Understanding endothelial cell apoptosis: what can the transcriptome, glycome and proteome reveal

Phil Trans R Soc B (2007) 362 1469ndash1487

doi101098rstb20072129

Understanding endothelial cell apoptosiswhat can the transcriptome glycome

and proteome reveal

Published online 14 June 2007

Muna Affara1dagger Benjamin Dunmore2dagger Christopher Savoie3 Seiya Imoto4

Yoshinori Tamada25 Hiromitsu Araki3 D Stephen Charnock-Jones2

Satoru Miyano4 and Cristin Print6

Electron1098rst

One con

Authodagger These

1Department of Pathology Cambridge University Tennis Court Road Cambridge CB2 1QP UK2Department of Obstetrics and Gynaecology Cambridge University The Rosie Hospital

Cambridge CB2 2SW UK3GNI Ltd Kasumigaseki IHF Building 3-5-1 Kasumigaseki Chiyoda-ku 100-0013 Toyko Japan

4Human Genome Centre Institute of Medical Science University of Tokyo 4-6-1 ShirokanedaiMinato-ku Tokyo 108-8639 Japan

5Bioinformatics Centre Institute for Chemical Research Kyoto University GokashoUji Kyoto 611-0011 Japan

6Department of Molecular Medicine and Pathology University of Auckland 85 Park RoadPrivate Bag 92019 Auckland New Zealand

Endothelial cell (EC) apoptosis may play an important role in blood vessel development homeostasisand remodelling In support of this concept EC apoptosis has been detected within remodellingvessels in vivo and inactivation of EC apoptosis regulators has caused dramatic vascular phenotypesEC apoptosis has also been associated with cardiovascular pathologies Therefore understanding theregulation of EC apoptosis with the goal of intervening in this process has become a current researchfocus The protein-based signalling and cleavage cascades that regulate EC apoptosis are well knownHowever the possibility that programmed transcriptome and glycome changes contribute to ECapoptosis has only recently been explored Traditional bioinformatic techniques have allowedsimultaneous study of thousands of molecular signals during the process of EC apoptosis Howeverto progress further we now need to understand the complex cause and effect relationships amongthese signals In this article we will first review current knowledge about the function and regulationof EC apoptosis including the roles of the proteome transcriptome and glycome Then we assess thepotential for further bioinformatic analysis to advance our understanding of EC apoptosis includingthe limitations of current technologies and the potential of emerging technologies such as generegulatory networks

Keywords endothelial cell apoptosis transcriptome gene network

1 INTRODUCTIONApoptosis is the programmed suicide of a cell (Wyllie amp

Arends 2003) It serves to remove supernumerary cells

during development and tissue remodelling (Meier et al2000) maintain homeostasis in adult tissues remove

self-reactive or non-reactive immune cells (Adams et al

1999) and delete damaged infected or transformed cells

It is an energy-requiring process throughout which

membrane integrity is maintained to avoid the initiation

of an inflammatory response This contrasts with

necrosis in which cells die relatively passively following

overwhelming damage leading to loss of membrane

integrity and to inflammation (Kerr et al 1972)

ic supplementary material is available at httpdxdoiorg10b20072129 or via httpwwwjournalsroyalsocacuk

tribution of 21 to a Theme Issue lsquoBioengineering the heartrsquo

r for correspondence (cprintaucklandacnz)authors contributed equally to this article

1469

Like other organs blood vessels and the heartappear to employ apoptosis for remodelling duringdevelopment and to maintain homeostasis duringadulthood Within blood vessels vascular endothelialcells (ECs) play an especially important role inregulating overall vessel structure Most ECs in adultblood vessels are relatively quiescent and resistant toapoptosis However they are thought to retain thelatent capacity for proliferation and apoptosis tomediate angiogenesis and regression respectively Inthis review we will discuss what is known about thefunction of EC apoptosis in health and cardiovasculardisease We will describe how this process appears to becontrolled at the level of proteins (the proteome)Then we will discuss the role that regulated RNAtranscript abundance profiles (the transcriptome) andcell surface polysaccharide profiles (the glycome) mayplay in the EC apoptotic programme To progress ourknowledge of the regulation of EC apoptosis furtherwe now need to understand the cause and effect

This journal is q 2007 The Royal Society

0 min 3 6 9 12 15

18 21 24 27 30 33

Figure 1 HUVEC were cultured to 70 confluence in their optimal medium washed and their media replaced with a low serum(2 charcoal-stripped FCS) medium Their morphology was followed over time in the presence of Hoechst 33342 and theirdeath imaged by time-lapse epifluorescence microscopy The following series of events was observed (i) loss of contact withmatrix and adjacent cells (ii) retraction of cytoplasm (iii) membrane blebbing and (iv) nuclear condensation (seen as bluecoloration representing the density of Hoechst fluorescence surpassing a pre-set threshold for display) Times are in minutes(adapted from Johnson et al 2004)

1470 M Affara et al Understanding endothelial cell apoptosis

relationships between the molecular signals thatcontrol this process In sect10 of this review we willassess whether a deeper analysis of the EC transcrip-tome using emerging technologies such as generegulatory network analysis may help us understandthe complex signals that control EC fate Ourconclusion is that these emerging technologies havegreat potential for EC research especially when theirresults are integrated with those of cell biology studiesproteomics and in silico databases However it is criticalthat the cause and effect relationships predicted usingtechnologies such as gene regulatory networks arethoroughly validated by laboratory experiments

2 THE FUNCTION OF EC APOPTOSIS IN NORMALPHYSIOLOGY(a) EC apoptosis may contribute to blood vessel

remodelling

Several lines of evidence support the view that ECapoptosis plays an important role in vascular biologyand pathology Firstly apoptosis can be detectedduring developmental vessel regression in the aorticarches (Fisher et al 2000) the abdominal aorta (Choet al 1995) and the ductus arteriosus (Kim et al 2000)EC apoptosis may play an especially important role inthe eye in development of the tunica vasculosa lentis ofthe ocular lens (Mitchell et al 1998) and the pupillarymembrane of the anterior chamber of the eye (Langet al 1994) Macrophages may promote this ECapoptosis (Lang et al 1994) possibly through Wntsignalling (Lobov et al 2005)

