.1 Mitochondrial transmembrane potential changes support the concept

of mitochondrial heterogeneity during apoptosis.........................................70

2 Intercellular level

2.1 Gap junctional communication and connexin43 expression in relation

to apoptotic cell death and survival of granulosa cells ................................78

iii

3 Burial level

3.1 Phagocytosis of necrotic cells by macrophages is

phosphatidylserine dependent and does not induce inflammatory

cytokine production.....................................................................................87

3.2 Analysis of NF- B activation in macrophages upon their co-culture

with apoptotic, primary and secondary necrotic cells ..................................99

3.3 Mechanisms of internalization of apoptotic and necrotic L929 cells

by a macrophage cell line studied by electron microscopy ........................105

3.4 Macrophages use different internalization mechanisms to clear

apoptotic and necrotic cells.......................................................................115

IV. General discussion and perspectives ..............................................127

V. Addenda

Addendum I: Transcriptional profiling of macrophages co-cultured with

apoptotic and necrotic cells...............................................................................141

Addendum II: Additional publication ..............................................................171

Addendum III: Curriculum vitae .....................................................................197

iv

Acknowledgements

Tempus fugit. The thesis you have in your hands is the result of four and half years of work. I

am a bit excited while writing these acknowledgements, not only for my research, but also for

the wonderful time that I have spent in Belgium, in Ghent, and especially in the Department

of Human Anatomy, Embryology and the Department for Molecular Biomedical Research.

During all these years, many people have helped and supported me. Without them, this

thesis would not have been completed. I am very glad that I now have the opportunity to

express my gratitude to all of them.

First and foremost, I would like to express my deepest gratitude to my promotor, Professor

Katharina D’Herde. Professor D’Herde, I am extremely grateful to you for your confidence in

me from the start and for the exceptional opportunities you have given me. Thank you for

your unwavering support, constant interest, encouragement and advice, from which I will

continue to benefit in my life.

I was very lucky that Professor Peter Vandenabeele, my second promotor, offered me the

opportunity to continue the second half of my research work in his Molecular Signalling and

Cell Death Unit of the Flanders Interuniversity Institute for Biotechnology. He has provided

me with a motivating, enthusiastic and critical atmosphere, which has made a deep

impression on me. Peter, it was both a pleasure and an educational experience working with

you. I also want to thank both promoters for all the valuable comments during the preparation

of the manuscript of this thesis.

I would like to address my sincere gratitude to the former Head of the Anatomy and

Embryology Section, Prof. Em. F. Roels, for his continuous support, especially at the

beginning of my research work, for his advice and comments along the way and for the

possibility to start to work in his department.

I am thankful to Prof. H. Thierens, Head of the Department of Human Anatomy, Embryology,

Histology and Medical Physics and Prof. Em. L. de Ridder for their support.

I would like to express my special acknowledgements to the reading committee, Prof. Z.

Zakeri, Prof. C. Cuvelier, Prof. W. Van den Broeck and Dr. W. Declercq, for taking the effort

to read my thesis and offering detailed evaluation on an earlier version. I would also like to

express my thanks to the all members of the examination jury, Prof. H. Thierens, Prof. L.

Leybaert, Prof. R. Beyaert and Prof. J. Van de Voorde, for their interest in my work. I am

particularly grateful to Prof. L. Leybaert for his collaboration in this work, for all the

suggestions and his willingness to participate as a jury member.

v Acknowledgements

I owe Prof. J. Grooten and T. Boonefaes from the Molecular Immunology Unit of the

Department for Molecular Biomedical Research of the Flanders Interuniversity Institute for

Biotechnology, for performing the macroarray investigation of macrophages. I especially

would like to thank Tom for sharing with me his experience and help. The incorporation of

these analyses provided an important extra value to this work and opened new interesting

perspectives for the future work.

I would like to acknowledge Dr. Sylvie Mussche for introducing me to granulosa cell isolation

techniques.

I want to thank my colleagues and ex-colleagues of the Department of Human Anatomy,

Embryology, Histology and Medical Physics for their friendly, practical and professional

support. In particular I would like to express my gratitude to Prof. G. Criel and B. De Prest for

their generous hospitality and for helping me to adapt living in Belgium, and to Prof. M.

Espeel, Dr. E. Barbaix and Prof. I. Kerckaert for their interest in my work. Many thanks also

to Prof. M. Cornelissen, for the advice she has offered. I am especially thankful to B. De

Bondt, D. Jacobus, H. Stevens, S. Van Hulle, G. Van Limbergen and A. De Smet for their

excellent technical assistance and to E. Roosen for her perfect and precise secretarial work.

I am thankful to T. Van Hoof, C. Pouders and M. Stevens for their company.

I am very grateful to have had many colleagues who have provided me with a warm and

friendly environment in the Department for Molecular Biomedical Research and in particular

in the Molecular Signalling and Cell Death Unit during my stay. I especially owe thanks to Dr.

Geertrui Denecker, Nele Festjens, Dr. Tom Vanden Berghe for their generosity, faith and

superb guidance and to Dr. Wim Declercq for always being ready to offer any kind of advice.

I am grateful to Ann Meeus and Wilma Burm for always being there to help me and for their

training in cell culture. I would like to extend my deepest gratitude to the following colleagues

and ex-colleagues who shared their experience with me: Dr. S. Cornelis, Dr. K. Dhondt, Dr.

X. Saelens, Dr. M. Lamkanfi, Dr. M. Kalai, Dr. V. Vandevoorde, K. Kersse, P. Ovaere, L.

Vande Walle, N. Vanlangenakker, E. Wirawan, E. Hoste, E. Parthoens, B. Gilbert, I.

Vanoverberghe. It has been a very pleasant experience to work with all of them.

Special acknowledgements go to Wim Drijvers for offering his unparalleled talents in making

the excellent drawing for this work.

I received extensive training as a medical student in the laboratory of Prof. Dr. Kevin Behrns

at the University of North Carolina in the USA. The work with Prof. Behrns was an important

bifurcation point in my career from which I have become more and more interested in the

apoptosis field.

vi

Während meine Leben war es ein großes Glück für mich, einen Teil meiner Zeit mit vielen

guten Freunden in Belgien und Deutschland verbringen zu dürfen. Besonders erwähnen

möchte ich Herr Kurt Hipp, Dr. Stephan Werle sowie Monika und Alfred Ullrich aus

Deutschland. Es heißt, Freundschaften sind der größte Schatz im Leben. Ihr Vertrauen und

ihre Fürsorge haben mir geholfen, die schwersten Momente während meine Leben

durchzustehen. Ich werde ihre Freundschaft für den Rest meines Lebens wertschätzen und

meine Erinnerungen an sie mit mir nehmen wo auch immer ich hingehe. Gerne erinnere ich

mich an die warmherzige Freundschaft zu meinen zuverlässigen Universitätsfreunden Mitya

Dovjanski und Alexander Gamaunov, deren Sympathie stets bei mir ist und sein wird und die

keine Grenzen kennt.

I would also like to take this opportunity to thank my mother and father, who not only

motivated me but continue to provide me with the standard by which to judge the

effectiveness of my life. I would like to extend my deepest gratitude to my father, who has

been teaching me throughout my life about the art of research, day after day, and who has

always been an example for me. My mother deserves high praise for the love I always feel,

despite the distance between us. My greatest pleasure is to thank my brother Anton and my

cousin Tatyana and their families for their love and encouragement.

Most of all, my deepest gratitude is due to Olga, my helpmate, playmate, soulmate, wife and

best friend. Her love, beauty, tenderness and charm are at the heart of my life.

Dmitri V. Krysko

Ghent

5 July 2006

vii

List of Abbreviations

ABC1 ATP-binding cassette transporter

AGA -glycyrrhetinic acid AIF Apoptosis-inducing factor Apaf-1 Apoptotic protease-activating factor-1 Apg Autophagy-defective ATP Adenosine triphosphate Aut Autophagocytosis Bax Bcl-2-associated X protein Bcl-2 B-cell lymphoma gene 2 BH3 Bcl-2 homology domain 3 BHA Butylated hydroxyanisole C.elegans Caenorhabditis elegans C1q Complement protein C1q caspase Cysteinyl aspartate-specific protease cDNA Complementary deoxyribonucleic acid Ced Cell death defective- CORD Chronic obstructive respiratory disease CRP C-reactive protein crq Croquemort Cvt Cytoplasm-to-vacuole targeting DAPk Death-associated protein kinase DC Dendritic cells DD Death domain DED Death effector domain DIABLO Direct inhibitor of apoptosis (IAP)-binding protein with low pI DISC Death-inducing signalling complex DMEM Dulbecco’s minimal essential medium DNA Deoxyribonucleic acid DR Death receptor EndoG Endonuclease G ER Endoplasmic reticulum ERK Extracellular signal-regulated kinase FACS Fluorescence-activated cell sorter FADD Fas receptor associated death domain FasL Fas ligand FKBP FK506-binding protein FRAP Fluorescence recovery after photo bleaching Gas6 Growth arrest-specific 6 G–CSF Granulocyte-colony stimulating factor GFP Green fluorescent protein HGF Hepatocyte growth factor HMGB-1 High mobility group box 1 protein HSP Heat shock protein IFN Interferon IL Interleukin iNOS inducible nitric-oxide synthase

I B Inhibitor of the enhancer in B cells

viii Abbreviations

LPS Lipopolysaccharide M0 Naïve macrophages M1 Classically activated macrophages M2 Alternatively activated macrophages MAPK Mitogen activated protein kinase MBL Mannose-binding lectin MEF Mouse embryonic fibroblasts Mf4/4 Macrophage cell line MFG-E8 Milk-fat globule epidermal growth factor-8 MIP-2 Macrophage-inflammatory protein 2 (MIP-2) mRNA messenger RNA

NF- B Nuclear factor-kappa B NO Nitric oxide PAF Platelet activating factor PBS Phosphate buffer saline PCD Programmed cell death PI Propidium iodide PI3K Phosphatidylinositol 3-kinase PLA2 Phospholipases A2 PS Phosphatidylserine PSR Phosphatidylserine receptor PTX3 Pentraxin-3 PUMA p53-upgelated modulator of apoptosis RIP Receptor Interacting Protein RNA ribonucleic acid ROCK Rho-associated coiled-coil–containing protein kinase ROS Reactive oxygen species

SIRP Signal regulatory protein SLE Systemic lupus erythematosus SMAC Second mitochondria-derived activator of caspases SREC Scavenger receptor found on endothelial cells tBid truncated Bid, C-terminal Bid fragment

TGF- 1 Transforming growth factor- 1TLR Toll like receptor TNFR1 Tumour necrosis factor-receptor 1

TNF Tumour necrosis factor TRAIL Tumour necrosis factor-related apoptosis-inducing ligand TUNEL Terminal deoxynucleotidyl transferase-mediated dUTP nick end labelling VEGF Vascular endothelial growth factor zVAD-fmk Benzyloxycarbonyl-Val-Ala-Asp-(OMe)-fluoromethylketone

m Mitochondrial membrane potential

1

Summary

Programmed cell death is an instrumental event during embryonic development which is

required for tissue homeostasis during adult life and plays a role in the pathophysiology of

many diseases. The most studied type of programmed cell death is apoptosis, characterized

by a peculiar pattern of chromatin condensation, DNA and protein cleavage and the

formation of membrane-enclosed apoptotic bodies. In mammalian cells, the apoptotic

response is mediated through either intrinsic or extrinsic cell death pathways, depending on

the origin of the death stimuli, and is always caspase-dependent. Mitochondrial dysfunction

revealed by release of mitochondrial intermembrane space proteins has been shown to be a

central coordinating event in the apoptotic pathway. Since apoptosis is an energy-dependent

process and one of the functions of mitochondria is to supply the cell with ATP, in the first

part of this study we investigated whether the response of mitochondria to apoptosis

induction is homogenous with regard to the loss of mitochondrial membrane potential. Our

approach with the use of an aldehyde-fixable fluorescent probe revealed in the granulosa

explant system that, after gonadotropin withdrawal, a subset of polarized mitochondria

exhibit normal mitochondrial membrane potential until the late stages of the apoptotic

process, when nuclear condensation is apparent, giving an additional argument in favour of

the mitochondrial heterogeneity concept.

The sequence of events involved in tissue homeostasis encompasses mechanisms within

single cells as well as interactions between cells within a population. Gap junctional

communication is thought to play a crucial role in these interactions by contributing to the

ability of cells to share cytoplasmic components that could modulate the fate of a cell’s life

and death. Although the molecular (intracellular) mechanisms leading to apoptotic cell death

were intensively unravelled during the last years, little is known about the role of intercellular

communication in the regulation of apoptotic cell death and survival. Therefore, in the second

part of our study, using a combined approach of FRAP analysis, connexin43

immunocytochemistry, experimental inhibition of gap junctional communication and apoptotic

index measurements, we demonstrated the involvement of gap junctional communication in

propagating a cell death message to neighbouring cells, while connexinx43, a major gap

junctional protein, is needed for survival independently of gap junctional communication.

The third part of this thesis is related to the end stage of apoptosis or the “burial” stage.

During this stage the death corpses are recognized and engulfed by either professional

phagocytes or neighbouring cells. While great progress has been achieved in the

understanding of molecular mechanisms of apoptotic cell engulfment, little is known about

the interactions of necrotic cells with macrophages. In order to study the interaction of

2 Summary

macrophages with apoptotic versus necrotic cells, an in vitro phagocytosis assay was used in

which a mouse macrophage cell line (Mf4/4) was cocultured with murine L929sAhFas cells

that were induced to die either in a necrotic way by TNFR1 or in an apoptotic way by Fas

stimulation. In order to have a quick and reliable method to quantify the uptake of dead cells,

we developed a dual color flow cytometric assay and found that shielding the exposed

phosphatidylserine on the target cells with recombinant annexin V influenced not only

apoptotic clearance but also necrotic disposal, suggesting that in both cases recognition

and/or uptake is phosphatidylserine-dependent. Importantly, necrotic cells neither induce

production of pro-inflammatory cytokines on mRNA and protein levels nor stimulate

activation of NF- B transcription factor in macrophages.

Finally, by use of two fluid phase tracers and ultrastructural investigation we showed that

macrophages employ morphologically different internalization mechanisms for apoptotic and

necrotic cells: the zipper-like mechanism versus the less selective macropinocytotic

mechanism. That different biochemical processes are associated with the two

ultrastructurally different ways of internalizing dead cells is suggested by the effect of

wortmannin, an inhibitor of phosphatidylinositol 3’-kinase, on the clearance of apoptotic and

necrotic cells. Wortmannin reduced the uptake of apoptotic cells, but the engulfment of

necrotic cells remained unaffected.

Insight into the mechanisms controlling cell death on intracellular and intercellular levels, as

well as understanding the molecular mechanisms of dead cells disposal, can contribute to

the development of strategies to manipulate and regulate cell death processes on multiple

levels: at the intracellular, intercellular and the clearance phase, and this may lead to the

discovery of new medical treatments for inflammatory and autoimmune diseases and, on the

other hand, the improvement of vaccine strategies against cancer and infection.

3

Samenvatting

Geprogrammeerde celdood is een onmisbaar proces tijdens de embryonale ontwikkeling, is

noodzakelijk voor de weefselhomeostase en speelt een rol in de pathophysiologie van vele

ziekten. Het meest bestudeerde type van geprogrameeerde celdood is apoptosis,

gekenmerkt door een karakteristiek patroon van chromatine condensatie, DNA fragmentatie,

specifieke proteineverknipping en daarnaast de vorming van apoptotische lichaampjes met

een intacte plasmamembraan. In zoogdierencellen wordt de apoptotische repons afhankelijk

van de oorsprong van het doodsignaal bewerkstelligd door een intrinsieke of een extrinsieke

signaaltransductieweg en is altijd caspase-afhankelijk. Het is aangetoond dat mitochondriale

dysfunctie gepaard met vrijstelling van de mitochondriale eiwitten uit de intermembranaire

ruimte een centrale en coördinerende rol speelt in de apotische cascade. Aangezien het

apoptose proces op zich een energie-afhankelijk proces is enerzijds, en de mitochondriën via

ATP productie de cel in aerobe omstandigheden voorzien van energie anderzijds,

onderzochten we in een eerste deel of de mitochondriale respons op apoptotische inductie

met betrekking tot daling van mictochondriale potentiaal homogeen is. Onze benadering met

gebruik van een aldehyde-fixeerbare fluorescente probe toonde in het granulosa explant

model aan dat na gonadotropine derving een subpopulatie van mitochondriën een normale

mitochondriale membraan potentiaal behoudt tot in de late stadia van het apoptose process,

wanneer de chromatine-condensatie duidelijk is. Deze bevinding is een bijkomend argument

dat het concept over de mitochondriale heterogeniteit ondersteunt.

De sequentie van gebeurtenissen betrokken bij weelselhomeostase spelen zich enerzijds af

op intracellulair niveau in individuele cellen, maar liggen ook op intercellulair niveau met

determinerende interacties van cellen binnen een populatie. Gap verbindigen (juncties) zijn

cruciaal bij intercellulaire communicatie en maakt het mogelijk dat cytoplasmatische signalen

die het lot van cellen kunnen beïnvloeden (overleving of celdood) kunnen doorgegeven

worden tussen naburige cellen. Ondanks het feit dat de laatste jaren belangrijke inzichten

werden verworven in de moleculaire (intracellulaire) mechanismen die leiden naar

apoptotische celdood, is nog weinig gekend over de rol van intercellulaire communicatie in

de regulatie van celdood en overleving. Daarom hebben we in een tweede luik van de thesis

via een gecombineerde benadering van FRAP analyse, connexine43 immuuncytochemie,

experimentele inhibitie van gap communicatie en apoptische index metingen aangetoond dat

gap-gemedieerde communicatie betrokken is in de verspreiding van een celdoodsignaal naar

naburige cellen, terwijl connexine43 per se een belangrijk gap verbindingseiwit nodig is voor

overleving.

4 Samenvatting

Het derde luik van de thesis heeft betrekking tot het eindstadium van apotosis of de

“opruimingsfase” wanneer de apototische lichaampjes worden gefagocyteerd. Bij deze

opruimingsfase kunnen zowel professionele phagocyten als naburige gezonde cellen worden

ingezet en dient een herkenningsfase en een opname fase onderscheiden te worden. Daar

waar een grote vooruitgang geboekt is met betrekking tot de inzichten in de moleculaire

mechanismen van apoptische cel opruiming is weinig gekend over de interacties van

necrotische cellen en fagocyterende cellen. Om deze interactie van macrofagen en

apopotische of necrotische cellen te bestuderen werd een in vitro fagocytosis test gebruikt

waarin een muizen macrofaag cellijn (Mf4/4) in cocultuur werd gebracht met muizen

L929sAhFas cellen die necrotisch sterven via TNFR1 inductie en apoptotisch via Fas

stimulatie. Om snel en betrouwbaar de opname van dode cellen te kwantificeren werd een

flow cytometrische assay op punt gesteld. Hiermee toonden we aantonen dat het

afschermen van fosphatidylserine op de buitenzijde van de plasmamembraan met behulp

van recombinant annexin V de opname van zowel apoptotische als ook necrotische target

cellen negatief beinvloedt. Deze waarneming suggereert dat in beide gevallen de

herkenningsfase en/of opnamefase phosphatidylserine-afhankelijk is. Bij de interactie tussen

necrotische cellen en macrofagen is er geen inductie van pro-inflammatoire cytokines en dit

noch op mRNA niveau (RNA protectie assay) als op proteine niveau (IL-1, IL-6, TNF), tevens

is er geen stimulatie van NF- B transcriptie factor in de macrofagen.

Tenslotte hebben we door gebruik van twee vloeistof-fase merkers voor pinocytose en

ulstrastructureel onderzoek aangetoond dat macrofagen twee morfologisch verschillende

internalisatie mechanismen gebruiken om apoptotische en necrotische cellen op te ruimen

namelijk het “zipper”-like mechanisme en het minder selectieve macropinocytosis

mechanisme. Dat verschillende biochemische processen geassocieerd zijn met deze

ultrastructureel verschillende wijze van opname wordt bevestigd door het effect van

wortmannin, een inhibitor van phosphatidylinositol 3’-kinase op de opruiming van apoptotisch

en necrotisch stervende cellen. Inderdaad wortmannin halveert de opname van apoptotische

cellen, maar heeft bijna geen effect op de opname van necrotische cellen.

Inzichten in de mechansimen die celdood reguleren op intracellulair en intercellulair niveau

alsook de moleculaire mechanismen betrokken bij celopruiming kunnen bijdragen tot het

ontwikkelen van strategieën om celdood te reguleren en te manipuleren op de verschillende

niveaus. Hieruit kunnen potentieel nieuwe behandelingen voor inflammatoire en

autoimmuunaandoeningen ontwikkeld worden naast het verbeteren van vaccinatie

strategieën tegen kanker en infecties.

5

Résumé

La mort cellulaire programmée est indispensable au cours del’embryogenèse, elle est

également nécessaire afin de maintenir un contrôle de l’homéostasie tissulaire et joue un

rôle dans la pathophysiologie de diverses maladies. Le type de mort cellulaire programmée

le plus étudié est l’apoptose, caracterisée par une condensation de la chromatine, une

fragmentation de l’ADN, une fragmentation de certaines protéines et la formation de corps

apoptotiques avec membrane plasmique intacte. Dans les cellules de mammifères la

réponse apoptotique est principalement médiée par voie extrinsique (la voie des récepteurs

à domaine de mort) ou par voie intrinsique (la voie mitochondriale), selon l’origine de

l’endommagement et implique toujours les caspases. La dysfonction mitochondriale, révélée

par libération des protéines intermembranaires, a été démontrée comme évènement central

et coordinateur. Comme l’apoptose est un processus nécessitant de l’énergie et qu’une des

fonctions des mitochondries est de pourvoir la cellule en ATP, nous avons analysé dans une

première partie si la réponse à l’induction de l’apoptose se fait de façon homogène en ce qui

conçerne la dégradation du potentiel membranaire mitochondrial. Notre étude, avec une

sonde mitochondriale fluorescente permettant la fixation à l’aldéhyde, nous a révélé que

dans le modèle des explants de granulosa privés de gonadotropines une sous-population

des mitochondries conserve un potentiel membranaire normal jusqu’à la fin du processus,

c’est-à-dire. lorsque la condensation nucléaire apparait. Cette observation nous procure un

argument de plus pour proposer le concept de l’hétérogénité mitochondriale.

La séquence des évènements déterminants pour l’homéostasie tissulaire surpasse les

mécanismes qui se déroulent au niveau des cellules individuelles, mais implique aussi les

interactions entre les cellules d’une population. La communication par jonction gap ou nexus

a un rôle crucial dans ces interactions, car les jonctions gap fournissent une possibilité pour

des cellules adjacentes d' échanger des molécules qui pourraient modifier le sort d’une

cellule, soit la mort soit la vie. Bien que les voies moléculaires (intracellulaires) qui

aboutissent à l’apoptose aientété largement découvertes ces dernières années, peu

d’informations sont connues sur le rôle de la communication intercellulaire dans la régulation

de l’apoptose ou de la survie. Pour cette raison nous avons combiné, dans une seconde

partie de notre étude, l’analyse fonctionelle des jonctions gap (dite analyse de FRAP), avec

l’immunocytochimie de connexin43, l’inhibiton expérimentale de la communication par

jonction gap et des déterminations de l’index apoptotique. Ainsi nous avons pu démontrer

l’intervention des jonctions gap dans la propagation d’un message de mort cellulaire aux

cellules avoisinantes et le rôle independant de la connexin43 dans la survie.

6 Résumé

La troisième partie de la thèse est liée à la phase finale de l’apoptose, dite la phase

d’enterrement. Dans cette phase les corps apoptotiques sont reconnus et envelopés par des

macrophages ou des cellules environnantes. Bien qu' un grand progrès ait été réalisé dans

le domaine des mécanismes moléculaires de l'incorporation de corps apoptotiques, peu de

donnés sont connues sur les rapports entre les cellules nécrotiques et les macrophages.

Afin d’étudier l’interaction entre les macrophages et les cellules mourantes de façon

apoptotique ou nécrotique, un modèle in vitro a eté utilisé dans lequel une lignée cellulaire

de macrophage a été cocultivée avec des cellules murines L929sAhFas qui sont stimulées

pour mourir de façon apoptotique ou nécrotique en utilisant soit le ligand Fas ou le TNFR1.

Pour pouvoir quantifier de façon rapide et fiable l’incorporation de cellules mortes nous

avons développé une technique de cytométrie de flux avec deux couleurs fluorescentes.

Cette technique nous a permis de démontrer que le blockage de la phosphatidylserine,

exposée à la face extérieure de la membrane plasmique en utilisant l’annexin V

recombinante, a un effet inhibiteur sur l’incorporation des cellules mortes par les

macrophages et ceci aussi bien pour les corps apoptotiques que pour les cellules

nécrotiques. Dans ce contexte, il est important de préciser que les cellules nécrotiques ne

stimulent ni les macrophages à produire des cytokines pro-inflammatoires au niveau de

l’ARNm ou au niveau de la traduction, ni la production du facteur de transcription NF- B.

Finalement, utilisant deux traceurs “fluid-phase” différents et des techniques de microscopie

électronique, nous avons pu démontrer que les macrophages utilisent deux mécanismes

d’incorporation pour éliminer des corps apoptotiques et des cellules nécrotiques, c’est à dire,

un mécanisme “zipperlike” (la phagocytose classique) et un mécanisme moins sélectif de

macropinocytose. La divergence morphologique associé à ces deux mécanimespeut

s’accompagnés de différents processus biochimiques est suggéréepar le fait que la

wormannin, un inhibituer de la phosphatidylinositol 3’-kinase, influence l’incorporation des

corps apoptotiques, mais pas celle de cellules nécrotiques.

La compréhension des voies qui contrôlent la mort cellulaire au niveau intracellulaire et

intercellulaire, ainsi que les mécanimes moléculaires liés à l’incorporation des cellules

mortes peut contribuer au développement de stratégies pour manipuler la mort cellulaire aux

differents niveaux stipulés auparavant. Ceci peut également mener à découvrir de nouveaux

traitements médicaux pour les maladies imflammatoires et autoimmunes et peut améliorer

les stratégies de vaccination contre le cancer et l’ infection.

Introduction

BrutusThen follow me, and give me audience, friends.Cassius, go you into the other street,And part the numbers.Those that will hear me speak, let 'em stay here;Those that will follow Cassius, go with him;And public reasons shall be renderedOf Caesar's death.First Citizen I will hear Brutus speak.Second Citizen I will hear Cassius; and compare their reasons,When severally we hear them rendered.

SCENE II. The ForumJulius Caesar the play by W Shakespear

The Secret, A Rodin

1. Types of cell death at a glance 2. Gap junctions and the propagation of

cell survival and cell death signals 3. Clearance of apoptotic and necrotic

cells

1. Types of cell death at a glance

DV Krysko, P Vandenabeele and K D’Herde In New Cell Apoptosis Research. Edited by F Columbus, Nova Science Publishers, Inc. Accepted for publication in the preliminary form.

7

1: Types of cell death at a glance

1.1. Historical overview of the cell death concept

Our life begins from a single cell that needs to divide and take in specific fates to form a

complex organism. It is almost counterintuitive that millions of cells would die during

development and life. However, cell death as an important aspect of the normal sculpting of

multicellular organisms has been recognized to some degree since antiquity. The fact that

some structures were transitory and eventually committed to disappear was known since

Galen of Pergamum (ca 130 – 200) observed the transitional state of the fetal arterial duct.

The disappearance of this structure at birth is a necessary physiological event. Obviously,

Galen did not directly address cell death and the first description of cell death was introduced

soon after the establishment of the cell theory (Schleiden and Schwann) by Carl Vogt in

1842, who was working on amphibian metamorphosis (Clarke and Clarke, 1995). Many other

scientists also described naturally occurring cell death over the next century (reviewed in

Clarke and Clarke, 1996), but most cell biologists were more interested in understanding the

life of a cell, not its death. The concept of programmed cell death (PCD) was introduced in

1965 by Lockshin and Williams (Lockshin and Williams, 1965) that occurred in predictable

places and at predictable times during embryogenesis, pointing to the fact that cells are

somehow programmed to die during the development of the organism. Later on Kerr and co-

authors (Kerr et al., 1972) described the morphological characteristics of cell death during

development and tissue homeostasis and coined the term apoptosis (derived from the Greek

word meaning ‘falling off”, as of leaves from a tree) to distinguish this type of cell death from

necrosis, viz. cell death as a consequence of physico-chemical insult. Thereafter, three

morphologically distinct types of physiological cell death in the tissue of mouse and rat

embryos using electron microscopy were identified (Schweichel and Merker, 1973): type I

(apoptotic cell death), type II (autophagic cell death), and type III (necrotic cell death). In the

following sections we will shortly discuss these and some other types of cell death.

1.2. Apoptotic cell death

Apoptosis or type I cell death occurs through a sequence of specific morphological changes

in the dying cell: condensation of the cytoplasm and margination of the nuclear chromatin

into one or several large masses with the subsequent formation of membrane-bounded

apoptotic bodies containing a variety of cytoplasmic organelles and nuclear fragments which

are engulfed by neighboring cells and by macrophages (Kerr et al., 1972; Schweichel and

Merker, 1973). There are two main pathways which can lead to apoptosis. In mammalian

8 Introduction: Part I

cells, the apoptotic response is mediated through either the intrinsic or the extrinsic pathway,

depending on the origin of the death stimuli. Both pathways finally converge on the

proteolytic activation of downstream effector caspases.

1.2.1. Extrinsic pathway of caspase activation

The extrinsic pathway is triggered at the cell surface by the binding of the extracellular death

ligand, such as FasL, tumor necrosis factor (TNF ), Apo3-ligand, or TRAIL (TNF-related

apoptosis inducing ligand/Apo-2 ligand), to its cell-surface death receptor, such as Fas and

tumor necrosis factor-receptor 1, death receptor (DR) 3 (APO-3/TRAMP), DR 4 (TRAIL-R1),

DR 5 (TRAIL-R2/TRICK2) and DR 6 (Chen and Goeddel, 2002; Nagata, 1999; Sheikh and

Fornace, 2000). Death receptors typically consist of an extracellular region containing

varying numbers of cystein-rich domains required for ligand binding, and an intracellular

region with a death domain (DD) motif, allowing homotypic protein-protein interactions. A

typical example for the extrinsic pathway is Fas-induced apoptosis (Fig. 1), where, following

the Fas ligand or agonistic antibodies binding to the homotrimeric Fas receptor, an apoptotic

signal is transduced by the death domain through homotypic interactions that recruit adaptor

molecules also containing a DD, such as FADD (Aravind et al., 1999; Hofmann, 1999). The

adaptor molecule FADD additionally contains a death effector domain (DED), which in its

turn allows the homotypic recruitment of apoptotic initiator caspases containing DED, such

as procaspase-8 in mice and procaspase-8 and -10 in humans (Muzio et al., 1996). These

homotypic interactions lead to the formation of an oligomeric death-inducing signaling

complex (DISC) (Kischkel et al., 1995; Peter and Krammer, 2003). This DISC forms a kind of

molecular platform in which the initiator caspase-8 is conformationally activated, which then

leads to autoproteolysis and proteolytic activation of the downstream effector caspases such

as caspase-3 and caspase-7. The extrinsic pathway can crosstalk with the intrinsic pathway

through caspase-8-mediated cleavage of Bid (a BH3-only member of the Bcl-2 family

proteins) (Luo et al., 1998; Yang et al., 1998), which then triggers the release of

mitochondrial proteins (Festjens et al., 2004). Two types of the Fas receptor-induced

apoptotic signaling have been established. Type I cells are characterized by high levels of

DISC formation and increased amounts of active caspase-8. In this case, activated caspase-

8 directly leads to the activation of downstream effector caspases without the need of a

mitochondrial amplification loop. In type II cells, there are lower levels of DISC formation and,

thus, lower levels of active caspase-8 (Scaffidi et al., 1998). In this case, signaling is

dependent on the mitochondrial amplification loop that involves the caspase-8-mediated

cleavage of Bid to generating truncated (t) Bid.

Types of cell death 9

Figure 1. An overview of the extrinsic and intrinsic pathways. Both pathways lead to activation of caspase-3 and give rise to apoptotic cell death. As an example of the extrinsic pathway, Fas-mediated signaling is shown. The transcriptional and cytoplasmic functions of p53 are depicted in the same pathway where p53 activates the expression of PUMA, which then serves to release cytoplasmic p53 from the inhibitory interaction with Bcl-xL. The released p53 is then free to directly activate Bax, leading to mitochondrial outer membrane permeabilization and apoptosis (Vousden, 2005).

This tBid induces the release of several mitochondrial factors such as cytochrome c, which

activates the mitochondrial apoptotic pathway (Korsmeyer et al., 2000). Type II induced

apoptosis is blocked by Bcl-2 overexpression, while type I is not.

1.2.2. Intrinsic pathway of caspase activation

The intrinsic pathway is triggered in response to a wide range of death stimuli that are

generated from within the cell, such as oncogene activation, DNA damage induced by

irradiation or chemotherapeutics and ER stress. The intrinsic pathway (Fig. 1) is mediated by

mitochondria (Wang, 2001) and it is regulated by Bcl-2 family of pro- and anti-apoptotic

Caspases-3/6/7

Caspase-9

Apaf-1

ApoptosomeCyt c

Bid

tBid

ActiveCaspase-8

Caspase-9

APOPTOSIS

Procaspase-8

FADD

FasL

Fas

Chemotherapy, radiation,growth factor withdrawal,

oxidative stress

SMAC

OmiAIP

ATP

Extrinsic pathway Intrinsic pathway

Type I

Type II

Caspases-3/7

p53

Nucelus

PUMABcl-xL

PUMA

p53

p53

Bax

Bcl-xL


proteins. Pro-apoptotic Bcl-2 proteins function to permeabilize the mitochondrial outer-

membrane, which is accompanied by release of several proteins from the intermembrane

space of mitochondria into the cytoplasm in response to apoptotic stimuli (Festjens et al.,

2004; Saelens et al., 2004; Wang, 2001). Recently, it has been shown that besides its role as

transcription factor in the nucleus, p53 also possesses an extranuclear function in that it

directly binds anti-apoptotic Bcl-2 proteins (Bcl-2 and Bcl-xL) and activates the pro-apoptotic

multi-domain Bcl-2 proteins (Bax and Bak) and thus regulates mitochondrial outer-membrane

permeabilization (Fig. 1) (Chipuk and Green, 2006; Chipuk et al., 2004). Some of the well-

characterized proteins which are released from mitochondria include cytochrome c, SMAC,

(second mitochondria-derived activator of caspases/DIABLO, a direct inhibitor of apoptosis

(IAP)-binding protein with low pI), AIF (apoptosis-inducing factor), EndoG (endonuclease G)

and OMI/HTRA2 (high-temperature-requirement protein A2) (van Loo et al., 2002). Probably

the most exciting one of these pro-apoptotic proteins is cytochrome c, which binds to and

activates the protein Apaf-1 in the cytoplasm (Li et al., 1997). Cytochrome c at its canonical

localization in the intermembrane space of mitochondria is involved in the transfer of

electrons between complex III and complex IV of the oxidative phosphorylation pathway,

which is essentially a survival function (production of ATP). Once released in the cytosol it

acquires a cell death function by its capacity to bind Apaf-1. This binding induces a

conformational change that allows Apaf-1 to bind to ATP/dATP and to form the apoptosome

complex (Jiang and Wang, 2000). This apoptosome functions as a platform that allows the

recruitment of caspase-9 through homotypic CARD-CARD interactions, inducing a

conformational change and activation of caspase-9 (Li et al., 1997; Rodriguez and Lazebnik,

1999; Saleh et al., 1999; Zou et al., 1999). Activated caspase-9 converges again on the

proteolytic activation of downstream effector caspases. Autoproteolysis of caspase-9 forms a

negative feedback loop resulting in the release and inactivation of caspase-9 from the

apoptosome platform (Twiddy et al., 2004; Twiddy et al., 2006). In addition, Lakhani et al.

(Lakhani et al., 2006), using caspase-3 and caspase-7 deficient embryonic fibroblasts,

provided evidence that these downstream caspases may amplify Bax translocation to

mitochondria as well as cytochrome c release in response to ultraviolet radiation. This

suggests that caspases-3 and -7 may participate in a feedback amplification loop to promote

mitochondrial cytochrome c release.

1.2.3. Other organelles than mitochondria in initiation of apoptosis

Accumulating evidence suggests that other organelles, including the endoplasmic reticulum

(ER), the Golgi apparatus and lysosomes, are also major points of integration of pro-

apoptotic and anti-apoptotic signaling or cellular damage sensing (Ferri and Kroemer, 2001).


The ER participates in the initiation of apoptosis by at least two different mechanisms,

namely, the unfolded protein response and Ca2+ signaling. Identification of caspase-12 on

the cytoplasmic side of the ER and demonstration that caspase-12 is processed in cells

treated with ER-stress agents has favored the idea that caspase-12 might be the initiator

caspase in ER-stress-mediated apoptosis (Nakagawa et al., 2000). However, later on it was

discovered that although caspase-12 is processed in ER-stress-mediated cell death in

B16/B16 melanoma cells, the cells die to the same extent in the absence of caspase-12

(Kalai et al., 2003). Similarly, it was shown that murine cells lacking caspase-12 expression

were not protected from apoptosis induced by ER stress agents (Di Sano et al., 2006; Obeng

and Boise, 2005). Besides, overexpression of the anti-apoptotic Bcl-2 family member, Bcl-xL

was capable of protecting cells from ER stress-induced cell death (Obeng and Boise, 2005).

Indeed, other ER-associated proapoptotic molecules have been reported, such as Bap31, a

polytopic integral protein of the ER membrane which can bind caspase-8 and Scotin, which

are implicated in p53-mediated apoptosis (Lamkanfi et al., 2004). However, the signal-

transducing events that connect ER stress to the cell death machinery are not clearly

understood and require further investigation.

Recent studies suggest that the Golgi complex can also sense and transduce apoptotic

signals. The discovery of a pool of caspase-2 localized to the cytoplasmic face of the Golgi

complex indicates that caspase-2 may play a key role in apoptotic signaling at the Golgi

complex. The golgin-160 can be cleaved by caspase-2, -3, and -7. The cleavage of golgin-

160 by caspase-2 occurs rapidly and precedes caspase-3 cleavage, indicating an early role

for caspase-2 activation at the Golgi complex (Mancini et al., 2000). Furthermore, HeLa cells

expressing a caspase-resistant mutant of golgin-160 were resistant to apoptosis induced by

ligation of death receptors and by drugs that induce ER stress, while sensitive to other

proapoptotic stimuli, including staurosporine, anisomycin, and etoposide (Maag et al., 2005).

These data indicate that some apoptotic signals may be sensed and integrated at Golgi

membranes. A common link between ligation of death receptors and the ER stress response

might be the Golgi complex.

Lysosomes, which have been given the epithet ‘suicide bags’ (De Duve, Nobel prize speech,

1974), contain a large number of catabolic hydrolases with an acidic pH optimum. They

contribute to type II (autophagic) cell death (discussed later). However, a number of

lysosomal proteases, cathepsins, have been implicated in the induction of apoptosis (Deiss

et al., 1996; Ferri and Kroemer, 2001).

In conclusion, each organelle may possess sensors that detect specific alterations, locally

activate signal transduction pathways and emit signals that ensure inter-organelle cross-talk

(Ferri and Kroemer, 2001).


1.3. Autophagy as a strategy for cell survival or for cell death?

In normal cells, there are two general mechanisms which are involved in the degradation and

recycling of the building blocks of organelles, proteins and other components of the

cytoplasm. In the large-scale degradation of components of the cytoplasm, short living

regulatory proteins are broken down to amino acids by the ubiquitin-proteasome system, and

long living structures and proteins are targeted to the lysosome for hydrolysis by autophagy.

Several forms of autophagy have been described (Baehrecke, 2005), but here we focus on

macroautophagy (hereafter referred to as autophagy) because of its association with type II

cell death. Autophagy involves the inducible sequestration of cytoplasm and organelles, or

organelles fragments, in a double-membrane structure, known as an autophagosome. The

autophagosome delivers its sequestered cargo to lysosomes by a fusion process that results

in autophagolysosomes. This compartment contains a range of hydrolases that are able to

degrade proteins, lipids, nucleic acids and carbohydrates that may lead to organelle

degradation (Klionsky and Emr, 2000). Several origins have been proposed for the wrapping

membrane structure. Among these are the ribosome-free regions of the rough endoplasmic

reticulum, the trans Golgi network and a membrane called the phagophore which is of

unknown origin (Petiot et al., 2002). The biogenesis and consumption of such vesicles has

been divided in four distinct steps: induction and cargo packaging, formation and completion,

docking and fusion, and breakdown.

Molecular mechanisms implicated in autophagy regulation in response to starvation were

initially discovered in the yeast Saccharomyces cerevisia. Screens for yeast mutants that

were starvation-sensitive or defective in the degradation of specific cytosolic proteins

produced Apg (autophagy-defective), Aut (autophagocytosis) and Cvt (cytoplasm-to-vacuole

targeting) mutants which partially overlap and which are collectively designated as Atg genes

(Baba et al., 1997; Harding et al., 1995; Thumm et al., 1994; Tsukada and Ohsumi, 1993).

The pro-survival function of autophagy is an evolutionary ancient process, conserved from

yeast to mammals. In response to starvation, C. elegans larvae enter dauer, a latent

developmental state, while inactivation with RNAi of autophagy genes (bec-1, atg8 and

atg18) disrupt normal dauer formation (Melendez et al., 2003). Moreover, mice that lack

Atg5, an acceptor molecule for the ubiquitin-like molecule Atg12, die during the neonatal

period, when the placental blood supply is interrupted and they undergo a form of starvation

(Kuma et al., 2004). Autophagy genes may also be critical for maintaining cellular survival

when cells are unable to take up external nutrients, e.g., during growth factor deprivation.

Growth factor withdrawal usually results in rapid apoptotic cell death, but recent studies in

apoptotic-deficient bax-/-, bak-/- cells unraveled an important role for autophagy genes in

maintaining cellular survival following IL-3 deprivation (Lum et al., 2005). IL-3-deprivied bax-/-,


bak-/- cells activate autophagy and ultimately succumb to death. Notably, at any point before

death, the addition of growth factor reverses the catabolic process and maintains cell viability

(Lum et al., 2005). Furthermore, RNAi against beclin1, atg5, atg10, and atg12 enhances

starvation-induced, but not staurosporine-induced, apoptotic cell death, which illustrates an

interesting mechanism by which autophagy genes promote survival during nutrient

deprivation by suppression of the apoptotic death pathway (Boya et al., 2005). Together,

these data suggest that autophagy is a critical survival mechanism to overcome stress

conditions.

Paradoxically, the presence of autophagic structures in dying cells may also indicate the

existence of type II autophagic cell death (Schweichel and Merker, 1973). Type II or

autophagic cell death is a type of cell death accompanied by massive autophagic

vacuolization of the cytoplasm (Kroemer et al., 2005; Okada and Mak, 2004). In autophagic

cell death, partial chromatin condensation (no DNA laddering) occurs late if at all (Kroemer et

al., 2005; Levine and Yuan, 2005). It is important to emphasize that an optimal identification

of autophagic cells death would require the use of transmission electron microscopy in order

to distinguish autophagosomes from other types of vesicles such as endosomes, lysosomes,

or macropinosomes. This type of cell death in most instances was found during the

destruction of organs and large cell units in mouse and rat embryos and fetuses (Schweichel

and Merker, 1973). Another example may be insect metamorphosis, where cell death is

typically autophagic, and blocking autophagy is pupariation lethal (Juhasz et al., 2003).

Autophagic cell death also has been reported to occur in adulthood of insects and

vertebrates, including humans; it is often associated with the elimination of large secretory

cells during the adjustment of sexual organs and ancillary tissues to seasonal reproduction

(Bursch et al., 2000).

However, it is important to stress that the issue of whether type II cell death represents a

separate form of cell death is currently sketchy (Lockshin and Zakeri, 2004), because in the

literature strong evidence exists in support of the notion that autophagic cell death may

overlap either with caspase-dependent or -independent cell death. In this regard, studies of

steroid-triggered salivary gland cell death during Drosophila development showed that

caspases function in the death of cells exhibiting an autophagic morphology. Caspase-3

activation and atg gene transcription immediately precede autophagic cell death in salivary

glands (Lee and Baehrecke, 2001; Lee et al., 2003; Martin and Baehrecke, 2004). Similarly,

it has been shown that the proapoptotic signaling molecule TNF-related apoptosis-inducing

ligand (TRAIL) regulates autophagy in an in vitro model of mammary gland formation (Mills et

al., 2004). In this study the increased level of TRAIL was detected during morphogenesis of

MCF-10A mammary epithelial cells in 3D basement-membrane cultures just prior to the

death of luminal cells that express active caspase-3, whereas these cells possess


autophagic cell death morphology. Additionally, it was shown that inactivation of lysosome-

associated membrane protein-2 (LAMP2) by RNAi or by homologous recombination in cells

cultured in nutrient-deficient conditions led to accumulation of autophagic vacuoles and then

succumbed to cell death with hallmarks of apoptosis with caspase activation and chromatin

condensation (Gonzalez-Polo et al., 2005). Another molecular link between autophagy and

apoptosis may the death-associated protein kinase (DAPk) and DAPk-related protein 1 which

regulate membrane blebbing during apoptosis, but can also promote autophagic vacuole

formation in dying cells (Inbal et al., 2002). Recent study suggests that Bcl-2 binds to Beclin

1 and disrupts its autophagic function, while in the absence of Bcl-2 binding, Beclin 1

mutants induce excessive autophagy and promote cell death (Pattingre et al., 2005). This

study demonstrates the Beclin 1-Bcl-2 complex may function as a rheostat that ensures that

autophagy levels remain within a homeostatic range rather than in a nonphysiological range

that triggers cell death. Taken together these studies indicate that complex interrelationships

may exist between autophagy and apoptotic cell death pathways.

In spite of the accumulating evidence for overlap with apoptosis it is of note that autophagy

may also be related with a caspase-independent cell death. In this line it was shown that cell

death induced by the human homologue of the Drosophila spin gene product (HSpin1)

(Gonzalez-Polo et al., 2005) is caspase-independent and autophagic (Yanagisawa et al.,

2003). In support of this idea are the experiments in which RNAi directed against two

autophagy genes, atg7 and beclin1, severely affected cell death in mouse L929 cells treated

with the caspase inhibitor zVAD-fmk (Yu et al., 2004). In this model, a new molecular

pathway was defined in which activation of RIP1 (a serine-threonine kinase) and Jun amino-

terminal kinase induced cell death with the morphology of autophagy. Inhibition of caspase-8

induces cell death with autophagic morphology (Yu et al., 2004). Similarly, the autophagic

cell death of Bax-/-Bak-/- mouse embryonic fibroblasts treated with either etoposide or

staurosporine was prevented by the knock-down of Beclin 1 and Atg5 (Shimizu et al., 2004),

suggesting that autophagic cell death occurs in a caspase-independent manner. In addition,

or alternatively, these data might be interpreted to mean that either apoptosis effectors such

as caspases, Bax or Bak actively suppress the autophagic components of cell death, or that

the autophagic and apoptotic effector mechanisms constitute backup mechanisms that come

into action when one or the other lethal pathway is inhibited.

The above-mentioned data indicate that autophagy in first instance acts fundamentally as a

cell survival mechanism which is activated in cells undergoing different forms of cellular

stress, while when stress continues autophagy may eventually become a cell death

mechanism. Despite recent advances in the understanding of molecular mechanisms and

biological functions of autophagic cell death it is currently unclear whether it is an alternative


way of dying, which is different from apoptotic and necrotic cell death or whether failure of

autophagy to rescue the cell can lead to cell death by either pathway.

1.4. Necrotic cell death

Type III cell death, according to the classification of Schweichel and Merker, is characterized

by swelling of the organelles (endoplasmic reticulum, mitochondria) and the cytoplasm with

subsequent collapse of the plasma membrane and lysis of the cells (Schweichel and Merker,

1973). The term oncosis was proposed to define the early stage of primary necrosis, during

which cells committed to death pass through a prelethal process in which they swell (Majno

and Joris, 1995). Notably, in the absence of phagocytosis apoptotic cells may lose their

cytoplasmic membrane integrity and proceed to a stage called secondary necrosis.

Secondary necrotic cells resemble the necrotic ones, but they have gone through an

apoptotic stage and this can still be recognized at the ultrastructural level by the typical

chromatin condensation pattern.

Necrotic cell death was often considered to be a passive process, lacking underlying

signaling events, which occurred under extreme physico-chemical conditions such as abrupt

anoxia, sudden shortage of nutrients, heat and detergents. Nevertheless, in several

physiological and pathological conditions necrosis is described as a form of caspase-

independent cell death. For example, necrosis has been found to be a potential substitute for

apoptosis during development. The loss of interdigital cells in the mouse embryo, a prototype

of programmed cell death, still occurs by necrosis either upon caspase inhibition by drugs, or

in mice bearing a mutation in the apaf-1 gene (Chautan et al., 1999). Caspase-independent

cell death is also involved in processes such as the negative selection of lymphocytes

(Doerfler et al., 2000; Jaattela and Tschopp, 2003; Smith et al., 1996), TNF-mediated liver

injury (Kunstle et al., 1999) and the death of chondrocytes controlling the longitudinal growth

of bones (Roach and Clarke, 2000). Necrotic cell death is also instrumental during cellular

turnover in the human large intestine (Barkla and Gibson, 1999) and in response to viral

infection (Chan et al., 2003).

Because the precise characterization of biochemical mechanisms of different forms of

caspase-independent cell deaths was still in progress, Leist and Jaattela proposed a

classification of cell death types based on the stage of chromatin condensation and the

nuclear morphology into four subclasses: classic apoptosis, apoptosis-like PCD, necrotic-like

PCD and accidental necrosis (Leist and Jaattela, 2001). Apoptosis-like PCD is characterized

by less compact, lumpy chromatin masses while necrotic-like PCD occurs either in the

complete absence of chromatin condensation or, at best, with chromatin clustering to form

loose speckles. Varying degrees of other apoptosis-like features, including externalization of


PS before lysis of the plasma membrane, can be found during apoptosis- and necrosis-like

PCD (Holler et al., 2000; Krysko et al., 2004; Mateo et al., 1999). The term PCD implies an

active process that is dependent on signaling events in the dying cell while accidental

necrosis is a passive process which occurs in response to exposure to high concentrations of

detergents, oxidants, and heat, and which can be prevented only by removal of the stimulus

(Leist and Jaattela, 2001). A broad array of cell death models that occur in the absence of

caspase activation falls into the category of either apoptosis- or necrotic-like PCD.

Next, we would like to consider some biochemical mechanisms which are implicated in

necrotic-like PCD. TNF can induce either apoptotic or necrotic cell death, depending on the

cell line (Laster et al., 1988). In fibrosarcoma L929, cell line TNF induces necrotic cell death

without the involvement of caspases, despite the fact that stimulation of Fas leads to

classical apoptosis in the same cells (Vercammen et al., 1998b). The necrotic pathway in

L929 cell line triggered by FasL can be detected only prior to the inactivation of the caspases

by zVAD-fmk (Vercammen et al., 1998a). Thus, both death domain-containing receptors,

TNFR1 and Fas, can initiate apoptotic and necrotic cell death depending on the cellular

context. This death receptor-induced caspase-independent pathway is not restricted to the

L929 cell line. Similarly, in the U937 or Jurkat T lymphocyte cell lines TNF, Fas and TRAIL

trigger necrotic-like cell death when caspases are inhibited (Holler et al., 2000; Khwaja and

Tatton, 1999; Matsumura et al., 2000). FADD is an important adaptor molecule, which serves

as a platform for the initiation of apoptotic (discussed in the subsection on apoptosis) as well

as necrotic cell death. Overexpression of a FADD containing only DD (FADD-DD) leads to

necrosis while overexpression of FADD containing only DED (FADD-DED) kills the L929

cells by apoptosis, and cell death can be redirected from apoptosis to necrosis in the

presence of zVAD-fmk (Boone et al., 2000; Vanden Berghe et al., 2004). These data suggest

that bifurcation between apoptosis and necrosis might be situated at the level of the adaptor

protein FADD. Binding of caspase-8 to DED of FADD leads to its activation and subsequent

apoptosis, while when the recruitment or the enzymatic activity of caspase-8 is prevented,

the presence of death domain of FADD in the receptor complex leads to necrosis (Boone et

al., 2000; Vanden Berghe et al., 2004). Another molecule involved in the necrotic signaling

cascade is RIP1. A study where FADD- and RIP-deficient Jurkat cells were used identified

the kinase RIP1 as a crucial component of Fas- and TNF-mediated necrotic cell death (Holler

et al., 2000). TNF signal is transduced by homotypic interaction of DD with RIP1, which

allows RIP1 to bind directly to the TNFR1 or indirectly through TRADD (Harper et al., 2003;

Stanger et al., 1995). It is interesting to note that RIP1, a key mediator in the necrotic cell

death can be cleaved by successful activation of caspase-8 (Lin et al., 1999; Martinon et al.,

2000), indicating that apoptotic and necrotic cell death pathways interfere with each other

where the initiation of apoptosis might actively suppress the necrotic pathway. The pathway


downstream of RIP1 which is responsible for Fas- and TNFR1-induced necrotic cell death is

unknown. Ample evidence suggests that excess formation of reactive oxygen species (ROS)

is involved. Inhibition of caspases (which sensitizes to necrosis) results in increased

formation of ROS, and the addition of the scavenger butylated hydroxyanisole (BHA) protects

L929 cells from TNF-induced necrosis, but not from Fas-mediated apoptosis (Goossens et

al., 1999; Vercammen et al., 1998b), while necrosis induced by Fas is also blocked by BHA

(Vercammen et al., 1998a). The strong protective effect of BHA may go beyond its oxygen

radical scavenging activity. Festjens et al. reported that BHA may inhibit directly several

mechanisms that are implicated in necrotic cell death such as complex I activity, PLA2

activation and lipoxygenase activity (Festjens et al., 2006).

The picture described above indicates a complexity of death receptor-induced necrotic

signaling networks that far exceeds that of the simple linear pathways originally indicated by

the discovery of the receptor-triggered caspase cascade and supports the notion that,

besides uncontrollable necrosis that may occur following massive mechanical aggression or

harsh chemical treatment of tissues, programmed necrotic-like cell death may exist with

certain molecular events. The challenge ahead is to explore whether signaling pathways of

necrotic cell death can be also extrapolated to in vivo models of necrosis. Remarkably, in

normal wild-type interdigital space 6% necrotic cells in addition to the 5% apoptotic cells

were also observed (Chautan et al., 1999).

1.5. Cell death as a consequence of aberrant mitosis

Cell death occurring during or shortly after deregulation or failed mitosis is called mitotic

catastrophe and can be accompanied by morphological alterations such as micronuclei

(which often are chromosomes or chromosomes fragments that have not been distributed

evenly between the daughter nuclei) and multinucleation (the presence of two or more nuclei

with similar or heterogenous sizes, resulting from deficient separation during cytokinesis)

(Kroemer et al., 2005). Mitotic catastrophe can be triggered by mitotic failure caused by

defective cell cycle checkpoints, e.g., by the inhibition of CHK2 (Castedo et al., 2004). Mitotic

catastrophe can, in particular, be inuduced by DNA damage, microtubule stabilizing or

destabilizing agents (Roninson et al., 2001). This type of cell death kills the cell during or

close to metaphase in a p53-independent manner. Although morphologically distinct from

apoptosis (Roninson et al., 2001), mitotic catastrophe has been reported to be accompanied

by the activation of the apoptotic machinery. This includes signs of mitochondrial membrane

permeabilization, such as the loss of the mitochondrial transmembrane potential ( m), the

mitochondrial release of cytochrome c and AIF, caspase activation and DNA fragmentation

(Castedo et al., 2004). However, others have argued that it is different from apoptosis, as


caspase inhibition or Bcl-2 overexpression fails to prevent catastrophic mitosis or the

development of giant multinucleated cells (Lock and Stribinskiene, 1996; Roninson et al.,

2001). Whether mitotic catastrophe represents a separate cell death type is therefore still a

matter of debate (Kroemer et al., 2005). However, the Nomenclature Committee on Cell

Death proposed not to consider mitotic catastrophe as a separate cell death type (Kroemer

et al., 2005).

1.6 Cell senescence

Cell senescence can be considered to be a type of “living cell death”, because although

senescent cells maintain the integrity of their plasma membrane, they undergo permanent

growth arrest and lose their clonogenicity. In the 1960s, L. Hayflick observed that non-

transformed cells explanted in culture undergo only a limited number of passages, and then

enter a state that he called replicative senescence (Hayflick, 1965). The so-called ‘Hayflick

limit’ was linked to the progressive shortening of chromosomal ends during cell division

(Harley et al., 1990). However, there is still disagreement concerning the question of whether

replicative senescence is a general property of normal mammalian cells. In recent studies, it

has been shown that rat glial cells were capable of apparently unrestricted growth in culture

under optimized conditions (Mathon et al., 2001; Tang et al., 2001). In addition, replicative

senescence of mouse cells (which, in contrast to human cells, is at least in part telomere-

independent) was proposed to be an artifact of suboptimal culture conditions (Wright and

Shay, 2001). Similarly, MEFs did not senesce in physiological (3%) oxygen levels, but

underwent a spontaneous event that allowed indefinite proliferation in 20% oxygen

(Parrinello et al., 2003). However, it has become apparent in recent years that terminal

growth arrest and phenotypic changes associated with senescence can be triggered by

factors other than cell doubling or telomere shortening. In particular, DNA damage of normal

human fibroblasts induces terminal growth arrest, which is accompanied by phenotypic

changes that bear a striking resemblance to cells that undergo replicative senescence (Di

Leonardo et al., 1994; Robles and Adami, 1998). This rapid process has been termed

accelerated senescence, which, like apoptosis, was proposed to be a programmed protective

response of the organism to potentially carcinogenic damage (Weinberg, 1997). Despite the

essential difference, the replicative and accelerated senescence share common phenotypic

markers, which include enlarged and flattened morphology, increased granularity and

expression of a relatively specific marker of senescence, senescence-associated -

galactosidase (Dimri et al., 1995; Roninson et al., 2001). Senescence has been proposed as

a mechanism to block immortalization and tumorogenesis. Moreover, in addition to


apoptosis, replicative and accelerated senescence were identified as an effective response

to chemotherapy (Blagosklonny, 2003).

Table1. Different types of cell death. At least 11 different types of cell death are known, 10 of which proceed according to genetically programmed mechanisms. *Programmed necrotic cell death may be classified under this category. WD, Wallerian degeneration; PLT, platelets; TG, transglutaminase; NO, nitric oxide; NCX, sodium calcium exchange channel; IAP, inhibitor of apoptosis proteins. Adapted from Melino et al. (2005).

At the moment it is possible to discriminate about 11 types of cell death (Table 1) which are

occurring in mammals (Melino et al., 2005). In this chapter we discussed several of them and

in particular we have focused on the apoptotic and necrotic cell deaths. The interest of the

latter form of cell death is increasing, but a precise description of the morphological and

biochemical events has not yet been given, as well as the importance of this cell death type

in embryology, ontogeny, physiology and pathophysiology. Moreover, until now no positive

discriminative marker for necrotic cell death, such as caspases for apoptotic cell death, has

yet been reported. In addition, the paradoxical roles of autophagy in cell survival and cell

death have been discussed. Many interesting, unresolved questions remain regarding

autophagic cell death; for example, it is yet unclear whether autophagy represents a

separate type of cell death or whether autophagy eventually results in apoptotic and necrotic

CaspasesChr. condens.

DNA fragment.

intactintactIntact

PS Exp

YesAnoikis

AP1,

caclium

TGLostLostYesLens

TG 1,3,5LostLostCrosslinked

lipid-

reassembly

IntactYesCornification

GATA2LostLostIntactIntactYesPLT

NO,

calcium

CalpainsLostLostIntactIntactYesExcitotoxicity

VPRIntact

lipid-

reassembly

IntactYesWD

Lysosomal

Beclin 1

Partial

chr.condens.

DNA fragm.

IntactIntact

PS exp

YesAutophagy

CalpainsIntactIntactIntact

PS exp

YesCaspase-

independent

cell death*

Bcl2 family,

IAP

Death RecCaspasesChr. condens.

DNA fragment.

IntactIntactIntcat

PS exp

YesApoptosis

NoneBlownLysedLysedNoneAccidental

Necrosis

RegulatorsReceptorsEnzymesNucleusMitosOrganellesMembraneGenetic

program

CaspasesChr. condens.

DNA fragment.

intactintactIntact

PS Exp

YesAnoikis

AP1,

caclium

TGLostLostYesLens

TG 1,3,5LostLostCrosslinked

lipid-

reassembly

IntactYesCornification

GATA2LostLostIntactIntactYesPLT

NO,

calcium

CalpainsLostLostIntactIntactYesExcitotoxicity

VPRIntact

lipid-

reassembly

IntactYesWD

Lysosomal

Beclin 1

Partial

chr.condens.

DNA fragm.

IntactIntact

PS exp

YesAutophagy

CalpainsIntactIntactIntact

PS exp

YesCaspase-

independent

cell death*

Bcl2 family,

IAP

Death RecCaspasesChr. condens.

DNA fragment.

IntactIntactIntcat

PS exp

YesApoptosis

NoneBlownLysedLysedNoneAccidental

Necrosis

RegulatorsReceptorsEnzymesNucleusMitosOrganellesMembraneGenetic

program


cell death. Future investigations of the molecular mechanisms of the whole spectrum of cell

death types will yield better insight into the evolution of cell-death programs, their

interrelationship with each other, and their potential for molecular targeting and manipulation

in many diseases.


References

Aravind L, Dixit VM, Koonin EV. 1999. The domains of death: evolution of the apoptosis

machinery. Trends Biochem Sci 24(2):47-53.

Baba M, Osumi M, Scott SV, Klionsky DJ, Ohsumi Y. 1997. Two distinct pathways for

targeting proteins from the cytoplasm to the vacuole/lysosome. J Cell Biol

139(7):1687-1695.

Baehrecke EH. 2005. Autophagy: dual roles in life and death? Nat Rev Mol Cell Biol

6(6):505-510.

Barkla DH, Gibson PR. 1999. The fate of epithelial cells in the human large intestine.

Pathology 31(3):230-238.

Blagosklonny MV. 2003. Cell senescence and hypermitogenic arrest. EMBO Rep 4(4):358-

362.

Boone E, Vanden Berghe T, Van Loo G, De Wilde G, De Wael N, Vercammen D, Fiers W,

Haegeman G, Vandenabeele P. 2000. Structure/Function analysis of p55 tumor

necrosis factor receptor and fas-associated death domain. Effect on necrosis in

L929sA cells. J Biol Chem 275(48):37596-37603.

Boya P, Gonzalez-Polo RA, Casares N, Perfettini JL, Dessen P, Larochette N, Metivier D,

Meley D, Souquere S, Yoshimori T, Pierron G, Codogno P, Kroemer G. 2005.

Inhibition of macroautophagy triggers apoptosis. Mol Cell Biol 25(3):1025-1040.

Bursch W, Ellinger A, Gerner C, Frohwein U, Schulte-Hermann R. 2000. Programmed cell

death (PCD). Apoptosis, autophagic PCD, or others? Ann N Y Acad Sci 926:1-12.

Castedo M, Perfettini JL, Roumier T, Yakushijin K, Horne D, Medema R, Kroemer G. 2004.

The cell cycle checkpoint kinase Chk2 is a negative regulator of mitotic catastrophe.

Oncogene 23(25):4353-4361.

Chan FK, Shisler J, Bixby JG, Felices M, Zheng L, Appel M, Orenstein J, Moss B, Lenardo

MJ. 2003. A role for tumor necrosis factor receptor-2 and receptor-interacting protein

in programmed necrosis and antiviral responses. J Biol Chem 278(51):51613-51621.

Chautan M, Chazal G, Cecconi F, Gruss P, Golstein P. 1999. Interdigital cell death can occur

through a necrotic and caspase-independent pathway. Curr Biol 9(17):967-970.

Chen G, Goeddel DV. 2002. TNF-R1 signaling: a beautiful pathway. Science

296(5573):1634-1635.

Chipuk JE, Green DR. 2006. Dissecting p53-dependent apoptosis. Cell Death Differ

13(6):994-1002.

Chipuk JE, Kuwana T, Bouchier-Hayes L, Droin NM, Newmeyer DD, Schuler M, Green DR.

2004. Direct activation of Bax by p53 mediates mitochondrial membrane

permeabilization and apoptosis. Science 303(5660):1010-1014.

Clarke PG, Clarke S. 1995. Historic apoptosis. Nature 378(6554):230.

Clarke PG, Clarke S. 1996. Nineteenth century research on naturally occurring cell death and

related phenomena. Anat Embryol (Berl) 193(2):81-99.

Deiss LP, Galinka H, Berissi H, Cohen O, Kimchi A. 1996. Cathepsin D protease mediates

programmed cell death induced by interferon-gamma, Fas/APO-1 and TNF-alpha.

Embo J 15(15):3861-3870.

Di Leonardo A, Linke SP, Clarkin K, Wahl GM. 1994. DNA damage triggers a prolonged

p53-dependent G1 arrest and long-term induction of Cip1 in normal human

fibroblasts. Genes Dev 8(21):2540-2551.

Di Sano F, Ferraro E, Tufi R, Achsel T, Piacentini M, Cecconi F. 2006. Endoplasmic

reticulum stress induces apoptosis by an apoptosome-dependent but caspase 12-

independent mechanism. J Biol Chem 281(5):2693-2700.


Dimri GP, Lee X, Basile G, Acosta M, Scott G, Roskelley C, Medrano EE, Linskens M,

Rubelj I, Pereira-Smith O, et al. 1995. A biomarker that identifies senescent human

cells in culture and in aging skin in vivo. Proc Natl Acad Sci U S A 92(20):9363-9367.

Doerfler P, Forbush KA, Perlmutter RM. 2000. Caspase enzyme activity is not essential for

apoptosis during thymocyte development. J Immunol 164(8):4071-4079.

Ferri KF, Kroemer G. 2001. Organelle-specific initiation of cell death pathways. Nat Cell

Biol 3(11):E255-263.

Festjens N, Kalai M, Smet J, Meeus A, Van Coster R, Saelens X, Vandenabeele P. 2006.

Butylated hydroxyanisole is more than a reactive oxygen species scavenger. Cell

Death Differ 13(1):166-169.

Festjens N, van Gurp M, van Loo G, Saelens X, Vandenabeele P. 2004. Bcl-2 family

members as sentinels of cellular integrity and role of mitochondrial intermembrane

space proteins in apoptotic cell death. Acta Haematol 111(1-2):7-27.

Gonzalez-Polo RA, Boya P, Pauleau AL, Jalil A, Larochette N, Souquere S, Eskelinen EL,

Pierron G, Saftig P, Kroemer G. 2005. The apoptosis/autophagy paradox: autophagic

vacuolization before apoptotic death. J Cell Sci 118(Pt 14):3091-3102.

Goossens V, De Vos K, Vercammen D, Steemans M, Vancompernolle K, Fiers W,

Vandenabeele P, Grooten J. 1999. Redox regulation of TNF signaling. Biofactors

10(2-3):145-156.

Harding TM, Morano KA, Scott SV, Klionsky DJ. 1995. Isolation and characterization of

yeast mutants in the cytoplasm to vacuole protein targeting pathway. J Cell Biol

131(3):591-602.

Harley CB, Futcher AB, Greider CW. 1990. Telomeres shorten during ageing of human

fibroblasts. Nature 345(6274):458-460.

Harper N, Hughes M, MacFarlane M, Cohen GM. 2003. Fas-associated death domain protein

and caspase-8 are not recruited to the tumor necrosis factor receptor 1 signaling

complex during tumor necrosis factor-induced apoptosis. J Biol Chem 278(28):25534-

25541.

Hayflick L. 1965. The Limited in Vitro Lifetime of Human Diploid Cell Strains. Exp Cell Res

37:614-636.

Hofmann K. 1999. The modular nature of apoptotic signaling proteins. Cell Mol Life Sci

55(8-9):1113-1128.

Holler N, Zaru R, Micheau O, Thome M, Attinger A, Valitutti S, Bodmer JL, Schneider P,

Seed B, Tschopp J. 2000. Fas triggers an alternative, caspase-8-independent cell death

pathway using the kinase RIP as effector molecule. Nat Immunol 1(6):489-495.

Inbal B, Bialik S, Sabanay I, Shani G, Kimchi A. 2002. DAP kinase and DRP-1 mediate

membrane blebbing and the formation of autophagic vesicles during programmed cell

death. J Cell Biol 157(3):455-468.

Jaattela M, Tschopp J. 2003. Caspase-independent cell death in T lymphocytes. Nat Immunol

4(5):416-423.

Jiang X, Wang X. 2000. Cytochrome c promotes caspase-9 activation by inducing nucleotide

binding to Apaf-1. J Biol Chem 275(40):31199-31203.

Juhasz G, Csikos G, Sinka R, Erdelyi M, Sass M. 2003. The Drosophila homolog of Aut1 is

essential for autophagy and development. FEBS Lett 543(1-3):154-158.

Kalai M, Lamkanfi M, Denecker G, Boogmans M, Lippens S, Meeus A, Declercq W,

Vandenabeele P. 2003. Regulation of the expression and processing of caspase-12. J

Cell Biol 162(3):457-467.

Kerr JF, Wyllie AH, Currie AR. 1972. Apoptosis: a basic biological phenomenon with wide-

ranging implications in tissue kinetics. Br J Cancer 26(4):239-257.


Khwaja A, Tatton L. 1999. Resistance to the cytotoxic effects of tumor necrosis factor alpha

can be overcome by inhibition of a FADD/caspase-dependent signaling pathway. J

Biol Chem 274(51):36817-36823.

Kischkel FC, Hellbardt S, Behrmann I, Germer M, Pawlita M, Krammer PH, Peter ME. 1995.

Cytotoxicity-dependent APO-1 (Fas/CD95)-associated proteins form a death-inducing

signaling complex (DISC) with the receptor. Embo J 14(22):5579-5588.

Klionsky DJ, Emr SD. 2000. Autophagy as a regulated pathway of cellular degradation.

Science 290(5497):1717-1721.

Korsmeyer SJ, Wei MC, Saito M, Weiler S, Oh KJ, Schlesinger PH. 2000. Pro-apoptotic

cascade activates BID, which oligomerizes BAK or BAX into pores that result in the

release of cytochrome c. Cell Death Differ 7(12):1166-1173.

Kroemer G, El-Deiry WS, Golstein P, Peter ME, Vaux D, Vandenabeele P, Zhivotovsky B,

Blagosklonny MV, Malorni W, Knight RA, Piacentini M, Nagata S, Melino G. 2005.

Classification of cell death: recommendations of the Nomenclature Committee on Cell

Death. Cell Death Differ 12 Suppl 2:1463-1467.

Krysko O, De Ridder L, Cornelissen M. 2004. Phosphatidylserine exposure during early

primary necrosis (oncosis) in JB6 cells as evidenced by immunogold labeling

technique. Apoptosis 9(4):495-500.

Kuma A, Hatano M, Matsui M, Yamamoto A, Nakaya H, Yoshimori T, Ohsumi Y, Tokuhisa

T, Mizushima N. 2004. The role of autophagy during the early neonatal starvation

period. Nature 432(7020):1032-1036.

Kunstle G, Hentze H, Germann PG, Tiegs G, Meergans T, Wendel A. 1999. Concanavalin A

hepatotoxicity in mice: tumor necrosis factor-mediated organ failure independent of

caspase-3-like protease activation. Hepatology 30(5):1241-1251.

Lakhani SA, Masud A, Kuida K, Porter GA, Jr., Booth CJ, Mehal WZ, Inayat I, Flavell RA.

2006. Caspases 3 and 7: key mediators of mitochondrial events of apoptosis. Science

311(5762):847-851.

Lamkanfi M, Kalai M, Vandenabeele P. 2004. Caspase-12: an overview. Cell Death Differ

11(4):365-368.

Laster SM, Wood JG, Gooding LR. 1988. Tumor necrosis factor can induce both apoptic and

necrotic forms of cell lysis. J Immunol 141(8):2629-2634.

Lee CY, Baehrecke EH. 2001. Steroid regulation of autophagic programmed cell death during

development. Development 128(8):1443-1455.

Lee CY, Clough EA, Yellon P, Teslovich TM, Stephan DA, Baehrecke EH. 2003. Genome-

wide analyses of steroid- and radiation-triggered programmed cell death in

Drosophila. Curr Biol 13(4):350-357.

Leist M, Jaattela M. 2001. Four deaths and a funeral: from caspases to alternative

mechanisms. Nat Rev Mol Cell Biol 2(8):589-598.

Levine B, Yuan J. 2005. Autophagy in cell death: an innocent convict? J Clin Invest

115(10):2679-2688.

Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES, Wang X. 1997.

Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates

an apoptotic protease cascade. Cell 91(4):479-489.

Lin Y, Devin A, Rodriguez Y, Liu ZG. 1999. Cleavage of the death domain kinase RIP by

caspase-8 prompts TNF-induced apoptosis. Genes Dev 13(19):2514-2526.

Lock RB, Stribinskiene L. 1996. Dual modes of death induced by etoposide in human

epithelial tumor cells allow Bcl-2 to inhibit apoptosis without affecting clonogenic

survival. Cancer Res 56(17):4006-4012.

Lockshin RA, Williams CM. 1965. Programmed Cell Death--I. Cytology of Degeneration in

the Intersegmental Muscles of the Pernyi Silkmoth. J Insect Physiol 11:123-133.


Lockshin RA, Zakeri Z. 2004. Apoptosis, autophagy, and more. Int J Biochem Cell Biol

36(12):2405-2419.

Lum JJ, Bauer DE, Kong M, Harris MH, Li C, Lindsten T, Thompson CB. 2005. Growth

factor regulation of autophagy and cell survival in the absence of apoptosis. Cell

120(2):237-248.

Luo X, Budihardjo I, Zou H, Slaughter C, Wang X. 1998. Bid, a Bcl2 interacting protein,

mediates cytochrome c release from mitochondria in response to activation of cell

surface death receptors. Cell 94(4):481-490.

Maag RS, Mancini M, Rosen A, Machamer CE. 2005. Caspase-resistant Golgin-160 disrupts

apoptosis induced by secretory pathway stress and ligation of death receptors. Mol

Biol Cell 16(6):3019-3027.

Majno G, Joris I. 1995. Apoptosis, oncosis, and necrosis. An overview of cell death. Am J

Pathol 146(1):3-15.

Mancini M, Machamer CE, Roy S, Nicholson DW, Thornberry NA, Casciola-Rosen LA,

Rosen A. 2000. Caspase-2 is localized at the Golgi complex and cleaves golgin-160

during apoptosis. J Cell Biol 149(3):603-612.

Martin DN, Baehrecke EH. 2004. Caspases function in autophagic programmed cell death in

Drosophila. Development 131(2):275-284.

Martinon F, Holler N, Richard C, Tschopp J. 2000. Activation of a pro-apoptotic

amplification loop through inhibition of NF-kappaB-dependent survival signals by

caspase-mediated inactivation of RIP. FEBS Lett 468(2-3):134-136.

Mateo V, Lagneaux L, Bron D, Biron G, Armant M, Delespesse G, Sarfati M. 1999. CD47

ligation induces caspase-independent cell death in chronic lymphocytic leukemia. Nat

Med 5(11):1277-1284.

Mathon NF, Malcolm DS, Harrisingh MC, Cheng L, Lloyd AC. 2001. Lack of replicative

senescence in normal rodent glia. Science 291(5505):872-875.

Matsumura H, Shimizu Y, Ohsawa Y, Kawahara A, Uchiyama Y, Nagata S. 2000. Necrotic

death pathway in Fas receptor signaling. J Cell Biol 151(6):1247-1256.

Melendez A, Talloczy Z, Seaman M, Eskelinen EL, Hall DH, Levine B. 2003. Autophagy

genes are essential for dauer development and life-span extension in C. elegans.

Science 301(5638):1387-1391.

Melino G, Knight RA, Nicotera P. 2005. How many ways to die? How many different models

of cell death? Cell Death Differ 12 Suppl 2:1457-1462.

Mills KR, Reginato M, Debnath J, Queenan B, Brugge JS. 2004. Tumor necrosis factor-

related apoptosis-inducing ligand (TRAIL) is required for induction of autophagy

during lumen formation in vitro. Proc Natl Acad Sci U S A 101(10):3438-3443.

Muzio M, Chinnaiyan AM, Kischkel FC, O'Rourke K, Shevchenko A, Ni J, Scaffidi C, Bretz

JD, Zhang M, Gentz R, Mann M, Krammer PH, Peter ME, Dixit VM. 1996. FLICE, a

novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95

(Fas/APO-1) death--inducing signaling complex. Cell 85(6):817-827.

Nagata S. 1999. Fas ligand-induced apoptosis. Annu Rev Genet 33:29-55.

Nakagawa T, Zhu H, Morishima N, Li E, Xu J, Yankner BA, Yuan J. 2000. Caspase-12

mediates endoplasmic-reticulum-specific apoptosis and cytotoxicity by amyloid-beta.

Nature 403(6765):98-103.

Obeng EA, Boise LH. 2005. Caspase-12 and caspase-4 are not required for caspase-

dependent endoplasmic reticulum stress-induced apoptosis. J Biol Chem

280(33):29578-29587.

Okada H, Mak TW. 2004. Pathways of apoptotic and non-apoptotic death in tumour cells. Nat

Rev Cancer 4(8):592-603.


Parrinello S, Samper E, Krtolica A, Goldstein J, Melov S, Campisi J. 2003. Oxygen

sensitivity severely limits the replicative lifespan of murine fibroblasts. Nat Cell Biol

5(8):741-747.

Pattingre S, Tassa A, Qu X, Garuti R, Liang XH, Mizushima N, Packer M, Schneider MD,

Levine B. 2005. Bcl-2 antiapoptotic proteins inhibit Beclin 1-dependent autophagy.

Cell 122(6):927-939.

Peter ME, Krammer PH. 2003. The CD95(APO-1/Fas) DISC and beyond. Cell Death Differ

10(1):26-35.

Petiot A, Pattingre S, Arico S, Meley D, Codogno P. 2002. Diversity of signaling controls of

macroautophagy in mammalian cells. Cell Struct Funct 27(6):431-441.

Roach HI, Clarke NM. 2000. Physiological cell death of chondrocytes in vivo is not confined

to apoptosis. New observations on the mammalian growth plate. J Bone Joint Surg Br

82(4):601-613.

Robles SJ, Adami GR. 1998. Agents that cause DNA double strand breaks lead to p16INK4a

enrichment and the premature senescence of normal fibroblasts. Oncogene

16(9):1113-1123.

Rodriguez J, Lazebnik Y. 1999. Caspase-9 and APAF-1 form an active holoenzyme. Genes

Dev 13(24):3179-3184.

Roninson IB, Broude EV, Chang BD. 2001. If not apoptosis, then what? Treatment-induced

senescence and mitotic catastrophe in tumor cells. Drug Resist Updat 4(5):303-313.

Saelens X, Festjens N, Vande Walle L, van Gurp M, van Loo G, Vandenabeele P. 2004.

Toxic proteins released from mitochondria in cell death. Oncogene 23(16):2861-2874.

Saleh A, Srinivasula SM, Acharya S, Fishel R, Alnemri ES. 1999. Cytochrome c and dATP-

mediated oligomerization of Apaf-1 is a prerequisite for procaspase-9 activation. J

Biol Chem 274(25):17941-17945.

Scaffidi C, Fulda S, Srinivasan A, Friesen C, Li F, Tomaselli KJ, Debatin KM, Krammer PH,

Peter ME. 1998. Two CD95 (APO-1/Fas) signaling pathways. Embo J 17(6):1675-

1687.

Schweichel JU, Merker HJ. 1973. The morphology of various types of cell death in prenatal

tissues. Teratology 7(3):253-266.

Sheikh MS, Fornace AJ, Jr. 2000. Death and decoy receptors and p53-mediated apoptosis.

Leukemia 14(8):1509-1513.

Shimizu S, Kanaseki T, Mizushima N, Mizuta T, Arakawa-Kobayashi S, Thompson CB,

Tsujimoto Y. 2004. Role of Bcl-2 family proteins in a non-apoptotic programmed cell

death dependent on autophagy genes. Nat Cell Biol 6(12):1221-1228.

Smith KG, Strasser A, Vaux DL. 1996. CrmA expression in T lymphocytes of transgenic

mice inhibits CD95 (Fas/APO-1)-transduced apoptosis, but does not cause

lymphadenopathy or autoimmune disease. Embo J 15(19):5167-5176.

Stanger BZ, Leder P, Lee TH, Kim E, Seed B. 1995. RIP: a novel protein containing a death

domain that interacts with Fas/APO-1 (CD95) in yeast and causes cell death. Cell

81(4):513-523.

Tang DG, Tokumoto YM, Apperly JA, Lloyd AC, Raff MC. 2001. Lack of replicative

senescence in cultured rat oligodendrocyte precursor cells. Science 291(5505):868-

871.

Thumm M, Egner R, Koch B, Schlumpberger M, Straub M, Veenhuis M, Wolf DH. 1994.

Isolation of autophagocytosis mutants of Saccharomyces cerevisiae. FEBS Lett

349(2):275-280.

Tsukada M, Ohsumi Y. 1993. Isolation and characterization of autophagy-defective mutants

of Saccharomyces cerevisiae. FEBS Lett 333(1-2):169-174.


Twiddy D, Brown DG, Adrain C, Jukes R, Martin SJ, Cohen GM, MacFarlane M, Cain K.

2004. Pro-apoptotic proteins released from the mitochondria regulate the protein

composition and caspase-processing activity of the native Apaf-1/caspase-9

apoptosome complex. J Biol Chem 279(19):19665-19682.

Twiddy D, Cohen GM, Macfarlane M, Cain K. 2006. Caspase-7 is directly activated by the

approximately 700-kDa apoptosome complex and is released as a stable XIAP-

caspase-7 approximately 200-kDa complex. J Biol Chem 281(7):3876-3888.

van Loo G, van Gurp M, Depuydt B, Srinivasula SM, Rodriguez I, Alnemri ES, Gevaert K,

Vandekerckhove J, Declercq W, Vandenabeele P. 2002. The serine protease

Omi/HtrA2 is released from mitochondria during apoptosis. Omi interacts with

caspase-inhibitor XIAP and induces enhanced caspase activity. Cell Death Differ

9(1):20-26.

Vanden Berghe T, van Loo G, Saelens X, Van Gurp M, Brouckaert G, Kalai M, Declercq W,

Vandenabeele P. 2004. Differential signaling to apoptotic and necrotic cell death by

Fas-associated death domain protein FADD. J Biol Chem 279(9):7925-7933.

Vercammen D, Beyaert R, Denecker G, Goossens V, Van Loo G, Declercq W, Grooten J,

Fiers W, Vandenabeele P. 1998a. Inhibition of caspases increases the sensitivity of

L929 cells to necrosis mediated by tumor necrosis factor. J Exp Med 187(9):1477-

1485.

Vercammen D, Brouckaert G, Denecker G, Van de Craen M, Declercq W, Fiers W,

Vandenabeele P. 1998b. Dual signaling of the Fas receptor: initiation of both apoptotic

and necrotic cell death pathways. J Exp Med 188(5):919-930.

Vousden KH. 2005. Apoptosis. p53 and PUMA: a deadly duo. Science 309(5741):1685-1686.

Wang X. 2001. The expanding role of mitochondria in apoptosis. Genes Dev 15(22):2922-

2933.

Weinberg RA. 1997. The cat and mouse games that genes, viruses, and cells play. Cell

88(5):573-575.

Wright WE, Shay JW. 2001. Cellular senescence as a tumor-protection mechanism: the

essential role of counting. Curr Opin Genet Dev 11(1):98-103.

Yanagisawa H, Miyashita T, Nakano Y, Yamamoto D. 2003. HSpin1, a transmembrane

protein interacting with Bcl-2/Bcl-xL, induces a caspase-independent autophagic cell

death. Cell Death Differ 10(7):798-807.

Yang X, Chang HY, Baltimore D. 1998. Essential role of CED-4 oligomerization in CED-3

activation and apoptosis. Science 281(5381):1355-1357.

Yu L, Alva A, Su H, Dutt P, Freundt E, Welsh S, Baehrecke EH, Lenardo MJ. 2004.

Regulation of an ATG7-beclin 1 program of autophagic cell death by caspase-8.

Science 304(5676):1500-1502.