Apoptosis in these regressing vessels is thought topromote their remodelling to meet the changing require-ments of the tissues they supply (Carmeliet 2000) Vesselregression may involve a classical series of events Forexample in one scenario following the inhibition ofVEGF-A signalling blood flow ceases followed by ECapoptosis and loss of pericytes leaving empty basementmembrane sheaves that can act as scaffolds for futurecapillary re-growth (Baffert et al 2006)

EC apoptosis appears to be less frequent in adultanimals than during development For example ECapoptosis appears to be much less common in adultrat kidney than in the developing rat kidney (Fierlbeck

Phil Trans R Soc B (2007)

et al 2003) Nevertheless adult ECs retain alatent capacity for apoptosis and EC apoptosis isdetected within specific adult tissues where substantialvessel remodelling occurs such as healing wounds(Desmouliere et al 1995 Greenhalgh 1998) regressingalveoli in the mammary gland after weaning (Matsumotoet al 1992 Djonov et al 2001) the developing placenta(Tertemiz et al 2005) and during the cyclical remodellingof female reproductive tissues (Modlich et al 1996Niswender et al 2000 Dickson et al 2001 Gaytan et al2002) Further support for a role of EC apoptosis invessel remodelling comes from in vitro studies Forexample incubation with pro-survival growth factors(Satake et al 1998) or retroviral-mediated anti-apoptoticBcl-2 expression (Pollman et al 1999) reduces theincidence of EC apoptosis and inhibits the regression ofvessel-like structures

It is important to note that two studies have beenunable to detect significant EC apoptosis withinregressing tissues where blood vessel remodellingwould be expected to occur (Hughes amp Chang-Ling2000 Arfuso amp Meyer 2003) In addition apoptosiswas not observed in one in vitro study where vesselswere disrupted by angiogenic inhibitors (Friis et al2006) However the majority of studies suggest thatEC apoptosis is frequently seen in remodelling vesselsimplying that EC apoptosis plays a role in theremodelling process

(b) EC apoptosis may contribute to blood vessel

morphogenesis and lumen formation

In addition to participating in vessel regression ECapoptosis may also contribute to angiogenesis andlumen formation Several EC types spontaneouslyform vessel-like structures in vitro (vascular morpho-genesis) when cultured on an extracellular matrix orwith matrix-producing cells Incubation of EC withanti-apoptotic caspase inhibitors or overexpression ofthe anti-apoptotic protein Bcl-2 appears to inhibitvascular morphogenesis in vitro (Segura et al 2002 andM Harris amp C Print 2003 unpublished results) Itinitially seems counterintuitive that EC apoptosis maycontribute to vascular morphogenesis However it ispossible that EC apoptosis contributes to this process

the apoptotic programme is initiated

apoptosis

proteincleavage uarrdarr

smallmolecules uarrdarr

proteintranslation uarrdarr

signallingcascades

matrix CHO uarrdarrRNA

transcription

RNAdegradation

proteinlocalization + interaction

uarrdarr

Figure 2 A schematic of the intracellular participants in the apoptotic programme RNA transcriptome-associated events areshown on the right and other events are shown on the left CHO denotes carbohydrates

Understanding endothelial cell apoptosis M Affara et al 1471

by deleting wrongly placed EC as nascent vesselsenlarge Furthermore the negatively charged mem-brane surface of apoptotic EC may promote angiogenicsprouting of adjacent vessels by causing localizedplasma membrane hyperpolarization (Weihua et al2005) Angiogenesis may also be promoted through theengulfment of apoptotic EC debris by viable neigh-bouring EC which may lead to the release of pro-angiogenic factors including VEGF-A (Golpon et al2004) However since in vitro angiogenesis modelsonly approximately recapitulate in vivo angiogenesisthese in vitro experiments should be interpreted withcaution until they are confirmed by in vivo studies

As well as supporting vessel regression and possiblypromoting angiogenesis EC apoptosis may in additionplay a role in lumen formation In support of thishypothesis ECs undergoing apoptosis are seen in thecentre of developing vascular structures (Duval et al2003) and newly formed lumen appear to containpost-apoptotic endothelial fragments (Bishop et al1999) Blocking apoptosis in the capillaries of thedeveloping rat kidney by inhibiting the pro-apoptoticfactor TGF-b1 appears to retard lumen formation(Fierlbeck et al 2003) In addition apoptotic EC canbe detected during lumen formation in placentalvessels in vivo (Tertemiz et al 2005)

(c) Mice in which regulators of EC apoptosis are

disrupted have vascular phenotypes

Additional evidence that EC apoptosis plays a role invascular biology is provided by the inactivation of genesencoding regulators of EC apoptosis in mice Forexample the growth factors vascular endothelialgrowth factor-A (VEGF-A) and angiopoietin-1(Ang-1) are important inhibitors of EC apoptosisInactivation of Ang-1 (Suri et al 1996) causes severevascular abnormalities early in development andinactivation of VEGF-A (using either a genetic-basedor protein inhibitor-based method) causes an increasein EC apoptosis and embryonic mortality (Gerber et al1999) An inherited mutation in the Tie2 receptor


causes vascular malformations by inhibiting ECapoptosis through constitutive activation of p52 ShcA(Morris et al 2006) Vascular endothelial (VE)cadherin is an ECndashEC adhesion molecule thattransduces anti-apoptoic signals into EC in conjunc-tion with VEGF-A Inactivation of VE-cadherin(Carmeliet et al 1999) causes embryonic lethality dueto failed vascular development Disruption of gapjunction formation by inactivation of connexin 43alters coronary vessel development in association withaltered apoptosis (Walker et al 2005) In mice in whichthe anti-apoptotic gene Bcl-2 has been inactivated inthe germ line retinal vascular development is retarded(Wang et al 2005) Although these phenotypes implyan important role for EC apoptosis since these signalsalso regulate EC proliferation it is difficult to assess theproportion of each mutant phenotype that resultsspecifically from the loss of EC apoptosis controlTherefore more specific studies to assess the role of ECapoptosis are underway in our laboratory and others Inthese ongoing studies EC apoptosis is being inhibiteddirectly by the EC-specific expression of anti-apoptotictransgenes We hope that the results of these studies willreveal the definitive functions of EC apoptosis in vivo

3 EC APOPTOSIS IN CARDIOVASCULARDISEASEAltered EC apoptosis is associated with severalcardiovascular pathologies in particular with athero-sclerosis and thrombosis EC apoptosis can be detectedin atherosclerotic plaques (Mehta et al 2002 Norataet al 2002) and may provide an important step in thetransition from a stable atherosclerotic plaque to aplaque with overlying thrombosis The tissue environ-ment of an atherosclerotic plaque may promote ECapoptosis For example oxidized LDL (ox-LDL)which is present in most atherosclerotic plaquesappears to downregulate EC expression of the anti-apoptotic protein cFLIP and subsequently activateFas-mediated EC death in vitro (Sata amp Walsh 1998)Ox-LDL also appears to activate the release of

macrophage chemokinesand phagocytosis

survival factor receptors

survivalgrowth factors

reduced of HLGAG sulphation

ECM components

ECMS

S

SSS

endothelial cell

CDC2 (darr24times)CDC6 (darr36times)kinesin-like spindle protein (darr35times)cyclin A2 (darr29times)cyclin B1 (darr19times)cyclin H (darr24times)cyclin E2 (darr34times)PCNA (darr34times)CDC45 (darr35times)MAD2-L1 (darr62)

flow-induced EGPR (darr49times)GP130 (darr58times)IL1 receptor-L1 (darr66times)Flt-1 (darr42times)

VEGF-C (darr45times)IL-8 (darr27times)

Gal-6-sulphotransferase (darr17times)heparan sulphate-sulphotransferase (darr17times)iduronate-2-sulphatase (darr24times)serglycin (darr31times)

collagen α1 type I (darr39times)collagen α1 type VII(darr43times)stromelysin-2 (darr91times)

ECM receptorsintegrinndashα2 (darr41times)Nr-CAM (darr53)integrinndashα3 (uarr29times)

caspase inhibitor MIHB (darr23times)IAP-recruiting protein TRAF-2 (darr23times)caspase 10(uarr18times)LARD (uarr20times)TRADD (uarr20times)TRAIL (uarr28times)TWEAK (uarr27times)decay acceleration factor (darr29times)

STAT2 (darr36times)ICAP-1a (darr33times)RAP2 (darr88times)MAPKP4 (uarr89times)SAK kinase (darr27times)NF κB p65 (uarr15times)I-κBα (uarr28times)I-κBβ (uarr28times)NF κB p100 (uarr48times)I-Rel (uarr35times)transcripts linked withG-protein signallingcaveolin 3(darr76times)lamin B1 (darr28times)connexin 43 (darr28times)chaperonin 10 (darr26times)Gu protein (darr66times)DEK (darr29times)

cell cycle and mitosisapoptotis regulators

signalling componentsand miscellaneous

MCP-1 (uarr48times)clusterin (uarr42times)

Figure 3 Summary of the regulated transcript abundance and carbohydrate sulphation changes that may play a role in preparingEC for apoptosis (adapted from Johnson et al 2004) Abbreviations ECM extracellular matrix HLGAG heparin-likeglycosaminoglycan


apoptosis-inducing factor (AIF) from coronary artery

EC mitochondria Suppression of AIF expression using

antisense oligonucleotides suppresses ox-LDL-

induced EC death (Zhang et al 2004) In contrast

high-density lipoprotein (HDL) cholesterol which is

anti-atherogenic appears to inhibit EC apoptosis in

part by promoting the activity of nitric oxide (NO)

synthase (Mineo et al 2006) Oxidative stress has also

been implicated in activating the apoptosis of EC in

atherosclerosis (Valgimigli et al 2003) The apoptosis

of EC overlying atherosclerotic plaques may cause

exposure of the underlying pro-thrombotic plaque

tissue and subsequent thrombosis For example

induction of EC apoptosis in vivo by the pro-apoptotic

drug staurosporine promotes both EC apoptosis and

local thrombosis This chain of events can be broken by

infusion of anti-apoptotic caspase inhibitors (Durand

et al 2004)

In addition apoptotic EC are thought to become

pro-coagulant by a process that does not require the

exposure of underlying plaque tissues (Bombeli et al1997) The coagulation cascade initiator tissue factor

(TF thromboplastin) may provide an especially

important link between EC apoptosis and thrombosis

TF activity is increased by phosphatidylserine (PS)

exposure on the surface of EC during apoptosis

(Tedgui amp Mallat 2003) Women with systemic lupus

erythematosus (SLE) are at increased risk of athero-

sclerosis and superimposed thrombosis This may be


due in part to an increased incidence of EC apoptosisand increased TF activity since elevated numbers ofPS-expressing post-apoptotic EC are detected in theblood of SLE patients

In atherosclerotic plaques vascular smooth musclecells (VSMC) become activated migrate into thetunica intima survive and proliferate This prolifer-ation may be promoted in part by factors derived fromapoptotic EC overlying the atherosclerotic plaqueRaymond et al (2004) suggest that proteolytic activityinitiated by caspases from apoptotic EC causesdegradation of proteoglycans which release unidenti-fied factors that activate VSMC ERK phosphorylationand Bcl-xL expression Other cardiovascular path-ologies with which EC apoptosis has been associatedinclude vasculitis (Woywodt et al 2003) and transplantrejection (Krupnick et al 2002 Hall amp Jevnikar 2003)EC apoptosis is also associated with inflammatorydisorders (Winn amp Harlan 2005) and non-cardiovas-cular pathologies such as neoplasia reviewed byCarmeliet amp Jain (2000) which are beyond the scopeof this review