Zou H, Li Y, Liu X, Wang X. 1999. An APAF-1.cytochrome c multimeric complex is a

functional apoptosome that activates procaspase-9. J Biol Chem 274(17):11549-

11556.

2. Gap junctions and the propagation of

cell survival and cell death signals

DV Krysko, L Leybaert, P Vandenabeele and K D’HerdeApoptosis 2005; 10(3): 459-69. Review.

Gap junctions and the propagation of cell survivaland cell death signals

D. V. Krysko, L. Leybaert, P. Vandenabeele∗ and K. D’Herde∗

Department of Human Anatomy, Embryology, Histology and Medical Physics (D. V. Krysko, K. D.’Herde),Molecular Signaling and Cell Death Unit, Department of Molecular Biomedical Research, VIB (D. V. Krysko, P.Vandenabeele) and Department of Physiology and Pathophysiology, Ghent University (L. Leybaert), 9000 Ghent,Belgium

Gap junctions are a unique type of intercellular channelsthat connect the cytoplasm of adjoining cells. Each gapjunction channel is comprised of two hemichannels orconnexons and each connexon is formed by the aggre-gation of six protein subunits known as connexins. Gapjunction channels allow the intercellular passage of small(<1.5 kDa) molecules and regulate essential processesduring development and differentiation. However, theirrole in cell survival and cell death is poorly understood.We review experimental data that support the hypothe-sis that gap junction channels may propagate cell deathand survival modulating signals. In addition, we explorethe hypothesis that hemichannels (or unapposed connex-ons) might be used as a paracrine conduit to spread fac-tors that modulate the fate of the surrounding cells. Fi-nally, direct signal transduction activity of connexins incell death and survival pathways is addressed.

Keywords: apoptosis; gap junctions; survival; chemichannels;connexins.

Introduction

Most studies on apoptosis regulation are focusing predom-inantly on intracellular signaling pathways within singlecell1–4 but it is conceivable that the decision which cellsdivide and which cells die must be regulated to some ex-tent by the organized microenvironment that cells livewithin. The pioneering work of Kanno and Loewenstein5

∗These authors share senior authorship.

This study was supported by Ghent University GOA grant no.12050502.

Correspondence to: Dmitri V. Krysko, Department of Hu-man Anatomy, Embryology, Histology and Medical Physics,Godshuizenlaan, 4, B-9000 Ghent, Belgium. Tel.: +32-9-2649232; Fax: +32-9-2259452; e-mail: [email protected];dmitri [email protected]

demonstrated that in most tissues, adjacent cells couldcommunicate in a direct way, without dilution of thesignaling molecules in extracellular compartments. Gapjunctions were first defined morphologically as specializedcontacts between cells.6,7 Although gap junctional cou-pling is present in all multicellular animals, vertebratesand invertebrate use completely unrelated gene familiesto encode gap junction proteins, mammals have connex-ins (Cxs) whereas Drosophila and C. elegans use 8 and 25innexin (invertebrate connexins) genes, respectively.8 Themammalian connexin family consists of 20 distinct pro-teins, which range in molecular mass from 20 to 62 kDa.9

Several connexins and tissues in which they are highlyexpressed are listed in Table 1. Gap junctions are chan-nels formed by two hemichannels, each formed by sixCx subunits contributed by each cell (Figures 1 and 2A).Gap junctions are usually constructed from more thanone type of connexin and thus connexon hemichannelsmay be homo- or heteromeric and gap junctions homo- orheterotypic10,11 (Figure 2B). Connexin proteins of verte-brates posses four hydrophobic membrane-spanning do-mains, two conserved, extracellular domains involved indocking with a connexin in the adjacent membrane, andthree cytoplasmic domains corresponding to the amino(N)-terminal region, a loop between transmembrane do-mains 2 and 3, and the carboxy C-terminal tail region12–15

(Figure 2C). Oligomerization of connexins into hemichan-nels starts in the endoplasmic reticulum and is com-pleted in the Golgi complex, where vesicles contain-ing hemichannels travel or are transported to fuse withthe plasma membrane.16,17 Depending on the connexintype, pore diameter ranges between approximately 6.5 and15 A, which is wide enough to allow the passage moleculesless than 1.5 kDa such as water, individual amino acids orshort peptides, all relevant cations and anions, includingCl− , Na+, K+, and Ca2+ and most second messengers asIP3, cAMP. The functional state of gap junction channelscan be dynamically regulated by covalent or non-covalentmodifications of the channels structure, a phenomenon

27

Table 1. Some of the mammalian connexins and organs ortissues in which they expressed. For original references seereviews.9,13,15,121

Connexin Organ or tissue distribution

Cx 26 Liver, mammary gland, skin and appendages,mucous membranes, pancreas, intestine,endometrium, lung, brain

Cx30 Brain, skin, cornea, cochlea

Cx30.3 Skin, kidney, placenta

Cx31 Skin, placenta, kidney, testes

Cx32 Brain, liver, thyroid, uterus, pancreas, kidney

Cx37 Endothelium, lung, ovary, heart, neutrophils,lymphocytes

Cx43 Brain, heart, gonads, lens, retina, skin, bone,pancreas, granulosa cells,

connective tissue, neutrophils, monocytes,macrophages, lymphocytes

Cx46 Heart, lens, kidney, lung, bone, testes

Cx50 Heart, lens, retina, cornea

Figure 1. Electron microscopic micrograph is showing gap junc-tion between two adjacent granulosa cells (arrowheads), whichwere prepared from ovarian follicles of adult regularly layingJapanese quail (Coturnix coturnix japonica). Different cellular or-ganelles can be observed: mitochondria (M) and endoplasmicreticulum (arrow). Scale bar 0.1 µm.

called gating. The best-studied gating mechanism, com-mon to most ion channels, is voltage dependency, butits physiological relevance is not well defined.18 Calcium(Ca2+) and hydrogen ions, as well as phosphorylation,have also been implicated in the modulation of chan-nel permeability.19–21 Gap junctional communication haswell-known effects on cell proliferation, differentiationand tumor-suppression,22–24 however, the mechanism(s)by which gap junctions coordinate spreading of cell deathand survival modulating signals across cell populationsis not very well understood. The aim of this review is tobring together recent reports on gap junction-dependentand hemichannel-dependent mechanisms responsible inthe propagation of signals modulating the fate of neigh-

boring cells. Finally the effect of the connexins themselves,the major building blocks of gap junctions and hemichan-nels, on cell death and survival pathway is addressed.

Gap junction channels—a conduit for spreadingcell fate modulating signals

One potential mechanism for the propagation of a cellfate modulator is via diffusion of an intracellular medi-ator(s) through gap junction channels. The capacity ofcells to kill each other through gap junctions has beennamed ‘bystander death’. This term was coined by Free-man et al.25 to describe the findings obtained with a par-ticular type of gene therapy for primary brain tumors.In their study, the thymidine kinase gene from herpessimplex virus (HSV-tk) was used in combination withthe nucleoside analogue, ganciclovir, thereby enzymati-cally converting the non-toxic prodrug into a lethal com-pound that inhibited the growth of gene transfected cells.Because of the low efficacy of viral vectors, only a frac-tion of the tumor mass contained the viral gene, but thetransfection resulted in cell death throughout the tumor.The therapeutically destructive effect induced in the suc-cessfully transfected cells had spread within the tumorto ‘bystanders’ cells that otherwise would have survivedand multiplied. Recent studies26–30 reinforce the obser-vations about the “bystander effect” made by Freeman etal.25 It has been shown that initiation of apoptosis whichis unsynchronized in serum-deprived cultures of a normaldiploid epithelial cell line (WB-F344)31 and granulosacell explants32 was accompanied with an overall increasein the level of gap junctional coupling in the cell popula-tion. It is remarkably that dying cells may still commu-nicate with healthy cells. In this regard, Wilson et al.31

studied gap junctional coupling of individual cells withdifferent apoptotic morphology and demonstrated thatalthough coupling of the dying cells gradually was de-creasing during the progression towards the late stagesof apoptosis, that early apoptotic cells were still coupledto healthy cells. This suggests that uncoupling of gapjunction channels is not an essential requirement priorto programmed cell death. Since cell coupling was notdisrupted prior to formation of apoptotic bodies,31 gapjunction channels may, serve a regulatory function duringthe initiation of apoptosis. In line with this idea Cotrinaet al.33 showed that astrocytes exposed to cell death stimuli(KCN and iodoacetat) remain coupled to adjacent viablecells until the dying cells lost the integrity of their plasmamembrane (visualized with propidium iodide, trypan blueexclusion, and loss of 5-(and) 6-carboxyfluorescein stain-ing). In this model system death proceeded according toa schedule that could not be characterized by the classicalcriteria as either typically necrotic or typically apoptotic.The work of Kalvelyte et al.34 demonstrated that transfec-

28 Introduction: Part II

tion of HeLa cells with Cx43 fused with enhanced greenfluorescence protein accelerates the progression from anapoptotic to a secondary necrotic stage, and that this pro-gression was inhibited by octanol, a gap junction channeland hemichanel blocker. This reflects the modulator role ofthese channels in cell death progression. It is conceivablethat more rapid osmotic swelling through hemichannel-mediated water import is implicated in secondarynecrosis.

The observation that only communicating cells die inclusters33,35 supports the notion that cell fate modula-tors may pass through gap junction channels. The oc-currence of apoptotic clusters may reflect either simulta-neous passage of a group of cells through differentiationwith synchronous entry into apoptosis36,37 or killing ofviable neighbor cells by primary apoptotic cells propagat-ing the cell fate modulator via gap junctions. However,the mechanism does not exclude the influence of extracel-lular modulating signals on the development of clusters.Gap junction-independent clustering of apoptotic cellsdue to the release of toxic compounds from an initiallysingle dying cell into the culture medium, has been re-ported, and hydrogen peroxide was suggested as the deathsignal.38

In addition, the idea that gap junctions mediate thepropagation of cell fate modulating signals is also sup-ported indirectly. Tumor promoters such as phenobarbi-tal, phorbol esters, peroxisome proliferators, which mayinhibit apoptosis can block gap junction channels. On theother hand tumor-suppressing chemicals such as retinoids,dexamethasone, which may augment apoptosis, can en-hance the permeability of gap junction channels.39,40 Fur-thermore, gap junctions mediate the spreading of celldeath in various insults including transient or permanentischemia.41 Gap junctions enhance neuronal vulnerabilityto brain traumatic injury42 and glutamate cytotoxicity.43

Neuronal cell death in hippocampal slice cultures 48 h af-ter oxygen/glucose deprivation is less pronounced in Cx43knockout mice than in slices from wild type mice.44 Inaddition, treatment with antisense oligodeoxynucleotidesfor Cx26 and Cx32 or for Cx43 reduces cell death causedby oxygen/glucose deprivation, and gap junction blockercarbenoxolone is neuroprotective.44 The infusion of gapjunction blockers limited the infarct size after focal orglobal brain ischemia in rodents.45–47

It is important to note that the mediators of bystanderdeath could be exogenous factors (e.g. ganciclovir, an ex-ogenous prototoxin) as well as compounds generated bythe dying cells as intermediates and second messengers inthe apoptotic process. The key question is what are thepossible signals that could pass through gap junctions andmediate either apoptosis or survival? This is obviously adifficult question to answer experimentally because thereare so many molecules and ions capable of passing throughgap junction channels, which could function both either

as a survival or death modulator depending on the cel-lular context. Cell fate modulators such as Ca2+, IP3,ATP, cAMP, which by traversing through gap junctionchannels or hemichannels may modulate cell death of theneighboring cells, might also act as survival modulators.Whether these cell fate modulators would act, as deathor survival signal probably depends on the cell type, andon the status and environmental context of the cell thatreceives the signal. In this regard the condition of the re-ceiving cell determines the outcome of the event: whetherthe signal transmitted through gap junction channels orhemichannels will favor survival or cell death. If the cellconcomitantly receives stress stimuli such as growth fac-tor deprivation, hypoxia, than the “survival” modulatorsmay turn into death modulators. The idea that proteinsand second messengers can have different functions de-pending on the cellular context and the time where theyact is well-known, e.g. Ca2+is a modulating signal for bothcell life and death. Ca2+ triggers a new life at fertilization,and controls the development and differentiation of thecells into specialized types, but it is also involved in celldeath in a well defined cellular context.48–50 Similar mod-ulating effects on cell life and death may also be exercisedby IP3, cAMP, ATP51–61.

Ca2+ may be one of the modulating signals, whichmight participate in the transduction of a cell death signalfrom one cell to another through gap junction channels.It is well known that Ca2+-dependent processes are in-terconnected with the mainstream apoptosis executioners– caspases and recent findings indicate that the accumu-lation of Ca2+ in the mitochondria through the seques-tration of Ca2+ uptake by the endoplasmic reticulum,can trigger the opening of the permeability transitionpores, with consequent release of apoptogenic mitochon-drial factors.50 Lin et al.62 showed that glioma cells thatresisted to apoptosis due to the enforced expression ofBcl-2 could nevertheless be killed by various types of in-jury (metabolic inhibition, oxidative stress and calciumoverload), when they were coupled via gap junctions withvulnerable non-transfected counterparts. The authors pro-posed that dysregulation of Ca2+ homeostasis may havea role in a ‘natural’ bystander death, but this does notimply that Ca2+ itself is the cell fate modulator. Thestudy of Rawanduzy et al.45 points towards a role of gapjunctions in spreading a cell death message because gapjunction blockers, in addition to reducing the infarct vol-ume, also blocked waves of spreading depression (slowlypropagating depolarization waves in the brain cortex). Re-cent data however indicate that the spreading depressionwave utilizes different propagation mechanisms as com-pared to intercellular Ca2+ waves,63,64 precluding firmconclusions on the role of Ca2+ as the cell fate modulatorfrom this study. Krutovskikh et al. 35proposed that the cal-cium ions themselves may function as an intercellular “celldeath message”. The authors demonstrated in a rat bladder

Gap junctions in cell survival & cell death 29

Figure 2.

Figure 3.


PlasmamembraneM4M3M2M1

E1 E2

CL CT

COOHNH2

Intracellular

Extracellular

Lipid bilayer

Lipid bilayer

gap of2-4 nm

Homotypic

two connexons forming an openchannel between adjacent cells

A:

Connexins

HeteromericHomomeric Heterotypic

C:

B:

Intercellular channelsConnexons

ERKs

MEK

SrcConnexins

Ca2+

APOPTOSIS SURVIVAL

Cyt c

Chemotherapeuticdrugs

Gap junction channelblockers

Gap junction channelblockers

Apoptoticstimuli

Apoptoticstimuli

Survivalsignals

?

ER

IPR

3

Ca

2+

Ca

2+

ER

IPR

3

Ca2

+

Ca2

+

cAMP

Bisphosphonates

Agents increasingproduction of cAMP Connexin hemichannel

1

3

8

4

5

6

7

9

10

11

13 12

NAD+

ATP

?

?

Bcl-2IP3

2

glutamate

4

3

SS

Ca2+

?

Figure 2. A: Schematic representation of a limited part of a gap junctional plaque. The diagram shows complete connexons composed ofhexamers. B: Possible arrangements of connexons to form gap junction channels. Connexons, which consist of six connexin subunits maybe homomeric (composed of six identical connexins subunits) or heteromeric (composed of more than one connexin isotype). Connexinsassociate end-to-end to form a double membrane gap junctions channel. The channel may be homotypic (if connexons are identical)or heterotypic (if the two connexons are different). C: Topology of connexins molecules. The carboxy- and amino-terminal domains (CTand NT, respectively) face the cytoplasm, as does the hydrophilic cytoplasmic loop (CL) domain between M2 and M3. The hydrophilicdomains between M1 and M2 and between M3 and M4 form two extracelullar loops, E1 and E2. The major cytoplasmic domains (i.e.CL and CT) are highly variable among the different Cxs and unique in both sequence and length, whereas the four transmembranesegments (M1–M4) and both E1 and E3 are strongly conserved.9

Figure 3. Gap junctions and the propagation of cell survival and cell death modulating signals. After stimulation with apoptosis inducingagents like the Fas ligand or Simvastatin (1), there is an increase in the IP3 production. One possibility is that IP3 triggers the releaseof Ca2+ from Ca2+ stores (2) such as the endoplasmic reticulum (ER), and that the, released Ca2+ passes through gap junctionsto induce apoptosis in the neighboring coupled cell (3). Another possibility is that IP3 itself moves through gap junctions and inducesapoptosis of the adjacent cell by triggering Ca2+ release from the stores (4). The involvement of Ca2+ and IP3 diffusion is discussed inthe text. Apoptotic stimuli like. staurosporine and ceramide induce the release of cytochrome c (cyt c), which binds to the IP3 receptor(IP3R) and blocks the Ca2+-dependent inhibition of this receptor (5). The released Ca2+ triggers additional cyt c release in a positivefeedback loop that may amplify the magnitude of the signal and spread of apoptosis (6). Caspase 3 may further promote Ca2+ releasefrom the stores by cleaving the IP3R and transforming it to an open channel (not shown). Increases of cAMP have also been observedwith some apoptotic agents (cholera toxin, forskolin, luteinizing hormone)122,123 and these changes can be transmitted to neighboringcells via gap junction channels and may be involved in spreading apoptosis (7). Hemichannels contribute to the release of differentsignaling molecules such as ATP, NAD+, glutamate that may consequently induce apoptosis of the adjacent cell (8). Expression ofconnexins increases the sensitivity to several common chemotherapeutic drugs such as etoposide, paclitaxel (Taxol

©R) and doxorubicin

(9). These connexin-mediated effects are associated with a decreased expression of Bcl-2 and are not dependent on gap junction channelfunctioning. The bisphosphonates (e.g. alendronate) induce conformational changes of connexin hemichannels with the subsequentactivation of Src, culminating in ERK phosphorylation and cell survival (10). Under certain conditions passage of cell fate modulators(SS) may affect survival (11). In this case gap junction channel blockers will induce apoptosis (12). Agents that block gap junctionchannels and inhibit the passage of cell death messages promote survival (13). This proposed scheme is summarized from the followingreferences.45–47,50–54,56,57,70–74,76,85,86,98,100–105,111

carcinoma cell line that reduction of gap junction per-meability with oleamide, did not affect the passage ofCa2+ ions and did not abrogate coordinated cell deathby clusters. This does however not mean that oleamideselectively restricts gap junction permeability to Ca2+; itrather means that pathways other than gap junctions, e.g.paracrine communication, are involved in the spread ofCa2+ signals. In line with this, recent data indicate thatpropagation of a Ca2+signal between cells is not only dueto the direct passage of Ca2+ through gap junctions butalso to the passage of other second messengers able to in-duce an increase in cytosolic Ca2+ concentration in the ad-jacent cells.65 Although Ca2+ can pass through gap junc-tions, their role as an intercellular messenger is probablylimited66 because Ca2+ diffusion in the cytosol is severelyhindered by binding to more slowly diffusing cytoplasmicCa2+ binding proteins, which makes their effective diffu-sion constant much lower as compared to that of a secondmessenger like inositol 1,4,5-trisphosphate (IP3).67,68 Astill somewhat controversial point is the apparent con-tradiction between Ca2+ signal propagation via IP3 dif-fusion through gap junctions and the well-known factthat an increase in intracellular Ca2+ concentration trig-gered by IP3 can block gap junctions. This blocking effecthowever depends on the magnitude, the duration and thesource of the calcium increase.69 IP3 is a second messen-ger produced primarily by phospholipase C metabolismof phosphoinositol-4,5-bisphosphate in response to thestimulation of G-protein-coupled receptors or receptors

tyrosine kinases. IP3 (molecular weight (MW) 414 inthe free anion from) can permeate through gap junctionchannels70,71 and triggers the rapid release of Ca2+ fromthe endoplasmic reticulum and other cellular membranesby binding to the IP3 receptor which amplifies/transducescellular signals, many of which are generated at the plasmamembrane.72 There is strong evidence pointing to a pos-sible role of IP3 as an apoptosis amplifying messengerthat can permeate through the gap junction channels. Inthis regard it has been reported that Fas-activation causesIP3 accumulation with subsequent damage of myocytes.51

Interestingly it has been shown that simvastatin (a ly-pophilic form of 3-hydroxy-3-methylglutaryl-coenzymeA reductase inhibitor) is cytotoxic to L6 myoblasts.52 Theauthors demonstrated that simvastatin induces tyrosinephosphorylation of phospholipase C (PLC) γ 1. ActivatedPLC-γ 1 catalyzes IP3 formation and thereby triggersCa2+ mobilization from the stores with consequent celldeath. These data are consistent with reports that PLC-γ 1 activation by tyrosine phosphorylation is necessary forsurface immunoglobulin M-induced B cell apoptosis.73

Moreover it was shown that IP3-induced Ca2+ spikescause permeability transition pore (PTP) opening, cy-tochrome c release, caspase activation and nuclear apopto-sis in cells exposed to proapoptotic stimuli.74 IP3-inducedopening of PTP is dependent on a Ca2+ signal trans-mission from IP3 receptors of endoplasmic reticulumto mitochondria (existence of mitochondria-ER calciumsynapse).74 Ca2+ release from the endoplasmic reticulum


in response to IP3 can be regulated by the anti-apoptoticprotein Bcl-2.75 Thus, IP3-mediated Ca2+ release can alsoserve as an efficient and selective activator of the mito-chondrial phase of the apoptotic process. Besides this, re-cent work has demonstrated that cytochrome c releasedby PTP opening binds to the IP3 receptor, blocks thewell-documented calcium-dependent inhibition of thisreceptor and thereby further amplifies calcium-relatedapoptosis.76 IP3 thus appears to be able to trigger cy-tochrome c release which on its turn activates a cascadethat results in more cytochrome c release,76 effectivelysetting up a vicious circle leading to apoptotic cell death.The recently reported truncation of IP3 type-1 receptorsby caspase-3 may add to this vicious circle.77

Another candidate second messenger is 3′ ,5′ cyclicadenosine monophosphate (cAMP), which with a MWof 329, can pass through gap junctions.78–80 Differentcell types, such as primary granulosa cells or neuronalcells undergo apoptosis53–55 after treatment with agentsthat elevate intracellular cAMP. Whether cAMP by pass-ing through gap junction channels can indeed function ascell fate modulator is unknown and additional studies arerequired to define this role of cAMP.

Experimental data suggest that gap junctions may al-low cell-to-cell propagation of cell death secondary toischemia-reperfusion in cardiac muscle cells. Reperfusedmyocardium infarcts consist almost exclusively of areasof contraction band necrosis formed by hypercontracteddead myocytes.81 Moreover, these authors showed thatpassage of Na+ through gap junctions from hypercon-tracting cell to adjacent ones, and subsequent exchangewith Ca2+ through reverse mode of Na/Ca exchange, re-sults in propagation of hypercontracture.82 In addition,it was demonstrated that this propagation contributessignificantly to the final infarct size, since its inhibitionwith intracoronary infusion of heptanol, a gap junctioninhibitor, reduced infarct size.83 Although gap junctionchannels have been shown to allow efficient sodium move-ment between cells84 additional studies are needed to in-vestigate in depth the role of Na+ in propagation of hy-percontracture and cell death.

Finally, in spite of the accumulating evidence for a roleplayed by gap junction channels in propagating cell deathunder various insults (discussed above) it is of note thatgap junction channels may mediate survival or be pro-tective for the adjacent cell. In mixed astrocyte neuroncultures, inhibition of astrocyte coupling with gap junc-tion blockers increases neuronal vulnerability to oxidativestress85 or glutamic acid toxicity.86 Cx32 and heterozy-gous Cx43 knockout mice showed enhanced sensitivityto cerebral ischemia, suggesting gap junction mediatedneuroprotection.86–88 Furthermore, results of Leung etal.89 demonstrated that fibroblast growth factor 2 up-regulates Cx43 in rat embryonic midbrain cultures andpromotes survival of dopaminergic neurons; oleamide-

induced cell uncoupling abolishes this survival promotingeffect. Recently it was demonstrated that functional gapjunctions are required for the survival of neural progeni-tor cells.90 Tazuke et al.91 showed that the zero populationgrowth locus of Drosophila encodes a germline-specific gapjunction protein, Innexin 4, is required for survival of dif-ferentiating early germ cells during gametogenesis in bothsexes. Thus, like other important mechanisms involved incell regulation, gap junction channels may have a life ordeath role in cell pathophysiology.

Other key questions are whether connexins distin-guish between modulating signals and whether any suchselectivity affects cell fate in either positive or negativeway. It has been reported that the restoration of normalcell growth of tumor cells in vivo depends on the Cx typetransfected.92 Recent work has also illustrated that thecomplexity of the Cx family may have evolved to servedifferent functions via transfer of molecules specific to aparticular Cx isotype. In support of this, in a study de-signed to determine what molecules pass through gapjunction channels, cell expressing Cx43 were shown toexchange ATP 300-fold and ADP and AMP 8-fold bet-ter than Cx32-expressing cells.93 Cells expressing Cx32exchanged adenosine 12-fold better that those containingCx43, which can not be easily explained by size and chargedifferences of the permeating molecules.93 Niessen et al.71

furthermore showed that permeability of Cx32 gap junc-tions to IP3 was also significantly above that of Cx26 andCx43. Thus, differences in the selectivity for second mes-sengers may play an important role in the extent to whichthe cell fate modulating signals are spread. Furthermorefunctional replacement of one gap junction channel genewith another in mice and flies demonstrated that cellularresponses require the correct type of intercellular channelsubunit. This suggest that the intercellular communica-tion is finely tuned to the physiological needs of a specifictissue and that the diverse gap junction channels are notfunctionally redundant.94

The most surprising finding of the past few years wasthat connexins have critical functions that do not involveintercellular channels. Recent studies demonstrated thatconnexins independent of their contribution to gap junc-tions are important in different cellular processes23,95–97

including the mediation of cell survival and death. Theroles of connexin that are independent of the gap junc-tion can be divided into two categories: first, mediatedby unopposed hemichannels and, second, connexin sub-units might have direct signal transduction activity bythemselves. These two possibilities differ fundamentally:hemichannel-mediated diffusion of intercellular messen-gers is attenuated by gap junction inhibitors but con-nexin mediated survival/apoptosis does not require nei-ther gap junction formation nor hemichannels and is thusinsensitive to uncoupling agents. This aspect will be con-sidered next.


Hemichannels in cell survival and cell death

Properties of hemichannels may well be crucial to the sur-vival of connexin expressing cells. In fact Cx43 hemichan-nels have been shown recently to play a role in the anti-apoptotic actions of bisphosphonates.98 Bisphosphonatesare stable analogs of pyrophoshates, which prevent os-teocyte and osteoblast apoptosis induced by glucocorti-coids. Bisphosphonates are widely used in the treatmentof bone disease and require activation of the extracel-lular signal-regulated kinases (ERKs). Plotkin et al.98

suggested that bisphosphonates induced transient open-ings of nonjunctional Cx43 hemichannels, resulting ina conformational change in Cx43 and subsequent ac-tivation of Src. Increased Src kinase activity results inactivation of the mitogen-activated protein kinase cas-cade, which culminates in ERK phosphorylation and cellsurvival.

In addition to the role of hemichannels in survival (dis-cussed above) it is important to note that hemichannelesmay play a role also in apoptosis. Hur et al.99 demon-strated that HeLa cells transfected with full-length Cx43(HeLa-Cx43) were more susceptible to staurosporine (abroad spectrum kinase inhibitor) treatment than HeLawild type and HeLa expressing C-terminal truncated formof Cx43. Moreover they showed that hemichannels of un-apposed HeLa-Cx43 cells open upon exposure to stau-rosporine, which might indicate that hemichannels mightbe involved in the release of modulating molecules thatact in a paracrine manner to propagate a death message.Several reports have indeed documented that hemichan-nels form a conductive pathway for the release of smallmolecules like ATP100–102 (MW 507), glutamate103 (MW292) and NAD+ (MW 663).104 Messengers like gluta-mate are since long known to trigger excitotoxic necrosisof susceptible neurons when released in an uncontrolledmanner and apoptosis has been recently added to the reper-toire of excess glutamate actions.105 Uncontrolled releaseof ATP might also result in induction of apoptosis ei-ther through the P2X7 purinergic receptor that forms socalled “permeabilizing pores56,106 or through the rapidaccumulation of adenosine the end-product of ATP degra-dation. Adenosine has been shown to induce apoptosis inthe neuroblastoma cell line N1E-115.57 Extensive releaseof ATP (1–2% of the total cellular content per stimulusevent) via hemichannels has been reported.107 In case ofrepeated stimulation, this can significantly influence theintracellular ATP concentration which is known as deci-sion point from apoptosis to necrosis.108 Taken togetherthese studies indicate that hemichannels may provide apathway for releasing modulating signals but additionalstudies should be undertaken in order to validate whetherthese released modulators could indeed spread cell death.

Connexins outside gap junction channels andhemichannels also affect cell survival and celldeath

Huang et al.109 demonstrated that transfection of Cx43into human glioblastoma cell lines (U251 and T98G) pro-foundly reduced cell proliferation and this effect was notassociated with the establishment of gap junction channelsin Cx43 transfected cells. In addition Qin et al.110 demon-strated in human breast tumor cells that Cx43 and Cx26induce their tumor-suppressing properties by a mecha-nism that is independent of significant gap junction chan-nels inhibition. Huang et al.111 showed that induction ofapoptosis by the tumor–suppressor gene, Cx43 in humanglioblastoma cells in response to chemotherapeutic drugsis mediated by down regulation of the apoptosis inhibitorprotein Bcl-2 and may be due to Cx43 action that is notdirectly related to gap junctional channel formation. Themechanism for the regulation of Bcl-2 expression by Cx43is currently unknown. The fact that transfected Cx43 waspredominantly localized in the nucleus109,112 indicate thatCx43 may directly regulate gene expression through bind-ing to cis elements in the promoter regions of regulatedgenes. Moreover, it has been reported that Cx43 is lo-calized in the nucleus and can bind to DNA, suggest-ing that it has distinct functions from its well-knowninvolvement in gap junctional communication.113 Alter-natively, the down regulation of Bcl-2 may result fromsignal transduction through secondary, down stream ele-ments since Cx43 exhibits SH2 and SH3 as well as ZO1binding sites.111,114–116

Another channel-independent function of Cxs is re-lated to resistance to injury. Recently, it was shown thatthe forced expression of Cx43 as well as two other mem-bers of the Cx family (Cx32 and Cx40) protects culturedglial cells against cell injury.117 Surprisingly, the pro sur-vival activity of these Cx proteins was not eliminated inlow-dense cultures or by using Cx channel blockers, sug-gesting that the survival function of Cxs is independentfrom function of gap junction channels. Furthermore, ex-ogenous expression of functionally inactive mutants of Cxsalso induced resistance to injury.117 It is important to notethat the protective effect could be related with a functionof hemichannels since their activity was not explicitly ex-amined in this study. Moreover it has been demonstratedusing cDNA microarrays that expression of pro-apoptoticand anti-apoptotic genes is affected in Cx43 null astro-cytes, which exhibited reduced growth rate in culture.118

The above-mentioned data indicate that Cxs, may partic-ipate in cell survival and cell death independently fromgap junctions and hemichannels, but additional studiesare required in order to discover the possible signal trans-duction pathways.


Conclusions and perspectives

It is important to point out that most studies use phar-macological inhibitors of gap junction channels andhemichannels in order to verify whether cell couplingis a means of propagating cell fate modulating signals.Indeed a broad spectrum of agents capable of disruptinggap junction channels are available but for the most partthey are not Cx specific and have limited efficacy in theconcentrations required to induce channel closure. Thenext goals include demonstration of involvement of gapjunction channels, hemichannels and non-channel activ-ities in the spreading of cell fate modulating signals invivo, in intact tissues. In order to work in vivo there is theneed to identify channels blockers that are more efficaciousfor producing long-lasting junctional inhibition, and todetermine by what route they should be administratedso as not to affect other ion channels, gap junctions orhemichannels in the brain, heart, liver, kidney and othervital organs. New methods of separating the intercellu-lar channels, hemichannels and non-channel activities ofCxs need to be developed. In this direction recently amutant connexin that induces hemichannel activity, butnot functional gap junction channels was reported.119 An-other possible approach might be the use of Cx mimeticpeptides, which when applied over a short time period,appear to be efficient blockers of hemichannels withoutaffecting gap junction channels.65,102,120 Further explo-ration of gap junction channels, hemichannels and Cxsfunction in the future will require development of morespecific approaches such as inducible dominant negativemutants, transgenic animal models with multiple con-nexin knockouts and mutant connexin knockins in orderto study their role in the propagating of cell fate modu-lating signals.

Mounting evidence reviewed here indicates that Cxsplay a role in determining the fate of a cell at three levels(summarized in the Figure 3): (1) the contribution of Cxsto gap junction channels implicate, that cells are coupledto neighboring cells and can receive cell fate modulatingsignals affecting either cell death or survival; (2) the con-tribution of Cxs to hemichannels implicate that cells canrelease to their environment signals which modulate cel-lular death or survival; (3) the expression of connexins perse might influence signal transduction pathways.

However, the major question concerning identifi-cation of the modulating molecules transferred bygap junctions/hemichannels and the biological pro-cesses they influence remains open. The identificationof modulating molecules transferred through gap junc-tions/hemichannels that are critical for promoting ei-ther cell survival or cell death is an exciting areawith many implications for future basic and clinicalresearch.

Acknowledgments

We thank Wim Drijvers for the artwork and HubertStevens for excellent technical assistance.

References

1. Hengartner MO. The biochemistry of apoptosis. Nature 2000;407: 770–776.

2. D’Herde K, De Prest B, Mussche S, et al. Ultrastructurallocalization of cytochrome c in apoptosis demonstrates mito-chondrial heterogeneity. Cell Death Differ 2000; 7: 331–337.

3. Krysko DV, Roels F, Leybaert L, D’Herde K. Mitochondrialtransmembrane potential changes support the concept of mi-tochondrial heterogeneity during apoptosis. J Histochem Cy-tochem 2001; 49: 1277–1284.

4. Saelens X, Festjens N, Vande Walle L, van Gurp M, van LooG, Vandenabeele P. Toxic proteins released from mitochondriain cell death. Oncogene 2004; 23: 2861–2874.

5. Kanno Y, Loewenstein WR. Intercellular Diffusion. Science1964; 143: 959–960.

6. Dewey MM, Barr L. A Study of the Structure and Distributionof the Nexus. J Cell Biol 1964; 23: 553–585.

7. Revel JP, Karnovsky MJ. Hexagonal array of subunits in in-tercellular junctions of the mouse heart and liver. J Cell Biol1967; 33: C7–C12.

8. Phelan P, Starich TA. Innexins get into the gap. Bioessays 2001;23: 388–396.

9. Willecke K, Eiberger J, Degen J, et al. Structural and func-tional diversity of connexin genes in the mouse and humangenome. Biol Chem 2002; 383: 725–737.

10. Laird DW, Puranam KL, Revel JP. Turnover and phosphoryla-tion dynamics of connexin43 gap junction protein in culturedcardiac myocytes. Biochem J 1991; 273(Pt 1): 67–72.

11. Goodenough DA, Goliger JA, Paul DL. Connexins, connex-ons, and intercellular communication. Annu Rev Biochem 1996;65: 475–502.

12. Munari-Silem Y, Rousset B. Gap junction-mediated cell-to-cell communication in endocrine glands–molecular and func-tional aspects: A review. Eur J Endocrinol 1996; 135: 251–264.

13. Trosko JE, Ruch RJ. Cell-cell communication in carcinogen-esis. Front Biosci 1998; 3: d208–d236.

14. Bruzzone R, White TW, Paul DL. Connections with connex-ins: the molecular basis of direct intercellular signaling. EurJ Biochem 1996; 238: 1–27.

15. Bruzzone R, White TW, Goodenough DA. The cellular In-ternet: On-line with connexins. Bioessays 1996; 18: 709–718.

16. Musil LS, Goodenough DA. Multisubunit assembly of an in-tegral plasma membrane channel protein, gap junction con-nexin43, occurs after exit from the ER. Cell 1993; 74: 1065–1077.

17. Martin PE, Blundell G, Ahmad S, Errington RJ, Evans WH.Multiple pathways in the trafficking and assembly of connexin26, 32 and 43 into gap junction intercellular communicationchannels. J Cell Sci 2001; 114: 3845–3855.

18. Bukauskas FF, Verselis VK. Gap junction channel gating.Biochim Biophys Acta 2004; 1662: 42–60.

19. White TW, Bruzzone R, Paul DL. The connexin family ofintercellular channel forming proteins. Kidney Int 1995; 48:1148–1157.

20. Kumar NM, Gilula NB. The gap junction communicationchannel. Cell 1996; 84: 381–388.


58. Seite P, Ruchaud S, Hillion J, et al. Ectopic expression of Bcl-2switches over nuclear signalling for cAMP-induced apopto-sis to granulocytic differentiation. Cell Death Differ 2000; 7:1081–1089.

59. Schulte G, Fredholm BB. Signalling from adenosine receptorsto mitogen-activated protein kinases. Cell Signal 2003; 15:813–827.

60. Conti M. Specificity of the cyclic adenosine 3’,5’-monophosphate signal in granulosa cell function. Biol Reprod2002; 67: 1653–1661.

61. Han PJ, Shukla S, Subramanian PS, Hoffman PN. Cyclic AMPelevates tubulin expression without increasing intrinsic axongrowth capacity. Exp Neurol 2004; 189: 293–302.

62. Lin JH, Weigel H, Cotrina ML, et al. Gap-junction-mediatedpropagation and amplification of cell injury. Nat Neurosci1998; 1: 494–500.

63. Peters O, Schipke CG, Hashimoto Y, Kettenmann H. Differ-ent mechanisms promote astrocyte Ca2+ waves and spread-ing depression in the mouse neocortex. J Neurosci 2003; 23:9888–9896.

64. Theis M, Jauch R, Zhuo L, et al. Accelerated hippocampalspreading depression and enhanced locomotory activity inmice with astrocyte-directed inactivation of connexin43. JNeurosci 2003; 23: 766–776.

65. Braet K, Vandamme W, Martin PE, Evans WH, LeybaertL. Photoliberating inositol-1,4,5-trisphosphate triggers ATPrelease that is blocked by the connexin mimetic peptide gap26. Cell Calcium 2003; 33: 37–48.

66. Churchill GC, Louis CF. Roles of Ca2+, inositol trisphosphateand cyclic ADP-ribose in mediating intercellular Ca2+ sig-naling in sheep lens cells. J Cell Sci 1998; 111( Pt 9): 1217–1225.

67. Allbritton NL, Meyer T, Stryer L. Range of messenger actionof calcium ion and inositol 1,4,5-trisphosphate. Science 1992;258: 1812–1815.

68. Sneyd J, Wetton BT, Charles AC, Sanderson MJ. Intercellularcalcium waves mediated by diffusion of inositol trisphosphate:A two-dimensional model. Am J Physiol 1995; 268: C1537–1545.

69. Lazrak A, Peracchia C. Gap junction gating sensitivity tophysiological internal calcium regardless of pH in Novikoffhepatoma cells. Biophys J 1993; 65: 2002–2012.

70. Saez JC, Connor JA, Spray DC, Bennett MV. Hepatocyte gapjunctions are permeable to the second messenger, inositol1,4,5-trisphosphate, and to calcium ions. Proc Natl Acad SciUSA 1989; 86: 2708–2712.

71. Niessen H, Harz H, Bedner P, Kramer K, Willecke K. Selec-tive permeability of different connexin channels to the secondmessenger inositol 1,4,5-trisphosphate. J Cell Sci 2000; 113(Pt 8): 1365–1372.

72. Patterson RL, Boehning D, Snyder SH. Inositol 1,4,5-trisphosphate receptors as signal integrators. Annu Rev Biochem2004; 73: 437–465.

73. Takata M, Homma Y, Kurosaki T. Requirement of phospho-lipase C-gamma 2 activation in surface immunoglobulin M-induced B cell apoptosis. J Exp Med 1995; 182: 907–914.

74. Szalai G, Krishnamurthy R, Hajnoczky G. Apoptosis drivenby IP(3)-linked mitochondrial calcium signals. Embo J 1999;18: 6349–6361.

75. Chen R, Valencia I, Zhong F, et al. Bcl-2 functionally inter-acts with inositol 1,4,5-trisphosphate receptors to regulatecalcium release from the ER in response to inositol 1,4,5-trisphosphate. J Cell Biol 2004; 166: 193–203.

76. Boehning D, Patterson RL, Sedaghat L, Glebova NO,Kurosaki T, Snyder SH. Cytochrome c binds to inositol (1,4,5)

trisphosphate receptors, amplifying calcium-dependent apop-tosis. Nat Cell Biol 2003; 5: 1051–1061.

77. Assefa Z, Bultynck G, Szlufcik K, et al. Caspase-3-inducedtruncation of type 1 inositol trisphosphate receptor acceler-ates apoptotic cell death and induces inositol trisphosphate-independent calcium release during apoptosis. J Biol Chem2004; 279: 43227–43236.

78. Lawrence TS, Beers WH, Gilula NB. Transmission of hor-monal stimulation by cell-to-cell communication. Nature1978; 272: 501–506.

79. Murray SA, Fletcher WH. Hormone-induced intercellularsignal transfer dissociates cyclic AMP-dependent protein ki-nase. J Cell Biol 1984; 98: 1710–1719.

80. Bedner P, Niessen H, Odermatt B, Willecke K, Harz H. Amethod to determine the relative cAMP permeability of con-nexin channels. Exp Cell Res 2003; 291: 25–35.

81. Solares J, Garcia-Dorado D, Oliveras J, et al. Contraction bandnecrosis at the lateral borders of the area at risk in reperfusedinfarcts. Observations in a pig model of in situ coronary oc-clusion. Virchows Arch 1995; 426: 393–399.

82. Ruiz-Meana M, Garcia-Dorado D, Hofstaetter B, Piper HM,Soler-Soler J. Propagation of cardiomyocyte hypercontractureby passage of Na(+) through gap junctions. Circ Res 1999;85: 280–287.

83. Garcia-Dorado D, Inserte J, Ruiz-Meana M, et al. Gap junc-tion uncoupler heptanol prevents cell-to-cell progression ofhypercontracture and limits necrosis during myocardial reper-fusion. Circulation 1997; 96: 3579–3586.

84. Rose CR, Ransom BR. Gap junctions equalize intracellularNa+ concentration in astrocytes. Glia 1997; 20: 299–307.

85. Blanc EM, Bruce-Keller AJ, Mattson MP. Astrocytic gap junc-tional communication decreases neuronal vulnerability to ox-idative stress-induced disruption of Ca2+ homeostasis and celldeath. J Neurochem 1998; 70: 958–970.

86. Ozog MA, Siushansian R, Naus CC. Blocked gap junc-tional coupling increases glutamate-induced neurotoxicity inneuron-astrocyte co-cultures. J Neuropathol Exp Neurol 2002;61: 132–141.

87. Siushansian R, Bechberger JF, Cechetto DF, Hachinski VC,Naus CC. Connexin43 null mutation increases infarct sizeafter stroke. J Comp Neurol 2001; 440: 387–394.

88. Nakase T, Sohl G, Theis M, Willecke K, Naus CC. Increasedapoptosis and inflammation after focal brain ischemia in micelacking connexin43 in astrocytes. Am J Pathol 2004; 164:2067–2075.

89. SiuYi Leung D, Unsicker K, Reuss B. Gap junctions modulatesurvival-promoting effects of fibroblast growth factor-2 oncultured midbrain dopaminergic neurons. Mol Cell Neurosci2001; 18: 44–55.

90. Cheng A, Tang H, Cai J, et al. Gap junctional communicationis required to maintain mouse cortical neural progenitor cellsin a proliferative state. Dev Biol 2004; 272: 203–216.

91. Tazuke SI, Schulz C, Gilboa L, et al. A germline-specific gapjunction protein required for survival of differentiating earlygerm cells. Development 2002; 129: 2529–2539.

92. Mesnil M, Krutovskikh V, Omori Y, Yamasaki H. Role ofblocked gap junctional intercellular communication in non-genotoxic carcinogenesis. Toxicol Lett 1995; 82–83: 701–706.

93. Goldberg GS, Moreno AP, Lampe PD. Gap junctions betweencells expressing connexin 43 or 32 show inverse permselec-tivity to adenosine and ATP. J Biol Chem 2002; 277: 36725–36730.

94. White TW. Nonredundant gap junction functions. News Phys-iol Sci 2003; 18: 95–99.


58. Seite P, Ruchaud S, Hillion J, et al. Ectopic expression of Bcl-2switches over nuclear signalling for cAMP-induced apopto-sis to granulocytic differentiation. Cell Death Differ 2000; 7:1081–1089.

59. Schulte G, Fredholm BB. Signalling from adenosine receptorsto mitogen-activated protein kinases. Cell Signal 2003; 15:813–827.

60. Conti M. Specificity of the cyclic adenosine 3’,5’-monophosphate signal in granulosa cell function. Biol Reprod2002; 67: 1653–1661.

61. Han PJ, Shukla S, Subramanian PS, Hoffman PN. Cyclic AMPelevates tubulin expression without increasing intrinsic axongrowth capacity. Exp Neurol 2004; 189: 293–302.

62. Lin JH, Weigel H, Cotrina ML, et al. Gap-junction-mediatedpropagation and amplification of cell injury. Nat Neurosci1998; 1: 494–500.

63. Peters O, Schipke CG, Hashimoto Y, Kettenmann H. Differ-ent mechanisms promote astrocyte Ca2+ waves and spread-ing depression in the mouse neocortex. J Neurosci 2003; 23:9888–9896.

64. Theis M, Jauch R, Zhuo L, et al. Accelerated hippocampalspreading depression and enhanced locomotory activity inmice with astrocyte-directed inactivation of connexin43. JNeurosci 2003; 23: 766–776.

65. Braet K, Vandamme W, Martin PE, Evans WH, LeybaertL. Photoliberating inositol-1,4,5-trisphosphate triggers ATPrelease that is blocked by the connexin mimetic peptide gap26. Cell Calcium 2003; 33: 37–48.

66. Churchill GC, Louis CF. Roles of Ca2+, inositol trisphosphateand cyclic ADP-ribose in mediating intercellular Ca2+ sig-naling in sheep lens cells. J Cell Sci 1998; 111( Pt 9): 1217–1225.

67. Allbritton NL, Meyer T, Stryer L. Range of messenger actionof calcium ion and inositol 1,4,5-trisphosphate. Science 1992;258: 1812–1815.

68. Sneyd J, Wetton BT, Charles AC, Sanderson MJ. Intercellularcalcium waves mediated by diffusion of inositol trisphosphate:A two-dimensional model. Am J Physiol 1995; 268: C1537–1545.

69. Lazrak A, Peracchia C. Gap junction gating sensitivity tophysiological internal calcium regardless of pH in Novikoffhepatoma cells. Biophys J 1993; 65: 2002–2012.

70. Saez JC, Connor JA, Spray DC, Bennett MV. Hepatocyte gapjunctions are permeable to the second messenger, inositol1,4,5-trisphosphate, and to calcium ions. Proc Natl Acad SciUSA 1989; 86: 2708–2712.

71. Niessen H, Harz H, Bedner P, Kramer K, Willecke K. Selec-tive permeability of different connexin channels to the secondmessenger inositol 1,4,5-trisphosphate. J Cell Sci 2000; 113(Pt 8): 1365–1372.

72. Patterson RL, Boehning D, Snyder SH. Inositol 1,4,5-trisphosphate receptors as signal integrators. Annu Rev Biochem2004; 73: 437–465.

73. Takata M, Homma Y, Kurosaki T. Requirement of phospho-lipase C-gamma 2 activation in surface immunoglobulin M-induced B cell apoptosis. J Exp Med 1995; 182: 907–914.

74. Szalai G, Krishnamurthy R, Hajnoczky G. Apoptosis drivenby IP(3)-linked mitochondrial calcium signals. Embo J 1999;18: 6349–6361.

75. Chen R, Valencia I, Zhong F, et al. Bcl-2 functionally inter-acts with inositol 1,4,5-trisphosphate receptors to regulatecalcium release from the ER in response to inositol 1,4,5-trisphosphate. J Cell Biol 2004; 166: 193–203.

76. Boehning D, Patterson RL, Sedaghat L, Glebova NO,Kurosaki T, Snyder SH. Cytochrome c binds to inositol (1,4,5)

trisphosphate receptors, amplifying calcium-dependent apop-tosis. Nat Cell Biol 2003; 5: 1051–1061.

77. Assefa Z, Bultynck G, Szlufcik K, et al. Caspase-3-inducedtruncation of type 1 inositol trisphosphate receptor acceler-ates apoptotic cell death and induces inositol trisphosphate-independent calcium release during apoptosis. J Biol Chem2004; 279: 43227–43236.

78. Lawrence TS, Beers WH, Gilula NB. Transmission of hor-monal stimulation by cell-to-cell communication. Nature1978; 272: 501–506.

79. Murray SA, Fletcher WH. Hormone-induced intercellularsignal transfer dissociates cyclic AMP-dependent protein ki-nase. J Cell Biol 1984; 98: 1710–1719.

80. Bedner P, Niessen H, Odermatt B, Willecke K, Harz H. Amethod to determine the relative cAMP permeability of con-nexin channels. Exp Cell Res 2003; 291: 25–35.

81. Solares J, Garcia-Dorado D, Oliveras J, et al. Contraction bandnecrosis at the lateral borders of the area at risk in reperfusedinfarcts. Observations in a pig model of in situ coronary oc-clusion. Virchows Arch 1995; 426: 393–399.

82. Ruiz-Meana M, Garcia-Dorado D, Hofstaetter B, Piper HM,Soler-Soler J. Propagation of cardiomyocyte hypercontractureby passage of Na(+) through gap junctions. Circ Res 1999;85: 280–287.

83. Garcia-Dorado D, Inserte J, Ruiz-Meana M, et al. Gap junc-tion uncoupler heptanol prevents cell-to-cell progression ofhypercontracture and limits necrosis during myocardial reper-fusion. Circulation 1997; 96: 3579–3586.

84. Rose CR, Ransom BR. Gap junctions equalize intracellularNa+ concentration in astrocytes. Glia 1997; 20: 299–307.

85. Blanc EM, Bruce-Keller AJ, Mattson MP. Astrocytic gap junc-tional communication decreases neuronal vulnerability to ox-idative stress-induced disruption of Ca2+ homeostasis and celldeath. J Neurochem 1998; 70: 958–970.

86. Ozog MA, Siushansian R, Naus CC. Blocked gap junc-tional coupling increases glutamate-induced neurotoxicity inneuron-astrocyte co-cultures. J Neuropathol Exp Neurol 2002;61: 132–141.

87. Siushansian R, Bechberger JF, Cechetto DF, Hachinski VC,Naus CC. Connexin43 null mutation increases infarct sizeafter stroke. J Comp Neurol 2001; 440: 387–394.

88. Nakase T, Sohl G, Theis M, Willecke K, Naus CC. Increasedapoptosis and inflammation after focal brain ischemia in micelacking connexin43 in astrocytes. Am J Pathol 2004; 164:2067–2075.

89. SiuYi Leung D, Unsicker K, Reuss B. Gap junctions modulatesurvival-promoting effects of fibroblast growth factor-2 oncultured midbrain dopaminergic neurons. Mol Cell Neurosci2001; 18: 44–55.

90. Cheng A, Tang H, Cai J, et al. Gap junctional communicationis required to maintain mouse cortical neural progenitor cellsin a proliferative state. Dev Biol 2004; 272: 203–216.

91. Tazuke SI, Schulz C, Gilboa L, et al. A germline-specific gapjunction protein required for survival of differentiating earlygerm cells. Development 2002; 129: 2529–2539.

92. Mesnil M, Krutovskikh V, Omori Y, Yamasaki H. Role ofblocked gap junctional intercellular communication in non-genotoxic carcinogenesis. Toxicol Lett 1995; 82–83: 701–706.

93. Goldberg GS, Moreno AP, Lampe PD. Gap junctions betweencells expressing connexin 43 or 32 show inverse permselec-tivity to adenosine and ATP. J Biol Chem 2002; 277: 36725–36730.

94. White TW. Nonredundant gap junction functions. News Phys-iol Sci 2003; 18: 95–99.


95. Goodenough DA, Paul DL. Beyond the gap: Functions ofunpaired connexon channels. Nat Rev Mol Cell Biol 2003; 4:285–294.

96. Ebihara L. New roles for connexons. News Physiol Sci 2003;18: 100-103.

97. Bennett MV, Contreras JE, Bukauskas FF, Saez JC. New rolesfor astrocytes: gap junction hemichannels have something tocommunicate. Trends Neurosci 2003; 26: 610–617.

98. Plotkin LI, Manolagas SC, Bellido T. Transduction of cellsurvival signals by connexin-43 hemichannels. J Biol Chem2002; 277: 8648–8657.

99. Hur KC, Shim JE, Johnson RG. A potential role for cx43-hemichannels in staurosporin-induced apoptosis. Cell CommunAdhes 2003; 10: 271–277.

100. Cotrina ML, Lin JH, Alves-Rodrigues A, et al. Connexinsregulate calcium signaling by controlling ATP release. ProcNatl Acad Sci USA 1998; 95: 15735–15740.

101. Stout CE, Costantin JL, Naus CC, Charles AC. Intercellularcalcium signaling in astrocytes via ATP release through con-nexin hemichannels. J Biol Chem 2002; 277: 10482–10488.

102. Braet K, Aspeslagh S, Vandamme W, et al. Pharmacologi-cal sensitivity of ATP release triggered by photoliberation ofinositol-1,4,5-trisphosphate and zero extracellular calcium inbrain endothelial cells. J Cell Physiol 2003; 197: 205–213.

103. Ye ZC, Wyeth MS, Baltan-Tekkok S, Ransom BR. Functionalhemichannels in astrocytes: a novel mechanism of glutamaterelease. J Neurosci 2003; 23: 3588–3596.

104. Bruzzone S, Guida L, Zocchi E, Franco L, De Flora A. Con-nexin 43 hemi channels mediate Ca2+-regulated transmem-brane NAD+ fluxes in intact cells. Faseb J 2001; 15: 10–12.

105. Zipfel GJ, Babcock DJ, Lee JM, Choi DW. Neuronal apop-tosis after CNS injury: the roles of glutamate and calcium. JNeurotrauma 2000; 17: 857–869.

106. Tsukimoto M, Harada H, Ikari A, Takagi K. Involvementof chloride in apoptotic cell death induced by activation ofATP-sensitive P2X7 purinoceptor. J Biol Chem 2004.

107. Braet K, Mabilde C, Cabooter L, Rapp G, Leybaert L. Elec-troporation loading and photoactivation of caged InsP3: toolsto investigate the relation between cellular ATP release inresponse to intracellular InsP3 elevation. J Neurosci Methods2004; 132: 81–89.

108. Nicotera P, Leist M, Ferrando-May E. Intracellular ATP, aswitch in the decision between apoptosis and necrosis. ToxicolLett 1998; 102–103: 139–142.

109. Huang RP, Fan Y, Hossain MZ, Peng A, Zeng ZL, BoyntonAL. Reversion of the neoplastic phenotype of human glioblas-toma cells by connexin 43 (cx43). Cancer Res 1998; 58: 5089–5096.

110. Qin H, Shao Q, Curtis H, et al. Retroviral delivery of connexingenes to human breast tumor cells inhibits in vivo tumor

growth by a mechanism that is independent of significant gapjunctional intercellular communication. J Biol Chem 2002;277: 29132–29138.

111. Huang RP, Hossain MZ, Huang R, Gano J, Fan Y, Boyn-ton AL. Connexin 43 (cx43) enhances chemotherapy-inducedapoptosis in human glioblastoma cells. Int J Cancer 2001; 92:130–138.

112. Dang X, Doble BW, Kardami E. The carboxy-tail of connexin-43 localizes to the nucleus and inhibits cell growth. Mol CellBiochem 2003; 242: 35–38.

113. de Feijter AW, Matesic DF, Ruch RJ, Guan X, Chang CC,Trosko JE. Localization and function of the connexin 43 gap-junction protein in normal and various oncogene-expressingrat liver epithelial cells. Mol Carcinog 1996; 16: 203–212.

114. Kanemitsu MY, Loo LW, Simon S, Lau AF, Eckhart W. Tyro-sine phosphorylation of connexin 43 by v-Src is mediated bySH2 and SH3 domain interactions. J Biol Chem 1997; 272:22824–22831.

115. Loo LW, Kanemitsu MY, Lau AF. In vivo association ofpp60v-src and the gap-junction protein connexin 43 in v-src-transformed fibroblasts. Mol Carcinog 1999; 25: 187–195.

116. Guerrier A, Fonlupt P, Morand I, et al. Gap junctions and cellpolarity: connexin32 and connexin43 expressed in polarizedthyroid epithelial cells assemble into separate gap junctions,which are located in distinct regions of the lateral plasmamembrane domain. J Cell Sci 1995; 108( Pt 7): 2609–2617.

117. Lin JH, Yang J, Liu S, et al. Connexin mediates gap junction-independent resistance to cellular injury. J Neurosci 2003; 23:430–441.

118. Iacobas DA, Urban-Maldonado M, Iacobas S, Scemes E, SprayDC. Array analysis of gene expression in connexin-43 nullastrocytes. Physiol Genomics 2003; 15: 177–190.

119. Bao X, Chen Y, Reuss L, Altenberg GA. Functional expressionin Xenopus oocytes of gap-junctional hemichannels formedby a cysteine-less connexin 43. J Biol Chem 2004; 279: 9689–9692.

120. Leybaert L, Braet K, Vandamme W, Cabooter L, Martin PE,Evans WH. Connexin channels, connexin mimetic peptidesand ATP release. Cell Commun Adhes 2003; 10: 251–257.

121. Wong CW, Christen T, Kwak BR. Connexins in leukocytes:Shuttling messages? Cardiovasc Res 2004; 62: 357–367.

122. Ruchaud S, Lanotte M. cAMP and ’death signals’ in a myeloidleukaemia cell: From membrane receptors to nuclear re-sponses: a review. Biochem Soc Trans 1997; 25: 410–415.

123. Sasson R, Amsterdam A. Stimulation of apoptosis in humangranulosa cells from in vitro fertilization patients and its pre-vention by dexamethasone: involvement of cell contact andbcl-2 expression. J Clin Endocrinol Metab 2002; 87: 3441–3451.


3. Clearance of apoptotic and necrotic cells and its immunological consequences

DV Krysko, K D’Herde and P Vandenabeele Accepted for publication in Apoptosis. Review.

38

3. Clearance of apoptotic and necrotic cells and its immunological

consequences

3.1 Abstract

The ultimate and most favorable fate of almost all dying cells is engulfment by neighboring or

specialized cells. Efficient clearance of cells undergoing apoptotic death is crucial for normal

tissue homeostasis and for the modulation of immune responses. Engulfment of apoptotic

cells is finely regulated by a highly redundant system of receptors and bridging molecules on

phagocytic cells that detect molecules specific for dying cells. Recognition of necrotic cells by

phagocytes is less well understood than recognition of apoptotic cells, but an increasing

number of recent studies, which are discussed here, are highlighting its importance. We

emphasize that clearance of dying cells is an important fundamental process serving multiple

functions in the regulation of normal tissue turnover and homeostasis, and is not just simple

anti- or pro-inflammatory responses. Here we review recent findings on genetic pathways

participating in apoptotic cell clearance and molecules involved in engulfment of apoptotic

versus necrotic cells, as well as their immunological consequences and relationships to

disease pathogenesis.

3.2. Introduction

A paradox of life is that it cannot continue without death. In the lifespan of a multicellular

organism, millions of cells die during development and other events in order to shape,

preserve and protect the organism as a whole (Baehrecke, 2002). The most favourable fate

of dying cells is phagocytosis by professional phagocytes or by neighbouring cells, because

cell corpses may release cytotoxic substances that can damage neighbouring cells.

Phagocytosis was described in the late 19th century by the Russian biologist Ilya Metchnikoff

(Nobel Prize in Physiology or Medicine 1908), who used a light microscope to observe

injured tadpole fins. This discovery established a fundamental role of the phagocyte in

cellular turnover and in host defence. Since then much has been learned about phagocytes

and phagocytosis. The last decade has witnessed significant progress in understanding the

mechanisms by which macrophages continuously monitor cell viability and engulf dying cells.

This review will discuss recent findings on (1) the genetic pathways involved in the uptake of

dying cells, in phylogenetic context, (2) molecules involved in recognition and clearance of

apoptotic versus necrotic cells, (3) the controversial role of the phosphatidylserine receptor in

phagocytosis, (4) the functional consequences for immunity of interactions of macrophages

39 Introduction: Part III

with apoptotic and necrotic cells, and (5) the role of impaired clearance of dead cells in

pathogenesis of diseases.

3.3. Phagocytosis of dying cells: a journey through evolution

Animal models such as the nematode Caenorhabditis elegans and the fruit fly Drosophila

melanogaster combined with the powerful approaches of forward and reverse genetics have

greatly contributed to disclosing the mechanisms involved in clearance of dying cells, and

indicated a high degree of evolutionary conservation in many of the signal transduction

pathways implicated in the uptake of dying cells. C. elegans has been widely used in forward

genetic approaches to identify the genetic pathways implicated in the phagocytosis of dead

cells. These studies led to the isolation of several mutations that affect engulfment (Ellis et

al., 1991; Gumienny and Hengartner, 2001). This mutational analysis approach revealed

seven genes that function in two genetic pathways for phagocytosis (Fig. 1). One pathway is

composed of members of the Rac pathways and involves ced-12, ced-2 and ced-5, which

are the nematode homologues of ELMO, CrkII and DOCK180, respectively (Reddien and

Horvitz, 2004). It has been demonstrated that assembly of this complex activates the

guanine nucleotide exchange activity of DOCK180 for Rac (ced-10). A recent study identified

upstream components of this pathway in both worms and mammalian cells: a GTPase,

RhoG/MIG-2, and its activator TRIO/UNC-73 (deBakker et al., 2004; Henson, 2005).

A second genetic pathway (Fig. 1) that affects engulfment in the nematode is composed of

membrane proteins ced-1 and ced-7 and an adaptor protein, ced-6. Ced-1 is homologous to

the scavenger receptor found on endothelial cells (SREC) (Zhou et al., 2001) and shares

sequence identity with the mammalian LDL receptor-related protein (LRP). It was recently

proposed that ced-1 is the functional orthologue of LRP (Su et al., 2002). The ced-7 gene

product is similar to mammalian ABC1, a member of the ATP binding cassette transporter

family, and it is required for ced-1 clustering (Wu et al., 2001; Zhou et al., 2001). The

engulfment gene ced-7 is essential both in the dying and the engulfing cell (Wu and Horvitz,

1998). Ced-6 encodes an adaptor protein with a phosphotyrosine binding (PTB) site, a

leucine-zipper motif for homodimerization, and a potential proline-rich motif, and it acts

downstream of ced-1 and ced-7 (Liu and Hengartner, 1998; Zhou et al., 2001). It was

recently shown that ced-10 (Rac) is also a downstream mediator of signalling by ced-1, -6

and –7, and results in the convergence of the two engulfment pathways (Kinchen et al.,

2005). How this signalling module leads to activation of ced-10 is unknown. Although studies

of engulfment are much less advanced in Drosophila melanogaster, evolutionary

conservation has also been shown in this organism. Apoptotic corpses are cleared by

Clearance of apoptotic & necrotic cells 40

neighbouring cells in C. elegans, but in Drosophila this is done by both amateur and

professional phagocytes named the haemocytes (Abrams et al., 1993; Franc, 2002).

Figure 1. Two genetic pathways are involved in the uptake of apoptotic cells in C. elegans. The first pathway leads from the cell surface via the putative receptor protein Ced-1 (CD91) to ced-6 (GULP), whereas the other signaling pathway is comprised of UNC-73 (TRIO), MIG-2 (RhoG), ced-2 (CrkII), ced-5 (DOCK-180), and ced-12 (ELMO). The two pathways cooperate to activate ced-10, which in turn promotes the cytoskeletal rearrangements that are required for the successful engulfment of dead cell. The nature of the ligand(s), the substrate(s) that are transported by ced-7 (ABC1), and the upstream activating receptor(s) of the second pathway, all of which have not been identified, are indicated by question marks. Importantly, all C. elegans engulfment genes that have been identified so far are highly conserved throughout evolution from nematodes to humans. The identities of the mammalian homologues of the corresponding C. elegans proteins are indicated inside the parentheses.

Thus, the greater similarity of Drosophila to mammals in this context makes it a model

organism of choice for unravelling the molecular mechanisms of phagocytosis of apoptotic

cells. The CD36-related receptor, called Croquemort, is specifically expressed on

UNC-73 (TRIO)

MIG-2 (RhoG)

Ligands ?

Bridging molecules

Receptors

Ced-10 (Rac)

Ced-12 (ELMO)

Ced-6 (GULP)

Ced-5 (DOCK180) Ced-2 (CrkII)

Ced-7 (ABC1)

Ced

-1(C

d91)

?

Apoptotic cell

Engulfing cell

?

?


macrophages during Drosophila embryogenesis, macrophages of crq-deficient embryos are

not defective in phagocytosis of bacteria, but their phagocytosis of apoptotic cells was

significantly reduced (Franc et al., 1999). This revealed that Drosophila macrophages have

distinct pathways for phagocytosis of apoptotic cells and bacteria. It also has demonstrated

conservation of the mechanisms of apoptotic cell phagocytosis, because various CD36-

family members also participate in this process in mammals. How Croquemort transduces

signals for phagocytosis and the identity of its ligand remain unknown. On the other hand,

DRAPER, a Drosophila homologue of ced-1 and encoded by draper, is implicated in

phagocytosis of apoptotic cells (Manaka et al., 2004). Thus the biochemical pathways

involved in clearance of dead cells are broadly similar in phylogenetically diverse multicellular

organisms, indicating that this process represents an important regulatory mechanism that

has been conserved through evolution.

3.4. Molecules involved in recognition and clearance of apoptotic versus necrotic cells

Many receptors, adaptors and chemotactic molecules are reportedly involved in removing

apoptotic cells in various tissues (Arur et al., 2003; de Almeida and Linden, 2005; Emoto et

al., 1997; Fadeel, 2003; Grimsley and Ravichandran, 2003). Among the changes on the

surface of the apoptotic cells that facilitate their recognition (summarized in Fig. 2A), the best

characterized is the loss of phospholipid asymmetry and the translocation of

phosphatidylserine (PS) from the inner to the outer leaflet of the lipid bilayer, which occurs

very early during the apoptotic process (Fadok et al., 1998a; Fadok et al., 2001b; Fadok et

al., 1992; Schlegel and Williamson, 2001). Although the precise mechanism involved in the

initiation of movement of PS to the external surface of the membrane is still unclear, some

mechanisms have been suggested, including a combination of inhibition of the flippase that

normally confines PS to the inner leaflet, and activation of a bidirectional nonspecific

phospholipid scramblase (Williamson and Schlegel, 2002). Moreover, the ATP-binding

cassette transporter 1 (ABC1), a structural orthologue of C. elegans ced-7 (Wu and Horvitz,

1998), is required for efficient transbilayer redistribution of PS in the phagocytic targets, and

its expression in macrophages is essential for fulfilling maximal phagocytic capacity (Hamon

et al., 2000). It has been shown that the presence of oxidized PS in conjunction with non-

oxidized PS on the cell surface is an important signal for clearance of apoptotic cells (Kagan

et al., 2002). Interestingly, living cells exposing PS, such as activated B cells or neutrophils in

Barth syndrome, are engulfed neither by amateur nor by professional phagocytes (Dillon et

al., 2001; Dillon et al., 2000; Kuijpers et al., 2004). Another group showed expression of PS

as mimicry in host parasite interaction (Wanderley et al., 2005; Wanderley et al., 2006).

Moreover, a recent study by Elliott et al. (Elliott et al., 2005) indicate that PS is exposed


constitutively at high levels on T lymphocytes that express low levels of the transmembrane

tyrosine phosphatase CD45RB. In this study it has been show that changes in PS distribution

in living cells could serve several different functions such as to facilitate T-cell migration

through an endothelial cell layer, to modulate Ca2+ and Na+ uptake through the P2X7 cation

channel itself and to regulate activity of the multidrug transporter P-glycoprotein (Elliott et al.,

2005). These remarkable findings were recently explained by the observation that living cells

protect themselves from engulfment by expressing inhibitory signals, such as CD31 (Brown

et al., 2002), or by disrupting interactions between CD47 on the target cell and SIRP-

(SHPS-1), a heavily glycosylated transmembrane protein on the engulfing cell (Gardai et al.,

2005). Macrophages may also have a threshold for sensitivity to externalized PS on the

surface, which enables them to distinguish apoptotic cells with elevated levels from normal

cells with low levels of externalized PS (Balasubramanian and Schroit, 2003; Borisenko et

al., 2003). In addition, reduced lateral mobility of PS in the plasma membrane of dying cells

might be another factor discriminating between living and dead cells (Appelt et al., 2005).

Various macrophage receptors, including the scavenger-receptor superfamily (Platt et al.,

1998), integrins (Fadok et al., 1998c) and complement receptors (Mevorach et al., 1998), are

involved in clearance of apoptotic cells. Moreover, it was shown that annexin I, which is also

externalized during apoptosis, co-localizes with PS and is required for efficient clearance of

apoptotic cells (Arur et al., 2003). It is possible that annexin I is involved in PS translocation

or in its aggregation, thus creating high local concentrations of PS in the outer leaflet of the

plasma membrane (Fadok and Henson, 2003). It has been reported that a broad range of

extracellular bridging molecules, such as 2 glycoprotein (Balasubramanian and Schroit,

1998), milk fat globule protein (MFG-E8) (Hanayama et al., 2002), protein S (Anderson et al.,

2003), growth arrest-specific 6 (Gas6) (Chen et al., 1997) and thrombospondin (Savill et al.,

1992) facilitate interactions between macrophages and exposed PS on apoptotic cells (Fig.

2A). In addition, bridging molecules can have higher specificity for oxidized PS than for non-

oxidized PS. It has been shown that MFG-E8 preferentially interacts with the oxidized form of

PS, and to a lesser degree with non-oxidized phosphatidylserine and phosphatidylcholine

(Borisenko et al., 2004).

In contrast to what is known about apoptotic cells, it is not clear how macrophages recognize

necrotic cells (Fig. 2B). Unexpectedly, it was shown that necrotic cells also externalize PS

(Krysko et al., 2004; Lecoeur et al., 2001) and that they can be recognized through a PS-

dependent mechanism (Bottcher et al., 2006; Brouckaert et al., 2004; Hirt and Leist, 2003),

although less effectively than apoptotic cells (Brouckaert et al., 2004). In addition, it was

demonstrated that recognition of apoptotic and necrotic cells occurs via distinct non-

competing mechanisms (Cocco and Ucker, 2001).


Figure 2. Phagocyte receptors, bridging molecules and ligands engaged in regulating the recognition and engulfment of apoptotic (A) and necrotic (B) cells. Clearance of dead cells is stimulated by various molecules displayed on the surface of macrophages and on dying cells. Note that serum factors frequently act as bridging proteins that engage receptors on the phagocyte. Despite possessing many molecules involved in stimulating engulfment, living cells can actively prevent their own clearance by expressing inhibitory signals, such as CD31, or by disrupting interactions between CD47 on the target cell and SHPS-1. There is considerable information on molecules involved in recognition and clearance of apoptotic cells by macrophages (partial list), but information related to necrotic cells is sparse (full list). ABC1-1, ATP-binding cassette transporter 1; PS, phosphatidylserine; Ox-PS, oxidized phosphatidylserine; MFG-E8, milk fat globule epidermal growth factor-8; Mer, myeloid epithelial reproductive tyrosine kinase; TSP-1, thrombospondin-1; iC3b inactivated complement

fragment C3b; C1q, complement protein C1q; Gas-6, growth arrest specific gene-6; SIRP , signal

regulatory protein ; PTX3, long pentaxin 3; CD14, lipopolysaccharide receptor; PSR, phosphatidylserine receptor; CD36, thrombospondin receptor; CD47, integrin-associated protein;

CD91, 2 macroglobulin receptor; ICAM-3 (CD50), intercellular adhesion molecule-3; FcR, Fc

fragment of immunoglobulin G receptor; v 3/5 vitronectic receptor integrins; SR-A, scavenger receptor class A; SR-B1; scavenger receptor class B1; MBL, mannose-binding lectin; HRG, histidine-rich glycoprotein; SAP, serum amyloid P component; CRP, C-reactive protein.

It has been suggested recently that several macrophage receptor systems known to be

involved in the engulfment of apoptotic cells also contribute to the uptake of necrotic cells

(Bottcher et al., 2006). Using different blocking antibodies, these authors demonstrated that

the thrombospondin-CD36- v 3 complex and CD14 are involved in the engulfment of heat-

induced necrotic peripheral blood lymphocytes by human monocyte-derived macrophages.

When used separately, these antibodies (anti-thrombospondin, anti-CD36, anti-CD14)

reduced engulfment by a maximum of only 30% as compared to controls. Necrotic cell

clearance appeared to be reduced rather than blocked by simultaneous inhibition of the

thrombospondin-CD36- v 3 system and PS, and did not result in a significant decrease of

apoptotic cells uptake. The inability to completely inhibit necrotic cell clearance, even when

multiple receptors were blocked, indicates redundant mechanisms for clearance, and implies

that other molecules may be involved in removal of necrotic cells.

FcR

Mer

Gas-6

YIgG

CD36

CD68

� �v ��

PSMFG-E8/Del1

Ox-PS

CRP/CAPFcR

ABC-1

ABC-1

Annexin I

CD14

CD91

C1q

Calreticulin

Collectins

CD31(disabled)

CD47

SR-A

SR-B1

SIRP�

TSP-1Protein S

� �v ��

FibronectinCollagen VI

ICAM-3

PTX3

iC3b

PS

A

CD36

CD14

PH antigen

C1q

TSP-1

PTX3

MBL

PS

C3/C4

HRG

PS

PS

B

PS

PSR ?


Binding of some bridging molecules depends on the stage and type of cell death. C1q and

mannose-binding lectin (MBL) bind to apoptotic cells rather late in the cell death process:

C1q binding to early apoptotic cells is much weaker than to late apoptotic cells, and binding

of MBL and pentraxin-3 (PTX3) was demonstrated exclusively on late apoptotic cells (Nauta

et al., 2003; Nauta et al., 2002; Rovere et al., 2000) and necrotic cells (Roos et al., 2004).

Similarly, deposition of C3 and C4 on apoptotic cells consequent to activation of the

complement cascade occurs late during apoptosis (Gaipl et al., 2001; Zwart et al., 2004).

Moreover, it has been shown that C3 and C4 bind to necrotic cells independently of the

method used to induce necrosis and the cell type (Ciurana et al., 2004). It was recently

demonstrated that C-reactive protein (CRP), an acute phase protein, preferentially binds to

secondary necrotic neutrophils (Hart et al., 2005). CRP is not unique in binding to secondary

necrotic cells; a similar phenomenon was reported for the unrelated serum protein

thrombospondin (Hart et al., 2000). It was also reported that histidine-rich glycoprotein

(HRG), a multidomain plasma protein, selectively binds to necrotic cells, thereby facilitating

their uptake by phagocytes (Jones et al., 2005). Moreover, PH2 antibody inhibited

macrophage uptake of late apoptotic and necrotic cells by, but not of early apoptotic cells

(Fujii et al., 2001). These data indicate that the putative PH2 antigen is a novel phagocytosis

marker that may translocate to the cell surface of necrotic cells and late apoptotic cells,

resulting in maximal recognition and engulfment by macrophages. A protein bound by PH2

was identified as the 170-kDa subunit of eukaryotic translation initiation factor 3 (Nakai et al.,

2005), but the physiological significance of this observation remains unclear. Taken together,

these findings underscore the complexity of necrotic cell clearance. Further studies are

needed to examine whether those molecules that are important in apoptotic cell clearance,

and their bridging counterparts, are also implicated in engulfment of necrotic cells, and to

identify other differences and similarities in the interaction of macrophages with apoptotic and

necrotic cells.

3.5 Phosphatidylserine receptor: unresolved controversies

A putative PS-specific receptor (PSR) was identified by phage display as the target of a

monoclonal antibody generated against stimulated human macrophages (Fadok et al., 2000).

The product of that screen, mAb 217G8E9, was used in PSR binding and blocking

experiments (with PS-containing liposomes), as well as in sub-cellular localization. From

these studies it was concluded that PSR is a transmembrane receptor critical in cell

engulfment (Fadok et al., 2000). Indeed, homologues of the mouse PSR have been found

not only in humans but also in Drosophila melanogaster, C. Elegans and D. rerio (Fadok et

al., 2000; Hong et al., 2004; Wang et al., 2003).


However, the important role of PSR in apoptotic cell clearance during development was

recently challenged by the phenotype of three PSR-deficient mice, independently produced

in different genetic backgrounds: a mixed C57BL/6 background (a disruption upstream of

exon 1 including translational start site of the PSR gene product) (Li et al., 2003), a chimeric

129 x C57BL/6 background (a disruption upstream of exon 1 and extending into exon 3)

(Kunisaki et al., 2004), and a pure C57BL/6 background (a disruption limited to exons 1 and

2) (Bose et al., 2004). All three PSR knockout mice died perinatally, and displayed

pathologies in different organs. Although two groups found defects in clearance of apoptotic

cells in these PSR-/- mice (Kunisaki et al., 2004; Li et al., 2003), it is not clear from the studies

of Kunisaki et al. (Kunisaki et al., 2004) whether this was the case in their PSR-/- mice. Based

on the observation that the embryonic liver and thymus of these PSR-/- mice exhibited

significantly higher frequencies of uncolocalised TUNEL-positive cells with F4/80 positive

macrophage, the authors concluded that phagocytosis of apoptotic cells was disturbed

during embryogenesis (Kunisaki et al., 2004). Furthermore, their data were puzzled by their

observation that the number of F4/80 positive macrophages and TUNEL-positive cells was

apparently reduced both in liver and thymus of PSR-/- embryos. Thus the question arises

whether the detected increase in non-engulfed apoptotic cells reflects functional defects in

phagocytic clearance of apoptotic cells, or a decrease in the overall phagocytic capacity of

the tissue as indicated by the decrease in F4/80-positive cells. Notably, in tissues in which

apoptosis is prominent during embryogenesis, such as remodelling of the genital ridge during

gonad morphogenesis and differentiation of the neural tube, the macrophages may act as

neighbouring bystander cells, not as recruited macrophages (Bose et al., 2004). Moreover,

one would expect a more striking phenotype in mice with a genetic deficiency (Kunisaki et

al., 2004; Li et al., 2003) that leads to real defect in apoptotic cell clearance accompanied by

development of autoimmunity, e.g. as observed in MFG-E8- and c-mer- deficient mice

(Cohen et al., 2002; Hanayama et al., 2004). Importantly, the third group (Bose et al., 2004)

could not find defects in the phagocytosis of apoptotic (TUNEL+) cells after extensive

histological analysis of numerous tissues. Their in vitro phagocytosis studies of fetal liver-

derived macrophages also found no defects in engulfment of PSR-deficient macrophages.

Surprisingly, mAb 217G8E9 still reacted with the surface of macrophages derived from PSR-

deficient mice (Bose et al., 2004), indicating that the antibody targeted something else. It is of

interest to explore whether mAB 217G8E9 blocks the uptake of apoptotic cells by

macrophages derived from PSR-deficient mice.

Li et al. (Li et al., 2003) performed adoptive transfer of PSR-/- fetal liver into lethally irradiated

wild type hosts. Four weeks after the transfer, thioglycollate-elicited macrophages were

isolated and then cocultured with apoptotic T cells. Phagocytosis was reduced by 50%, and

PS-inhibitable engulfment by PSR-deficient macrophages was completely abolished. In order


to further address the controversial role of PSR in apoptotic cell clearance, the phagocytic

capacity of mouse embryonic fibroblasts (MEF) derived from PSR-/- mice (Li et al., 2003) was

investigated (Mitchell et al., 2006). These mice were chosen because this study (Li et al.,

2003) provided the best genetic evidence implicating PSR in apoptotic cell clearance.

Importantly, PSR-deficient MEFs were able to engulf both apoptotic and necrotic cells

(Mitchell et al., 2006). In addition, the authors showed that overexpression of PSR in 293T

cells did not enhance their engulfing activity, which contrasts with the enhancement obtained

by overexpression of v 5 integrin, another receptor involved in phagocytosis (Albert et al.,

2000). It was demonstrated using mAB 217G8E9 after cell permeabilization that PSR is

localized predominantly in the nucleus (Mitchell et al., 2006), and this was corroborated by

two other independent studies using expression of GFP-tagged PSR in mammalian and

hydra cells (Cikala et al., 2004; Cui et al., 2004). Moreover, no PSR was detected in the

plasma membrane at the moment of engulfment of apoptotic interstitial cells by epithelial

cells in hydra (Cikala et al., 2004). Indeed, it was shown that PSR has a chromatin binding

motif (Clissold and Ponting, 2001), which explains its localization in the nucleus. Using a

PSR-peptide array it was found that mAb 217G8E9 can bind weakly to a PSR peptide (Bose

et al., 2004), which explains the original isolation of PSR cDNA clones by phage display

(Fadok et al., 2000). Thus, a mere role of this protein as a membrane-bound receptor for PS

molecules is challenged by recent publications demonstrating the nuclear localization of PSR

(Cikala et al., 2004; Cui et al., 2004). Importantly, however, nuclear localization does not

preclude a role in cell clearance, for instance in regulation of other genes involved in

recognition or engulfment of dying cells. It is also conceivable that PSR may re-localize to

nucleus after recognition of PS. Alternatively, the protein could cycle between different

cellular compartments and might be present only transiently in the plasma membrane, as has

been shown for another transmembrane protein, macrosialin, the mouse homolog of human

CD68 (Kurushima et al., 2000).

3.6. Interactions of macrophages with apoptotic and necrotic cells: functional

consequences for immunity

3.6.1. Interaction with apoptotic cells and secondary necrotic cells

A characteristic of apoptotic cell clearance is that no pro-inflammatory macrophage

responses are induced. Several groups proposed that phagocytosis of apoptotic cells

induces an anti-inflammatory reaction by leading to production of transforming growth factor

(TGF- ), prostaglandin E2 and platelet activating factor (PAF) (Fadok et al., 1998b; Huynh

et al., 2002). In fact direct in vivo setting up of apoptotic cells into inflammatory lesions in


lungs of LPS-stimulated mice accelerated inflammation resolution mediated by increased

production of TGF- (Huynh et al., 2002). Moreover, it has been shown that TGF- prevents

pro-inflammatory cytokine production by inhibiting p38 MAPK phosphorylation and NF- B

activation (Xiao et al., 2002). Thus, one of the leading roles was ascribed to a model in which

subsequent inflammatory down-regulation is dependent on initial TGF- stimulation (Fadok

et al., 1998b; Huynh et al., 2002). In those studies the macrophage responses were

assessed after long periods of target cell interaction ( 18 h) (Fadok et al., 1998b; Huynh et

al., 2002). However, it has been shown by analysis at earlier time points that TGF- was not

involved in the initial anti-inflammatory responses by apoptotic cells (Cvetanovic and Ucker,

2004). In contrast, apoptotic cells exert their anti-inflammatory effect directly upon binding to

the macrophages, independently of subsequent engulfment and soluble factor involvement

(Cvetanovic and Ucker, 2004). Similarly, it has been demonstrated that cell-cell contact with

apoptotic cells is sufficient to profoundly inhibit production of the inflammatory cytokine IL-12

by activated macrophages (Kim et al., 2004). Indeed, it is conceivable that additional

mechanisms might contribute to the anti-inflammatory relationship between the phagocytes

and apoptotic cells, namely that apoptotic cells sustain an anti-inflammatory phenotype

throughout the process of cellular dissolution. Apoptotic cells themselves can produce

immunomodulatory factors, such as IL-10 (Gao et al., 1998) and TGF- (Chen et al., 2001).

Furthermore, when apoptotic neutrophils contact activated monocytes they can switch from a

pro-inflammatory to an anti-inflammatory state (Byrne and Reen, 2002).

According to the present concept, if apoptotic cells are not cleared within a certain time, they

become late apoptotic cells that have lost membrane integrity and leak large amounts of

intracellular contents; these are also called secondary necrotic cells, as they behave like

necrotic cells and induce an immune response (Cohen et al., 2002; Savill et al., 2002). In

contrast to what we would have anticipated, it has been shown that secondary necrotic

neutrophils are efficiently taken up by macrophages without release of pro-inflammatory

mediators such as IL-8 and TNF- (Ren et al., 2001). Moreover, engulfment of secondary

necrotic cells by J774A.1 macrophages leads to inhibition of LPS-induced release of the pro-

inflammatory cytokines TNF- and IL-6 (Cocco and Ucker, 2001). Corroborating this notion

is a recent study showing that late apoptotic (secondary necrotic) cells were just as effective

as early apoptotic cells in terms of inhibiting ERK1/2 activity and stimulating JNK1/2 and p38,

whereas necrotic cells had no detectable effect on c-Jun N-terminal kinase and p38 (Patel et

al., 2006). This view is supported by studies in which targeted deletion of CD14 or MBL in

mice leads to in vivo persistence of apoptotic cells in the absence of inflammation and

autoimmunity (Devitt et al., 2004; Stuart et al., 2005). A reasonable conclusion is that the

onset of secondary necrosis does not always carry inflammatory consequences. Additionally,


secondary and primary necrotic (lysed) cells may differ in cellular content. In secondary

necrotic cells, the contents had been exposed to activated caspases, which have dismantled

crucial cellular components and processes, such as proteasomes, translation and the cell

cycle, but this is not the case with lysed necrotic cells. In this regard primary necrotic cells

may elicit responses other than those elicited by secondary necrotic cells. Similarly, it has

been shown that, in contrast to secondary necrotic neutrophils, lysed neutrophils release

their formidable arsenal of proteases into tissue, prolonging the inflammatory response

(Fadok et al., 2001a).

Engulfment of apoptotic cells by macrophages is often considered as the most

immunologically favourable way of removing cells, simplistically contrasting it to the less

favourable, pro-inflammatory role of necrotic cell clearance. However, many studies indicate

that apoptotic cell clearance can also have inflammatory consequences, which challenges

the well-accepted anti-inflammatory paradigm of apoptotic cells. In this regard macrophages

that had ingested bacterially-induced apoptotic neutrophils exhibited markedly increased

production of the pro-inflammatory cytokine TNF- , but not the anti-inflammatory cytokine

TGF- (Zheng et al., 2004). In vivo evidence shows that the clearance of apoptotic cells can

activate macrophages. Irradiation-induced apoptotic cells can stimulate pro-inflammatory

response by splenic macrophages, as a consequence of their recognition and clearance

(Lorimore et al., 2001). In addition, apoptotic cells that become opsonized with antibody,

particularly with IgG, can be recognized via Fc receptors, leading to a pro-inflammatory

response (Gregory and Devitt, 2004; Hart et al., 2004). Delayed or impaired clearance of

apoptotic cells was observed in mice in which genetic deletion of complement protein C1q,

MER receptor tyrosine kinase or MFG-E8 lead in the context of inflammatory signal to the

development of systemic autoimmune disease (Cohen et al., 2002; Hanayama et al., 2004;

Mitchell et al., 2002).

Taken together, the above-mentioned data indicate that the response of macrophages to

apoptotic and secondary necrotic cells may be either anti- or pro-inflammatory. Such

differential reaction of macrophages towards apoptotic cells may depend largely on the

activating state of macrophages, source of target cells, and death inducing stimuli, as well as

on the receptors involved in the uptake. Additional work is required to understand in detail

the complex interactions between apoptotic cells and macrophages.

3.6.2. Interaction with necrotic cells

It has been shown that lysed neutrophils stimulate production of macrophage-inflammatory

protein 2 (MIP-2), IL-8, TNF- and IL-10, but lymphocytes do not (Fadok et al., 2001a).

Importantly these authors demonstrated that the membranes of apoptotic and necrotic


neutrophils have similar anti-inflammatory effects. The signals emanating from necrotic cells

are important because different ways of recognizing and taking up dying cells may have

major effects on immunologic parameters. For instance, it has been shown that the trigger

provoking necrotic death has an important impact on the kind of immunological response of

phagocytes (Hirt and Leist, 2003). The authors demonstrated that uptake of ‘programmed

necrotic’ (ATP depleted or lysed) Jurkat cells clearly inhibited the E. coli-induced TNF-

secretion by human monocyte derived macrophages to a comparable extent as the uptake of

apoptotic cells, whereas heat-killed cells failed to inhibit the pro-inflammatory reaction by

macrophages.

It is note worthy that necrotic cells are not necessarily always pro-inflammatory. We recently

showed that L929 cells (a mouse fibrosarcoma cell line) subjected to TNF-induced necrosis

fail to induce pro-inflammatory cytokine production (Brouckaert et al., 2004). These findings

are in line with the occurrence of necrotic cell death in the absence of an inflammatory

response in the apaf-1 deficient mice (Chautan et al., 1999). In addition, it was shown that

necrotic cells cannot activate macrophages themselves but can enhance their activation by

LPS, whereas apoptotic cells inhibit phlogistic macrophage responses (Cocco and Ucker,

2001).

The release of cytokines or other factors from necrotic cells themselves may be crucial for an

inflammatory response. It has been reported that necrotic cells may stimulate inflammatory

responses via the extracellular activity of the DNA-binding protein, high mobility group box 1

protein (HMGB-1), which can leak from necrotic cells and incite inflammatory responses from

macrophages, but it remains bound in apoptotic cells even when they are undergoing

secondary necrosis, which prevents induction of inflammatory responses (Bianchi and

Manfredi, 2004; Scaffidi et al., 2002). Macrophage activation by HMGB-1 occurs through

TLR2 and TLR4, as is the case for LPS (Park et al., 2004). Necrotic cells can release heat-

shock proteins (Basu et al., 2000) and large amounts of uric acid (Shi et al., 2003), both of

which can have a pro-inflammatory effect through TLR2 and TLR4 (Binder et al., 2004; Tsan,

2006).

In addition to release of HMGB1 and HSPs following cell damage or necrosis, other

substances that are normally confined within cells might also be released. For example, RNA

released from or associated with necrotic cells form double-stranded structures that can

stimulate TLR3 on dendritic cells, leading to interferon secretion (Alexopoulou et al., 2001;

Kariko et al., 2004). This inflammatory response to necrotic cells can be abolished by

pretreating necrotic cells with RNAse (Kariko et al., 2004). Moreover genomic DNA (double-

stranded) released from necrotic cells exerts immunostimulatory effects on macrophages

and dendritic cells (Ishii et al., 2001). Nucleotides (ATP) and nucleosides (adenosine) can

also be released from necrotic cells (Krysko et al., 2005; Skoberne et al., 2004), upon which


they stimulate maturation of dendritic cells (Wilkin et al., 2001) and activate transcription

factor NF- B (Ferrari et al., 1997). Nucleotides and nucleosides are unstable and therefore

might act only transiently in the local environment. In addition, in a recent study it was shown

that pre-treatment of primary and Bac1 murine macrophages with LPS sensitizes them to

activation of caspase-1, a converting enzyme required for the cleavage and maturation of the

pro-inflammatory cytokines IL-1 and IL-18, to ATP (Kahlenberg et al., 2005). This link

between release of ATP from necrotic cells and release of pro-inflammatory cytokines must

be explored further.

In addition, it was found that cells undergoing necrosis are characterized by NF- B- and

p38MAPK-mediated upregulation and secretion of IL-6 but these events are strongly reduced

or absent in the same cell types when they are dying by apoptosis (Vanden Berghe et al.,

2006). These findings demonstrate that necrotic cells induce an inflammatory response not

only by spilling their contents after lysis, which occurs mainly in the late stages, but also by

actively secreting inflammatory cytokines. Inflammatory cytokine production in necrotic cells

is in accordance with the notion that necrotic cells retain their ability to synthesize proteins

(Clemens et al., 2000; Saelens et al., 2005; Saelens et al., 2001).

3.6.3. Clearance of apoptotic and necrotic cells promotes cell life, cell growth and cell

death

In addition to the role of dead cells in modulating immune responses, it has recently become

clear that engulfment of dying cells may also promote survival and provide growth signals to

surrounding cells. Engulfment of apoptotic cells can promote survival of murine peritoneal

and bone marrow derived macrophages; this can be the result of activation of Akt

simultaneously with inhibition of the mitogen-activated protein kinases extracellular signal-

regulated kinase (ERK) 1 and ERK2 (Reddy et al., 2002). In addition, it was recently shown

that engulfment of apoptotic neutrophils actively regulates granulopoiesis by suppressing IL-

23 secretion by phagocytes, which in turn reduces IL-17 and G–CSF (Stark et al., 2005).

Furthermore, in Drosophila melanogaster apoptotic cells express the secretory factor

wingless, which can directly stimulate cell proliferation (Ryoo et al., 2004). In an elegant

study by Golpon et al. (Golpon et al., 2004) it was demonstrated that phagocytosis of

apoptotic Jurkat T cells by mouse epithelial cells (HC-11) and peritoneal macrophages leads

to the secretion of growth and survival factors, such as vascular endothelial growth factor

(VEGF), which subsequently promote growth of endothelial and epithelial cells in a paracrine

way. The authors also detected increased transcription of hepatocyte growth factor (HGF)

when phagocytes were exposed to apoptotic cells but not when they were exposed to

necrotic cells. HGF is a growth factor for epithelial cells that is important during epithelial


repair after tissue injury (Ware and Matthay, 2002). In line with this study, it has been shown

in mice that during bacterial pneumonia alveolar macrophages produce HGF after

engulfment of apoptotic neutrophils (Morimoto et al., 2001). Moreover, using microarray

analysis it was demonstrated that several genes involved in angiogenesis, such as

enthothelin-1, plasminogen activator inhibitor and nitric oxide synthase 2 (Ramanathan et al.,

2003; Salani et al., 2000a; Salani et al., 2000b; Stefansson et al., 2003) were up-regulated

following engulfment of apoptotic cells, while inhibitors of angiogenesis, endostatin and von

Hippel-Lindau tumour suppressor gene (which controls VEGF expression) were down-

regulated (Golpon et al., 2004).

Necrotic cells may also induce expression of VEGF and hence participate in stimulation of

angiogenesis. Necrotic cells can activate NF- B and induce expression of genes involved in

inflammation and tissue-repair responses, including the neutrophil-specific chemokine genes

KC, macrophage-inflammatory protein-2 and VEGF (Li et al., 2001). Interestingly, VEGF

expression could be detected in areas adjacent to necrotic cells in various disease models

(Sharkey et al., 1993; Shinohara et al., 1996), and in situ analysis of tumour specimens has

also demonstrated the localization of VEGF-producing cells near necrotic foci (Shweiki et al.,

1992). In this context it is worth noting that induced cell death often leads to replenishment

and repair, in contrast to the removal of excess cells seen during development. Which

molecules in apoptotic and necrotic cells lead to the changes in the phagocyte gene

expression pattern and which receptors participate in these responses is unknown.

Several studies have addressed the ability of macrophages to induce not only cell life and

growth but also to actively influence cell death (Diez-Roux et al., 1999; Duffield et al., 2000;

Lang and Bishop, 1993). For example, engulfment of cells can promote death and ensure

that it is irreversible. It has been demonstrated by Reddien et al (Reddien et al., 2001) and

Hoeppner et al .(Hoeppner et al., 2001) that mutations that block clearance in C. elegans

strongly enhance the ability of partial loss-of-function mutations in the pro-apoptotic genes

ced-1 or ced-4 to rescue cells destined to die. Several potential mechanisms, including the

production of oxidants, nitric oxide (NO), TNF- and Fas-Fas-ligand interaction (Brown and

Savill, 1999), have been proposed. Dendritic cells can also induce rapid cell death, but in this

case death is caspase-independent, and they engulf the dead cells (Trinite et al., 2005).

These observations point to a close connection between clearance of apoptotic and necrotic

cells on the one hand and fundamental processes such as tissue repair, cell death, and cell

growth on the other.


3.6.4. Phagocytosis in anti-tumour defence

Macrophages and their specialized derivatives, myeloid dendritic cells (DC), are well-

characterized as antigen-presenting cells. Engulfment of necrotic tumour cells induces in

vitro DC maturation (Basu et al., 2000; Kacani et al., 2005; Sauter et al., 2000; Wong et al.,

2005), antigen presentation by macrophages (Barker et al., 2002), and endothelial cell

activation, but engulfment of apoptotic tumour cells does not induce these processes (Chen

et al., 2006). These results indicate that necrotic cells may be used in the development of

anti-tumour vaccines; however, several recent in vivo studies demonstrated that apoptotic

rather than necrotic cells are suitable for development of anti-tumour vaccines. In this regard

irradiation-induced apoptotic CT26-HA, CT26-wt (murine colonic epithelial tumour) or

RENCA (spontaneous murine renal cell carcinoma) prevented tumour outgrowth in 100%,

75% and 100% of mice, respectively, for more than 30 days after lethal dose of tumour

challenge (Scheffer et al., 2003). In contrast, the authors showed that injection of freeze-

thaw-induced necrotic tumour cells led to protection of no more than 30% of mice (Scheffer

et al., 2003). Similarly, mice injected with dendritic cells pre-loaded with apoptotic melanoma

cells were protected against an intracutaneous challenge with B16, a poorly immunogenic

murine melanoma of spontaneous origin. But injection with freeze-thaw-induced necrotic

cells granted no protection. Remarkably, 80% of the mice remained tumour-free 12 weeks

after challenge (Goldszmid et al., 2003). A recent study showed that mice inoculated

simultaneously with viable and killed in vitro by doxorubicin (caspase activation, chromatin

condensation, TUNEL negative) colon carcinoma CT26 cell line, but not killed with mitomycin

C (caspase activation, chromatin condensation, TUNEL positive), reduced the frequency of

tumours by 45% on day 30 and by 30% on day 120 (Casares et al., 2005). Importantly

disrupting the integrity of the membrane of doxorubicin treated tumor cells by freeze-thawing

was sufficient to abolish their protective property. Furthermore, the fact that slow acting drugs

such as doxorubicin induce genotoxic stress and activate NF- B, which is important for the

activation of inflammatory gene products, may explain the caspase-dependent

immunogenicity of cells that underwent doxorubicin-induced cell death (Janssens et al.,

2005). In addition, animals that did not develop tumours were also protected against a

second inoculation with live CT26 cells, but were not immune against the unrelated murine

mammary adenocarcinoma TS/A tumour. Thus, the induced immune response was specific

and not a general inflammatory-like protective response like that elicited by Coley toxins

(Lillehei et al., 1999). Moreover, BALB/c mice vaccinated with cells from freshly resected or

cryopreserved CT26 tumours elicited an effective anti-tumour immune response only if

doxorubicin was added during the in vitro incubation, although cryopreserved cells were less

effective than freshly removed tumour cells. However, in all these studies (Casares et al.,


2005; Goldszmid et al., 2003; Scheffer et al., 2003) the anti-tumour effects of vaccines

derived from apoptotic cells were compared to the effects of vaccines derived from necrotic

cells killed by freezing and thawing. This way of killing cells physically disrupts membranes

and organelles, and is not associated with signal transduction (Leist and Jaattela, 2001).

These data delineate a potential approach for generating anti-cancer vaccines and

stimulating anti-neoplastic immune responses in vivo. Further studies are needed to resolve

the controversy related to the possible anti-tumour properties of apoptotic and necrotic cells

in vitro and in vivo.

Data on whether DC and macrophages can mature and present antigens derived from

apoptotic and/or necrotic cells are controversial. Nevertheless, one could envision novel

therapeutic strategies, e.g. involving vaccination with DCs that have phagocytosed apoptotic

or necrotic tumour cells, in combination with adjuvants. In this regard, the protection elicited

against the tumour may be enhanced by adding irradiated lymphoma cells coupled with

Annexin V (Bondanza et al., 2004). Notably, 90% of the mice vaccinated with Annexin V-

coupled tumour cells rejected the tumour after challenge, compared to 25% of the animals

vaccinated with tumour cells without Annexin V. The authors found that in vivo there was a

decrease in the clearance of Annexin V-coupled tumour cells by thioglycollate-elicited

macrophages. Importantly Annexin V preferentially targets irradiated tumour cells to CD8+

dendritic cells for in vivo clearance, pointing to the involvement of CD8+ dendritic cells in the

anti-tumour immunity. This may be a promising strategy for developing strong anti-tumour

immunity. Another possible approach is forcing phagocytosis of living cancer cells in the

absence of a death signal that induces phagocytosis (Fadeel et al., 2004). This might be

done by disrupting interactions between the “repelling protein” CD47 on the target cell and

SIRP on the engulfing cell, which would then allow the uptake of viable cells (Gardai et al.,

2005). In other words, cancer cells may be buried alive when they expose appropriate

macrophage recognition signals.

The above mentioned data provide strong evidence that clearance of apoptotic and necrotic

cells represents a fundamental process not only in the sense of simple anti- and pro-

inflammatory responses, but a process with multiple different impacts on surrounding tissues

including neutral, pro- or anti-inflammatory response, modulation of cell survival, cell death,

cell proliferation, or affecting anti-tumour immunity (summarized in Fig. 3).

Disturbance of the balance between these responses during clearance of dying cells may

lead to different diseases, some of which will be discussed below. It is likely that progression

of inflammatory responses and development of diseases is profoundly influenced by a

combination of the above-mentioned responses.


Figure 3. Possible multiple impacts of apoptotic and necrotic cell clearance on surrounding cells and tissues.

3.7. Impairment of phagocytosis of dead cells and its role in the development of

diseases

Defective or inefficient clearance of dying cells may contribute to persistence of inflammation

and excessive tissue injury, as well as several different human pathologies, including

systemic lupus erythematosus (SLE) (Herrmann et al., 1998; Munoz et al., 2005; Ren et al.,

2003), cystic fibrosis (Vandivier et al., 2002), chronic obstructive pulmonary disease (Hodge

et al., 2003) and diabetes (O'Brien et al., 2002). For example, in SLE patients it was found

that the major autoantigens were nucleosomal proteins released from the DNA before or at

the time of internucleosomal cleavage during apoptosis. Thus, defective phagocytic

clearance of apoptotic cells leads to increased exposure of nucleosomes to the immune

system and to generation of nucleosome specific T cells and antinucleosome autoantibodies

(Baumann et al., 2002; Berden, 2003). Moreover, during apoptosis enzymes such caspases

might increase the immunogenicity of apoptotic cells by generating neoantigens that can

Apoptoticcell

Necroticcell

Anti-inflammatory:

TGF-�1, E2, PAF

Engulfment

Engulfment

Pro-inflammatory:IL-6, HMBG1, HSP70,

HSP90, uric acid

NeutralPro-inflammatory:

IL-8, IL- 10, TNF-�

Antigen presentation &anti-tumour immunity

Tissue repair:VEGF, MIP-2

Anti-inflammatory:

TGF-�1

Antigen presentation &anti-tumour immunity

Tissue repair:VEGF, HGF

Cell death:

FasL, TNF-�, NO

Regulation of cell growth

Regulation of cell growth


trigger autoimmune responses (Casciola-Rosen et al., 1995; Casiano et al., 1996). In this

context, it has been shown that the presence of large numbers of apoptotic cells can evoke

an immune response. Intravenous injection of apoptotic thymocytes induced the production

of autoantibodies to nuclear antigens in most normal mice (Mevorach et al., 1998). Moreover,

in a recent study by Donnelly et al. (Donnelly et al., 2006) it was shown that neutrophils from

SLE patients are less likely to be recognized and removed by the C1q/calreticulin/CD91-

mediated apoptotic pathway, despite the presence of the main apoptotic recognition

partners. This finding indicates that an unidentified component acts as a C1q binding partner

on apoptotic cells, and that it may be absent in cells isolated from SLE patients. Interestingly,

an MFG-E8 mutant, designated D89E and carrying a point mutation in an RGD motif,

inhibited the uptake of apoptotic cells by phagocytes and also inhibited the enhanced

production of IL-10 by thioglycollate-elicited peritoneal macrophages (Asano et al., 2004).

After intravenous injection of D89E in mice, production of autoantibodies, including

antiphospholipids and antinuclear antibodies, were detected, with deposition of IgG in the

glomeruli.

Evidence that disturbed phagocytic uptake of apoptotic cells might be a factor in the

pathogenesis of human autoimmune disease was demonstrated in the case of SLE

(Herrmann et al., 1998). In vitro phagocytosis of apoptotic cells by differentiated

macrophages from SLE patients is impaired. A mouse model in a C57BL/6 background, in

which the membrane tyrosine kinase c-MER or MFG-E8 is deleted, exhibits impaired

clearance of apoptotic cells and autoimmunity (Cohen et al., 2002; Hanayama et al., 2004;

Mitchell et al., 2002). However, these data contrast with the findings that macrophages from

Balb/c CD14-deficient mice are less able to bind and internalize apoptotic cells both in vitro

and in vivo, but exhibit no increase in anti-nuclear antibodies, and no obvious changes in the

secretion of pro- or anti-inflammatory cytokines (Devitt et al., 2004). In the same way, MBL-

deficient mice in a lupus-prone genetic background (129 x C57BL/6) also exhibit reduced

apoptotic cell clearance (as measured by FACS analysis), but no increase in antinuclear

antibodies or proteinurea, even at advanced age (Stuart et al., 2005). Interpretation of results

obtained from different mouse strains used for modelling autoimmune diseases becomes

difficult because the genetic background is very crucial for the development of the

phenotype. For example, it has been shown that expression of autoimmunity in mice

deficient in complement component C1q is strongly influenced by backgrounds genes

(Mitchell et al., 2002). In C1q-deficient C57BL/6 mice, no evidence of an autoimmune

phenotype was found, and C1q deficiency in both the C57BL/6.lpr/lpr and MRL/MP-lpr/lpr

strains did not modify the autoimmune phenotype observed in wild type controls. However,

impaired apoptotic cell clearance in vivo in C1q-deficient MRL/MP+/+ animals was

accompanied by acceleration of the titers of antinuclear antibodies and glomerulonephritis


was seen. Although data obtained from CD-14 and MBL-deficient mice indicate that

apoptotic cell clearance is a highly complex process, these studies indicate that a defect in

apoptotic cell clearance alone is not sufficient to generate the autoimmune phenotype.

Conclusions

Efficient clearance of cells undergoing apoptotic or necrotic death is a fascinating dynamic

and critical biological process that is crucial for normal tissue homeostasis and for

modulation of immune responses. The machinery for clearing apoptotic cells is conserved

from worms to mammals. Molecules participating in recognition and engulfment of apoptotic

cells have been identified in recent years, but studies are needed to identify the molecules

involved in the recognition of necrotic cells, and the molecular pathways in macrophages that

lead to the morphologically different mechanisms for internalizing apoptotic and necrotic

cells. Moreover it will be important to resolve the controversy of whether the

phosphatidylserine receptor is indeed involved in cell clearance and/or it is a protein localized

in the nucleus and performing a function is distinct from removal of dead cells. Mounting

evidence reviewed here indicates that the interaction between macrophages and dead

(apoptotic and necrotic) cells is a complicated one. Increasing understanding of the

molecular mechanisms involved may lead to development of strategies for manipulating

phagocytic clearance of apoptotic and necrotic cells for the treatment of inflammatory and

autoimmune diseases, and for improving vaccine strategies against cancer.

Acknowledgments

We thank Wim Drijvers for the artwork and Dr. Amin Bredan for editing the manuscript.


References

Abrams JM, White K, Fessler LI, Steller H. 1993. Programmed cell death during Drosophila

embryogenesis. Development 117(1):29-43.

Albert ML, Kim JI, Birge RB. 2000. alphavbeta5 integrin recruits the CrkII-Dock180-rac1

complex for phagocytosis of apoptotic cells. Nat Cell Biol 2(12):899-905.

Alexopoulou L, Holt AC, Medzhitov R, Flavell RA. 2001. Recognition of double-stranded

RNA and activation of NF-kappaB by Toll-like receptor 3. Nature 413(6857):732-738.

Anderson HA, Maylock CA, Williams JA, Paweletz CP, Shu H, Shacter E. 2003. Serum-

derived protein S binds to phosphatidylserine and stimulates the phagocytosis of

apoptotic cells. Nat Immunol 4(1):87-91.

Appelt U, Sheriff A, Gaipl US, Kalden JR, Voll RE, Herrmann M. 2005. Viable, apoptotic

and necrotic monocytes expose phosphatidylserine: cooperative binding of the ligand

Annexin V to dying but not viable cells and implications for PS-dependent clearance.

Cell Death Differ 12(2):194-196.

Arur S, Uche UE, Rezaul K, Fong M, Scranton V, Cowan AE, Mohler W, Han DK. 2003.

Annexin I is an endogenous ligand that mediates apoptotic cell engulfment. Dev Cell

4(4):587-598.

Asano K, Miwa M, Miwa K, Hanayama R, Nagase H, Nagata S, Tanaka M. 2004. Masking of

phosphatidylserine inhibits apoptotic cell engulfment and induces autoantibody

production in mice. J Exp Med 200(4):459-467.

Baehrecke EH. 2002. How death shapes life during development. Nat Rev Mol Cell Biol

3(10):779-787.

Balasubramanian K, Schroit AJ. 1998. Characterization of phosphatidylserine-dependent

beta2-glycoprotein I macrophage interactions. Implications for apoptotic cell clearance

by phagocytes. J Biol Chem 273(44):29272-29277.

Balasubramanian K, Schroit AJ. 2003. Aminophospholipid asymmetry: A matter of life and

death. Annu Rev Physiol 65:701-734.

Barker RN, Erwig LP, Hill KS, Devine A, Pearce WP, Rees AJ. 2002. Antigen presentation

by macrophages is enhanced by the uptake of necrotic, but not apoptotic, cells. Clin

Exp Immunol 127(2):220-225.

Basu S, Binder RJ, Suto R, Anderson KM, Srivastava PK. 2000. Necrotic but not apoptotic

cell death releases heat shock proteins, which deliver a partial maturation signal to

dendritic cells and activate the NF-kappa B pathway. Int Immunol 12(11):1539-1546.

Baumann I, Kolowos W, Voll RE, Manger B, Gaipl U, Neuhuber WL, Kirchner T, Kalden

JR, Herrmann M. 2002. Impaired uptake of apoptotic cells into tingible body

macrophages in germinal centers of patients with systemic lupus erythematosus.

Arthritis Rheum 46(1):191-201.

Berden JH. 2003. Lupus nephritis: consequence of disturbed removal of apoptotic cells? Neth

J Med 61(8):233-238.

Bianchi ME, Manfredi A. 2004. Chromatin and cell death. Biochim Biophys Acta 1677(1-

3):181-186.

Binder RJ, Vatner R, Srivastava P. 2004. The heat-shock protein receptors: some answers and

more questions. Tissue Antigens 64(4):442-451.

Bondanza A, Zimmermann VS, Rovere-Querini P, Turnay J, Dumitriu IE, Stach CM, Voll

RE, Gaipl US, Bertling W, Poschl E, Kalden JR, Manfredi AA, Herrmann M. 2004.

Inhibition of phosphatidylserine recognition heightens the immunogenicity of

irradiated lymphoma cells in vivo. J Exp Med 200(9):1157-1165.

Borisenko GG, Iverson SL, Ahlberg S, Kagan VE, Fadeel B. 2004. Milk fat globule

epidermal growth factor 8 (MFG-E8) binds to oxidized phosphatidylserine:


implications for macrophage clearance of apoptotic cells. Cell Death Differ 11(8):943-

945.

Borisenko GG, Matsura T, Liu SX, Tyurin VA, Jianfei J, Serinkan FB, Kagan VE. 2003.

Macrophage recognition of externalized phosphatidylserine and phagocytosis of

apoptotic Jurkat cells--existence of a threshold. Arch Biochem Biophys 413(1):41-52.

Bose J, Gruber AD, Helming L, Schiebe S, Wegener I, Hafner M, Beales M, Kontgen F,

Lengeling A. 2004. The phosphatidylserine receptor has essential functions during

embryogenesis but not in apoptotic cell removal. J Biol 3(4):15.

Bottcher A, Gaipl US, Furnrohr BG, Herrmann M, Girkontaite I, Kalden JR, Voll RE. 2006.

Involvement of phosphatidylserine, alphavbeta3, CD14, CD36, and complement C1q

in the phagocytosis of primary necrotic lymphocytes by macrophages. Arthritis

Rheum 54(3):927-938.

Brouckaert G, Kalai M, Krysko DV, Saelens X, Vercammen D, Ndlovu, Haegeman G,

D'Herde K, Vandenabeele P. 2004. Phagocytosis of necrotic cells by macrophages is

phosphatidylserine dependent and does not induce inflammatory cytokine production.

Mol Biol Cell 15(3):1089-1100.

Brown S, Heinisch I, Ross E, Shaw K, Buckley CD, Savill J. 2002. Apoptosis disables CD31-

mediated cell detachment from phagocytes promoting binding and engulfment. Nature

418(6894):200-203.

Brown SB, Savill J. 1999. Phagocytosis triggers macrophage release of Fas ligand and

induces apoptosis of bystander leukocytes. J Immunol 162(1):480-485.

Byrne A, Reen DJ. 2002. Lipopolysaccharide induces rapid production of IL-10 by

monocytes in the presence of apoptotic neutrophils. J Immunol 168(4):1968-1977.

Casares N, Pequignot MO, Tesniere A, Ghiringhelli F, Roux S, Chaput N, Schmitt E, Hamai

A, Hervas-Stubbs S, Obeid M, Coutant F, Metivier D, Pichard E, Aucouturier P,

Pierron G, Garrido C, Zitvogel L, Kroemer G. 2005. Caspase-dependent

immunogenicity of doxorubicin-induced tumor cell death. J Exp Med 202(12):1691-

1701.

Casciola-Rosen LA, Anhalt GJ, Rosen A. 1995. DNA-dependent protein kinase is one of a

subset of autoantigens specifically cleaved early during apoptosis. J Exp Med

182(6):1625-1634.

Casiano CA, Martin SJ, Green DR, Tan EM. 1996. Selective cleavage of nuclear autoantigens

during CD95 (Fas/APO-1)-mediated T cell apoptosis. J Exp Med 184(2):765-770.



Chen J, Carey K, Godowski PJ. 1997. Identification of Gas6 as a ligand for Mer, a neural cell

adhesion molecule related receptor tyrosine kinase implicated in cellular

transformation. Oncogene 14(17):2033-2039.

Chen Q, Stone PR, McCowan LM, Chamley LW. 2006. Phagocytosis of necrotic but not

apoptotic trophoblasts induces endothelial cell activation. Hypertension 47(1):116-

121.

Chen W, Frank ME, Jin W, Wahl SM. 2001. TGF-beta released by apoptotic T cells

contributes to an immunosuppressive milieu. Immunity 14(6):715-725.

Cikala M, Alexandrova O, David CN, Proschel M, Stiening B, Cramer P, Bottger A. 2004.

The phosphatidylserine receptor from Hydra is a nuclear protein with potential Fe(II)

dependent oxygenase activity. BMC Cell Biol 5:26.

Ciurana CL, Zwart B, van Mierlo G, Hack CE. 2004. Complement activation by necrotic cells

in normal plasma environment compares to that by late apoptotic cells and involves

predominantly IgM. Eur J Immunol 34(9):2609-2619.


Clemens MJ, Bushell M, Jeffrey IW, Pain VM, Morley SJ. 2000. Translation initiation factor

modifications and the regulation of protein synthesis in apoptotic cells. Cell Death

Differ 7(7):603-615.

Clissold PM, Ponting CP. 2001. JmjC: cupin metalloenzyme-like domains in jumonji, hairless

and phospholipase A2beta. Trends Biochem Sci 26(1):7-9.

Cocco RE, Ucker DS. 2001. Distinct modes of macrophage recognition for apoptotic and

necrotic cells are not specified exclusively by phosphatidylserine exposure. Mol Biol

Cell 12(4):919-930.

Cohen PL, Caricchio R, Abraham V, Camenisch TD, Jennette JC, Roubey RA, Earp HS,

Matsushima G, Reap EA. 2002. Delayed apoptotic cell clearance and lupus-like

autoimmunity in mice lacking the c-mer membrane tyrosine kinase. J Exp Med

196(1):135-140.

Cui P, Qin B, Liu N, Pan G, Pei D. 2004. Nuclear localization of the phosphatidylserine

receptor protein via multiple nuclear localization signals. Exp Cell Res 293(1):154-

163.

Cvetanovic M, Ucker DS. 2004. Innate immune discrimination of apoptotic cells: repression

of proinflammatory macrophage transcription is coupled directly to specific

recognition. J Immunol 172(2):880-889.

de Almeida CJ, Linden R. 2005. Phagocytosis of apoptotic cells: a matter of balance. Cell

Mol Life Sci 62(14):1532-1546.

deBakker CD, Haney LB, Kinchen JM, Grimsley C, Lu M, Klingele D, Hsu PK, Chou BK,

Cheng LC, Blangy A, Sondek J, Hengartner MO, Wu YC, Ravichandran KS. 2004.

Phagocytosis of apoptotic cells is regulated by a UNC-73/TRIO-MIG-2/RhoG

signaling module and armadillo repeats of CED-12/ELMO. Curr Biol 14(24):2208-

2216.

Devitt A, Parker KG, Ogden CA, Oldreive C, Clay MF, Melville LA, Bellamy CO, Lacy-

Hulbert A, Gangloff SC, Goyert SM, Gregory CD. 2004. Persistence of apoptotic cells

without autoimmune disease or inflammation in CD14-/- mice. J Cell Biol

167(6):1161-1170.

Diez-Roux G, Argilla M, Makarenkova H, Ko K, Lang RA. 1999. Macrophages kill capillary

cells in G1 phase of the cell cycle during programmed vascular regression.

Development 126(10):2141-2147.

Dillon SR, Constantinescu A, Schlissel MS. 2001. Annexin V binds to positively selected B

cells. J Immunol 166(1):58-71.

Dillon SR, Mancini M, Rosen A, Schlissel MS. 2000. Annexin V binds to viable B cells and

colocalizes with a marker of lipid rafts upon B cell receptor activation. J Immunol

164(3):1322-1332.

Donnelly S, Roake W, Brown S, Young P, Naik H, Wordsworth P, Isenberg DA, Reid KB,

Eggleton P. 2006. Impaired recognition of apoptotic neutrophils by the

C1q/calreticulin and CD91 pathway in systemic lupus erythematosus. Arthritis Rheum

54(5):1543-1556.

Duffield JS, Erwig LP, Wei X, Liew FY, Rees AJ, Savill JS. 2000. Activated macrophages

direct apoptosis and suppress mitosis of mesangial cells. J Immunol 164(4):2110-

2119.

Elliott JI, Surprenant A, Marelli-Berg FM, Cooper JC, Cassady-Cain RL, Wooding C, Linton

K, Alexander DR, Higgins CF. 2005. Membrane phosphatidylserine distribution as a

non-apoptotic signalling mechanism in lymphocytes. Nat Cell Biol 7(8):808-816.

Ellis RE, Jacobson DM, Horvitz HR. 1991. Genes required for the engulfment of cell corpses

during programmed cell death in Caenorhabditis elegans. Genetics 129(1):79-94.


Emoto K, Toyama-Sorimachi N, Karasuyama H, Inoue K, Umeda M. 1997. Exposure of

phosphatidylethanolamine on the surface of apoptotic cells. Exp Cell Res 232(2):430-

434.

Fadeel B. 2003. Programmed cell clearance. Cell Mol Life Sci 60(12):2575-2585.

Fadeel B, Orrenius S, Pervaiz S. 2004. Buried alive: a novel approach to cancer treatment.

Faseb J 18(1):1-4.

Fadok VA, Bratton DL, Frasch SC, Warner ML, Henson PM. 1998a. The role of

phosphatidylserine in recognition of apoptotic cells by phagocytes. Cell Death Differ

5(7):551-562.

Fadok VA, Bratton DL, Guthrie L, Henson PM. 2001a. Differential effects of apoptotic

versus lysed cells on macrophage production of cytokines: role of proteases. J

Immunol 166(11):6847-6854.

Fadok VA, Bratton DL, Konowal A, Freed PW, Westcott JY, Henson PM. 1998b.

Macrophages that have ingested apoptotic cells in vitro inhibit proinflammatory

cytokine production through autocrine/paracrine mechanisms involving TGF-beta,

PGE2, and PAF. J Clin Invest 101(4):890-898.

Fadok VA, Bratton DL, Rose DM, Pearson A, Ezekewitz RA, Henson PM. 2000. A receptor

for phosphatidylserine-specific clearance of apoptotic cells. Nature 405(6782):85-90.

Fadok VA, de Cathelineau A, Daleke DL, Henson PM, Bratton DL. 2001b. Loss of

phospholipid asymmetry and surface exposure of phosphatidylserine is required for

phagocytosis of apoptotic cells by macrophages and fibroblasts. J Biol Chem

276(2):1071-1077.

Fadok VA, Henson PM. 2003. Apoptosis: giving phosphatidylserine recognition an assist--

with a twist. Curr Biol 13(16):R655-657.

Fadok VA, Voelker DR, Campbell PA, Cohen JJ, Bratton DL, Henson PM. 1992. Exposure

of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific

recognition and removal by macrophages. J Immunol 148(7):2207-2216.

Fadok VA, Warner ML, Bratton DL, Henson PM. 1998c. CD36 is required for phagocytosis

of apoptotic cells by human macrophages that use either a phosphatidylserine receptor

or the vitronectin receptor (alpha v beta 3). J Immunol 161(11):6250-6257.

Ferrari D, Wesselborg S, Bauer MK, Schulze-Osthoff K. 1997. Extracellular ATP activates

transcription factor NF-kappaB through the P2Z purinoreceptor by selectively

targeting NF-kappaB p65. J Cell Biol 139(7):1635-1643.

Franc NC. 2002. Phagocytosis of apoptotic cells in mammals, caenorhabditis elegans and

Drosophila melanogaster: molecular mechanisms and physiological consequences.

Front Biosci 7:d1298-1313.

Franc NC, Heitzler P, Ezekowitz RA, White K. 1999. Requirement for croquemort in

phagocytosis of apoptotic cells in Drosophila. Science 284(5422):1991-1994.

Fujii C, Shiratsuchi A, Manaka J, Yonehara S, Nakanishi Y. 2001. Difference in the way of

macrophage recognition of target cells depending on their apoptotic states. Cell Death

Differ 8(11):1113-1122.

Gaipl US, Kuenkele S, Voll RE, Beyer TD, Kolowos W, Heyder P, Kalden JR, Herrmann M.

2001. Complement binding is an early feature of necrotic and a rather late event

during apoptotic cell death. Cell Death Differ 8(4):327-334.

Gao Y, Herndon JM, Zhang H, Griffith TS, Ferguson TA. 1998. Antiinflammatory effects of

CD95 ligand (FasL)-induced apoptosis. J Exp Med 188(5):887-896.

Gardai SJ, McPhillips KA, Frasch SC, Janssen WJ, Starefeldt A, Murphy-Ullrich JE, Bratton

DL, Oldenborg PA, Michalak M, Henson PM. 2005. Cell-surface calreticulin initiates

clearance of viable or apoptotic cells through trans-activation of LRP on the

phagocyte. Cell 123(2):321-334.


Goldszmid RS, Idoyaga J, Bravo AI, Steinman R, Mordoh J, Wainstok R. 2003. Dendritic

cells charged with apoptotic tumor cells induce long-lived protective CD4+ and CD8+

T cell immunity against B16 melanoma. J Immunol 171(11):5940-5947.

Golpon HA, Fadok VA, Taraseviciene-Stewart L, Scerbavicius R, Sauer C, Welte T, Henson

PM, Voelkel NF. 2004. Life after corpse engulfment: phagocytosis of apoptotic cells

leads to VEGF secretion and cell growth. Faseb J 18(14):1716-1718.

Gregory CD, Devitt A. 2004. The macrophage and the apoptotic cell: an innate immune

interaction viewed simplistically? Immunology 113(1):1-14.

Grimsley C, Ravichandran KS. 2003. Cues for apoptotic cell engulfment: eat-me, don't eat-me

and come-get-me signals. Trends Cell Biol 13(12):648-656.

Gumienny TL, Hengartner MO. 2001. How the worm removes corpses: the nematode C.

elegans as a model system to study engulfment. Cell Death Differ 8(6):564-568.

Hamon Y, Broccardo C, Chambenoit O, Luciani MF, Toti F, Chaslin S, Freyssinet JM,

Devaux PF, McNeish J, Marguet D, Chimini G. 2000. ABC1 promotes engulfment of

apoptotic cells and transbilayer redistribution of phosphatidylserine. Nat Cell Biol

2(7):399-406.

Hanayama R, Tanaka M, Miwa K, Shinohara A, Iwamatsu A, Nagata S. 2002. Identification

of a factor that links apoptotic cells to phagocytes. Nature 417(6885):182-187.

Hanayama R, Tanaka M, Miyasaka K, Aozasa K, Koike M, Uchiyama Y, Nagata S. 2004.

Autoimmune disease and impaired uptake of apoptotic cells in MFG-E8-deficient

mice. Science 304(5674):1147-1150.

Hart SP, Alexander KM, Dransfield I. 2004. Immune complexes bind preferentially to Fc

gamma RIIA (CD32) on apoptotic neutrophils, leading to augmented phagocytosis by

macrophages and release of proinflammatory cytokines. J Immunol 172(3):1882-1887.

Hart SP, Alexander KM, MacCall SM, Dransfield I. 2005. C-reactive protein does not

opsonize early apoptotic human neutrophils, but binds only membrane-permeable late

apoptotic cells and has no effect on their phagocytosis by macrophages. J Inflamm

(Lond) 2:5.

Hart SP, Ross JA, Ross K, Haslett C, Dransfield I. 2000. Molecular characterization of the

surface of apoptotic neutrophils: implications for functional downregulation and

recognition by phagocytes. Cell Death Differ 7(5):493-503.

Henson PM. 2005. Engulfment: ingestion and migration with Rac, Rho and TRIO. Curr Biol

15(1):R29-30.

Herrmann M, Voll RE, Zoller OM, Hagenhofer M, Ponner BB, Kalden JR. 1998. Impaired

phagocytosis of apoptotic cell material by monocyte-derived macrophages from

patients with systemic lupus erythematosus. Arthritis Rheum 41(7):1241-1250.

Hirt UA, Leist M. 2003. Rapid, noninflammatory and PS-dependent phagocytic clearance of

necrotic cells. Cell Death Differ 10(10):1156-1164.

Hodge S, Hodge G, Scicchitano R, Reynolds PN, Holmes M. 2003. Alveolar macrophages

from subjects with chronic obstructive pulmonary disease are deficient in their ability

to phagocytose apoptotic airway epithelial cells. Immunol Cell Biol 81(4):289-296.

Hoeppner DJ, Hengartner MO, Schnabel R. 2001. Engulfment genes cooperate with ced-3 to

promote cell death in Caenorhabditis elegans. Nature 412(6843):202-206.

Hong JR, Lin GH, Lin CJ, Wang WP, Lee CC, Lin TL, Wu JL. 2004. Phosphatidylserine

receptor is required for the engulfment of dead apoptotic cells and for normal

embryonic development in zebrafish. Development 131(21):5417-5427.

Huynh ML, Fadok VA, Henson PM. 2002. Phosphatidylserine-dependent ingestion of

apoptotic cells promotes TGF-beta1 secretion and the resolution of inflammation. J

Clin Invest 109(1):41-50.


Ishii KJ, Suzuki K, Coban C, Takeshita F, Itoh Y, Matoba H, Kohn LD, Klinman DM. 2001.

Genomic DNA released by dying cells induces the maturation of APCs. J Immunol

167(5):2602-2607.

Janssens S, Tinel A, Lippens S, Tschopp J. 2005. PIDD mediates NF-kappaB activation in

response to DNA damage. Cell 123(6):1079-1092.

Jones AL, Poon IK, Hulett MD, Parish CR. 2005. Histidine-rich glycoprotein specifically

binds to necrotic cells via its amino-terminal domain and facilitates necrotic cell

phagocytosis. J Biol Chem 280(42):35733-35741.

Kacani L, Wurm M, Schwentner I, Andrle J, Schennach H, Sprinzl GM. 2005. Maturation of

dendritic cells in the presence of living, apoptotic and necrotic tumour cells derived

from squamous cell carcinoma of head and neck. Oral Oncol 41(1):17-24.

Kagan VE, Gleiss B, Tyurina YY, Tyurin VA, Elenstrom-Magnusson C, Liu SX, Serinkan

FB, Arroyo A, Chandra J, Orrenius S, Fadeel B. 2002. A role for oxidative stress in

apoptosis: oxidation and externalization of phosphatidylserine is required for

macrophage clearance of cells undergoing Fas-mediated apoptosis. J Immunol

169(1):487-499.

Kahlenberg JM, Lundberg KC, Kertesy SB, Qu Y, Dubyak GR. 2005. Potentiation of

caspase-1 activation by the P2X7 receptor is dependent on TLR signals and requires

NF-kappaB-driven protein synthesis. J Immunol 175(11):7611-7622.

Kariko K, Ni H, Capodici J, Lamphier M, Weissman D. 2004. mRNA is an endogenous

ligand for Toll-like receptor 3. J Biol Chem 279(13):12542-12550.

Kim S, Elkon KB, Ma X. 2004. Transcriptional suppression of interleukin-12 gene expression

following phagocytosis of apoptotic cells. Immunity 21(5):643-653.

Kinchen JM, Cabello J, Klingele D, Wong K, Feichtinger R, Schnabel H, Schnabel R,

Hengartner MO. 2005. Two pathways converge at CED-10 to mediate actin

rearrangement and corpse removal in C. elegans. Nature 434(7029):93-99.

Krysko DV, Leybaert L, Vandenabeele P, D'Herde K. 2005. Gap junctions and the

propagation of cell survival and cell death signals. Apoptosis 10(3):459-469.

Krysko O, De Ridder L, Cornelissen M. 2004. Phosphatidylserine exposure during early



Kuijpers TW, Maianski NA, Tool AT, Becker K, Plecko B, Valianpour F, Wanders RJ,

Pereira R, Van Hove J, Verhoeven AJ, Roos D, Baas F, Barth PG. 2004. Neutrophils

in Barth syndrome (BTHS) avidly bind annexin-V in the absence of apoptosis. Blood

103(10):3915-3923.

Kunisaki Y, Masuko S, Noda M, Inayoshi A, Sanui T, Harada M, Sasazuki T, Fukui Y. 2004.

Defective fetal liver erythropoiesis and T lymphopoiesis in mice lacking the

phosphatidylserine receptor. Blood 103(9):3362-3364.

Kurushima H, Ramprasad M, Kondratenko N, Foster DM, Quehenberger O, Steinberg D.

2000. Surface expression and rapid internalization of macrosialin (mouse CD68) on

elicited mouse peritoneal macrophages. J Leukoc Biol 67(1):104-108.

Lang RA, Bishop JM. 1993. Macrophages are required for cell death and tissue remodeling in

the developing mouse eye. Cell 74(3):453-462.

Lecoeur H, Prevost MC, Gougeon ML. 2001. Oncosis is associated with exposure of

phosphatidylserine residues on the outside layer of the plasma membrane: a

reconsideration of the specificity of the annexin V/propidium iodide assay. Cytometry

44(1):65-72.




Li M, Carpio DF, Zheng Y, Bruzzo P, Singh V, Ouaaz F, Medzhitov RM, Beg AA. 2001. An

essential role of the NF-kappa B/Toll-like receptor pathway in induction of

inflammatory and tissue-repair gene expression by necrotic cells. J Immunol

166(12):7128-7135.

Li MO, Sarkisian MR, Mehal WZ, Rakic P, Flavell RA. 2003. Phosphatidylserine receptor is

required for clearance of apoptotic cells. Science 302(5650):1560-1563.

Lillehei KO, Liu Y, Kong Q. 1999. Current perspectives in immunotherapy. Ann Thorac Surg

68(3 Suppl):S28-33.

Liu QA, Hengartner MO. 1998. Candidate adaptor protein CED-6 promotes the engulfment of

apoptotic cells in C. elegans. Cell 93(6):961-972.

Lorimore SA, Coates PJ, Scobie GE, Milne G, Wright EG. 2001. Inflammatory-type

responses after exposure to ionizing radiation in vivo: a mechanism for radiation-

induced bystander effects? Oncogene 20(48):7085-7095.

Manaka J, Kuraishi T, Shiratsuchi A, Nakai Y, Higashida H, Henson P, Nakanishi Y. 2004.

Draper-mediated and phosphatidylserine-independent phagocytosis of apoptotic cells

by Drosophila hemocytes/macrophages. J Biol Chem 279(46):48466-48476.

Mevorach D, Mascarenhas JO, Gershov D, Elkon KB. 1998. Complement-dependent

clearance of apoptotic cells by human macrophages. J Exp Med 188(12):2313-2320.

Mitchell DA, Pickering MC, Warren J, Fossati-Jimack L, Cortes-Hernandez J, Cook HT,

Botto M, Walport MJ. 2002. C1q deficiency and autoimmunity: the effects of genetic

background on disease expression. J Immunol 168(5):2538-2543.

Mitchell JE, Cvetanovic M, Tibrewal N, Patel V, Colamonici OR, Li MO, Flavell RA, Levine

JS, Birge RB, Ucker DS. 2006. The Presumptive Phosphatidylserine Receptor Is

Dispensable for Innate Anti-inflammatory Recognition and Clearance of Apoptotic

Cells. J Biol Chem 281(9):5718-5725.

Morimoto K, Amano H, Sonoda F, Baba M, Senba M, Yoshimine H, Yamamoto H, Ii T,

Oishi K, Nagatake T. 2001. Alveolar macrophages that phagocytose apoptotic

neutrophils produce hepatocyte growth factor during bacterial pneumonia in mice. Am

J Respir Cell Mol Biol 24(5):608-615.

Munoz LE, Gaipl US, Franz S, Sheriff A, Voll RE, Kalden JR, Herrmann M. 2005. SLE--a

disease of clearance deficiency? Rheumatology (Oxford) 44(9):1101-1107.

Nakai Y, Shiratsuchi A, Manaka J, Nakayama H, Takio K, Zhang JT, Suganuma T, Nakanishi

Y. 2005. Externalization and recognition by macrophages of large subunit of

eukaryotic translation initiation factor 3 in apoptotic cells. Exp Cell Res 309(1):137-

148.

Nauta AJ, Raaschou-Jensen N, Roos A, Daha MR, Madsen HO, Borrias-Essers MC, Ryder

LP, Koch C, Garred P. 2003. Mannose-binding lectin engagement with late apoptotic

and necrotic cells. Eur J Immunol 33(10):2853-2863.

Nauta AJ, Trouw LA, Daha MR, Tijsma O, Nieuwland R, Schwaeble WJ, Gingras AR,

Mantovani A, Hack EC, Roos A. 2002. Direct binding of C1q to apoptotic cells and

cell blebs induces complement activation. Eur J Immunol 32(6):1726-1736.

O'Brien BA, Fieldus WE, Field CJ, Finegood DT. 2002. Clearance of apoptotic beta-cells is

reduced in neonatal autoimmune diabetes-prone rats. Cell Death Differ 9(4):457-464.

Park JS, Svetkauskaite D, He Q, Kim JY, Strassheim D, Ishizaka A, Abraham E. 2004.

Involvement of toll-like receptors 2 and 4 in cellular activation by high mobility group

box 1 protein. J Biol Chem 279(9):7370-7377.

Patel VA, Longacre A, Hsiao K, Fan H, Meng F, Mitchell JE, Rauch J, Ucker DS, Levine JS.

2006. Apoptotic Cells, at All Stages of the Death Process, Trigger Characteristic

Signaling Events That Are Divergent from and Dominant over Those Triggered by


Necrotic Cells: IMPLICATIONS FOR THE DELAYED CLEARANCE MODEL OF

AUTOIMMUNITY. J Biol Chem 281(8):4663-4670.

Platt N, da Silva RP, Gordon S. 1998. Recognizing death: the phagocytosis of apoptotic cells.

Trends Cell Biol 8(9):365-372.

Ramanathan M, Giladi A, Leibovich SJ. 2003. Regulation of vascular endothelial growth

factor gene expression in murine macrophages by nitric oxide and hypoxia. Exp Biol

Med (Maywood) 228(6):697-705.

Reddien PW, Cameron S, Horvitz HR. 2001. Phagocytosis promotes programmed cell death

in C. elegans. Nature 412(6843):198-202.

Reddien PW, Horvitz HR. 2004. The engulfment process of programmed cell death in

caenorhabditis elegans. Annu Rev Cell Dev Biol 20:193-221.

Reddy SM, Hsiao KH, Abernethy VE, Fan H, Longacre A, Lieberthal W, Rauch J, Koh JS,

Levine JS. 2002. Phagocytosis of apoptotic cells by macrophages induces novel

signaling events leading to cytokine-independent survival and inhibition of

proliferation: activation of Akt and inhibition of extracellular signal-regulated kinases

1 and 2. J Immunol 169(2):702-713.

Ren Y, Stuart L, Lindberg FP, Rosenkranz AR, Chen Y, Mayadas TN, Savill J. 2001.

Nonphlogistic clearance of late apoptotic neutrophils by macrophages: efficient

phagocytosis independent of beta 2 integrins. J Immunol 166(7):4743-4750.

Ren Y, Tang J, Mok MY, Chan AW, Wu A, Lau CS. 2003. Increased apoptotic neutrophils

and macrophages and impaired macrophage phagocytic clearance of apoptotic

neutrophils in systemic lupus erythematosus. Arthritis Rheum 48(10):2888-2897.

Roos A, Xu W, Castellano G, Nauta AJ, Garred P, Daha MR, van Kooten C. 2004. Mini-

review: A pivotal role for innate immunity in the clearance of apoptotic cells. Eur J

Immunol 34(4):921-929.

Rovere P, Peri G, Fazzini F, Bottazzi B, Doni A, Bondanza A, Zimmermann VS, Garlanda C,

Fascio U, Sabbadini MG, Rugarli C, Mantovani A, Manfredi AA. 2000. The long

pentraxin PTX3 binds to apoptotic cells and regulates their clearance by antigen-

presenting dendritic cells. Blood 96(13):4300-4306.

Ryoo HD, Gorenc T, Steller H. 2004. Apoptotic cells can induce compensatory cell

proliferation through the JNK and the Wingless signaling pathways. Dev Cell

7(4):491-501.

Saelens X, Festjens N, Parthoens E, Vanoverberghe I, Kalai M, van Kuppeveld F,

Vandenabeele P. 2005. Protein synthesis persists during necrotic cell death. J Cell Biol

168(4):545-551.

Saelens X, Kalai M, Vandenabeele P. 2001. Translation inhibition in apoptosis: caspase-

dependent PKR activation and eIF2-alpha phosphorylation. J Biol Chem

276(45):41620-41628.

Salani D, Di Castro V, Nicotra MR, Rosano L, Tecce R, Venuti A, Natali PG, Bagnato A.

2000a. Role of endothelin-1 in neovascularization of ovarian carcinoma. Am J Pathol

157(5):1537-1547.

Salani D, Taraboletti G, Rosano L, Di Castro V, Borsotti P, Giavazzi R, Bagnato A. 2000b.

Endothelin-1 induces an angiogenic phenotype in cultured endothelial cells and

stimulates neovascularization in vivo. Am J Pathol 157(5):1703-1711.

Sauter B, Albert ML, Francisco L, Larsson M, Somersan S, Bhardwaj N. 2000. Consequences

of cell death: exposure to necrotic tumor cells, but not primary tissue cells or apoptotic

cells, induces the maturation of immunostimulatory dendritic cells. J Exp Med

191(3):423-434.

Savill J, Dransfield I, Gregory C, Haslett C. 2002. A blast from the past: clearance of

apoptotic cells regulates immune responses. Nat Rev Immunol 2(12):965-975.


Savill J, Hogg N, Ren Y, Haslett C. 1992. Thrombospondin cooperates with CD36 and the

vitronectin receptor in macrophage recognition of neutrophils undergoing apoptosis. J

Clin Invest 90(4):1513-1522.

Scaffidi P, Misteli T, Bianchi ME. 2002. Release of chromatin protein HMGB1 by necrotic

cells triggers inflammation. Nature 418(6894):191-195.

Scheffer SR, Nave H, Korangy F, Schlote K, Pabst R, Jaffee EM, Manns MP, Greten TF.

2003. Apoptotic, but not necrotic, tumor cell vaccines induce a potent immune

response in vivo. Int J Cancer 103(2):205-211.

Schlegel RA, Williamson P. 2001. Phosphatidylserine, a death knell. Cell Death Differ

8(6):551-563.

Sharkey AM, Charnock-Jones DS, Boocock CA, Brown KD, Smith SK. 1993. Expression of

mRNA for vascular endothelial growth factor in human placenta. J Reprod Fertil

99(2):609-615.

Shi Y, Evans JE, Rock KL. 2003. Molecular identification of a danger signal that alerts the

immune system to dying cells. Nature 425(6957):516-521.

Shinohara K, Shinohara T, Mochizuki N, Mochizuki Y, Sawa H, Kohya T, Fujita M, Fujioka

Y, Kitabatake A, Nagashima K. 1996. Expression of vascular endothelial growth

factor in human myocardial infarction. Heart Vessels 11(3):113-122.

Shweiki D, Itin A, Soffer D, Keshet E. 1992. Vascular endothelial growth factor induced by

hypoxia may mediate hypoxia-initiated angiogenesis. Nature 359(6398):843-845.

Skoberne M, Beignon AS, Bhardwaj N. 2004. Danger signals: a time and space continuum.

Trends Mol Med 10(6):251-257.

Stark MA, Huo Y, Burcin TL, Morris MA, Olson TS, Ley K. 2005. Phagocytosis of apoptotic

neutrophils regulates granulopoiesis via IL-23 and IL-17. Immunity 22(3):285-294.

Stefansson S, McMahon GA, Petitclerc E, Lawrence DA. 2003. Plasminogen activator

inhibitor-1 in tumor growth, angiogenesis and vascular remodeling. Curr Pharm Des

9(19):1545-1564.

Stuart LM, Takahashi K, Shi L, Savill J, Ezekowitz RA. 2005. Mannose-binding lectin-

deficient mice display defective apoptotic cell clearance but no autoimmune

phenotype. J Immunol 174(6):3220-3226.

Su HP, Nakada-Tsukui K, Tosello-Trampont AC, Li Y, Bu G, Henson PM, Ravichandran KS.

2002. Interaction of CED-6/GULP, an adapter protein involved in engulfment of

apoptotic cells with CED-1 and CD91/low density lipoprotein receptor-related protein

(LRP). J Biol Chem 277(14):11772-11779.

Trinite B, Chauvin C, Peche H, Voisine C, Heslan M, Josien R. 2005. Immature CD4-

CD103+ rat dendritic cells induce rapid caspase-independent apoptosis-like cell death

in various tumor and nontumor cells and phagocytose their victims. J Immunol

175(4):2408-2417.

Tsan MF. 2006. Toll-like receptors, inflammation and cancer. Semin Cancer Biol 16(1):32-

37.

Vanden Berghe T, Kalai M, Denecker G, Meeus A, Saelens X, Vandenabeele P. 2006.

Necrosis is associated with IL-6 production but apoptosis is not. Cell Signal

18(3):328-335.

Vandivier RW, Fadok VA, Hoffmann PR, Bratton DL, Penvari C, Brown KK, Brain JD,

Accurso FJ, Henson PM. 2002. Elastase-mediated phosphatidylserine receptor

cleavage impairs apoptotic cell clearance in cystic fibrosis and bronchiectasis. J Clin

Invest 109(5):661-670.

Wanderley JL, Benjamin A, Real F, Bonomo A, Moreira ME, Barcinski MA. 2005. Apoptotic

mimicry: an altruistic behavior in host/Leishmania interplay. Braz J Med Biol Res

38(6):807-812.


Wanderley JL, Moreira ME, Benjamin A, Bonomo AC, Barcinski MA. 2006. Mimicry of

apoptotic cells by exposing phosphatidylserine participates in the establishment of

amastigotes of Leishmania (L) amazonensis in mammalian hosts. J Immunol

176(3):1834-1839.

Wang X, Wu YC, Fadok VA, Lee MC, Gengyo-Ando K, Cheng LC, Ledwich D, Hsu PK,

Chen JY, Chou BK, Henson P, Mitani S, Xue D. 2003. Cell corpse engulfment

mediated by C. elegans phosphatidylserine receptor through CED-5 and CED-12.

Science 302(5650):1563-1566.

Ware LB, Matthay MA. 2002. Keratinocyte and hepatocyte growth factors in the lung: roles

in lung development, inflammation, and repair. Am J Physiol Lung Cell Mol Physiol

282(5):L924-940.

Wilkin F, Duhant X, Bruyns C, Suarez-Huerta N, Boeynaems JM, Robaye B. 2001. The

P2Y11 receptor mediates the ATP-induced maturation of human monocyte-derived

dendritic cells. J Immunol 166(12):7172-7177.

Williamson P, Schlegel RA. 2002. Transbilayer phospholipid movement and the clearance of

apoptotic cells. Biochim Biophys Acta 1585(2-3):53-63.

Wong OH, Huang FP, Chiang AK. 2005. Differential responses of cord and adult blood-

derived dendritic cells to dying cells. Immunology 116(1):13-20.

Wu YC, Horvitz HR. 1998. The C. elegans cell corpse engulfment gene ced-7 encodes a

protein similar to ABC transporters. Cell 93(6):951-960.

Wu YC, Tsai MC, Cheng LC, Chou CJ, Weng NY. 2001. C. elegans CED-12 acts in the

conserved crkII/DOCK180/Rac pathway to control cell migration and cell corpse

engulfment. Dev Cell 1(4):491-502.

Xiao YQ, Malcolm K, Worthen GS, Gardai S, Schiemann WP, Fadok VA, Bratton DL,

Henson PM. 2002. Cross-talk between ERK and p38 MAPK mediates selective

suppression of pro-inflammatory cytokines by transforming growth factor-beta. J Biol

Chem 277(17):14884-14893.

Zheng L, He M, Long M, Blomgran R, Stendahl O. 2004. Pathogen-induced apoptotic

neutrophils express heat shock proteins and elicit activation of human macrophages. J

Immunol 173(10):6319-6326.

Zhou Z, Hartwieg E, Horvitz HR. 2001. CED-1 is a transmembrane receptor that mediates

cell corpse engulfment in C. elegans. Cell 104(1):43-56.

Zwart B, Ciurana C, Rensink I, Manoe R, Hack CE, Aarden LA. 2004. Complement

activation by apoptotic cells occurs predominantly via IgM and is limited to late

apoptotic (secondary necrotic) cells. Autoimmunity 37(2):95-102.

Aim of the study

How can my Muse want subject to invent,While thou dost breathe, that pour'st into my verseThine own sweet argument, too excellentFor every vulgar paper to rehearse?

If my slight Muse do please these curious days,The pain be mine, but thine shall be the praise.

Sonnet 38, How can my Muse want subject to inventW Shakespear

The Thinker, A Rodin

67

Aim of the study

Apoptotic cell death plays a fundamental role in the development and maintenance of tissue

homeostasis (Meier et al., 2000; Opferman and Korsmeyer, 2003). Therefore, insight into the

regulation of apoptosis is crucial to understand both the development of a number of

pathologies as well as the possible therapeutic intervention targets. As the control of

apoptotic cell death can be allocated to several levels, we addressed at each respective level

the following issues:

1. Apoptosis is regulated at the intracellular level by activation of either intrinsic or extrinsic

cell death pathways where mitochondria play a central role by releasing multiple proteins

from mitochondrial space (Newmeyer and Ferguson-Miller, 2003). One of the released

mitochondrial intermembrane space proteins is cytochrome c, an important component of

the mitochondrial respiratory chain. The concept of heterogeneity in the mitochondrial

population with respect to cytochrome c release remains controversial (D'Herde et al.,

2000; Goldstein et al., 2005; Goldstein et al., 2000). Therefore we undertook an

additional study to further analyse the heterogeneity issue at a functional level, namely by

investigating the loss of mitochondrial transmembrane potential ( m). m was

evaluated in a granulosa explant model by applying an aldehyde–fixable transmembrane

potential probe together with DAPI staining, and the appropriate controls for

mitochondrial staining and apoptosis induction. Assessment of m changes was done

qualitatively by epifluorescence microscopy and quantified using digital imaging

microscopy. In this part we address the following issue:

How homogenous is the mitochondrial response to apoptotic signals with

regard to mitochondrial transmembrane potential? (Krysko et al., Journal of

Histochemistry and Cytochemistry, 2001)

2. That the decision of which cells die and which cells live must be regulated to some extent

by the organizing microenvironment (Jacobson et al., 1997), imposes us to think of a

regulation level of apoptosis by intercellular communication. At the intercellular level one

of the possible control mechanisms of cell death can be achieved via direct

communication between cells, mediated by gap junctional communication. Only recently,

the exploration of the involvement of gap junctions in cell death has emerged and

therefore their role in the propagation of cell death remains unclear. Thus, in the second

part we would like to deal with the following questions by using FRAP analysis,

68 Aim of the study

connexin43 immunocytochemistry and experimental inhibition of gap junctional

communication for apoptotic and control granulosa explants:

What is the level of gap junctional coupling during initiation of apoptosis?

(Krysko et al., Journal of Histochemistry and Cytochemistry, 2004)

Does gap junctional coupling contribute to the propagation of cell survival and

cell death? (Krysko et al., Journal of Histochemistry and Cytochemistry, 2004)

3. In the third part of the thesis we focus on interactions between macrophages and two

fundamentally distinct types of dying cells; namely, apoptotic versus necrotic cells. The

phagocytic process can be considered as the final level of regulation of the cell death

process. It is well known that swift and efficient removal of dying cells is essential for the

embryonic morphogenesis and for the daily control of tissue homeostasis. Moreover,

disturbances in the clearance of dead cells may lead to a myriad of pathologies.

Interestingly, intensive studies of the last decade have provided a deep inside view into

the key signals displayed by apoptotic cells, the bridging and receptor molecules on the

phagocyte while, in contrast, at the time when this study was started, only a few

investigators studied the disposal of the necrotic cells, another morphological and

biochemical type of cell death. The process of interaction of macrophages with dying

cells can be divided into different stages, starting with the recognition phase, followed by

a signalling phase and ending with internalization (Gregory and Devitt, 2004).

In this part of our study we used an in vitro phagocytosis assay in which a macrophage

cell line (Mf4/4) was cocultured with apoptotic and necrotic L929sAhFas cells. We used

transmission and scanning electron microscopy, fluorescence microscopy, a dual color

flow cytometric and RNAase protection assays and Mf4/4 containing a NF- B regulated

luciferase reporter gene in order to elucidate the following issues:

On the binding and signalling stage:

To determine the role of phosphatidylserine (PS) in the uptake of apoptotic

versus necrotic cells (Brouckaert G, Kalai M, Krysko DV et al., Molecular Biology of

the Cell, 2004)

To examine whether either type of cell death induces anti- or pro-inflammatory

responses (Brouckaert G, Kalai M, Krysko DV et al., Molecular Biology of the Cell,

2004)

69

To investigate whether either type of cell death induces activation of NF- B

transcription factor1

On the internalization stage:

To explore the morphological mechanisms of internalization used by the

macrophages to engulf apoptotic versus necrotic cells (Krysko et al., Journal of

Morphology, 2003; Krysko et al., Cell Death and Differentiation, 2006)

1These are unpublished results

References: D'Herde K, De Prest B, Mussche S, Schotte P, Beyaert R, Coster RV, Roels F. 2000.

Ultrastructural localization of cytochrome c in apoptosis demonstrates mitochondrial heterogeneity. Cell Death Differ 7(4):331-337.

Goldstein JC, Munoz-Pinedo C, Ricci JE, Adams SR, Kelekar A, Schuler M, Tsien RY, Green DR. 2005. Cytochrome c is released in a single step during apoptosis. Cell Death Differ 12(5):453-462.

Goldstein JC, Waterhouse NJ, Juin P, Evan GI, Green DR. 2000. The coordinate release of cytochrome c during apoptosis is rapid, complete and kinetically invariant. Nat Cell Biol 2(3):156-162.

Gregory CD, Devitt A. 2004. The macrophage and the apoptotic cell: an innate immune interaction viewed simplistically? Immunology 113(1):1-14.

Jacobson MD, Weil M, Raff MC. 1997. Programmed cell death in animal development. Cell 88(3):347-354.

Meier P, Finch A, Evan G. 2000. Apoptosis in development. Nature 407(6805):796-801. Newmeyer DD, Ferguson-Miller S. 2003. Mitochondria: releasing power for life and

unleashing the machineries of death. Cell 112(4):481-490. Opferman JT, Korsmeyer SJ. 2003. Apoptosis in the development and maintenance of the

immune system. Nat Immunol 4(5):410-415.

Results

1. Intracellular level

2. Intercellular level

3. Burial level

Mitochondrial heterogeneityduring apoptotic cell death

How much more praise deserved thy beauty's use,If thou couldst answer 'This fair child of mineShall sum my count and make my old excuse,'Proving his beauty by succession thine!This were to be new made when thou art old,And see thy blood warm when thou feel'st it cold.

Sonnet 2, When forty winters shall besiege thy browW Shakespear

The Centaur, A Rodin

1. Intracellular level

Mitochondrial transmembrane potential changes support the concept of mitochondrial heterogeneity during apoptosis

DV Krysko, F Roels, L Leybaert and K D’Herde The Journal of Histochemistry and Cytochemistry 2001; 49(10): 1277–1284.

Mitochondrial Transmembrane Potential Changes Supportthe Concept of Mitochondrial Heterogeneity During Apoptosis

Dmitri V. Krysko, Frank Roels, Luc Leybaert, and Katharina D’Herde

Department of Human Anatomy, Embryology, Histology and Medical Physics (DVK,FR,KD), and Department of Physiology and Pathophysiology (LL), Ghent University, Ghent, Belgium

SUMMARY Dissipation of mitochondrial membrane potential (

��m) and release of cyto-chrome c from mitochondria appear to be key events during apoptosis. The precise rela-tionship (cause or consequence) between both is currently unclear. We previously showedin a model of serum-free cultured granulosa explants that cytochrome c is retained in asubset of respiring mitochondria until late in the apoptotic process. In this study we furtherinvestigated the issue of heterogeneity by using the

��m-sensitive probe CM-H2TMRos incombination with a DNA fluorochrome. Changes of

��m were assessed qualitatively byepifluorescence microscopy and were quantified using digital imaging microscopy. This ap-proach yielded the following results: (a) CM-H2TMRos staining is a reliable and specific pro-cedure to detect

��m changes in granulosa cells explants; (b) dissipation of transmem-brane potential is an early event during apoptosis preceding nuclear changes but isconfined to a subpopulation of mitochondria within an individual cell; (c) in frankly apop-totic cells a few polarized mitochondria can be detected. These findings support the hy-pothesis that ATP needed for completion of the apoptotic cascade can be generated dur-ing apoptosis in a subset of respiring mitochondria and is not necessarily derived fromanaerobic glycolysis. (J Histochem Cytochem 49:1277–1284, 2001)

Since mitochondria were discovered in 1840,they have provided fertile ground for scientific in-quiry. Given the importance of mitochondria for celllife, it comes now as no surprise that mitochondrialdysfunction and failure lead to apoptotic and necroticcell death. Apoptosis and necrosis are two forms ofcell death with clearly distinguishing morphologicaland biochemical features (Wyllie et al. 1980). Recentadvances suggest, however, that apoptosis and necro-sis have some common steps. Bcl-2 and caspase inhib-itors had been believed to specifically inhibit apopto-sis, but preventive effects of Bcl-2 on necrosis were alsoproved by the demonstration that Bcl-2 and its rela-tive, Bcl-xL, inhibit necrotic cell death induced by ox-ygen depletion, respiratory chain inhibitors such asKCN and antimycin A, or by glutathione depletion

(Tsujimoto 1997). Necrotic cell death was also re-tarded by caspase inhibitors, including tetrapeptide in-hibitors and a serpin, CrmA, derived from cowpox vi-rus (Tsujimoto 1997). Moreover, the mitochondrialpermeability transition (MPT) represents a pathwaythat is shared both in apoptosis and necrosis (Lemas-ters et al. 1999).

There is general agreement that apoptosis, in con-trast to necrosis, is an active, energy-requiring process.Richter et al. (1996) were the first to propose that thecellular ATP level is an important determining factorfor cell death, either by apoptosis or necrosis. Theyhypothesized that a cell stays alive as long as a certainATP level is maintained. When ATP falls below thislevel apoptosis ensues, provided that enough ATP isstill available for energy-requiring apoptotic processessuch as enzymatic hydrolysis of macromolecules, nu-clear condensation, and bleb formation. Only lowATP concentrations can switch an apoptotic death to-wards a necrotic fate (Richter et al. 1996). This hy-pothesis has been confirmed by later studies of Leist etal. (1997). These authors demonstrated that demise of

Correspondence to: Dmitri V. Krysko, Dept. of Human Anat-omy, Embryology, Histology and Medical Physics, Faculty of Medi-cine, Ghent University, Godshuizenlaan 4, B-9000 Ghent, Belgium.E-mail: [email protected]

Received for publication January 24, 2001; accepted May 16,2001 (1A5473).

KEY WORDSmitochondrial membrane

potential

apoptosis

granulosa cells

CM-H2TMRos

cytochrome c

70

human T-cells caused by two classic apoptotic triggers(staurosporine and CD95 stimulation) changed fromapoptosis to necrosis when cells were depleted of ATP.There seems to be an apparent contradiction betweenthe obligatory induction of the MPT associated with acessation of mitochondrial ATP synthesis, on the onehand, and a need for the maintenance of intracellularATP levels during the development of apoptosis, onthe other. This controversy could be resolved if theapoptosis-inducing changes were restricted to a subsetof mitochondria or if the energy level can be main-tained by anaerobic glycolysis. In a previous study wehave characterized by ultrastructural localization ofcytochrome c the persistence of a subset of respiringmitochondria until late in the apoptotic process in amodel of granulosa explants cultured under serum-free conditions, which elicits apoptosis by gonadotro-pin withdrawal (D’Herde et al. 2000). The aim of thepresent study was to investigate whether apoptosis isaccompanied by an early drop of

��m and whetherthe detected heterogeneity of the mitochondrial popu-lation with respect to cytochrome c loss is paralleledby a similar heterogeneity at the level of the mitochon-drial membrane potential. To measure

��m we useda chloromethyl rosamine-derived probe, CM-H2TMRos,which becomes fluorescent only when it is oxidized inthe cell. This dye has an alkylating chloromethyl moi-ety attached. Owing to their membrane potential,functional mitochondria take up the dye. Once theprobe accumulates in the mitochondria, the chloro-methyl group can react with accessible thiol groups ofpeptides and proteins to form an aldehyde-fixableconjugate (Macho et al. 1996; Poot et al. 1996).

Materials and MethodsIsolation and Culture of Granulosa Cell SheetsGranulosa cell (GC) sheets were prepared from ovarian folli-cles of adult regularly laying Japanese quail (Coturnixcoturnix japonica). The animals were reared under continu-ous artificial illumination, with food (fresh lettuce and com-plete breeding food; Biofor AVEVE, Belgium) and water adlibitum. Animal care procedures were conducted in accor-dance with the guidelines set by the European CommunityCouncil Directives 86/6091 EEC. The monolayered granu-losal epithelium of the largest preovulatory follicle (F1) wasisolated from the surrounding thecal covering as previouslydescribed (Gilbert et al. 1977). This method provides largesheets of vital GCs sandwiched between their basementmembrane and vitelline membrane. In the avian granulosa,three regions were defined by us (Farioli–Vecchioli et al.2000). The transition region (animal-vegetative pole) andthe animal pole region are both in close contact with the ger-minal disc that contains several cell proliferatory factors,which could influence survival of GCs (Tischkau and Bahr1996). Therefore these regions were selectively discardedfrom the rest of the GC sheet. The remaining vegetative pole

region of

�5.8 cm2 was divided into smaller squares of

�4mm2, followed by culture in 35-mm petri dishes under se-rum-free conditions for up to 72 hr in humidified room airat 38C. The culture medium was M199 (Sigma; Bornem,Belgium) phenol red-free supplemented with 0.1% w/v bo-vine serum albumin fraction V (Sigma), 6.0 g/liter HEPES(Acros; Geel, Belgium), 1% v/v penicillin–streptomycin(Gibco BRL, Paisley, UK) at pH 7.4. To inhibit the apop-totic process, the culture medium of controls was supple-mented with luteinizing hormone (LH, 100 ng/ml; Sigma)and insulin-like growth factor-I (IGF-I, 10 ng/ml; Sigma)(Onagbesan and Peddie 1995).

Mitochondrial Staining by CM-H2TMRosCM-H2TMRos (Molecular Probes; Leiden, The Nether-lands) was stored desiccated at

�20C (following the instruc-tions from the manufacturer) and dissolved in dimethylsul-foxide (DMSO; UCB, Leuven, Belgium) to give a 1 mMstock solution before use. Staining media were prepared im-mediately before use by adding the dye stock solution to cul-ture medium to obtain the desired final dye concentration of200 nM. Living GC sheets were incubated with the dye inphenol red-free medium in serum-free conditions for 15, 20,30, and 60 min at 37C shaking in the dark. Specific mito-chondrial staining was obtained after a minimum of 30 minof incubation and was totally absent at shorter incubationtimes; at 60 min of incubation the signal-to-noise ratio fur-ther improved. Therefore, all images shown, as well as quan-titative measurements of CM-H2TMRos, are derived fromexperiments with a 60-min incubation period. After the in-cubation the GC sheets were gently rinsed in PBS at 37C be-fore fixation in freshly prepared 4% paraformaldehyde inPBS at 37C for 15 min (following the manufacturer’s in-structions). Thereafter, GC sheets were rinsed in PBS andpermeabilized by incubation in ice-cold acetone at

�20C for10 min (following the manufacturer’s instructions). This ac-etone permeabilization step appeared to improve signal re-tention (data not shown). Finally, GC sheets were rinsed inPBS, mounted in a drop of bidest water, and allowed to air-dry. Samples were examined on a Leica DM IRB/E invertedmicroscope equipped with epifluorescence optics, suitable fil-ters for FITC and TRITC detection, and an MPS-60 camera.

Double Staining (CM-H2TMRos and DAPI)and Quantification of Apoptotic NucleiGC sheets were stained with CM-H2TMRos according tothe standard procedure (see above), fixed with paraformal-dehyde, permeabilized with acetone, and rehydrated withPBS. Then GC sheets were stained with 2

�,6

�-diamidino-2-fenylindole (DAPI; Sigma) at a concentration of 1 mg/ml inPBS at room temperature for 10 min. Then the GC sheetswere mounted on slides and samples were observed under anepifluorescence microscope Leica (see above). Apoptoticcells were identified as cells showing condensed chromatinmasses and/or fragmented nuclei. Small groups of apoptoticbodies were counted as remnants of one apoptotic cell. Apop-tosis was expressed as the number of apoptotic nuclei pernumber of total nuclei counted in the same microscopic fieldand expressed as the percent apoptotic nuclei. This apop-totic index (AI) was averaged for 10 fields, giving a total

71 Results: Part I

number of about 1500 cells counted for each independentexperiment.

Cytochrome c StainingLiving GC sheets were stained at 37C in culture medium,pH 7.4, for 40–90 min with 2 mg/ml 3-3

� diaminobenzidine-tetrahydrochloride dihydrate (DAB). For light microscopicevaluation, whole mounts of the sheets were counterstainedwith methyl green. The DAB technique used in the presentstudy to localize cytochrome c in individual mitochondriawas described previously by us (D’Herde et al. 2000). Be-cause it depends on cytochrome oxidase activity, it also givesevidence for an active respiration.

Pretreatment with Mitochondrial PoisonsValinomycin (Sigma) was dissolved in 95% ethanol to ob-tain a 10 mM stock solution. The stock solution was storedin small aliquots at

�20C and diluted in culture medium im-mediately before adding to the cells to obtain a final concen-tration of 1 mM.

Carbonyl cyanide p-trifluoromethoxyphenylhydrazone(FCCP; Sigma) was dissolved in ethanol. The stock solutionwas stored in small aliquots at

�20C and diluted in culturemedium immediately before adding to the cells to obtain afinal concentration of 50

�M.To investigate whether mitochondrial staining with CM-

H2TMRos is dependent on the

��m, uncultured GC sheetsand GC explants cultured for 72 hr supplemented with IGF-Iand LH were analyzed in which the

��m was dissipatedwith valinomycin (5 min at 37C) and with FCCP (15 min at37C).

Quantitative Measurements of CM-H2TMRos FluorescenceQuantitative measurements of CM-H2TMRos fluorescencewere performed by digital imaging epifluorescence micros-copy. GCs were viewed with an inverted Nikon Eclipse TE300 epifluorescence microscope using a

�40 oil-immersionlens. CM-H2TMRos fluorescence images were obtained byexcitation at 546 nm, reflection off a dichroic mirror with acut-off wavelength at 564 nm, and longpass emission filter-ing at 590 nm. Images were captured with an intensifiedCCD camera (Extended Isis camera; Photonic Science, EastSussex, UK) and were stored in a PC equipped with an im-age acquisition and processing board (Data Translation,type DT3155; Marlboro, MA).

To study the effect of mitochondrial poisons on CM-H2TMRos fluorescence, images were acquired in the two ex-perimental groups: uncultured GC sheets and GC explantscultured for 72 hr supplemented with IGF-I and LH.

To investigate

��m changes during apoptosis inductionby gonadotropin withdrawal, fluorescence intensity gener-ated by CM-H2TMRos was quantified in 72-hr cultures ofGCs compared to control cultures supplemented with IGF-Iand LH. All images were corrected for background fluores-cence (culture medium without sheets).

Statistical AnalysisData are expressed as mean

� SEM with n denoting thenumber of experiments on different animals. Statistical sig-

nificance was tested using a Student’s t-test for unpaired ob-servations. p values less than 0.05 were considered as statis-tically significant.

ResultsSensitivity of CM-H2TMRos to Mitochondrial Poisons

To confirm that the mitochondrial CM-H2TMRos dyeaccumulation was dependent on mitochondrial trans-membrane potential, uncultured (i.e., freshly isolated)GC sheets and 72-hr IGF-I- and LH-supplementedcultures were treated with the mitochondrial uncou-pling agents FCCP and valinomycin. Pretreatment ofGC sheets with FCCP or valinomycin before probeloading resulted in absence of punctiform mitochon-drial staining and showed only diffuse cytoplasmicstaining on visual inspection (Figures 1A, 1E, and 1F).Quantification of the CM-H2TMRos signal underthese conditions showed significantly reduced fluores-cence intensity with FCCP or valinomycin comparedto control (Figures 2A and 2B). Both mitochondrialpoisons had no effect on the CM-H2TMRos fluores-cence when added after loading of the mitochondrialprobe.

Subcellular Distribution of Mitochondria During Apoptosis

In uncultured GC sheets, staining of polarized mito-chondria was intense and the pattern was uniformthroughout the sheet. In images of GC sheets stainedwith CM-H2TMRos in combination with DAPI, mito-chondria appeared as threadlike and granular struc-tures homogeneously distributed throughout the GCs.The mitochondrial staining revealed by CM-H2TMRoswas similar to the appearance of respiring mitochon-dria identified by cytochrome c staining (Figures 1Aand 1B).

Double staining (CM-H2TMRos–DAPI) of 72-hrIGF-I- and LH-supplemented cultures, i.e., the controlcondition in which the apoptotic process is inhibited(AI of circa 0.64

� 0.27% SEM; n

�4), showed a ho-mogeneous pattern in contrast to the serum-free cul-tures. All normal cells showed a similar number of po-larized mitochondria localized in compact masses atone side of the nucleus (Figure 1C).

Seventy-two-hour cultured GC explants cultured inthe absence of IGF-I and LH (AI of circa 29

� 8.03%SEM; n

�4) showed a non-uniform pattern of CM-H2TMRos staining throughout the GC explant celllayer (Figure 1D). Some cells with normal nucleiwere surrounded by very few stained mitochondria,whereas other adjacent cells displayed a sizable num-ber of polarized mitochondria localized around nucleiwith normal morphology. Apoptotic cells were ob-served with condensed and fragmented nuclei but stillcontaining polarized mitochondria (Figure 1G). In the

Mitochondrial heterogeneity 72

Figure 1 (A,C–G) Epifluorescence microscopy of paraformaldehyde-fixed GC sheetsdouble stained with CM-H2TMRos and DAPI. Polarized mitochondria are stained as sep-arate red dots. (A) Large numbers of polarized mitochondria in all cells of freshly iso-lated GC sheet. (B) Brightfield light microscopy; cytochrome c localization (brown) infreshly isolated living GC sheet, methyl green counterstaining of nuclei. Large numbersof reactive mitochondria; image similar to A. (C) Polarized mitochondria are concen-trated in compact masses at one side of the nucleus in GC sheet cultured for 72 hr andsupplemented with IGF-I and LH. (D) Few polarized mitochondria when GC sheet is cul-tured for 72 hr in serum-free conditions; in comparison to A and C, a smaller area of thecytoplasm is taken in by punctiform CM-H2TMRos staining, while the largest part dis-plays only background fluorescence. (E) Mitochondria depolarized by treatment withFCCP before loading with CM-H2TMRos do not take up the probe; there are no gran-ules. GCs cultured for 72 hr with IGF-I and LH. (F) Polarized mitochondria stained withCM-H2TMRos before treatment with FCCP in freshly isolated GC sheet. Note necroticcells with absence of polarized mitochondria. (G) Mitochondrial heterogeneity in GCsheet cultured for 72 hr in serum-free conditions. Remark persistence of polarized mito-chondria surrounding apoptotic chromatin bodies (arrowheads). Other apoptotic nucleishow no stained granules (arrow). Bars: A,B,E,F � 10 �m; C,D,G � 5 �m.