4 THE CONTROL OF EC APOPTOSIS BYPROTEIN-BASED SIGNALLING ANDCLEAVAGE CASCADESWhether an EC lives or whether it dies by apoptosis isdetermined by the interplay of numerous signals It isthe integration of all these signals rather than any

(a)

(b)

(c)

30

20

10

10 15min

10 15min

healthcells

(mole )

48 serumdeprivation(mole )

12

11

1

1

2

3

5

532

di-Smono-S

tri-S

8

7

76

10

saccharide buildingblocks

1 (DU2SHNS6S) 458

102

101

0

148

25

75

92

173

284

780

0

365

00

101

0

4 (DU2SHNAc6S)

7 (DUHNAc6S)

8 (DUHNAc6SGHNS3S6S)

6 (DU2SHNAc)

2 (DU2SHNS)3 (DUHNS6S)

5 (DUHNS)

mA

Um

AU

Figure 4 Compositional analysis of heparin sulphatesisolated from control (grown in optimal media) and 48 hSFD HUVEC cultures (a) CE electropherogram depictingthe profile of cell surface HSGAGs from healthy ECs and (b)from SFD cells There is a loss of the highly sulphateddisaccharides (peak 1) and an increase in mono-sulphateddisaccharides (peaks 5 and 7) (c) The relative contribution ofeach HSGAG to the total pool investigated is shown for fourseparate experiments each using a different set of threepooled HUVEC isolates (adapted from Johnson et al 2004)


individual signal that ultimately determines an ECrsquos

fate The signals that regulate EC apoptosis are closely

integrated with the signals that regulate other activities

such as cell division stress responses activation and

migration (Blagosklonny 2003) The most widely

studied mechanisms that regulate EC apoptosis arein the realm of proteins (the proteome) These

include regulated protein translation phosphorylation

re-localization cleavage and oligomerization We have

reviewed these mechanisms in detail previously (Duval

et al 2003) and they can be broadly divided into thosethat regulate (i) the mitochondrial and (ii) the death


receptor apoptosis pathways When triggered thesetwo pathways converge to activate the lsquoexecutionerrsquocaspases 3 6 and 7 and nuclease enzymes that alongwith other enzymes such as cathepsins (Madge et al2003) and calpains (Porn-Ares et al 2003) cleavestructural proteins metabolic enzymes and nucleicacids This results in a series of biochemical andmorphological changes including (in approximateorder of appearance) loss of cellndashcell and cellndashmatrixcontact cytoplasmic retraction re-localization ofcytochrome c from mitochondria to cytoplasm acti-vation of caspases phosphatidyl serine exposure on theplasma membrane membrane blebbing and chromatincondensation (Granville et al 1999) This process canbe seen in the time-lapse sequence in figure 1 (Johnsonet al 2004) The end result is post-apoptotic ECcorpses that retain their membrane integrity and areeither dispersed in the bloodstream or occasionallyphagocytosed by neighbouring cells Interestinglycomponents of EC corpses that are dispersed in thebloodstream may be taken up by endothelial progeni-tors and promote their proliferation and differentiation(Hristov et al 2004)

The mitochondrial apoptosis pathway is initiated byintracellular damage or by the loss of extracellularsurvival signals for example by growth factor depri-vation when vessels become occluded during theprocess of regression (Lang amp Bishop 1993 Meesonet al 1999) This activates caspase 9 and releases pro-apoptotic molecules such as AIF from the mito-chondria thus initiating the apoptotic programmedescribed above

A plethora of extracellular survival signals normallysuppress the mitochondrial apoptosis pathway Theseinclude integrin-mediated cell adhesion to basementmembrane (Meredith et al 1993 Re et al 1994 Frischet al 1996) adhesion to adjacent cells mediated byVE-cadherin (Carmeliet et al 1999) and platelet ECadhesion molecule-1 (PECAM-1 Gao et al 2003Limaye et al 2005) haemodynamic shear forces fromblood flow (Meeson et al 1996) vascular EC-specificphosphotyrosine phosphatase (VE-PTP Baumer et al2006) VEGF-A (Lobov et al 2002) interleukin (IL)-8(Li et al 2003) basic fibroblast growth factor (FGFFuks et al 1994) and survival signals derived frompericytes or VSMC (Benjamin et al 1998) Theseextracellular survival signals are transduced into thecytoplasm where they activate intracellular signallingintermediates including pro- and anti-apoptoticmembers of the Bcl-2 family Overexpression of anti-apoptotic Bcl-2 or Bcl-xL proteins inhibits ECapoptosis (Schechner et al 2000 Zheng et al 2000)while antisense oligonucleotide-mediated inhibition ofthe pro-survival Bcl-xL protein induces EC apoptosisin vitro (Ackermann et al 1999) The radio-protectivedrug amifostine may act in part by elevating theexpression of anti-apoptotic members of the Bcl-2family (Khodarev et al 2004) Interestingly immedi-ately after birth the ratio of pro-apoptotic Bax to pro-survival Bcl-2 is elevated in regressing vessels (such asthe ductus arteriosus) relative to non-regressing vessels(Kim et al 2000) A central hub in the inhibition of ECapoptosis is the kinase AKT For example inactivationof AKT in mice markedly alters blood vessel


development possibly by reducing the expression ofthrombospondin (TSP)-1 and -2 (Chen et al 2005)