73 Results: Part I

majority of the apoptotic cells, however, CM-H2TMRosfluorescence was no longer detected (Figure 1G). Innecrotic cells, often present at the edge of the explantsand recognized by their small pyknotic nuclei, polar-ized mitochondria were never observed (Figure 1F).

Quantification of the CM-H2TMRos signal in the72-hr cultured GC explants revealed that the fluores-cence intensity was significantly lower (Figure 3) whenIGF-I and LH were absent in the culture medium com-pared to the control condition (with both survival fac-tors).

DiscussionCationic lipophilic fluorochromes such as 3,3

�-dihex-iloxocarbocyanine iodide (DiOC6), rhodamine 123(R123), and 5,5

�,6,6

�-tetrachloro-1,1

�,3,3

�-tetraethyl-benzimidazolocarbocyanine iodide (JC-1) have beenwidely used to assess the functionality of mitochon-dria in diverse biological scenarios, including differen-tiation (Mancini et al. 1997), aging (Hagen et al.1997), and apoptosis (Zamzami et al. 1995a,b).

Compared to DiOC6 and R123, JC-1 is a reliableand sensitive probe, as shown in several studies. Itsloading is impaired by several drugs able to collapse

��m, which is not the case for DiOC6 and R123 ow-ing to a high sensitivity to changes of plasma mem-brane potential for DiOC6 and energy-independentbinding sites for R123 (Salvioli et al. 1997). More-over, this dichromatic probe permitted detection of

the coexistence of mitochondria in different energeticconditions within an individual cell. A disadvantage ofJC-1 is that it cannot be combined with other fluores-cent probes in the FITC/TRITC emission spectra. Inaddition, the use of DiOC6, R123, and JC-1 is incom-patible with chemical fixation.

Macho et al. (1996) described the main characteris-tics of a class of fixable fluorescent probes: the Mi-toTracker probes, focusing on MitoTracker Red (boththe oxidized and the reduced form CMXRos andH2-TMRos respectively). In the present study weused MitoTracker Orange (in its reduced form:CM-H2TMRos) to study mitochondrial transmem-brane potential changes during apoptosis in relationto nuclear morphology as revealed by DAPI staining.CM-H2TMRos enters the mitochondria due to thenegative mitochondrial membrane potential and onlyafter that, in a second step when it is oxidized, bindsto SH-groups, which makes this probe aldehyde-fix-able. Uncoupling treatments were ineffective afterprobe loading (Figures 1F, 2A, and 2B), indicatingthat binding of the probe to SH-groups occurs only af-ter entering of CM-H2TMRos in mitochondria due totheir membrane potential. The first step of our workwas to demonstrate the usefulness of CM-H2TMRosin measuring

��m. We analyzed fluorescence inten-sity after impairing mitochondrial function with com-monly used, mechanism-specific mitochondrial poi-sons (FCCP, a proton translocator, and valinomycin,a potassium ionophore) in uncultured GC sheets andin 72-hr cultures supplemented with IGF-I and LH. Pre-treatment with FCCP and valinomycin before probeloading gave no punctiform mitochondrial staining(Figure 1E) and significantly reduced the intensity offluorescence (Figures 2A and 2B). A possible explana-

Figure 2 Effect of mitochondrial poisons on CM-H2TMRos fluores-cence. GCs were either pretreated with FCCP (FCCP+CM-H2TMRos),pretreated with valinomycin (Valin+CM-H2TMRos), or treated withFCCP after probe loading (CM-H2TMRos+FCCP), treated with vali-nomycin after probe loading (CM-H2TMRos+Valin). Intensity ofCM-H2TMRos fluorescence was expressed as the whole field fluo-rescence. *p0.001 compared to control; #, changes were not sig-nificant compared to control. (A) Uncultured GC sheets. The datarepresent the mean � SEM of four independent experiments. Ineach experiment, fluorescence intensity of CM-H2TMRos was mea-sured in 10 fields. (B) GC explants cultured for up to 72 hr supple-mented with IGF-I and LH. Data represent mean � SEM of five in-dependent experiments. In each experiment, fluorescence intensityof CM-H2TMRos was measured in five fields.

Figure 3 Effect of IGF-I and LH supplementation or absence onCM-H2TMRos fluorescence in serum-free GC explants cultured forup to 72 hr. Data represent the mean � SEM of five independentexperiments. In each experiment, fluorescence intensity of CM-H2TMRos was measured in five fields. *p0.0001 compared to con-trol. Intensity of CM-H2TMRos fluorescence was expressed as thewhole field fluorescence divided by the number of nuclei in thefield to correct for decreased cell mass.


tion for the presence of the diffuse cytoplasmic stain-ing could be that cells still oxidized the probe in theircytoplasm but, as mitochondrial respiration is alreadyuncoupled, the dye is not sequestered in the mitochon-dria but is distributed in the cytoplasm.

The second argument to support the mitochondrialspecificity of the probe can be found in the similarityof the ��m staining and cytochrome c localization inrespiring mitochondria (Figures 1A and 1B). Our dataare in agreement with results of Poot et al. (1996),who reported that uncoupling mitochondrial respira-tion with carbonyl cyanide m-chlorophenylhydrazone(CCCP) before staining with MitoTracker Red (H2-CMXRos) significantly decreased fluorescence gener-ated by the dye. Moreover, Poot et al. (1996) showedcomplete co-localization in the mitochondria of amonoclonal antibody which is specific for subunit I ofhuman oxidase and H2-CMXRos staining. By con-trast, the results of Scorrano et al. (1999) indicated thatstaining with oxidized MitoTracker Orange (CMTM-Ros) is characterized by two different mechanisms.One is energy-independent, in that it is also observedin mitochondria de-energized by FCCP. It can besignificantly prevented by N-ethylmaleimide or phe-nylarsine oxide (inhibitors of SH-groups). However,the data obtained with compounds such as phenylar-sine oxide should be interpreted with much care be-cause such compounds can induce mitochondrialmembrane permeabilization (Costantini et al. 2000)and thereby prevent staining. The second componentof CMTMRos staining is energy-dependent. Interpre-tation of this energy-dependent staining componentshould also be done with care because Scorrano et al.(1999) have mentioned that mitochondria sometimesspontaneously release the probe within minutes, sug-gesting that they became depolarized. This effectcould be explained by the experiments of Minami-kawa et al. (1999), which showed that CMXRos has astrong photosensitizing action in living cells. Photoir-radiation of intact living cells loaded with CMXRosinduces depolarization of the inner mitochondrialmembrane and swelling of mitochondria, resultingin apoptosis. Based on our own findings and theabove-mentioned literature data, we propose to useonly the reduced chloromethylrosamine-derived probeCM-H2TMRos (MitoTracker Orange) and to evaluate��m after fixation.

The second part of our work was to study ��mchanges in the GC explants cultured up to 72 hr underserum-free conditions, which elicits apoptosis. Boththe images and the fluorescence measurements showthat, in our model system, dissipation of transmem-brane potential in individual mitochondria is an earlyevent in GCs not displaying nuclear manifestations ofapoptosis. Moreover, we have shown the presence of asubset of polarized mitochondria that exhibited still

normal ��m until late in the apoptotic process (Fig-ure 1G). These data confirm our previous conclusionson mitochondrial heterogeneity during apoptosis witha subset of respiring mitochondria retaining cyto-chrome c function until the stage of chromatin con-densation and nuclear fragmentation (D’Herde et al.2000). The presence of polarized cytochrome c-con-taining and respiring mitochondria until late in the ap-optotic process preserves the possibility for ATP pro-duction even in the stage of chromatin condensationand fragmentation. Bradham et al. (1998) observed byconfocal microscopy in an apoptosis model consistingof TNF- treatment in rat hepatocytes expressing anIkB super-repessor that a gradual onset of mitochon-drial depolarization in a subpopulation of mitochon-dria occurred using the tetramethylrhodamine methylester (TMRM) probe, in conjunction with calcein, aprobe marking MPT in individual mitochondria. Theauthors concluded that, for several hours during theapoptotic response, hepatocytes contain both polar-ized and depolarized mitochondria, the latter beingthe presumptive source of released cytochrome c.These results are perfectly in line with our observa-tions. However, because these authors used a non-fix-able mitochondrion-selective probe, TMRM, it wasimpossible to check whether vital cells with apoptoticnuclear morphology still contain polarized mitochon-dria as was done in the present study by using the fix-able CM-H2TMRos probe and DAPI staining.

Heiskanen et al. (1999) monitored mitochondrialdepolarization in individual mitochondria of single cy-tochrome c–green fluorescent protein (GFP)-trans-fected PC6 cells undergoing apoptosis after staurospo-rine treatment. For this purpose, TMRM-loaded cellswere observed at various time points after treatmentby scanning laser confocal microscopy. Transfectionwith cytochrome c–GFP enabled the authors to studythe relation between mitochondrial depolarizationand cytochrome c release. In agreement with our ob-servations, they demonstrated that mitochondrial de-polarization did not occur homogenously over the en-tire mitochondrial population and that cytochrome crelease is initially also restricted to a subpopulationyielding a pattern of mixed punctate and diffuse cyto-chrome c–GFP localization. Again, persistence of po-larized mitochondria in cells with clearly apoptoticmorphology was not documented in this study.

Goldstein et al. (2000) studied the kinetics of cyto-chrome c release during apoptosis, using cytochrome ctagged with GFP. With this approach they reportedfor HeLa cells induced to apoptosis by a variety ofagents that, once initiated, the release of cytochrome ccontinues until all of the protein is released from allthe mitochondria in an individual cell, within about 5min. To reconcile these findings with the need for ATPproduction, Goldstein et al. (2000) proposed that

75 Results: Part I

some cytochrome c either remains in the mitochondriaor may re-enter the intermembrane space to partici-pate in electron transport. Indeed, one has to take intoaccount that GFP-tagged cytochrome c represents only1% of the endogenous cytochrome c content of themitochondria. In any case, heterogeneity in the mito-chondrial population with respect to cytochrome closs was not mentioned, although their time-lapse im-ages (Goldstein et al. 2000; Figure 3A, upper panel)show the maintenance of some cytochrome c-contain-ing mitochondria 10 min after release from the firstmitochondria.

Recently, Salvioli et al. (2000) reported for stauro-sporine-treated HL60 cells that the collapse in ��m isa heterogeneous phenomenon both at the single or-ganelle level and at the cellular level, and therefore isnot unequivocally related to the execution of the deathprocess but rather is an ancillary event. Several reportsdescribed an early ��m decrease before the exposureof phosphatidylserine on the cell surface or DNA frag-mentation (Castedo et al. 1995; Zamzami et al.1995a,b). In contrast, a late decrease in ��m wasfound in several other models of apoptosis (Deck-werth and Johnson 1993; Ankarcrona et al. 1995;Cossarizza et al. 1996; Bossy–Wetzel et al. 1998). Thisdiscrepancy may be due to the use of different celltypes or apoptosis-inducing stimuli.

In several experimental models, cell stress is accom-panied by an early hyperpolarization of mitochondria(Banki et al. 1999; Diaz et al. 1999; Scarlett et al.2000; Khaled et al. 2001), while subsequent mito-chondrial depolarization precedes mitochondrial dis-ruption and is dependent on the stimulus and cell-typecaspase activation.

Several authors reported mitochondrial clusteringin various apoptosis paradigms as an early event, inde-pendent of ��m changes or cytochrome c release(De Vos et al. 1998; Li et al. 1998; Diaz et al. 1999;Esposti et al. 1999). In serum-free cultured granulosaexplants in which apoptosis is not synchronized(D’Herde and Leybaert 1997), we did not detect thismitochondrial clustering. Instead, a heterogeneousdistribution of polarized mitochondria throughout theGC sheet could be observed (Figure 1G), with variousnumbers of polarized mitochondria in non-apoptoticcells and sometimes the maintenance of a few polarizedmitochondria in frankly apoptotic cells. However, incontrol cultures supplemented with gonadotropinsand IGF-I, an unexpected mitochondrial clustering inpacked masses at one side of the nucleus was revealed(Figure 1C). It was previously shown by Aharoni et al.(1993) that mitochondrial clustering can be induced inGCs under the combined action of gonadotropins andIGF-I in parallel with stimulation of steroidogenesis.

In this study we have shown that (a) CM-H2TMRosstaining evaluated after aldehyde fixation is a reliable

and specific probe to detect ��m changes in the gran-ulosa explant system, (b) dissipation of transmem-brane potential is an early event during apoptosis, pre-ceding nuclear changes, and is again confined to asubpopulation of mitochondria within an individualcell, and (c) the majority of frankly apoptotic cells aredevoid of polarized mitochondria, but in some apop-totic cells a few polarized mitochondria can be de-tected. Necrotic cells do not display polarized mito-chondria.

The present findings support the hypothesis thatATP needed for completion of the apoptotic cascadecan be generated during apoptosis in a subset of re-spiring mitochondria and is not necessarily derivedfrom anaerobic glycolysis.

AcknowledgmentsSupported by the BOF (Bijzonder Onderzoeksfonds) to

KD 01115099. DVK is a recipient of a predoctoral studentgrant from Universiteitsvermogen R.U.G. P 0211 (FR). LLwas supported by the Fund for Scientific Research, Flanders,Belgium (3G023599 and G.0012.01), the Belgian Society forScientific Research in Multiple Sclerosis (WOMS) (grant no.51F06700), and Ghent University (grant no. 01115099).

We thank Prof Dr M. Espeel, Ms S. Van Hulle, and MrH. Stevens for excellent photographic assistance, and Ms B.De Prest for compiling the bibliography and for the cyto-chrome c staining.

Literature CitedAharoni D, Dantes A, Amsterdam A (1993) Cross-talk between

adenylate cyclase activation and tyrosine phosphorylation leadsto modulation of the actin cytoskeleton and to acute pro-gesterone secretion in ovarian granulosa cells. Endocrinology133:1426–1436

Ankarcrona M, Dypbukt JM, Bonfoco E, Zhivotovsky B, OrreniusS, Lipton SA, Nicotera P (1995) Glutamate-induced neuronaldeath: a succession of necrosis or apoptosis depending on mito-chondrial function. Neuron 15:961–973

Banki K, Hutter E, Gonchoroff NJ, Perl A (1999) Elevation of mito-chondrial transmembrane potential and reactive oxygen interme-diate levels are early events and occur independently from activa-tion of caspases in Fas signaling. J Immunol 162:1466–1479

Bossy–Wetzel E, Newmeyer DD, Green DR (1998) Mitochondrialcytochrome c release in apoptosis occurs upstream of DEVD-spe-cific caspase activation and independently of mitochondrialtransmembrane depolarization. EMBO J 17:37–49

Bradham CA, Qian T, Streetz K, Trautwein C, Brenner DA, Lemas-ters JJ (1998) The mitochondrial permeability transition is re-quired for tumor necrosis factor alpha-mediated apoptosis andcytochrome c release. Mol Cell Biol 18:6353–6364

Castedo M, Macho A, Zamzami N, Hirsch T, Marchetti P, Uriel J,Kroemer G (1995) Mitochondrial perturbations define lympho-cytes undergoing apoptotic depletion in vivo. Eur J Immunol25:3277–3284

Cossarizza A, Franceschi C, Monti D, Salvioli S, Bellesia E, Riva-bene R, Biondo L, Rainaldi G, Tinari A, Malorni W (1996) Pro-tective effect of N-acetylcysteine in tumor necrosis factor-alpha-induced apoptosis in U937 cells: the role of mitochondria. ExpCell Res 220:232–240

Costantini P, Belzacq AS, Vieira HL, Larochette N, de-Pablo MA,Zamzami N, Susin SA, Brenner C, Kroemer G (2000) Oxidation


of a critical thiol residue of the adenine nucleotide translocatorenforces Bcl-2-independent permeability transition pore openingand apoptosis. Oncogene 13:307–314

D’Herde K, De Prest B, Mussche S, Schotte P, Beyaert R, VanCoster R, Roels F (2000) Ultrastructural localization of cyto-chrome c in apoptosis demonstrates mitochondrial heterogene-ity. Cell Death Differ 7:331–337

D’Herde K, Leybaert L (1997) Intracellular free calcium related toapoptotic cell death in quail granulosa cell sheets kept in serum-free culture. Cell Death Differ 4:59–65

De Vos K, Goossens V, Boone E, Vercammen D, VancompernolleK, Vandenabeele P, Haegeman G, Fiers W, Grooten J (1998) The55-kDa tumor necrosis factor receptor induces clustering of mi-tochondria through its membrane-proximal region. J Biol Chem273:9673–9680

Deckwerth TL, Johnson EM Jr (1993) Temporal analysis of eventsassociated with programmed cell death (apoptosis) of sympa-thetic neurons deprived of nerve growth factor. J Cell Biol123:1207–1222

Diaz G, Setzu MD, Zucca A, Isola R, Diana A, Murru R, Sogos V,Gremo F (1999) Subcellular heterogeneity of mitochondrialmembrane potential: relationship with organelle distribution andintracellular contacts in normal, hypoxic and apoptotic cells. JCell Sci 112:1077–1084

Esposti MD, Hatzinisiriou I, McLennan H, Ralph S (1999) Bcl-2and mitochondrial oxygen radicals. J Biol Chem 274:29831–29837

Farioli–Vecchioli S, Raes S, Espeel M, Roels F, D’Herde K (2000)Inverse expression of peroxisomes and connexin-43 in the granu-losa cells of the quail follicle. J Histochem Cytochem 48:167–177

Gilbert AB, Evans AJ, Perry MM, Davidson MH (1977) A methodfor separating the granulosa cells, the basal lamina and the thecaof the preovulatory ovarian follicle of the domestic fowl (Gallusdomesticus). J Reprod Fertil 50:179–181

Goldstein JC, Watehouse NJ, Juin P, Evan GI, Green DR (2000)The coordinate release of cytochrome c during apoptosis is rapid,complete and kinetically invariant. Nature Cell Biol 2:156–162

Hagen TM, Yowe DL, Bartholomew JC, Wehr CM, Do KL, ParkJY, Ames BN (1997) Mitochondrial decay in hepatocytes fromold rats: membrane potential declines, heterogeneity and oxi-dants increase. Proc Natl Acad Sci USA 94:3064–3069

Heiskanen KM, Bhat MB, Wang H-W, Ma J, Nieminen A-L (1999)Mitochondrial depolarization accompanies cytochrome c releaseduring apoptosis in PC6 cells. J Biol Chem 274:5654–5658

Khaled AR, Reynolds DA, Young HA, Thompson CB, Muegge K,Durum SK (2001) Interleukin-3 withdrawal induces an early in-crease in mitochondrial membrane potential unrelated to the Bcl-2family. J Biol Chem 276:6453–6462

Leist M, Single B, Castoldi AF, Kuhnle S, Nicotera P (1997) Intra-cellular adenosine triphosphate (ATP) concentrations: a switch inthe decision between apoptosis and necrosis. J Exp Med 185:1481–1486

Lemasters JJ, Qian T, Bradham CA, Brenner DA, Cascio WE, TrostLC, Nishimura Y, Nieminen A-L, Herman B (1999) Mitochon-drial dysfunction in the pathogenesis of necrotic and apoptoticcell death. J Bioenerg Biomembr 31:305–319

Li H, Zhu H, Hu CJ, Yuan J (1998) Cleavage of BID by caspase 8mediates the mitochondrial damage in the Fas pathway of apop-tosis. Cell 94:491–501

Macho A, Decaudin D, Castedo M, Hirsch T, Susin SA, ZamzamiN, Kroemer G (1996) Chloromethyl-x-rosamine is an aldehyde-fixable potential-sensitive fluorochrome for the detection of earlyapoptosis. Cytometry 25:333–340

Mancini M, Anderson BO, Caldwell E, Sedghinasab M, Paty PB,Hockenbery DM (1997) Mitochondrial proliferation and para-doxical membrane depolarization during terminal differentiationand apoptosis in human colon carcinoma cell line. J Cell Biol138:449–469

Minamikawa T, Sriratana A, Williams DA, Bowser DN, Hill JS,Nagley P (1999) Chloromethyl-x-rosamine (Mitotracker Red)photosensitises mitochondria and induces apoptosis in intact hu-man cells. J Cell Sci 112:2419–2430

Onagbesan OM, Peddie MJ (1995) Effects of insulin-like growthfactor I and interactions with transforming growth factor andLH on proliferation of chicken granulosa cells and production ofprogesterone in culture. J Reprod Fertil 104:259–265

Poot M, Zhang Y-Z, Kraemer JA, Wells KS, Jones LJ, Hanzel DK,Lugade AG, Singer VL, Haugland RP (1996) Analysis of mito-chondrial morphology and function with novel fixable fluores-cent stains. J Histochem Cytochem 44:1363–1372

Richter C, Schweizer M, Cossarizza A, Franceschi C (1996) Controlof apoptosis by the cellular ATP level. FEBS Lett 378:107–110

Salvioli S, Ardizzoni A, Franceschi C, Cossarizza (1997) JC-1, butnot DiOC6 (3) or rhodamine 123, is a reliable fluorescent probeto assess �� changes in intact cells: implications for studies onmitochondrial functionality during apoptosis. FEBS Lett 411:77–82

Salvioli S, Dobrucki J, Moretti L, Troiano L, Fernandez MG, PintiM, Pedrazzi J, Franceschi C, Cossarizza A (2000) Mitochondrialheterogeneity during staurosporine-induced apoptosis in HL60cells: analysis at the single cell and single organelle level. Cytom-etry 40:189–197

Scarlett JL, Sheard PW, Hughes G, Ledgerwood EC, Ku HH, Mur-phy MP (2000) Changes in mitochondrial membrane potentialduring staurosporine-induced apoptosis in Jurkat cells. FEBSLett 475:267–272

Scorrano L, Petronilli V, Colonna R, Di Lisa F, Bernardi P (1999)Chloromethyltetramethylrosamine (Mitotracker OrangeTM) in-duces the mitochondrial permeability transition and inhibits re-spiratory complex. Implications for the mechanism of cyto-chrome c release. J Biol Chem 274:24657–24663

Tischkau SA, Bahr JM (1996) Avian germinal disc region secretesfactors that stimulate proliferation and inhibit progesterone pro-duction by granulosa cells. Biol Reprod 54:865–870

Tsujimoto Y (1997) Apoptosis and necrosis: intracellular ATP levelas a determination for cell death modes. Cell Death Differ4:429–434

Wyllie AH, Kerr JFR, Currie AR (1980) Cell death: the significanceof apoptosis. Int Rev Cytol 68:251–306

Zamzami N, Marchetti P, Castedo M, Decaudin D, Macho A,Hirsch T, Susin SA, Petit PX, Mignotte B, Kroemer G (1995a)Sequential reduction of mitochondrial transmembrane potentialand generation of reactive oxygen species in early programmedcell death. J Exp Med 182:367–377

Zamzami N, Marchetti P, Castedo M, Zanin C, Vayssiere JL, PetitPX, Kroemer G (1995b) Reduction in mitochondrial potentialconstitutes an early irreversible step of programmed lymphocytedeath in vivo. J Exp Med 181:1661–1672

77 Results: Part I

Gap junctional communications in cell death and cell survival

Then of thy beauty do I question make,That thou among the wastes of time must go,Since sweets and beauties do themselves forsakeAnd die as fast as they see others grow;And nothing 'gainst Time's scythe can make defenceSave breed, to brave him when he takes thee hence.

Sonnet 12, When I do count the clock that tells the timeW Shakespear

The kiss, A Rodin

2. Intercellular level

Gap junctional communication and connexin43 expression in relation to apoptotic cell death and survival of granulosa cells.

DV Krysko, S Mussche, L Leybaert and K D’Herde The Journal of Histochemistry and Cytochemistry 2004; 52(9): 1199-207.

Gap Junctional Communication and Connexin43 Expression in Relation to Apoptotic Cell Death and Survival of Granulosa Cells

Dmitri V. Krysko,1 Sylvie Mussche,1 Luc Leybaert, and Katharina D’Herde

Department of Human Anatomy, Embryology, Histology and Medical Physics (DVK,SM,KD), and Department of Physiology and Pathophysiology (LL), Ghent University, Ghent, Belgium

SUMMARY Ovarian follicular atresia in all vertebrates is mediated via apoptosis that isinitiated in the granulosa cell layer. Here we investigated the relation between connexinexpression, cell coupling, and apoptosis in avian granulosa cells. Results from qualitativeand quantitative immunocytochemical analysis and Western blotting of connexin43 (Cx43)and electron microscopic observations of gap junctions were compared with functionaldata on gap junctional coupling obtained by fluorescence recovery after photobleaching infour experimental groups: a control group of freshly isolated granulosa cells, 24-hr serum-free cultures as the apoptosis-inducing condition, and two other groups in which apoptosiswas inhibited by either hormone substitution or exposure to elevated extracellular calcium.Our work shows that apoptosis induction in granulosa cells is accompanied by an increasedlevel of cell coupling and that decreasing cell coupling with the gap junction blocker

-glycyr-rhetinic acid dose-dependently inhibits apoptosis. The level of Cx43 expression was in-versely related to the apoptotic index, suggesting that Cx43 expression plays a role in gran-ulosa cell survival. Our study supports the hypothesis that gap junctional coupling plays arole in propagating a cell death message and suggests a role for Cx43 expression per se ingranulosa cell survival. (J Histochem Cytochem 52:1199–1207, 2004)

The majority of the female germ line dies through-out postnatal life, because most follicles never ovulatebut rather degenerate by a process called follicularatresia. It has been established that apoptotic celldeath of granulosa cells is instrumental in the processof follicular atresia in the ovary of all vertebrates, in-cluding humans (Tilly 1997; Jiang et al. 2003). Al-though the vast majority of research to elucidate themolecular ordering of cell signaling during the processof granulosa cell apoptosis has been conducted inmammalian model systems, there is ample evidence todemonstrate that many of the proteins, enzymes, andcell signaling pathways are common to ovarian folli-cles from avian species (Johnson 2000).

Most studies on apoptosis regulation, including

ours, are focusing predominantly on intracellular sig-naling pathways (D’Herde et al. 2000; Krysko et al.2001), while tissue function obviously involves cell-to-cell communication. Granulosa cells can communicateeither through local production of intraovarian factorssuch as cytokines (Adashi 1992) and growth factors(Adashi et al. 1991), which act as paracrine and/or au-tocrine modulators, or through gap junctions. Gapjunctions are channels formed by two hemichannels,each formed by six connexin subunits contributed byeach cell. All known connexins are transmembraneproteins with four transmembrane domains, two ex-tracellular loops, one intracellular loop, and cyto-plasmic amino and carboxyl termini (Bruzzone et al.1996; Munari-Silem and Rousset 1996; Trosko andRuch 1998). Cx43 is the major gap junctional proteinexpressed in granulosa cells that makes a significantcontribution to intercellular coupling (Dekel 1987;Grazul-Bilska et al. 1997; Farioli-Vecchioli et al. 2000;Sommersberg et al. 2000; Rosenfeld et al. 2001; Git-tens et al. 2003). It has been shown that gap junctionsplay an important role in granulosa cell development,

Correspondence to: Dmitri V. Krysko, Dept. of Human Anat-omy, Embryology, Histology and Medical Physics, Faculty of Medi-cine, Ghent University, Godshuizenlaan 4, B-9000 Ghent, Belgium.E-mail: [email protected]

Received for publication December 16, 2003; accepted April 18,2004 [DOI: 10.1369/jhc.3A6227.2004].

1These authors contributed equally to this work.

KEY WORDSgranulosa cells

apoptosis

gap junctions

connexin43

FRAP

78

The

Jour

nal o

f His

toch

emis

try

& C

ytoc

hem

istr

y

differentiation, and luteinization (Kidder and Mhawi2002). Classically, it is believed that gap junctions pro-vide a way for cell-to-cell diffusion of hydrophilic mol-ecules with a molecular mass of less than 1 kD, such ascAMP, Ca2

�, IP3, and ATP (Kumar and Gilula 1996).More recently, however, several studies demonstratedthat Cx43 could affect signaling pathways indepen-dently of its contribution to gap junction channels(Moorby and Patel 2001; Plotkin and Bellido 2001).Despite the increased interest of research in this field,little is known regarding the involvement of gap junc-tional intercellular communication and Cx43 expressionduring apoptosis. In a previous study, we observed byelectron microscopy (EM) in 24-hr serum-free culturedgranulosa cell explants, in which apoptosis is elicitedby gonadotropin withdrawal, an increased number ofhuge gap junctions in comparison with unculturedgranulosa cells (D’Herde and Leybaert 1997). This ob-servation prompted us to investigate the relation be-tween connexin expression, cell coupling, apoptosis,and survival of the granulosa cell explants.

Materials and MethodsIsolation and Culture of Granulosa Cell SheetsGranulosa cell sheets were prepared from ovarian follicles ofregularly laying adult Japanese quails (Coturnix coturnixjaponica). The animals were reared under an artificial lightcycle of 18 hr per day, with food (fresh lettuce and completebreeding food; Biofor AVEVE, Belgium) and water ad libi-tum. Animal care procedures were conducted in accordancewith the guidelines set by the local ethical committee ECP00/05 (2000). The monolayered granulosa epithelium ofthe largest preovulatory follicle (F1) was isolated in Krebs–Ringer solution as previously described (Gilbert et al. 1977).This method provides large sheets of vital granulosa cellssandwiched between their basement and vitelline mem-branes (Mussche et al. 2000; Krysko et al. 2001). The gran-ulosa explants are cultured in suspension in phenol red-freeM199 medium (Sigma; Bornem, Belgium) supplemented with0.1% w/v bovine serum albumin (BSA) fraction V (Sigma),6.0 g/liter HEPES (Acros; Geel, Belgium), 1% v/v penicillin–streptomycin (GIBCO; Paisley, UK) at pH 7.4. To inhibit theapoptotic process, the culture medium of controls was sup-plemented either with a combination of luteinizing hormone(LH, 100 ng/ml; Sigma-Aldrich, Bornem, Belgium) and in-sulin-like growth factor-I (IGF-I, 10 ng/ml; Sigma-Aldrich;(Onagbesan and Peddie 1988; Krysko et al. 2001) or with acontinuous load of increased extracellular Ca2

�, a protocolthat promotes survival through activation of a calcium sens-ing receptor (Mussche et al. 2000; and Mussche et al. un-published data). The concentration of all vehicle solutionswas always below 0.1% and had no effects on the viabilityof granulosa cell explants (data not shown).

Quantification of Apoptosis by DAPI StainingGranulosa cell explants were rinsed in 10 mM PBS, pH 7.4,and fixed in 4% buffered formaldehyde. Thereafter, granu-

losa cell explants were rinsed again in PBS and stained with a1:1000 solution of DAPI (2’,6’-diamidino-2-phenylindole;Sigma-Aldrich) in PBS. Mounted granulosa cell explants wereexamined in a Leica DM IRB/E inverted microscope and ap-optotic cells were identified by their characteristic fragmentedand condensed chromatin masses. Small groups of apoptoticbodies were counted as remnants of one apoptotic cell. Apop-totic indexes (AIs) are expressed as the number of apoptoticnuclei per number of total nuclei counted in the same micro-scopic field. This AI was averaged for 10 fields, giving a totalnumber of

�1500 cells counted per experiment.

AntibodiesA mouse monoclonal antibody (IgG1) raised against the pep-tide sequence corresponding to the amino acid residues 252–270 of the native rat Cx43 was used for immunocytochemis-try (ICC) and Western blotting and was purchased fromProBio (Poole, UK) and BD Biosciences (Erembodegem, Bel-gium), respectively. These monoclonal antibodies were re-ported to crossreact with avian Cx43 (Beyer 1990; Musil etal. 1990). Affinity-purified secondary goat anti-mouse (IgG)antibody conjugated to horseradish peroxidase (HRP) wereobtained from Cell Signaling Technology (Westburg, Leus-den, The Netherlands) and secondary rabbit–FITC anti-mouse (IgG) antibody was purchased from DAKO (Glos-trup, Denmark).

Immunocytochemistry of Connexin43After rinsing of granulosa cell explants with PBS, they werefixed for 20 min in 3% PBS-buffered paraformaldehyde.Next, explants were incubated for 20 min in 3% BSA andthen for 10 min in 1% glycine. After 45-min incubation in10% normal rabbit serum, the explants were incubated,directly and without a rinsing step, overnight at 4C withmouse monoclonal anti-Cx43 antibody (1:400 dilution of 1mg/ml stock; or for the controls the corresponding IgG1 atthe same concentration; DAKO). Then the granulosa explantswere incubated for 45 min with rabbit–FITC anti-mouseIgG (1:40). For double staining, the granulosa explants werestained with DAPI as described above, rinsed with 0.9%NaCl, and mounted with fluorescent mounting medium. Un-less otherwise described, explants were rinsed thoroughlywith PBS between all steps.

Quantification of Cx43 ImmunofluorescenceQuantitative measurements of Cx43 immunofluorescence(Laing et al. 1997) were performed by determining the aver-age fluorescence intensity in images obtained by confocal la-ser scanning microscopy (CLSM) of granulosa cell sheets orexplants immunostained as described above. Images wereacquired with a custom-made video-rate CLSM (Sandersonand Parker 2003) built around a Nikon Eclipse TE300 (Ana-lis; Ghent, Belgium) equipped with a

�40 oil immersion ob-jective (CFI Plan Fluor; Nikon) and with 488-nm excitationfrom an argon laser.

Protein Extraction and Western Blotting Analysisof Cx43 ExpressionGranulosa cell explants were lysed in 50

�l of ice-cold mod-ified RIPA buffer (50 mM NaCl, 0.5% NP40, 0.5% sodium

79 Results: Part II

The

Jour

nal o

f His

toch

emis

try

& C

ytoc

hem

istr

y

deoxycholate, 0.1% sodium dodecyl sulfate, 250 mM

-gly-cerophosphate, 25 mM Tris, pH 8.2) containing a standardphosphatase inhibitor cocktail 2 (20

�l/ml; Sigma-Aldrich)and complete EDTA-free protease inhibitor tablet (20

�l/ml;Roche Diagnostics, Brussels, Belgium) for 10 min and soni-cated on ice three times for 10 sec using a Vibracell VC-130sonicator. The protein concentration of each sample wasmeasured according to Bradford (1976) using the BioRadprotein assay kit (BioRad Laboratories; Nazareth/Eke, Bel-gium) and a microtiter plate reader model 550 (Bio-RadLaboratories). Protein samples were solubilized in Laemllisample buffer and were heated at 100C for 10 min and cen-trifuged at 13,000

� g for 7 min to remove cell debris. Ratbrain lysate was obtained from BD Biosciences (Erembode-gem, Belgium) and was used as a positive control for Cx43protein expression on Western blots. Approximately 55

�gof total protein from each sample was separated on 12%Tris-HCl precast gels (BioRad Laboratories). After separa-tion, proteins were transferred to a nitrocellulose membrane(VWR; Zaventem, Belgium), blocked for 1 hr in TBS con-taining 5% non-fat dried milk and 0.1% Tween (blockingbuffer), and incubated overnight at 4C with monoclonal an-tibodies to Cx43 diluted 1:250 in blocking buffer. After ex-tensive washes in blocking buffer, membranes were probedfor 1 hr with HRP-conjugated secondary goat anti-mouseIgG antibody diluted in blocking buffer (1:2000) and devel-oped by enhanced chemiluminescence (Cell Signaling Tech-nology; Westburg) according to the manufacturer’s instruc-tions. The time of exposure (typically 10 min) to X-ray films(Kodak Biomax Light-1; Sigma-Aldrich) was identical for allexperimental conditions. The intensity of the bands afterWestern blotting was determined by scanning of the X-rayfilms followed by quantitative analysis using Total-lab soft-ware (Phoretix).

Transmission Electron MicroscopyThe granulosa explants were fixed by immersion for 2 hr in2% glutaraldehyde buffered with 0.1 M Na-cacodylate con-taining 1 mM CaCl2 (pH 7.4). After a rinse in 0.1 M Na-cacodylate containing 7.5% sucrose, the explants were os-micated in 2% OsO4 in 0.1 M Na-cacodylate and embeddedin LX medium (Ladd; Burlington, VT). Ultrathin sectionswere stained with uranyl acetate and lead citrate and wereanalyzed on a Jeol 1200 EXII electron microscope.

Fluorescence Recovery After Photobleaching (FRAP)Gap junctional coupling was investigated in granulosa cellsheets or explants using the FRAP technique (Braet et al.2003). For this purpose, the sheets or explants were loadedwith the gap junction-permeable fluorescent tracer 5-carboxy-fluorescein diacetate (CFDA; Molecular Probes, Leiden, TheNetherlands) by incubation in Hank’s buffered salt solution(HBSS-HEPES) containing 50

�M CFDA for 60 min atroom temperature and under continuous shaking. The prep-arations were then rinsed three times with HBSS-HEPES andtransferred to the microscope stage of the CLSM describedabove. The loading and rinsing solutions contained in addi-tion 1 mM probenecid (Sigma) to prevent rapid dye ex-trusion from the cells (Di Virgilio et al. 1988; D’Herde andLeybaert 1997). Fluorescence within a single cell was pho-

tobleached by spot exposure to the 488-nm line of an argonlaser over a period of 10 sec. Spot exposure was done by re-ducing the scan amplitude of the horizontal and verticalscanning mirrors. CFDA fluorescence before and after pho-tobleaching was recorded in confocal imaging mode withnormal scanning mirror settings. Recovery of fluorescencebecause of dye influx from neighboring cells was quantifiedat 4 min after the start of photobleaching. A correction wasmade for the baseline bleaching effect associated with re-peated exposure to the 488-nm imaging light. This was doneby recording the fluorescence profile in cells not exposed tothe spot bleach and correcting the recovery signal in propor-tion to the baseline fluorescence decay.

Gap Junction Blocking by 18

-Glycyrrhetinic AcidEighteen

-glycyrrhetinic acid (3

-hydroxy-11-oxo-18

,20

-olean-12-en-29-oic acid, AGA; Sigma) was dissolvedin dimethylsulfoxide (DMSO; UCB, Leuven, Belgium). Tostudy the effect of gap junction blocking on cell survival,granulosa cell explants were incubated for 24 hr with AGAdoses ranging from 25

�M to 250

�M and AIs were calcu-lated on DAPI- stained granulosa cell explants as describedabove. Cell coupling after treatment with AGA (250

�M, 24hr) was investigated using FRAP.

Statistical AnalysisThe data are expressed as mean

� SEM, with n denoting thenumber of experiments. Statistical significance was testedusing a t-test for unpaired observations and using a p valueof less than 0.05. Multiple groups were compared usingvariance analysis, followed by the Dunnett test for multiplecomparisons to a control group or the Student-Newman-Keuls test for comparison of all groups among each other.Curve fittings for dose–response relations were performedwith non-linear least-square procedures available in the pro-gram Inplot.

ResultsApoptosis and Cx43 ICC

Freshly isolated preparations were used for control pur-poses, 24-hr serum-free (24h-SF) cultures as the apop-

Figure 1 Effect of the culture conditions on the AI in granulosacell explants. Culturing granulosa cell explants under 24h-SF condi-tion induced apoptosis, while culturing granulosa cell explants for24h-SF-H or 24h-SF-Ca supplemented conditions inhibited this in-duced apoptosis. The AI in fresh granulosa cell sheets was negligi-ble and therefore is not shown. *p0.05 compared with 24h-SF;n�60 in six different animals.


The

Jour

nal o

f His

toch

emis

try

& C

ytoc

hem

istr

y

tosis-inducing condition (D’Herde and Leybaert 1997),and 24h-SF cultures supplemented either with hor-mones (IGF-I and LH; 24h-SF-H) (Onagbesan andPeddie 1988; Krysko et al. 2001) or with an increasedconcentration of extracellular calcium (24h-SF-Ca)(Mussche et al. 2000; and Mussche et al. unpublisheddata) as the apoptosis-inhibiting conditions. The AIsobserved under these experimental conditions are il-lustrated in Figure 1.

Staining of freshly isolated granulosa cell sheetswith anti-Cx43 antibodies demonstrated immunoreac-tivity appearing as a fine and continuous trace, liningthe cell borders and involving both places of cell–cellcontact as well as non-junctional regions (Figures 2Aand 2E). Granulosa cell explants in 24h-SF culturesdisplayed gap junctional plaques in the form of largedots or lines at cell–cell interfaces (Figures 2B, 2F, and2I). On average, each cell contained 5.24

� 0.11(n

�100 cells) of these immunopositive dots or lines.Apoptotic cells, easily recognized by their nuclearmorphology by DAPI staining, did not show Cx43 im-munolabeling at their plasma membrane (Figure 2I).However, occasionally apoptotic bodies demonstrated

very weak dot-like cytoplasmic staining (Figure 2I).Interestingly, DAPI staining revealed apoptotic granu-losa cells that were lying in clusters or streaks (Figure2J). Granulosa cell explants in 24h-SF-H culture dis-played a staining pattern that was regular and moreintense than in 24h-SF culture condition. The gapjunctions appeared to be huge as well (Figures 2C and2G) with on average 6.30

� 0.14 (n

�100 cells) ofthese gap junctional immunopositivities per cell (sig-nificantly higher compared with 24h-SF granulosa cellexplants). In granulosa cell explants cultured under24h-SF-Ca conditions, the immunostaining appearedas a dotted pattern of many short gap junctions sur-rounding each cell (Figures 2D and 2H). Quantifica-tion of the level of Cx43 immunoreactivity, deter-mined from the average intensity of confocal imagesacquired in the four conditions, demonstrated that theimmunoreactivity in 24h-SF granulosa cell explantswas significantly below the level observed in freshlyisolated granulosa cells. The Cx43 immunoreactivitywas significantly higher under the 24h-SF-H and 24h-SF-Ca, i.e., apoptosis-inhibiting conditions, comparedwith 24h-SF (Figure 3A).

Figure 2 Epifluorescence micro-graphs of granulosa cells. (A–D)Stained with anti-Cx43 antibodyalone or (E–I) in combination with DAPI.(J) Stained with DAPI alone. (A,E) Infreshly isolated granulosa cell sheets,gap junctions appear as a fine andcontinuous line bordering the cells.Immunoreactivity covers both junc-tional and non-junctional regions.Cx43 immunoreactivity is also presentin the cytoplasm. (B,F,I) In granulosacell explants cultured under 24h-SFconditions, gap junctions appear asbig dots or lines and are less apparentin the cytoplasm. (I) Note apoptoticcells with absence of Cx43 immunola-beling (arrows). However, occasion-ally Cx43 immunoreactivity in the cy-toplasm of the apoptotic cells isobserved (arrowheads). (C,G) In gran-ulosa cell explants cultured under24h-SF-H supplementation, gap junc-tions stain more intensely and arelarger compared with 24h-SF. (D,H) Ingranulosa cell explants cultured un-der 24h-SF-Ca supplementation, apunctate pattern of many small gapjunctions is observed. Fragmentationof gap junctions is apparent. (J) Apop-totic granulosa cells are organized inclustered or streaks. Bars � 10 �m.

81 Results: Part II

The

Jour

nal o

f His

toch

emis

try

& C

ytoc

hem

istr

y

Level of Cx43 Expression as Studiedby Western Blotting

Granulosa cell explants, both cultured and uncultured,express a 43-kD protein that reacts with Cx43-specificantibodies on Western blots. Immunoblots of wholecell lysates showed that 24h-SF granulosa cells ex-plants contained only 39% of what is expressed infreshly isolated granulosa cells sheets, while the levelof Cx43 expression recovered to 74% and 76% underthe 24h-SF-H and 24h-SF-Ca conditions, respectively(Figure 3B), which is in accordance with the data ob-tained by ICC analysis. Combination of the quanti-

tative data from ICC analysis and Western blottingshowed that Cx43 expression was inversely related tothe AIs (Figure 3C).

Electron Microscopy

Electron microscopic analysis of granulosa cells dem-onstrated rare gap junctions in freshly isolated granu-losa cells despite the positive Cx43 immunoreactivitydescribed above, and showed huge gap junctions in24h-SF and 24h-SF-H conditions (Figure 4). In 24h-SF-Ca conditions, many short gap junctions were ob-served in accordance with the ICC punctate pattern(Figure 4). Internalization of gap junctions did notoccur, as evidenced by the absence of annular junc-tions.

Figure 3 (A) Cx43 immunofluorescence intensity as determinedfrom the average intensity in confocal laser scanning micrographs.Immunofluorescence intensity was lower in 24h-SF cultured granu-losa cell explants compared with freshly isolated granulosa cellsheets (Fresh), and was again higher in the 24h-SF-H and 24h-SF-Caconditions. Cx43 immunofluorescence is expressed relative to a100% control level of freshly isolated granulosa cell sheets. *, #p0.001 compared with all other groups; n�36 in one animal. Allfields investigated contained equal numbers of cells. (B) Immuno-blot probed with mouse monoclonal anti-Cx43 antibody, showingthe expression level of Cx43 in granulosa cells explants. Lane 1,freshly isolated; Lane 2, 24-SF; Lane 3, 24-SF-H; Lane 4, 24-SF-Ca;Lane 5, positive control with rat cerebrum lysates. This immunoblotis representative of three similar experiments. (C) Inverse relationof Cx43 immunoreactivity and apoptotic indexes. The Cx43immmunoreactivity expressed in abcisssa are average values calcu-lated from the experiments presented in A and B.

Figure 4 Electron microscopic micrographs. (A) Granulosa cell ex-plants cultured in 24h-SF condition. Note a huge gap junction (ar-rows). (B,C) Granulosa cell explants cultured in 24h-SF-Ca supple-mented condition. Fragmented dot-like gap junctions (arrows). (B)An overview of C. Bars: A,C � 0.5 �m; B � 1 �m.


The

Jour

nal o

f His

toch

emis

try

& C

ytoc

hem

istr

y

Functional Coupling as Studied by FRAP

The level of gap junctional coupling in granulosa cellexplants cultured in 24h-SF conditions was signifi-cantly higher compared with the freshly isolated gran-ulosa cell sheets (Figure 5). In 24h-SF-H treated gran-ulosa cell explants, the level of gap junctional couplingwas comparable to the freshly isolated conditions, whilegranulosa cell explants kept under 24h-SF-Ca displayedthe highest level of coupling (Figure 5).

Effect of Gap Junction Blockade Experimentson Apoptotic Indexes

To investigate the effect of gap junction blocking oncell survival, granulosa cell explants were incubatedfor 24 hr in serum-free condition with increasingdoses of AGA (25

�M to 250

�M range) and AIs wereevaluated by DAPI staining. These results showed thatthe AI was dose-dependently decreased by AGA, witha half-maximal effect occurring at

�40

�M (Figure 6).FRAP experiments showed that 24 hr exposure of thegranulosa cell explants to AGA caused a drastic reduc-tion of gap junctional coupling, with 250

�M AGAdecreasing the fluorescence recovery from a controllevel of 45.8

� 2.6% to 3.2

� 0.6% (n

�8 in one ani-mal; p

0.00001).

DiscussionThe purpose of the present study was to investigatethe role of cell coupling and level of connexin43 ex-pression in apoptosis and survival of granulosa cell ex-plants. We demonstrated that apoptosis induction ingranulosa cell explants is accompanied by an increasedlevel of gap junctional coupling, that apoptosis wasdose-dependently inhibited by the gap junction blockerAGA, and that apoptosis induction is accompanied bya decreased level of Cx43 expression, which is re-gained by the two applied survival protocols.

Granulosa cell explants provide a model systemclosely mimicking the apoptotic process that occurs invivo during spontaneous follicular atresia (D’Herdeand Leybaert 1997). Contributing factors to explainthe relatively low apoptotic rates in 24h-SF granulosacell explants might be the presence of native extracel-lular matrix, i.e., the vitelline and basement mem-branes, which are still present in the used modelsystem and which are known to promote survival(Amsterdam et al. 1989; Boudreau et al. 1995). Sec-ond, as reviewed by Johnson (2000), death-suppress-ing members of the Bcl-2 family (Bcl-xLong, Bcl-2) areexpressed in granulosa cells from preovulatory folli-cles. Another contributing factor is the expression ofIAP-1, an inhibitor of apoptosis protein found in pre-ovulatory hen granulosa cells (Johnson et al. 1998).

The data of the present study demonstrated that, infreshly isolated granulosa cells, Cx43 immunolabelingwas observed all around the cells, even in those re-gions of the plasma membrane where visible contactwith any apposing plasma membrane of neighboringcells was missing. In the absence of any apposing cells,immunolabeling in the non-junctional plasma mem-brane regions is assumed to represent hemichannels ofCx43 (Musil and Goodenough 1991; Zampighi et al.1999; Quist et al. 2000). Quantification of the Cx43immunoreactivity (as revealed by ICC and Westernblotting) in freshly isolated granulosa cell sheets showeda higher amount of Cx43 protein compared with 24h-SF or 24h-SF-H supplemented cultures in which atEM level long gap junctions between neighboring cellsare easily detected. Despite the detected Cx43 immu-nolabeling and coupling of the cells as assessed byFRAP, extremely rare gap junctions were detected byEM analysis in these freshly isolated granulosa cells.This discrepancy between ultrastructural detection ofgap junctions and ICC detection of their major com-

Figure 5 Gap junctional coupling as studied with FRAP. Gap junc-tional coupling was more prominent (larger extent of fluorescencerecovery) in 24h-SF conditions compared with freshly isolated con-ditions (Fresh). Under 24h-SF-H conditions, gap junctional couplingwas no different from the freshly isolated condition, whereas the24h-SF-Ca condition was associated with a very large coupling de-gree. *p0.05, **p0.01 compared with the fresh condition; n�25in at least three different animals.

Figure 6 Dose–response relation between inhibition of gap junc-tional coupling with increasing doses of AGA and the resulting ef-fect on AI. AGA dose-dependently inhibited apoptosis, with half-maximal inhibition occurring with concentrations in the order of40 �M. *p0.05 compared with control (con).

83 Results: Part II

The

Jour

nal o

f His

toch

emis

try

& C

ytoc

hem

istr

y

ponent Cx43 might be explained by the fact that im-munoreactivity is not confined only to gap junctionsbut also reveals Cx43 present in the cytoplasm or or-ganized as connexin hemichannels. Moreover, one shouldconsider the fact that clustering of gap junctions intoplaques is the primary basis for their ultrastructuralidentification (Goodenough and Revel 1970; McNuttand Weinstein 1970).

In 24h-SF cultured granulosa cell explants, the apop-tosis-inducing condition, well-organized gap junctionswere observed (as revealed by ICC and EM), provid-ing a better coupling of the granulosa cells than infreshly isolated granulosa cells as measured by FRAPanalysis. However, a lower immunoreactive signalwas detected in these 24h-SF granulosa cells (as re-vealed by quantitative analysis of both the confocalimages and immunoblots) compared with freshly iso-lated granulosa cells. These apparently inconsistentfindings underscore the fact that the abundance of Cx43does not necessarily correlate with the amount of in-tracellular communication (Lau et al. 1992; Kurataand Lau 1994; Laing et al. 1997; Sia et al. 1999).

In agreement with the data from the present study,Wilson et al. (2000) found an average increase of gapjunctional coupling during initiation of apoptosis inserum-deprived (24-hr) cultures of a normal diploidepithelial cell line (WB-F344), assessed by the scrapeload/dye transfer method. In the same study, individ-ual cells were analyzed by the FRAP assay to comparenuclear morphology to differences in dye transfer dur-ing serum deprivation. The authors demonstrated thatmost apoptotic bodies did not communicate withneighboring cells. This is in line with the present ICCfindings revealing that granulosa cells with clear apop-totic nuclear morphology do not show Cx43 immuno-labeling at their plasma membrane. Kalvelyte et al.(2003) reported that expression of Cx43 in HeLa-transfected cells accelerates the transition of cells froman early phase of apoptosis to the late phase.

To verify whether cell coupling is indeed a means ofpropagating a cell death message, experiments usingthe gap junctional intercellular communication inhibi-tor AGA were carried out. This compound was shownto block cell coupling (Davidson et al. 1986), presum-ably by disassembling gap junctional channels and af-fecting their dephosphorylation (Guan et al. 1996). Inthe present model system, gap junctional coupling wassignificantly inhibited after long-term AGA exposure.Moreover, it was demonstrated that, by blocking gapjunctional coupling with AGA, AIs were dose-depen-dently decreased (Figure 6). In agreement with the presentdata, Krutovskikh et al. (2002) showed that in BC31cells (a rat bladder carcinoma cell line) cell couplingand apoptosis were significantly inhibited by treat-ment with AGA. In addition, Lin et al. (1998) re-ported that gap junctions achieved by transfection of

Cx43 could mediate the propagation of a death signalbetween dying and healthy glial cells in a co-culturesystem.

A further argument in support of a role of cell cou-pling in the propagation of a death signal amonggranulosa cells comes from DAPI-stained granulosaexplants in the 24h-SF condition (Figure 2J), whichdemonstrates that granulosa cells did not die ran-domly but, rather, in tracks of closely associated cells.This finding supports the observation described byCotrina et al. (1998) and Krutovskikh et al. (2002)that communicating cells die by clusters. Althoughapoptosis in granulosa cell explants is associated withincreased cell coupling (as discussed above), it is im-portant to note that increased cell coupling does notnecessarily induce apoptosis per se, as illustrated inthe Ca2�-rescue protocol. Therefore, our findings andthose of others (Lin et al. 1998; Sai et al. 2001; Kru-tovskikh et al. 2002; Kalvelyte et al. 2003) are con-sistent with the hypothesis that gap junctional inter-cellular communication is indeed needed for cell–celltransmission of a death signal and thus may contrib-ute to the regulation of the apoptotic process in di-verse model systems.

It is known that gap junctions are permeable to sev-eral second messengers, such as inositol-1,4,5-triphos-phate, ATP, cAMP, and calcium.

In a rat bladder carcinoma line, it has been demon-strated that a reduction of gap junction permeabilitywith oleamide, which appears not to affect the passageof Ca2� ions, did not abrogate coordinated cell deathby clusters, suggesting that Ca2� ions are involved asthe signal propagating cell death (Krutovskikh et al.2002).

Interestingly, Cx43 immunoreactivity (as studiedby ICC and immunoblotting) was inversely correlatedwith apoptosis (Figure 3C), suggesting that Cx43 ex-pression plays a role in the survival process. Sassonand Amsterdam (2002) reported that apoptosis induc-tion by LH and forskolin was accompanied by in-creased expression of Cx43 in human luteinized gran-ulosa cells, a discrepancy that might be related to theuse of different cell models and apoptosis-inducingstimuli. Several reports point to the fact that Cx43plays a role in promoting survival in response to ex-tracellular cues independently of gap junctional inter-cellular communication (Plotkin and Bellido 2001;Plotkin et al. 2002). Huang et al. (2001) demonstratedthat regulation of apoptosis by Cx43 in human glio-blastoma cells is mediated by downregulation of theapoptosis inhibitor protein Bcl-2 and hence is notlinked to gap junction channel function. Recently, Linet al. (2003) demonstrated that forced expression ofCx43 as well as two other members of the connexinfamily (Cx32 and Cx40) increased the resistance of as-trocytes to injury and that the anti-death activity of


The

Jour

nal o

f His

toch

emis

try

& C

ytoc

hem

istr

y

connexin proteins was independent of the functionalstatus of gap junctions.

In conclusion, the present study suggests that gapjunctions mediate the propagation of a cell death mes-sage, while independently Cx43 expression is neededfor survival. Future studies will focus on the experi-mental modulation of Cx43 expression and its effectson the major pro- and anti-apoptotic signaling path-ways.

AcknowledgmentsSupported by Ghent University (GOA grant no. 12050502

to KD) and the Fund for Scientific Research Flanders, Bel-gium (FWO grant nos. 3G023599, 3G001201, and G.0335.03to LL).

We are very grateful to Barbara De Bondt, Dominique Ja-cobus, and Hubert Stevens for excellent technical assistance.

Literature CitedAdashi EY (1992) The potential relevance of cytokines to ovarian

physiology. J Steroid Biochem Mol Biol 43:439–444Adashi EY, Resnick CE, Hurwitz A, Ricciarelli E, Hernandez ER,

Roberts CT, Leroith D, et al. (1991) Insulin-like growth factors:the ovarian connection. Hum Reprod 6:1213–1219

Amsterdam A, Rotmensch S, Furman A, Venter EA, Vlodavsky I(1989) Synergistic effect of human chorionic gonadotropin andextracellular matrix on in vitro differentiation of human granu-losa cells: progesterone production and gap junction formation.Endocrinology 124:1956–1964

Beyer EC (1990) Molecular cloning and developmental expressionof two chick embryo gap junction proteins. J Biol Chem 265:14439–14443

Boudreau N, Sympson CJ, Werb Z, Bissell MJ (1995) Suppressionof ICE and apoptosis in mammary epithelial cells by extracellularmatrix. Science 267:891–893

Bradford MM (1976) A rapid and sensitive method for the quanti-tation of microgram quantities of protein utilizing the principleof protein-dye binding. Anal Biochem 72:248–254

Braet K, Vandamme W, Martin PE, Evans WH, Leybaert L (2003)Photoliberating inositol-1,4,5-trisphosphate triggers ATP releasethat is blocked by the connexin mimetic peptide gap 26. Cell Cal-cium 33:37–48

Bruzzone R, White TW, Goodenough DA (1996) The cellular inter-net: on-line with connexins. Bioessays 18:709–718

Cotrina ML, Kang J, Lin JHC, Bueno E, Hansen TW, He L, Liu Y,et al. (1998) Astrocytic gap junctions remain open during ische-mic conditions. J Neurosci 18:2520–2537

Davidson JS, Baumgarten IM, Harley EH (1986) Reversible inhibi-tion of intercellular junctional communication by glycyrrhetinicacid. Biochem Biophys Res Commun 134:29–36

Dekel N (1987) Interaction between the oocyte and the granulosacells in the preovulatory follcile. In Armstrong DT, Freisen HG,Leung PCK, Moger W, Ruf KB, eds. Endocrinology and Physiol-ogy of Reproduction. New York, Plenum Publishing, 197–209

D’Herde K, De Prest B, Mussche S, Schotte P, Beyaert R, CosterRV, Roels F (2000) Ultrastructural localization of cytochrome cin apoptosis demonstrates mitochondrial heterogeneity. CellDeath Differ 7:331–337

D’Herde K, Leybaert L (1997) Intracellular free calcium related toapoptotic cell death in quail granulosa cell sheets kept in serum-free culture. Cell Death Differ 4:59–65

Di Virgilio F, Steinberg TH, Swanson JA, Silverstein SC (1988) Fura-2secretion and sequestration in macrophages. A blocker of organicanion transport reveals that these processes occur via a membranetransport system for organic anions. J Immunol 140:915–920

Farioli-Vecchioli S, Raes S, Espeel M, Roels F, D’Herde K (2000)Inverse expression of peroxisomes and connexin-43 in the granu-losa cells of the quail follicle. J Histochem Cytochem 48:167–178

Gilbert AB, Evans AJ, Perry MM, Davidson MH (1977) A methodfor separating the granulosa cells, the basal lamina and the thecaof the preovulatory ovarian follicle of the domestic fowl (Gallusdomesticus). J Reprod Fertil 50:179–181

Gittens JE, Mhawi AA, Lidington D, Ouellette Y, Kidder GM(2003) Functional analysis of gap junctions in ovarian granulosacells: distinct role for connexin43 in early stages of folliculogene-sis. Am J Physiol 284:C880–887

Goodenough DA, Revel JP (1970) A fine structural analysis of inter-cellular junctions in the mouse liver. J Cell Biol 45:272–290

Grazul-Bilska AT, Reynolds LP, Redmer DA (1997) Gap junctionsin the ovaries. Biol Reprod 57:947–957

Guan X, Wilson S, Schlender KK, Ruch RJ (1996) Gap-junctiondisassembly and connexin 43 dephosphorylation induced by 18beta-glycyrrhetinic acid. Mol Carcinog 16:157–164

Huang RP, Hossain MZ, Huang R, Gano J, Fan Y, Boynton AL(2001) Connexin 43 (cx43) enhances chemotherapy-inducedapoptosis in human glioblastoma cells. Int J Cancer 92:130–138

Jiang JY, Cheung CK, Wang Y, Tsang BK (2003) Regulation of celldeath and cell survival gene expression during ovarian folliculardevelopment and atresia. Front Biosci 8:d222–237

Johnson AL (2000) Granulosa cell apoptosis: conservation of cellsignaling in an avian ovarian model system. Biol Signals Recept9:96–101

Johnson AL, Bridgham JT, Digby MR, Lowenthal JW (1998) Ex-pression of the inhibitor of T-cell apoptosis (ita) gene in hen ova-rian follicles during development. Biol Reprod. 58:414–420

Kalvelyte A, Imbrasaite A, Bukauskiene A, Verselis VK, BukauskasFF (2003) Connexins and apoptotic transformation. BiochemPharmacol 66:1661–1672

Kidder GM, Mhawi AA (2002) Gap junctions and ovarian follicu-logenesis. Reproduction 123:613–620

Krutovskikh VA, Piccoli C, Yamasaki H, Yamasali H (2002) Gapjunction intercellular communication propagates cell death incancerous cells. Oncogene 21:1989–1999

Krysko DV, Roels F, Leybaert L, D’Herde K (2001) Mitochondrialtransmembrane potential changes support the concept of mito-chondrial heterogeneity during apoptosis. J Histochem Cy-tochem 49:1277–1284

Kumar NM, Gilula NB (1996) The gap junction communicationchannel. Cell 84:381–388

Kurata WE, Lau AF (1994) p130gag-fps disrupts gap junctionalcommunication and induces phosphorylation of connexin43 in amanner similar to that of pp60v-src. Oncogene 9:329–335

Laing JC, Tadros PN, Westphale EM, Beyer EC (1997) Degradationof connexin43 gap junctions involves both the proteasome andthe lysosome. Exp Cell Res 236:482–492

Lau AF, Kanemitsu MY, Kurata WE, Danesh S, Boynton AL (1992)Epidermal growth factor disrupts gap-junctional communicationand induces phosphorylation of connexin43 on serine. Mol BiolCell 3:865–874

Lin JH, Weigel H, Cotrina ML, Liu S, Bueno E, Hansen AJ, HansenTW, et al. (1998) Gap-junction-mediated propagation and am-plification of cell injury. Nature Neurosci 1:494–500

Lin JHC, Yang J, Liu S, Takano T, Wang X, Gao Q, Willecke K, etal. (2003) Connexin mediates gap junction-independent resis-tance to cellular injury. J Neurosci 23:430–441

McNutt NS, Weinstein RS (1970) The ultrastructure of the nexus.A correlated thin-section and freeze-cleave study. J Cell Biol 47:666–688

Moorby C, Patel M (2001) Dual functions for connexins: Cx43 reg-ulates growth independently of gap junction formation. Exp CellRes 271:238–248

Munari-Silem Y, Rousset B (1996) Gap junction-mediated cell-to-cell communication in endocrine glands-molecular and func-tional aspects: a review. Eur J Endocrinol 135:251–264

Musil LS, Beyer EC, Goodenough DA (1990) Expression of the gap

85 Results: Part II

junctional protein connexin-43 in embryonic chick lens: molecu-lar cloning, ultrastructural localization and post-translationalphosphorylation. J Membr Biol 116:163–175

Musil LS, Goodenough DA (1991) Biochemical analysis of con-nexin43 intracellular transport, phosphorylation, and assemblyinto gap junctional plaques. J Cell Biol 115:1357–1374

Mussche S, Leybaert L, D’Herde K (2000) First and second messen-ger role of calcium. Survival versus apoptosis in serum-free cul-tured granulosa explants. Ann NY Acad Sci 926:101–115

Onagbesan OM, Peddie MJ (1988) Induction of ovulation and ovi-position in female quail with luteinizing hormone, luteinizinghormone releasing hormone, or progesterone. Gen Comp Endo-crinol 71:124–131

Plotkin LI, Bellido T (2001) Bisphosphonate-induced, hemichannel-mediated, anti-apoptosis through the Src/ERK pathway: a gapjunction-independent action of connexin43. Cell Commun Ad-hes 8:377–382

Plotkin LI, Manolagas SC, Bellido T (2002) Transduction of cellsurvival signals by connexin-43 hemichannels. J Biol Chem 277:8648–8657

Quist AP, Rhee SK, Lin H, Lal R (2000) Physiological role of gap-junctional hemichannels. Extracellular calcium-dependent isos-motic volume regulation. J Cell Biol 148:1063–1074

Rosenfeld CS, Wagner JS, Roberts RM, Lubahn DB (2001) Intra-ovarian actions of oestrogen. Reproduction 122:215–226

Sai K, Kang KS, Hirose A, Hasegawa R, Trosko JE, Inoue T (2001)Inhibition of apoptosis by pentachlorophenol in v-myc-transfected

rat liver epithelial cell: relation to down-regulation of gap junc-tional intercellular communication. Cancer Lett 173:163–174

Sanderson MJ, Parker I (2003) Video-rate confocal microscopy. InMarriott G, Parker I, eds. Biophotonics. Methods in Enzymol-ogy. Vol 360. San Diego, Academic Press, 447–481

Sasson R, Amsterdam A (2002) Stimulation of apoptosis in humangranulosa cells from in vitro fertilization patients and its preven-tion by dexamethasone: involvement of cell contact and bcl-2 ex-pression. J Clin Endocrinol Metab 87:3441–3451

Sia MA, Woodward TL, Turner JD, Laird DW (1999) Quiescentmammary epithelial cells have reduced connexin43 maintain ahigh level of gap junction intercellular communication. DevGenet 24:111–122

Sommersberg B, Bulling A, Salzer U, Frohlich U, Garfield RE, Am-sterdam A, Mayerhofer A (2000) Gap junction communicationand connexin 43 gene expression in a rat granulosa cell line: regu-lation by follicle-stimulating hormone. Biol Reprod 63:1661–1668

Tilly JL (1997) Apoptosis and the ovary: a fashionable trend orfood for thought? Fertil Steril 67:226–228

Trosko JE, Ruch RJ (1998) Cell-cell communication in carcinogene-sis. Front Biosci 3:D208–236

Wilson MR, Close TW, Trosko JE (2000) Cell population dynamics(apoptosis, mitosis, and cell-cell communication) during disrup-tion of homeostasis. Exp Cell Res 254:257–268

Zampighi GA, Loo DD, Kreman M, Eskandari S, Wright EM(1999) Functional and morphological correlates of connexin50expressed in Xenopus laevis oocytes. J Gen Physiol 113:507–524


Phagocytosis of apoptotic and necrotic cells

Hand of God, A Rodin

Ten times thyself were happier than thou art,If ten of thine ten times refigured thee:Then what could death do, if thou shouldst depart,Leaving thee living in posterity?Be not self-will'd, for thou art much too fairTo be death's conquest and make worms thine heir.

Sonnet 6, Then let not winter's ragged hand defaceW Shakespear

3. Burial level

3.1 Phagocytosis of necrotic cells by macrophages is phosphatidylserine dependent and does not induce inflammatory cytokine production

G Brouckaert, M Kalai, DV Krysko, X Saelens, D Vercammen, M Ndlovu, G Haegeman, K D’Herde, P VandenabeeleMolecular Biology of the Cell 2004; 15, 1089–1100.

Phagocytosis of Necrotic Cells by Macrophages IsPhosphatidylserine Dependent and Does Not InduceInflammatory Cytokine ProductionGreet Brouckaert,* Michael Kalai,*† Dmitri V. Krysko,‡ Xavier Saelens,*Dominique Vercammen,*§ ‘Matladi Ndlovu,� Guy Haegeman,�Katharina D’Herde,‡ and Peter Vandenabeele*†¶

*Molecular Signalling and Cell Death Unit, Department of Molecular Biomedical Research, VIB,Ghent University, Ghent Belgium; ‡Department of Human Anatomy, Embryology, Histology, andMedical Physics, Ghent University, Ghent, Belgium; and �Laboratory of Eukaryotic Gene Expressionand Signal Transduction, Department of Molecular Biology, Ghent University, Ghent, Belgium

Submitted September 16, 2003; Revised November 25, 2003; Accepted November 25, 2003Monitoring Editor: David Drubin

Apoptotic cells are cleared by phagocytosis during development, homeostasis, and pathology. However, it is still unclearhow necrotic cells are removed. We compared the phagocytic uptake by macrophages of variants of L929sA murinefibrosarcoma cells induced to die by tumor necrosis factor-induced necrosis or by Fas-mediated apoptosis. We show thatapoptotic and necrotic cells are recognized and phagocytosed by macrophages, whereas living cells are not. In both cases,phagocytosis occurred through a phosphatidylserine-dependent mechanism, suggesting that externalization of phospha-tidylserine is a general trigger for clearance by macrophages. However, uptake of apoptotic cells was more efficient bothquantitatively and kinetically than phagocytosis of necrotic cells. Electron microscopy showed clear morphologicaldifferences in the mechanisms used by macrophages to engulf necrotic and apoptotic cells. Apoptotic cells were taken upas condensed membrane-bound particles of various sizes rather than as whole cells, whereas necrotic cells wereinternalized only as small cellular particles after loss of membrane integrity. Uptake of neither apoptotic nor necrotic L929cells by macrophages modulated the expression of proinflammatory cytokines by the phagocytes.

INTRODUCTION

During embryonic development, tissue homeostasis, im-mune regulation, and other physiological processes, super-fluous and harmful cells are eliminated through an orderedcellular process called apoptosis that involves proteolyticactivation or inactivation of proteins by caspases (Earnshawet al., 1999; Lamkanfi et al., 2002a,b). During apoptosis, theplasma membrane blebs, chromatin is condensed, the nu-cleus is fragmented, DNA is degraded, and apoptotic bodiesare formed (Earnshaw et al., 1999). Nevertheless, the plasmamembrane of the dying cell typically remains intact until thelate stages of apoptosis, preventing leakage of cell contents(Denecker et al., 2001a). Rapid recognition, uptake, and deg-radation of apoptotic cells by phagocytes limits potentialtissue injury by avoiding the spilling of cellular contents and

prevents the occurrence of (auto)immune responses to intra-cellular autoantigens associated with dying cells (Ren andSavill, 1998). Apoptotic cells undergo specific surfacechanges that signal professional or nonprofessional phago-cytes to bind and engulf them. Among these markers is mostnotably the surface exposure of phosphatidylserine (PS) dueto loss of plasma membrane asymmetry (Henson et al., 2001;Scott et al., 2001). PS exposure is crucial for the uptake ofapoptotic cells (Fadok et al., 2001b) and precedes DNA deg-radation, zeiosis, and cell lysis (Verhoven et al., 1999;Denecker et al., 2000), indicating that it is an early event.When apoptotic cells are phagocytosed in an early stage ofthe cell death process, all of the subsequent apoptotic eventsoccur inside the phagocytes, including the DNA-fragmenta-tion by the macrophage lysosomal DNaseII instead of thecaspase-activated DNase CAD (McIlroy et al., 2000).

Necrotic cell death has classically been described in severeand acute physicochemical injury. However, necrosis canalso occur when a cell death stimulus is given under condi-tions where the apoptotic pathway is blocked by absence ofcaspase-8 and overexpression of Bcl-2, or by the presence ofviral or synthetic caspase-inhibitors (Vercammen et al., 1998;Matsumura et al., 2000; Kalai et al., 2002). Typically duringnecrosis, plasma membrane integrity is lost, the cell andorganelles swell, and eventually the cytosolic content isreleased in the surrounding tissue (Kitanaka and Kuchino,1999; Denecker et al., 2001a; Kalai et al., 2002), possiblyinitiating inflammatory or autoimmune responses (Fadok etal., 2001a). Several studies demonstrated that necrosis occurs

Article published online ahead of print. Mol. Biol. Cell 10.1091/mbc.E03–09–0668. Article and publication date are available atwww.molbiolcell.org/cgi/doi/10.1091/mbc.E03–09–0668.

† These authors share senior authorship.§ Present address: Department of Plant System Biology, VIB,

Ghent University, Ghent, Belgium.¶ Corresponding author. E-mail address: peter.vandenabeele@

dmbr.ugent.be.Abbreviations used: CTGr, Cell Tracker Green; CTOr, Cell TrackerOrange; FACS, fluorescence-activated cell sorter; FasL, Fas ligand;IFN, interferon; IL, interleukin; LPS, lipopolysaccharide; PI, pro-pidium iodide; PS, phosphatidylserine; TGF-�, tumor growth factor-�;TNF, tumor necrosis factor; TNF-R1, TNF-receptor 1; zVAD-fmk,benzyloxycarbonyl-Val-Ala-Asp(OMe)-fluoromethylketone.

87

during embryonic development in a programmed, nonacci-dental manner. Caspase-independent necrotic cell death ofthe interdigital cells can replace apoptosis and correctlyshape the fingers of Apaf-1–deficient mice or mice receivingzVAD-fmk, suggesting that necrosis can function appropri-ately in a developmental context (Chautan et al., 1999). Ne-crosis also occurs in pathological conditions such as bacterialinfection (Francois et al., 2000), inflammatory disease (Wanget al., 1996), brain ischemia (Beilharz et al., 1995), myocardialinfarction, and Alzheimer’s and Parkinson’s diseases(Kitanaka and Kuchino, 1999). Moreover, caspase-indepen-dent necrotic cell death in T cells induced by Fas ligand isexecuted through a specific pathway requiring the presenceof RIP1 (Holler et al., 2000). Down-modulation of RIP1 bygeldanamycin prevents necrosis and results in caspase-de-pendent apoptosis in L929 cells treated with tumor necrosisfactor (TNF) (Vanden Berghe et al., 2003), suggesting that thecomposition of the death receptor complex may be decisivein determining the type of cell death.

Uptake of necrotic cells was reported to induce matura-tion of dendritic cells (Sauter et al., 2000) and production ofproinflammatory cytokines in phagocytes (Fadok et al.,2001a). Most studies on uptake of necrotic cells are based onmodel systems in which necrosis is caused by physical dam-age of the target cells, leading to rupture of the membraneand rapid release of the intracellular content. We developeda cellular system in which L929sAhFas cells die by apoptosiswhen stimulated with agonistic anti-Fas antibodies or solu-ble Fas ligand, and by necrosis when treated with TNF(Vercammen et al., 1997). Here, we used this cellular systemto study the molecular mechanisms involved in the phago-

cytosis of necrotic cells in comparison with the mechanismof clearance of apoptotic cells.

MATERIALS AND METHODS

CellsThe L929sA murine fibrosarcoma cell line, selected for its sensitivity to thecytotoxic activity of TNF (Vanhaesebroeck et al., 1991), was stably transfectedwith the cDNAs encoding the human TNF receptor (Vercammen et al., 1995)and the human Fas receptor (Vercammen et al., 1997, 1998), to produceL929sAhR55 and L929sAhFas cells, respectively. L929rTA cells (Vanhoe-nacker et al., 1999) were stably cotransfected with three plasmids: pPHT,pTet-tTS (BD Biosciences Clontech, Palo Alto, CA), and pUHD10–3-FKBP2-link-FADD coding for hygromycin resistance, a tetracyclin-controlled tran-scriptional silencer (Freundlieb et al., 1999), and a FKBP2-link-FADD fusionprotein, respectively.

Mf4/4 (Desmedt et al., 1998) and P388D1 are mouse macrophage cell linesand J774 is a mouse monocyte-derived macrophage cell line. Cells weregrown in RPMI medium (Invitrogen, Eggenstein, Germany), supplementedwith 10% fetal calf serum, penicillin (100 U/ml), glutamax I (200 �M), �-mer-captoethanol (2 � 10�5 mM), and sodium pyruvate (1 mM), and kept underlipopolysaccharide (LPS)-free conditions.

Antibodies, Cytokines, and ReagentsRecombinant murine TNF was produced in-house (Tavernier et al., 1987). Thereceptor-specific mutein of hTNF (R32WS86T) has been described previously(Van Ostade et al., 1993). Agonistic antihuman Fas antibody (IgG3, clone 2R2)was from BioCheck (Munster, Germany). Soluble FLAG-tagged Fas ligand(FasL) was produced and purified in-house, as described previously (Schnei-der et al., 1998). Recombinant annexin V (BD PharMingen, San Diego, CA)was used at 10 �g per 105 cells. Annexin V-fluorescein isothiocyanate (FITC)conjugate (BD PharMingen) was used at 1 �g/ml. Cell Tracker Green andOrange were from Molecular Probes (Eugene, OR). Fluorescent cell linkersPKH26 (Red) and PKH67 (Green) were from Sigma-Aldrich (St. Louis, MO).Propidium iodide (BD Biosciences, San Jose, CA) was used at 30 �M. Purifiedrat antimouse CD16/CD32 (Fc III/II receptor) IgG2b�� antibody (BD PharM-

Figure 1. Mf4/4 macrophage-like cells takeup apoptotic and necrotic L929sAhFas cells.(A) Dot plot representation of a flow fluoro-cytometrical analysis of CTOr-stained Mf4/4cells after 30-min coincubation with CTGr-stained target cells treated as follows: un-treated (control); 2 h, anti-Fas 250 ng/ml (ap-optotic); and 18 h, TNF 104 IU/ml (necrotic).Mf4/4 cells accumulate in squares I (no up-take) and II (uptake) and free target cells insquare III. The calculated percentage of dou-ble-positive Mf4/4 cells (% phagocytic) wasused as a measure of the engagement of mac-rophages in phagocytosis, and the increase inthe mean green fluorescence (� mean Gr FL)was used as an estimate of the amount clearedby the phagocytes. (B and C) Confocal micro-scopic visualization of uptake of apoptotic (B)and necrotic cells (C). Mf4/4 cells were la-beled with PKH26 (red) and target cells withPKH67 (green). Note in B an engulfed apo-ptotic cell nucleus. Bars (B and C), 0.5 �m.

88 Results: Part III

ingen) was used at 5 �g/ml. zVAD-fmk was from Enzyme Systems Products(Livermore, CA). Doxycyclin (Duchefa Biochemicals, Haarlem, The Nether-lands) was used at 1 �g/ml. The FKBP dimerizer AP1510 was kindly pro-vided by ARIAD Pharmaceuticals (Cambridge, MA) and used at 1 �M(Amara et al., 1997). Phycoerythrin-conjugated F4/80 rat-antibody was fromSerotec (Oxford, United Kingdom).

Induction of Apoptotic and Necrotic Cell DeathL929sAhR55 and L929sAhFas target cells were seeded at 2.5 � 105 cells/wellin uncoated 24-well tissue culture plates (Sarstedt, Newton, NC). The nextday, anti-Fas antibody (250 ng/ml), Fas ligand (150 ng/ml), or mTNF (10,000IU/ml) were added to L929sAhFas cells, and hTNF R32WS86T (100 ng/ml) toL929sAhR55 cells. Cells were harvested at indicated times and put on icebefore analysis with a FACSCalibur flow cytometer (BD Biosciences). Loss ofcell membrane integrity as a measure of cell death was determined bypropidium iodide (PI) fluorescence (excitation 535/emission 617). PS expo-sure was monitored by annexin V-FITC staining (excitation 494/emission 518)(BD PharMingen).

In Vitro Phagocytosis AssayTarget cells were stained with 5 �M Cell Tracker Green (excitation 492/emission 516) and seeded at 2.5 � 105 cells/well in uncoated 24-well suspen-sion tissue culture plates. Macrophage-like cells were stained with 5 �M CellTracker Orange (excitation 540/emission 566) and seeded in adherent 24-wellplates at 2.5 � 105 cells/well. Cells were cultured overnight at 37°C in 5%CO2. Target cells were induced to undergo apoptosis or necrosis and collectedfor coincubation with phagocytes at the indicated time points. For coincuba-tion, 200 �l of a 2.5 � 106/ml suspension of target cells was added to eachwell, resulting in a 1:1 ratio of macrophage-like cells to target cells. Cells werecocultured at 37°C in 5% CO2 for the indicated times, after which the cellswere detached from the plates with enzyme-free cell dissociation buffer(Invitrogen), washed, and resuspended in phosphate-buffered saline (PBS).Data acquisition was performed on a FACSCalibur flow cytometer, by usingCell Quest software (BD Biosciences). At each time point, 5000 cells wereanalyzed. Quantitative analysis was done using Flowjo software (Tree Star,Ashland, OR).

In Vivo Phagocytosis AssayL929rTA and L929rTAFADD cells stained with 5 �M CTGr were treated withdoxycyclin and AP1510, 10 and 5 h before injection, respectively. Afterwashing in PBS, 2 � 106 target cells were injected into the peritoneum of eachmouse. Mice were sacrificed 4 h later, and their peritoneal cells were recov-ered by flushing with 10 ml of PBS. The recovered cells were counted. The celltype composition was determined microscopically from cytospin slidesstained with May-Grunwald/Giemsa (Sigma-Aldrich). Phagocytosis was de-termined by flow cytometry as described above, after the staining of themacrophage population with a phycoerythrin-conjugated F4/80 antibody(McKnight et al., 1996).

Transmission Electron MicroscopyNonlabeled adherent cocultures of macrophages and target cells (prepared asdescribed above) were fixed by immersion in 2% glutaraldehyde containing 1mM CaCl2, buffered with 0.1 M Na-cacodylate (pH 7.4). After a rinse in 100mM Na-cacodylate containing 7.5% sucrose, the cocultures were osmicatedovernight in 1% OsO4 in the same buffer (without sucrose) and embedded inLX medium after dehydration in an ethanol ascending series (Ladd, Burling-ton, VT). Ultrathin 60-nm sections, mounted on Formvar-coated copper grids,were stained with uranyl acetate and lead citrate and examined with a JEOL1200 EXII electron microscope.

RNase Protection AssayL929sAhR55 and L929sAhFas target cells were seeded at 5 � 106 cells/platein uncoated 9-cm-diameter tissue culture dishes (Bibby Sterilin, Staffordshire,United Kingdom). Mf4/4 cells were seeded at 5 � 106 in adherent 9-cm-diameter culture dishes (Nunclon, Roskilde, Denmark). The next day,L929sAhFas cells were stimulated with Fas ligand (150 ng/ml) for 1 h andL929sAhR55 cells with hTNF R32WS86T (100 ng/ml) for 14 h. Mf4/4 cellswere left untreated or treated with 100 ng/ml LPS 2 h before addition ofcontrol or treated target cells. Target cells were added to the phagocytes andcoincubated for 1 h and then washed away. Total RNA was prepared usingRNeasy (QIAGEN, Westburg B.V., The Netherlands), 4 h later. CytokinemRNA levels were measured by RNase protection assay by using the Ribo-quant multiprobe set (BD PharMingen). In brief, 15 �g of RNA per samplewas hybridized overnight with a 32P-labeled RNA multiprobe template set(mck-3b; BD PharMingen). Single-stranded RNA and free probe were di-gested by RNase A and T1. Protected RNA was separated on a 6% denaturingpolyacrylamide gel and visualized and quantified using PhosphorImagersoftware (Molecular Dynamics, Sunnyvale, CA). Probes specific for mRNAsof the housekeeping genes L32 and GAPDH were included as negativecontrols. For quantification, cytokine values were expressed as a percentage ofthe mean values of L32 for each lane.

Interleukin (IL)-6 and TNF Bioactivity MeasurementIL-6 secreted in the incubation medium was quantified using a bioassay, viz.,IL-6–dependent growth of 7TD1 cells (Poupart et al., 1987). TNF concentra-tion was determined in a standardized cytotoxicity assay by using L929sAcells.

RESULTS

Flow Fluorocytometrical Analysis of Phagocytosis ofDying CellsTo quantify phagocytosis of apoptotic and necrotic cellsby macrophages, we designed a two-parameter flow cy-tometry phagocytosis assay in which Cell Tracker Green(CTGr) labeled apoptotic, necrotic, or viable controlL929sAhFas target cells were incubated at a ratio of 1:1with a monolayer of Mf4/4 macrophages labeled withCell Tracker Orange (CTOr). Target cells were labeled andthen treated with agonistic anti-Fas antibodies to elicitapoptotic cell death or with TNF to initiate necrotic celldeath. After 30 min of coincubation, macrophages andtarget cells were detached from the plates with enzyme-free dissociation buffer, a treatment that also detachesadhering target cells from the macrophages, as verified bylight microscopy. The extent of phagocytosis was quanti-fied by fluorescence activated cell sorter (FACS) analysis(Figure 1A). The percentage of double-stained macro-phages (square II) out of the whole macrophage popula-tion (squares I � II) measures the fraction of the macro-phage population involved in phagocytosis of target cells(% phagocytic). The difference in the mean green fluores-cence values of the macrophage population before and

Figure 2. Uptake of apoptotic and necrotic cells after PS exposure.Uptake of target cells stimulated with anti-Fas (250 ng/ml) (A) orTNF (10,000 IU/ml) (B) for the indicated time periods. Bars repre-sent percentage of double-positive Mf4/4 cells (% phagocytic), tri-angles percentage PS positive, and circles percentage of PI positivecells as a measure of membrane permeability. Error bars representthe SD of at least three independent experiments.

PS dependent clearance of necrotic cells 89

after coincubationwith the CTGr-labeled target cells (�mean Gr FL) determines the amount of target cell materialactually taken up by the macrophages.

These results were confirmed by confocal laser scanningmicroscopy, by using cell linker dye PKH26 (red) for Mf4/4cells, and PKH67 (green) for target cells (Figure 1, B and C).When apoptotic cells were coincubated with Mf4/4 cells,uptake consisted of apoptotic bodies of various sizes, up toseveral per phagocyte (Figure 1B). Occasionally, whole cellsor parts containing the nucleus of a dying cell were engulfed(Figure 1B). Phagocytosis of necrotic target cells was limitedto the engulfment of one or more small cytosolic particlesper phagocyte (Figure 1C).

Phagocytosis of Apoptotic and Necrotic Cells Occursthrough a PS-dependent MechanismApoptotic and necrotic cell death can be distinguished bythe differential appearance of two characteristics, which arePS exposure, as measured by annexin V-FITC labeling, andmembrane permeability, assessed by PI staining. In Fas-mediated apoptosis, PS exposure precedes loss of membraneintegrity, whereas in TNF-induced necrosis both processesoccur simultaneously (Denecker et al., 2001b). We deter-mined in parallel the kinetics of these parameters on targetcells and compared them with the extent of macrophage-mediated uptake after 30-min coincubation with apoptoticor necrotic cells. Macrophages specifically recognized dying

Figure 3. Pretreatment with recombinantannexin V inhibits the uptake of apoptoticand necrotic targets cells by Mf4/4 cells. Con-trol, apoptotic (2 h, anti-Fas) and necrotic (20h, TNF) L929hFas cells were treated with un-labeled recombinant annexin V (10 �g/105

cells) and stained with annexin V-FITC (A), orcoincubated for 30 min with Mf4/4 cells (B).Annexin V staining and phagocytosis wereanalyzed by flow cytometry. (C) Althoughearly apoptotic cells present a lower PS expo-sure level than necrotic or late apoptotic cells,as demonstrated by the relative mean in-crease in annexin V fluorescence with time [�mean Ann V (Fl)], they present better targetsfor phagocytosis. Empty bars represent per-centage of double-positive Mf4/4 cells (%phagocytic), whereas gray bars represent thepercentage of PS-positive target cells. Dia-monds represent the � mean Ann V (Fl). Errorbars represent the standard deviations of twoindependent experiments.


cells, because no uptake of living control cells was observed(Figure 1A, 2). Uptake of apoptotic cells coincided with therapid exposure of PS (Figure 2A). With necrotic cells, phago-cytosis occurred concomitantly with PI and PS positivity(Figure 2B). These results demonstrate that phagocytosis ofapoptotic and necrotic cells by macrophages, coincides withPS exposure. Therefore, we examined whether the recogni-tion and uptake of both types of dying cells is PS dependent,by shielding the PS exposed on the surface of apoptotic,necrotic and viable control target cells with recombinantunlabeled annexin V. This treatment not only protected thecells from annexin V-FITC staining (Figure 3A) but also ledto a drastic, nearly complete inhibition of phagocytosis ofboth apoptotic and necrotic cells (Figure 3B). These datademonstrate that a PS-dependent mechanism mediates theuptake of necrotic as well as apoptotic cells.

Phagocytosis of Necrotic Cells Is Less Efficient andRequires More TimeAlthough the levels of annexin V staining were higher withnecrotic and late apoptotic cells, the best targets for phago-cytosis were early apoptotic cells with a relatively low levelof mean annexin V staining (Figure 3C). Indeed, exposure ofmacrophages for 30 min to early apoptotic cells was suffi-

cient to induce phagocytic activity in most of the macro-phages (Figure 2). However, only about one-half of themacrophages engaged in uptake of necrotic cells. To excludethe possibility that early apoptotic cells were more efficientthan necrotic cells in inducing the phagocytic activity ofmacrophages due to Fc receptor anti-Fas antibody interac-tion, we compared the uptake of target cells by untreatedMf4/4 cells with others pretreated with a neutralizing ratanti-mouse Fc receptor antibody. Shielding the Fc receptoron the phagocytes had no effect on their ability to engulfanti-Fas–stimulated apoptotic cells (Figure 4A), suggestingthat IgG3–Fc receptor interactions did not contribute tophagocytosis of the apoptotic cells. Similar results wereobtained when apoptosis was induced by recombinant hu-man FasL (Figure 4B), excluding the involvement of Fcreceptors and effects dependent on antibody opsonization.

To exclude the possibility that murine TNF transferredfrom the necrotic cells affected the phagocytic activity of themacrophages, we tested uptake of necrotic cells induced todie by other stimuli. In L929sAFas cells, Fas-activation in thepresence of the pan caspase inhibitor benzyloxycarbonyl-Val-Ala-Asp(OMe)-fluoromethylketone (zVAD-fmk) resultsin necrosis (Vercammen et al., 1998). Comparison of theengagement of macrophages in phagocytosis of L929sAFas

Figure 4. Phagocytosis of L929sAhFas andL929sAhR55 cells induced to die by differentstimuli, via Fas or TNF receptor pathway.CTGr-labeled target cells were coincubatedwith CTOr-labeled Mf4/4 cells for 30 minbefore analysis. Percentage of double-positiveMf4/4 cells (bars) was determined for uptakeof apoptotic and necrotic targets. In parallel,nonlabeled target cells were double stainedwith annexin V-FITC and PI for assessment ofPS exposure (triangles) and membrane per-meability (circles), respectively. (A) Anti-Fas–treated L929sAhFas were coincubated withuntreated Mf4/4 cells (left), or with Mf4/4cells pretreated with an anti-Fc receptor neu-tralizing antibody (right). (B) L929sAhFascells were untreated (left) or pretreated with25 �M zVAD-fmk for 2 h (right) and thenstimulated with Fas ligand for the indicatedtime before coincubation with Mf4/4 cells.Negative controls of L929sAhFas cells weretreated with zVAD-fmk alone (bottom). (C)Necrotic cells were prepared fromL929sAhFas cells (left) or L929sAhR55 cells(right) treated with mTNF and R32WS68T,respectively.


cells treated with FasL in the presence or absence of zVAD-fmk confirmed that phagocytosis of necrotic target cells isless efficient than phagocytosis of apoptotic cells (Figure 4B).Similar results were obtained with necrotic L929sAFas cellsinduced to die by wild-type human TNF, and necroticL929sAhR55 cells expressing the human TNF-receptor 1 (R1)that were treated with R32SW68T, a human TNF mutantthat solely activates human TNF-R1 (Loetscher et al., 1993)(Figure 4C). Moreover, in 16 independent experiments theefficiency of apoptotic cells in inducing phagocytosis de-creased when the target cells lost their membrane integrityand became PI positive, a phenomenon designated second-ary necrosis (Figure 2A). Together, these results excludeinterference by ligand transfer and suggest that irrespectiveof the stimulus, the efficiency of uptake of necrotic cells wasconsistently lower than that of apoptotic cells. The onset ofuptake under all of these conditions coincided with theexposure of PS. Similar results were obtained using twoother murine macrophage cell lines (J774 and P388D1), ex-tending the observation that macrophages internalize cellu-lar particles of necrotic cells.

Next, we tested the effect of coincubation time on theefficiency of phagocytosis. Increasing the coincubation timefrom 30 to 90 min improved the uptake of early apoptoticcells without significantly increasing the number of Mf4/4cells involved in phagocytosis, probably because most ofthem were already phagocytic after a short time of incuba-tion (Figure 5A). In contrast, prolonging the coincubationtime of necrotic cells with macrophages from 30 to 120 minincreased both the number of Mf4/4 cells involved inphagocytosis and the amount of material cleared, indicatingthat uptake of necrotic particles is more difficult and requiresmore time (Figure 5B).

Next, we tested whether apoptotic and necrotic cells cancompete with each other for uptake. We performed a phago-cytosis assay using fixed amounts of labeled apoptotic ornecrotic cells and increasing numbers of unlabeled apoptoticor necrotic competitors (Figure 6). Unlabeled apoptotic cellswere able to compete with labeled apoptotic and necroticcells in a dose-dependent manner (Figure 6). The effect wasapparent both on the number of macrophages engaged inphagocytosis of labeled cells (Figure 6A) and on the amountof labeled material that was engulfed (Figure 6B). Con-versely, unlabeled necrotic cells were poor competitors withboth types of labeled target cells, decreasing the proportionof the macrophage population engaged in the phagocytosisof labeled cells only slightly (Figure 6A), and hardly affect-ing the amount of labeled material that was actually re-moved (Figure 6B). These results suggest that macrophageshave a preference for clearing apoptotic compared withnecrotic cells.

The Mechanisms of Phagocytosis of Apoptotic andNecrotic Cells Are Morphologically DifferentOur previous studies (Vercammen et al., 1998; Kalai et al.,2002; Vanden Berghe et al., 2003) and confocal microscopyanalysis (Figure 1) showed that apoptotic L929 cells breakdown into many small apoptotic bodies, suggesting that theapoptotic process increases the number of easy to engulftargets. In contrast, necrotic cells swell and increase in sizeand may become harder to engulf. We used electron micros-copy to examine and compare the subcellular structuresinvolved in the phagocytic process of apoptotic and necroticcells by macrophages (Figure 7). Although macrophageswere previously reported to engulf small-sized, dying cellssuch as neutrophils or T cells as single entities (Fadok et al.,1998; Hirt et al., 2000; Cocco and Ucker, 2001), we did not

observe uptake of whole necrotic and only rarely of wholeapoptotic cells. This might be due to the fact that the targetcells and Mf4/4 cells were of similar size (Figure 7, A and B)or due to the larger cytoplasm/nucleus size ratio of thetarget cells that led to more apoptotic body formation. Ap-optotic cells showed a compacted nucleus, with dense chro-matin distributed along the nuclear envelope, and con-densed cytoplasm typically broken up into apoptotic bodies(Figure 7, C, E, and G). In contrast, necrotic cells contained amottled-looking nucleus and very translucent cytoplasm,surrounded by an ill defined, damaged cell membrane (Fig-ure 7, D, F, and H). Mf4/4 cells engaged in phagocytosisstretched out by moving most of their vacuole-containingcytoplasm toward the dying cells, in contrast to restingmacrophages that showed a more symmetrical morphology.In dealing with apoptotic cells, macrophages formed thinprotrusions that easily engulfed the well-defined, mem-brane-enclosed apoptotic bodies (Figure 7, C and E). Inter-nalized apoptotic material showed up as dark, condensedareas within the lighter staining macrophage cytoplasm

Figure 5. Time kinetics of uptake of apoptotic and necrotic targetcells. (A) Apoptotic target cells. L929hFas cells were stimulated withFas ligand for 1 h and then coincubated with Mf4/4 cells for 30 or90 min. (B) Necrotic target cells. L929hR55 cells were stimulatedwith R32WS68T (100 ng/ml) and coincubated with Mf4/4 cells forthe indicated time periods and then phagocytosis was determinedby flow cytometry. PI positivity and annexin V positivity weredetermined on the target cells just before coincubation. Error barsrepresent the SD of at least three independent experiments.


(Figure 7G). Phagocytes dealing with necrotic cells stretchedout by sending thick protrusions into the swollen “ghost”-like structures of the target cells, grasping only small vol-umes of the cellular debris (Figure 7, D and F). On internal-ization, this necrotic material showed up as a lighter andprobably less condensed material compared with the mac-rophage cytoplasm (Figure 7H). These results suggest thatphagocytosis of necrotic and apoptotic cells occur by differ-ent mechanisms and that the former process may be moredifficult.

Phagocytosis of Necrotic Cells by Macrophages Occurs InVitro and In Vivo and Does Not Stimulate Macrophagesto Express Proinflammatory CytokinesOverexpression of FADD in L929 cells leads to necrosis(Boone et al., 2000). To study phagocytosis of necrotic cells invivo, while excluding effects due to transfer of the death-inducing stimulus, we used stably transfected L929rTA forinducible expression of a FKBP-FADD fusion protein(L929rTAFADD). Treatment of these cells with doxycyclinfollowed by the FKBP dimerizer AP1510 induces the expres-sion and forced oligomerization of FADD and results in atypical necrotic cell death (Figure 8A). We compared the invitro with the in vivo phagocytosis of necroticL929rTAFADD cells by primary macrophages. For that pur-

pose, we tested the uptake of the cells by Mf4/4 cells (Figure8A). In parallel, we injected CTGr-labeled viable L929rTA ornecrotic L929rTAFADD into the peritonea of mice, recov-ered the peritoneal cells after 4 h, stained the macrophagepopulation with phycoerythrin-conjugated macrophage-specific antibody (McKnight et al., 1996), and determinedtheir phagocytic efficiencies (Figure 8B). Phagocytosis ofL929rTAFADD cells treated with doxycyclin and AP1510 byMf4/4 cells was as efficient as that of R32WS86T-treatedL929hR55 cells (Figure 8A). Similar results were obtained invivo (Figure 8B), indicating that phagocytosis of necroticcells by macrophages occurs to the same extent both in vivoand in vitro. Interestingly, analysis of the peritoneal cells byMay Grunwald/Giemsa staining and microscopy demon-strated that the cell-type composition was not significantlyaffected by exposure to the control or necrotic cells (Figure8C). These results suggest that exposure to necrotic cells for4 h did not induce a strong proinflammatory response.Fadok and coworkers (Fadok et al., 1998, 2001a) suggestedthat in contrast to apoptotic cells, necrotic cells induce proin-flammatory cytokine production in macrophages. Therefore,we examined the effects of apoptotic and necrotic cells on thetranscriptional expression profile of several inflammatorycytokines in Mf4/4 macrophages by using an RNase protec-tion assay. To assess the expression of cytokines by phago-

Figure 6. Apoptotic cells are better compet-itors than necrotic cells for uptake of apopto-tic or necrotic cells by Mf4/4 cells. CellTracker Green-labeled early apoptoticL929hFas cells treated for 1 h with Fas ligand,and necrotic L929hR55 cells treated for 14 hwith R32WS86T, were incubated with CellTracker Orange-labeled Mf4/4 cells in thepresence or absence of unlabeled competitortarget cells. Labeled targets (2.5 � 105 cells/well) were mixed with unlabeled apoptotic ornecrotic competitor cells at the indicated ra-tios and added to the macrophages. The ex-tent of phagocytosis was determined after 30min. (A) Percentage of double-positive Mf4/4cells (% phagocytic). (B) The increase in themean green fluorescence of the macrophagepopulation (� mean Gr FL).


cytic macrophages, total RNA was extracted from Mf4/4cells 4 h after their exposure for 1 h to control, apoptotic, ornecrotic target cells. Mf4/4 macrophages alone expressedlow levels of mRNAs for TNF- and transforming growthfactor (TGF)�1, and only traces of mRNAs of IL-6 and in-terferon-� (IFN�) (Figure 9). No changes were observed inthe expression of the mRNAs of the tested cytokines bymf4/4 cells, after coincubation of these macrophages withcontrol, apoptotic, or necrotic target cells. Target cells ontheir own, with or without the apoptotic or necrotic treat-ments, expressed the mRNAs of TGF�1, TNF-, IL-6, orIFN�. To assess the possibility that apoptotic or necrotic cellscan modulate the levels of mRNA of preexisting proinflam-matory cytokines in activated macrophages, we pretreatedmacrophages with LPS for 2 h before exposure to apoptoticor necrotic target cells. Total RNA was prepared 5 h afteraddition of the target cells and analyzed for the presence ofcytokine mRNAs. On LPS-stimulation, Mf4/4 macrophagesshowed a marked increase in the expression of the mRNAsof TNF-, IL-6, and IFN�, indicating the reliability of thisanalytical method, and demonstrating that the macrophageswere capable of producing these cytokines (Figure 9). More-over, the levels of mRNAs for TGF�1, TNF-, IL-6, and IFN�expressed by LPS-stimulated macrophages were not affected

by uptake of either apoptotic or necrotic cells. Furthermore,no modulation of TNF- or IL-6 activity was detected insupernatants of Mf4/4 cells, whether or not they were ex-posed to apoptotic or necrotic target cells. LPS stimulationstrongly induced the expression of these cytokines (Figure10). These results indicate that phagocytosis of apoptotic andnecrotic L929 cells apparently does not modulate the expres-sion pattern of proinflammatory cytokines by macrophagesat the mRNA and the protein levels. Interestingly, pretreat-ment of macrophages with LPS for 2 h before exposure totarget cells resulted in an increase in the percentage of themacrophage population involved in phagocytosis from 71 to92% for apoptotic cells, and from 47 to 63% for necrotic cells,suggesting that a proinflammatory stimulus may prime themacrophages for phagocytosis of dying cells.

DISCUSSION

Although the molecular mechanisms involved in phagocy-tosis of apoptotic cells are becoming better understood, littleis known about the recognition and uptake of necroticallydying cells. Here, we used differentially labeled macro-phages and target cells to study the uptake of apoptotic and

Figure 7. Electron microscopy micrographsof phagocytic uptake of apoptotic and ne-crotic L929sAhFas cell particles by Mf4/4cells. (A) A control Mf4/4 macrophage-likecell with abundant cytoplasm containingwell-developed lysosomal apparatus. (B) Acontrol unstimulated L929sAhFas cell withwell preserved cytoplasmic organelles,smoothly outlined nuclei containing hetero-chromatin, and numerous microvilli protrud-ing from the entire surface. (C, E, and G)Uptake of early apoptotic L929sAhFas cells byMf4/4 cells after 30 min of coculture. Targetcells were treated with agonistic anti-Fas for1 h before coculture. (D, F, and H) Uptake ofnecrotic L929sAhFas cells by Mf4/4 cells after90 min of coculture. Target cells treated withTNF for 18 h before coculture. (C) Engulfmentof apoptotic bodies by an Mf4/4 cell. (E) En-largement of C, showing the protrusions ofthe phagocyte (arrowheads) toward the apo-ptotic bodies. (D) Uptake of necrotic materialby an Mf4/4 cell (arrowheads). (F) Enlarge-ment of D, showing a labyrinth of surfaceprotrusions of the phagocyte surroundingdisintegrated cytoplasm of a necrotic targetcell. (G) An Mf4/4 cell with engulfed apopto-tic bodies of different sizes (arrowheads), sur-rounded by two apoptotic cells (asterisks)with condensed chromatin at the margins ofthe nuclei (bottom). (H) An Mf4/4 cell con-taining engulfed necrotic particles (arrow-heads) displaying surface protrusions (ar-rows) toward the cytoplasm of adisintegrating necrotic cell. Bars (A, B, C, D,G, and H), 1 �m; (E and F), 2 �m.


necrotic cells in a quantitative flow cytometry phagocytosisassay. The L929 fibrosarcoma cellular system provided uswith well-characterized models of apoptosis and necrosismediated by a death domain receptor (Denecker et al.,2001b), distinct from cell death caused by physical damagesuch as rupture of the cell membrane by douncing, freezethawing, or exposure to high temperature (Sauter et al., 2000;Cocco and Ucker, 2001; Fadok et al., 2001a).

Previous studies showed that PS is a prerequisite for theuptake of apoptotic cells by macrophages. PS exposed on theapoptotic cell membrane is recognized directly by the mac-

rophage PS receptor (Fadok et al., 2000) or indirectly viabinding of PS to a soluble intermediate (milk fat globule-epidermal growth factor-factor 8), which is secreted by mac-rophages, followed by recognition of the complex by thev�3 integrin on the phagocyte’s surface (Hanayama et al.,2002). Our results demonstrate that both apoptotic and ne-crotic target cells are recognized and phagocytosed by mac-rophages, whereas viable control cells are not. Uptake ofapoptotic and necrotic cells occurred irrespective of the stim-ulus inducing cell death. Phagocytosis of necrotic cells is afeature common to different macrophage cell lines and alsooccurs in vivo, as demonstrated in Figure 8. Apoptotic cellsare taken up very efficiently as soon as PS exposure becomesevident, at a time when their membrane is still intact. Up-take of necrotic cells takes place when cells loose the integ-rity of the membrane and become PS positive. Moreover,recombinant annexin V can inhibit phagocytosis of bothtypes of dying cells, suggesting that externalization of PS isa general signal indicating to phagocytes that a dying cellshould be cleared, irrespective of the way the cell has died.Indeed, PS exposure is an old “eat me signal” also used forthe recognition and uptake of aged erythrocytes (Schlegeland Williamson, 2001). Although the recognition of dyingL929sA cells seems to be dependent on PS for both apoptoticand necrotic cells, the efficiency of uptake is higher for theformer. Moreover, when given the choice, macrophages pre-fer to clear early apoptotic cells, as demonstrated by thecompetition experiment (Figure 6), even though these targetcells present a lower mean annexin V staining level. Thelower efficiency of uptake of necrotic cells might be due tophysical constraints. It is possible that the PS exposed onnecrotic and late apoptotic cells is less accessible to themacrophages. Moreover, although apoptotic cells fragmentinto many small-contained particles that are easy to recog-nize and engulf, necrotic cells swell and remain as largesingle entities for a long time. Similar observations weremade in Caenorhabditis elegans, where necrotic corpses lingerfor much longer than apoptotic ones, probably due to theirlarger volume, although uptake of both types of dying cellsrequires the same engulfment genes (Chung et al., 2000).Indeed, electron microscopy analysis presented here sug-gests that the mechanism used by macrophages to engulfnecrotic cells is morphologically different from that used forapoptotic cells. Macrophages that engulf necrotic cells pro-trude into the swollen ghost-like structures of the dyingcells, grasping only small volumes of the cellular debris,whereas apoptotic cells are readily engulfed as containeddistinct apoptotic bodies (Krysko et al., 2003).

Phagocytosis of apoptotic cells does not induce inflamma-tion and is often referred to as a silent event (Fadok et al.,1998; Cocco and Ucker, 2001; Fadok et al., 2001a). A possibleconsequence of the greater difficulty apparently encounteredby macrophages in clearing necrotic cells is spillage of thecellular contents of the dying cells, leading to more histo-toxicity and development of proinflammatory (Haslett, 1992;Haslett et al., 1994; Wiegand et al., 2001; Medan et al., 2002) orautoimmune responses (Rovere et al., 2000; Chan et al., 2001;Magnus et al., 2001). Indeed, apoptotic neutrophils and PS-exposing membranes were reported to elicit an antiinflam-matory effect, whereas incubation of macrophages withphysically lysed neutrophils, but not with lysed lympho-cytes, significantly stimulated the production of macro-phage-inflammatory protein 2, IL-8, TNF-a, and IL-10(Fadok et al., 2001a). However, the latter effect was attributedto the cleavage of PS-receptor on the surface of the macro-phages by elastase, a protease released upon the lysis ofneutrophils. The cleavage of PS-receptor probably also im-

Figure 8. Phagocytosis of necrotic cells by peritoneal macrophagesin vivo. (A) CTOr-labeled Mf4/4 cells were coincubated with CTGr-labeled untreated or necrotic L929hR55 cells, or L929rTAFADDcells. Percentage of double-positive Mf4/4 cells (bars) was deter-mined 4 h later. Necrosis in L929hR55 and L929rTAFADD cells wasinduced by R32WS86T and by doxycyclin � AP1510, respectively.PI (circles) and PS (triangles) positivity were determined on thetarget cells just before coincubation. (B) In parallel, CTGr-labeledviable L929rTA, or necrotic L929rTAFADD cells treated with doxy-cyclin � AP1510, were injected into the peritoneum of mice. Peri-toneal cells were recovered 4 h later, and macrophages were stainedwith a phycoerythrin-conjugated F4/80 antibody; phagocytosis wasdetermined by flow cytometry. (C) The peritoneal cell-type compo-sition was determined microscopically following May-Grunwald/Giemsa (Sigma-Aldrich) staining.


pairs the phagocytic capacity of the macrophages and maycause them some stress (Vandivier et al., 2002). Our resultsshow that phagocytosis of apoptotic or necrotic cells by

Mf4/4 macrophages does not induce the expression of IFN�,TGF�, TNF, or IL-6, neither at the mRNA nor at the proteinlevel. On the other hand, exposure of the same macrophages

Figure 9. Analysis of the effects of uptake ofapoptotic or necrotic cells on the expression ofinflammatory cytokine mRNA by the phago-cytes. Cytokine mRNA levels were measuredby RNase protection assay by using a Ribo-quant multiprobe set (BD PharMingen).Mf4/4 cells were left unstimulated or werestimulated with LPS for 2 h and then eithercoincubated or not with untreated or FasL-treated apoptotic L929hFas cells, or with un-treated or R32WS86T-treated necrotic L929hR55cells. RNA was isolated 5 h later. Controls ofuntreated and FasL- or R32WS86T-treated tar-get and Mf4/4 cells were also taken. A mousepool RNA included in the kit as a positive con-trol, and the undigested 32P-labeled RNAprobe, are shown on the right. Relative quanti-fication of intensities is expressed as percentageof L32 expression.

Figure 10. Analysis of the effect of phagocy-tosis of apoptotic or necrotic cells on the ex-pression of biologically active IL-6 and TNF.Mf4/4 cells were coincubated for the indi-cated time with untreated or FasL-treated ap-optotic L929hFas cells, or with untreated orR32WS86T-treated necrotic L929hR55 cells.Incubation medium was collected 24 h laterand the concentrations of biologically activeIL-6 and TNF were determined. Positive con-trols of Mf4/4 cells treated for 2 and 24 h withLPS and negative controls of untreated andFasL- or R32WS86T-treated target, and Mf4/4cells are also presented. Measurements arepresented in logarithmic scale. Error bars rep-resent the SD of at least three independentexperiments.


to LPS strongly increased the expression of TNF and IL-6and moderately increased the levels of IFN� mRNA. More-over, phagocytosis of apoptotic or necrotic L929 cells did notaffect this LPS-induced response. Several reports are inagreement with these findings. Cocco and Ucker (2001) ob-served that heat-killed necrotic and late apoptotic thymomaand T-cell hybridoma target cells did not induce J774A.1 andRAW 264.7 macrophages to secrete TNF or IL-6. Further-more, no signs of inflammation accompanied the necroticinterdigital cell death occurring in Apaf-1–deficient miceand further development proceeded normally (Chautan etal., 1999). Although inflammation often accompanies necro-sis, the above-mentioned observations and our results indi-cate that this is not due to the induction of expression ofproinflammatory cytokines in the macrophages clearing thedying cells. Therefore, it is conceivable that the release ofcytokines or other factors from the necrotic cells themselvesmay be crucial for an inflammatory response. Additionally,our results demonstrate that the clearance of primary andprobably also secondary necrotic cells is clearly less efficientand more difficult and time consuming than that of apopto-tic cells. This process may cause the macrophages to remainat the same site longer, thus heightening the inflammatorystate. These results suggest that prevention of necrosis andsecondary necrosis, and promotion of apoptosis may allow amore rapid and efficient clearance of the dying cells anddecrease the damage to the surrounding tissue both in injuryand in antitumor cancer treatment.

ACKNOWLEDGMENTS

We thank Ann Meeus, Wilma Burm, Dominique Jacobus, Hubert Stevens, andWim Van Molle for excellent technical assistance and Alex Raeymaekers forpreparation of TNF. J774 cells were kindly provided by Dr. E. Smits (DevGenN.V.). This work was supported by the Interuniversitaire Attractiepolen(IUAP-V), the Fonds voor Wetenschappelijk Onderzoek Vlaanderen (grants31.5189.00 and 3G.0006.01), the EC-RTD (grant QLRT-CT-1999-00739), theRUG-cofinanciering EU project (011C0300), and GOA project (12050502). G.B.is a doctoral fellow with the Fond for Scientific Research-Flanders (FWO).D.V.K. is a doctoral fellow with the GOA project (120505502).

REFERENCES

Amara, J.F., Clackson, T., Rivera, V.M., Guo, T., Keenan, T., Natesan, S.,Pollock, R., Yang, W., Courage, N.L., Holt, D.A., and Gilman, M. (1997). Aversatile synthetic dimerizer for the regulation of protein-protein interactions.Proc. Natl. Acad. Sci. USA 94, 10618–10623.

Beilharz, E.J., Williams, C.E., Dragunow, M., Sirimanne, E.S., and Gluckman,P.D. (1995). Mechanisms of delayed cell death following hypoxic-ischemicinjury in the immature rat: evidence for apoptosis during selective neuronalloss. Brain Res. Mol. Brain Res. 29, 1–14.

Boone, E., Vanden Berghe, T., Van Loo, G., De Wilde, G., De Wael, N.,Vercammen, D., Fiers, W., Haegeman, G., and Vandenabeele, P. (2000). Struc-ture/function analysis of p55 tumor necrosis factor receptor and Fas-associ-ated death domain. Effect on necrosis in L929sA cells. J. Biol. Chem. 275,37596–37603.

Chan, A., Magnus, T., and Gold, R. (2001). Phagocytosis of apoptotic inflam-matory cells by microglia and modulation by different cytokines: mechanismfor removal of apoptotic cells in the inflamed nervous system. Glia 33, 87–95.

Chautan, M., Chazal, G., Cecconi, F., Gruss, P., and Golstein, P. (1999).Interdigital cell death can occur through a necrotic and caspase-independentpathway. Curr. Biol. 9, 967–970.

Chung, S., Gumienny, T.L., Hengartner, M.O., and Driscoll, M. (2000). Acommon set of engulfment genes mediates removal of both apoptotic andnecrotic cell corpses in C. elegans. Nat. Cell. Biol. 2, 931–937.

Cocco, R.E., and Ucker, D.S. (2001). Distinct modes of macrophage recognitionfor apoptotic and necrotic cells are not specified exclusively by phosphatidyl-serine exposure. Mol. Biol. Cell 12, 919–930.

Denecker, G., Dooms, H., Van Loo, G., Vercammen, D., Grooten, J., Fiers, W.,Declercq, W., and Vandenabeele, P. (2000). Phosphatidyl serine exposure

during apoptosis precedes release of cytochrome c and decrease in mitochon-drial transmembrane potential. FEBS Lett. 465, 47–52.

Denecker, G., Vercammen, D., Declercq, W., and Vandenabeele, P. (2001a).Apoptotic and necrotic cell death induced by death domain receptors. CellMol. Life Sci. 58, 356–370.

Denecker, G., et al. (2001b). Death receptor-induced apoptotic and necrotic celldeath: differential role of caspases and mitochondria. Cell Death Differ. 8,829–840.

Desmedt, M., Rottiers, P., Dooms, H., Fiers, W., and Grooten, J. (1998).Macrophages induce cellular immunity by activating Th1 cell responses andsuppressing Th2 cell responses. J. Immunol. 160, 5300–5308.

Earnshaw, W.C., Martins, L.M., and Kaufmann, S.H. (1999). Mammaliancaspases: structure, activation, substrates, and functions during apoptosis.Annu. Rev. Biochem. 68, 383–424.

Fadok, V.A., Bratton, D.L., Guthrie, L., and Henson, P.M. (2001a). Differentialeffects of apoptotic versus lysed cells on macrophage production of cytokines:role of proteases. J. Immunol. 166, 6847–6854.

Fadok, V.A., Bratton, D.L., and Henson, P.M. (2001b). Phagocyte receptors forapoptotic cells: recognition, uptake, and consequences. J. Clin. Investig. 108,957–962.

Fadok, V.A., Bratton, D.L., Konowal, A., Freed, P.W., Westcott, J.Y., andHenson, P.M. (1998). Macrophages that have ingested apoptotic cells in vitroinhibit proinflammatory cytokine production through autocrine/paracrinemechanisms involving TGF-beta, PGE2, and PAF. J. Clin. Investig. 101, 890–898.

Fadok, V.A., Bratton, D.L., Rose, D.M., Pearson, A., Ezekewitz, R.A., andHenson, P.M. (2000). A receptor for phosphatidylserine-specific clearance ofapoptotic cells. Nature 405, 85–90.

Francois, M., Le Cabec, V., Dupont, M.A., Sansonetti, P.J., and Maridonneau-Parini, I. (2000). Induction of necrosis in human neutrophils by Shigellaflexneri requires type III secretion, IpaB and IpaC invasins, and actin poly-merization. Infect. Immun. 68, 1289–1296.

Freundlieb, S., Schirra-Muller, C., and Bujard, H. (1999). A tetracycline con-trolled activation/repression system with increased potential for gene trans-fer into mammalian cells. J. Gene Med. 1, 4–12.

Hanayama, R., Tanaka, M., Miwa, K., Shinohara, A., Iwamatsu, A., andNagata, S. (2002). Identification of a factor that links apoptotic cells to phago-cytes. Nature 417, 182–187.

Haslett, C. (1992). Resolution of acute inflammation and the role of apoptosisin the tissue fate of granulocytes. Clin. Sci. 83, 639–648.

Haslett, C., Savill, J.S., Whyte, M.K., Stern, M., Dransfield, I., and Meagher,L.C. (1994). Granulocyte apoptosis and the control of inflammation. Phil.Trans. R. Soc. Lond. B. Biol. Sci. 345, 327–333.

Henson, P.M., Bratton, D.L., and Fadok, V.A. (2001). Apoptotic cell removal.Curr. Biol. 11, R795–R805.

Hirt, U.A., Gantner, F., and Leist, M. (2000). Phagocytosis of nonapoptoticcells dying by caspase-independent mechanisms. J. Immunol. 164, 6520–6529.

Holler, N., Zaru, R., Micheau, O., Thome, M., Attinger, A., Valitutti, S.,Bodmer, J.L., Schneider, P., Seed, B., and Tschopp, J. (2000). Fas triggers analternative, caspase-8-independent cell death pathway using the kinase RIP aseffector molecule. Nat. Immunol. 1, 489–495.

Kalai, M., Van Loo, G., Vanden Berghe, T., Meeus, A., Burm, W., Saelens, X.,and Vandenabeele, P. (2002). Tipping the balance between necrosis andapoptosis in human and murine cells treated with interferon and dsRNA. CellDeath Differ. 9, 981–994.

Kitanaka, C., and Kuchino, Y. (1999). Caspase-independent programmed celldeath with necrotic morphology. Cell Death Differ. 6, 508–515.

Krysko, D.V., Brouckaert, G., Kalai, M., Vandenabeele, P., and D’Herde, K.(2003). Mechanisms of internalization of apoptotic and necrotic L929 cells bya macrophage cell line studied by electron microscopy. J. Morphol. 258,336–345.

Lamkanfi, M., Declercq, W., Depuydt, B., Kalai, M., Saelens, X., and Vandena-beele, P. (2002a). The caspase family. In: Caspases: Their Role in Cell Deathand Cell Survival, eds. M. Los and H. Walczak, Georgetown, TX: LandesBioscience, Kluwer Academic Press, 1–40.

Lamkanfi, M., Declercq, W., Kalai, M., Saelens, X., and Vandenabeele, P.(2002b). Alice in caspase land. A phylogenetic analysis of caspases from wormto man. Cell Death Differ. 9, 358–361.

Loetscher, H., Stueber, D., Banner, D., Mackay, F., and Lesslauer, W. (1993).Human tumor necrosis factor alpha (TNF alpha) mutants with exclusivespecificity for the 55-kDa or 75-kDa TNF receptors. J. Biol. Chem. 268, 26350–26357.


Magnus, T., Chan, A., Grauer, O., Toyka, K.V., and Gold, R. (2001). Microglialphagocytosis of apoptotic inflammatory T cells leads to down-regulation ofmicroglial immune activation. J. Immunol. 167, 5004–5010.

Matsumura, H., Shimizu, Y., Ohsawa, Y., Kawahara, A., Uchiyama, Y., andNagata, S. (2000). Necrotic death pathway in Fas receptor signaling. J. CellBiol. 151, 1247–1256.

McIlroy, D., Tanaka, M., Sakahira, H., Fukuyama, H., Suzuki, M., Yamamura,K., Ohsawa, Y., Uchiyama, Y., and Nagata, S. (2000). An auxiliary mode ofapoptotic DNA fragmentation provided by phagocytes. Genes Dev. 14, 549–558.

McKnight, A.J., Macfarlane, A.J., Dri, P., Turley, L., Willis, A.C., and Gordon,S. (1996). Molecular cloning of F4/80, a murine macrophage-restricted cellsurface glycoprotein with homology to the G-protein-linked transmembrane7 hormone receptor family. J. Biol. Chem. 271, 486–489.

Medan, D., Wang, L., Yang, X., Dokka, S., Castranova, V., and Rojanasakul, Y.(2002). Induction of neutrophil apoptosis and secondary necrosis duringendotoxin-induced pulmonary inflammation in mice. J. Cell. Physiol. 191,320–326.

Poupart, P., Vandenabeele, P., Cayphas, S., Van Snick, J., Haegeman, G.,Kruys, V., Fiers, W., and Content, J. (1987). B cell growth modulating anddifferentiating activity of recombinant human 26-kd protein (BSF-2, HuIFN-beta 2, HPGF). EMBO J. 6, 1219–1224.

Ren, Y., and Savill, J. (1998). Apoptosis: the importance of being eaten. CellDeath Differ. 5, 563–568.

Rovere, P., et al. (2000). The long pentraxin P.T.X3 binds to apoptotic cells andregulates their clearance by antigen-presenting dendritic cells. Blood 96,4300–4306.

Sauter, B., Albert, M.L., Francisco, L., Larsson, M., Somersan, S., and Bhard-waj, N. (2000). Consequences of cell death: exposure to necrotic tumor cells,but not primary tissue cells or apoptotic cells, induces the maturation ofimmunostimulatory dendritic cells. J. Exp. Med. 191, 423–434.

Schlegel, R.A., and Williamson, P. (2001). Phosphatidylserine, a death knell.Cell Death Differ. 8, 551–563.

Schneider, P., Holler, N., Bodmer, J.L., Hahne, M., Frei, K., Fontana, A., andTschopp, J. (1998). Conversion of membrane-bound Fas(CD95) ligand to itssoluble form is associated with downregulation of its proapoptotic activityand loss of liver toxicity. J. Exp. Med. 187, 1205–1213.

Scott, R.S., McMahon, E.J., Pop, S.M., Reap, E.A., Caricchio, R., Cohen, P.L.,Earp, H.S., and Matsushima, G.K. (2001). Phagocytosis and clearance ofapoptotic cells is mediated by MER. Nature 411, 207–211.

Tavernier, J., Franssen, L., Maremout, A., Van der Heyden, J., Mueller, J.,Ruysschaert, M. R., Van Vliet, A., Bauden, R., and Fiers, W. (1987). Isolationand expression of the genes coding for mouse and human tumor necrosis

factor (TNF) and biological properties of recombinant TNF. In: MolecularCloning and Analysis of Lymphokines. Vol. 13, Lymphokines, eds. D.R.Webb, and D.V. Goeddel. Orlando, FL; Academic Press, 181–198.

Van Ostade, X., Vandenabeele, P., Everaerdt, B., Loetscher, H., Gentz, R.,Brockhaus, M., Lesslauer, W., Tavernier, J., Brouckaert, P., and Fiers, W.(1993). Human TNF mutants with selective activity on the p55 receptor.Nature 361, 266–269.

Vanden Berghe, T., Kalai, M., van Loo, G., Declercq, W., and Vandenabeele, P.(2003). Disruption of HSP90 function reverts tumor necrosis factor-inducednecrosis to apoptosis. J. Biol. Chem. 278, 5622–5629.

Vandivier, R.W., Fadok, V.A., Hoffmann, P.R., Bratton, D.L., Penvari, C.,Brown, K.K., Brain, J.D., Accurso, F.J., and Henson, P.M. (2002). Elastase-mediated phosphatidylserine receptor cleavage impairs apoptotic cell clear-ance in cystic fibrosis and bronchiectasis. J. Clin. Investig. 109, 661–670.

Vanhaesebroeck, B., Van Bladel, S., Lenaerts, A., Suffys, P., Beyaert, R., Lucas,R., Van Roy, F., and Fiers, W. (1991). Two discrete types of Tumor NecrosisFactor-resistant cells derived from the same cell line. Cancer Res. 51, 2469–2477.

Vanhoenacker, P., Gommeren, W., Luyten, W.H.M.L., Leysen, J.E., and Hae-geman, G. (1999). Optimized expression of serotonin receptors in mammaliancells using inducible expression systems. Gene Ther. Mol. Biol. 3, 301–310.

Vercammen, D., Brouckaert, G., Denecker, G., Van de Craen, M., Declercq, W.,Fiers, W., and Vandenabeele, P. (1998). Dual signaling of the Fas receptor:initiation of both apoptotic and necrotic cell death pathways. J. Exp. Med. 188,919–930.

Vercammen, D., Vandenabeele, P., Beyaert, R., Declercq, W., and Fiers, W.(1997). Tumour necrosis factor-induced necrosis versus anti-Fas-induced ap-optosis in L929 cells. Cytokine 9, 801–808.

Vercammen, D., Vandenabeele, P., Declercq, W., Van de Craen, M., Grooten,J., and Fiers, W. (1995). Cytotoxicity in L929 murine fibrosarcoma cells aftertriggering of transfected human p75 tumour necrosis factor (TNF) receptor ismediated by endogenous murine TNF. Cytokine 7, 463–470.

Verhoven, B., Krahling, S., Schlegel, R.A., and Williamson, P. (1999). Regula-tion of phosphatidylserine exposure and phagocytosis of apoptotic T lym-phocytes. Cell Death Differ. 6, 262–270.

Wang, J.H., Redmond, H.P., Watson, R.W., Duggan, S., McCarthy, J., Barry,M., and Bouchier-Hayes, D. (1996). Mechanisms involved in the induction ofhuman endothelial cell necrosis. Cell. Immunol. 168, 91–99.

Wiegand, U.K., Corbach, S., Prescott, A.R., Savill, J., and Spruce, B.A. (2001).The trigger to cell death determines the efficiency with which dying cells arecleared by neighbours. Cell Death Differ. 8, 734–746.


3.2 Analysis of NF- B activation in macrophages upon their co-culture with apoptotic, primary and secondary necrotic cells

99

ANALYSIS OF NF- B ACTIVATION IN MACROPHAGES UPON THEIR CO-CULTURE

WITH APOPTOTIC, PRIMARY AND SECONDARY NECROTIC CELLS

Introduction

In vertebrates, the NF- B family of transcription factors comprises five members, called p50, p52,

p65 (RelA), c-Rel, and RelB. These subunits can homodimerize and heterodimerize in various

combinations. Most commonly, NF- B consists of two polypeptides of 50 kDa (p50) and 65 kDa

(p65) (Zingarelli, 2005). In resting cells, NF- B resides in the cytoplasm in an inactive form

physically associated with inhibitor proteins known as inhibitor B proteins. Activation of NF-

B requires phosphorylation of its physiologic inhibitors (i.e. Bs) at specific serine residues. This

phosphorylation event is mediated by a B kinase complex and triggers the subsequent poly-

ubiquitination and subsequent degradation by the 26S proteasome The proteolytic degradation of

B proteins liberates NF- B and allows its translocation into the nucleus, where it regulates the

expression of hundreds of genes important in immune and inflammatory responses as well as in

anti-apoptotic function (Beinke and Ley, 2004; Caamano and Hunter, 2002; Zingarelli, 2005).

Our previous results already suggested that neither the uptake of apoptotic nor the uptake of

necrotic L929 cells by macrophages modulates the expression of pro-inflammatory cytokines as

determined by RNase protection assay (Brouckaert et al., 2004). In contrast to our data, Li et al.

showed that activation of NF- B took place in mouse embryonic fibroblasts exposed to thymocytes

lysed by douncing (“necrotic” cells) with subsequent induction of genes involved in inflammation (Li

et al., 2001). It has been demonstrated that lysed neutrophils, but not lymphocytes, significantly

stimulate production of macrophage-inflammatory protein 2 (MIP-2), IL-8, TNF- and IL-10 (Fadok

et al., 2001).

Because the regulation of cytokine gene transcription signaled through TLR4 and other

inflammatory receptors is effected by critical transcription activators such as NF- B, the aim of the

present study was to investigate whether either apoptotic or primary and secondary necrotic cells

during their recognition and uptake by macrophages might modulate NF- B activation.

Materials and Methods

Expression plasmids

pWTP-NF- B-Luc is a lentiviral expression plasmid containing the luciferase (Luc) reporter gene

driven by a minimal chicken conalbumin promoter preceded by three copies of a mouse Ig oligo

(DMBR collection number 3249) (Lamkanfi et al., 2004; Van Huffel et al., 2001). In brief, the cDNA


for the NF- B-Luc fragment was generated by PCR from the pNF-conLuc plasmid (Lamkanfi et al.,

2004) by using the forward primer: 5’-CCG CTC GAG GCT CTC CCT TAT GCG ACT CCT GC-3’

and the reverse primer 5’-CGG GGT ACC TTA CAA TTT GGA CTT TCC GC-3’, introducing XhoI

and KpnI sites, respectively. This fragment was cloned into the lentiviral pWPT vector. The pWPT

lentiviral vector was a gift from Dr. D. Trono (School of life Sciences, Swiss Institute of Technology

Lausanne, Switzerland) and the pNF-conLuc was a kind gift from Dr A. Israel (Institute Pasteur,

Paris, France). The PCR product after cloning in the pWPT lentiviral vector was checked by

sequencing to ensure that no errors had been introduced by PCR.

Transfections and luciferase assays

In order to test the functionality of pWPT-NF- B-Luc, a human embryonal kidney carcinoma 293T

cell line was transfected with 100 ng of pWPT-NF- B-luciferase and pEF6-LacZ- -galactosidase

reporter plasmids using a calcium phosphate precipitation method (O'Mahoney and Adams, 1994).

Briefly, cells were seeded the day before transfection at 2 x 105 cells/well, transfected for 4 h,

washed, and incubated overnight. To screen for NF- B-mediated induction of luciferase activity,

cells were treated for 6 hours with 1000 IU/ml of hTNF- and lysates were prepared using 200 l of

ice-cold Nonidet P-40 lysis buffer (1% Nonidet P-40; 10 mM TrisHCl pH 7; 200 mM NaCl; 5 mM

EDTA; 10% glycerol supplemented with 1 mM dithiothreitol, 1 mM Na3VO4, 1 mM

phenylmethylsulfonyl fluoride). After the addition of 25 l of substrate buffer (658 M luciferin, 378

mM co-enzyme A, and 742 M ATP) to 25 l of cell lysates, luciferase activity was assayed by

luminescence in a TopCount NXT microplate scintillation reader (PerkinElmer Life and Analytical

Sciences). To normalize transfection efficiency, cell lysates were also subjected to -galactosidase

colorimetric assay. In brief, 25 l of cell lysates were incubated 5 min at room temperature with 155

l of a solution containing 0.9 mg/ml -nitrophenyl- -D-galactopyranoside, 1 mM MgCl2, 45 mM -

mercaptoethanol, and 100 mM sodium phosphate, pH 7.5. The optical density was read at a

wavelength of 595 nm. Results were expressed as relative luciferase units per second/optical

density for -galactosidase activity.

Lentiviral transduction of macrophages (Mf4/4)

Viruses were produced by transient cotransfection of 293T cells with a three-plasmid combination.

Briefly, 25 cm2 Falcon containing 1.2 x 106 cells/ml were transfected with 15 g of pMDG (envelop

plasmid), 30 g of pCMV-D8-9 (packaging plasmid) and 30 g of pWPT-NF- B-Luc by the calcium

phosphate DNA precipitation method. After 6 hours of incubation with DNA precipitate, the 293T

cells were carefully washed and incubated in fresh medium overnight. The viral supernatant was

first harvested and filtrated through a 0.45 M filter. For transduction, macrophages were

centrifuged in the presence of lentiviral supernatant for 1 hour at 1800 rpm at 32°C. The next day

NF- B 101

the transduction was repeated on the same macrophages and after 8 hours the medium was

refreshed and the macrophages were left, for safety reasons, for 10 days in P2 facilities. Thereafter

the pool of macrophages was subcloned by under-limiting dilution (0.3 cells per well in adherent

96-well plate). Clones were tested for NF- B-luciferase activity in response to LPS (1 g/ml) using

the luciferase assay described above. The best-responding clone was isolated and expanded for

the future studies.

Analysis of the effect of apoptotic, primary and secondary necrotic cells on NF- B

activation in macrophages

L929sAhFas target cells were kept in suspension by seeding 2.5 x 105 cells per well in uncoated

24-well tissue culture plates (Sarstedt). Apoptotic cell death was induced by adding anti-Fas

antibody (250 ng/ml) for 1 h (early apoptotic condition) or for 10 h (secondary necrotic condition).

Necrotic cell death was induced by adding mTNF (10.000 IU/ml) for 10 h (necrotic condition). The

cells were harvested at the times indicated and kept on ice until analysis with a FACSCalibur flow

cytometer (BD Biosciences). Loss of cell membrane integrity as a measure of cell death was

determined by propidium iodide (PI) fluorescence (ex 535/em 617). Phosphatidylserine (PS)

exposure was monitored by AnnexinV-fluorescein isothiocyanate (FITC) staining (ex 494/em 518).

AnnexinV-FITC conjugate (BD Pharmingen, San Diego, CA, USA) was used at 1 �g/ml. Propidium

iodide (BD Biosciences, San Jose CA, USA) was used at 30 µM. More than 70% of early apoptotic

cells were PS positive while remained PI negative. In contrast, 80% of primary and secondary

necrotic cells were positive for both PS and PI.

Macrophages were seeded in adherent 24-well plates at 2 x 105 cells per well and were incubated

overnight at 37°C in 5% CO2. At the indicated time points, target cells were induced to undergo

apoptosis or primary and secondary necrosis, washed once, and added for coincubation with

macrophages resulting in a 1:3 ratio of macrophages to target cells. At the indicated time points,

macrophages were washed once in phosphate buffered saline (PBS) and lysed in 200 l of ice-

cold Nonidet P-40 lysis buffer. Luciferase activity was measured as described above.

Results and Discussion

We generated a lentiviral vector containing a NF- B regulated luciferase reporter gene and

transduced Mf4/4 with this vector. This approach revealed that neither apoptotic nor primary or

secondary necrotic cells induce NF- B activation in macrophages upon their uptake (Figure 1).

This confirms our previously published data (Brouckaert et al., 2004) that neither secondary nor

primary necrotic cells induce a pro-inflammatory response in macrophages.


Figure 1. A reporter of NF- B dependent transcription reveals that neither apoptotic nor primary and

secondary necrotic cells modulate NF- B activity in macrophages. Macrophages were incubated with viable, apoptotic and primary and secondary necrotic targets (prepared as described in Materials and Methods).Data are presented as luciferase activity in macrophages cocultured with target cells relative to the control population, which were cultured without any target cells (“fold-induction”).

In contrast, cells produced by dounce-lysis or freeze-thaw lysis induced activation of NF- B in

macrophages (Li et al., 2001) and induced release of pro-inflammatory cytokines (Fadok et al.,

2001). It is important to point out that physical disruption of cells results in passive, uncontrolled

necrosis, which may be completely different from necrosis as a result of TNF-mediated signal

transduction. For instance, it has been shown that the triggering insult provoking necrotic death

has an essential impact on the kind of immunological response of phagocytes (Hirt and Leist,

2003). The authors demonstrated that uptake of signal transduction mediated necrotic

(staurosporine treated under ATP depletion) Jurkat cells clearly inhibited the E. coli-induced TNF-

secretion by human monocyte-derived macrophages to a comparable extent as the uptake of

apoptotic cells, while heat-killed cells failed to inhibit the proinflammatory reaction by

macrophages.

Importantly, it has been shown in the same model system as that used in the present study that

uptake of apoptotic and necrotic cells by Mf4/4 is PS-dependent (Brouckaert et al., 2004). Thus

one of the possible explanations of the absence of any pro-inflammatory effects of dead necrotic

Viable

Time of coculture (h)

0

2

4

6

8

10

12

14

16

18

1 2 3 4 5 6 7

NF

-B

-De

pe

nd

en

t L

uc

ife

ras

e A

cti

vit

y

(fo

ld o

f in

du

cti

on

)

Apoptotic

Secondary Necrotic

Primary Necrotic

LPS

NF- B 103

cells may be attributed to PS exposure by both types of dying cells. In this regard it has been

shown that PS can inhibit macrophage production of pro-inflammatory cytokines and NO, and even

more, it can block macrophage killing of intracellular parasites (Aramaki, 2000; Aramaki et al.,

1996; Gilbreath et al., 1985a; Gilbreath et al., 1985b). Furthermore, blocking PS on apoptotic cells

with annexin V has been shown to eliminate the inhibitory effect of the apoptotic cells on the

humoral response (Stach et al., 2000). Thus, in the future it would be of high interest to investigate

whether exposed PS on necrotic cells is indeed responsible for the absence of pro-inflammatory

response of macrophages.


References

Aramaki Y. 2000. Liposomes as immunomodulator--inhibitory effect of liposomes on NO

production from macrophages. Biol Pharm Bull 23(11):1267-1274.

Aramaki Y, Arima H, Hara T, Tsuchiya S. 1996. Liposomal induction of a heat-stable macrophage

priming factor to induce nitric oxide in response to LPS. Pharm Res 13(9):1389-1392.

Beinke S, Ley SC. 2004. Functions of NF-kappaB1 and NF-kappaB2 in immune cell biology.

Biochem J 382(Pt 2):393-409.

Brouckaert G, Kalai M, Krysko DV, Saelens X, Vercammen D, Ndlovu, Haegeman G, D'Herde K,

Vandenabeele P. 2004. Phagocytosis of necrotic cells by macrophages is phosphatidylserine

dependent and does not induce inflammatory cytokine production. Mol Biol Cell

15(3):1089-1100.

Caamano J, Hunter CA. 2002. NF-kappaB family of transcription factors: central regulators of

innate and adaptive immune functions. Clin Microbiol Rev 15(3):414-429.

Fadok VA, Bratton DL, Guthrie L, Henson PM. 2001. Differential effects of apoptotic versus lysed

cells on macrophage production of cytokines: role of proteases. J Immunol 166(11):6847-

6854.

Gilbreath MJ, Nacy CA, Hoover DL, Alving CR, Swartz GM, Jr., Meltzer MS. 1985a. Macrophage

activation for microbicidal activity against Leishmania major: inhibition of lymphokine

activation by phosphatidylcholine-phosphatidylserine liposomes. J Immunol 134(5):3420-

3425.

Gilbreath MJ, Swartz GM, Jr., Alving CR, Nacy CA, Hoover DL, Meltzer MS. 1985b. Differential

inhibition of macrophage microbicidal activity by liposomes. Infect Immun 47(2):567-569.



Lamkanfi M, Kalai M, Saelens X, Declercq W, Vandenabeele P. 2004. Caspase-1 activates nuclear

factor of the kappa-enhancer in B cells independently of its enzymatic activity. J Biol Chem

279(23):24785-24793.

Li M, Carpio DF, Zheng Y, Bruzzo P, Singh V, Ouaaz F, Medzhitov RM, Beg AA. 2001. An

essential role of the NF-kappa B/Toll-like receptor pathway in induction of inflammatory

and tissue-repair gene expression by necrotic cells. J Immunol 166(12):7128-7135.

O'Mahoney JV, Adams TE. 1994. Optimization of experimental variables influencing reporter gene

expression in hepatoma cells following calcium phosphate transfection. DNA Cell Biol

13(12):1227-1232.

Stach CM, Turnay X, Voll RE, Kern PM, Kolowos W, Beyer TD, Kalden JR, Herrmann M. 2000.

Treatment with annexin V increases immunogenicity of apoptotic human T-cells in Balb/c

mice. Cell Death Differ 7(10):911-915.

Van Huffel S, Delaei F, Heyninck K, De Valck D, Beyaert R. 2001. Identification of a novel A20-

binding inhibitor of nuclear factor-kappa B activation termed ABIN-2. J Biol Chem

276(32):30216-30223.

Zingarelli B. 2005. Nuclear factor-kappaB. Crit Care Med 33(12 Suppl):S414-416.

3.3 Mechanisms of internalization of apoptotic and necrotic L929 cells by a macrophage cell line studied by electron microscopy

DV Krysko, G Brouckaert, M Kalai, P Vandenabeele and K D’Herde Journal of Morphology 2003; 258: 336–345.

Mechanisms of Internalization of Apoptotic and NecroticL929 Cells by a Macrophage Cell Line Studied byElectron MicroscopyDmitri V. Krysko,1* Greet Brouckaert,2 Michael Kalai,2 Peter Vandenabeele,2 andKatharina D’Herde1

1Department of Human Anatomy, Embryology, Histology and Medical Physics, Ghent University,9000 Ghent, Belgium2Molecular Signaling and Cell Death Unit, Department of Molecular Biomedical Research, VIB, Ghent University,9000 Ghent, Belgium

ABSTRACT Rapid and efficient phagocytic removal ofdying cells is a key feature of apoptosis. In necroticcaspase-independent modes of death, the role and extentof phagocytosis is not well documented. To address thisissue, we studied at the ultrastructural level the phago-cytic response to dying cells in an in vitro phagocytosisassay with a mouse macrophage cell line (Mf4/4). As tar-get cells, murine L929sAhFas cells were induced to die byTNFR1-mediated necrosis or by Fas-mediated apoptosis.Apoptotic L929sAhFas cells are taken up by completeengulfment of apoptotic bodies as single entities forming atight-fitting phagosome, thus resembling the “zipper”-likemechanism of internalization. In contrast, primary andsecondary necrotic cells were internalized by a macropi-nocytotic mechanism with formation of multiple ruffles bythe ingesting macrophage. Ingestion of necrotic cellularmaterial was invariably taking place after the integrity ofthe cell membrane was lost and did not occur as discreteparticles, in contrast to apoptotic material that is sur-rounded by an intact membrane. Although nuclei of ne-crotic cells have been observed in the vicinity of macro-phages, no uptake of necrotic nuclei was observed. Thepresent report provides a basis for future studies aimed atdiscovering molecular pathways that precede these di-verse mechanisms of uptake. J. Morphol. 258:336–345,2003. © 2003 Wiley-Liss, Inc.

KEY WORDS: apoptosis; necrosis; macropinocytosis;phagocytosis; ultrastructure

One of the most conspicuous features of apoptoticcell death is the efficient and swift removal of dyingcells by professional macrophages or by neighboringcells (Garcia-Porrero and Ojeda, 1979; Parnaik etal., 2000). Apoptotic cells exhibit numerous signalsfor phagocytosis, including surface exposure of phos-phatidylserine and alteration of membrane carbohy-drates. Multiple ligands and receptors have beenimplicated in the recognition and uptake of apopto-tic cells before membrane lysis, thus preventing re-lease into the tissue of potentially toxic and immu-nogenic intracellular substances (Henson et al.,2001a; Somersan and Bhardawaj, 2001). In addi-

tion, the binding and/or uptake of apoptotic cells notonly fails to induce secretion of inflammatory medi-ators by the macrophage but inhibits proinflamma-tory cytokine production following stimulation(Huynh et al., 2002). In the dichotomic classificationof cell death, necrosis was previously invariably de-fined as a disordered mode of cell death, occurringeither in cases of severe and acute injuries such assudden shortage of nutrients and abrupt anoxia orin extreme injuries such as heat, detergents, strongbases, and irradiation. More recently, the existenceof a necrotic-like cell death pathway regulated by anintrinsic death program distinct from that of apopto-sis becomes more and more documented (Kitanakaand Kuchino, 1999; Denecker et al., 2001;Proskuryakov et al., 2002).

Rabinovitch (1967) was the first to describe twodistinct stages in the phagocytic process. The first isthe attachment step, in which the solids adhere tothe surface of the phagocyte, followed by internal-ization of the particle. Two subtypes of the internal-ization process (endocytosis) were observed so far.One is the efficient antigen uptake mechanism in-volving the internalization of large particles calledphagocytosis and the other is fluid phase uptake ofsmall molecules. Fluid phase uptake can be per-formed via two distinct mechanisms, either by mi-cropinocytosis, which is the ingestion of small vesi-cles via clathrin-coated pits or macropinocytosis bypinosomes (�0.2 �m diameter), which are formed bymembrane ruffling (Swanson and Baer, 1995; Swan-

Contract grant sponsor: the FWO-Vlaanderen (doctoral grant toGB); Contract grant number: GOA 120505502.

*Correspondence to: Dmitri V. Krysko, Department of HumanAnatomy, Embryology, Histology and Medical Physics, Godshuizen-laan, 4, B-9000 Ghent, Belgium. E-mail: [email protected]

105

Figure 1


son and Watts, 1995; Akari et al., 1996; Jones andWillingham, 1999; Somersan and Bhardwaj, 2001;Toril et al., 2001).

Recently, it has been proposed that macropinocy-tosis is the mechanism for uptake of apoptotic cells(Henson et al., 2001b; Hoffmann et al., 2001). Asimilar mechanism was suggested to occur for theuptake of cell debris during removal of damaged andnecrotic cells (Ogden et al., 2001). However, the in-ternalization mechanisms of apoptotic and necroticcells have not been studied at the ultrastructurallevel. To address this question, we used a well-characterized cell death model system in which mu-rine L929sAhFas cells die by apoptosis when stim-ulated with agonistic anti-Fas antibodies or humanFas ligand, while treatment with TNF elicits ne-crotic cell death (Vercammen et al., 1997, 1998).Apoptotically or necrotically dying L929sAhFas cellswere used as target cells in an in vitro phagocytosisassay using a mouse macrophage cell line (Mf4/4) asa professional phagocyte to discover the ultrastruc-tural characteristics of the internalization mecha-nisms for apoptotic and necrotic cells.

MATERIALS AND METHODSCells

The mouse fibrosarcoma cell L929sAh was selected for its sen-sitivity to the cytotoxic activity of TNF (Vanhaesebroeck et al.,1991). L929sAh was stably transfected with the human Fas re-ceptor to produce L929sAhFas cells, as described previously (Ver-cammen et al., 1998). Mf4/4 is a mouse macrophage cell line thathas been characterized by Desmedt et al. (1998).

All cells were grown in RPMI medium (Gibco/BRL, Eggenstein,Germany), supplemented with 10% fetal calf serum, penicillin(100 U/ml), glutamax I (200 �M), �-mercaptoethanol (2 � 10-5

mM), sodium pyruvate (1 mM), and kept in endotoxin-free condi-tions.

Reagents

Recombinant murine TNF (mTNF) was produced in Esche-richia coli and purified to at least 99% homogeneity and specific

activity of 6 � 107 IU/ml as determined in a standardized cyto-toxicity assay using L929sA cells. Agonistic antihuman Fas an-tibodies (clone 2R2) were purchased from BioCheck (Munster,Germany). Fc block (purified rat antimouse CD16/CD32 monoclo-nal antibody) was purchased from Pharmingen (San Diego, CA)and used at 5 �g/ml.

Induction of Apoptotic and Necrotic CellDeath

L929sAhFas target cells were kept in suspension by seeding2.5 � 105 cells per well in uncoated 24-well tissue culture plates(Sarstedt). Apoptotic cell death was induced by adding anti-Fasantibody (250 ng/ml) for 1 h (early apoptotic condition) or for 18 h(late apoptotic condition). Necrotic cell death was induced byadding mTNF (10.000 IU/ml) for 7 h (early necrotic condition) orfor 18 h (late necrotic condition).

In Vitro Phagocytosis Assay

Macrophages were seeded in adherent 24-well plates at 2.5 �105 cells per well and incubated at 37°C and 5% CO2. Macro-phages were pretreated with Fc block for 15 min before coincu-bation with target cells to prevent the anti-Fas antibody fromaffecting uptake. For coincubation 200 �l of a suspension of eithercontrol or apoptotic or necrotic target cells (2.5 � 106/ml) wasadded to each well, resulting in a 1:1 ratio of macrophage to targetcells. Macrophages and control untreated, apoptotic, or necrotictarget cells were cocultured at 37°C, with 5% CO2 for either 30 or90 min. Studies by G. Brouckaert et al. (data not shown) revealedthat uptake of dying L929sAhFas cells is not due to mTNF effectson the macrophages, since similar results were obtained withhuman TNF-R1 expressing L929sAhR55 cells induced to die bynecrosis with R32SW68T, an hTNF mutant that does not interactwith murine TNF-R1 but specifically activates human TNF-R1(Loetscher et al., 1993).

Transmission Electron Microscopy

Adherent cocultures of macrophages and target cells (preparedas described above) were fixed by immersion in 2% glutaralde-hyde containing 1 mM CaCl2, buffered with 0.1 M Na-cacodylate(pH 7.4). Following a rinse in 100 mM Na-cacodylate containing7.5% sucrose, cocultures were osmicated overnight in 1% OsO4 inthe same buffer (without sucrose) and embedded in LX mediumafter dehydration in a graded series of ethanol (70%, 85%, 95%,100%; 10 min each) (Ladd, Burlington, VT, USA). Semithin sec-tions of 2 �m were contrasted with Toluidine blue and examinedusing a Leitz Aristoplan light microscope equipped with a Leitzorthomat E photocamera. Ultrathin sections of 60 nm were cutwith a diamond knife on a Reichert Jung Ultracut E ultramic-rotome (Austria), mounted on formvar-coated copper grids andstained with uranyl acetate and lead citrate. Samples wereviewed on a Jeol EXII transmission electron microscope at 80 kVaccelerating voltage.

RESULTSUnstimulated L929sAhFas Cells CoculturedWith Macrophages (Control): UltrastructuralCharacteristics

In control experiments, untreated L929sAhFascells were cocultured with macrophages as describedabove. The spindle-shaped control L929sAhFas cellswere rounded up during the harvesting process.These cells showed microvilli protruding from theentire surface, well-preserved cytoplasmic or-ganelles, smoothly outlined nuclei, and chromatin

Fig. 1. A: Control. Unstimulated L929sAhFas cocultured for90 min with Mf4/4 macrophage. Scale bar 1 �m. A’: Detail ofrectangle of A. Well-preserved mitochondria (arrowhead). Scalebar 0.5 �m. B: Control. Mf4/4 macrophage cocultured for 90min with unstimulated L929sAhFas. Scale bar 1 �m. C: Mac-rophage with engulfed apoptotic bodies of different sizes (arrow-heads), surrounded by three apoptotic cells with marginated con-densed chromatin (filled arrows). Note as well a single secondarynecrotic L929sAhFas cell (unfilled arrow). Scale bar 1 �m. C’:Detail of rectangle of C. Heterogeneous population of cellularorganelles can be observed: normal mitochondria (arrowhead) vs.swollen mitochondria (arrow). Scale bar 0.5 �m. D: Lightmicroscopy. A L929sAhFas cell is in an advanced stage of theapoptotic cell death process showing sealing of the plasma mem-brane and formation of a number of membrane-bound apoptoticbodies (arrowhead). Scale bar 100 �m. E: Ultrastructural fea-tures of the budding process of apoptosis. Scale bar 1 �m.C,D,E: Target cells treated with agonistic anti-Fas for 1 h andthen cocultured for 30 min with Mf4/4.

Internalization mechanisms 107

either finely dispersed or partly in the form of het-erochromatin (Fig. 1A,A’). Control macrophagesshowed the presence of abundant cytoplasm contain-ing numerous lysosomes (Fig. 1B). No ruffles (Fig.1B) and no uptake of living control target cells wereobserved.

Ultrastructural Features of theInternalization Mechanisms Used by theMacrophages

For the uptake of apoptotic bodies. After in-duction of apoptosis in L929sAhFas cells by 1 h ofanti-Fas treatment (early apoptotic condition) andcoincubation with macrophages for either 30 or 90min, typical signs of apoptosis were observed in thetarget cells. Microvilli had mostly disappeared(Fig. 1C). Nuclear chromatin was compacted intosharply circumscribed, uniformly dense masses thatabutted on the nuclear envelope (Fig. 1C); the cyto-plasm was condensed and contained both well-preserved and swollen mitochondria (Fig. 1C’). Sur-face protuberances separated with sealing of theplasma membrane, converting the cell into a num-ber of membrane-bound apoptotic bodies of varyingsize containing different cytoplasmic organelles;some of these bodies lacked a nuclear component,whereas others contained one or more nuclear frag-ments in which compacted chromatin was distrib-uted either in a peripheral crescent or throughouttheir cross-sectional area (Fig. 1D,E). Analysis of alarge number of ultrathin sections revealed the var-ious stages of apoptotic body internalization inwhich apoptotic bodies were either adhering to themacrophage surface or were partially surrounded byextended narrow surface projections or pseudopods(Fig. 2B), while others were completely engulfed bymacrophages. Apoptotic bodies inside the macro-phage formed a tight-fitting phagosome (Figs. 2A,3D). In the initial stages a double membrane struc-ture formed by both phagosomal and the apoptoticbody membranes was easily observed (Figs. 2A, 3D).Macrophages containing multiple phagosomes of dif-ferent sizes were often observed (Fig. 1C).

For the uptake of primary and secondarynecrotic material. Cells induced to necrosis bymTNF treatment demonstrated typical ultrastruc-tural features of necrotic cell death such as swellingof the organelles (endoplasmic reticulum, mitochon-dria) and the cytoplasm with subsequent destruc-tion of the plasma membrane. During internaliza-tion of necrotic material, macrophages showed thegreatest number of membrane ruffles over their en-tire surface, creeping up the “ghost”-like structure ofthe dead cells (Fig. 3A–C).

These membrane ruffles formed into the macro-phage spacious macropinosomes containing en-gulfed necrotic material, which was flowing in co-ingested extracellular fluid (Fig. 3A,B). In contrastto the narrow surface protrusions or pseudopods

simply fusing during apoptotic body internalization,these protrusive flaps of cytoplasm or ruffles oftenformed elaborate labyrinths with protrusions thatseem to recede back into the cytoplasm (Fig. 3B,E).The macropinosomes were heterogeneous in size,some fused with each other, and formed macropino-somes as large as 4 �m in diameter (Fig. 3B). Anumber of images showed the diversity and scope ofthe internalization mechanism. Several indepen-dent macrophages, not necessarily in contact, occa-sionally attempted to internalize the same “ghost”-like dead cell by means of multiple rufflesemanating from each macrophage to take up differ-ent amounts of material from the target cell (Fig.4B). Very often necrotic nuclei were observed lyingin the close vicinity to the macrophages (Fig. 4A).Hereby, the plasma membrane of the macrophagelined up to the membrane of the necrotic nucleus(Fig. 3C). However, attempts of the macrophages toengulf the necrotic nucleus either through formationof pseudopods (Fig. 4C,D) or ruffles (Fig. 3C) werenever observed to be successful, i.e., engulfment ofnecrotic nuclei was never observed.

When apoptosis was induced by 18 h of anti-Fastreatment (late apoptotic condition) followed by co-incubation with macrophages for 30 or 90 min, tar-get cells showed pronounced signs of secondary ne-crosis (Majno and Joris, 1995; Van Cruchten andVan den Broeck, 2002). Chromatin was condensedand still showed margination or was lost from thenucleus, while the cytoplasm underwent an exten-sive microvacuolation with clearly damaged or-ganelles (Fig. 5A,B). Interestingly, when confrontedwith such target cells macrophages internalized thelate apoptotic material via pronounced membraneruffling and the uptake was restricted to small par-ticles, as seen with necrotic target cells (Fig. 3D,E).Notably, in these experimental conditions uptake oflarge-diameter particles was observed (Fig. 3D).Moreover, a single macrophage was able to performboth mechanisms of internalization simultaneously:at one side phagocytosis of an apoptotic body byformation of pseudopods and at another side inter-nalization of “secondary” necrotic material by mac-ropinocytosis with pronounced membrane ruffling(Fig. 3D).

DISCUSSION

The present observations demonstrate that mu-rine L929sAhFas cells when stimulated with agonis-tic anti-Fas antibodies show classical ultrastruc-tural features characteristic of apoptosis (Fig.1C,C’). The heterogeneity in the ultrastructural ap-pearance of mitochondria as shown in Figure 1C’ arein line with earlier reported heterogeneity duringapoptosis induction with regard to cytochrome c re-lease and loss of mitochondrial transmembrane po-tential (D’Herde et al., 2000; Krysko et al., 2001).Apoptotic cells that undergo budding are swiftly in-


Fig. 2. A: Detail of an apoptotic body (white asterisk) inter-nalized by a macrophage demonstrating the formation of atightly fitting phagosome. Note a double membrane structure(arrows) formed by the phagosomal and apoptotic body mem-branes. Nucleus of the macrophage (black asterisk). B: Processof engulfment of apoptotic bodies by the macrophage. Observethe macrophage’s tiny protrusions towards the apoptotic bodies(arrowhead). Target cells treated with agonistic anti-Fas duringeither 18 h (A) or 1 h (B) before coculture with Mf4/4 for 30 min.Scale bars 1 �m.


Figure 3


ternalized by phagocytosis as well-enclosed distinctapoptotic bodies. In the initial stages ingestion en-tails two component activities, the extension and theclosure (fusion) of pseudopodia (Fig. 3D) along theapoptotic body, leading to the formation of an intra-cellular phagosome with a clear double membranestructure (Figs. 2A, 3D). However, only one mem-brane was seen in the advanced stages of digestion(Fig. 1C). These findings obtained in an in vitrosystem are in agreement with the observations fromin vivo systems reported in a landmark article bySchweichel and Merker (1973). They characterizedultrastructurally three types of cell death in thetissue of mouse and rat embryos and fetuses andshowed that the cells that demonstrated morpholog-ical criteria of type-I cell death (apoptotic cell death)were quickly surrounded by thin, flat processes fromneighboring cells and taken up, followed by the for-mation of intracellular phagosomes with initially adouble membrane structure. Moreover, our resultsdemonstrate that during phagocytosis of apoptoticbodies, the phagocytic plasma membrane envelopsthe apoptotic body as a closely fitting sleeve (Figs.2A, 3D), thus resembling the zipper model forphagocytosis proposed by Griffin et al. (1975a,b). Acorollary of the zipper model is that the advance of apseudopod requires continual ligation of cell surfacereceptors and therefore the phagosomal membraneshould be closely apposed to the particle surface(Swanson and Baer, 1995). As a consequence, theintegrity of the apoptotic body’s plasma membraneis an important prerequisite for this mechanism.

After 7 h of mTNF treatment L929sAhFas cellsdemonstrate typical signs of necrosis such as cellu-lar swelling and organelle disintegration, condensa-tion of nuclear chromatin in ill-defined smallmasses, and loss of the plasma membrane integrity.This is in agreement with ultrastructural features ofnecrosis in a wide range of cells (Wyllie et al., 1980).

This type of cell death was defined by Schweicheland Merker (1973) as physiological cell death type-III. In their studied prenatal tissues no reaction ofneighboring nonprofessional phagocytes to necrotictargets was discovered. However, more recently theactive engulfment of necrotic cells by phagocyteswas documented by several groups (Sauter et al.,2000; Subklewe et al., 2001). Macrophages attempt-ing to engulf necrotic cells demonstrated numerousruffles and folds of their plasma membrane creepingup the necrotic material (Fig. 3A–C). The formationof membrane ruffles was proposed recently to be aspecific feature of macropinocytosis, a type of endo-cytosis (Ridley et al., 1992; Lamaze and Schmid,1995; Toril et al., 2001). Macropinosomes are classi-fied as phase-bright organelles with diameters �0.2�m (Jones and Willingham, 1999). In the presentstudy macropinosomes are spacious and heteroge-neous in size, forming underneath the plasma mem-brane as ruffles that fold back against the cell sur-face, trapping necrotic material with extracellularfluid (Fig. 3B).

Our observations demonstrate that both internal-ization mechanisms entail the formation of surfaceprotrusions: apoptotic cells as targets elicit narrowpseudopods (Figs. 2B, 3D), while necrotic target cellselicit ruffles that are in most cases disproportional tothe engulfed necrotic material (Fig. 3A–C) and inabsolute terms are often wider then the pseudopods.Moreover, the tightly fitting engulfment distin-guishes phagocytosis from macropinocytosis (Figs.2A, 3D, vs. 3A–C) (Swanson and Baer, 1995; Swan-son and Watts, 1995). In the latter mechanism, clo-sure of the macropinocytotic ruffles into an intracel-lular vesicle does not always follow (Swanson andWatts, 1995; Jones and Willingham, 1999). Usingthis mechanism, macrophages were able to grasponly small volumes from the dead cells and, thus,the process of uptake is less efficient than that ofapoptotic cells. We never observed engulfment ofTNF-challenged L929sAhFas cells with an intactplasma membrane. Interestingly, the same macro-phage is able to perform both mechanisms of inter-nalization simultaneously: i.e., phagocytosis of anapoptotic body by the zipper-like mechanism andinternalization of “secondary” necrotic material bymacropinocytosis (Fig. 3D,E). This suggests that thetype of cell death at the contact site between a dyingtarget cell and a phagocyte determines themembrane-associated events in the latter cell.

Recently, it has been proposed that macropinocy-tosis is the mechanism for internalization of apopto-tic cellular material (Henson et al., 2001b; Hoff-mann et al., 2001). Indeed, it has been shown bylight and immunofluorescence microscopy that up-take of apoptotic cells initially involves formation ofspacious phagosomes and the concurrent ingestionof extracellular fluid (Hoffmann et al., 2001; Ogdenet al., 2001). This is in marked contrast with theinternalization mechanism of apoptotic bodies in our

Fig. 3. A: Internalization of necrotic cell material by mem-brane ruffling of the macrophage (arrow). B: Enlargement of A. Alabyrinth of surface ruffles of the macrophage surround the dis-integrated cytoplasm of a necrotic target cell. Closure of theruffles into an intracellular macropinosome (arrow). Fusion oftwo macropinosomes (unfilled arrowhead). Necrotic particlesflowing freely in the macropinosome (asterisk). C: A macrophagedisplays surface ruffles toward disintegrated necrotic cytoplasm(arrowheads). Nuclear membrane of karyolytic nucleus in closeapposition to the macrophage. D: Overview. The process of inter-nalization of an apoptotic body is seen at the bottom of themacrophage. Fusion of two pseudopods (arrow) by the formationof tightly fitting phagosome during apoptotic body internaliza-tion. Homogenous space (asterisk) is an artifact due to cell prep-aration for electron microscopy. At the left upper corner second-ary necrotic material is taken up by multiple ruffles. Note amacropinosome containing freely flowing secondary necrotic par-ticles (arrowhead). E: Detail of D left side. Numerous surfaceruffles surround secondary necrotic material (asterisks). Scalebars 1 �m. A–C: Target cells treated with mTNF for 18 h beforecoculture with Mf4/4 for 90 min. D,E: Target cells treated withagonistic anti-Fas for 18 h before coculture with Mf4/4 for 90 min.


model system, where there is no evidence for fluiduptake into the phagosome (Figs. 2A, 3D). The sim-plest explanation is that the cells they used wereindeed late apoptotic cells exhibiting morphologicalcharacteristics of secondary necrosis. Another expla-nation may be that the way by which dead cells areinternalized is dependent on the trigger used to killthem (apoptosis induction: anti-Fas antibody in thepresent study vs. UV irradiation in Hoffmann et al.,

(2001) and Ogden et al., (2001)). Indeed, it has beensuggested that the apoptotic trigger can determinethe extent by which dying cells are recognized andremoved by phagocytes (Wiegand et al., 2001). Thus,it might be that apoptotic cells are not necessarilyequally marked to be internalized by the samemechanism. However, dead cells exhibiting mor-phological characteristics of secondary necrosis(Fig. 3D,E) or primary necrosis (Fig. 3A–C) are

Fig. 4. Target cells treated with mTNF for 7 h before cocul-ture with Mf4/4 for 30 min. A,B: Light microscopy. A: Necroticnuclei attached to the macrophages (filled arrowheads) and notattached necrotic nuclei (unfilled arrowheads). Scale bar 100�m. B: Several macrophages attempt to internalize the samenecrotic cell (arrow) by formation of multiple ruffles. Scale bar 100 �m. C: Macrophage attached to necrotic nuclear ghost(arrow). Scale bar 1 �m. D: Detail of C. Note the presence ofa narrow pseudopod-like protrusion (arrowhead). Scale bar 1 �m.


internalized by the macropinocytotic mechanism,as suggested but not documented by Hoffmann etal. (2001) and Ogden et al. (2001).

Of course, one may question whether in vivophagocytes indeed use different internalizationmechanisms in response to the presented dying cell.Two different mechanisms of internalization presentin in vivo and in vitro have been described for opso-nized sheep red blood cells by mouse peritoneal mac-rophages (Orenstein and Shelton, 1977). One is in-ternalization by multiple surface protrusions, whichare frequently of unequal size and have overlappingmargins, thus resembling the macropinocytoticmechanism for uptake of necrotic cells by membraneruffling, as proposed in this study. Another is inter-nalization of erythrocytes by a single cup-like orfunnel-like pseudopod, which is similar to the“zipper”-like internalization of apoptotic bodies.Moreover, a mechanism of “sinking-in craters” forinternalization of apoptotic cells in early embryo-genesis of the chick lens rudiment has also beendocumented (Polliack and Grodon, 1975; Garcia-Porrero et al., 1984). This mechanism can be consid-ered as a variation of the zipper model (Swansonand Baer, 1995), which is shown in the presentstudy as the mechanism of internalization for apo-ptotic bodies in vitro. All these data suggest thatdifferent internalization mechanisms do exist invivo. Another remarkable feature is that necroticnuclei are often situated in the vicinity of the phago-cytic membrane, but were never detected inside themacrophage (Fig. 4A,C,D). One might wonderwhether the membrane of the necrotic nucleus is not

eliciting an “eat me” signal, strong enough for thezipper-like mechanism to occur.

In conclusion, the electron microscopical data pre-sented here demonstrate that both apoptotic andnecrotic cells are removed by the macrophages.Moreover, from our observations the existence ofdifferent internalization mechanisms for apoptoticand necrotic cells is likely. The present report, there-fore, provides a basis for further studies aimed todiscover molecular pathways that precede these di-verse mechanisms of uptake and to study the func-tional consequences of the differential phagocyticuptake on innate, inflammatory, and immune re-sponses.

ACKNOWLEDGMENTS

We thank Erika Roosen, Barbara De Bondt, Simo-nne Van Hulle, Dominique Jacobus, Aron De Smet,and Hubert Stevens for excellent technical assis-tance.

LITERATURE CITED

Akari N, Johnson MT, Swanson JA. 1996. A role for phosphoino-sitide 3-kinase in the completion of macropinocytosis andphagocytosis by macrophages. J Cell Biol 135:1294–1260.

Denecker G, Vercammen D, Declercq W, Vandenabeele P. 2001.Apoptotic and necrotic cell death induced by death domainreceptors. Cell Mol Life Sci 58:356–370.

Desmedt M, Rottiers P, Dooms H, Fiers W, Grooten J. 1998.Macrophages induce cellular immunity by activating Th1 cellresponses and suppressing Th2 cell responses. J Immunol 160:5300–5308.

Fig. 5. Target cells treated with agonistic anti-Fas for 18 h before coculture with Mf4/4 for 90 min. A: A secondary necrotic cell withmargination of chromatin, loss of plasma membrane integrity, vacuolization of the cytoplasm, and clearly damaged organelles.B: Detail of rectangle of A. Note mitochondria in various stages of disintegration (arrows), numerous dense bodies, and tertiarylysosomes (arrowhead) and the absence of Golgi lamellae and rough endoplasmic reticulum. Note also the absence of cytoplasmiccondensation. Scale bars 1 �m.


D’Herde K, De Prest B, Mussche S, Schotte P, Beyaert R, CosterRV, Roels F. 2000. Ultrastructural localization of cytochrome cin apoptosis demonstrates mitochondrial heterogeneity. CellDeath Differ 7:331–337.

Garcia-Porrero JA, Ojeda JL. 1979. Cell death and phagocytosisin the neuroepithelium of the developing retina. A TEM andSEM study. Experientia 35:375–376.

Garcia-Porrero JA, Colvee E, Ojeda JL. 1984. The mechanisms ofcell death and phagocytosis in the early chick lens morphogen-esis: a scanning electron microscopy and cytochemical ap-proach. Anat Rec 208:123–136.

Griffin FM, Griffin JA, Leider JE, Silverstein SC. 1975a. Studieson the mechanism of phagocytosis. I. Requirements for circum-ferential attachment of particle-bound ligands to specific recep-tors on the macrophage plasma membrane. J Exp Med 142:1263–1282.