The death receptor apoptosis pathway is activated whentumour necrosis factor (TNF)-family proteins (Slowiket al 1997) such as TNFa Fas ligand and vascularendothelial growth inhibitor (VEGI Yu et al 2001)bind to TNF-family lsquodeath receptorsrsquo on the ECsurface Death receptors activate cytoplasmic adapterproteins which in turn activate caspase 8 and the p38mitogen-activated protein kinase (MAPK) pathway(Cardier amp Erickson-Miller 2002 Pru et al 2003)Although ECs express death receptors on their surface(Richardson et al 1994 Sata et al 2002) TNFa andFas ligand only activate apoptosis in a subset of EC(Biancone et al 1997 Sata et al 2002 Filippatos et al2004) The apparent resistance of many ECs todeath receptor-induced apoptosis may be in part dueto their expression of the caspase 8-antagonist cFLIP(Aoudjit amp Vuori 2001 Bannerman et al 2001) It mayalso be due to a second action of the TNF-receptorfamilymdashactivation of the transcription factor NFkB(Yasumoto et al 1992 Rath amp Aggarwal 1999) whichupregulates the expression of RNAs encoding apopto-sis inhibitors (Bach et al 1997) including A1 (Duriezet al 2000) A20 and xIAP (Stehlik et al 1998) Thereare additional lsquodeath ligandsrsquo for ECs that are notmembers of the TNF superfamily For example adultcapillary and venous EC but not arterial EC aresensitive to bone morphogenetic protein 4 (BMP4)-induced apoptosis The relatively higher resistanceof arterial EC to apoptosis may be due in part totheir expression of inhibitory (I) Smads After theknockdown of I-Smad expression using short interfer-ing RNA the resistant arterial EC become sensitiveto BMP4-induced death In contrast ectopicexpression of I-Smads in BMP4-sensitive cellssuppressed BMP4-induced death Furthermore intra-venous administration of BMP4 into mice causedhaemorrhage of capillary EC in brain and lung(Kiyono amp Shibuya 2006)

5 COULD CHANGES TO THE TRANSCRIPTOMEAND GLYCOME POTENTIATE EC APOPTOSISIn addition to the extensively studied protein-basedevents described above EC apoptosis may also beregulated by altered RNA transcript abundance profiles(an altered transcriptome) altered cell surface poly-saccharide profiles (an altered glycome) or by alteredconcentrations of intracellular small molecules Thepotential interplay between the intracellular events thatform the apoptotic programme is shown schematicallyin figure 2

To map the transcriptome and glycome changes thataccompany EC apoptosis our laboratory recentlymodelled the apoptosis of EC that may occur duringvessel regression by culturing primary human umbilicalvein EC (HUVEC) in conditions of partial survivalfactor deprivation (SFD) Time-lapse microscopy ofthe last events in this process is shown in figure 1Within 1 h of the onset of SFD HUVEC showed signsof stress After 28 h progression through the cell cyclehad ceased and only approximately 60 of original cellnumbers remained At this time approximately 85ndash


90 of the cultures consisted of stressed cells thatwithout intervention were destined to die over the next3 days and approximately 10ndash15 of the culturesconsisted of cells actively undergoing apoptosis Sincealmost all post-apoptotic cells appeared to detach andfloat into the medium within 3 h of death theremaining adherent cultures which were subsequentlyanalysed contained very few post-apoptotic corpsesTo determine whether SFD-induced EC apoptosis wasassociated with transcriptome changes we usedAffymetrix human U95A gene arrays to compare anabundance of 12 600 transcripts in HUVEC culturedin their optimal medium with the correspondingabundance in HUVEC after 28 and 48 h of SFDEach experiment was replicated five times each timeusing primary HUVEC derived from a differentindividual We found that most of the 12 600transcripts were not significantly affected by SFDHowever Bayesian Studentrsquos t-tests identified 171transcripts that were consistently upregulated and 495downregulated in all five experiments (p001) Inde-pendent component analysis (ICA a machine learningmethod for the analysis of noisy data that makes very fewstatistical assumptions) (Saidi et al 2004) confirmed theselection of regulated transcripts made by the BayesianStudentrsquos t-tests (Johnson et al 2004)

Interestingly very few of the apoptosis-associatedtranscriptome changes indicated a protective stressresponse Instead most changes appeared to be eitherdirectly pro-apoptotic or to indirectly prime cells forfuture apoptosis They are summarized in figure 3some are of special interest For example transcriptsencoding the death receptor LARD (DR3) and thedeath receptor ligands TRAIL and TWEAK wereupregulated RNAs encoding survival signals and theirreceptors such as VEGF-C IL-8 flow-inducedG-protein-coupled receptor GP130 IL1 receptorcomponent-L1 and Flt-1 were downregulated Tran-scripts encoding intracellular inhibitors of apoptosiswere also downregulated including cIAP1 TRAF-2STAT2 and the integrin-associated kinase ICAP-1aBax a pro-apoptotic member of the Bcl-2 family wasupregulated The final stage of the apoptotic pro-gramme is engulfment of apoptotic corpses byphagocytes Transcript abundance changes may assistthis process since RNAs encoding the chemokinemonocyte chemoattractant protein-1 (MCP-1) andclusterin (apolipoprotein J thought to promote theuptake of apoptotic bodies by non-professional phago-cytes) were also upregulated

We also observed changes in the abundance of RNAsencoding enzymes that determine the structure ofheparin-like cell surface glycosaminoglycans(HSGAG) Highly sulphated HSGAG are required topresent extracellular matrix-binding survival factorssuch as FGF and VEGF-A to their receptors on the ECsurface (Park et al 1993 Venkataraman et al 1999)After 28 h apoptosis the RNA encoding N-deacetylaseN-sulphotransferase (which promotes HSGAG sulpha-tion) was downregulated the RNA encoding iduronate-2-sulphatase (which removes sulphate groups fromHSGAG) was upregulated and serglycin (a proteoglycancore protein secreted by EC) was downregulated In linewith these transcript abundance changes when we


analysed HSGAG profiles we found that during ECapoptosis there was a relative loss of tri-sulphateddisaccharides and a relative increase in mono-sulphateddisaccharides (figure 4) This apoptosis-associatedreduction in HSGAG sulphation may further reducethe ability of stressed or apoptotic EC to respond tosurvival factors such as VEGF-A In agreement with thishypothesis we found that the ability of EC to mount aresponse to VEGF-A (as measured by the degree ofVEGF-induced transcriptome change) was dramati-cally reduced following SFD (Johnson et al 2004)

Based on these results we suggest that the cellbiology of stressdeprivation-induced EC apoptosis isunderpinned by synergy between previously describedprotein-based changes and the changes to the tran-scriptome and glycome identified in this study Thisprovides a new paradigm for the regulation of ECapoptosis and the possibility that a small number of theapoptosis-associated transcriptome and glycomechanges identified by this study may provide futuredrug targets for the regulation of EC apoptosis

6 WHAT IS THE TIME COURSE OF TRANSCRIP-TOME CHANGE DURING EC APOPTOSISTo understand the kinetics of transcriptome changeduring EC apoptosis we have performed a time courseexperiment HUVEC (a pooled culture from 10 donors)were exposed to SFD conditions identical to those usedin the study described aboveRNAwas prepared from thecultures before the induction of apoptosis (time 0) andafter 05 15 3 6 9 12 and 24 h hybridized toCodeLink Human Uniset 20K gene arrays Thisexperiment was repeated three times independentlyThe regulation of transcripts during EC apoptosis canbe visualized in the scatterplots shown in figure 5andashd Wethen selected a subset of 276 transcripts that wereconsistently regulated over the time course of ECapoptosis for further analysis To do this we excludedall transcripts from our analysis that werenot regulated byZS15s at 3 or more adjacent time points in all threeexperimental replicates (see Appendix for methods)These data are available in file 1 in the electronicsupplementary material From these 276 transcriptsk-means clustering was used to select eight groups oftranscripts (in addition to an unclassified group) withsimilar time-course profiles (figure 6) From theseprofiles and from the scatterplots shown in figure 5andashdit can be seen that a number of transcripts start to beregulated form 15 h and that in general the rate ofchange appears to low after 12 h When we plotted theindividual transcripts identified in the previous study wenoted that in general the transcripts encoding growthfactors (such as Ang-2 and IL-8) were among the earliesttranscripts to be regulated while transcripts associatedwith the cell cycle (such as cyclins A2 H and E CDC6CDC28 and kinesin-like spindle protein) were regulatedlater (table 1) The regulation of many of these transcriptsfollowing 28 h SFD has been validated by our previousAffymetrix study These data suggest that somefunctional classes follow common patterns of regulationduring the SFD-induced apoptosis of EC cultures Moresophisticated analysis (see below) suggests commonupstream regulators of these transcripts


7 PROGRAMMED CHANGES TO THE TRAN-SCRIPTOME MAY POTENTIATE APOPTOSIS INOTHER CELL TYPESContribution of the transcriptome and glycome to the

apoptotic programme is not unique to EC It has beenknown for some time that regulation of RNA transcript

abundance may turn the apoptotic programme on oroff during development For example ecdysone-

dependent expression of RNAs encoding the caspasesDRONC (Dorstyn et al 1999) and DRICE (Kilpatrick

et al 2005) appears to prime Drosophila tissues for

histolysis at specific developmental stages Interest-ingly the strategy of regulated transcript abundance

activating the apoptotic programme during develop-ment may also be employed in plants In the tissues of

barley undergoing developmental cell death RNAs

encoding plant homologues of the endonucleases that acleave DNA during mammalian apoptosis are upregu-

lated (Zaina et al 2003)Further support for the hypothesis that transcrip-

tome change may play a role in apoptotic programmescomes from experiments in which apoptosis of some

cell types can be blocked by inhibiting RNA synthesis

(Galli et al 1995 Schulz et al 1996) In additionseveral studies have identified programmed changes in

RNA transcript abundance as cultured cells undergoapoptosis For example infection with group A

Streptococcus pyogenes induces apoptosis in some

human epithelial cells Microarray analysis suggeststhat this apoptosis is associated with altered gene

expression including induction of RNA transcriptsencoding caspases and pro-apoptotic members of the

Bcl-2 family (Nakagawa et al 2004) Ionizing radi-ation-induced apoptosis of lymphoblastoid cells is

accompanied by the induction of RNAs encoding

pro-apoptotic proteins such as Fas and Bak (Akermanet al 2005) 5-Fluorouracil-induced apoptosis of colon

carcinoma cells is accompanied by a programme ofaltered gene expression including the induction of

RNAs encoding p53 Fas and TNF receptor-2 (Zhang

et al 2003) Apoptosis of human lung cancer cells isassociated with the upregulation of a number of genes

including pro-apoptotic members of the Bcl-2 family(Robinson et al 2003) The dexamethasone-induced

apoptosis of lymphoid cells is also accompanied byalterations in gene expression including the induction

of BimEL a pro-apoptotic member of the Bcl-2 family

(Medh et al 2003) In neurons a set of RNAtranscripts appears to be upregulated during apoptosis

including those encoding caspase 3 the death receptorDR6 (Chiang et al 2001) Id2 (Gleichmann et al 2002)

and the pro-apoptotic Bcl-2 family member Bax

(Wullner et al 1998)As would be expected different cell lineages appear

to regulate different cohorts of RNA transcripts duringtheir apoptotic programmes However some apopto-

sis-associated RNA changes are conserved across

different cell lineages (including EC) such as theinduction of transcripts encoding pro-apoptotic

members of the Bcl-2 family Taken together thesedata suggest that in many cell types the altered

abundance during or immediately before apoptosisof transcripts encoding apoptotic regulators or

(a)