Griffin FM, Griffin JA, Silverstein SC. 1975b. Studies on themechanism of phagocytosis. II. The interaction of macrophageswith anti-immunoglobulin IgG-coated bone marrow-derivedlymphocytes. J Exp Med 144:788–809.

Henson PM, Bratton DL, Fadok VA. 2001a. The phosphatidylser-ine receptor: a crucial molecular switch? Nat Rev Mol Cell Biol2:627–633.

Henson PM, Bratton DL, Fadok VA. 2001b. Apoptotic cell re-moval. Curr Biol 11:R795–R805.

Hoffmann PR, deCathelineau AM, Ogden CA, Leverrier Y, Brat-ton DL, Daleke DL, Ridley AJ, Fadok VA, Henson PM. 2001.Phosphatidylserine (PS) induces PS receptor-mediated mac-ropinocytosis and promotes clearance of apoptotic cells. J CellBiol 4:649–659.

Huynh MN, Fadok VA, Henson PM. 2002. Phosphatidylserine-dependent ingestion of apoptotic cells promotes TGF-� secre-tion and the resolution of inflammation. J Clin Invest 109:41–50.

Jones NL, Willingham MC. 1999. Modified LDLs are internalizedby macrophages in part via macropinocytosis. Anat Rec 255:57–68.

Kitanaka C, Kuchino Y. 1999. Caspase-independent programmedcell death with necrotic morphology. Cell Death Differ 6:508–515.

Krysko DV, Roels F, Leybaert L, D’Herde K. 2001. Mitochondrialtransmembrane potential changes support the concept of mito-chondrial heterogeneity during apoptosis. J Histochem Cyto-chem 49:1277–1284.

Lamaze C, Schmid SL. 1995. The emergence of clathrin-independent pinocytic pathways. Curr Opin Cell Biol 7:573–580.

Loetscher H, Stueber D, Banner D, Mackay F, Lesslauer W. 1993.Human tumor necrosis factor alpha (TNF alpha) mutants withexclusive specificity for the 55-kDa or 75-kDa TNF receptors.J Biol Chem 268:26350–26357.

Majno G, Joris I. 1995. Apoptosis, oncosis and necrosis. An over-view of cell death. Am J Pathol 146:3–15.

Ogden CA, deCathelineau A, Hoffmann PR, Bratton D, Ghebre-hiwet B, Fadok VA, Henson PM. 2001. C1q and mannose bind-ing lectin engagement of cell surface calreticulin and CD91initiates macropinocytosis and uptake of apoptotic cells. J ExpMed 194:781–795.

Orenstein JM, Shelton E. 1977. Membrane phenomena accompa-nying erythrophagocytosis. A scanning electron microscopestudy. Lab Invest 36:363–374.

Parnaik R, Raff MC, Scholes J. 2000. Differences between theclearance of apoptotic cells by professional and non-professionalphagocytes. Curr Biol 10:857–860.

Polliack A, Grodon S. 1975. Scanning electron microscopy of mu-rine macrophages. Surface characteristics during maturation,activation, and phagocytosis. Lab Invest 33:469–477.

Proskuryakov SYa, Gabai VL, Konoplyannikov AG. 2002. Necro-sis is an active and controlled form of programmed cell death.Biochemistry (Moscow) 67:387–408.

Rabinovitch M. 1967. The dissociation of the attachment andingestion phases of phagocytosis by macrophages. Exp Cell Res46:19–28.

Ridley AJ, Paterson HF, Johnston CL, Diekmann D, Hall A. 1992.The small GTP-binding protein rac regulates growth factor-induced membrane ruffling. Cell 70:401–410.

Sauter B, Albert ML, Francisco L, Larsson M, Somersan S,Bhardwaj N. 2000. Consequences of cell death: exposure tonecrotic tumor cells, but not primary tissue cells or apoptoticcells, induces the maturation of immunostimulatory dendriticcells. J Exp Med 191:423–434.

Schweichel JU, Merker HL. 1973. The morphology of varioustypes of cell death in prenatal tissues. Teratology 7:253–266.

Somersan S, Bhardwaj N. 2001. Tethering and tickling: a newrole for the phosphatidylserine receptor. J Cell Biol 155:501–504.

Subklewe M, Paludan C, Tsang ML, Mahnke K, Steinman RM,Munz C. 2001. Dendritic cells cross-present latency gene prod-ucts from Epstein-Barr virus-transformed B cells and expandtumor-reactive CD8(�) killer T cells. J Exp Med 193:405–411.

Swanson JA, Baer SC. 1995. Phagocytosis by zippers and trig-gers. Trends Cell Biol 5:89–93.

Swanson JA, Watts C. 1995. Macropinocytosis. Trends Cell Biol5:424–428.

Toril I, Morikawa S, Nagasaki M, Nokano A, Morikawa K. 2001.Differential endocytotic characteristics of a novel human B/DCcell line HBM-Noda: effective macropinocytic and phagocyticfunction rather than scavenging function. Immunology 103:70–80.

Van Cruchten S, Van den Broeck W. 2002. Morphological andbiochemical aspects of apoptosis, oncosis and necrosis. AnatHistol Embryol 31:214–223.

Vanhaesebroeck B, Van Bladel S, Lenaerts A, Suffys P, BeyaertR, Lucas R, Van Roy F, Fiers W. 1991. Two discrete types oftumor necrosis factor-resistant cells derived from the same cellline. Cancer Res 51:2469–2477.

Vercammen D, Vandenabeele P, Beyaert R, Declercq W, Fiers W.1997. Tumor necrosis factor-induced necrosis versus anti-Fas-induced apoptosis in L929 cells. Cytokine 9:801–808.

Vercammen D, Brouckaert G, Denecker G, Van de Craen M,Declercq W, Fiers W, Vandenabeele P. 1998. Dual signaling ofthe Fas receptor: initiation of both apoptotic and necrotic celldeath pathways. J Exp Med 188:919–930.

Wiegand UK, Corbach S, Prescott AR, Savill J, Spruce BA. 2001.The trigger to cell death determines the efficiency with whichdying cells are cleared by neighbors. Cell Death Differ 8:734–746.

Wyllie AH, Kerr JFR, Currie AR. 1980. Cell death: the signifi-cance of apoptosis. Int Rev Cytol 68:251–306.


3.4 Macrophages use different internalization mechanisms to clear apoptotic and necrotic cells

DV Krysko, G Denecker, N Festjens, S Gabriels, E Parthoens, K D’Herde and P Vandenabeele Cell Death and Differentiation advance online publication, 21 April 2006; doi:10.1038/sj.cdd.4401900

Macrophages use different internalizationmechanisms to clear apoptotic and necrotic cells

DV Krysko1,2, G Denecker2, N Festjens2, S Gabriels1,

E Parthoens3, K D’Herde1,4 and P Vandenabeele*,2,4

1 Department of Human Anatomy, Embryology, Histology and Medical Physics,Ghent University, Ghent 9000, Belgium

2 Molecular Signalling and Cell Death Unit, Department of Molecular BiomedicalResearch, VIB – Ghent University, Ghent 9052, Belgium

3 Microscopy Core Facility, Department for Molecular Biomedical Research, VIB– Ghent University, Ghent 9052, Belgium

4 These authors share senior authorship.* Corresponding author: P Vandenabeele, Molecular Signalling and Cell DeathUnit, Department for Molecular Biomedical Research, VIB – Ghent University,Technologiepark 927, 9052 Ghent, Belgium. Tel: 32-9-33-13-760;Fax: 32-9-33-13-609; E-mail: [email protected]

Received 13.7.05; revised 24.1.06; accepted 13.2.06Edited by B Zhivorivsky

AbstractThe present study characterized two different internalizationmechanisms used by macrophages to engulf apoptotic andnecrotic cells. Our in vitro phagocytosis assay used a mousemacrophage cell line, and murine L929sAhFas cells that areinduced to die in a necrotic way by TNFR1 and heat shock orin an apoptotic way by Fas stimulation. Scanning electronmicroscopy (SEM) revealed that apoptotic bodies were takenup by macrophages with formation of tight fitting phago-somes, similar to the ‘zipper’-like mechanism of phagocy-tosis, whereas necrotic cells were internalized by amacropinocytotic mechanism involving formation of multipleruffles directed towards necrotic debris. Two macropinocy-tosis markers (Lucifer Yellow (LY) and horseradish perox-idase (HRP)) were excluded from the phagosomes containingapoptotic bodies, but they were present inside the macro-pinosomes containing necrotic material. Wortmannin (phos-phatidylinositol 30-kinase (PI3K) inhibitor) reduced the uptakeof apoptotic cells, but the engulfment of necrotic cellsremained unaffected. Our data demonstrate that apoptoticand necrotic cells are internalized differently by macro-phages.Cell Death and Differentiation advance online publication, 21 April2006; doi:10.1038/sj.cdd.4401900

Keywords: phagocytosis; macropinocytosis; apoptosis; necro-

sis; internalization mechanisms

Abbreviations: PS, phosphatidylserine; PI, propidium iodide;

FACS, fluorescence activated cell sorter; SEM, scanning electron

microscopy; TEM, transmission electron microscope; LY, Lucifer

Yellow; HRP, horseradish peroxidase; PI3K, phosphatidylinositol

30-kinase; mTNF, recombinant murine tumor necrosis factor; anti-

Fas antibodies, agonistic antihuman Fas antibodies; DAPI, 2,6-

diamidino-2-fenylindole; DAB, diaminobenzidine; PARP, poly

(ADP-ribose) polymerase

Introduction

Apoptosis is a highly regulated process in which the cellactivates an intrinsic suicide mechanism that rapidly leads tocell shrinkage, chromatin condensation, membrane blebbing,and the formation of one or more apoptotic bodies.1 Apoptoticcells are cleared rapidly and efficiently as intact cells orapoptotic bodies by professional phagocytes or neighboringcells. A broad array of phagocyte receptors, apoptotic-cell-associated ligands and intermediate molecules has beenproposed to mediate phagocytosis of apoptotic cells. Thesereceptors on the phagocyte include integrins (e.g. aVb3),scavenger receptors (e.g. CD36), immunoglobulin super-family molecules (e.g. CD31) and receptors for complement(e.g. CD91/calreticulin).2–5 Additional molecules, such asthrombospondin, Clq, mannose-binding lectin, C-reactiveprotein and milk fat globule EGF factor 8 may serve asintermediates that bridge macrophage and apoptotic-cellsurfaces or act as opsonins.3,5–7 The best-studied surfacechange on apoptotic cells is the externalization of phospha-tidylserine (PS) due to loss of plasmamembrane phospholipidasymmetry.8 Apoptotic cell clearance is characterized by thelack of induction of proinflammatory macrophage responsesand can exert anti-inflammatory effects on macrophages asevidenced by their release of soluble anti-inflammatorymediators, such as transforming growth factor b, prostaglan-din E2 and platelet activating factor, in response to apoptoticcells.9 Moreover apoptotic cell clearance can trigger tolero-genic responses in the adaptive immune system.10 Additionalmechanismsmay contribute to the anti-inflammatory reaction,for example apoptotic cells themselves can produce immuno-modulatory factors such as IL-1011 and TGF-b.12 Since thepersistence of apoptotic cells can have harmful conse-quences, such as inflammation and autoimmune disease,their efficient removal is of significant biological relevance.In contrast, necrotic cell death typically occurs followingexposure to high concentrations of endogenous or exogenoustoxins, heat treatment, freeze thawing or other immediatelydisruptive insults. Leakage of noxious contents from necroticcells can cause injury to the surrounding tissue and initiatesthe activation or potentiation of proinflammatory responses.Nevertheless, some reports describe a necrotic-like cell deathpathway regulated by an intrinsic death program distinct fromthat of apoptosis.13,14 Unexpectedly, was shown that necroticcells also externalize PS15 and can be recognized through aPS-dependent mechanism.16,17

Two types of internalization processes (endocytosis) havebeen reported so far. Phagocytosis is the efficient antigenuptake by internalization of large particles, and comprisesseveral separate but usually linked events: attachment,internalization, and digestion of the phagocytosed particles.

115

The other process is fluid phase uptake of small molecules,and is performed by two distinct mechanisms: micropino-cytosis is the ingestion of small vesicles via clathrin-coatedpits, and macropinocytosis is ingestion of fluid via pinosomesformed by membrane ruffling.18

It has been proposed that apoptotic cells are taken up bymacropinocytosis.5,19 The authors suggested that a similarmechanism is used for the uptake of necrotic cells. However,in a previous ultrastructural study it had been suggested thatmacrophages use different internalization mechanisms, de-pending on how the target cell has died: phagocytosis forapoptotic cells, and macropinocytosis for necrotic cells.20 Inthe present work, we sought to further validate this workinghypothesis by an in vitro phagocytosis assay, using apopto-tically or necrotically dying L929sAhFas cells as target cells,and a mouse macrophage cell line (Mf4/4) as professionalphagocytes. We observed by scanning electron microscopy(SEM) that macrophages taking up apoptotic or necroticcells have different surface characteristics. Using two fluidphasemarkers, Lucifer Yellow (LY) and horseradish peroxidase(HRP), we also demonstrated the existence of differentinternalization mechanisms: ‘zipper’-like phagocytosis forapoptotic cells and macropinocytosis for necrotic cells.Finally, we demonstrated that wortmannin, a phosphatidyl-inositol 30-kinase (PI3K) inhibitor, differentially influences theuptake of apoptotic and necrotic cells.

Results

Description of apoptotic versus necrotic cell death

L929sAhFas, when stimulated by anti-Fas, die rapidly byapoptosis, but necrotically when exposed to recombinantmurine tumor necrosis factor (mTNF). In order to assessapoptotic versus necrotic cell death, several cell deathparameters were studied, such as caspase activation,poly(ADP-ribose) polymerase (PARP) cleavage, and PSexposure at the surface and cell membrane permeabilization.Apoptotic cell death is characterized by DEVDase activityand consecutive cleavage of PARP, while in the case of TNF-mediated necrotic cell death there is absence of DEVDaseactivity and PARP proteolysis (Figure 1a and b). PS exposure(Annexin-V-FITC labeling) and loss of membrane integrity(propidium iodide (PI) staining) of target cells was determinedby fluorescence activated cell sorter (FACS) assay beforeusing them in the in vitro phagocytosis assay (Figure 1c).

SEM analysis of surface characteristics ofmacrophages during internalization of apoptoticand necrotic cells

In order to further characterize the distinct internalizationmechanisms used by macrophages to engulf apoptotic andnecrotic cells, we used the in vitro phagocytosis assaydescribed by Brouckaert et al.17 and studied surfacecharacteristics of macrophages and target cells by SEM.Mouse macrophages (Mf4/4) were used as professionalphagocytes, and L929sAhFas cells treated with anti-Fas ormTNF to induce apoptotic or necrotic cell death, respectively,were used as target cells. All coculture experiments were

carried out in the presence of heat-inactivated serum, as thepresence of heat-inactivated serum in our experimental setupdid not significantly affect the phagocytic uptake of apoptoticand necrotic cells, which is in agreement with findings of Hartet al.21 (Supplementary Figure 1 is available online at http://www.nature.com/cdd/index.html).In control conditions, the spindle-shaped L929sAhFas cells

rounded up during the short coculture time and the harvestingprocess (Figure 2a). Most of the control macrophages wereattached, flattened, and spread onto the surface of plates(Figure 2a). Macrophages demonstrated marked variations insize, shape, and degree of spreading. Small ruffles wereoccasionally observed on the surface of macrophages, but nouptake of healthy control target cells was detected (Figure 2a).After induction of apoptosis in L929sAhFas cells by

treatment with anti-Fas for 1 h and coculture with macro-phages, apoptotic L929sAhFas cells formed rounded surfaceprotuberances characteristic of budding phenomena(Figure 2b), a well-known feature of apoptosis. Sometimesas many as three apoptotic bodies were seen attached to thecell surface or in the process of being internalized by a singlephagocyte (Figures 2b and b0). Macrophages internalizingapoptotic bodies displayed narrow pseudopods extendingover the surface of apoptotic bodies and enclosing them on allsides (Figures 2b and b0). Spherical or hemispherical craterswith rolled edges containing remnants of apoptotic bodieswere occasionally observed (Figures 2c and c0). A moreprominent feature of macrophages cocultured with apoptoticor necrotic cells, but not frequently seen in macrophagescultured with unstimulated control cells, was the presence of

0

1000

2000

3000

4000

5000

6000

0 0.5 1 1.5 2 2.5 3 4 5 6 7 8Time (h)

∆F/m

inmTNF

7h

10

30

50

70

90

Control%

Cel

lsAnti-Fas

1h55˚C0.5h

a

b

c

Figure 1 (a) Immunoblots for PARP show the appearance of an 85-kDacleaved fragment of PARP in apoptosis but not in necrosis. (b) Time kineticsanalysis of caspase activity was measured by Ac-DEVD-amc cleavage. Blackbars represent apoptotic and white bars necrotic cell death. (c) PS exposure wasdetermined by Annexin V binding and cells were simultaneously stained with PI toassess cell permeability. White bars represent PS positive and PI negative cellsand black bars represent PS and PI positive cells for anti-Fas induced apoptosis,mTNF- and heat-induced necrosis


multiple fine cytoplasmic microvilli emanating from the entirecytoplasm of the macrophages (Figures 2b and b0).TNF-induced necrotic cells had irregular surface changes

that were not observed during apoptotic cell death (Figure3b0). Macrophages internalizing necrotic material had broadmembrane ruffles (Figures 3a and b) terminating in long fineprotrusions (Figures 3b0). Necrotic material can be observedbetween these protrusions (Figures 3b0). Comparison of SEMpictures with transmission electron microscope (TEM) imagessuggests that these protrusions, when retracting, take up thenecrotic material inside the macrophages. In general, macro-phages cocultured with necrotic cells had shapes that were farmore varied and irregular than macrophages cocultured withapoptotic cells, and displayed prominent ruffles.

Analysis of internalization mechanisms using fluidphase markers: lucifer yellow and horseradishperoxidase

In the next part of our work, we used two different fluid phasemarkers to discriminate between the distinct mechanisms of

internalization used by macrophages to engulf apoptotic andnecrotic cells. Fluid-phase pinocytosis is usually studied bydetecting the cellular accumulation of different soluble andimpermeant probes, such as LY and HRP.22–24 Intracellulardistribution of LY and HRP was followed in cocultures ofmacrophages and target cells (unstimulated, apoptotic andnecrotic): LY with fluorescence microscopy, and HRP withlight and transmission electron microscopy. In cocultures withunstimulated living target cells, a few macrophages showedevidence of vacuoles containing LY or HRP, demonstratingthat Mf4/4 were engaged in constitutivemacropinocytosis, butno uptake of living control target cells was observed (Figures5a and 6a).When macrophages were cocultured with apoptotically

dying cells in the presence of LY or HRP, uptake of apoptoticbodies was not accompanied by uptake of LY or HRP (Figures4a, b, 5b, 6b and c). Although both by SEM (Figures 2) andTEM we observed regularly the attachment of severalapoptotic bodies to one macrophage, we never foundinternalization of more than one apoptotic body per macro-phage (Figures 5b and 6b). In contrast, when primary (TNF-induced necrotic) (Figures 4c, 5c, 6d and e) or secondary

Figure 2 SEM micrographs. (a) Control. Unstimulated L929sAhFas cells (C) were cocultured with Mf4/4 macrophages (M). No uptake of healthy cells or membraneruffles were observed. Note absence of microvilli on the surface of the macrophage. (b) Overview, process of internalization of apoptotic bodies (a) by a macrophage.Budding of apoptotic cells (arrowheads). (b0) Enlargement of rectangle in (b) Formation of pseudopods during the internalization of an apoptotic body. Note numerousmicrovilli on the surface of the macrophage. (c) Overview, formation of craters on the surface of a macrophage during coculture with apoptotic cells. (c0) Detail of (c).Remnants of engulfed material can be seen inside the crater. (a, b, b0 c) Scale bars 2 mm. (c0) Scale bar 1 mm


necrotic cells (Figures 4a and b) were taken up by themacrophages, both fluid phase tracers were colocalized withengulfed necrotic material. At the ultrastructural level, macro-

pinosomes were initially irregular in size and shape, containedboth HRP and necrotic material, and assumed a rounderappearance when they were localized more centrally in thecell. HRP-labeled areas were often seen at the cell periphery,near lamellar membrane projections corresponding to rufflingzones, and in cytoplasmic regions that were essentially devoidof other organelles. The reaction product of HRP was neverseen attached to the surface of the macrophages or to thenecrotic cells (Figure 6d). Similarly, when macrophageswere cocultured with necrotic cells, which were killed by heatshock, engulfed necrotic material was colocalized withLY (Figure 4d). In contrast to TNF-induced necrosis, heat-induced necrotic cells were often unfragmented and weretaken up as large entities (Figure 6f). However, in this case themacrophage membrane forms numerous ruffles at the inter-face with the necrotic corpses with formation underneath ofspacious macropinosomes with coingestion of necroticmaterial and HRP (Figure 6g).

Analysis of internalization mechanisms using thePI3K inhibitor wortmannin

PI3K are activated during phagocytosis induced by severalreceptors, and studies using inhibitors such as wortmanninshowed that it is required for phagocytosis of apoptoticcells.25,26 However, the requirement for PI3K in phagocytosisof necrotic cells has not been tested. To investigate whetherthe uptake of apoptotic and necrotic cells by macrophages isPI3K dependent, macrophages were pretreated for 30minwith increasing doses of wortmannin (10–100 nm range) andpercent of uptake was evaluated by FACS analysis. Theseresults showed that wortmannin decreased the percent ofuptake of apoptotic cells by the macrophages in a dose-dependent manner, but did not affect the uptake of necroticcells (Figure 7).

Discussion

Insights into the morphological features of the process ofclearing dead cells have been largely based upon the two-dimensional views furnished by light and TEM,20,27–30 butrelatively little is known concerning the surface topography ofinteraction of macrophages with dying target cells. In thepresent study, we used SEM to visualize in 3-D the differencesin the surface properties of macrophages ingesting eitherapoptotic or necrotic L929sAhFas cells. Apoptotic cells thatunderwent budding (Figure 2b) were swiftly internalized byphagocytosis as well-enclosed distinct apoptotic bodies(Figures 2b and b0). SEM examination showed that duringphagocytosis of apoptotic bodies the phagocytic plasmamembrane envelopes the apoptotic bodies as a closely fittingsleeve (Figure 2b0), which is similar to the zipper model ofphagocytosis proposed by Griffin et al.31,32 According to thezipper model of phagocytosis, pseudopod advance over aparticle is guided by interactions of cell surface receptors, andso the phagosomal membrane should be closely apposedto the particle surface.33 The integrity of the apoptoticbody’s plasma membrane is an important requirement forthis mechanism. These data are in agreement with the

Figure 3 SEM micrographs of TNF-induced necrotic cell internalization. (a)Macrophage (M) demonstrating broad ruffling towards the necrotic cell (arrow).(b) Clearance of necrotic material (arrow). Macrophage (M) displays flatmembrane ruffle directed towards the necrotic debris. (b0) Detail of the rectanglein (b). Necrotic material (arrow) seems to be grasped by tiny protrusionsemanating from the membrane ruffle of the macrophage. Note that the necroticcell (N) has an irregular surface. Scale bars 2mm


Figure 4 Fluorescence microscopy. Coculture of macrophages with apoptotic (a and b), TNF- (c) and heat-induced necrotic (d) cells in the presence of LY. Target cellswere pre-stained with Cytotracker Red (Cytotracker). Uptake of apoptotic cytoplasmic and nuclear (DAPI stained) material by the macrophages without coingestion of LY(arrows in a and b) while secondary necrotic material is engulfed with coingestion of LY (arrowheads in a and b). Ingestion of TNF- (c) and heat-induced (d) necroticmaterial by the macrophages was accompanied by uptake of LY (arrowheads in c and d), which colocalized in the same macropinosomes. Scale bars 8 mm


observations of Giles et al.,34 who used SEM to demonstrate azipper-like interaction between a human monocyte-derivedmacrophage and an apoptotic neutrophil. In addition, in thepresent study craters were occasionally observed on thesurface of macrophages cocultured with apoptotic cells(Figures 2c and c0). A mechanism of ‘sinking-in craters’ forinternalization of apoptotic cells in early embryogenesis of thechick lens rudiment has been documented by Garcia-Porreroet al.35 This mechanism can be considered as a variation ofthe zipper model.33 L929sAhFas cells treated with mTNFunderwent necrosis; SEM showed that their surfaces wereirregular, pointing to loss of the integrity of the plasmamembrane (Figure 3). Macrophages attempting to engulfnecrotic cells demonstrated numerous broad membraneruffles that ended in long fine protrusions creeping up necroticmaterial (Figure 3b0). It was recently proposed that formationof the ruffles is a specific feature of macropinocytosis.36 OurSEM data revealed distinct surface features of macrophages:apoptotic target cells induce narrow pseudopods, whereasnecrotic target cells induce broad membrane ruffles that aredisproportional to the engulfed necrotic material.The microvilli present at the surface of the macrophage

disappear upon completion of the internalization process ascan be seen when comparing Figure 2b with c. Indeed onFigure 2c the macrophage is already finalizing the process ofengulfment while on Figure 2b apoptotic bodies are stillattaching to the surface of the macrophage and are in theprocess of engulfment. It has been shown that macrophagesmay possess a membrane reserve in their microvilli, whichcan be utilized for spreading and initiating cell locomotion.37,38

In addition, Dini et al.39 showed that reduction in microvilli isparalleled by the progressive decrease of binding and uptakecapacity of Kupffer cells. Another explanation may be thatthese microvilli facilitate recognition of the apoptotic cells andparticipate in attachment of apoptotic bodies to the surface ofthe macrophage during the internalization process. In thatcontext, it has been shown that microvilli express differentreceptors such as the selectins,40 the integrins a4b7 anda4b1.41,42

A second approach in the present study consisted of usingthe fluid phase markers LY and HRP as tracers, addingfunctional data in support of the SEM. A critical requirementfor any marker solute is that it can be taken up in the fluidphase, without adsorption to the cell surface. Cytochemically,the HRP reaction product was never seen attached to thesurface of the macrophages or the necrotic cells, therebyexcluding the possibility of nonspecific entrapment of HRP bythe necrotic cells (Figure 6d). This observation is in line withthe findings of Steinman and Cohn.22 These authors could notdetect HRP attached to the surface of macrophages, but incells exposed to the enzyme for just 5–10min it was easilyobserved in intracellular vesicles. These features of HRPuptake constitute an additional argument for an uptake in thefluid phase that is clearly different from uptake of materialsthat bind to the cell surface before engulfment. When LY andHRP were added to cocultures of apoptotic cells andmacrophages, apoptotic bodies inside the macrophagesformed tightly enclosed phagosomes devoid of the LY andHRP reaction products (Figures 4a, b, 5b, 6b and c). Incontrast, during internalization of TNF-induced necrotic and

Figure 5 Light microscopy. (a) Coculture of macrophages with unstimulated L929sAhFas cells in the presence of HRP (control). Note the presence of HRP positivevacuoles of uniform dimension in the macrophages (arrows). (b) Coculture of macrophages with apoptotic target cells in the presence of HRP. Notice apoptotic cells withmarginated condensed chromatin (arrowheads). Internalized apoptotic body and phagosome devoid of HRP (arrow). (c) Coculture of macrophages and TNF-inducednecrotic cells in the presence of HRP. Macrophage in the vicinity of necrotic cell (arrowhead) contains several HRP positive vacuoles of variable dimension (arrow). Notea necrotic cell (arrowhead). Scale bars 10 mm


secondary necrotic material, numerous LY and HRP positivemacropinosomes containing engulfed necrotic material wereobserved inside the macrophages (Figures 4c, 5c, 6d and e).Although cells killed by heat shock, producing unfragmentedcells, were internalized as big entities, we never observed theformation of tightly fitting phagosomes as in the case during

the uptake of apoptotic bodies (Figures 6f and g versus 6b andc). In contrast, macrophages displayed numerous ruffles atthe contact site with heat-induced necrotic cells, with theformation of macropinosomes containing both HRP reactionproduct and necrotic debris at the base of the ruffles(Figure 6g). As was the case for clearance of TNF-induced

Figure 6 Transmission electron micrographs. (a) Macrophages cocultured with unstimulated L929sAhFas cells in the presence of HRP (control). Note HRP positivevacuoles in the macrophages as a signature of constitutive endocytosis. (b) Coculture of macrophages and apoptotic cells in the presence of HRP. The phagosome ofthe macrophage (M) is devoid of HRP and contains an engulfed apoptotic body (asterisk). (c) Detail of rectangle in (b). An apoptotic body internalized by a macrophagedemonstrating the absence of HRP and formation of a tightly fitting phagosome. Note a double membrane structure (arrows) formed by the phagosomal and apoptoticbody membranes. (d, e) Macrophages cocultured with TNF-induced necrotic cells in the presence of HRP. (d) Note absence of the HRP reaction products at the surfaceof the macrophage (M) and remnants of the necrotic cell (arrow). (e) Detail of rectangle in (d). Colocalization of HRP and necrotic material in the macropinosomes(arrows). Note that necrotic material is located in the spacious macropinosomes. (f, g) Macrophages cocultured with heat-induced necrotic cells in the presence of HRP.(f) Nearly complete encircling by the macrophage of a necrotic cell corpse. (g) Detail of rectangle in (f). Note formation of ruffles at the surface of the macrophage(asterisks) with underneath the creation of spacious macropinosomes with coingestion of HRP and necrotic material (arrows). Scale bars 1 mm


necrotic cells, heat-induced necrotic material was colocalizedwith LY (Figure 4d). In addition, in order to address thecontroversial issue of cell specificity, we have used in our invitro phagocytosis assay another cell type as target cell. Forthis purpose, we applied a well-characterized model systemfor apoptotic or necrotic cell death, Jurkat E cells stimulatedwith anti-Fas or FADD-deficient Jurkat cells stimulated withhTNF43,44 (Supplementary materials and methods are avail-able online at http://www.nature.com/cdd/index.html). Wecould confirm the data obtained for L929sAhFas cells, namelythat the fluid phase marker HRP was coingested in the samemacropinosomes with necrotic material, revealing a macro-pinocytotic mechanism of internalization, while tracer wasabsent in the phagosomes containing apoptotic bodies,pointing to the zipper-like mechanism of internalization(Supplementary Figure 2 is available online at http://www.nature.com/cdd/index.html). All these findings contrast withthe results of Hoffmann et al.19 and Ogden et al.5 Theseauthors demonstrated by fluorescence microscopy thatuptake of apoptotic cells initially involves formation ofspacious phagosomes with coingestion of LY, implying amacropinocytotic mechanism for the internalization of apop-totic cellular material.5,19 This discrepancy might reflect theuse of different triggers to induce death in target cells (Fas-mediated apoptosis in the present study versus UV irradiationin Hoffmann et al.19 and Ogden et al.5), which can determinethe extent to which cells are recognized and removed byphagocytes.45 Another more likely explanation for theseopposing data might be the use of late apoptotic cells in thestudy of Ogden et al. (2001), which display the morphologicalfeatures of secondary necrosis. An additional argumentsuggesting that Ogden et al.5 used late apoptotic cells is theirprovision of evidence that C1q and mannose-binding lectin(MBL), a member of the collectin family of proteins, bind toapoptotic cells and stimulate ingestion by the macrophages. Ithas been shown that C1q and MBL bind to apoptotic cellsrather late in the cell death process; C1q binding to earlyapoptotic cells is much weaker than to late apoptotic cells, and

binding of MBL was demonstrated exclusively on lateapoptotic cells.46,47

In addition, Brouckaert et al.17 showed in the same modelsystem as used in the present study that uptake of necroticcells is quantitatively and kinetically less efficient than theuptake of apoptotic cells, and so uptake by macropinocytosisseems to be inefficient compared to zipper-like phagocytosis.As Cocco and Ucker48 reported that the macrophage cell lineJ774A.1 engulfs necrotic cells to the same extent as apoptoticcells, there is a clear need to study the phagocytosis issue inan in vivo model system.Next, we tested whether uptake of apoptotic and necrotic

cells can be differentially affected by wortmannin, a potentfungal toxin that is a specific potent inhibitor of PI3K. It wasreported that nanomolar concentrations of wortmannin do notinhibit the activities of any other known protein or lipidkinase.49 We found that wortmannin inhibits the uptake ofapoptotic cells by macrophages in a dose dependent manner,but has no effect on the uptake of necrotic cells (Figure 7).Remarkably, the extent of engulfment of apoptotic cells wasnever less than that of necrotic cells, which implies that one oftwo engulfment mechanisms for removal of apoptotic materialmight come into play, depending on the stage of apoptosis,only one of which can be blocked by wortmannin. Theseresults suggest that different biochemical processes areassociated with the two ultrastructurally different ways ofinternalizing dead cells, and that PI3K might preferentiallyparticipate in clearance of apoptotic cells. Our results are inagreement with the findings of Leverrier and Ridley,25 and Huet al.,26 who showed that blocking PI3K in macrophagesinhibits uptake of apoptotic cells. Moreover, Schpetner et al.49

showed that wortmannin did not block the uptake of thesoluble phase marker LY, which is taken up by themacropinocytotic mechanism. Their data correlate with ourfinding that wortmannin had no effect on the uptake of necroticcells, which were internalized by macropinocytosis withcoingestion of the fluid phase markers LY and HRP.If different internalization mechanisms exist, then a ques-

tion arises about the biological relevance of the mechanismsused by macrophages for the uptake of apoptotic and necroticcells. It has been shown that the mechanism of antigenuptake is decisive in determining the type of immunologicalresponses to the antigens.50 In addition, antigen presentationby macrophages was enhanced by the uptake of necrotic butnot apoptotic cells.51 In this regard, the zipper-like mechanismof phagocytosis of apoptotic cells is precise and selective,which is in agreement with indications that apoptotic cellremoval is immunologically silent. In contrast, necroticmaterial is coingested with extracellular fluid present in thevicinity of the budding macropinosomes. The mechanisms ofuptake might thus contribute to the difference in immuneresponses consequent to exposure to and uptake of apoptoticand necrotic cells.The data presented here demonstrate that macrophages

use different mechanisms to internalize apoptotic and necroticcells: zipper-like phagocytosis for apoptotic cells and macro-pinocytosis for necrotic cells (the main features are summar-ized in Table 1). The challenge ahead is to understand thebasic cellular machineries required for internalizing apoptoticand necrotic cells. This understanding might generate useful

0

5

10

15

20

25

30

35

40

0 10 50 100Concentration of wortmannin nM

% o

f u

pta

ke

Control Apoptotic Necrotic

**

##

Figure 7 Macrophages were pretreated for 30 min with the indicated doses ofwortmannin and cocultured with control (unstimulated), apoptotic or TNF-inducednecrotic cells. A two-parameter flow cytometry phagocytosis assay was used toquantify the percentage of uptake as described in ‘Materials and Methods.’Wortmannin decreased the percentage of uptake of apoptotic cells in a dosedependent manner but had no effect on the uptake of necrotic cells. Significantlydifferent (*) and not significantly different (#) from condition without wortmannin(0 nM), Po0.001 (Banferroni’s multiple comparison test). Error bars representthe SD of at least three independent experiments, each one performed induplicate


clues for unraveling the underlying control processes in-volved, and provide the basis for novel strategies tomanipulate clearance of apoptotic and necrotic cells fortherapeutic purposes.


Cells

The mouse fibrosarcoma cell line L929sA was selected for its sensitivityto the cytotoxic activity of TNF.52 L929sA were stably transfected withthe human Fas receptor to produce L929sAhFas cells, as describedpreviously.52 L929sAhFas cells were grown in DMEM mediumsupplemented with 10% fetal calf serum (FCS), penicillin (100 U/ml),streptomycin (0.1 mg/ml) and glutamax I (200mM).Mf4/4, a mouse macrophage cell line characterized by Desmedt

et al.,53 was grown in RPMI medium (Gibco/BRL, Eggenstein, Germany)supplemented with 10% FCS, penicillin (100 U/ml), streptomycin (0.1 mg/ml), glutamax I (200mM), b-mercaptoethanol (2� 10�5 mM), and sodiumpyruvate (1 mM), and kept in LPS-free conditions.

Antibodies, cytokines and reagents

mTNF was produced in Escherichia coli and purified to at least 99%homogeneity and specific activity of 6� 107 IU/ml as determined in astandardized cytotoxicity assay using L929sA cells (Department forMolecular Biomedical Research, VIB-Ghent University, Ghent, Belgium).Anti-Fas antibodies (anti-Fas, IgG3, clone 2R2) were purchased fromBioCheck GmbH (Munster, Germany). AnnexinV-FITC conjugate (BDPharmingen, San Diego, CA, USA) was used at 1 mg/ml. Cell TrackerGreen and Orange were from Molecular Probes (Eugene, Oregon, USA),and were used at 0.8 and 10 mM, respectively. Propidium iodide (BDBiosciences, San Jose CA, USA) was used at 30 mM.

Induction of apoptotic and necrotic cell death

L929sAhFas target cells were seeded at 2.5� 105 cells per well inuncoated 24-well tissue culture plates (Sarstedt, Newton, NC, USA). Thenext day, for in vitro phagocytosis assay anti-Fas antibody (250 ng/ml) wasadded for 1 h or mTNF (10 000 IU/ml) for 7 h L929sAhFas cells were killedby heat shock incubating at 551C for 30 min.54 The cells were harvested atthe indicated times and kept on ice until analysis with a FACSCalibur flowcytometer (BD Biosciences). Loss of cell membrane integrity as a measureof cell death was determined by PI fluorescence (ex 535/em 617). PS

exposure was monitored by AnnexinV-fluorescein isothiocyanate (FITC)staining (ex 494/em 518).

Caspase enzymatic activity

The fluorogenic substrate assay for caspase activity was carried out asdescribed previously.55 Briefly, 2.5� 105 cells/ml were treated with TNF oranti-Fas. Then cells were washed in cold phosphate buffer and lysed in75 ml of caspase lysis buffer (supplemented with glutathione at a finalconcentration of 1 mM). Cell debris was removed by centrifugation andcaspase activity was determined by incubating 25 ml of the soluble fractionwith 50 mM of Ac-DEVD-amc in 150ml of cell-free system buffer,containing 220mM mannitol, 68 mM sucrose, 2mM MgCl2, 2 mM NaCl,2.5 mM KH2PO4, 0.5 mM EDTA, 0.5 mM sodium pyruvate, 0.5 mML-glutamine, 10mM HEPES-NaOH pH 7.4 and 10mM dithiothreitol. Therelease of fluorescent 7-amino-4-methylcoumarin was measured for60 min at 2-min intervals by fluorometery (excitation at 360 nmand emission at 480 nm) (Cytofluor; PerSeptive Biosystems, Cambridge,MA, USA); the maximal rate of increase in fluorescence was calculated(DF/min).

Western blot analysis

Protein extracts from cell lines were prepared by lysis in Laemmli buffer,separated in 12.5% SDS-PAGE gels, and transferred to nitrocellulose.Antibodies (monoclonal antibody) against PARP and anti-murine HRP asa secondary antibody (BIOMOL Research Laboratories Inc. Technology)were used to probe the blots. Immunoreactive proteins were visualizedusing chemiluminescence and signals were captured by exposure to film(Amersham Biosciences).

In vitro phagocytosis assay

In order to quantify uptake of apoptotic and necrotic cells by macrophages,we used a previously described17 two-parameter flow cytometryphagocytosis assay, in which apoptotic, necrotic, and viable controlL929sAhFas target cells labeled with Cell Tracker Green (MolecularProbes) were incubated with a monolayer of Mf4/4 macrophages labeledwith Cell Tracker Orange (Molecular probes). Target cells were stainedwith 0.8 mM Cell Tracker Green (ex 492/em 516), and seeded at 2.5� 105

cells per well in uncoated 24-well suspension tissue culture plates.Macrophages were stained with 10 mM Cell Tracker Orange (ex 540/em566), and seeded in adherent 24-well plates at 2.5� 105 cells per well.Both target cell and macrophage cultures were incubated overnight at

Table 1 Comparative characteristics of the mechanisms of apoptotic and necrotic L929sAhFas cell uptake by the Mf4/4, a macrophage cell line

Apoptotic cells uptake(zipper-like mechanism)

Primary and secondary necrotic materialuptake (macropinocytotic mechanism)

Structure inside the Mfc Tight-fitting phagosomes Spacious macropinosomesProportionality of Mf extensions versusparticle sizec

Proportional Disproportional

Integrity of plasma membranea,c,b Required Not requiredEfficiencyb Efficient InefficientPS dependencyb Dependent DependentCoingestion of extracellular fluid andfluid phase markers (HRP and LY)a

No Yes

PI3Ka Dependent Independent

Both PS dependency and effect of PI3Kwas tested for apoptotic and TNF-induced necrotic cells but not for heat-induced necrotic cells. The data are summarized fromthe presenta and previous reports (bBrouckaert et al.22; cKrysko et al.27)


371C in 5% CO2. At the indicated time points, target cells were induced toundergo apoptosis or necrosis, washed once, and added for coincubationwith macrophages resulting in a 1 : 1 ratio of macrophages to target cells(apoptotic and TNF-induced necrotic) and 1 : 3 ratio of macrophages toheat-killed L929sAhFas cells. The cocultures were done in mediumcontaining heat-inactivated serum at 371C in 5% CO2 for 2 h, after whichthe cells were detached from the plate with Enzyme Free Cell DissociationBuffer (Gibco BRL), washed, and resuspended in ice-cold PBS. Data wasobtained on a FACScalibur flow cytometer, using Cell Quest software(Becton Dickinson, San Jose, CA, USA). In all, 3000 cells were analyzedat each time point. Control experiments in previous studies revealedthat there is no effect of stimulus, mTNF or anti-Fas antibodies, with theprotocol used.17 The percentage of double-stained macrophages out ofthe whole macrophage population measures the fraction of themacrophage population involved in phagocytosis of target cells (%phagocytosis).

Scanning electron microscopy analysis of surfacecharacteristics of macrophages duringinternalization of apoptotic and necrotic cells

Adherent cocultures of macrophages and target cells (prepared asdescribed above) were fixed by immersion for 1 h in prewarmed (371C)2% glutaraldehyde containing 0.1 M sucrose buffered with 0.1 MNa-cacodylate (pH 7.2). Following several rinses in 0.15 M Na-cacodylateHCl buffer, cocultures were postfixed for 90min in 1% OsO4 in the samebuffer at room temperature and dehydrated in a graded series of ethanol(70, 85, 95 and 100%; 10min each). Samples were viewed on a Jeol JSM-840 SEM. For quantitative control, FACS assay was performed asdescribed above.

Analysis of internalization mechanisms usingLucifer Yellow

Lucifer Yellow CH, lithium salt (Molecular Probes, 1 mg/ml) was addedfor 2 h at 371C to the nonlabeled adherent cocultures of macrophagesand target cells labeled with Cell Tracker Orange (10 mM) (prepared asdescribed above). Following several extensive rinses in PBS buffer,adherent cocultures were fixed in 3.8% paraformaldehyde beforemounting using Vectashield Mounting medium for fluorescence with DAPIH-1200 (Vector Burlingame) on coverslips. DIC and fluorescence imageswere obtained at 371C using a Leica DM IRE2 inverted microscopeequipped with a HCX PLAPO 63x/1.30 glycerin corrected 371C lens and acool snap HQ camera. The camera is controlled by the Leica ASMDWacquisition software. Lucifer yellow was measured using a standard GFPfilter set, and cell tracker red and DAPI were measured using a combinedB/G/R filter set. In order to get 3D information, images were taken atdifferent z-levels. Blind de-convolution (MLE-algorithm) and 3D rotationswere carried out using the Leica Deblur software. As primary andsecondary necrotic cells had showed background staining, we quantifiedLY fluorescence of primary and secondary necrotic cells by determiningthe average fluorescence intensity, and then all images were corrected forbackground staining using the Metamorph 5.0 software.

Ultrastructural analysis of internalizationmechanisms using horseradish peroxidase

HRP (1 mg/ml) was added for 2 h at 371C to the nonlabeled adherentcocultures of macrophages and target cells (prepared as described

above). Following several rinses in PBS buffer, adherent cocultures ofmacrophages and target cells were fixed, on ice, by immersion in 2%glutaraldehyde containing 1mM CaCl2 and 0.1 M sucrose buffered with0.1 M Na-cacodylate (pH 7.4) for 1 h. The fixative was removed withseveral washes in 0.1 M Na-cacodylate buffer (pH 7.4), and the presenceof HRP was revealed by incubation of cocultures at 371C in Tris-buffer (pH7.6) containing 0.05M DAB and 0.1% H2O2. Aminotriazole was added toblock endogenous catalase activity. After 1 h the reaction was blocked bywashing in Tris-buffer containing 7.5% sucrose, followed by osmicationovernight in 2% OsO4 in the same buffer (without sucrose). Thereafter thespecimens were dehydrated in a graded series of ethanol (70, 85, 95,100%; 10min each) (Ladd, Burlington, VT, USA), and then embedded inLX medium. Semi-thin sections of 2 mmwere contrasted with toluidine blueand examined with a Leitz Aristoplan light microscope equipped with aLeitz orthomat E photo camera. Ultrathin sections of 60 nm were cut with adiamond knife on a Reichert Jung Ultracut E ultramicrotome (Austria),mounted on formvar-coated copper grids, and stained with uranyl acetateand lead citrate. Samples were viewed on a Jeol 1200 EXII TEM at 80 kVaccelerating voltage. For quantitative control, FACS assay was performedas described above. The localization of HRP was revealed cytochemicallyby the following oxidation reaction:

HRPþH2O2 þ DAB ! HRPþ 2H2O

þoxidizedDAB ðbrownish stainingÞ

Analysis of internalization mechanisms using thePI3K inhibitor wortmannin

Wortmannin (Sigma-Aldrich N.V.) was dissolved at 200 mM in DMSO, andstored as 7-ml aliquots at -201C. Wortmannin aliquots were thawed andadded directly to Mf 4/4 to achieve the final concentration indicated in eachexperiment. As wortmannin is photosensitive and unstable in aqueoussolutions, it was routinely thawed, diluted, and added to cells within 10 min.

Acknowledgements

We thank Dominique Jacobus, Barbara De Bondt, and Hubert Stevens forexcellent technical assistance. We are grateful to Amin Bredan for copyediting the manuscript. This study was supported by Ghent UniversityGOA Grant no. 12050502.

References

1. Kerr JF, Wyllie AH and Currie AR (1972) Apoptosis: a basic biologicalphenomenon with wide-ranging implications in tissue kinetics. Br. J. Cancer 26:239–257.

2. Savill J, Dransfield I, Hogg N and Haslett C (1990) Vitronectin receptor-mediated phagocytosis of cells undergoing apoptosis. Nature 343: 170–173.

3. Savill J, Hogg N, Ren Y and Haslett C (1992) Thrombospondin cooperates withCD36 and the vitronectin receptor in macrophage recognition of neutrophilsundergoing apoptosis. J. Clin. Invest. 90: 1513–1522.

4. Brown S, Heinisch I, Ross E, Shaw K, Buckley CD and Savill J (2002)Apoptosis disables CD31-mediated cell detachment from phagocytespromoting binding and engulfment. Nature 418: 200–203.

5. Ogden CA, deCathelineau A, Hoffmann PR, Bratton D, Ghebrehiwet B, FadokVA and Henson PM (2001) C1q and mannose binding lectin engagement ofcell surface calreticulin and CD91 initiates macropinocytosis and uptake ofapoptotic cells. J. Exp. Med. 194: 781–795.

6. Gershov D, Kim S, Brot N and Elkon KB (2000) C-Reactive protein binds toapoptotic cells, protects the cells from assembly of the terminal complement


components, and sustains an antiinflammatory innate immune response:implications for systemic autoimmunity. J. Exp. Med. 192: 1353–1364.

7. Hanayama R, Tanaka M, Miwa K, Shinohara A, Iwamatsu A and Nagata S(2002) Identification of a factor that links apoptotic cells to phagocytes. Nature417: 182–187.

8. Fadeel B and Kagan VE (2003) Apoptosis and macrophage clearanceof neutrophils: regulation by reactive oxygen species. Redox. Rep. 8:143–150.

9. Fadok VA, Bratton DL, Konowal A, Freed PW, Westcott JY and Henson PM(1998) Macrophages that have ingested apoptotic cells in vitro inhibitproinflammatory cytokine production through autocrine/paracrine mechanismsinvolving TGF-beta, PGE2, and PAF. J. Clin. Invest. 101: 890–898.

10. Savill J, Dransfield I, Gregory C and Haslett C (2002) A blast from the past:clearance of apoptotic cells regulates immune responses. Nat. Rev. Immunol.2: 965–975.

11. Gao Y, Herndon JM, Zhang H, Griffith TS and Ferguson TA (1998)Antiinflammatory effects of CD95 ligand (FasL)-induced apoptosis. J. Exp.Med. 188: 887–896.

12. Chen W, Frank ME, Jin W and Wahl SM (2001) TGF-beta released byapoptotic T cells contributes to an immunosuppressive milieu. Immunity 14:715–725.

13. Denecker G, Vercammen D, Steemans M, Vanden Berghe T, Brouckaert G,Van Loo G, Zhivotovsky B, Fiers W, Grooten J, Declercq W and VandenabeeleP (2001) Death receptor-induced apoptotic and necrotic cell death: differentialrole of caspases and mitochondria. Cell. Death Differ. 8: 829–840.

14. Proskuryakov SY, Konoplyannikov AG and Gabai VL (2003) Necrosis: aspecific form of programmed cell death? Exp. Cell Res. 283: 1–16.

15. Krysko O, De Ridder L and Cornelissen M (2004) Phosphatidylserine exposureduring early primary necrosis (oncosis) in JB6 cells as evidenced byimmunogold labeling technique. Apoptosis 9: 495–500.

16. Hirt UA and Leist M (2003) Rapid, noninflammatory and PS-dependentphagocytic clearance of necrotic cells. Cell. Death Differ. 10: 1156–1164.

17. Brouckaert G, Kalai M, Krysko DV, Saelens X, Vercammen D, Ndlovu M,Haegeman G, D’Herde K and Vandenabeele P (2004) Phagocytosis of necroticcells by macrophages is phosphatidylserine dependent and does not induceinflammatory cytokine production. Mol. Biol. Cell 15: 1089–1100.

18. Swanson JA and Watts C (1995) Macropinocytosis. Trends Cell Biol. 5:424–428.

19. Hoffmann PR, deCathelineau AM, Ogden CA, Leverrier Y, Bratton DL, DalekeDL, Ridley AJ, Fadok VA and Henson PM (2001) Phosphatidylserine (PS)induces PS receptor-mediated macropinocytosis and promotes clearance ofapoptotic cells. J. Cell Biol. 155: 649–659.

20. Krysko DV, Brouckaert G, Kalai M, Vandenabeele P and D’Herde K (2003)Mechanisms of internalization of apoptotic and necrotic L929 cellsby a macrophage cell line studied by electron microscopy. J. Morphol. 258:336–345.

21. Hart SP, Smith JR and Dransfield I (2004) Phagocytosis of opsonized apoptoticcells: roles for ‘old-fashioned’ receptors for antibody and complement. Clin.Exp. Immunol. 135: 181–185.

22. Steinman RM and Cohn ZA (1972) The interaction of soluble horseradishperoxidase with mouse peritoneal macrophages in vitro. J. Cell Biol. 55: 186–204.

23. Steinman RM, Silver JM and Cohn ZA (1974) Pinocytosis in fibroblasts.Quantitative studies in vitro. J. Cell Biol. 63: 949–969.

24. Norbury CC, Chambers BJ, Prescott AR, Ljunggren HG and Watts C (1997)Constitutive macropinocytosis allows TAP-dependent major histocompatibilitycomplex class I presentation of exogenous soluble antigen by bone marrow-derived dendritic cells. Eur. J. Immunol. 27: 280–288.

25. Leverrier Y and Ridley AJ (2001) Requirement for Rho GTPases and PI 3-kinases during apoptotic cell phagocytosis by macrophages. Curr. Biol. 11:195–199.

26. Hu B, Punturieri A, Todt J, Sonstein J, Polak T and Curtis JL (2002) Recognitionand phagocytosis of apoptotic T cells by resident murine tissue macrophagesrequire multiple signal transduction events. J. Leukoc. Biol. 71: 881–889.

27. Goldman R and Bursuker I (1976) Differential effects of lectins mediatingerythrocyte attachment and ingestion by macrophages. Exp. Cell Res. 103:279–294.

28. Dini L, Pagliara P and Carla EC (2002) Phagocytosis of apoptotic cells by liver:a morphological study. Microsc. Res. Tech. 57: 530–540.

29. Geiser M (2002) Morphological aspects of particle uptake by lung phagocytes.Microsc. Res. Tech. 57: 512–522.

30. Monks J, Rosner D, Jon Geske F, Lehman L, Hanson L, Neville MC and FadokVA (2005) Epithelial cells as phagocytes: apoptotic epithelial cells are engulfedby mammary alveolar epithelial cells and repress inflammatory mediatorrelease. Cell. Death Differ. 12: 107–114.

31. Griffin Jr FM, Griffin JA, Leider JE and Silverstein SC (1975) Studies on themechanism of phagocytosis. I. Requirements for circumferential attachment ofparticle-bound ligands to specific receptors on the macrophage plasmamembrane. J. Exp. Med. 142: 1263–1282.

32. Griffin Jr FM, Griffin JA and Silverstein SC (1976) Studies on the mechanism ofphagocytosis. II. The interaction of macrophages with anti-immunoglobulin IgG-coated bone marrow-derived lymphocytes. J. Exp. Med. 144: 788–809.

33. Swanson JA and Baer SC (1995) Phagocytosis by zippers and triggers. TrendsCell Biol. 5: 89–93.

34. Giles KM, Hart SP, Haslett C, Rossi AG and Dransfield I (2000) An appetitefor apoptotic cells? Controversies and challenges. Br. J. Haematol. 109:1–12.

35. Garcia-Porrero JA, Colvee E and Ojeda JL (1984) The mechanisms of celldeath and phagocytosis in the early chick lens morphogenesis: a scanningelectron microscopy and cytochemical approach. Anat. Rec. 208: 123–136.

36. Torii I, Morikawa S, Nagasaki M, Nokano A and Morikawa K (2001) Differentialendocytotic characteristics of a novel human B/DC cell line HBM-Noda:effective macropinocytic and phagocytic function rather than scavengingfunction. Immunology 103: 70–80.

37. Follett EA and Goldman RD (1970) The occurrence of microvilli duringspreading and growth of BHK21-C13 fibroblasts. Exp. Cell Res. 59: 124–136.

38. Erickson CA and Trinkaus JP (1976) Microvilli and blebs as sources of reservesurface membrane during cell spreading. Exp. Cell Res. 99: 375–384.

39. Dini L, Ruzittu M, Carla EC and Falasca L (1998) Relationship between cellularshape and receptor-mediated endocytosis: an ultrastructural and morphometricstudy in rat Kupffer cells. Liver 18: 99–109.

40. Bruehl RE, Springer TA and Bainton DF (1996) Quantitation of L-selectindistribution on human leukocyte microvilli by immunogold labeling and electronmicroscopy. J. Histochem. Cytochem. 44: 835–844.

41. Berlin C, Bargatze RF, Campbell JJ, von Andrian UH, Szabo MC, Hasslen SR,Nelson RD, Berg EL, Erlandsen SL and Butcher EC (1995) alpha 4 integrinsmediate lymphocyte attachment and rolling under physiologic flow. Cell 80:413–422.

42. Erlandsen SL, Hasslen SR and Nelson RD (1993) Detection and spatialdistribution of the beta 2 integrin (Mac-1) and L-selectin (LECAM-1) adherencereceptors on human neutrophils by high-resolution field emission S.E.M.J. Histochem. Cytochem. 41: 327–333.

43. Kalai M, Van Loo G, Vanden Berghe T, Meeus A, Burm W, Saelens X andVandenabeele P (2002) Tipping the balance between necrosis and apoptosis inhuman and murine cells treated with interferon and dsRNA. Cell. Death Differ.9: 981–994.

44. Saelens X, Festjens N, Parthoens E, Vanoverberghe I, Kalai M, van KuppeveldF and Vandenabeele P (2005) Protein synthesis persists during necrotic celldeath. J. Cell Biol. 168: 545–551.

45. Wiegand UK, Corbach S, Prescott AR, Savill J and Spruce BA (2001) Thetrigger to cell death determines the efficiency with which dying cells are clearedby neighbours. Cell. Death Differ. 8: 734–746.

46. Nauta AJ, Trouw LA, Daha MR, Tijsma O, Nieuwland R, Schwaeble WJ,Gingras AR, Mantovani A, Hack EC and Roos A (2002) Direct binding of C1q toapoptotic cells and cell blebs induces complement activation. Eur. J. Immunol.32: 1726–1736.

47. Nauta AJ, Raaschou-Jensen N, Roos A, Daha MR, Madsen HO, Borrias-Essers MC, Ryder LP, Koch C and Garred P (2003) Mannose-binding lectinengagement with late apoptotic and necrotic cells. Eur. J. Immunol. 33: 2853–2863.

48. Cocco RE and Ucker DS (2001) Distinct modes of macrophage recognition forapoptotic and necrotic cells are not specified exclusively by phosphatidylserineexposure. Mol. Biol. Cell 12: 919–930.

49. Shpetner H, Joly M, Hartley D and Corvera S (1996) Potential sites of PI-3kinase function in the endocytic pathway revealed by the PI-3 kinase inhibitor,wortmannin. J. Cell Biol. 132: 595–605.

50. Peppelenbosch MP, DeSmedt M, Pynaert G, van Deventer SJ and Grooten J(2000) Macrophages present pinocytosed exogenous antigen via MHC class I


whereas antigen ingested by receptor-mediated endocytosis is presented viaMHC class II. J. Immunol. 165: 1984–1991.

51. Barker RN, Erwig LP, Hill KS, Devine A, Pearce WP and Rees AJ (2002)Antigen presentation by macrophages is enhanced by the uptake of necrotic,but not apoptotic, cells. Clin. Exp. Immunol. 127: 220–225.

52. Vercammen D, Brouckaert G, Denecker G, Van de Craen M, Declercq W,Fiers W and Vandenabeele P (1998) Dual signaling of the Fas receptor:initiation of both apoptotic and necrotic cell death pathways. J. Exp. Med. 188:919–930.

53. Desmedt M, Rottiers P, Dooms H, Fiers W and Grooten J (1998) Macrophagesinduce cellular immunity by activating Th1 cell responses and suppressing Th2cell responses. J. Immunol. 160: 5300–5308.

54. Cvetanovic M and Ucker DS (2004) Innate immune discrimination of apoptoticcells: repression of proinflammatory macrophage transcription is coupleddirectly to specific recognition. J. Immunol. 172: 880–889.

55. Vanden Berghe T, Denecker G, Brouckaert G, Vadimovisch Krysko D, D’HerdeK and Vandenabeele P (2004) More than one way to die: methods to determineTNF-induced apoptosis and necrosis. Methods Mol. Med. 98: 101–126.

Supplementary Information accompanies the paper on Cell Death and Differentiation website (http://www.nature.com/cdd)


General discussion andperspectives

Now stand you on the top of happy hours,And many maiden gardens yet unsetWith virtuous wish would bear your living flowers,Much liker than your painted counterfeit:

Sonnet 16, But wherefore do not you a mightier wayW Shakespear

The Cathedral, A Rodin

127

General Discussion and Perspectives

In Chapter III, several aspects of cell death regulation were addressed at three different

levels. We focused first on the intracellular level, investigating the function of mitochondria

during the course of apoptosis with regard to the mitochondrial transmembrane potential.

Secondly, regulation on the intercellular level was addressed and here the role of gap

junctional communication in apoptosis was investigated. Finally, the ultimate fate of dying

cells was considered, namely, the engulfment of apoptotic and necrotic cells.

1. Mitochondrial transmembrane potential changes support the concept of

mitochondrial heterogeneity during apoptosis

Mitochondria have been long thought to play a critical role in many processes essential for

cell survival, including energy production, redox control, and calcium homeostasis, to give

just a few examples. One of the major physiological features of mitochondria is the

generation of a transmembrane potential across the mitochondrial inner membrane. This is a

direct consequence of the biochemical reactions that constitute the respiratory chain. The

mitochondrial respiratory chain, composed of four complexes coupled to the F0F1-ATPase,

functions through the transfer of electrons from the NADH-FADH2 reducing equivalents to

molecular oxygen. The process also pumps protons across the mitochondrial inner

membrane, generating a proton gradient motive force that is largely expressed as a

electrochemical potential over the inner mitochondrial membrane between 150-180 mV

negative to the matrix, also defined as transmembrane potential ( m). This is the stored

force that underpins the production of ATP (Bras et al., 2005). Over the last decade,

mitochondria have emerged as central players in the regulation of apoptotic cell death

(Green and Kroemer, 2004; Saelens et al., 2004). The complete apoptotic program involves

energy-requiring steps, one of which is, for example, at the level of the formation of the

protein complex between Apaf-1, cytochrome c and pro-caspases (Nicotera and Melino,

2004; Skulachev, 2006).

At the beginning of this doctoral study it remained uncertain how homogenous the

mitochondrial response is to apoptotic signals as revealed by m. By using the fixable

m-sensitive probe CM-H2TMRos in combination with a DNA fluorochrome, we found that

in serum-free cultured granulosa explants a subset of polarized mitochondria exhibit normal

m until late in the apoptotic process when nuclear condensation is apparent (Krysko et al.,

2001). These data are in agreement with previous results on mitochondrial heterogeneity

during apoptosis, showing that a subset of respiring mitochondria retains cytochrome c

128 Discussion & Perspectives

function until the stage of chromatin condensation and nuclear fragmentation in a model of

granulosa explants cultured under serum-free conditions (D'Herde et al., 2000). These data

support the hypothesis that dysfunction of a fraction of mitochondria can provoke activation

of the apoptotic program while the other fraction of mitochondria, with canonically localized

cytochrome c, preserve the ability of cellular respiration and ATP production until the stage of

chromatin condensation and fragmentation. In agreement with our observation, it has been

shown that release of cytochrome c is never complete from mitochondria fraction of anti-Fas-

induced apoptotic L929 cells (Denecker et al., 2001) and from mitochondria treated in vitro

with tBid. However, our findings are in contrast to the results of Dr. Green’s lab, which used

in situ fluorescent labeling of cytochrome c with green fluorescent protein (Goldstein et al.,

2000) and tetracysteine-containing sequence (Goldstein et al., 2005) to show that in several

cell types (NCI-H1299 and HeLa treated either with UV or with actinomycin D and TNF plus

cycloheximide) all mitochondria within a single cell release all their cytochrome c within a

time span of five minutes. The authors suggested a model of a ‘single step’ release of

cytochrome c. On one hand, this difference in kinetics of the mitochondrial response to

apoptosis induction can be dependent on whether the mitochondria are organized in a

continuous or discontinuous network in a cell. Several lines of evidence have been provided

that mitochondria can form physically interconnected networks. In COS-7 cells and cardiac

myocytes, large networks of electrically connected mitochondria have been demonstrated by

targeting depolarizing stimuli to a small area of the mitochondria which led to the collapse of

m within the network over the length of tens of micrometers (Amchenkova et al., 1988; De

Giorgi et al., 2000). Moreover, evidence for the existence of a largely interconnected

mitochondrial network in HeLa cells was presented by applying the fluorescence recovery

after photobleaching technique (Rizzuto et al., 1998). Photobleaching of mito-GFP within a

small area of the cell resulted in the sequential redistribution of the mito-GFP from unaffected

portions of the cell, suggesting luminal continuity of the mitochondria with networks over

broad areas of the cell. The disintegration of this network is a first step in the cellular demise.

In this regard, downregulation of Fis (involved in fission of mitochondria) prevents cell death

(Lee et al., 2004), indicating that fission is implicated in apoptosis.

On the other hand, several studies showed that mitochondria could be morphologically and

functionally independent within cells. The functional heterogeneity of mitochondria in several

cell lines, including HeLa, COS-7, hepatocytes, cortical astrocytes and neuronal cells, was

observed with respect to m, the rate of permeability transition pore permeabilization and

the dynamics of Ca2+ sequestration (Collins et al., 2002; Collins and Bootman, 2003). Thus,

depending on whether mitochondria in a given cell represent a continuous or non-continuous

network, its response may be different to apoptotic stimuli including cytochrome c release

129

and drop of m. A non-continuous mitochondrial network could allow the possibility that

certain mitochondria are involved in apoptotic signaling, whilst others continue to supply ATP

during apoptosis.

Different responses of mitochondria within the same cell type could be related to distinct

culture conditions affecting the status of the mitochondrial networks. In this regard it was

shown in living Saccharomyces cerevisiae expressing mito-GFP, using a high-resolution

beam-scanning multifocal multiphoton confocal microscope, that the number of branch points

of the mitochondria was regulated by the growth conditions and was increased about four

times when the yeast was grown on glucose (Egner et al., 2002).

In addition, the mitochondrial response during apoptosis induction may be dependent on the

strength of the cell death stimulus. If the cell death signal (e.g. growth factor depletion) has to

accumulate over 24 hours, then mitochondrial response can be heterogeneous and

asynchronic and this response will be different from apoptosis induction, e.g., by death

domain signaling (see e.g. model of Dr. Greein’s lab Goldstein et al., 2005). In addition, the

stimuli such as actinomycin D and TNF which are used in combination with cycloheximide

applied by Goldstein et al. (Goldstein et al., 2005) in fact are very strong stimuli which

activate the extrinsic and intrinsic cell death pathways, while in our model system intrinsic

pathway gradually builds up.

In conclusion, mitochondria are dynamic organelles that can change in shape, morphology

and number and these are determined by precisely regulated rates of organelle fusion and

fission processes (Scorrano, 2005). A growing number of “mitochondrial-shaping” proteins

have been discovered, which, by regulating their level and function, control the mitochondrial

fusion and fission processes and influence apoptotic cell death (Karbowski and Youle, 2003;

Youle and Karbowski, 2005). Additional studies are warranted in order to understand how

culture conditions influence the fusion/fission balance. In this regard, one of our future plans

is to analyse cytochrome c release and the decrease of m during induction of apoptotic

cell death in suspension and in adherent primary cultures of granulosa cells, and to find out

whether the data obtained in the granulosa explant model can be transferred to primary

granulosa cells growing under different culture conditions. It would also be interesting to

investigate the respiration level of granulosa cells and to determine how long the cells could

retain the capacity of cellular respiration. Finally, we may not forget that the complexity of

mitochondrial behavior was already mirrored in the Greek name of the organelle -“mitos” for

thread, and “chondros” for gain.


2. Gap junctional communication and connexin43 expression in relation to apoptotic

cell death and survival of granulosa cells

The age-old statement “no man is an island” is applied to human society, signifying that each

of us is immeasurably affected by the activities of those around us; it goes without saying

that each of our own individual activities may have tremendous effects on the activities and

functions of society as a whole. Similarly, in the multicellular organism the maintenance of

the homeostatic balance is controlled by the global exchange of signals and information

between individual cells, thus making it possible for multicellular organism to function as a

well balanced and regulated system. One of the possible mechanisms to communicate

between neighbouring cells is via gap junctions. In the next part of our work we investigated

the impact of intercellular communication on apoptosis and survival (Krysko et al., 2004a).

We found that apoptosis induction was accompanied by an increased level of gap junctional

coupling. In addition, apoptosis was decreased by blocking the gap junctions with AGA while

increased cell coupling does not per se induce apoptosis, as was shown in the Ca2+-rescuing

protocol. It is important to point out that in the gap junction inhibition experiments we have to

rely on AGA, while the more specific connexin mimetic peptide (gap 27) was not blocking

gap junctional coupling in our experimental set-up. Moreover, Cx43 immunoreactivity was

inversely correlated with apoptosis, indicating that Cx43 plays a role in the survival process.

In addition, several attempts were done in order to explore the possible role of gap junctional

coupling in transmission of death signals from apoptotic cells to healthy cells (Cusato et al.,

2006; Udawatte and Ripps, 2005). In this regard, it was shown in one of these studies that

confluent baby hamster kidney cells stably transfected with Cx43 do spread cell death

signals from apoptotic to healthy cells after they were scrape loaded with cytochrome c

(Udawatte and Ripps, 2005). Similarly, electroporation of cytochrome c in a defined area of

confluent C6 glioma cells is accompanied by apoptosis induction in the surrounding cells in

Cx43 transfected cultures, but not in wild type cultures, indicating the necessity of gap

junctions for the propagation of cell death messages (Decrock and Leybaert, unpublished

data). It is important to point out that although the molecule of cytochrome c (MW=12,327) is

too large to traverse by itself through gap junctional channels, cell death signals may be

conveyed by other intermediate molecules, e.g., Ca2+, cAMP, or IP3, which may be generated

as a results of apoptosis induction (Krysko et al., 2005). Although there is substantial

evidence in support of the role of gap junctional coupling in propagating signals that may

modulate cell survival and cell death (reviewed in Krysko et al., 2005; Vinken et al., 2006), a

challenge we now face is to unravel the precise molecule(s) that may pass from one cell to

another via gap junctional (hemi-) channels and modulate the cell’s fate.

131

3. Phagocytosis of necrotic cells by macrophages is phosphatidylserine dependent

and does not induce inflammatory cytokine production

In many pathological situations, apoptosis and necrosis occur simultaneously. Accidental

necrosis is a passive death process which typically occurs following exposure to high

concentrations of endogenous or exogenous toxins, heat treatment, freeze thawing or other

immediately disruptive insults (Leist and Jaattela, 2001). Morphologically similar to but

mechanistically different from necrosis is necrotic-like programmed cell death (Leist and

Jaattela, 2001) which may occur in vivo under different circumstances (Barkla and Gibson,

1999; Doerfler et al., 2000; Jaattela and Tschopp, 2003; Kunstle et al., 1999). A number of

elegant studies have shown that cells dying by apoptosis are specifically recognized and

taken up by macrophages before their plasma membranes lyse (Gregory and Devitt, 2004). It

has been shown that necrotic cells also externalize PS (Krysko et al., 2004b), but it was

unclear whether PS is involved in necrotic cells clearance. Therefore we developed a

method, a dual colour flow cytometry based phagocytosis assay, which allowed us to study

the relationship between macrophages and apoptotic versus necrotic cells (Brouckaert et al.,

2004). We showed here that recombinant annexin V can inhibit the uptake of both

programmed necrotic as well as apoptotic cells; remarkably, however, cells which were dying

by programmed necrosis were taken up less efficiently than apoptotic cells. These data

demonstrate that the externalization of PS may be a common signal for the engulfment of

apoptotic and necrotic cells used by macrophages. This is in line with the study of Hirt and

Leist (Hirt and Leist, 2003), who demonstrated that PS is essential for the uptake of

programmed necrotic cells. Similarly it was shown that heat-, methanol- or ethanol-induced

necrotic peripheral blood-derived lymphocytic cells are engulfed by human monocyte-derived

macrophages in PS dependent manner (Bottcher et al., 2006). However, a study of Cocco

and Ucker (Cocco and Ucker, 2001) suggested that the uptake of heat-induced necrotic cells

is not specified exclusively by PS exposure.

Indeed, many receptors and adaptor molecules have been shown to contribute to the

recognition and uptake of apoptotic cells by phagocytes and, as we can see now, some of

them are unexpectedly shared in apoptotic and necrotic cell clearance. Questions still

remain, however, for future research as which receptors and bridging molecules also

contribute to necrotic cell uptake and how these molecules define the mechanism of uptake.


4. Analysis of NF- B activation in macrophages upon their co-culture with apoptotic

and necrotic cells

The signals emanating from necrotic cells are of great interest, since different ways of

recognition of dying cells have major implications for immunological parameters. For

instance, we provided evidence that necrotic cells are not necessarily pro-inflammatory.

Necrotic cells, like their apoptotic counterparts, fail to induce pro-inflammatory cytokine

production (Brouckaert et al., 2004). This is line with the work of Cocco and Ucker (Cocco

and Ucker, 2001) demonstrating that necrotic cells are unable to activate macrophages. NF-

B is a master regulator of genes involved in immune and inflammatory responses. Induction

of these genes typically occurs in response to stimulation of cells with conserved microbial

patterns such as LPS or with pro-inflammatory cytokines. In order to investigate whether

necrotic cells influence the activity of NF- B, we generated a macrophage cell line containing

a NF- B-dependent luciferase reporter gene and found that neither apoptotic nor primary or

secondary necrotic cells trigger the activation of this transcription factor. These data

confirmed our previously published data where we showed that neither apoptotic nor necrotic

cells during their engulfment by macrophages induce expression of pro-inflammatory

cytokines, either at the mRNA or at the protein level (Brouckaert et al., 2004). One of the

possible future directions of study is to investigate whether apoptotic cells could modulate the

NF- B activity of LPS pre-treated macrophages, as well as to explore the possible

differences in modulating the activity of cells that died by programmed necrosis and

accidental necrosis on the NF- B activity of macrophages.

5. Macrophages use different internalization mechanisms to clear apoptotic and

necrotic cells

Two major subtypes of internalization are known. One is the efficient uptake of large

particles, called phagocytosis, and the other is fluid phase uptake of small molecules, called

pinocytosis, which is further classified as micropinocytosis, the ingestion of small vesicles via

clathrin-coated pits (85-110 nm in diameter) or macropinocytosis, ingestion of pinosomes of

>0.2 µM diameter (Swanson and Baer, 1995; Swanson and Watts, 1995). It has been known

for 25 years from ultrastructural studies that the mode of internalization depends on which

phagocytic receptor is mediating internalization. For example, during Fc R-mediated

phagocytosis, sequential interaction of IgG-coated particles with receptors stimulates the

extension of pseudopods from the macrophage surface that surround the target (Griffin et al.,

1975; Griffin et al., 1976; Swanson and Baer, 1995). The advance of the pseudopodia is

133

limited by the availability of ligands, which results in a close fitting phagosome with almost no

solute taken up during internalization. This observation led to the “zipper-like” hypothesis of

phagocytosis. Important implications of this “zipper-like” interaction between receptors and

ligands are that phagocytosis is locally and temporally controlled, that it is restricted to the

adhering particle, that the engulfing pseudopod follows the contour of the particle, and that

both surfaces are tightly apposed. In contrast, during phagocytosis of complement-coated

particles through CR3- m 2 integrin few protrusions form, and the particles appear to sink

into the cells (Allen and Aderem, 1996; Kaplan and Bertheussen, 1977). Moreover, it has

been shown that the morphological differences between uptake via FcR and CR are

accompanied by distinct molecular mechanisms and diverse immunological consequences

(Caron and Hall, 1998; Giles et al., 2000). In another form of phagocytosis, bacteria release

protein messengers into the cytosol of targeted cells, causing cell surface ruffling and

formation of macropinosomes-like membrane invaginations (Alpuche-Aranda et al., 1994).

These membrane invaginations form spacious macropinosomes, wherein large amounts of

external solute are internalized along with the bacterium (Swanson and Baer, 1995;

Swanson and Watts, 1995; Torii et al., 2001). Through several mechanisms for internalizing

microorganisms have been proposed (Rittig et al., 1998; Rittig et al., 1999), information on

the mechanisms for internalizing apoptotic and necrotic cells is sparse.

Therefore, the purpose of our work was to investigate the mechanisms of internalization of

apoptotic versus necrotic cells in an in vitro phagocytosis assay. Our morphological and

functional data (Krysko et al., 2003; Krysko et al., 2006) revealed that apoptotic cells are

taken up by complete engulfment of apoptotic bodies as single entities, forming tight-fitting

phagosomes without coingestion of fluid phase markers, thus resembling the “zipper-like”

mechanism of internalization (Fig. 1).

In contrast, we observed that primary and secondary necrotic cells are engulfed by formation

of spacious macropinosomes which were accompanied by the formation of multiple ruffles by

the ingesting macrophages and by co-uptake of fluid phase tracers (Fig. 1). Moreover, our

conclusion is independent of the cell type used as a target (Krysko et al., 2006). However,

this view is at odds with several reports (Fiorentini et al., 2001; Hoffmann et al., 2001;

Jersmann et al., 2003; Ogden et al., 2001). In this regard it has been shown by light and

immunofluorescence microscopy that uptake of apoptotic cells initially involves formation of

spacious phagosomes and the concurrent ingestion of extracellular fluid (Hoffmann et al.,

2001; Ogden et al., 2001). Fiorentini et al. demonstrated that a bacterial protein toxin, the

Escherichia coli cytotoxic necrotizing factor 1, allowed epithelial cells to engulf apoptotic cells

by a macropinocytic mechanism (Fiorentini et al., 2001). Importantly, the results of this study

should be interpreted with care, because it has been shown that several bacterial pathogens


use macropinocytosis as a mechanism of cell invasion (Alpuche-Aranda et al., 1994; Rittig et

al., 1999; Swanson and Watts, 1995).

Moreover, Jersmann et al. (Jersmann et al., 2003) demonstrated that bovine fetuin and

human alpha2-HS glycoprotein significantly increased the engulfment of apoptotic cells as

well as labeled dextran 70000 by monocyte-derived macrophages. They reported that the

uptake of dextran occurs by macropinocytosis and extrapolated these findings to the uptake

of apoptotic cells, concluding that apoptotic cells might be also internalized by the same

macropinocytotic mechanism. However, it is important to note that neither of these studies

(Fiorentini et al., 2001; Jersmann et al., 2003) used fluid phase tracers, which are necessary

for distinguishing macropinocytosis from “zipper-like” phagocytosis. Moreover, the

contradiction may be due to the use of cells that were already in a late apoptotic stage and

exhibiting morphological features of secondary necrosis (Fiorentini et al., 2001; Hoffmann et

al., 2001; Jersmann et al., 2003; Ogden et al., 2001).

Figure 1. A scheme of “zipper-like” and macropinocytotic mechanisms used by macrophages to engulf apoptotic and necrotic cells, respectively. Apoptotic cells are internalized by formation of tight-fitting phagosomes without coingestion of lucifer yellow, a fluid phase marker. In contrast, necrotic cell material is internalized by macrophages with formation of spacious macropinosomes and co-localization of lucifer yellow (a green staining).

Phagocytosis“zipper”- like mechanism

Macropinocytotic mechanism

Apoptosis Necrosis

Secondarynecrosis

135

In a recent study, Xu et al. (Xu et al., 2006) demonstrated that pre-treatment of M-CSF-

stimulated human mononuclear cells with DMA (5-(N, N-Dimethyl)amiloride hydrochloride)

inhibits not only the uptake of apoptotic cells, but also the uptake of lucifer yellow. Thus, the

authors suggested on the basis of FACS analysis that apoptotic cells are internalized by

macropinocytosis. However, it is important to mention that the use of fluid phase tracers is

not sufficient by itself. Microscopic examination (e.g. fluorescence or/and transmission

electron microscopy) is needed to co-localize fluid phase markers with ingested material

within the same compartment, which would indicate macropinocytosis (Krysko et al., 2006).

This was not done in the study of Xu et al. (Xu et al., 2006). Electron transmission

microscopy is the method of choice for determining whether ingested material is localized in

spacious macropinosomes or in tightly fitting phagosomes (Krysko et al., 2003).

If different modes for internalization of apoptotic and necrotic cells do exist, then questions

arise about the existence of differences at the molecular level, and the biological

consequences of those different mechanisms. In this context, it was reported that C. elegans

uses the same set of engulfment genes for removal of both apoptotic and necrotic cell

corpses (Chung et al., 2000), which conflicts with data on internalization mechanisms

(Krysko et al., 2003; Krysko et al., 2006). This apparent paradox may be due to recognition

of apoptotic and necrotic cells by distinct mechanisms (Cocco and Ucker, 2001).

Interestingly, Dictyostelium, a free-living soil amoeba and a professional phagocyte, can

internalize fluid and particles at higher rates than macrophages and neutrophils. It has been

widely used in studies to investigate molecular mechanisms of phagocytosis and

macropinocytosis, which indicates that some molecular mechanisms of macropinocytosis

and phagocytosis are indeed distinct (Cardelli, 2001).

Interestingly, necrotic nuclei in close vicinity to the plasma membrane of the macrophages

very often were observed by TEM (Krysko et al., 2003), which were hardly seen engulfed by

the macrophages. Recently, it was demonstrated that the nuclei which are expelled from the

erythroid precursor cells are engulfed by the macrophages in a PS-dependent manner

(Yoshida et al., 2005). Although the authors suggested that released nuclei quickly expose

PS on their surface, it is not possible to completely exclude the fact that PS exposure

happens on the surface of a thin rim of the plasma membrane which remains around the

nuclei during enucleation. Thus, one of our future plans is to analyse whether necrotic nuclei

can be engulfed by macrophages and whether it happens in a PS-dependent manner. To

better understand the mechanisms of clearance of nuclear material derived from apoptotic

and necrotic cells is of utmost importance, since deficient clearance of nuclear components

is involved in the generation of anti-DNA antibodies during systemic lupus erythematosus

(Munoz et al., 2005).


Next we demonstrated that the PI3K inhibitor wortmannin significantly reduces the uptake of

apoptotic cells, while it has no effect on the clearance of necrotic cells. Although several

limitations may exist to use wortmannin in vivo, the idea that it is possible to differentially

regulate the uptake of dying cells in a mixed population of apoptotic and necrotic cells, in

particular towards the clearance of necrotic cells, may have several important clinical

applications, e.g., in immunotherapy of cancer. Although there are some discrepancies

between in vitro and in vivo studies on the question whether apoptotic or necrotic cells are

better suitable for use in vaccination against cancer (as discussed in the introduction of this

thesis), it was demonstrated that engulfment of necrotic tumour cells but not apoptotic

tumour cells induced in vitro DC maturation (Basu et al., 2000; Kacani et al., 2005; Sauter et

al., 2000; Wong et al., 2005), antigen presentation by macrophages (Barker et al., 2002), and

endothelial cell activation (Chen et al., 2006), suggesting that necrotic tumour cells may be

used in development of anti-tumour vaccines. It is known that frequently combined

chemotherapy produces in vivo mixed populations with an enormous amount of apoptotic

and necrotic cells. The uptake of apoptotic cells is frequently but not unanimously viewed as

immunologically silent (leading to ignorance by the immune system) or even as tolerogenic

(actively down-regulating the specific anti-tumour immune response) (Steinman and

Mellman, 2004; Zitvogel et al., 2004). Therefore, the development of strategies which will

allow co-application blockage of the uptake of apoptotic tumour cells on the one hand, and

stimulation of necrotic cells engulfment on the other hand, may provide an additional

mechanism for enhancement of immunogenicity of necrotic cells.

Indeed, knowledge gathered in the present study on different mechanisms controlling cell

death, namely on intracellular and intercellular levels, as well as understanding of molecular

mechanisms of dead cell disposal can contribute to the development of strategies to

manipulate and regulate cell death processes on multiple levels. This may lead ultimately to

the discovery of new medical treatments for inflammatory and autoimmune diseases and to

the improvement of anti-cancer vaccines.

“I do not know what I may appear to the world; but to myself I seem to have been only like a boy playing on the seashore, and diverting myself in now and then finding a smoother pebble or a prettier shell than ordinary, while the great ocean of truth lay all undiscovered before me.”

Sir Isaac Newton

137

References

Allen LA, Aderem A. 1996. Molecular definition of distinct cytoskeletal structures involved

in complement- and Fc receptor-mediated phagocytosis in macrophages. J Exp Med

184(2):627-637.

Alpuche-Aranda CM, Racoosin EL, Swanson JA, Miller SI. 1994. Salmonella stimulate

macrophage macropinocytosis and persist within spacious phagosomes. J Exp Med

179(2):601-608.

Amchenkova AA, Bakeeva LE, Chentsov YS, Skulachev VP, Zorov DB. 1988. Coupling

membranes as energy-transmitting cables. I. Filamentous mitochondria in fibroblasts

and mitochondrial clusters in cardiomyocytes. J Cell Biol 107(2):481-495.

Barker RN, Erwig LP, Hill KS, Devine A, Pearce WP, Rees AJ. 2002. Antigen presentation

by macrophages is enhanced by the uptake of necrotic, but not apoptotic, cells. Clin

Exp Immunol 127(2):220-225.

Barkla DH, Gibson PR. 1999. The fate of epithelial cells in the human large intestine.

Pathology 31(3):230-238.

Basu S, Binder RJ, Suto R, Anderson KM, Srivastava PK. 2000. Necrotic but not apoptotic

cell death releases heat shock proteins, which deliver a partial maturation signal to

dendritic cells and activate the NF-kappa B pathway. Int Immunol 12(11):1539-1546.

Bottcher A, Gaipl US, Furnrohr BG, Herrmann M, Girkontaite I, Kalden JR, Voll RE. 2006.

Involvement of phosphatidylserine, alphavbeta3, CD14, CD36, and complement C1q

in the phagocytosis of primary necrotic lymphocytes by macrophages. Arthritis

Rheum 54(3):927-938.

Bras M, Queenan B, Susin SA. 2005. Programmed cell death via mitochondria: different

modes of dying. Biochemistry (Mosc) 70(2):231-239.




Mol Biol Cell 15(3):1089-1100.

Cardelli J. 2001. Phagocytosis and macropinocytosis in Dictyostelium: phosphoinositide-

based processes, biochemically distinct. Traffic 2(5):311-320.

Caron E, Hall A. 1998. Identification of two distinct mechanisms of phagocytosis controlled

by different Rho GTPases. Science 282(5394):1717-1721.

Chen Q, Stone PR, McCowan LM, Chamley LW. 2006. Phagocytosis of necrotic but not

apoptotic trophoblasts induces endothelial cell activation. Hypertension 47(1):116-

121.

Chung S, Gumienny TL, Hengartner MO, Driscoll M. 2000. A common set of engulfment

genes mediates removal of both apoptotic and necrotic cell corpses in C. elegans. Nat

Cell Biol 2(12):931-937.



Cell 12(4):919-930.

Collins TJ, Berridge MJ, Lipp P, Bootman MD. 2002. Mitochondria are morphologically and

functionally heterogeneous within cells. Embo J 21(7):1616-1627.

Collins TJ, Bootman MD. 2003. Mitochondria are morphologically heterogeneous within

cells. J Exp Biol 206(Pt 12):1993-2000.

Cusato K, Ripps H, Zakevicius J, Spray DC. 2006. Gap junctions remain open during

cytochrome c-induced cell death: relationship of conductance to 'bystander' cell

killing. Cell Death Differ.


De Giorgi F, Lartigue L, Ichas F. 2000. Electrical coupling and plasticity of the mitochondrial

network. Cell Calcium 28(5-6):365-370.

Denecker G, Vercammen D, Steemans M, Vanden Berghe T, Brouckaert G, Van Loo G,

Zhivotovsky B, Fiers W, Grooten J, Declercq W, Vandenabeele P. 2001. Death

receptor-induced apoptotic and necrotic cell death: differential role of caspases and

mitochondria. Cell Death Differ 8(8):829-840.

D'Herde K, De Prest B, Mussche S, Schotte P, Beyaert R, Coster RV, Roels F. 2000.

Ultrastructural localization of cytochrome c in apoptosis demonstrates mitochondrial

heterogeneity. Cell Death Differ 7(4):331-337.

Doerfler P, Forbush KA, Perlmutter RM. 2000. Caspase enzyme activity is not essential for

apoptosis during thymocyte development. J Immunol 164(8):4071-4079.

Egner A, Jakobs S, Hell SW. 2002. Fast 100-nm resolution three-dimensional microscope

reveals structural plasticity of mitochondria in live yeast. Proc Natl Acad Sci U S A

99(6):3370-3375.

Fiorentini C, Falzano L, Fabbri A, Stringaro A, Logozzi M, Travaglione S, Contamin S,

Arancia G, Malorni W, Fais S. 2001. Activation of rho GTPases by cytotoxic

necrotizing factor 1 induces macropinocytosis and scavenging activity in epithelial

cells. Mol Biol Cell 12(7):2061-2073.

Giles KM, Hart SP, Haslett C, Rossi AG, Dransfield I. 2000. An appetite for apoptotic cells?

Controversies and challenges. Br J Haematol 109(1):1-12.

Goldstein JC, Munoz-Pinedo C, Ricci JE, Adams SR, Kelekar A, Schuler M, Tsien RY,

Green DR. 2005. Cytochrome c is released in a single step during apoptosis. Cell

Death Differ 12(5):453-462.

Goldstein JC, Waterhouse NJ, Juin P, Evan GI, Green DR. 2000. The coordinate release of

cytochrome c during apoptosis is rapid, complete and kinetically invariant. Nat Cell

Biol 2(3):156-162.

Green DR, Kroemer G. 2004. The pathophysiology of mitochondrial cell death. Science

305(5684):626-629.

Gregory CD, Devitt A. 2004. The macrophage and the apoptotic cell: an innate immune

interaction viewed simplistically? Immunology 113(1):1-14.

Griffin FM, Jr., Griffin JA, Leider JE, Silverstein SC. 1975. Studies on the mechanism of

phagocytosis. I. Requirements for circumferential attachment of particle-bound

ligands to specific receptors on the macrophage plasma membrane. J Exp Med

142(5):1263-1282.

Griffin FM, Jr., Griffin JA, Silverstein SC. 1976. Studies on the mechanism of phagocytosis.

II. The interaction of macrophages with anti-immunoglobulin IgG-coated bone

marrow-derived lymphocytes. J Exp Med 144(3):788-809.



Hoffmann PR, deCathelineau AM, Ogden CA, Leverrier Y, Bratton DL, Daleke DL, Ridley

AJ, Fadok VA, Henson PM. 2001. Phosphatidylserine (PS) induces PS receptor-

mediated macropinocytosis and promotes clearance of apoptotic cells. J Cell Biol

155(4):649-659.

Jaattela M, Tschopp J. 2003. Caspase-independent cell death in T lymphocytes. Nat Immunol

4(5):416-423.

Jersmann HP, Dransfield I, Hart SP. 2003. Fetuin/alpha2-HS glycoprotein enhances

phagocytosis of apoptotic cells and macropinocytosis by human macrophages. Clin

Sci (Lond) 105(3):273-278.

139

Kacani L, Wurm M, Schwentner I, Andrle J, Schennach H, Sprinzl GM. 2005. Maturation of

dendritic cells in the presence of living, apoptotic and necrotic tumour cells derived

from squamous cell carcinoma of head and neck. Oral Oncol 41(1):17-24.

Kaplan G, Bertheussen K. 1977. The morphology of echinoid phagocytes and mouse

peritoneal macrophages during phagocytosis in vitro. Scand J Immunol 6(12):1289-

1296.

Karbowski M, Youle RJ. 2003. Dynamics of mitochondrial morphology in healthy cells and

during apoptosis. Cell Death Differ 10(8):870-880.

Krysko DV, Brouckaert G, Kalai M, Vandenabeele P, D'Herde K. 2003. Mechanisms of

internalization of apoptotic and necrotic L929 cells by a macrophage cell line studied

by electron microscopy. J Morphol 258(3):336-345.

Krysko DV, Denecker G, Festjens N, Gabriels S, Parthoens E, D'Herde K, Vandenabeele P.

2006. Macrophages use different internalization mechanisms to clear apoptotic and

necrotic cells. Cell Death Differ.

Krysko DV, Leybaert L, Vandenabeele P, D'Herde K. 2005. Gap junctions and the

propagation of cell survival and cell death signals. Apoptosis 10(3):459-469.

Krysko DV, Mussche S, Leybaert L, D'Herde K. 2004a. Gap junctional communication and

connexin43 expression in relation to apoptotic cell death and survival of granulosa

cells. J Histochem Cytochem 52(9):1199-1207.

Krysko DV, Roels F, Leybaert L, D'Herde K. 2001. Mitochondrial transmembrane potential

changes support the concept of mitochondrial heterogeneity during apoptosis. J

Histochem Cytochem 49(10):1277-1284.

Krysko O, De Ridder L, Cornelissen M. 2004b. Phosphatidylserine exposure during early



Kunstle G, Hentze H, Germann PG, Tiegs G, Meergans T, Wendel A. 1999. Concanavalin A

hepatotoxicity in mice: tumor necrosis factor-mediated organ failure independent of

caspase-3-like protease activation. Hepatology 30(5):1241-1251.

Lee YJ, Jeong SY, Karbowski M, Smith CL, Youle RJ. 2004. Roles of the mammalian

mitochondrial fission and fusion mediators Fis1, Drp1, and Opa1 in apoptosis. Mol

Biol Cell 15(11):5001-5011.



Munoz LE, Gaipl US, Franz S, Sheriff A, Voll RE, Kalden JR, Herrmann M. 2005. SLE--a

disease of clearance deficiency? Rheumatology (Oxford) 44(9):1101-1107.

Nicotera P, Melino G. 2004. Regulation of the apoptosis-necrosis switch. Oncogene

23(16):2757-2765.

Ogden CA, deCathelineau A, Hoffmann PR, Bratton D, Ghebrehiwet B, Fadok VA, Henson

PM. 2001. C1q and mannose binding lectin engagement of cell surface calreticulin

and CD91 initiates macropinocytosis and uptake of apoptotic cells. J Exp Med

194(6):781-795.

Rittig MG, Burmester GR, Krause A. 1998. Coiling phagocytosis: when the zipper jams, the

cup is deformed. Trends Microbiol 6(10):384-388.

Rittig MG, Wilske B, Krause A. 1999. Phagocytosis of microorganisms by means of

overshooting pseudopods: where do we stand? Microbes Infect 1(9):727-735.

Rizzuto R, Pinton P, Carrington W, Fay FS, Fogarty KE, Lifshitz LM, Tuft RA, Pozzan T.

1998. Close contacts with the endoplasmic reticulum as determinants of mitochondrial

Ca2+ responses. Science 280(5370):1763-1766.

Saelens X, Festjens N, Vande Walle L, van Gurp M, van Loo G, Vandenabeele P. 2004.

Toxic proteins released from mitochondria in cell death. Oncogene 23(16):2861-2874.


Sauter B, Albert ML, Francisco L, Larsson M, Somersan S, Bhardwaj N. 2000. Consequences

of cell death: exposure to necrotic tumor cells, but not primary tissue cells or apoptotic

cells, induces the maturation of immunostimulatory dendritic cells. J Exp Med

191(3):423-434.

Scorrano L. 2005. Proteins that fuse and fragment mitochondria in apoptosis: con-fissing a

deadly con-fusion? J Bioenerg Biomembr 37(3):165-170.

Skulachev VP. 2006. Bioenergetic aspects of apoptosis, necrosis and mitoptosis. Apoptosis.

Steinman RM, Mellman I. 2004. Immunotherapy: bewitched, bothered, and bewildered no

more. Science 305(5681):197-200.

Swanson JA, Baer SC. 1995. Phagocytosis by zippers and triggers. Trends Cell Biol 5(3):89-

93.

Swanson JA, Watts C. 1995. Macropinocytosis. Trends Cell Biol 5(11):424-428.

Torii I, Morikawa S, Nagasaki M, Nokano A, Morikawa K. 2001. Differential endocytotic

characteristics of a novel human B/DC cell line HBM-Noda: effective macropinocytic

and phagocytic function rather than scavenging function. Immunology 103(1):70-80.

Udawatte C, Ripps H. 2005. The spread of apoptosis through gap-junctional channels in BHK

cells transfected with Cx32. Apoptosis 10(5):1019-1029.

Vinken M, Vanhaecke T, Papeleu P, Snykers S, Henkens T, Rogiers V. 2006. Connexins and

their channels in cell growth and cell death. Cell Signal 18(5):592-600.

Wong OH, Huang FP, Chiang AK. 2005. Differential responses of cord and adult blood-

derived dendritic cells to dying cells. Immunology 116(1):13-20.

Xu W, Roos A, Schlagwein N, Woltman AM, Daha MR, van Kooten C. 2006. IL-10-

producing macrophages preferentially clear early apoptotic cells. Blood.

Yoshida H, Kawane K, Koike M, Mori Y, Uchiyama Y, Nagata S. 2005. Phosphatidylserine-

dependent engulfment by macrophages of nuclei from erythroid precursor cells.

Nature 437(7059):754-758.

Youle RJ, Karbowski M. 2005. Mitochondrial fission in apoptosis. Nat Rev Mol Cell Biol

6(8):657-663.

Zitvogel L, Casares N, Pequignot MO, Chaput N, Albert ML, Kroemer G. 2004. Immune

response against dying tumor cells. Adv Immunol 84:131-179.

Addenda

Addendum I: Transcriptional profiling of macrophages co-cultured with apoptotic and necrotic cells

141

TRANSCRIPTIONAL PROFILING OF MACROPHAGES CO-CULTURED

WITH APOPTOTIC AND NECROTIC CELLS

Abstract

Removal of apoptotic and necrotic cells by phagocytes is essential for the maintenance of

tissue homeostasis. Data on the impact of apoptotic or necrotic cells on the transcriptional

level of macrophages are sparse and therefore in the present study we applied a cDNA

macroarray analysis to reveal transcriptional profiles of differently activated macrophage

populations (M0 “naïve” macrophages; M1 “killer” macrophages or M2 “healer”

macrophages) after confrontation with either apoptotic or necrotic cells in an in vitro

phagocytosis assay. Our in vitro phagocytosis assay used a mouse thyoglycollate-elicited

peritoneal macrophages, and murine L929 cells that in a Tet-On inducible way die either by

apoptosis or by necrosis. The macroarray data confirm the existence of specific transcript

profile in differentially activated macrophages representing three major functional classes of

macrophages: M0 (naïve), M1 (killer) and M2 (healer) macrophages. When these

differentially activated macrophages are exposed to apoptotic cells, they exhibit a differential

transcription profile. Exposure to apoptotic cells specifically favours an M2-like transcription

profile, suggesting that apoptotic cells promote the regeneration function of macrophages.

Importantly, neither macrophage type specifically responded to necrotic cells, although their

uptake was more efficient than in case of apoptotic cells. Although we fully realize that our

macroarray cDNA hybridization results should be validated by real-time PCR, we believe that

the conclusions formulated in the present study form a firm basis for formulating challenging

working hypotheses and exploring further experiments.

Introduction

Clearance of dead cells is fundamental to animal development, immune function and cellular

homeostasis. Our previous study suggests that neither apoptotic nor necrotic L929 cells

modulate the expression of pro-inflammatory cytokines on the mRNA (TNF- , IL-6, IFN-

and protein level (TNF- , IL-6) (Brouckaert et al., 2004), nor do they increase NF- B

activation, a critical transcription activator of pro-inflammatory genes, in macrophages.

Although we did not see any effect of either apoptotic or necrotic cells on the induction of IL-

6, TNF and IFN- in macrophages (Brouckaert et al., 2004), it is conceivable that dying cells

may have broader effects in these macrophages, such as modulation of genes involved in

tissue repair, antigen presentation, or anti-tumour immunity. Data on the impact of apoptotic

142 Addendum I

or necrotic cells on the transcriptional level of macrophages are sparse and therefore we

applied cDNA macroarray analysis to reveal transcriptional profiles in macrophages after

confrontation with either apoptotic or necrotic cells. For transcription analysis, we made use

of a membrane-based cDNA array system with radioactive detection, developed by Tom

Boonefaes in the Unit for Molecular Immunology, Department for Molecular Biomedical

Research (Prof. Dr. Johan Grooten, VIB, Ghent University). This medium-scale array

consists of a set of 562 genes with relevance to macrophage biology (list of all genes, which

were tested see in Table 5), including cytokines and cytokine receptors, chemokines and

chemokine receptors, macrophage differentiation markers, pattern recognition and

scavenger receptors, adhesion and matrix remodelling molecules, transcriptional regulators,

feedback regulators of signalling pathways and several enzymes involved in macrophage

effector functions.

For the cell death system we used L929 cells that in a Tet-On inducible way (Gossen and

Bujard, 1992; Gossen et al., 1995) die either by apoptosis or by necrosis (Vanden Berghe et

al. 2004). L929rTA-FKBP-FADD-DED and L929rTA-FKBP-FADD, stably transfected cells,

contain the Tet-On regulated transactivator (TA) and were transfected with deletion mutants

of the cell death-inducing adaptor molecule, Fas associated death domain (FADD), preceded

by a three-fold repeat of a dimerization motif, FKBP (FK506-binding protein). This motif

allows dimerization by administration of a ligand (AP1510, Ariad Pharmaceuticals Inc.

Cambridge, MA). The FADD and the FADD-DED are cloned after a minimal cytomegalovirus

(CMV) promotor proceeded by the Tet-On regulated operator sequence. FADD is an adaptor

molecule that consists of two death domain motifs, a death domain (DD) and a death effector

domain (DED) (Aravind et al., 1999; Muzio et al., 1996). The DED motif of FADD is involved

in the recruitment of caspase-8 and leads to apoptosis (Muzio et al., 1996), while the death

domain of FADD in certain cells recruits the RIP1 kinase and leads to necrotic cell death

(Vanden Berghe et al., 2004). When the wild type FADD or FADD-DED are induced and

dimerized in the stably transfected L929rTA-FKBP-FADD or L929rTA-FKBP-FADD-DED

cells, necrotic or apoptotic cell death is observed, respectively. The morphology and the

biochemistry of these types of cell death in this inducible system have been described

previously (Vanden Berghe et al., 2004; Tom Vanden Berghe, doctoral thesis, 2005). We

have chosen for this inducible system to avoid effects due to transfer of the death-inducing

stimulus such as Fas-ligand or TNF. As a scavenger cells, we used thyoglycollate-elicited

peritoneal macrophages, which were either left untreated (“M0” macrophages) or pretreated

with IFN- and LPS (“M1” macrophages) or with IL-4 and IL-10 (“M2” macrophages).

Macroarray analysis 143

Figure 1. M1 and M2 macrophages, the extreme of a continuum. Essential properties of polarized

macrophage populations are shown. For M1 cells, molecules induced by IFN- and LPS are shown in green. For M2 cells, molecules induced by IL-4 and IL-13 are shown in yellow, those induced by IL-10 in red and those induced both by IL-4 and IL-13, and IL-10, in blue. Macrophages exposed to the

classic activation signals, IFN- and LPS, express opsonic receptors (e.g. Fc RIII), whereas M2 macrophages are characterized by abundant levels of non-opsonic receptors (e.g. MR). Components of the IL-1 system are differentially regulated in activated macrophage populations. IL-4 and IL-13 induce expression of the IL-1 type II decoy receptor, and IL-1R accessory protein (IL-1RacP). IL-4 and IL-13 induce IL-1ra production and inhibit IL-1. Therefore, pro- and anti-inflammatory components of the IL-1 system are co-ordinately regulated by signals that polarize macrophages in a type 1 or type 2 direction. IL-10 up regulates the CC chemokine receptors CCR1, CCR2 and CCR5. By contrast, CXCR2 and CXCR4 are partially downregulated under the same conditions. M1 macrophages characterized by increased production of pro-inflammatory cytokines (TNF, IL-1, IL-6, IL-12), inducible nitric-oxide synthase (iNOS), reactive oxygen species production (ROS) while M2 macrophages demonstrated increased expression of arginase with generation of ornithine and polyamines. Adapted from Mantovani et al. (Mantovani et al., 2002).

TLR2, TLR4

FCg-RI,II,III

CD80, CD86

TNF

IL-1

IL-6

IL-12

Type I IFN

Scavenger receptors A,B

MR

CD14

Cytokines

Membrane receptors

Chemokines

Chemokine receptors

Effector molecules

IL-1ra

Cytokine receptorsIL-1 R type I

CXL2, CXL3,CCL4, CCL5

CXL6

CXCL9, CXCL10,CXCL11

CXL18

CXL16

CCR7

CCR2

iNOS

ROI

Arginase

IFN- and LPS IL-10 IL-4 and IL-13, IL-10

M1 M2

Cd163

FC -RII

IL-10

Decoy IL-1 R type II

CCL17, CCL22

CCL24

CXCR1, CXCR2

IL-4 and IL-13

144 Addendum I

These pretreatments lead to the development of distinct phenotypes and physiological

activities in these macrophages. M1 are the classically activated macrophages that show

increased production of pro-inflammatory cytokines (TNF, IL-1, IL-6, IL-12), inducible nitric-

oxide synthase (iNOS), reactive oxygen species production (ROS) and increased ability of

antigen presentation (Gordon and Taylor, 2005). Thus, these classically activated M1

macrophages are potent effector (‘killer’) cells, which destroy microorganisms and tumour

cells and produce plentiful amounts of pro-inflammatory cytokines.

By contrast, M2 macrophages result from an alternative activation characterized by

increased expression of mannose receptor, dectin 1 and arginase with generation of

ornithine and polyamines (Gordon and Taylor, 2005; Mantovani et al., 2004). M1 and M2

macrophages differ in terms of receptor expression, effector function and cytokine and

chemokine production (Fig. 1) (Mantovani et al., 2004). In addition, M2 macrophages, in

contrast to M1, promote angiogenesis, tissue remodelling, and repair. Therefore, they are

often called ‘healer’ macrophages. Alternatively activated macrophages (M2) are found

during the healing phase of acute inflammatory reactions in chronic inflammation diseases,

such as rheumatoid arthritis and psoriasis (Gratchev et al., 2001).

In the present study we undertook a cDNA macroarray analysis in order to unravel the

influence of apoptotic and necrotic cells on the transcriptional profile of differently activated

macrophage populations: M0 “naïve” macrophages, M1 “killer” macrophages or M2 “healer”

macrophages in an in vitro phagocytosis assay.


Isolation and activation of peritoneal macrophages

C57BL/6N@CrlBR mice (Charles River France) were treated with 2 ml intraperitoneal

injection of 3% thyoglycollate broth (Sigma, St. Louis, MO), 4 days later sacrificed and

peritoneal macrophages were isolated by peritoneal lavage with cold sterile LPS-free

phosphate buffered saline (PBS, BioWhittaker, Cambrex Bio Science, Verviers, Belgium).

Cells were washed, re-suspended in RPMI 1640 medium [(Gibco/BRL, Eggenstein,

Germany) supplemented with 10% FCS, penicillin (100 U/ml), streptomycin (0.1 mg/ml),

glutamax I (200 µM), -mercaptoethanol (2 x 10-5 mM), and sodium pyruvate (1 mM)],

seeded at 106 cells per well in 12-well tissue culture plates and incubated at 37 °C 5% CO2,

in a humidified chamber to allow macrophages to attach. After 30 min, non-adherent cells

were removed and sticking macrophages were left overnight in the incubator, followed by

additional wash steps to remove cells that had lost adherence (mostly myeloid dendritic cells,

personal communication by Boonefaes and Grooten). Due to the high number of cells


seeded, a nearly confluent monolayer of macrophages was obtained, which was crucial in

preventing the adherence of L929 cells upon co-culture.

For the next 24h, macrophages were either left untreated (naive M0 macrophages) or were

polarized into classically (M1) or alternatively activated macrophages (M2) with LPS + IFN-

or IL4 + IL10, respectively. LPS (E. coli O111:B4 LPS, Sigma) was used at 1 µg/ml,

recombinant murine IFN- at 500 units/ml, recombinant murine IL-4 at 50 ng/ml and

recombinant murine IL-10 at 10 ng/ml (all cytokines obtained from Peprotech, London, UK).

Induction of apoptotic and necrotic cell death

The mouse fibrosarcoma cell lines L929rTA-FKBP-FADD-DED and L929rTA-FKBP-FADD

(Vanden Berghe et al., 2004) were cultured in DMEM supplemented with 10% fetal calf

serum, penicillin (100 units/ml), streptomycin (0.1 mg/ml), and L-glutamine (0.03%).

L929rTA-FKBP-FADD-DED and L929rTA-FKBP-FADD target cells were seeded in 175 cm2

adherent flasks (BD Biosciences). The next day, L929rTA-FKBP-FADD-DED cells were

treated with doxycycline (Duchefa Biochemicals, The Netherlands, 1 mg/ml) for 9h or 15h in

order to induce either early or late apoptotic cell death, respectively (Vanden Berghe et al.,

2004). To induce necrotic cell death, L929rTA-FKBP-FADD cells (Vanden Berghe et al.,

2004) were treated first with doxycycline (1 mg/ml) for 4h followed by 9h of AP1510 (0.5 µM),

a dimerizer of FKBP (Amara et al., 1997) kindly provided by ARIAD Pharmaceuticals Inc.

(Cambridge, MA). Non-adherent cells in suspension, mostly representing dead cells, were

either subjected to analysis with a FACSCalibur flow cytometer (BD Biosciences) or used in

coculture experiments with macrophages. Loss of cell membrane integrity was determined

by PI fluorescence (ex 535/em 617) as a measure of cell death. Phosphatidylserine exposure

was monitored by AnnexinV-fluorescein isothiocyanate (FITC, BD PharMingen) staining (ex

494/em 518).

In vitro phagocytosis assay

In order to quantify uptake of apoptotic and necrotic cells by macrophages, we used a

previously described phagocytosis assay (Brouckaert et al., 2004) based on two-parameter

analysis by flow cytometry. Target L929 cells were labelled with Cell Tracker Green

(Molecular Probes), while the minelayers of peritoneal macrophages were labelled with Cell

Tracker Orange (Molecular Probes). At the time points indicated, target cells were induced to

undergo early apoptotic (9h doxycycline), late apoptotic (15h doxycycline) or necrotic (4h

doxycycline followed with 9h AP1510) cell death as described above, harvested, washed

once, and co-incubated with macrophages at 1:1 ratio. The co-cultures were done in RPMI

146 Addendum I

1640 medium (as described above) in 5% CO2, at 37°Cfor 3 or 24 hours. Then cells were

rinsed three times in order to remove non-ingested target cells. Next, adherent macrophages

were either detached from the plate with Enzyme Free Cell Dissociation Buffer (Gibco BRL)

and analyzed on a FACSCalibur flow cytometer (BD Biosciences), or total RNA from the cells

was prepared.

Preparation of RNA, cDNA synthesis and macroarray hybridization

RNA was extracted after indicated time points (after 3 and 24 hours co-culture) using Aurum

total RNA mini kit (BioRad) following manufacturer’s instructions. RNA preparations were

quantified by spectrophotometry. For each sample, radioactively labelled cDNA probes were

synthesized by oligo(dT) primed reverse transcription of 1 µg of total RNA using SuperScript

II (Invitrogen) in the presence of 50 µCi [ -33P] dCTP (Amersham Biosciences). The reaction

mixture was incubated at 42°C for 1h and terminated by addition of EDTA (50 mM final

concentration). RNA was finally degraded by addition of NaOH (200 mM final concentration)

(10 min, 37°C), followed by neutralization with addition of TrisHCl buffer. The labelled cDNA

probes were purified using ProbeQuant G-50 micro columns (Amersham Biosciences),

quantified by liquid scintillation counting and denatured at 95°C.

The macrophage-oriented macroarray filters have been developed by Tom Boonefaes (Prof.

Dr. Johan Grooten, Molecular Immunology Unit, Department for Molecular Biomedical

Research, VIB, Ghent University). The macroarray consists of a set of 562 PCR fragments

(representing murine genes relevant to macrophage biology listed in Table 5) spotted in

duplicate onto Hybond-N membranes. The macroarray filters were hybridized with 5 x 106

cpm/ml of cDNA probe in 2 ml of NorthernMax Hybridization Solution (Ambion) at 42°C for

20h under continuous rotation. After hybridization, the arrays were rinsed at room

temperature in 2x SSC (Saline Sodium Citrate; stock buffer is 20x SSC containing 3 M NaCl,

0.3 M sodium-citrate), and washed four times at 68°C for 30 min (twice in 2x SSC, 1% SDS

and twice in 0,6x SSC, 1% SDS).

The arrays were exposed to PhosporImager screens (BioRad) for 15h and the signals were

digitized at 50 µm resolution with a Molecular-FX PhosphorImager (BioRad) using the

Quantity One software (BioRad). The resulting 16-bit (65536 grey-level) TIFF images were

analyzed with ArrayAnJ, a set of java plug-ins written by Tom Boonefaes for the open-source

image-analysis software ImageJ (NIH, http://rsb.info.nih.gov/ij/). In short, the array-image

was corrected for the local background signal, the spot locations and grid geometry were

determined, and the spot intensities were measured. The average of the duplicate spots was

calculated and a quality factor Q was defined as the ratio of the lowest to the highest values.

The error on the average of the duplicates (Avg) was calculated as E = Avg x (1-Q) +


Sqrt(Avg). All arrays were normalized by multi-pair wise detection of least differential genes,

yielding a normalized expression matrix of intensity values and errors for all genes, in which

the intensity values could be directly compared between any two samples.

The following expression ratios were calculated: for each macrophage population (M0, M1,

M2) at both time-points (3h/24h), the influence of each co-culture condition (+ early apoptotic,

+ late apoptotic, + necrotic) was calculated as the ratio to the corresponding macrophages

cultured in the absence of dead L929 cells (macrophage alone). In addition, the macrophage

polarization (the extent of differentiation in the M1 or M2 direction) was calculated as M1 =

(M1 alone)/(M0 alone), M2 = (M2 alone)/(M0 alone) and M2/M1 = (M2 alone)/(M1 alone).

Differences in expression were considered significant if they were lager than two fold and the

difference was lager than the sum of the errors of both expression levels.

Quantitative Real-Time PCR

RNA was extracted after 3 or 6 hours using Aurum total RNA mini kit (BioRad) as per

manufacturer’s instructions and quantified by spectrophotometry. Next, cDNA was prepared

using SuperScript II (Invitrogen) with oligo(dT) primers (Promega) following the guidelines of

the manufacturer. Quantitative Real-Time PCR was performed using qPCR Core Kit for

SYBR Green I (Eurogentec, Seraing, Belgium) according to the supplier’s instructions using

an iCycler Thermal Cycler (BioRad). Primers used were: Hmbs (housekeeping gene which

encodes an enzyme called hydroxymethylbilane synthase): 5’-

GAAACTCTGCTTCGCTGCATT and 5’-TGCCCATCTTTCATCACTGTATG; F4/80: 5’-

ATGGCAGTTTTGTTTTTAGCTCAAA and 5’-AGGACGTGCTCTTTTCAAAACAT;

Hygromycin resistance gene (selection marker in the transfected L929 cells): 5’-

AATAGCTGCGCCGATGGTTTC and 5’-GGCGGGAGATGCAATAGGTCAG.

Results

Optimization of the experimental procedure

In order to correctly analyze the macrophage responses at the transcriptional level, the RNA

samples should be practically free of RNA derived from the L929 cells, because otherwise it

would be extremely difficult to discriminate between gene induction in the macrophage or

differential L929 content of the samples. In order to have a more pure population of dead

cells, we induced cell death in adherent L929 cultures and harvested the detached, mostly

dead cells. This method of separation of dead cells from viable cells allowed us to have up to

80% of dead cells (Fig. 2).

148 Addendum I

Figure 2. PS exposure was determined by Annexin V binding and simultaneously cell permeability was assessed by PI staining and analysed by FACS assay. For the analysis, detached cells during induction of cell death were harvested. Early and late apoptotic cell death was induced by treatment with doxycycline for 9h and 15h, respectively. Necrosis was induced by 4h treatment with doxycycline followed by 15h with AP1510. Open bars: double negative living cells; grey bars: single positive PS positive cells; black bars: double positive PS and PI positive cells.

To further prevent the adherence of residual living L929 cells, red-labelled (Cell Tracker

Orange, Molecular Probes) macrophages were seeded at high density resulting in a nearly

confluent monolayer. Macrophages were co-cultured with green-labelled dead or living L929

cells (Cell Tracker Green, Molecular Probes) at a ratio of 1:1. In order to further decrease the

possible influence of non-ingested L929 cells on the macroarray analysis, we decided to use

extra washing steps to remove non-ingested dead L929 cells after the co-culture with

macrophages. For this, after 3h or 6h non-adherent cells were collected by three washing

steps with PBS resulting in a “non-adherent” fraction of mostly non-ingested L929 cells (Fig.

3). The remaining “adherent fraction” was detached with a cell dissociation buffer,

representing macrophages which either or not ingested dead L929 cells (Fig. 3). The ”total

co-culture” fraction was reconstituted by pooling equal parts of non-adherent and adherent

fractions in order to obtain the complete picture of cell composition in the co-culture. The

proportion of the L929 cells (green), macrophages (red) and double positive macrophages

(which had ingested L929 cells) in the total, non-adherent and adherent fractions was

80

0

10

20

30

40

50

60

70

NecroticApoptotic Late

apoptotic

PS- PI-

PS+ PI- PS+ PI+


determined by FACS analysis after the separation of co-cultures of macrophages with either

viable or dead L929 cells and the cell composition is illustrated in Figure 3.

A:

0%

20%

40%

60%

80%

100%

L929 54% 57% 56% 57% 77% 71% 64% 68% 21% 16% 30% 17%

Mf + L929 5% 5% 3% 3% 5% 6% 6% 4% 4% 3% 4% 1%

Mf 41% 38% 41% 40% 18% 24% 30% 28% 75% 81% 66% 82%

3h 6h 3h 6h 3h 6h 3h 6h 3h 6h 3h 6h

L929 DED L929 FADD L929 DED L929 FADD L929 DED L929

FADD

total coculture non-adherent fraction adherent fraction

B:

0%

20%

40%

60%

80%

100%

L929 46% 40% 31% 22% 76% 64% 75% 61% 3% 1% 3% 1%

Mf + L929 15% 15% 35% 24% 13% 18% 15% 22% 19% 12% 49% 25%

Mf 38% 46% 34% 54% 11% 18% 11% 17% 78% 87% 49% 74%

3h 6h 3h 6h 3h 6h 3h 6h 3h 6h 3h 6h

L929 DED L929 FADD L929 DED L929 FADD L929 DED L929

FADD

total coculture non-adherent fraction adherent fraction

Figure 3. The cell composition of non-adherent, adherent and total fractions after separation of co-cultures of macrophages with viable (A) or dead (B) cells. Macrophages were cocultured during 3h and 6h with (A) living L929rTA-FADD-DED (L929 DED) and L929rTA-FADD (L929 FADD) cells or (B)

with apoptotic L929rTA-FADD-DED (L929 DED) and necrotic L929rTA-FADD (L929 FADD) cells. The data represent the percentage of non-ingested L929 cells (L929, green colour) and macrophages which were engaged (Mf+L929, hatched bars) and not-engaged (Mf, red colour) in the uptake of L929 cells. Notice that the adherent fraction contains only 3% and 1% of non-ingested L929 cells at 3h and 6h, respectively. In order to calculate the percentage of uptake from these data, the value of “Mf+L929” should be divided by the sum of “Mf + (Mf+L929)” and multiplied by 100.

150 Addendum I

From Figure 3A, it can be seen that viable L929 cells adhered firmly in the co-culture

conditions. It constituted 21% to 30% of the adherent fraction after 3h and 16% to 17% after

6h of co-culture (Fig. 3A). These results indicate that it is impossible to use macrophages

cocultured with living L929 cells as controls in the transcription profiling experiment, because

a substantial amount of transcript profile would be L929-derived. From Figure 3B, it is

apparent that the separation of non-adherent and adherent cells after the co-culture of

macrophages with dead L929 cells resulted in an adherent fraction containing only 3% and

1% of non-ingested L929 cells at 3h and 6h, respectively (Fig. 3B). This is an acceptable

level of contamination of the adherent macrophages that may allow transcript profiling of the

macrophages without too much interference of the RNA from the L929 cells. The extent of

contamination was further assessed on the RNA level by Quantitative Real-Time PCR (Q-

PCR) for the L929-specific hygromycin resistance gene transcript (an antibiotic selection

marker used during transfection) and for the macrophage marker F4/80 (Fig. 4). The rate of

contamination for the adherent fraction with L929-derived RNA was between 1.3% and 5.9%

(based on the ratios of the quantitative PCR’s of hygromycin resistance gene transcript/F4-

80 gene transcript), which is in agreement with our FACS data (Fig 3B).

Figure 4. The rate of contamination by RNA derived from L929 cells. The extent of contamination was assessed by Quantitative Real-Time PCR (Q-PCR) for the L929-specific hygromycin resistance gene transcript (an antibiotic selection marker used during transfection) and for the macrophage marker F4/80. Macrophages were cocultured for 3h or 6h with apoptotic (L929-DED) or necrotic (L929-FADD) cells or were cultured alone. Non-adherent and adherent cells were separated; RNA was isolated from both fractions and analyzed by Quantitative Real-Time PCR. Note that in the adherent fraction L929-derived RNA was less than 5.9%. These results suggest a low level of L929-derived RNA contamination, indicating that RNA from dead L929 cells constitutes only a minor fraction of the total RNA isolated from the adherent fraction.

Thus, the above-mentioned optimization steps involving the separation between the non-

adherent and the adherent cells allowed us to reduce significantly the rate of contamination

5,9

1,35,0

2,0 1,0

0

10

20

30

40

50

60

70

3h 6h 3h 6h 6h

L929-DED L929-FADD Mf alone

adherent fraction

non-adherent fraction


with L929-derived RNA and it was considered to be sufficiently pure for the macroarray

analysis. Moreover, the PCR results suggest that the RNA contribution from ingested dead

L929 cells is also low, most probably due to RNA degradation during the cell death process

or during phagocytosis.

Transcription analysis of macrophages co-cultured with apoptotic, late apoptotic and

necrotic cells

In the present study, we examined the transcriptional profile of differently activated

macrophages which were exposed to early apoptotic, late apoptotic and necrotic cells. To

this end naïve (M0: untreated), classically activated (M1: LPS + IFN- stimulated) and

alternatively activated (M2: IL4 + IL10 stimulated) macrophages were pretreated for 24h prior

to co-culture with early apoptotic, late apoptotic or necrotic L929 cells. Macrophages were

labelled with Cell Tracker Orange, while the dead target cells had been labelled with Cell

Tracker Green. PS exposure (Annexin-V-FITC labelling) and loss of membrane integrity (PI

staining) of target cells was determined by FACS analysis before using them in the in vitro

phagocytosis assay (Figs. 5A and 5B). As shown in Figures 5A and 5B, early apoptotic cells

are mainly single positive (PS+ PI-), while late apoptotic and necrotic cells are mainly double

positive (PS+ PI+). After 3h and 24h of co-culture, the extent of phagocytosis was determined

by FACS. Percentages of uptake of early apoptotic, late apoptotic and necrotic cells by M0,

M1 and M2 macrophages are presented in Figures 5C and 5D. These results showed that

after 3h of co-culture, necrotic materials were taken by all types of macrophages (M1, M0

and M2) and the number involved in uptake cells was bigger than in co-culture with apoptotic

cells (Fig 5C). In addition, the pretreatment of macrophages with either LPS + IFN- (M1) or

IL4 + IL10 (M2) strongly stimulates the macropynocytosis of necrotic cell-derived material

while the engulfment of apoptotic cells was only slightly increased.

In the parallel experiment, RNA was prepared from the adherent fraction of macrophages for

transcription profile analysis in order to avoid contamination of residual L929 cells, as

discussed in the previous session. RNA extracts from both living and dead populations of the

L929 clones were also included as controls. A schematic representation of the RNA samples

analyzed is given in Figure 6.

152 Addendum I

Figure 5. A and B: PS exposure was determined by Annexin V binding and cells were simultaneously stained with PI to assess cell permeability as analysed by FACS. Grey bars represent PS positive and PI negative cells and black bars represent PS and PI positive cells for apoptotic and late apoptotic L929rTA-FADD-DED (A) and for necrotic L929rTA-FADD (B) cells. C and D: Macrophages (M0, M1, and M2) were co-cultured with control (viable), early apoptotic, late apoptotic and necrotic cells for 3h

(C) and 24 h (D). Macrophages were labelled with Cell Tracker Orange while the dead target cells had been labelled with Cell Tracker Green. These data represent the total population (non-adherent and adherent fractions). The percentage of double-stained macrophages quantifies the fraction of the macrophage population involved in phagocytosis of target cells (% of phagocytosis). In this particular experiment, the phagocytosis of necrotic material by all types of macrophages (M1, M0, M2) was more efficient than the engulfment of apoptotic cells. M1 and M2 macrophages took up necrotic material more efficiently than naïve macrophages (M0).

% o

f u

pta

ke

Necrotic

C

Early apoptotic

PS+ PI- PS+ PI+

A B

D

Control DED

Control FADD

Late apoptotic

0

20

40

60

80

M0 M1 M20

20

40

60

80

M0 M1 M2

% o

f u

pta

ke

Control0

20

40

60

80

Necrotic

% C

ell

s

Early

apoptotic

0

20

40

60

80

Control Late

apoptotic

% C

ell

s


Figure 6. A schematic representation of the samples analyzed on the cDNA array in the present study.

Transcriptional profile of differently activated macrophages cultured alone (M1 and

M2)

It is necessary to mention that after 24h pretreatment with either LPS + IFN- (M1) or IL4 +

IL10 (M2) and following further 3 and 24 hours culture alone, M1 and M2 macrophages differ

in terms of gene expression profile, which is shown in Figure 7, which is in agreement with

previously published data (Mantovani et al., 2004; Mantovani et al., 2002).

0,0001

0,001

0,01

0,1

1

10

100

1000

10000

0,0001 0,001 0,01 0,1 1 10 100 1000 10000

Ratio M2/M1 at 3h

Ra

tio

M2

/M1 a

t 24

h

M1

markers

M2

markers

Figure 7. Differences in gene induction between M1 and M2 macrophages. Following the 24h activation period, macrophages were cultured alone for a further 3 and 24 hours. Differential expression ratios between M2 and M1 macrophages (M2/M1) were calculated for both time points and are plotted in the x-axis for 3h and in the y-axis for 24h. Specific colour codes are given to markers specific for M1 (yellow) and M2 (blue) macrophages which will be used throughout this part. Elements in black are not differentially expressed between M1 and M2 macrophages.

In particular, in M1 macrophages cultured alone for 24h we observed gene induction of

chemokines such as CCL2 (2,8 folds), CCL4 (3.7 folds) specific for M1 macrophages. In

contrast to M1 macrophages, M2 macrophages cultured alone for 24h demonstrated an

154 Addendum I

increased level of transcripts for surface molecules (e.g. MRC1 - 11,3 folds), adhesion

molecules (e.g. Fn1 - 48,7 folds), and chemokines (e.g. CCL24 - 7.5 folds).

First of all, it is important to mention a general trend which becomes apparent from the cDNA

analysis: (1) a low number of upregulated genes by co-culture with dying cells was detected

(56 genes out of 562 analyzed) and it was impossible to identify downregulated genes

(discussed later); (2) apoptotic cells upregulate more genes in macrophages after 24h than

after 3h co-culture, and more in the case of M2 macrophages (14 genes) than in the case of

M1 (6 genes) and M0 macrophages (4 genes) (Fig. 8 and Table 1); (3) the impact of the late

apoptotic cells on gene expression was similar to that of the apoptotic cells, though with a

lower extent of gene induction (Tables 1, 2, 3, 4); (4) necrotic cells hardly modulate any

specific gene expression levels in either type of macrophages (Tables 2, 3, 4). In the

following paragraphs, we would like to consider the positive effects of dying cells on the

transcriptional profile of M1, M0 and M2 macrophages. The transcripts which showed a more

than two-fold upregulation following 3h and 24h co-culture of macrophages with target cells

are summarized in Tables 1, 2, 3 and 4.

0

5

10

15

20

25

30

0,001 0,1 10 1000

0

5

10

15

20

25

30

0,001 0,1 10 1000

0

5

10

15

20

25

30

0,001 0,1 10 1000

0

5

10

15

20

25

30

0,001 0,1 10 1000

0

5

10

15

20

25

30

0,001 0,1 10 1000

0

5

10

15

20

25

30

0,001 0,1 10 1000

M1 M0 M2

3h

24h

0

5

10

15

20

25

30

0,001 0,1 10 1000

0

5

10

15

20

25

30

0,001 0,1 10 1000

0

5

10

15

20

25

30

0,001 0,1 10 1000

0

5

10

15

20

25

30

0,001 0,1 10 1000

0

5

10

15

20

25

30

0,001 0,1 10 1000

0

5

10

15

20

25

30

0,001 0,1 10 1000

M1 M0 M2

3h

24h

Figure 8. Influence of apoptotic cells on the gene expression profile of naïve (M0, central graphs) and activated (M1 and M2, left and right graphs respectively) macrophages at 3h (top) and 24h (bottom). The differences induced by the uptake of apoptotic cells are plotted in the y-axis in function of the differential expression between M2 and M1 cells (x-axis, identical for each graph, compare with Fig. 8). M1 and M2 markers are shown in yellow and blue, respectively.

In Tables 1, 2, 3 and 4, we indicated the so called “marker” genes, and which transcriptions

are specifically up regulated in M1 or M2 cells. The following expression ratios were

calculated in these tables: for each macrophage population (M1, M0, M2) at both time-points

(3h/24h), the influence of each co-culture condition (+ Apoptotic, + Late Apoptotic, +


Necrotic.) was calculated as the ratio to the corresponding macrophages cultured in the

absence of dead L929 cells (macrophage alone).

Effect of apoptotic cells engulfment on transcriptional profile of classically activated

macrophages (M1)

As it can be seen from Table 1 and Figure 10, only a small set of genes (6 genes) were

specifically upregulated in M1 macrophages by apoptotic cells and among them were

chemokines (Cxcl1, Ccl3) and cytokines (IL-1 ). IFN- and LPS primed macrophages (M1)

interacting with early apoptotic, late apoptotic or necrotic cells demonstrated upregulated

levels of transcripts for vascular endothelial growth factor (VEGF-A, Table 2). A typical

autoradiography pattern of VEGF-A transcript expression between M1, M0 and M2

macrophages upon their interaction with dead cells is shown in Figure 9.

Figure 9. Autoradiograph in which radiographic signal at each spot of the array gives a measure of VEGF gene activity. Note that transcription of VEGF was highly upregulated in M1 macrophages after 24h co-culture with apoptotic (14.9 fold), late apoptotic (6.2 fold) and necrotic cells (9.6 fold) and to a lower extent in M2 macrophages by apoptotic cells at 3 h (3.13 folds) and 24 h (2.13 folds). This figure illustrates how expression ratios were calculated in the tables. For this, for each macrophage population (M1, M0, M2) at both time-points (3h/24h), the influence of each co-culture condition (+ Apopt., + Late Ap., + Necr.) was calculated as the ratio to the corresponding macrophages cultured in the absence of dead L929 cells (macrophage alone). For the detail ratios of VEGF-A transcripts see Tables 2-4.

This upregulation of VEGF-A transcription confirms previous observations in mouse epithelial

cells (HC 11) engulfing apoptotic human Jurkat T cells (UV-induced) (Golpon et al., 2004).

However, they are in contrast with the observation in the same publication that necrotic cells

(induced by multiple freeze thaw cycles) do not influence the expression level and release of

VEGF-A. In favour of our observation that both apoptotic and necrotic cells induce

upregulation of VEGF-A transcripts, however, is the report that in hearts with myocardial

infarction, the intensity of the VEGF signal was much higher in smooth muscle cells of

arterioles adjacent to necrosis and in infiltrating macrophages than in myocytes around the

site of the necrosis (Shinohara et al., 1996). Our data may suggest that pre-treatment with

156 Addendum I

IFN- and LPS of thyoglycollate-elicited macrophages strongly sensitises the expression of

VEGF-A mRNA upon interaction with early apoptotic, late apoptotic and necrotic cells. VEGF

can significantly contribute to macrophage recruitment by promoting migration and survival

as well as by stimulating angiogenesis (Golpon et al., 2004; Moldovan and Moldovan, 2005).

Effect of apoptotic cells engulfment on the transcriptional profile of naïve

macrophages (M0)

A low number of genes were specifically upregulated in naïve macrophages (M0) exposed to

early apoptotic cells (4 genes), while the most genes in M0 were also commonly upregulated

in M2 macrophages (15 genes) and some in M1 macrophages (5 genes, Table 1 and Fig.

10). This set of commonly upregulated genes in M0, M1 and M2 include CD14 and Gas6 (M0

and M1), and Fn1 and Mrc1 (M0 and M2) during interaction with early as well as with late

apoptotic cells (Table 2). Several of these molecules have been shown to be involved in

phagocytosis of apoptotic cells. For example, CD14 is a cell surface receptor which plays a

role in the recognition of apoptotic cells by macrophages (Devitt et al., 1998), and Gas6 is a

bridging molecule and a ligand for Mer receptor (Nakano et al., 1997). Adhesion of

macrophages to fibronectin, one of the extracellular matrix components, may facilitate the

uptake of apoptotic cells (McCutcheon et al., 1998). Several transcripts encoding

chemokines, which are involved in mononuclear cell chemoattraction, were increased when

naïve macrophages had been exposed to early and late apoptotic cells. Remarkably,

chemokine CCL2/Monocyte Chemoattractant Protein (MCP)-1 was upregulated in all types of

macrophages (M1, M0 and M2), and CCL6 (in M0 and M1) macrophage migration inhibitory

factor (MIF) (in M0 and M2), which may regulate monocyte and lymphocyte recruitment,

thereby facilitating the clearance of apoptotic cells (Li et al., 2005). In line with our findings,

increased production of MIF mRNA was also detected in macrophages exposed to

etoposide-induced apoptotic P388 cells while the protein level production of MIF was only

slightly enhanced by co-culturing with apoptotic cells (Kawagishi et al., 2001).

Effect of apoptotic cells engulfment on transcriptional profile of alternatively activated

macrophages (M2)

Interaction of alternatively activated macrophages (M2) with early apoptotic cells gave the

most extensive set of specifically upregulated genes (14 genes, Table 1 and Fig. 10). After

interaction with apoptotic cells, alternatively activated macrophages demonstrated the broad

range of genes for cytokines, chemokines, surface markers and genes involved in tissue

remodelling (Tables 1 and 4). Several specific chemokines such as Ccl4, Ccl7, and Ccl24


were upregulated only in M2 macrophages by apoptotic cells, and some by late apoptotic

cells (Tables 1 and 4). Strongly upregulated expression exhibited genes which encoding

matrix metalloproteinases (MMP) such as MMP8 (collagenase-2) and MMP12 (macrophage

metalloelastase) in M0 and M2 macrophages, while MMP13 (collagenase-3) was

upregulated specifically in M2 macrophages up to 25 folds (Table 1). It is known that MMP is

required for macrophage-mediated extracellular matrix proteolysis and tissue invasion. It has

been shown that the macrophages of MMP12-deficient (MMP-/-) mice have a markedly

diminished capacity to degrade extracellular matrix components and MMP-/- macrophages

are essentially unable to penetrate reconstituted basement membranes in vitro and in vivo. In

contrast to our findings, exposure of murine resident peritoneal macrophages to apoptotic

thymocytes in vitro did not induce production of MMP9 transcripts (Parks and Shapiro, 2001).

Importantly, it is apparent from Table 1 that apoptotic cells further increase transcription of

genes which were already upregulated in M2 macrophages after exposure to IL-4 and IL-10.

158 Addendum I

Offcialsymbol

3h 24h 3h 24h 3h 24h

Egr1 Egr1 Transcription 2,69 0,79 2,30 0,36 5,89 0,91 Ratio :

Fos Fos Transcription 2,28 4,65 5,00 2,33 2,22 1,60 M2 2 - 4

Ccl2 MCP-1 Chemokine 2,48 1,86 3,05 6,75 4,69 12,15 M1 4 - 8

Csf1 CSF-1 Cytokine 4,12 0,43 3,44 1,00 1,27 2,04 > 8

Cd83 CD83 Surface marker 0,83 3,09 2,07 1,08 2,83 2,00 M1

Alox15 12-Lipoxygenase Eicosanoid synthesis 1,50 3,67 2,46 4,14 1,92 2,26 M2

Cdkn1a p21 Waf1/CIP1 Cell cycle 1,37 2,35 2,45 2,73 3,24 2,54 M1

Tnfaip3 A20 Signaling 0,63 2,39 2,26 3,13 2,05 1,63 M1

Ndrg1 Ndrg1 1,70 2,46 0,98 2,17 2,57 3,47 M2

Ccl6 CCL6 / c10 Chemokine 0,45 2,26 2,00 5,07 1,22 1,33 M2

Cd14 CD14 PRR 0,92 8,22 3,86 3,18 1,60 1,84 M1

Gas6 GAS6 bridging molecule 0,70 2,50 1,58 2,80 0,68 1,00 M2

Hmox1 HO-1 Anti-oxidative 2,24 0,61 1,03 2,18 1,55 1,84 M1

Zfp36 TTP (tistetraprolin) Translational regulation 0,88 2,19 2,78 1,36 1,81 0,57 M2

Mif MIF Chemokine 0,96 0,96 2,37 6,82 1,20 3,22 M1

Mrc1 MR Surface marker 1,56 1,00 0,42 3,32 0,85 5,06 M2

Fcgr2b Fcg-RIIb Surface marker 0,76 0,70 0,89 3,94 0,66 2,01 M2

Fcgr3 Fcg-RIII Surface marker 0,67 1,12 0,86 2,23 1,02 3,03 M2

Emp1 Surface marker 1,82 1,00 1,91 2,57 3,50 3,12 M2

Pla2g7 PAF-AH 0,84 0,89 1,12 2,28 1,78 6,13 M1

Arg1 Arginase 1 Enzyme 0,71 1,61 2,82 6,89 2,83 6,69 M2

Cbr2 Enzyme 1,14 0,59 1,20 2,69 1,27 2,39 M2

Fn1 Fibronectin-1 Matrix remodeling 1,49 0,91 3,45 6,11 0,70 2,58 M2

Mmp12 MMP-12 Matrix remodeling 1,20 0,93 0,61 2,32 0,57 5,82 M2

Mmp8 MMP-8 Matrix remodeling 0,82 0,64 0,72 8,21 0,91 2,48 M2

Timp1 TIMP-1 Matrix remodeling 0,75 1,17 2,38 1,16 1,39 3,07

9030624J02Rik unknown 1,03 1,85 1,84 6,11 1,06 2,25 M2

2700094F01Rik unknown 1,40 1,89 2,58 3,67 1,96 2,11

A530088I07Rik unknown 0,88 0,21 1,31 2,51 1,03 2,31 M1

Spp1 Osteopontin Chemokine ? 1,08 0,80 1,08 5,09 0,80 1,91 M2

Lyzs Lysozyme Anibacterial 1,03 1,20 2,00 2,92 0,80 1,06 M2

Inpp5d SHIP-1 signaling 0,79 0,85 2,64 1,84 0,93 1,98 M2

Tmem14c Surface marker 0,73 1,43 1,72 3,37 0,61 1,55 M2

Cxcl1 KC / GRO1 Chemokine 6,68 1,03 1,48 0,62 0,96 1,19

Ccl3 MIP-1 Chemokine 0,76 2,52 0,52 0,67 1,61 1,85 M1

Il1b IL-1b Cytokine 2,22 3,25 0,88 0,39 0,67 1,27 M1

Inhba Inhba Cytokine 2,79 1,86 0,59 0,56 0,63 1,95 M1

Ptgs2 COX-2 Eicosanoid synthesis 0,62 4,29 1,05 0,28 0,55 1,00 M1

Socs3 SOCS3 Signaling 1,44 2,83 0,97 0,50 0,85 1,08 M2

Ccl7 MCP-3 Chemokine 0,77 0,56 0,94 0,53 4,50 12,07 M1

Ccl12 MCP-5 Chemokine 1,39 0,98 0,89 0,64 1,47 2,08 M2

Ccl4 MIP-1 Chemokine 1,20 1,38 1,00 1,55 1,71 2,55

Ccl24 Eotiaxin-2 Chemokine 0,59 0,73 0,63 0,34 2,43 19,63 M2

Mmp13 MMP-13 Matrix remodeling 0,82 0,38 1,44 1,52 1,58 25,72

Mgl1 Mgl1 Surface marker 1,58 0,65 1,41 1,79 0,89 4,97 M2

Mgl2 Mgl2 Surface marker 1,27 1,86 0,48 0,72 0,72 2,57 M2

Gadd45a GADD45 Signaling 0,96 0,38 1,17 1,22 1,00 2,20 M2

Stat5a STAT5a Transcription 0,89 0,98 1,02 1,71 0,96 3,46 M2

Nfkb1 NFkB p50/p105 Transcription 1,18 0,72 0,60 0,81 2,08 0,79

Emb Embigin ? 0,92 1,26 0,68 1,46 0,77 2,19 M2

Ramp2 Ramp2 Signaling 1,33 1,11 1,71 0,48 1,16 2,13

Tgfbi TGF induced unknown 0,92 1,30 0,98 1,19 4,44 3,23 M1

Pgm2 Pgm2 enzyme 1,49 1,04 0,98 1,56 1,76 2,36

Vegfa VEGF-A Angiogenesis 1,33 14,87 1,15 1,38 3,13 2,13 M1

Clec7a Dectin-1 PRR 1,09 2,52 0,56 0,54 1,60 3,66 M2

Sc5d Sc5d Sterol metabolism 1,63 2,29 0,69 0,44 1,92 3,27 M2

1200016E24Rik unknown 0,75 2,93 1,53 0,97 0,82 2,65 M1

Ma

rke

r

in all macrophages

in M0 and M1 macrophages


M1

+Apoptotic

M0

+Apoptotic

M2

+Apoptotic

in M0 macrophages only




Table 1. A comparison of impacts of early apoptotic cells uptake on transcriptional profile in M1, M0 and M2 macrophages. Note that only a minor fraction of genes (9) upregulated in all macrophages.


Offcicial

symbol 3h 24h 3h 24h 3h 24h

Ccl2 MCP-1 2,48 1,86 2,07 1,64 0,83 0,99 M1 Ratio :

Ccl6 CCL6 0,45 2,26 0,54 1,84 0,47 1,11 M2 2 - 4

Ccl3 MIP-1 0,76 2,52 1,48 2,03 1,57 0,53 M1 4 - 8

Cxcl1 KC / GRO1 6,68 1,03 4,36 1,09 0,88 0,76 > 8

Vegfa VEGF-A 1,33 14,87 1,41 6,22 1,02 9,63 M1

Il1b IL-1 2,22 3,25 1,26 0,70 1,11 1,15 M1

Inhba Inhibin beta-A 2,79 1,86 1,16 1,69 1,41 1,35 M1

Csf1 CSF-1 4,12 0,43 2,94 0,33 0,47 0,30

Cd14 CD14 0,92 8,22 0,87 2,93 0,48 2,30 M1

Clec7a Dectin-1 1,09 2,52 1,41 3,29 1,21 1,07 M2

Cd83 CD83 0,83 3,09 0,59 3,58 0,59 2,93 M1

Gas6 GAS6 0,70 2,50 0,88 1,94 0,93 1,75 M2

Fos Fos 2,28 4,65 1,76 1,65 1,41 1,89 M2

Egr1 Egr1 2,69 0,79 1,77 0,84 1,23 4,21

Cdkn1a p21 Waf1/CIP1 1,37 2,35 1,99 1,33 2,52 1,17 M1

Tnfaip3 A20 0,63 2,39 0,82 1,69 0,80 1,34 M1

Socs3 SOCS3 1,44 2,83 1,36 1,47 0,93 1,58 M2

Zfp36 TTP (tristetraprolin) 0,88 2,19 0,95 1,50 0,79 1,89 M2

Ndrg1 Ndrg1 1,70 2,46 2,70 1,78 1,70 1,67 M2

Ptgs2 COX-2 0,62 4,29 1,06 3,02 0,73 1,75 M1

Hmox1 Heme oxygenase 1 2,24 0,61 2,72 0,74 2,16 1,13 M1

Alox15 12-Lipoxygenase 1,50 3,67 1,56 4,03 1,26 3,05 M2

Sc5d Sterol C-5 desaturase 1,63 2,29 0,73 2,25 0,96 1,21 M2

1200016E24Rik 0,75 2,93 0,77 1,86 0,70 2,18 M1

Macrophage effector functions

unknown

Cytokines / Chemokines

Surface markers

Gene regulation / signaling

+ Apoptotic + Late

Apoptotic

+ Necrotic

Ma

rke

r

Table 2. Changes in induction of genes of M1 macrophages upon interaction with early apoptotic, late apoptotic and necrotic cells.

Figure 10. Distribution of upregulated genes in response to apoptotic cells among M1, M0 and M2 macrophages. For details of expression levels see Table 1.

160 Addendum I

Official


Ccl2 MCP-1 3,05 6,75 2,76 2,84 0,81 0,53 M1 Ratio :

Ccl6 CCL6 2,00 5,07 1,97 7,06 1,17 1,90 M2 2 - 4

Mif MIF 2,37 6,82 0,98 4,08 1,07 1,53 M1 4 - 8

Spp1 Osteopontin 1,08 5,09 0,82 3,05 0,60 2,61 M2 > 8

Csf1 CSF-1 3,44 1,00 1,56 1,04 1,81 0,43

Cd14 CD14 3,86 3,18 3,57 1,99 1,14 1,05 M1

Mrc1 MR 0,42 3,32 1,64 3,28 0,45 0,72 M2

Fcgr2b Fcg-RIIb 0,89 3,94 0,74 2,30 0,64 1,09 M2

Fcgr3 Fcg-RIII 0,86 2,23 0,75 2,00 1,19 1,37 M2

Cd83 CD83 2,07 1,08 1,60 0,98 0,63 0,75 M1

Emp1 Emp1 1,91 2,57 1,32 1,33 0,96 0,75 M2

Tmem14c Tmem14c 1,72 3,37 0,78 2,53 2,03 1,33 M2

Gas6 GAS6 1,58 2,80 1,39 1,91 1,70 0,95 M2

Fn1 FN1 3,45 6,11 0,91 2,40 1,62 2,09 M2

Mmp12 MMP-12 0,61 2,32 0,88 1,55 0,78 1,10 M2

Mmp8 MMP-8 0,72 8,21 1,00 6,90 0,72 1,36 M2

Timp1 TIMP-1 2,38 1,16 1,25 0,53 1,31 0,49

Fos Fos 5,00 2,33 2,63 1,02 1,11 1,27 M2

Egr1 Egr1 2,30 0,36 0,87 0,50 0,78 0,58

Cdkn1a p21 Waf1/CIP1 2,45 2,73 1,08 1,98 0,91 1,38 M1

Tnfaip3 A20 2,26 3,13 2,65 2,35 2,00 1,65 M1

Inpp5d SHIP-1 2,64 1,84 1,45 1,80 1,67 1,64 M2

Zfp36 TTP (tristetraprolin) 2,78 1,36 1,15 1,00 1,78 1,00 M2

Ndrg1 Ndrg1 0,98 2,17 0,69 0,88 0,69 0,82 M2

Arg1 Arginase 1 2,82 6,89 2,27 1,03 0,82 0,40 M2

Lyzs Lysozyme 2,00 2,92 1,09 2,22 1,94 1,39 M2

Hmox1 HO-1 1,03 2,18 2,23 1,91 1,52 1,15 M1

Alox15 12-LO 2,46 4,14 2,02 2,63 2,09 1,27 M2

Pla2g7 PAF-AH 1,12 2,28 0,55 1,21 0,86 0,84 M1

Cbr2 Cbr2 1,20 2,69 1,09 1,67 0,80 0,64 M2

9030624J02Rik 1,84 6,11 1,48 4,12 1,05 1,15 M2

A530088I07Rik 1,31 2,51 1,19 1,93 1,30 1,64 M1

2700094F01Rik 2,58 3,67 0,90 2,72 1,50 1,59

+ Apoptotic + Late

Apoptotic

+ Necrotic

Ma

rke

rMacrophage effector functions

unknown


Surface markers

Matrix remodeling


Table 3. Changes in inductions of genes of M0 macrophages upon interaction with early apoptotic, late apoptotic and necrotic cells.


Official


Ccl2 MCP-1 4,69 12,15 3,92 10,08 0,93 1,77 M1

Ccl7 MCP-3 4,50 12,07 4,58 5,47 1,33 0,80 M1

Ccl12 MCP-5 1,47 2,08 1,19 1,67 0,89 0,94 M2

Ccl4 MIP-1 1,71 2,55 1,12 1,05 0,51 1,00

Ccl24 Eotaxin-2 2,43 19,63 1,73 8,59 1,14 1,60 M2

Mif MIF 0,99 3,45 0,88 2,53 0,54 1,88 M1

Vegfa VEGF-A 3,13 2,13 1,13 1,39 0,56 0,61 M1

Csf1 CSF-1 1,27 2,04 1,93 1,12 1,00 1,32

Ratio :

Clec7a Dectin-1 1,60 3,66 1,74 2,32 0,93 1,08 M2 2 - 4

Mrc1 MR 0,85 5,06 1,09 3,03 0,77 1,52 M2 4 - 8

Fcgr2b Fc -RIIb 0,66 2,01 0,71 1,35 0,63 1,51 M2 > 8

Fcgr3 Fcg-RIII 1,02 3,03 0,77 2,44 0,64 2,11 M2

Mgl1 Mgl1 0,89 4,97 0,80 1,67 0,53 1,68 M2

Mgl2 Mgl2 0,72 2,57 0,59 2,10 0,66 1,33 M2

Cd83 CD83 2,83 2,00 2,50 1,92 0,94 0,33 M1

Emp1 Emp1 3,50 3,12 4,19 1,63 1,50 1,19 M2

Emb Embigin 0,77 2,19 0,93 1,81 1,14 0,72 M2

Fn1 FN1 0,70 2,58 0,86 1,33 0,61 1,47 M2

Mmp13 MMP-13 1,58 25,72 2,25 2,76 1,67 0,97

Mmp12 MMP-12 0,57 5,82 1,21 1,77 0,97 1,10 M2

Mmp8 MMP-8 0,91 2,48 0,93 1,97 0,83 1,25 M2

Timp1 TIMP-1 1,39 3,07 0,78 3,37 0,69 1,98

Fos Fos 2,22 1,60 2,33 0,81 0,66 0,56 M2

Egr1 Egr1 5,89 0,91 5,78 1,09 2,33 1,00

Cdkn1a p21 Waf1/CIP1 3,24 2,54 1,56 1,65 1,01 0,98 M1

Nfkb1 NFkB p50/p105 2,08 0,79 1,59 0,96 1,29 0,90

Tnfaip3 A20 2,05 1,63 1,88 0,83 0,60 1,00 M1

Stat5a STAT5 0,96 3,46 0,95 1,45 0,86 0,90 M2

Gadd45a GADD45 1,00 2,20 1,10 1,80 0,99 0,79 M2

Ramp2 Ramp2 1,16 2,13 1,42 1,53 1,00 1,27

Ndrg1 Ndrg1 2,57 3,47 0,70 1,87 0,41 1,04 M2

Arg1 Arginase 1 2,83 6,69 2,77 3,18 1,93 2,87 M2

Alox15 1,92 2,26 1,12 2,06 1,38 1,52 M2

Pla2g7 PAF-AH 1,78 6,13 2,17 3,58 1,49 2,73 M1

Cbr2 Cbr2 1,27 2,39 0,96 1,63 0,85 1,52 M2

Tgfbi TGFb induced 4,44 3,23 3,63 1,42 2,06 1,35 M1

Pgm2 Pgm2 1,76 2,36 1,42 0,94 0,93 0,90

Sc5d Sterol C-5 desaturase 1,92 3,27 2,12 2,01 1,63 1,01 M2

1200016E24Rik 0,82 2,65 1,034 0,834 0,648 0,748 M19030624J02Rik 1,06 2,25 0,82 2,12 0,83 1,60 M2

A530088I07Rik 1,03 2,31 1,09 1,61 1,10 2,09 M1

2700094F01Rik 1,96 2,11 0,95 1,34 1,12 1,39

+ Apoptotic + Late

Apoptotic

+ Necrotic

Ma

rke

r

Macrophage effector functions

unknown


Surface markers

Matrix remodeling


Table 4. Changes in transcriptional profile of M2 macrophages upon interaction with early apoptotic, late apoptotic and necrotic cells. Note that the most genes which are upregulated by apoptotic cells in M2 macrophages are already upregulated in cultured alone M2 macrophages and apoptotic cells increase further their expression (see the column “Marker”). This suggests that apoptotic cells synergize their effects with IL-4 and IL-10 on gene expression profile of macrophages.

162 Addendum I

Discussion

In the present study we used a macrophage-oriented cDNA array system which consists of a

set of 562 genes (list of all genes, which were tested see in Table 5) in order to analyze the

impact of early apoptotic, late apoptotic and necrotic cells on the transcriptional profile of

differently activated macrophages (M1, M0 and M2). We found that early apoptotic cells

induce gene expression in macrophages more strongly after 24h co-culture with

macrophages, and that the effect was more prominent in M2 than in M0 and M1

macrophages. Exposure of macrophages to early apoptotic cells mostly enhanced the

transcription of M2 marker genes, suggesting that apoptotic cells act synergistically with

alternatively activated macrophages, promoting tissue repair function. We observed that late

apoptotic cells had similar effects to early apoptotic cells on macrophage gene expression,

although these effects were less prominent. This is in agreement with the discussed view in

the third part of the Introduction (“Clearance of apoptotic and necrotic cells and its

immunological consequences”) that late apoptotic (secondary necrotic) cells, like apoptotic

cells, may elicit the same responses in macrophages. However, one should note that during

24h of co-culture the cell death process in the target cells is further progressing, and

therefore it is likely that a 24h co-incubation condition with early apoptotic cells may reflect a

situation including late apoptotic cells.

Surprisingly, while the uptake of necrotic cells in this set-up was more efficient (Figs. 5C and

5D), we hardly see any specific modulatory effects of necrotic cells on gene expression

profile of M1, M0 and M2 macrophages. In fact, this is in agreement with our previously

published data in which we demonstrated that macrophage cell line (Mf 4/4) upon interaction

with necrotic L929hR55 cells do not produce TNF- and IL-6 on mRNA and protein levels

(Brouckaert et al., 2004). These findings are also in line with the observations that in the

apaf-1 deficient mice necrotic cell death occurs without inflammatory response (Chautan et

al., 1999). In addition, it was shown that necrotic cells are unable to activate macrophages

themselves but can enhance their activation by LPS, whereas apoptotic cells inhibit

phlogistic macrophage responses (Cocco and Ucker, 2001). In this respect, one could think

of studying the influence of pre-exposure of macrophages to apoptotic or necrotic cells on

the M1- or M2-type activation. The question would be how dying cells can influence the

further activation or differentiation of naïve macrophages.

Although it has been demonstrated that exposure of M1 macrophages to apoptotic cells

inhibits LPS-induced NF- B activity (Cvetanovic and Ucker, 2004), we were unable to detect

any genes which were consistently downregulated by dying cells. This may partially be

explained by the fact that the peritoneal macrophages used represent a heterogeneous

population. Indeed, only 30% of the macrophage population is engaged in phagocytosis of


apoptotic or necrotic cells. It is conceivable that gene regulation depends on the issue of

whether the macrophages are involved in phagocytosis or not. Since our macroarray data

reflect a net result of engaged and non-engaged macrophages, gene induction levels may

have been influenced by the heterogeneity of the response. Genes that are strongly induced

in 30% of the cells are scored as inducible, although the net ratio is reduced (e.g., x10 in

30% of cells gives final net ratio of x3.7). However, a total shut-down of a gene in 30% of the

cells will only result in a 30% reduction of the total signal in total population, or gives a ratio

of x0.7, which would not have been scored as relevant because of less than two-fold

difference. One could solve this aspect by including a sorting procedure between engaged

and non-engaged cells prior to RNA extraction.

Several genes, such as MCP-1, Egr1, Fos (in all macrophages), and HO-1 (in M0 and M2)

were induced by apoptotic cells. Induction of the same set of genes has been shown to occur

after exposure of different types of cells, including monocytes, leukocytes and endothelial

cells, to oxidized phospholipids (Leitinger, 2003). Apoptotic cells can be one of the sources of

oxidized phospholipids, e.g., apoptotic cells externalize oxidized phosphatidylserine (Matsura

et al., 2005) and discard membrane parts which contain biologically active oxidized

phospholipids (Huber et al., 2002). Moreover, externalization of oxidized phospholipids was

shown to be required for macrophage clearance of apoptotic cells (Kagan et al., 2002). One

of our future plans is to analyse whether increased levels of, e.g., chemokines, MMPs, VEGF

mRNA’s is effectively accompanied by a concomitant increase of the corresponding proteins

and/or activity. It is obvious that the final catalytic activity, for example, of MMPs is the net

result of regulatory mechanism at several levels: gene expression, compartmentalization,

enzyme activation, enzyme inactivation, substrate availability and affinities (Parks and

Shapiro, 2001). Therefore, an important validation of the macroarray data lies in the control

of the findings on the protein level. Another important aspect of the macroarray results is

their reliability. In an additional paragraph we provide supplementary information on the

reproducibility and the reliability of the macroarray results (see below).

Despite these considerations, the macroarray data provided in this chapter allow us to

formulate the following conclusions: (1) Our macroarray data confirm the existence of

specific transcript profiles in differentially activated macrophages representing three major

functional classes of macrophages: M0 (naïve), M1 (killer) and M2 (healer) macrophages

(Tables 2, 3, 4). (2) When these differentially activated macrophages are exposed to

apoptotic cells, they exhibit a differential transcription profile. The differential responsiveness

of the macrophages to apoptotic cells is therefore likely to dependent largely on the balance

of pro- versus anti-inflammatory signals in the phagocyte’s immediate vicinity. (3) Exposure

to apoptotic cells specifically favours an M2-like transcription profile, suggesting that

apoptotic cells promote the regeneration function of macrophages (Tables 3, 4). (4)

164 Addendum I

Importantly, neither macrophage type specifically responded to necrotic cells (Tables 2, 3,

and 4), although their uptake was more efficient than in the case of apoptotic cells (Figs. 5C

and 5D). (5) Once more, although we fully realize that our macroarray cDNA hybridization

results should be validated for the genes discussed by real-time PCR we think that the

conclusions formulated above form a firm basis for formulating challenging working

hypotheses and exploring further experiments. In the next section we discuss briefly the

reliability of the macroarray data.

Supplementary information

Control of reproducibility of the array results

As each PCR fragment is spotted in duplicate on the macroarray, the variation between

these duplicate spots is a first measure for the accuracy of the obtained expression levels. A

representative example in Figure 11A illustrates the correlation between the duplicate spot

intensities which was significant (R2=0.985). When all duplicates on all 28 macroarray

experiments in the dataset are considered (Fig. 10B), 95% of duplicate spots are varied less

than 1.88-fold and 99% less than 3-fold. We decided to use a threshold of 2-fold difference

(96.33% of duplicate variation) in combination with a restriction on the error (see “Materials

and Methods”) to exclude false differences caused by variability.

A second quality measure is the inter-macroarray variability of spot intensities when two

independent macroarrays are hybridized in duplicate with the same labelled cDNA probe. As

illustrated in Figure 12, the inter-array reproducibility was quite high and no false differences

greater than two-fold were observed. As a next quality measure, we should mention the high

reproducibility of the differential expression ratios obtained by analyzing samples from

independent experiments in which peritoneal thyoglycollate-elicited macrophages were

treated with 1 µg/ml of LPS for 6 hours or left untreated (Fig. 13).

The last internal quality control is reproducibility of the given macroarray data of gene

expression of activated macrophages (M1 and M2). Moreover, when M1 and M2

transcription profiles were compared with and without target cells the same marker genes are

expressed, demonstrating the reproducibility and reliability of the macroarray results. This

transcriptional profile is demonstrated in Figure 5. The same genes were equally expressed

in differently activated macrophages (LPS + IFN- (M1) versus IL-4 + IL-10 (M2)) after their

culture for 3 and 24 hours, and the gene expression pattern of these macrophages was

comparable to the previously observed gene profile of the M1 and M2 macrophages

(Boonefaes and Grooten, unpublished data).


A: B:

Figure 11. A: A scatter plot illustrating the reproducibility of the duplicate spot signals within one array (condition naïve macrophages (M0) co-cultured with apoptotic cells during 24h). For intensities above 100, all differences were smaller than 1.4-fold. Below the intensity of 100, most variation was within a two-fold range, while few outliers could be observed (red dots), mainly due to artefacts on the membrane. Red lines mark two-fold differences of the signal intensity, green lines 1.4-fold differences. B: Graph represents the variability between duplicate spots on all arrays hybridized for this experiment. The y-axis represents the percentage of all duplicates that vary less than the ratio indicated on the x-axis. Note that 96.33% of all duplicates have less than a two-fold variation.

1,132

1,29

1,59

1,88

2

2,25

3

4,1

9,6

0 2 4 6 8 10 12

50,00%

75,00%

90,00%

95,00%

96,33%

97,50%

99,00%

99,50%

99,90%

% o

f d

up

licate

sp

ots

ratio between duplicate spots

R2 = 0,985

1

10

100

1000

10000

100000

1 10 100 1000 10000 100000

Intensity of spot 1

Inte

ns

ity

of

sp

ot

2

R2 = 0,988

1

10

100

1000

10000

100000

1 10 100 1000 10000 100000

Intensity on Array A

Inte

nsit

y o

n A

rray B

Figure 12. Inter-array reproducibility. Elements are plotted as a function of their expression level in two arrays. Notice no false differences greater than two-fold could be observed. Mouse thyoglycollate-elicited macrophages were treated with 1 µg/ml of LPS for 6h (Boonefaes and Grooten, unpublished results).

Taking into account the above-mentioned quality measures, our data on changes of gene

expression profile of differently activated macrophages can be considered as reliable, but

166 Addendum I

this does not exclude the fact that only validation by quantitative PCR, and eventually by

western blotting, is required to allow to draw firm conclusions on the intercellular

consequences of macrophages differentially exposed to dying cells.

0,01

0,1

1

10

100

1000

0,01 0,1 1 10 100 1000

Ratio +LPS 6h (A)

Ra

tio

+L

PS

6h

(B

)

Figure 13. The data represent the reproducibility among different biological samples. Mouse thyoglycollate-elicited macrophages were treated with 1 µg/ml of LPS for 6 hours or left untreated in duplicate. cDNA from all samples was analyzed by macroarray hybridisation. The obtained differential expression ratios between LPS-treated and untreated cells are represented for comparison A in the x-axis and for comparison B in the y-axis. Dots on the black line are equally induced in both experiments. Dots on the red lines are induced two-fold higher in experiment A (lower red line) or in experiment B (upper red line).


Table 5. List of all genes (official symbols) tested in the macroarray experiment.

0610007C21Rik B2m Ccr3 Cul3 Foxo1 Igf1

1100001G20Rik B2m Ccr4 Cx3cl1 Foxo3 Igf2r

1110007F12Rik Bax Ccr5 Cx3cr1 Frmd4b Ikbkb

1110012D08Rik Bcl10 Ccr8 Cxcl1 Frmd4b Ikbke

1110020C13Rik Bcl2 Cd14 Cxcl10 Fth Il10

1110032A03Rik Bcl2l Cd163 Cxcl16 Ftl1 Il10ra

1200016E24Rik Bcl3 Cd164 Cxcl9 G3bp Il12a

2010106G01Rik Bcl6 Cd2 Cxcr3 Gadd45a Il12b

2400001E08Rik Becn1 CD209d Cxcr4 Gadd45b Il12rb1

2610507B11Rik Birc6 Cd209e Cybb Gadd45g Il12rb2

2700094F01Rik Bmp2k Cd276 Cysltr2 Gal Il13

2810417H13Rik Bnip1 Cd28 D11Ertd759e Galnt1 Il15

2810474O19Rik Bnip2 Cd68 Ddit3 Galnt2 Il15ra

2810474O19Rik Btla Cd79b Ddx26 Galnt7 Il18

3110001I20Rik C3 Cd80 Ddx48 Gas6 Il18bp

3110023E09Rik C4 Cd83 Ddx5 Gata3 Il1a

4933407C03Rik C4bp Cd84 Ddx6 Gatm Il1b

9030624J02Rik C6 Cd86 Def6 Gfi1 Il1rl1

A530088I07Rik Cacna1d Cd97 Dmxl2 Gla Il1rn

Abcd4 Cai Cdk7 Dnajc5 Gm2a Il2

Actb Calu Cdkn1a Dnm Gp49a Il23a

Acvrl1 Capg Ceacam1 Dpm2 Gpi1 Il24

Adfp Capza1 Cebpa Ear11 Gpnmb Il27

AI256775 Car4 Cebpb Ear2 Gpt2 Il27ra

AI316828 Card11 Cebpd Eef1a1 Grn Il2ra

AI427809 Card15 Cebpe Eef1g H2-Aa Il2rb

AI597013 Casp1 Cebpg Egr1 H2-ab1 Il2rg

AK129128 Casp2 Cggbp1 Egr1 H2-K Il3

Ak2 Casp3 Chi3l3 eif4a1 Hmga2 Il31ra

Ak4 Casp4 Chi3l3 Eif4ebp1 Hmox1 Il4

Akr1a4 Casp6 Chi3l3 Elf3 Hnrpa2b1 Il4ra

Akr1a4 Casp7 Chuk Elk1 Hnrpc Il5

Akt1 Casp8 Cish Elk3 Hnrpd Il6

Aldo1 Cbfb Clec7a Elovl1 Hnrpl Il7

Alox15 Cbr2 Clecsf9 Emb Hrmt1l2 Il7r

Alox15b Ccl1 COPD_2_c5 Emilin1 Hsf1 Il9

Alox5 Ccl12 Creb1 Emp1 Hspa8 Inhba

Ankfy1 Ccl17 Csf1 Ets2 Icam1 Inpp5d

Anp32e Ccl2 Csf2 Evi5 Icosl Iqgap1

Anpep Ccl22 Csf3r F13a Icosl Irak1

Anxa1 Ccl24 Cspg2 F7 Idh2 Irak3

Ap1b1 Ccl3 Csrp1 Fbxo6b Ifi1 Irak4

Ap2a2 Ccl4 Ctbs Fcer1g Ifi205 Irf1

Apg7l Ccl5 Ctsb Fcgr1 Ifi30 Irf2

Arg1 Ccl6 Ctsc Fcgr2b Ifit2 Irf3

Arhgap19 Ccl7 Ctsd Fcgr3 Ifnar1 Irf8

Atf1 Ccl8 Ctsk Fcgrt Ifnar2 Itga4

Atf2 Ccl9 Ctsl Fn1 Ifnb1 Itgal

Atf3 Ccnd1 Ctsl Fos Ifng Itgb2

Atp6v1a1 Ccnl1 Ctss Fosl2 Ifngr1 Itm1

168 Addendum I

Ivd Mmp1b Pgm2 Runx1 Tbx21 Vps11

Jun Mmp2 Pi4k2a Rxra Tbxa2r Vsig4

Junb Mmp7 Pilra S100a11 Tbxas1 Vtcn1

Jund1 Mmp8 Pla2g4a S100a8 Tcf7 VUB_10H

Klf1 Mmp9 Pla2g7 Saa3 Tcf7 VUB_4G1

Klf10 Mrc1 Plat Sat1 Tebp VUB_5F1

Klf2 Mrps17 Plau Sc5d Tegt VUB_Fcgr

Klf4 Msn Pld3 Sdc1 Tep1 Wdr68

Klrd1 Mta1 Plod1 Sdc3 Tert Xcl1

Klrk1 Mt-Cytb Pmp22 Sdha Tgfb1 Xlkd1

Lamp2 Mxd4 Pparbp Sec23b Tgfbi Ywhaz

Lasp1 Myc Ppard Seh1l Tgtp Zfp36

Lass2 Ndrg1 Ppard Sepp1 Tia1

Lat Nfam1 Pparg Sfpi1 Tial1

Ldh1 Nfatc1 Ppid Sftpa Timp1

Lef1 Nfatc2 Ppm2c Sgk Timp2

Lifr Nfe2l2 Prdx1 Sidt2 Timp3

Lilrb4 Nfkb1 Prg4 Sigirr Timp4

Lime1 Nfkb2 Prkwnk1 Sigirr Tlr2

Litaf Nfkbia Psap Siva Tlr4

Lman1 Nfkbib Pten Slamf7 Tlr5

LPS_1_a4 Nfkbie Ptgdr Slc10a6 Tlr9

LPS_1_d8 Nfkbiz Ptger1 Slc11a1 Tmem14c

LPS_2_d9 Nfyb Ptger2 Slc25a24 Tnf

LPS_6_h10 Nkrf Ptger4 Slc36a2 Tnfaip3

Lrg1 Nos2 Ptges Slc7a2 Tnfrsf11a

Lta Npm1 Ptges2 Slc7a2 Tnfrsf11b

Ltb4dh Nr4a1 Ptgir Slpi Tnfrsf18

Ltb4r2 Nrp Ptgis Smad3 Tnfrsf1a

Ltbp1 Nudt7 Ptgs1 Soat1 Tnfrsf1b

Ltbp2 Oasl1 Ptgs2 Socs1 Tnfrsf5

Ltbp3 Oasl2 Rab3d Socs3 Tnfrsf6

Ltc4s OVA_2_a7 Rai14 Sp1 Tnfrsf8

Ly6i OVA_2_e3 Ramp2 Spcs3 Tnfsf11

Lyzs OVA_3_c11 Rap2b Spen Tnfsf18

Malt1 P2rx4 Rbmx Spp1 Tnfsf4

Map2k6 Pbef1 Rel Src Tnfsf5

Map3k8 Pcdh7 Rela Srf Tnfsf6

Mapk14 Pcna Rest Stat1 Tnfsf8

MGC6827 Pdcd1 Retnla Stat1 Tnfsf9

Mgl1 Pdcd1lg1 Retnla Stat1 Tnip1

Mgl2 Pdcd1lg1 Rgs2 Stat3 Tnip1

Mif Pdcd1lg2 Rin2 Stat4 Tnip2

Mif Pdgfa Rnase4 Stat5a Tns

Mllt7 Pdia6 Rnasel Stat5b Trf

Mmp12 Pdk1 Rnf128 Stat6 Trp53

Mmp12 Peli1 Rnf13 Stmn1 Twist2

Mmp13 Peli2 Rnf141 Tank Unc119

Mmp14 Pep4 Rps3 Taok1 Usp18

Mmp19 Pfkfb2 Rsu1 Tbc1d20 Vegfa


References

Amara JF, Clackson T, Rivera VM, Guo T, Keenan T, Natesan S, Pollock R, Yang W,

Courage NL, Holt DA, Gilman M. 1997. A versatile synthetic dimerizer for the

regulation of protein-protein interactions. Proc Natl Acad Sci U S A 94(20):10618-

10623.