100

10

1

(b)

100

10

1

15 h 6 h

(c)

100

100

10

10

1

(d )

100

10

1

1

9 h 24 h

0 h 0 h

0 h

100101

0 h

Figure 5 A time course of transcriptome change during EC apoptosis Scatterplots show each transcript represented by a dotwith abundance at time 0 shown on each x -axis and abundance at other time points shown on the y-axes (a) 15 h (b) 6 h (c)9 h and (d ) 24 h SFD Those transcripts that are not regulated remain approximately on the diagonal Transcripts that appearedto be upregulated when cultures were exposed to SFD conditions for 24 h compared with healthy cultures at time 0 arehighlighted in white The gradual upregulation of these transcripts over time can be seen


apoptotic machinery may play a lsquofeed-forwardrsquo role toassist the apoptotic programme

8 WHAT ARE THE LIMITATIONS OF TRADITIONALGENE ARRAYBIOINFORMATIC TECHNIQUES FORTHE STUDY OF EC APOPTOSISAs discussed above traditional bioinformatic tech-niques have revealed interesting new paradigms for thecontrol of EC apoptosis However they do have reallimitations While these limitations should not deterprospective EC researchers from using transcriptomeprofiling studies they require consideration

(i) Owing to the false discovery introduced by noiseor individual differences (Schoenfeld et al 2004)in addition to any annotation or contaminationerrors inherent in the gene array platform itself aproportion of the results of every gene arrayexperiment will be incorrect Researchers andclinicians using microarray techniques mustaccept this low rate of error However acceptinga low rate of error is not as difficult as it soundsAfter all deciding on an acceptable probability oferror forms the basis of most statistical analysesof scientific experiments

(ii) The results obtained when the same RNA isanalysed using different gene array platforms are


not always concordant (Jordan 2004) This has

shaken the confidence of some researchers in

gene array technology (Tan et al 2003) This

problem is illustrated by the microarray data we

describe in this paper Our previous study of the

effect of 28 h SFD on HUVEC used Affymetrix

U95A genechips and highlighted 79 regulated

transcripts (listed in file 3 in the electronic

supplementary material of Johnson et al 2004)

The CodeLink genechip time-course study

presented in this paper used the same cell type

and SFD conditions and contained a 24 h time

point We compared the transcripts apparently

regulated by 28 h SFD in the Affymetrix study

with the transcripts apparently regulated by 24 h

SFD in the CodeLink study Of the 79 transcripts

highlighted by the Affymetrix study 48 could be

identified by reference to Unigene IDs in the

CodeLink gene arrays but only 30 of these

appeared to be regulated by 15-fold or more in a

congruent direction to the Affymetrix study

Although the lack of full concordance is con-

cerning some of the differences between the

results of our two studies may be due to

experimental differences Other differences may

be due to occasional misannotation of one or

both gene array platforms since it was recently

Table 1 The kinetics of SFD-dependent regulation of transcript abundance Values represent median fold change (relative to0 h) of normalized transcript abundance of the triplicate time-course data Negative values represent downregulation Positivevalues represent upregulation

05 h 15 h 3 h 6 h 9 h 12 h 24 h

cell cycle and mitosisCCNA2 K103 101 K108 K119 K121 K167 K362CCNE 101 103 K105 K132 K191 K329 K425CCNH 102 105 K107 K141 K143 K168 K169CDC6 120 142 130 K103 K143 K227 K411CDC28 K119 K123 K116 K112 K121 K117 K257KNSL1 K104 K004 104 114 104 K129 K349

apoptosis regulatorsTRAIL K103 187 174 291 328 459 1162LARD K103 K109 125 235 268 317 256

growth factorsANGPT2 K108 102 K164 K242 K311 K449 K193IL-8 K151 K429 K392 K330 K155 107 104

signalling componentsRELB 102 120 102 138 317 573 302chaperonin 10 103 K102 K103 K109 K126 K151 K245

10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01

set 7 14 genes 14 in list







set 8 7 genes 7 in list unclassified 19685 genes 0 in list2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9norm

aliz

ed in

tens

ity(l

og s

cale

)no

rmal

ized

inte

nsity

(log

sca

le)

norm

aliz

ed in

tens

ity(l

og s

cale

)

Figure 6 k-means clustering revealed eight groups of transcripts (from a shortlist of 685 highly regulated transcripts) thatfollowed distinct temporal patterns of regulation


shown that matching of gene array probes

between platforms by actual sequences rather

than by Unigene IDs dramatically improved the

apparent concordance (Mecham et al 2004)

This microarray cross-platform discordance

problem urgently requires further study The

safest approach when possible may be to focus on

results that have been confirmed by two inde-

pendent gene array platforms or by one platform

and quantitative polymerase chain reactions


(iii) Finally there is an important but hidden

limitation of traditional bioinformatic tech-

niques Although techniques such as clustering

(Eisen et al 1998) ICA (Saidi et al 2004) and

self organizing maps (Tamayo et al 1999) have

been extremely valuable in revealing hidden

patterns of transcript co-regulation in general

they are not good at revealing the causalrelationships among molecular signals within

cells However it is the combination of the cause


and effect relationships between all the signalsoperating in a cell rather than the isolatedactions of individual signals that drive andregulate processes such as apoptosis

9 THE CONTRIBUTION AND LIMITATIONS OFTHE PROTEOMEAn obvious limitation of gene array data is that it doesnot give direct information about proteins or post-transcriptional regulation Although the use of genearray technology has helped to enrich our knowledge ofthe key regulatory genes involved in processes such asEC apoptosis understanding regulation at the level ofthe proteome will be of more direct value Largeamounts of useful information have already emergedfrom proteomic studies of EC (Bruneel et al 2003Kamino et al 2003 Nylund amp Leszczynski 2004Sprenger et al 2004 Yu et al 2004) Notable amongthese is a subtractive proteomics study that identifiedtumour-associated EC surface proteins which maybe targets for anti-tumour therapy In fact radio-immunotherapy directed against one of the EC surfaceproteins identified annexin A1 is effective in vivo (Ohet al 2004) Another notable study from the sameresearch group illustrated the differences betweenin vitro and in vivo lung microvascular ECs andhighlighted the need for replicated in vivo confirmationof in vitro results (Durr et al 2004) A study of theproteomic changes that occur during etoposide-induced EC apoptosis has suggested that multipleprotein-based signalling cascades including the mito-chondrial and the ER stress pathways may play a role(Bruneel et al 2005)