Aravind L, Dixit VM, Koonin EV. 1999. The domains of death: evolution of the apoptosis

machinery. Trends Biochem Sci 24(2):47-53.




Mol Biol Cell 15(3):1089-1100.





Cell 12(4):919-930.

Cvetanovic M, Ucker DS. 2004. Innate immune discrimination of apoptotic cells: repression

of proinflammatory macrophage transcription is coupled directly to specific

recognition. J Immunol 172(2):880-889.

Devitt A, Moffatt OD, Raykundalia C, Capra JD, Simmons DL, Gregory CD. 1998. Human

CD14 mediates recognition and phagocytosis of apoptotic cells. Nature

392(6675):505-509.

Golpon HA, Fadok VA, Taraseviciene-Stewart L, Scerbavicius R, Sauer C, Welte T, Henson

PM, Voelkel NF. 2004. Life after corpse engulfment: phagocytosis of apoptotic cells

leads to VEGF secretion and cell growth. Faseb J 18(14):1716-1718.

Gordon S, Taylor PR. 2005. Monocyte and macrophage heterogeneity. Nat Rev Immunol

5(12):953-964.

Gossen M, Bujard H. 1992. Tight control of gene expression in mammalian cells by

tetracycline-responsive promoters. Proc Natl Acad Sci U S A 89(12):5547-5551.

Gossen M, Freundlieb S, Bender G, Muller G, Hillen W, Bujard H. 1995. Transcriptional

activation by tetracyclines in mammalian cells. Science 268(5218):1766-1769.

Gratchev A, Schledzewski K, Guillot P, Goerdt S. 2001. Alternatively activated antigen-

presenting cells: molecular repertoire, immune regulation, and healing. Skin

Pharmacol Appl Skin Physiol 14(5):272-279.

Huber J, Vales A, Mitulovic G, Blumer M, Schmid R, Witztum JL, Binder BR, Leitinger N.

2002. Oxidized membrane vesicles and blebs from apoptotic cells contain biologically

active oxidized phospholipids that induce monocyte-endothelial interactions.

Arterioscler Thromb Vasc Biol 22(1):101-107.

Kagan VE, Gleiss B, Tyurina YY, Tyurin VA, Elenstrom-Magnusson C, Liu SX, Serinkan

FB, Arroyo A, Chandra J, Orrenius S, Fadeel B. 2002. A role for oxidative stress in

apoptosis: oxidation and externalization of phosphatidylserine is required for

macrophage clearance of cells undergoing Fas-mediated apoptosis. J Immunol

169(1):487-499.

Kawagishi C, Kurosaka K, Watanabe N, Kobayashi Y. 2001. Cytokine production by

macrophages in association with phagocytosis of etoposide-treated P388 cells in vitro

and in vivo. Biochim Biophys Acta 1541(3):221-230.

Leitinger N. 2003. Oxidized phospholipids as modulators of inflammation in atherosclerosis.

Curr Opin Lipidol 14(5):421-430.

170 Addendum I

Li P, Garcia GE, Xia Y, Wu W, Gersch C, Park PW, Truong L, Wilson CB, Johnson R, Feng

L. 2005. Blocking of monocyte chemoattractant protein-1 during tubulointerstitial

nephritis resulted in delayed neutrophil clearance. Am J Pathol 167(3):637-649.

Mantovani A, Sica A, Sozzani S, Allavena P, Vecchi A, Locati M. 2004. The chemokine

system in diverse forms of macrophage activation and polarization. Trends Immunol

25(12):677-686.

Mantovani A, Sozzani S, Locati M, Allavena P, Sica A. 2002. Macrophage polarization:

tumor-associated macrophages as a paradigm for polarized M2 mononuclear

phagocytes. Trends Immunol 23(11):549-555.

Matsura T, Togawa A, Kai M, Nishida T, Nakada J, Ishibe Y, Kojo S, Yamamoto Y, Yamada

K. 2005. The presence of oxidized phosphatidylserine on Fas-mediated apoptotic cell

surface. Biochim Biophys Acta 1736(3):181-188.

McCutcheon JC, Hart SP, Canning M, Ross K, Humphries MJ, Dransfield I. 1998. Regulation

of macrophage phagocytosis of apoptotic neutrophils by adhesion to fibronectin. J

Leukoc Biol 64(5):600-607.

Moldovan L, Moldovan NI. 2005. Role of monocytes and macrophages in angiogenesis.

Exs(94):127-146.

Muzio M, Chinnaiyan AM, Kischkel FC, O'Rourke K, Shevchenko A, Ni J, Scaffidi C, Bretz

JD, Zhang M, Gentz R, Mann M, Krammer PH, Peter ME, Dixit VM. 1996. FLICE, a

novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95

(Fas/APO-1) death--inducing signaling complex. Cell 85(6):817-827.

Nakano T, Ishimoto Y, Kishino J, Umeda M, Inoue K, Nagata K, Ohashi K, Mizuno K, Arita

H. 1997. Cell adhesion to phosphatidylserine mediated by a product of growth arrest-

specific gene 6. J Biol Chem 272(47):29411-29414.

Parks WC, Shapiro SD. 2001. Matrix metalloproteinases in lung biology. Respir Res 2(1):10-

19.

Shinohara K, Shinohara T, Mochizuki N, Mochizuki Y, Sawa H, Kohya T, Fujita M, Fujioka

Y, Kitabatake A, Nagashima K. 1996. Expression of vascular endothelial growth

factor in human myocardial infarction. Heart Vessels 11(3):113-122.

Vanden Berghe T, van Loo G, Saelens X, Van Gurp M, Brouckaert G, Kalai M, Declercq W,

Vandenabeele P. 2004. Differential signaling to apoptotic and necrotic cell death by

Fas-associated death domain protein FADD. J Biol Chem 279(9):7925-7933.

Addendum II: Additional publication

More than one way to die: methods to determine TNF-induced apoptosis and necrosis.

T Vanden Berghe, G Denecker, G Brouckaert, DV Krysko, K D’Herde and P VandenabeeleIn Methods in Molecular Medicine, Vol. 98, 2004: Tumor Necrosis Factor. Edited by: A Corti and P Ghezzi, Humana Press Inc.

More Than One Way to Die 101

More Than One Way to DieMethods to Determine TNF-Induced Apoptosis and Necrosis

Tom Vanden Berghe, Geertrui Denecker, Greet Brouckaert, DmitriVadimovisch Krysko, Katharina D’Herde, and Peter Vandenabeele

SummaryIn most cellular systems tumor necrosis factor (TNF) induces apoptotic cell death. How-

ever, in some particular cell lines, such as the L929sA fibrosarcoma, TNF induces necrotic celldeath. This effect is not the result of an inability to die apoptotically, because triggering of Fasin L929sAhFas cells leads to apoptosis. Moreover, TNFR-1-induced necrosis can be revertedto apoptosis when cells are pretreated with geldanamycin, an Hsp90 inhibitor. In contrast, addi-tion of caspase-inhibitors (zVAD-fmk) prevents Fas-induced apoptosis and switches it tonecrosis. These results demonstrate that depending on the cellular context, the same stimuluscan induce either apoptosis or necrosis. Apoptosis and necrosis are clearly distinguished bytheir morphology, although in the absence of phagocytosis, the late stage of apoptosis is asso-ciated with secondary necrotic cell death, which is hard to distinguish from necrotic cell death.Necrosis is described mostly in negative terms as cell death that is characterized by the absenceof apoptotic parameters, such as caspase activation, cytochrome c release, and DNA fragmen-tation. Here we describe a selection of techniques used to distinguish both modes of TNFR-1-induced cell death, namely apoptotic or necrotic cell death.

Key Words: Tumor necrosis factor; Fas; apoptosis; necrosis; caspases.

1. IntroductionIn contrast to the overflow of reports on the molecular mechanisms of

apoptotic (type I) programmed cell death (PCD), reports on the molecularmechanisms of autophagic (type II) and necrotic (type III) PCD are sparse. Allthree distinct types of cell death have been observed in the developing embryo(1). In brief, these morphological changes include cell shrinkage and extensivechromatin condensation in type I PCD; formation of autophagic vacuoles inside

171

the dying cell in type II PCD; and rapid loss of plasma membrane integrity andspillage of the intracellular content in type III PCD (2). Depending on the stimu-lus and the cellular context, one distinct cell-death program will becomeapparent, most probably because every cell-death program is a result of self-propagating signals and of signals that suppress the other cell death programs (3).

Apoptosis is morphologically characterized by membrane blebbing, shrink-ing of the cell and its organelles, internucleosomal degradation of DNA, anddisintegration of the cell, after which the fragments are phagocytized by neigh-boring cells (4,5). Necrosis is characterized by swelling of the cell and theorganelles, and results in disruption of the cell membrane and lysis (6,7). Oneof the cell lines intensively studied in our laboratory is the L929 fibrosarcomacell line. In these cells, apoptotic as well as necrotic cell death can be induced.Stimulation of TNFR-1 in L929sA cells leads to necrotic cell death. This pro-cess is strongly enhanced by pretreatment with the pancaspase inhibitor zVAD-fmk or overexpression of CrmA (8). In L929sAhFas cells transfected withhuman Fas, the apoptotic cell-death pathway can be induced by clustering ofFas with agonistic anti-Fas antibodies. This apoptotic cell-death process can bereverted to necrosis by inhibiting caspase activity (9).

The signal transduction pathways of TNFR-1- and Fas-mediated apoptosishave been extensively studied (10). Triggering of TNFR-1 and Fas leads to aFADD-mediated recruitment and activation of caspase-8. The downstreamexecutioner caspase-3 is activated directly by caspase-8, which results inso-called type I apoptotic cell death (11). In the case of type II apoptotic celldeath, caspase-3 activation is propagated by a mitochondrial loop. In this lattercase the mitochondrial cytochrome c contributes to the activation of caspase-3through the formation of the apoptosome complex leading to the activation ofcaspase-9 and caspase-3 (12). In both type I and type II apoptotic cell-deathpathways there is release of cytochrome c, but in the former pathway it is notrequired for the activation of caspase-3, while in the latter pathway it is. TypeI apoptotic cell death is not sensitive to Bcl-2 overexpression, while type II is(11). A molecular link connecting the death-receptor-mediated activation ofcaspase-8 with the release of mitochondrial factors is provided by the caspase-8-mediated cleavage of Bid (13). The C-terminal part of Bid translocates to themitochondria, where it induces release of cytochrome c (14) and other mito-chondrial factors, such as endonuclease G, Smac/DIABLO, and the serine pro-tease Omi/HtrA2 (15). These mitochondrial factors contribute in different waysto the activation of the downstream executioner caspases (16), eventually lead-ing to the cleavage of different substrates and causing the typical morphologi-cal and biochemical features of apoptotic cell death (17).

As mentioned above, one way to induce caspase-independent necrotic celldeath is to treat cells with an apoptotic stimulus in the presence of caspase

172 Addendum II

inhibitors (9). A. Kawahara and collaborators (18) showed that enforced dimer-ization of FADD in Jurkat cells treated with zVAD-fmk or in Jurkat cells defi-cient for caspase-8 resulted in a necrotic cell death, indicating that FADD mayalso function as an adapter to necrotic cell death. N. Holler and colleagues (19)demonstrated that necrotic cell death induced by FasL in the presence ofcaspase inhibitors is absent in Jurkat cells deficient for RIP, suggesting a rolefor the latter as adapter in necrotic cell death. Moreover, pretreatment of wild-type Jurkat cells with geldanamycin (GA), an inhibitor of Hsp90, protects thesecells from FasL-mediated caspase-independent necrotic cell death in the pres-ence of zVAD-fmk (19). Pretreatment of cells with GA results in the downmodulation of RIP, confirming its possible role in necrotic cell death. We dem-onstrated that in L929 cells TNFR-1-induced necrosis switches to apoptosis bypretreatment with Hsp90 inhibitors (20). We conclude that necrotic cell deathis also governed by defined signal transduction pathways leading to mitochon-drial production of reactive oxygen radicals (6) and the release of lysosomalcathepsins (Festjens, N., et al., in preparation). Remarkably, both types of celldeath are interchangeable depending on the cellular constellation of the L929sAcells: Anti-Fas-induced apoptosis is reverted to necrosis in the presence ofcaspase inhibitors (9), while TNFR-1-induced necrosis is reverted to apoptosisin the presence of Hsp90 inhibitors (20).

The problem with the detection of necrotic cell death is that other than thedistinct morphology, no distinct biochemical parameter has been identified thatunambiguously and positively differentiates necrotic from apoptotic death.Even the morphological evaluation of dying cells can be misleading, becauseapoptotic cells in the absence of phagocytosis proceed to a stage called second-ary necrosis. Secondary necrotic cells resemble the necrotic ones, but they havegone through an apoptotic stage; therefore, it is generally advisable in cell-death research to perform time-kinetics experiments of the cell-death param-eters. Necrotic cell death is described mostly in negative terms as cell deaththat is characterized by the absence of apoptotic parameters, such as caspaseactivation, cytochrome c release, and DNA fragmentation. Here we describe aselection of techniques used to distinguish both modes of TNFR-1-inducedcell death, namely apoptotic or necrotic cell death.

2. Materials1. Recombinant human TNF: produced in Escherichia coli and purified to at least

99% homogeneity. Specific biological activity 2.3 × 107 IU/mg as determined ina standardized cytotoxicity assay on L929sA cells.

2. Agonistic anti-Fas antibody (BioCheck GmbH): prepared as a 500-µg/mL stocksolution; use at 250 ng/mL.

3. Geldanamycin (Sigma): prepared as a 1 mM stock solution in DMSO; use at 1 µM.4. Propidium iodide (Sigma): prepared as a 3 mM stock solution in PBS; use at 30 µM.

Additional publication 173

5. Annexin V-binding buffer: 10 mM HEPES NaOH, pH 7.4, 150 mM NaCl, 5 mMKCl, 1 mM MgCl2, 1.8 mM CaCl2.

6. Annexin V-fluorescein isothiocyanate (FITC, Molecular Probes): use at 1 µg/mL.7. Acridine orange (Sigma): prepared as a 10 mg/mL stock solution; use at 5 µg/mL.8. PBS: 8 g/L NaCl, 0.2 g/L KCl, 2.89 g/L Na2HPO4•12H2O, 0.2 g/L KH2PO4.9. Caspase lysis buffer: 1% NP40, 10 mM Tris-HCl, pH 7.4, 200 mM NaCl, 5 mM

EDTA, 10% glycerol, 1 mM PMSF, 0.3 mM aprotinin, 1 mM leupeptin.10. Cell-free system buffer (CFS-buffer): 10 mM HEPES NaOH, pH 7.4, 220 mM

mannitol, 68 mM sucrose, 2 mM MgCl2, 2 mM NaCl, 2.5 mM KH2PO4, 0.5 mMEGTA, 0.5 mM sodium pyruvate, 0.5 mM L-glutamine.

11. Antibodies: anti-cytochrome oxidase subunit IV (COX) (Molecular Probes), anti-cytochrome c (Pharmingen), anti-Bid (R&D Systems), anti-caspase-2 (gift from Dr.Sharad Kumar, Adelaide, Australia), anti-human caspase-3 (BioSource), anti-humancaspase-6, -7, and -9 (StressGen Biotechnologies Corp.), antihuman caspase-8(clone 12F5, BioSource), antimouse caspase-8 (Santa Cruz), antimouse caspase-9(Cell Signaling Technology). Rabbit polyclonal antibodies raised against recom-binant murine caspase-3, -6, and -7 were prepared at the Centre d’EconomieRurale (Laboratoire d’Hormonologie Animale, Marloie, Belgium). HRP-coupledsecondary antibodies were obtained from Amersham Life Science (Amersham).

12. 5X LaemmLi buffer: 312.5 mM Tris-HCl, pH 6.8, 10% SDS, 50% glycerol, 20%-mercaptoethanol.

13. Cathepsin and calpain extraction buffer: 20 mM HEPES NaOH, pH 7.5, 0.02%digitonin, 250 mM sucrose, 10 mM KCl, 1.5 mM MgCl2, 1 mM EDTA, 1 mMEGTA, 1 mM pefablock (Boehringer-Mannheim).

14. Cathepsin reaction buffer: 50 mM sodium acetate, pH 6.0, 4 mM EDTA, 8 mMDTT, 1 mM pefablock (Boehringer-Mannheim).

15. Cathepsin L reaction buffer: 4M urea, 20 mM sodium acetate, pH 5.0, 4 mMEDTA, 8 mM DTT, 1 mM pefablock.

16. Calpain reaction buffer: 50 mM Tris-HCl, pH 7.5, 8 mM DTT.17. Caspase fluorogenic substrates: Ac-DEVD-AMC, Ac-LEHD-AMC, Ac-YVAD-

AMC, Ac-IETD-AMC, and Ac-WEHD-AMC (Peptide Institute, Osaka, Japan)prepared as 100 mM stock solution in DMSO; use at 50 µM.

18. Cathepsin and calpain fluorogenic substrates: zRR-AMC (Calbiochem), zFR-AFC (Enzyme Systems Products) and succinyl-LLVY-AMC (Bachem) preparedas 10 mM stock solutions in DMSO; use at 50 µM.

19. Transmission electron microscopic fixation buffer (TEM fixation buffer): 0.1 Msodium cacodylate, pH 7.4, 2% glutaraldehyde, 1 mM CaCl2.

20. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT): preparedas a 5 mg/mL stock solution in PBS; use at 500 µg/mL.

21. SDS-HCl buffer: 10% SDS, 0.01 N HCl.

3. MethodsThe methods described below outline a variety of techniques that have been

used to distinguish between apoptotic and necrotic cell death in the same cellu-

174 Addendum II

lar context. These techniques rely on specific morphological and biochemicalchanges associated with these two processes. As an example, we will use theL929sA fibrosarcoma cell line. In these cells, apoptotic as well as necrotic celldeath can be induced. Stimulation of TNFR-1 in L929sA cells leads to necroticcell death. When the cells are pretreated with geldanamycin (GA) prior to expo-sure to TNF, the cells die in an apoptotic way (20). In L929sA cells transfectedwith human Fas (L929sAhFas), the apoptotic cell-death pathway is induced byclustering of Fas with agonistic anti-Fas antibodies (21). This apoptotic cell-death process is reverted to necrosis by inhibiting caspase activity (9). Thus,both death receptors TNFR-1 and Fas can induce apoptotic and necrotic celldeath, depending on the cellular constitution.

3.1. The Use of Transmission Electron Microscopy to StudyApoptotic and Necrotic Cells

Transmission electron microscopic (TEM) analysis has been considered a“golden standard” in the field of cell-death research. Using Fourier methods, itprovides a two- and three-dimensional image of the inner cell and therebyallows the understanding of biological structure–function relationships at(sub)cellular and molecular levels (22,23). Compared to light microscopy,TEM is time consuming and requires more expensive equipment, but it offersmuch higher resolving power (0.1–0.4 nm). For that reason, TEM providesmuch more detailed information about cellular morphology and is thereforethe most accurate method for distinguishing apoptosis and necrosis in cell cul-tures. Preparation of dying cells for electron microscopy may be difficultbecause dying cells typically detach from their substrate, and spinning downthese floating cells may cause damage to their original morphology. We havecircumvented this problem by using macrophages to capture dying apoptoticand necrotic cells and in this way attaching them to the bottom of the tissueculture plate while they are being phagocytized.

1. Seed macrophages in adherent 6-well plates at 5 × 105 cells per well and targetcells in uncoated 6-well suspension plates at 5 × 105 cells per well (see Note 1).

2. Induce target cells (e.g., L929sAhFas) to undergo apoptosis with, for example,agonistic anti-Fas antibody (250 ng/mL, for at least 1 h), or necrosis with, forexample, hTNF (10,000 IU/mL, for at least 7 h) and collect them for co-incuba-tion with the macrophages. The chosen time points depend on the kinetics of celldeath.

3. Coculture macrophages and target cells (in a 1:1 ratio) at 37°C, 5% CO2, for atleast 1 h.

4. Fix cocultures of macrophages and target cells in the 6-well plate by immersionin TEM fixation buffer overnight at 4°C.

5. Rinse 3X for 5 min in 0.1M sodium cacodylate containing 7.5% w/v sucrose.


6. Postfix in 1% w/v OsO4 in the same buffer (without sucrose) overnight at 4°C.7. Rinse 3X for 5 min in 100 mM sodium cacodylate containing 7.5% w/v sucrose.8. Dehydrate in a graded series of ethanol: 50% for 15 min, 70% for 20 min, 90%

for 30 min, and 100% for 90 min.9. Infiltrate with 100% ethanol: LX-112 resin (Ladd, Burlington, VT) (1:1 for 30

min, 1:2 for 30 min, and 100% LX-112 resin for 120 min).10. Polymerize for at least 48 h at 60°C.11. Break off the plastic of the polymerized block and saw into pieces to fit the ultra-

microtome holders. Remove any remaining plastic bits from the cutting surface.12. Make ultrathin sections of 60 nm with a diamond knife and mount on 100-mesh

formvar-coated copper grids.13. Stain with uranyl acetate (7.5% in double-distilled water, 1 drop per grid for 20

min) and Reynold’s lead citrate (1 drop per grid for 10 min).14. Examine the sections by TEM.

The earliest classic ultrastructural changes detectable in apoptosis are thecondensation of chromatin to form uniformly dense masses that lay against thenuclear envelope (24) and the persistence of nucleolar structure until the verylate stages (25). Further features displayed by most apoptotic cells include lossof specialized surface structures, such as microvilli and cell–cell contacts. Thecell volume decreases and the cell density increases due to the marked conden-sation of cytoplasm. Membrane-bound apoptotic bodies of varying size areformed that contain well-preserved but compacted cytoplasmic organelles and/or nuclear fragments (24,26) (Fig. 1B). In contrast, no cell fragmentationappears in the course of necrosis, but a general cell hydration occurs, causingswelling of the organelles (endoplasmic reticulum, mitochondria). The cyto-plasm becomes increasingly translucent and contains a mottled-looking nucleusdue to condensation of nuclear chromatin in ill-defined masses. Finally, thereis loss of plasma membrane integrity, and cell contents starts to leak out pro-voking inflammation (Fig. 1C).

3.2. The Use of Flow Fluorocytometry (FACS) to Study Apoptoticor Necrotic Cell Death

In this section, we will describe the use of flow fluorocytometry to distin-guish between apoptotic and necrotic cell death by analyzing cell membranepermeability, cell morphology, phosphatidylserine (PS) exposure, DNA frag-mentation, and lysosomal destabilization (see Note 2). For the different proto-cols described below, we first describe incubation and stimulation conditions.

1. Seed cells at 1.5 × 105 cells/mL per well the day before analysis in uncoated 24-wellsuspension plates, to avoid attachment of the cells to the bottom (see Note 3).

2. The next day, stimulate the cells with TNF (10,000 IU/mL), geldanamycin (1 µM,

176 Addendum II

Fig. 1. Ultrastructural features of apoptotic and necrotic mouse L929sAhFas fibro-sarcoma cells. Unstimulated cells (A) and cells exposed either to agonistic anti-Fas for1 h (B) or to hTNF for 18 h (C). (A) The cell shows microvilli protruding from theentire surface, a smoothly outlined nucleus with chromatin in the form of heterochro-matin, and well-preserved cytoplasmic organelles. (B) Apoptotic cell with sharplydelineated masses of condensed chromatin, convolution of the cellular surface, andformation of apoptotic bodies. Note the nucleolus (arrow) present near a cup-shapedchromatin margination. (C) Necrotic cell in close vicinity to a macrophage (asterisk),showing ill-defined edges of the clumps of compacted chromatin, swollen mitochon-dria with matrix densities, dissolution of membranes, and loss of plasma membraneintegrity. Scale, bars: 1 µm.

18 h pretreatment) plus TNF (10,000 IU/mL), or anti-Fas (250 ng/mL), and ana-lyze the cell samples from the suspension cultures at regular time intervals on aflow fluorocytometer (see Note 4).


3.2.1. Analysis of Cell Morphology vs Cell Permeability

Many studies report that apoptotic dying cells are smaller and denser thantheir living counterparts (4,27). Initially, the membrane of the cell starts toform blebs, which become separated from the main cell body. This phenom-enon is followed by the shrinkage of the entire nucleus or, in other cases, bud-ding outward of chromatin into smaller apoptotic bodies (28). These changesare accurately detected in the majority of cells by their light-scattering proper-ties as measured by flow fluorocytometry. The forward-scatter reflects cellsize, while the sideward-scatter reveals the degree of granularity of the cell.The protocol described below uses flow fluorocytometry to access changes incell size and granularity as well as the loss of cell membrane integrity mea-sured by propidium iodide (PI) uptake (see Note 5).

1. Transfer 300 µL of cells in suspension from the 24-well suspension plate to a 5-mLpolypropylene round-bottom tube, and add 3 µL PI from a 3-mM stock solution.Keep cells on ice for 3–10 min (see Note 6).

2. Set up the flow fluorocytometer for forward-scatter and sideward-scatter both atlinear scale on a dot plot to determine cell size and granularity (Fig. 1). Create ahistogram in FL3 to detect PI uptake at 610 nm, e.g., on a FACS Calibur flowfluorocytometer (Becton Dickinson) equipped with a water-cooled argon-ionlaser at 488 nm (see Note 7). To clearly see the dead cells on the dot plot, one cangate PI positive cells in red, for example, in the FL3 histogram.

3. After measurement, the tubes with the remaining suspension cells (at least 150µL) or freshly prepared tubes (300 µL, see step 1) can be used to determine thepercentage of cells with hypoploid DNA (see Subheading 3.2.2.).

4. The region of analysis is gated (R1) (see Note 8). At time point zero, the viablecell population is detected within the region of analysis; these cells are PI nega-tive. In the case of TNFR-1-induced apoptosis in L929 cells (GA/TNF, Fig. 2A-D),the membrane of the cells starts to bleb at 2 h, causing increased diffraction of thelaser beam, revealed by spreading of the dots (Fig. 2B). The release of apoptoticparticles is detected by the occurrence of a population of small-sized PI-negativedots (negative because of the absence of DNA) in the lower-left corner of the dotplot (Fig. 2B). The condensation of the nucleus, shrinkage of the cells, and for-mation of DNA-containing apoptotic bodies is visualized by the population ofPI-positive gray dots, appearing in the lower-left corner on the dot plot (Fig. 2C).The population in Fig. 2D illustrates the end stage of the apoptotic process,namely secondary necrosis. This population clearly colocalizes with the necroticcell death population induced by TNF (Fig. 2H).

3.2.2. Analysis of DNA Fragmentation

DNA fragmentation is a hallmark of apoptosis and involves the formation ofhigh molecular weight (>50 kbp) and nucleosome-sized (200 bp) DNA frag-ments (29,30). An easy and quantitative way to analyze DNA fragmentation is

178 Addendum II

More Than One Way to Die 109

Fig. 2. Analysis of cell morphology versus cell permeability by flow fluorocytom-etry. Cells were analyzed by flow fluorocytometry for changes in cell size (side-scat-ter) and granularity (forward-scatter) as well as the loss of membrane integrity bypropidium iodide uptake (FL3). L929 fibrosarcoma cells were pretreated for 18 h withor without geldanamycine (GA), followed by treatment with human TNF, as indi-cated. The differential changes in morphology between apoptotic and necrotic cellscan be followed in time on the dot plots (side- vs forward-scatter). The appearance ofPI-positive cells in time is indicated in gray on the dot plots.


by adding PI to the dying cell population and applying one freeze–thaw cycleto permeabilize cells and cell fragments. PI intercalates in the DNA, and thesize of DNA fragments appears as a hypoploid DNA histogram.

1. Transfer 300 µL of cells in suspension from the 24-well suspension plate to a 5-mLpolypropylene round-bottom tube, and add 3 µL PI from a 3 mM stock solution.

2. Subject the tubes to one freeze–thaw cycle by putting them shortly on liquidnitrogen. As a result of the freeze–thaw cycle, cells are permeabilized and becomePI positive. Alternatively, use the remaining suspension cells (at least 150 µL)that were used for measurement of PI uptake in Subheading 3.2.1.

3. Analyze samples after thawing (see Note 9). Use the same experimental set up asdescribed in Subheading 3.2.1.

Viable nondividing cells exhibit a clear diploid DNA peak (2n, G1), whereasdividing cells give rise to a tetraploid peak (4n, G2). A nonsynchronized popu-lation of cells shows a typical biphasic peak of 2n (G1) and 4n (G2) cells.During apoptosis, the DNA is fragmented and is partially lost as a result of theformation of apoptotic bodies. Therefore, apoptotic cells typically show a lowerDNA fluorescence pattern as compared to necrotic cells that maintain theirentire DNA content in the nucleus. Hence apoptotic cells with fragmented DNAappear as cells with a hypoploid DNA content and are represented as a hypo-ploid “sub-G1” peak. An example of a hypoploid DNA pattern as a result ofTNFR-1-induced apoptosis in the presence of GA is shown in Fig. 3B. InTNFR-1-induced necrosis, this typically apoptotic event is absent, as illustratedin Fig. 3D.

3.2.3. Analysis of PS Exposure vs Cell Permeability

Plasma membranes of viable cells exhibit significant phospholipid asym-metry, with most of the phosphatidylserine (PS) residing on the inner, cyto-plasmic membrane leaflet. PS is kept in the inner membrane by the rapidinward-moving action of an ATP-dependent phospholipid transporter, theaminophospholipid translocase. Early in the apoptotic process, this translocaseis inhibited, while another, bidirectional aminophospholipid transporter, calledscramblase, becomes activated (31). As a result, phospholipid asymmetry islost, and cells increasingly expose PS on their outer membrane leaflet (32).Annexin V, a Ca2+-dependent phospholipid-binding protein, preferentiallybinds to negatively charged phospholipids like phosphatidylserine and there-fore is a useful tool to detect cells at the early stages of the apoptotic process(33). Using FITC-conjugated Annexin V in combination with PI uptake, thedifferential appearance between PS exposure and the loss of membrane integ-rity allows one to distinguish between apoptotic and necrotic cell death (6) (seeNote 10). Apoptotic cells are characterized by a lag period between PS positiv-ity and PI positivity. In necrotic cell death, both events coincide.

180 Addendum II

1. Transfer 1.5–3 × 105 cells to an Eppendorf tube, centrifuge the cells for 5 min at250g at 4°C, and resuspend in 1 mL ice-cold Annexin V-binding buffer (see Notes6 and 11).

2. Centrifuge the cells for 5 min at 250g at 4°C and resuspend in 300 µL of AnnexinV-binding buffer containing 1 µg/mL of Annexin V-FITC. Incubate for 5 min onice and protect from light.

3. Add 3 µL from a 3-mM PI stock solution to the same 300-µL cell suspension for3–10 min for analysis.

4. Analyze samples by setting up the flow cytometer in the same way as describedin Subheading 3.2.1. Create an extra dot plot showing Annexin V-FITC (FL1)versus PI (FL3), and set up quadrants dividing Annexin V-positive, AnnexinV-negative, PI-positive, and PI-negative populations (see Note 12).

Membrane changes leading to PS exposure occur rapidly in TNFR-1-induced apoptotic cell death. In the presence of GA, the cell population shiftsfrom the lower-left quadrant (PI-negative and Annexin V-negative cells, Fig.4A) to the lower-right quadrant (PI-negative and Annexin V-positive cells,Fig. 4B). After PS exposure to the outer leaflet of the cell membrane, cells start

Fig. 3. Analysis of DNA fragmentation by flow fluorocytometry. Cells were ana-lyzed by flow fluorocytometry for the detection of DNA hypoploidy. L929 fibrosar-coma cells were pretreated for 18 h with or without geldanamycin (GA), followed bytreatment with human TNF, as indicated. The percentage of cells containing hypop-loid DNA is the population of cells with lower fluorescence than the G1, 2n peak.


112 Vanden Berghe et al.

Fig. 4. Analysis of PS exposure versus cell permeability by flow fluorocytometry.Cells were analyzed by flow fluorocytometry for loss of membrane integrity by PIuptake and translocation of phophatidylserine to the outer leaflet of the plasma membraneby staining with Annexin V-FITC. L929 fibrosarcoma cells were pretreated for 18 h withor without geldanamycin (GA), followed by treatment with human TNF, as indicated.

182 Addendum II

losing their membrane integrity, and the population shifts to the upper-rightquadrant (PI-positive and Annexin V-positive cells, Fig. 4C). During TNFR-1-induced necrotic cell death, PS exposure coincides with the permeabilizationof the outer cell membrane. Consequently, cells immediately move from thelower-left quadrant (PI-negative and Annexin V-negative cells, Fig. 4D) to theupper-right quadrant of the dot plot (PI-positive and Annexin V-positive cells,Fig. 4E,F) without passing through the intermediate PI-negative and AnnexinV-positive stage.

3.2.4. Analysis of Lysosomal Destabilization

During the last few years several reports have reprised the old concept thatlysosomes might be involved in cell death, both by necrosis as well as byapoptosis (34,35). The role of lysosomes in cell death is a topic of increasinginterest; therefore, we would like to describe a method to measure lysosomalstability. This observation is performed by following the uptake of the weakbase acridine orange (AO) (36). Due to proton trapping, this vital dye mainlyaccumulates in the acidic vacuolar apparatus, preferentially in secondary lyso-somes (see Note 13).

1. Stain the cells in the 24-well suspension plate by adding 5 µL of a 10-mg/mLstock solution of AO to 1.5–3 × 105 cells/mL, and incubate at 37°C for 15 min.

2. Wash with 1 mL PBS, centrifuge at 250g for 5 min and resuspend in PBS.3. Set up the flow fluorocytometer as described in Subheading 3.2.1. Create a his-

togram in FL3 to detect impairment of AO uptake. To clearly see the cells withdestabilized lysosomes on the dot plot, one can gate AO-negative cells in the FL3histogram (see Note 13).

Exciting AO with green light results in red fluorescence at high (lysosomal)concentrations. This effect is revealed on a histogram by the peak of cells withstill intact lysosomes (Fig. 5A). During TNFR-1-induced necrosis, the popula-tion of cells that have lost the capacity to accumulate AO in the lysosomesincreases (Fig. 5B–D). The percentage of cells with low intensity of red fluo-rescence is used as a marker for the extent of lysosomal destabilization (impair-ment of AO uptake). At the end stage, the lysosomes of all cells have lost thecapacity to retain AO (Fig. 5D).

3.3. Analysis of the Apoptotic Signal TransductionCascade by Western Blot

In this section we will describe the use of Western blot to detect the activa-tion of caspases (see Note 14), the cleavage of Bid to its truncated pro-apoptoticform (tBid), and finally by the release of cytochrome c from the mitochondria.These events are all absent in TNFR-1-induced necrotic cell death (6).


184 Addendum II

CARD, caspase recruitment domain), or (2) by cleavage via upstream proteasesin an intracellular cascade. The net result of these proteolytic activities is theseparation of the prodomain from the p20 and p10 subunits, which form anactive heterotetramer (17).

1. Follow the steps of Subheading 3.3.2. Centrifuge the cells for 5 min at 250g, aspirate PBS, and lyse the cells in 150 µL

cold caspase lysis buffer (see Note 6).3. Remove cell debris by centrifuging the lysate (10 min at 20,800g at 4°C) and

transfer the remaining cytosol to another Eppendorf tube. Add 1:5 volume of 5XLaemmLi buffer to the cell lysate and boil for 5–10 min at 98°C.

4. Load equal volumes per lane on 12.5% SDS-polyacrylamide gels (see Note 15).After separation, transfer to a nitrocellulose membrane and detect the differentcaspases with the appropriate antibodies. Reveal with chemiluminescence reagentplus (Nycomed Amersham plc).

Proteolytic activation of procaspases leads to the appearance of a typicalp20 and p10 subunit (see Note 16). Figure 6A illustrates the activation ofcaspase-3 and -7 by the appearance of the p20 subunit in Fas-induced apoptosis(procaspase-3, black arrow; p20 subunit, white arrow). In TNFR-1-inducednecrosis, p30 caspase-3 and -7 are not activated, and thus there is no appear-ance of a p20 subunit (Fig. 6B).

3.3.2. Analysis of Bid Cleavage and Cytochrome c Release

During death receptor-mediated apoptosis, the mitochondria are induced torelease cytochrome c through the proteolytic activation of Bid to truncated Bid(tBid), which translocates to the mitochondria. To establish the role of the mito-chondrial apoptotic pathway, one can detect the engagement of the mitochon-dria at the different stages: Bid proteolytic activation, tBid translocation, andcytochrome c release. In order to detect the release of cytochrome c in thecytosol, the organelle and cytosolic fraction of the cell should be separatedwithout the spontaneous release of cytochrome c from the mitochondria due toartifacts of the organelle preparation. Therefore, a mild detergent, digitonin, isused (see Note 17).

1. Follow the steps of Subheading 3.3.2. Centrifuge the cells for 5 min at 250g, remove PBS, and lyse in 100 µL 0.02%

digitonin dissolved in cell-free system buffer (CFS-buffer); leave on ice for 1min (see Note 6).

3. Centrifuge the lysate (10 min at 20,800g at 4°C) and transfer the cytosol to aseparate Eppendorf tube. Add 1:5 volume of 5X LaemmLi buffer to the cytosoliclysate and 100 µL 1X LaemmLi buffer to the remaining organelle fraction andboil for 5–10 min at 98°C.

4. Load equal volumes of the organelle fraction or cytosolic cell lysate per lane on


116 Vanden Berghe et al.

Fig. 6. Analysis of caspase activation, Bid cleavage, and cytochrome c release byWestern blot. (A) and (B) Time kinetic analysis of caspase-3 and -7 activation duringanti-Fas-induced apoptosis and TNFR-1-induced necrosis. Cytosolic fractions wereprepared and analyzed by PAGE and Western blot. Open and solid arrows indicateprocessed p20 subunit and precursor procaspase, respectively. (C) and (D) Timekinetic analysis of the proteolytic activation and delocalization of Bid from the cytosolto the organelle fraction during anti-Fas-induced apoptosis and TNFR-1-inducednecrosis. (E) and (F) Cellular distribution of cytochrome c in the cytosol and in theorganelle fraction. Cytosolic and organelle fractions were prepared as described andwere analyzed by Western blot. Solid and open arrows indicate full-length Bid andtBid, respectively.

186 Addendum II

12.5% SDS-polyacrylamide gels (see Note 15). After separation, transfer to anitrocellulose membrane by electroblotting, and detect Bid cleavage and cyto-chrome c release with the appropriate antibodies or antisera. Reveal with chemi-luminescence reagent plus (Nycomed Amersham plc).

Anti-Fas treatment results in an early cleavage of full-length Bid to its pro-apoptotic truncated form (tBid). Only for a short period and to a minor extent istBid detectable in the cytosolic fraction. The vast majority of tBid becomesrapidly associated with the organelle fraction (Fig. 6C). Therefore, the safestway to detect the engagement of Bid one should look both at the disappearanceof Bid from the cytosol and the appearance of tBid in the organelle fraction. Innecrotic cells, no cleavage of p22 Bid is detectable either in the cytosol or inthe organelle fraction (Fig. 6D). Only late in the necrotic process does Biddisappear from the cytosol due to cytosolic leakage. In addition, a small amountof tBid becomes detectable in the organelle fraction, probably because of alter-native Bid cleavage by lysosomal proteases (37). The appearance of tBid in theorganelle fraction during anti-Fas induced apoptosis coincides with the releaseof cytochrome c to the cytosol (Fig. 6C,E). We found that during secondarynecrosis of anti-Fas-stimulated cells, cytochrome c also accumulates in theculture supernatant (Fig. 6E). TNFR-1-induced necrosis does not result in adetectable release of cytochrome c in the cytosol. However, in the late necroticphase of TNF-stimulated cells, cytochrome c accumulates in the culture super-natant from the moment that cells lose their plasma membrane integrity (Fig. 6F).

3.4. Use of Fluorogenic Substrates to Distinguish Apoptoticor Necrotic Cell Death

To measure the actual activity of caspases instead of their proteolytic acti-vation state, one can use fluorogenic substrates for caspases, cathepsins, orcalpains. In the case of caspases, these substrates contain the minimal aminoacid composition corresponding with the cleavage site of a typical substrate,coupled to 7-amino-4-methylcoumarin (AMC) or 7-amino-4-trifluoromethyl-coumarin (AFC). Hydrolysis of the peptide substrate results in the release offree AMC or AFC, which are fluorescent and can be measured in a fluorometer.

3.4.1. Analysis of Caspase Activity Using Fluorogenic Substrates

The first synthetic substrates designed and used to measure and analyzecaspase activities were based on the cleavage sites of the initially identifiedcellular protein substrates (17,38). These synthetic substrates were usually atetrapeptide with aspartate at the position P1 (XXXD) conjugated to afluorogenic AMC or AFC (see Note 18); for example, Ac-DEVD-AMC is derivedfrom the caspase-3 cleavage site in poly-ADP ribosyl polymerase (PARP) andAc-YVAD-AMC is derived from the caspase-1 cleavage site in prol-IL-1 .



cold caspase lysis buffer (see Note 19). Add glutathione at a final concentrationof 1 mM (see Notes 6 and 20).

3. Remove cell debris by centrifuging the lysate (10 min at 20,800g at 4°C) andtransfer the cytosol to another Eppendorf tube (“cytosolic cell lysate”).

4. Measure caspase activity by incubating 25 µg cytosolic cell lysate with 50 µM ofthe fluorogenic substrate (e.g., Ac-DEVD-AMC) in 150 µL CFS-buffer. Adddithiothreitol (DTT) at a final concentration of 10 mM (see Note 20).

5. Monitor the release of fluorescent AMC for 1 h at 37°C at 2-min time intervals ina fluorometer (e.g., CytoFluor, PerSeptive Biosystems) using a filter with an exci-tation wavelength of 360 nm and a filter with an emission wavelength of 460 nm.Express data as the increase in fluorescence as a function of time ( fluorescence/min) (6,20).

6. Caspase activity is only detected in apoptotic conditions.

3.4.2. Analysis of Cytosolic Cathepsin or Calpain Activity UsingFluorogenic Substrates

Several reports show an involvement of calpain activity in necrotic orcaspase-independent cell death, for example, increased µ-calpain auto-pro-teolysis in necrotic fibers; a high Ca2+ requirement for necrotic cell death in C.elegans; and the activation of calpain I converting excitotoxic neuron deathinto caspase-independent cell death (39–41). Other reports show release ofcathepsin B in cytosol during TNFR-1-induced apoptosis, and during necrosisor under conditions in which caspases are blocked (34,42), or show theinvolvement of lysosomal cathepsins in an alternative cleavage of PARP (43)and Bid (37).


cathepsin and calpain extraction buffer; incubate cells for 5 min on ice (see Note 6).3. Analyze the enzyme activities by adding 50 µM zRR-AMC or zFR-AFC for

cathepsin B and L, respectively, or 50 µM succinyl-LLVY-AMC plus 2 mM Ca2+

for calpain measurements in appropriate reaction buffers: cathepsin reactionbuffer, cathepsin L reaction buffer, and calpain reaction buffer, respectively.

4. Monitor the release of AFC (excitation 409 nm, emission 505 nm) or AMC (exci-tation 360 nm, emission 460 nm) for 1 h at 30°C at 2 min time intervals in afluorometer (e.g., CytoFluor, PerSeptive Biosystems). Express data as theincrease in fluorescence as a function of time ( fluorescence/min).

3.5. Analysis of Cell Death on a Large Scale

In this part we describe two general and easy methods for determining theextent of cell death irrespective of the type of cell death. These methods allow

188 Addendum II

large screenings of conditions in 96-well plates, or optimization of concentra-tions of inhibitors or sensitizing reagents.

1. Seed cells the day before at 2 × 104 cells in 100 µL per well in 96-well plates.2. Expose the cells to a serial dilution of TNF (horizontal dilution added to the cells

in a 50 µL volume) in the absence or presence of different concentrations ofinhibitor or sensitizing agent (vertical dilution added to the cells in a 50 µL vol-ume) (see Note 21).

3.5.1. Lactate Dehydrogenase (LDH) Assay

Lactate dehydrogenase (LDH) is a cytosolic enzyme present within all mam-malian cells. The intact plasma membrane is impermeable to LDH, and loss ofits integrity is detectable by the release of LDH into the supernatant (44), whereits enzymatic activity is measured. In vitro release of LDH from cells providesan accurate measure of cell membrane integrity and cell viability.

1. Remove 50 µL supernatant of cells under study to a 96-well plate.2. Add 50 µL reconstituted substrate mixture (CytoTox 96™ Assay, Promega) (see

Note 22).3. Incubate for 30 min at room temperature.4. Add 50 µL Stop solution to each well and analyze by spectrophotometry at 492 nm.5. Results may be expressed either as optical absorbance values or as international

units of enzyme (calculated from a standard curve obtained with the positiveLDH control, which is provided with the system).

3.5.2. 3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide(MTT) Assay

MTT is a pale yellow substrate that is cleaved by living cells to yield a darkblue formazan product, which is precipitated out as crystals. This process iscatalyzed by succinate dehydrogenase complex II of the inner mitochondrialmembrane, an enzyme involved in oxidative phosphorylation. The assayreflects both the number of cells as well as the activity status of the mitochon-dria. Damaged mitochondria in dead cells are affected in their capacity tocleave significant amounts of MTT (see Note 23). The colorimetric assaydescribed thus measures both cytostatic as well as cytotoxic effects.

1. 18 h after stimulation add 20 µL of MTT solution to each well containing cells.2. Incubate the plate in a CO2 incubator at 37ºC for 4–6 h.3. Add 80 µL of SDS-HCl buffer to each well to dissolve the formazan crystals.4. Put the plate into the 37ºC incubator for an additional 6–8 h in order to dissolve

the crystals.5. Transfer the plate to the plate reader and measure absorbance at 595 nm (put

reference filter at 655 nm).6. The percentage of cell death (Pcd) is calculated using Equation 1:


P

A

At

A

Am

A

Au

A

Am

cd = ×1 100

595

655

595

655

595

655

595

655

where t indicates the absorbance ratio of treated cells, u indicates the absorbanceratio of untreated cells, and m indicates the absorbance ratio of the medium.

4. Notes1. The target cells are grown in suspension plates in order to easily transfer the

population of dying cells to the well of attached macrophages.2. We will not describe the use of mitochondrial membrane-potential-sensitive dyes

such as tetramethylrhodamine methyl, rhodamine123, JC-1 (Molecular Probes)or the use of dihydrorhodamine123 for measurement of oxygen radical species(ROS), because these methods do not distinguish the mode of cell death (45,46).In both cell-death pathways the membrane potential of the mitochondria drops,and the cells start producing ROS (6,47). In case of TNFR-1-induced cell deathin L929 cells, it was shown that although ROS production occurs in both celldeath pathways, it is only a crucial event in necrotic cell death (8,48).

3. To perform FACS analysis it is advisable to grow adherent cells in suspension ascommented in Note 1. In this way it is possible to work faster during kinetics andavoid interfering manipulations, such as trypsin/EDTA treatment, which couldlead to damage or permeabilization of the cells. Therefore, adherent cells areseeded on special uncoated suspension tissue culture plates (Sarstedt). It is impor-tant to test in advance whether the adherent cells allow growing in suspensionconditions.

4. As a general rule it is important to perform time kinetics when analyzing cell-death parameters. The end stage of apoptosis, the so-called secondary necrosis,displays the same features as necrosis, i.e., loss of membrane integrity and leak-age of proteins into the supernatant (6). Therefore, it is advisable to take samplesat regular time intervals in order to detect the differential appearance of apoptoticor necrotic features.

5. Propidium iodide (PI) is a membrane-impermeable dye that stains fluorescent byintercalating into nucleic acid molecules. In this way PI can be used to measurethe loss of cell-membrane integrity, as it enters the cell only if the cell membranebecomes permeable. When bound to nucleic acids, the absorption maximum forPI is 535 nm and the fluorescence emission maximum is 617 nm. PI can be excitedwith a xenon or mercury-arc lamp, or with the 488 line of an argon-ion laser. Inpractical terms, PI fluorescence is detected in the FL2 or FL3 channel of flowfluorocytometers. The broad emission spectrum of PI is a problem when combin-ing it with other fluorochromes. Only fluorochromes with a clearly distinct emis-sion spectrum or with a small emission-spectrum overlap can be used. The smalloverlap can be compensated for during FACS measurements.

190 Addendum II

6. Cells are kept on ice to stop further biochemical activity of the cell.7. A typical FACS Calibur flow fluorocytometer can analyze five different param-

eters. Two detectors detect the light scatter: (1) forward scattered light (FSC,indicates cell size) and (2) side scattered light (SSC, indicates granularity). Threeremaining photomultiplier tubes (PMTs) detect the fluorescent signals: (1) FL1(green fluorescence; used for FITC, R123, GFP), (2) FL2 (red fluorescence; usedfor PI, CMTMros, PE), and (3) FL3 (far red fluorescence, used for PI,Cy-Chrome).

8. The region of analysis (R1) must be large enough that the dying cell populationcan also be measured. This population shifts to the left on the dot plot as com-pared to the living cells.

9. After the freezing procedure in liquid nitrogen, the samples can be stored indefi-nitely at –20°C, until analyzed.

10. With the methods presented, it is not possible to distinguish whether a positiveAnnexin V result during necrosis is due to extracellular PS exposure or to intra-cellular detection of PS, because Annexin V can also enter the cell as a conse-quence of loss of cell membrane integrity (32). This problem can only be solvedusing immunogold staining methods and transmission electron microscopy analy-sis. However, the PS-dependent recognition of necrotic cells by phagocytic cellsstrongly argues for the occurrence of PS at the surface of necrotic dying cells(Brouckaert G, et al., in press).

11. The staining protocol absolutely must be carried out in Annexin V-binding bufferbecause the binding of this protein to the phosphatidylserine is calcium dependent.

12. Include also single-stained cells in your experiment, i.e., cells stained only withAnnexin V-FITC (no PI added) and cells stained only with PI (no AnnexinV-FITC added), to set up compensation and quadrant markers. The absorptionmaximum for FITC is 494 nm and the fluorescence emission maximum is 518 nm.

13. The absorption maximum for AO is 489 nm and the fluorescence emission maxi-mum is 520 nm. AO is also a metachromatic fluorophore. When excited by bluelight (440–470 nm) it shows red fluorescence at high (lysosomal) concentrationsand green fluorescence at low (nuclear and cytosolic) concentrations. If, how-ever, green excitation light (515–560 nm) is used, only concentrated (lysosomal)AO is observed by its red or orange fluorescence. Qualitative analysis is done byfluorescence or confocal microscopy. For this purpose cells need to be seeded onglass coverslips.

14. In addition to the activation pattern of different caspases, the cleavage pattern ofa variety of substrates can be detected, such as PARP and ICAD (17). Alterna-tively, caspase activation can be measured using fluorogenic caspase substrates(see Subheading 3.4.1.).

15. Different cell samples are lysed and processed in a standardized way with equalvolumes, which should reflect the identical input of the cells at the beginning ofthe experimental setup. One should not correct for protein content because atlater time points during the cell death process cells start to leak cytosolic proteinsinto the supernatant (6).


16. Depending on the antibody used, all proteolytic fragments or only one of themcan be detected. Antibodies are also available that specifically recognize the acti-vated form of caspases (PharMingen). These antibodies can also be used in FACSanalysis to detect caspase activity in cells; alternatively, cell-permeablefluorogenic substrates can be used (49). These antibodies also allow immuno-detection of active caspase in immunohistochemistry (50).

17. In L929 cells, 0.02% digitonin is capable of permeabilizing the plasma mem-brane, but it leaves the mitochondria and lysosomes intact. The concentration ofdigitonin used as well as the required lysis period is cell-type dependent. There-fore, a series of different concentrations of digitonin should be tested, to selectthe condition in which the outer membrane is permeabilized (by detection ofcytosolic proteins such as actin), while the mitochondrial or lysosomal mem-branes remain intact (no detection of mitochondrial proteins such as COX, orlysosomal proteins such as hexosaminidase).

18. Acetyl(Ac)-DEVD-AMC is a substrate for caspase-3 and caspase-7 and is mostcommonly used. Other available substrates are Ac-LEHD-AMC for caspase-5,Ac-YVAD-AMC for caspase-1 and -4, Ac-IETD-AMC for caspase-8 and -6, andAc-WEHD-AMC for caspase-1, -4, and -5 (17). The specificity of the substratesis not always restricted to the ones mentioned by the companies (51). The speci-ficity problem is due to the fact that depending on the concentration of the caspase,the substrate specificity is lost. High concentrations of caspase-3 will also cleaveAc-IETD-AMC, the caspase-8 and -6 substrate. Therefore it is appropriate to com-bine enzymatic measurements of caspase activities with Western blot analysis toidentify the actual presence and activation status of the caspase (Subheading 3.3.1.).

19. For analysis of caspase activity: Do not use pefablock protease inhibitors of othertotal protease inhibitor mixes as these interfere with caspase activity.

20. Caspases have a reactive cysteine in their catalytic site. To block enzymatic activ-ity during the lysis, 1 mM oxidized gluthathion is added, which blocks caspaseactivity and avoids further auto-activation or proteolytic caspase cascades duringlysis. When caspase activity is measured in CSF buffer, 10 mM dithiotreitol isadded to neutralize GSH and liberate the catalytic cysteine.

21. It is often advisable to add the inhibitors or modulators of TNF responsivity 2 hbefore the administration of the TNF serial dilution to allow the inhibitors ormodulators to penetrate the cell and to interfere with their intracellular moleculartargets. In the case of geldanamycin, an Hsp90 inhibitor, a 16–18 h pretreatmentis required to switch the cells from TNFR-1-induced necrosis to TNFR-1-inducedapoptosis (20).

22. CytoTox 96 Assay is based on a coupled enzymatic assay involving the conver-sion of a tetrazolium salt, 2-p-(iodophenyl)-3-(p-nitrophenyl)-5-phenyltetrazoliumchloride (INT), into a formazan product. The reaction is catalyzed by lactate dehy-drogenase released from cells and by diaphorase present in the assay substratemixture.

Lactate + NAD+ Pyruvate + NADHNADH + INT NAD+ + Formazan

192 Addendum II

23. A problem with the MTT assay, especially at early stages of apoptotic cell death,is that these cells often have intact complex-II-activity MTT. Therefore, themodulatory effect of some compounds can be missed; moreover, some com-pounds may influence complex II activity without affecting cell death per se.One should watch carefully for these types of false negative or false positiveeffects in the MTT assay.

References1. Schweichel, J. U. and Merker, H. J. (1973) The morphology of various types of

cell death in prenatal tissues. Teratology 7, 253–266.2. Clarke, P. G. (1990) Developmental cell death: morphological diversity and mul-

tiple mechanisms. Anat. Embryol. (Berl.) 181, 195–213.3. Clarke, P. G. (2002) Apoptosis: from morphological types of cell death to inter-

acting pathways. Trends Pharmacol. Sci. 23, 308–309; author reply 310.4. Kerr, J. F., Wyllie, A. H., and Currie, A. R. (1972) Apoptosis: a basic biological

phenomenon with wide-ranging implications in tissue kinetics. Br. J. Cancer 26,239–257.

5. Wyllie, A. H., Kerr, J. F., and Currie, A. R. (1980) Cell death: the significance ofapoptosis. Int. Rev. Cytol. 68, 251–306.

6. Denecker, G., Vercammen, D., Steemans, M., Vanden Berghe, T., Brouckaert,G., Van Loo, G., et al. (2001) Death receptor-induced apoptotic and necrotic celldeath: differential role of caspases and mitochondria. Cell Death Differ. 8, 829–840.

7. Fiers, W., Beyaert, R., Declercq, W., and Vandenabeele, P. (1999) More than oneway to die: apoptosis, necrosis and reactive oxygen damage. Oncogene 18, 7719–7730.

8. Vercammen, D., Beyaert, R., Denecker, G., Goossens, V., Van Loo, G., Declercq,W., et al. (1998) Inhibition of caspases increases the sensitivity of L929 cells tonecrosis mediated by tumor necrosis factor. J. Exp. Med. 187, 1477–1485.

9. Vercammen, D., Brouckaert, G., Denecker, G., Van de Craen, M., Declercq, W.,Fiers, W., et al. (1998) Dual signaling of the Fas receptor: initiation of bothapoptotic and necrotic cell death pathways. J. Exp. Med. 188, 919–930.

10. Beyaert, R., Van Loo, G., Heyninck, K., and Vandenabeele, P. (2002) Signalingto gene activation and cell death by tumor necrosis factor receptors and Fas. Int.Rev. Cytol. 214, 225–272.

11. Scaffidi, C., Fulda, S., Srinivasan, A., Friesen, C., Li, F., Tomaselli, K. J., et al.(1998) Two CD95 (APO-1/Fas) signaling pathways. EMBO J. 17, 1675–1687.

12. Li, P., Nijhawan, D., Budihardjo, I., Srinivasula, S. M., Ahmad, M., Alnemri, E.S., et al. (1997) Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 91, 479–489.

13. Kischkel, F. C., Hellbardt, S., Behrmann, I., Germer, M., Pawlita, M., Krammer,P. H., et al (1995) Cytotoxicity-dependent APO-1 (Fas/CD95)-associated proteinsform a death-inducing signaling complex (DISC) with the receptor. EMBO J. 14,5579–5588.


14. Luo, X., Budihardjo, I., Zou, H., Slaughter, C., and Wang, X. (1998) Bid, a Bcl2interacting protein, mediates cytochrome c release from mitochondria in responseto activation of cell surface death receptors. Cell 94, 481–490.

15. Van Loo, G., Demol, H., van Gurp, M., Hoorelbeke, B., Schotte, P., Beyaert, R.,et al. (2002). A matrix-assisted laser desorption ionization post-source decay(MALDI-PSD) analysis of proteins released from isolated liver mitochondriatreated with recombinant truncated Bid. Cell Death Differ. 9, 301–308.

16. Van Loo, G., Saelens, X., van Gurp, M., MacFarlane, M., Martin, S. J., andVandenabeele, P. (2002) The role of mitochondrial factors in apoptosis: a Russianroulette with more than one bullet. Cell Death Differ. 9, 1031–1042.

17. Lamkanfi, M., Declercq, W., Depuydt, B., Kalai, M., Saelens, X., andVandenabeele, P. The caspase family. In Caspases: Their Role in Cell Death andCell Survival (Los, M. and Walczak, H., eds.). Landes Bioscience, Kluwer Aca-demic, Georgetown, TX, 2003

18. Kawahara, A., Ohsawa, Y., Matsumura, H., Uchiyama, Y., and Nagata, S. (1998)Caspase-independent cell killing by Fas-associated protein with death domain. J.Cell Biol. 143, 1353–1360.

19. Holler, N., Zaru, R., Micheau, O., Thome, M., Attinger, A., Valitutti, S., et al.(2000) Fas triggers an alternative, caspase-8-independent cell death pathway usingthe kinase RIP as effector molecule. Nat. Immunol. 1, 489–495.

20. Vanden Berghe, T., Kalai, M., van Loo, G., Declercq, W., and Vandenabeele, P.(2003) disruption of HSP90 function reverts tumor necrosis factor-induced necro-sis to apoptosis. J. Biol. Chem. 278, 5622–5629.

21. Vercammen, D., Vandenabeele, P., Beyaert, R., Declercq, W., and Fiers, W.(1997) Tumour necrosis factor-induced necrosis versus anti-Fas-inducedapoptosis in L929 cells. Cytokine 9, 801–808.

22. Unwin, P. N. and Zampighi, G. (1980) Structure of the junction between commu-nicating cells. Nature 283, 545–549.

23. Unwin, P. N. and Ennis, P. D. (1984) Two configurations of a channel-formingmembrane protein. Nature 307, 609–613.

24. Cummings, M. C., Winterford, C. M., and Walker, N. I. (1997) Apoptosis. Am. J.Surg. Pathol. 21, 88–101.

25. Falcieri, E., Gobbi, P., Cataldi, A., Zamai, L., Faenza, I., and Vitale, M. (1994)Nuclear pores in the apoptotic cell. Histochem. J. 26, 754–763.

26. Kerr, J. F., Winterford, C. M., and Harmon, B. V. (1994) Apoptosis. Its signifi-cance in cancer and cancer therapy. Cancer 73, 2013–2026.

27. Kitanaka, C. and Kuchino, Y. (1999). Caspase-independent programmed celldeath with necrotic morphology. Cell Death Differ. 6, 508–515.

28. Earnshaw, W. C. (1995) Nuclear changes in apoptosis. Curr. Opin. Cell Biol. 7,337–343.

29. Nagata, S. (2002) Breakdown of chromosomal DNA. Cornea 21, S2–S6.30. Nagata, S. (2000) Apoptotic DNA fragmentation. Exp. Cell Res. 256, 12–18.31. Bevers, E. M., Comfurius, P., Dekkers, D. W., Harmsma, M., and Zwaal, R. F.

(1998) Regulatory mechanisms of transmembrane phospholipid distributions and

194 Addendum II

pathophysiological implications of transbilayer lipid scrambling. Lupus 7(Suppl2), S126–S131.

32. Martin, S. J., Reutelingsperger, C. P., McGahon, A. J., Rader, J. A., van Schie, R.C., LaFace, D. M., et al. (1995) Early redistribution of plasma membranephosphatidylserine is a general feature of apoptosis regardless of the initiatingstimulus: inhibition by overexpression of Bcl-2 and Abl. J. Exp. Med. 182, 1545–1556.

33. Denecker, G., Dooms, H., Van Loo, G., Vercammen, D., Grooten, J., Fiers, W., etal. (2000) Phosphatidyl serine exposure during apoptosis precedes release of cyto-chrome c and decrease in mitochondrial transmembrane potential. FEBS Lett. 465,47–52.

34. Foghsgaard, L., Wissing, D., Mauch, D., Lademann, U., Bastholm, L., Boes, M.,et al. (2001) Cathepsin B acts as a dominant execution protease in tumor cellapoptosis induced by tumor necrosis factor. J. Cell Biol. 153, 999–1010.

35. Brunk, U. T., Neuzil, J., and Eaton, J. W. (2001) Lysosomal involvement inapoptosis. Redox. Rep. 6, 91–97.

36. Neuzil, J., Weber, T., Schroder, A., Lu, M., Ostermann, G., Gellert, N., et al.(2001) Induction of cancer cell apoptosis by alpha-tocopheryl succinate: molecu-lar pathways and structural requirements. FASEB J. 15, 403–415.

37. Stoka, V., Turk, B., Schendel, S. L., Kim, T. H., Cirman, T., Snipas, S. J., et al.(2001) Lysosomal protease pathways to apoptosis. Cleavage of bid, not pro-caspases, is the most likely route. J. Biol. Chem. 276, 3149–3157.

38. Thornberry, N. A., Rano, T. A., Peterson, E. P., Rasper, D. M., Timkey, T., Garcia-Calvo, M., et al. (1997) A combinatorial approach defines specificities of mem-bers of the caspase family and granzyme B. Functional relationships establishedfor key mediators of apoptosis. J. Biol. Chem. 272, 17,907–17,911.

39. Spencer, M. J., Croall, D. E., and Tidball, J. G. (1995) Calpains are activated innecrotic fibers from mdx dystrophic mice. J. Biol. Chem. 270, 10,909–10,914.

40. Xu, K., Tavernarakis, N., and Driscoll, M. (2001) Necrotic cell death in C. elegansrequires the function of calreticulin and regulators of Ca(2+) release from theendoplasmic reticulum. Neuron 31, 957–971.

41. Lankiewicz, S., Marc Luetjens, C., Truc Bui, N., Krohn, A. J., Poppe, M., Cole,G. M., et al. (2000) Activation of calpain I converts excitotoxic neuron death intoa caspase-independent cell death. J. Biol. Chem. 275, 17,064–17,071.

42. Nakayama, M., Ishidoh, K., Kayagaki, N., Kojima, Y., Yamaguchi, N., Nakano,H., et al. (2002) Multiple pathways of TWEAK-induced cell death. J. Immunol.168, 734–743.

43. Gobeil, S., Boucher, C. C., Nadeau, D., and Poirier, G. G. (2001) Characterizationof the necrotic cleavage of poly(ADP-ribose) polymerase (PARP-1): implicationof lysosomal proteases. Cell Death Differ. 8, 588–594.

44. Rae, T. (1977) Tolerance of mouse macrophages in vitro to barium sulfate used inorthopedic bone cement. J. Biomed. Mater. Res. 11, 839–846.

45. Metivier, D., Dallaporta, B., Zamzami, N., Larochette, N., Susin, S. A., Marzo, I.,et al. (1998) Cytofluorometric detection of mitochondrial alterations in early


CD95/Fas/APO-1-triggered apoptosis of Jurkat T lymphoma cells. Comparisonof seven mitochondrion-specific fluorochromes. Immunol. Lett. 61, 157–163.

46. Kroemer, G., Dallaporta, B., and Resche-Rigon, M. (1998) The mitochondrialdeath/life regulator in apoptosis and necrosis. Annu. Rev. Physiol. 60, 619–642.

47. Lemasters, J. J., Nieminen, A. L., Qian, T., Trost, L. C., Elmore, S. P., Nishimura,Y., et al. (1998). The mitochondrial permeability transition in cell death: a com-mon mechanism in necrosis, apoptosis and autophagy. Biochim. Biophys. Acta1366, 177–196.

48. Goossens, V., Grooten, J., De Vos, K., and Fiers, W. (1995) Direct evidence fortumor necrosis factor-induced mitochondrial reactive oxygen intermediates andtheir involvement in cytotoxicity. Proc. Natl. Acad. Sci. USA 92, 8115–8119.

49. Sanchez, I., Mahlke, C., and Yuan, J. (2003). Pivotal role of oligomerization inexpanded polyglutamine neurodegenerative disorders. Nature 421, 373–379.

50. Lippens, S., Kockx, M., Knaapen, M., Mortier, L., Polakowska, R., Verheyen, A.,et al. (2000) Epidermal differentiation does not involve the pro-apoptotic execu-tioner caspases, but is associated with caspase-14 induction and processing. CellDeath Differ. 7, 1218–1224.

51. Talanian, R. V., Quinlan, C., Trautz, S., Hackett, M. C., Mankovich, J. A., Banach,D., et al. (1997) Substrate specificities of caspase family proteases. J. Biol. Chem.272, 9677–9682.

196 Addendum II

Addendum III:

Curriculum vitae

197

Curriculum Vitae

Personalia

Last name Krysko

First Name Dmitri

Date of birth 21 July 1975

Place of birth Saratov, USSR

Nationality Russian

Marital status Married to Olga Krysko

Address Henleykaai 48, 9000 Ghent, Belgium

Home Phone +32-(0)9-2256553

Mobile Phone +32-(0)479-527323

E-mail [email protected]; [email protected]

Languages English, German, learning Dutch and French

Current Status

Jan 2006 - Dec 2006 A scientific co-worker at the Molecular Signaling and Cell Death Unit, Department for Molecular Biomedical Research, VIB – Ghent University, Technologiepark 927, 9052 Ghent, Belgium.

2005- Editorial Board Member of APOPTOSIS: An International Journal on Programmed Cell Death

Education

1982-1992 Gymansium N 3, Saratov, Russian Federation.

1992-1998 Attended and graduated from the Saratov State Medical University, Russian Federation, with congratulations of the Jury.

1998-2000 Resident at the Department of General Surgery of Saratov State Medical University, Russian Federation.

2000-2001 Predoctoral training in Medical Sciences and Health Sciences, Faculty of Medicine and Health Sciences, Ghent University, Belgium.

2001-2003 A Ph.D. student at the Department of Anatomy, Embryology, Histology and Medical Physics, Faculty of Medicine and Health Sciences, Ghent University, Belgium.

2004-2005 A Ph.D. student at the Molecular Signaling and Cell Death Unit, Department for Molecular Biomedical Research, Flemish Interuniversity for Biotechnology – Ghent University, Belgium.

198 Addendum III

Travel Grants for Training Abroad

1. 01.07.1993-31.08.1993 – For training in medical care in the State Hospital of Ludwigshafen/Rhein, Germany.

2. 11.07.1994-25.08.1994 – For training in General Surgery in the State Hospital of Worms, Germany.

3. 01.07.1995-31.08.1995 – For training in General Surgery in the State Hospital of Worms, Germany.

4. 08.07.1996-16.08.1996 – For training in Orthopaedics in the Hospital of Heidelberg University, Germany.

5. 14.07.1997-14.08.1997 – For training in General Surgery in the Hospital of Heidelberg University, Germany.

6. 01.07.1998-31.08.1998 – For training in Orthopaedics in the Hospital of Heidelberg University, Germany.

7. 14.04.1997-30.05.1997 – For clinical service and scientific training on apoptosis research in the Department of Surgery of the University of North Carolina, Chapel Hill, USA.

8. 23.02.2000-31.05.2000 – For scientific training in the Department of Anatomy, Embryology, Histology and Medical Physics, Faculty of Medicine and Health Sciences, Ghent University, Belgium.

Travel Awards for Participation in Conferences

1. Award from Belgian Society for Cell and Developmental Biology to attend the 14th Euroconference on Apoptosis: “Death or Survival? Fate in Sardinia”, 29 September - 4 October, 2006, Sardinia, Italy.

2. Full scholarship from organizing committee (covering registration, accommodations and travel expenses) to attend the 13th Euroconference on Apoptosis: "Survival at the Danuber”, 1-4 October, 2005, Budapest, Hungary.

3. Aventis [i]lab award (from Pharma Deutschland GmbH) to attend the Gordon Research Conference on “Apoptotic Cell Recognition and Clearance”, 19-24 June, 2005, New London, USA.

4. Full scholarship from organizing committee (covering registration, accommodations and travel expenses) to attend the 12th Euroconference on Apoptosis: “Cell death next to the sea", 17-20 September 2004 Chania, Crete, Greece.

Scientific Awards

1. 1st place in the contest of scientific works at the 57th of Annual Meetings of young scientists in Saratov State Medical University, 1996, Saratov, Russian Federation.

2. 1st place in the contest of scientific works at the 59th of Annual Meetings of young scientists in Saratov State Medical University, 1998, Saratov, Russian Federation.

3. 1st place and scientific award of Regional Public Support Fund for talented students and youth after N.I. Vavilov for the best scientific report in "medicine" at the conference "Youth and Science at the beginning of the XXI century" in the Saratov State University. 1998, Saratov, Russian Federation.

CV 199

List of Publications

Publications in Journals (A1)

1. DV Krysko, K D’Herde and P Vandenabeele Clearance of apoptotic and necrotic cells and its immunological consequences. Review. Accepted for publication in Apoptosis.

2. DV Krysko, G Denecker, N Festjens, S Gabriels E Parthoens, K D’Herde and P Vandenabeele Macrophages use different internalization mechanisms to clear apoptotic and necrotic cells. Cell Death and Differentiation 2006; Apr 21; [Epub ahead of print].

3. DV Krysko, L Leybaert, P Vandenabeele and K D’Herde Gap junctions and the propagation of cell survival and cell death signals. Invited Review. Apoptosis 2005; 10(3): 459-69.

4. DV Krysko, S Mussche, L Leybaert and K D’Herde Gap junctional communication and connexin43 expression in relation to apoptotic cell death and survival of granulosa cells. The Journal of Histochemistry and Cytochemistry 2004; 52(9): 1199-207.

5. G Brouckaert, M Kalai, DV Krysko, X Saelens, D Vercammen, M Ndlovu, G Haegeman, K D’Herde and P Vandenabeele Phagocytosis of apoptotic cells by macrophages is more efficient than that of necrotic cells. Molecular Biology of the Cell 2004; 15, 1089–1100.

6. DV Krysko, G Brouckaert, M Kalai, P Vandenabeele and K D’Herde Mechanisms of internalization of apoptotic and necrotic L929 cells by a macrophage cell line studied by electron microscopy. Journal of Morphology 2003; 258: 336–345.

7. DV Krysko, F Roels, L Leybaert and K D’Herde Mitochondrial transmembrane potential changes support the concept of mitochondrial heterogeneity during apoptosis TheJournal of Histochemistry and Cytochemistry 2001; 49(10): 1277–1284.

Chapters in Books

1. T Vanden Berghe, G Denecker, G Brouckaert, DV Krysko, K D’Herde, and P Vandenabeele More Than One Way to Die Methods to Determine TNF-Induced Apoptosis and Necrosis. In Methods in Molecular Medicine, Vol. 98, 2004: Tumour Necrosis Factor. Edited by: A Corti and P Ghezzi, Humana Press Inc.

2. DV Krysko, P Vandenabeele, K D’Herde. Cell death at a glance. In New Cell Apoptosis Research. Edited by F Columbus, Nova Science Publishers, Inc. Accepted for publication in the preliminary form.

Abstracts Published in Journals

1. DV Krysko, F Roels and K D’Herde Persistence of actively respiring mitochondria until late in the apoptotic process in serum-free cultured granulosa explants: additional evidence. Conference "A new life of mitochondria" 2000, Leuven, Belgium. Archives of Biochemistry and Physiology, 110, B10, 2000.

2. DV Krysko, S Mussche, L Leybaert and K D’Herde Apoptosis of granulosa cell explants: role of gap junctions. FEBS Meeting “Signal Transduction: from Membrane to Gene Expression, from Structure to Disease,” July 3-8, 2003, Brussels, Belgium. EuropeanJournal of Biochemistry, 270, suppl. 1, July, PS 01-350, 2003.

200 Addendum III

Invited Lectures

1. Macrophages use different internalization mechanisms for the clearance of apoptotic and necrotic cells. Gordon Research Conference on “Apoptotic Cell Recognition and Clearance”, 19-24 June, 2005, New London, USA.

2. Mechanisms of recognition. 48 Symposium of the Society for Histochemistry on: “Histochemistry of cell damage and death”,

th

7-10 September, 2006, Stressa, Italy.

Oral Presentations Done Personally

1. DV Krysko, K D’Herde and P Vandenabeele Macrophages use different internalization mechanisms to clear apoptotic and necrotic cells. 27th June, 2006, Pasteur Institute, Brussels, Belgium.

2. DV Krysko, G Brouckaert, M Kalai, P Vandenabeele and K D’Herde Clearance mechanisms of apoptotic and necrotic L929 cells by a macrophage cell line studied by electron microscopy. BVM-SBM Annual Meeting, 23d May, 2003, Antwerp, Belgium.

3. DV Krysko, S Mussche, L Leybaert and K D’Herde Role of gap junctions in survival of granulosa cell explants. BVM-SBM Annual Meeting, 23d May, 2003, Antwerp, Belgium.

4. DV Krysko, F Roels, L Leybaert and K D'Herde Mitochondrial heterogeneity during the apoptotic process revealed by probing changes of mitochondrial transmembrane potential. Annual BSM Meeting at GlaxoSmithKline Biologicals in Rixensart, 1st of December 2001, Belgium.

Published in the Abstract Form and Presented as Posters Personally

1. DV Krysko and K D'Herde Mitochondrial transmembrane potential changes support the concept of mitochondrial heterogeneity during apoptosis. Published in abstract form at the Advanced FEBS Course "Mitochondria in Cell Life and Death", 2-7 September, Moscow State University, 2001, Moscow, Russia.

2. DV Krysko, F Roels, L Leybaert and K D'Herde Mitochondrial heterogeneity during the apoptotic process revealed by probing changes of mitochondrial transmembrane potential. Published in abstract form and presented as an oral communication at the Annual BSM Meeting at GlaxoSmithKline Biologicals in Rixensart, Belgium, 1st of December, 2001, Belgium.

3. DV Krysko, L Leybaert, K D’Herde Up-regulation of gap junctional intracellular communication during initiation of apoptosis in serum-free cultured granulosa explants. Annual Cell Biology and Histochemistry Meeting “Localizing Molecules in Cells”, 2002, London, United Kingdom.

4. DV Krysko, G Brouckaert, M Kalai, P Vandenabeele and K D’Herde Ultrastructural data on phagocytosis of apoptotic and necrotic L929 cells in an in vitro assay. 10th

Euroconference on Apoptosis: “Charming to Death,” October 11-13, 2002, Paris, France.

5. DV Krysko, S Mussche, L Leybaert and K D’Herde Levels of Connexin-43 expression and gap junctional intracellular communication: role in survival of granulosa cell explants. Wetenschapsdag, Universitair Ziekenhuis Gent, 17th January, 2002, Ghent, Belgium.

6. DV Krysko, G Brouckaert, M Kalai, P Vandenabeele and K D’Herde Mechanisms of internalization of apoptotic and necrotic L929 cells by a macrophage cell line studied by

CV 201

electron microscopy. Wetenschapsdag, Universitair Ziekenhuis Gent, 17th January, 2002, Ghent, Belgium.

7. DV Krysko, S Mussche, L Leybaert and K D’Herde Survival of granulosa cell explants: role of levels of connexin43 expression and gap junctional intracellular communication. European Conference on Apoptosis: “From signalling pathways to therapeutic tools,” January 29th to February 1st, 2003, Kirchenberg, Luxembourg.

8. DV Krysko, G Brouckaert, M Kalai, P Vandenabeele and K D’Herde Mechanisms of internalization of apoptotic and necrotic L929 cell line studied by electron microscopy. European Conference on Apoptosis: “From signalling pathways to therapeutic tools,” January 29th to February 1st, 2003, Kirchenberg, Luxembourg.

9. DV Krysko, G Brouckaert, M Kalai, P Vandenabeele and K D’Herde Different internalization mechanisms of apoptotic and necrotic L929 cells. Annual BSCB Meeting “Endocytosis and Signalling” 16-17 May, 2003, Brussels, Belgium.

10. DV Krysko, G Brouckaert, M Kalai, P Vandenabeele and K D’Herde Ultrastructural data on internalization mechanisms of apoptotic and necrotic L929 cells in an in vitro assay. Gordon Research Conference on “Clearance of Apoptotic Cells by Phagocytes: Mechanisms and Consequences,” August 3-8, 2003, New London, USA.

11. DV Krysko, S Mussche, L Leybaert and K D’Herde Function of gap junctions in apoptosis of granulosa cell explants. 11th Euroconference on Apoptosis: “Cell Death Under the Three Towers,” October 25-28, 2003, Ghent, Belgium.

12. DV Krysko, G Brouckaert, M Kalai, P Vandenabeele and K D’Herde Electron microscopical study of uptake mechanisms of apoptotic and necrotic L929 cells by a macrophage cell line. 11th Euroconference on Apoptosis: “Cell Death Under the Three Towers,” October 25-28, 2003, Ghent, Belgium.

13. DV Krysko, G Denecker, P Vandenabeele and K D’Herde Macrophages use different internalization mechanisms for the clearance of apoptotic and necrotic L929 cells. 12th

Euroconference on Apoptosis: "Cell death next to the sea," September 17 - 20, 2004, Chania, Crete, Greece.

14. DV Krysko, G Denecker, S Gabriels, P Vandenabeele and K D’Herde Macrophage use different internalization mechanisms for the clearance of apoptotic and necrotic cells. VIB Seminar, March 3-4, 2005, Blankenberge, Belgium.

15. DV Krysko, G Denecker, P Vandenabeele and K D’Herde Macrophages use different internalization mechanisms for the clearance of apoptotic and necrotic cells. Gordon Research Conference on “Apoptotic Cell Recognition and Clearance”, June 19-24, 2005, New London, USA.

16. DV Krysko, G Denecker, P Vandenabeele and K D’Herde Apoptotic and necrotic cells

internalised differently by a macrophage cell line without activation of NF- B transcription factor. 12th Euroconference on Apoptosis. “Survival at the Danuber,” October 1-4, 2005, Budapest, Hungary.

Other Abstract and Poster Presentations

1. K D'Herde, DV Krysko, F Roels, and L Leybaert Mitochondrial transmembrane potential changes support the concept of mitochondrial heterogeneity during apoptosis. XI-th International Congress of Histochemistry and Cytochemistry, 2000, 3-8 September, minisymposium M13, York, United Kingdom.

2. DV Krysko, F Roels, L Leybaert, and K D'Herde Dissipation of transmembrane potential is confined to a subpopulation of mitochondria during initiation of apoptosis. Published in

202 Addendum III

abstract from at the "Gordon Research Conference on Cell Death", July 2001, Oxford, United Kingdom.

3. DV Krysko, G Brouckaert, M Kalai, P Vandenabeele and K D’Herde Phagocytosis of apoptotic and necrotic L929 cells by macrophages cell lines studied by electron microscopy. The 4th International Cell Death Symposium on “The Mechanisms of Cell Death”, 2002, Noosaville, Australia.

4. G Brouckaert, M Kalai, DV Krysko, D Vercammen, X Saelens, K D’Herde and P Vandenabeele Differences and similarities in phagocytosis of apoptotic and necrotic cells by macrophages. 10th Euroconference on Apoptosis “Charming to Death,” October 11-13, 2002, Paris, France.

5. M Kalai, G Brouckaert, DV Krysko, X Saelens, D Vercammen, M Ndlovu, G Haegeman, K D’Herde, P Vandenabeele Phagocytosis of necrotic by macrophages is phosphatidylserine-dependent and does not induce inflammatory cytokine production. 11th Euroconference on Apoptosis: “Cell Death Under the Three Towers,” October 25-28, 2003, Ghent, Belgium.

Oral Presentation During Medical Studies

1. DV Krysko The mathematical model of interaction of electromagnetic waves of millimetre frequency with physical objects and her computer model. Abstract and oral presentationat the 58-th Annual Meeting of young scientists of Saratov State Medical University, 1997, Saratov, Russian Federation.

2. DV Krysko The role of apoptosis in pathology. Abstract and oral presentation at the 58-th Annual Meeting of young scientists of Saratov State Medical University, 1998, Saratov, Russia Federation.

3. DV Krysko Modification by Electromagnetic waves of the millimetre frequency the pharmacological conditions of medical drugs. Abstract and oral presentation at the 59-th Annual Meeting of young scientists of Saratov State Medical University, 1998, Saratov, Russia Federation.

4. DV Krysko Influence of millimetre waves on drugs and anaesthetics. Abstract and oralpresentation at the Conference: "Youth and Science at the beginning of the XXI century“. Saratov State University, 1998, Saratov, Russian Federation.

Abstracts During Medical Studies

1. DV Krysko A new method of registration of molecular condition of objects and it’s clinical application. Abstract and oral presentation at the 57-th Annual Meeting of young scientists of Saratov State Medical University, 1996, Saratov, Russian Federation.

2. V Petrosyan, Y Gulaev, V Krysko, DV Krysko, E Zhiteneva, V Yelkin, N Sinitcin Theory of resonant transmission. Wave propagation in a medium of coupled harmonic oscillators. International conference “The doors to future“. 1996, Lodz-Dobieszkow, Poland.

3. DV Krysko, G Brill, V Petrosyan, L Martynov, E Zhiteneva, N Sinitcin Mathematical modelling of interplay between millimetre waves and lipid membranes. Abstract on the international conference "Modelling of Ecosystems and optimization in the condition of technogenesis“. 1996, Soligorsk, Belarus.

4. V Petrosyan, Y Gulaev, E Zhiteneva, A Gulaev, L Martynov, DV Krysko, N Sinitcin Resonant transmission and state diagnostics of physical and biological objects under the

CV 203

action of millimetre waves. Abstract on the international conference: "Modelling of Ecosystems and optimization in the condition of technogenesis". 1996, Soligorsk, Belarus.

5. V Petrosyan, N Devyatkov, Y Gulaev, N Sinitcin, E Zhiteneva, V Krysko, DV Krysko, M Skobelev Effects of resonance interaction of millimetre waves with aqueous and biological objects. Abstract on the 11-th Annual international symposium: “Millimetre waves in biology and medicine“. 1997, Moscow, Russian Federation.

Supervision

2003-2004 – Supervision of master thesis entitled “Internalization mechanisms used by macrophages to engulf apoptotic and necrotic cells” of S. Gabriels at the Faculty of Biomedical Sciences, Ghent University, Belgium.

Date post:	17-May-2023
Category:	Documents
Upload:	independent
View:	0 times
Download:	0 times

Gap junctions and the propagation of cell survival and cell death signals

Documents