The information derived from these proteomicsstudies could not have been derived from transcrip-tome studies since it is very difficult to correlate geneexpression levels with protein abundance (Gygi et al1999 Lian et al 2002) Although technologies such asprotein microarrays are rapidly developing and nowenable relatively specific and sensitive analysis ofproteins (Letarte et al 2005) we believe that at presenttranscriptome studies still remain extremely viableThe best approach in the future may be to combine theresults of proteomics with that of transcriptomeanalysis using systems biology techniques (see below)

10 INFERENCE OF CAUSAL RELATIONSHIPSFROM TRANSCRIPTOME DATAMethods to infer cause and effect relationships basedon transcriptome data are emerging Of these methodssome researchers believe the technique of generegulatory network analysis may hold particularpromise (Brazhnik et al 2002 Schlitt amp Brazma2005) Practitioners of gene regulatory networktranscriptome analysis claim to produce lsquocircuitdiagramsrsquo that show the hierarchical networks ofcause and effect transcription control relationshipsthat underlie a biological process like EC apoptosis orinterventions like drug treatment These gene regulat-ory networks are specific to the cell type under studyUnlike the traditional post-translational signallingnetwork charts that most biologists are used to seeing


the nodes on gene regulatory network graphs representtranscripts rather than proteins The links between thenodes (called edges) represent putative cause and effectrelationships between transcripts where the abun-dance of one transcript through one or moreintervening protein-based steps can regulate theabundance of a second transcript Therefore althoughproteins are not explicitly shown in gene regulatorynetwork graphs they provide the hidden engine thatunderlies the network structure

Several mathematical methods have been used toinfer gene regulatory networks such as differentialequation models (Chen et al 1999 de Hoon et al2003) state space models (Rangel et al 2004) Booleannetwork models (Akutsu et al 1998 Shmulevich et al2002) and Bayesian network models (Friedman et al2000 Imoto et al 2002) In some situations acombination of these techniques is helpful Forexample Boolean and Bayesian network algorithmscan be combined (Imoto et al 2003b) to infer genenetworks for response to a drugmdashthe Boolean networkapproach is used initially to identify the set oftranscripts regulated in abundance by the drug and aBayesian network approach is then used to map theupstream regulators of the drug-affected genes whichprovide potential lsquodruggablersquo targets This combinationof approaches has been used successfully in yeast(Aburatani et al 2003) to determine that the proteinregulated by the oral antifungal drug griseofulvin isCIK1 (Savoie et al 2003)

Some features of gene regulatory network inferencetechniques appear especially important when inferringnetworks from microarray data For example tran-scripts are not always regulated in a linear mannertherefore capturing nonlinear relationships betweentranscripts can greatly increase the power of networkanalysis This can be achieved by the use of non-parametric additive regression models (Imoto et al2002) In addition most gene expression data containmany outlying data points and show heteroscedasticity(the dispersion of distributions differs between tran-scripts) Therefore modelling which takes the effects ofoutliers and heteroscedasticity into account isespecially well suited to inferring networks frommicroarray data (Imoto et al 2003a) Like allbioinformatics techniques both the quality andquantity of the available data determine the statisticalpower of gene network techniques

To illustrate the principles of applying gene regulat-ory networks to endothelial biology we have used thedynamic Bayesian network inference method (Kimet al 2003) to generate a gene network based on themedian of the HUVEC apoptosis time-course genearray data which was described above A graphrepresenting this network is shown in figure 7a Inthis network the regulation of abundance of mostparents and their multiple children over the apoptosistime course appears to be correlated with the up- ordownregulation of the putative children often laggingbehind that of their parents (figure 7b and file 2 in theelectronic supplementary material) In this gene net-work the parent node with the most children is GABAreceptor-associated protein (GABARAP) Accordingto this network GABARAP is potentially acting as a


novel transcriptional lsquohubrsquo during the process of ECapoptosis Interestingly GABARAP has recently beenshown in other laboratories to induce apoptosis byinteracting with DEAD box polypeptide 47 (Lee et al2005) and appears to be encoded by a tumoursuppressor gene mutated in breast cancer (Klebiget al 2005) To illustrate the potential function ofGABARAP in the regulation of EC apoptosis weknocked down the expression of GABARAP andluciferase (a negative control) in HUVEC usingsiRNA RT-PCR analysis indicated that 24 h afteranti-GABARAP siRNA treatment the abundance ofGABARAP mRNA was reduced to less than 20 of itsformer levels (data not shown) siRNA-mediatedreduction of GABARAP expression appeared to inhibitHUVEC apoptosis The median incidence of apoptosisinduced in HUVEC by 24 h of SFD (performed as inJohnson et al 2004) was 117 in cells treated withanti-luciferase siRNA (control) but only 58 in cellstreated by anti-GABARAP siRNA (figure 7c) Analysisof this data using the MannndashWhitney two-tailed test anon-parametric statistical method gave a statisticallysignificant result of UZ4265 and p0001 Inaddition the morphological changes induced inHUVEC by 24 h of SFD appeared to be reduced bythe inhibition of GABARAP expression (figure 7d )This suggests that GABARAP may indeed be aregulator of EC apoptosis as is suggested by itsprominent position in the gene network Anothernode that appears to act as transcription hub duringapoptosis is interferon regulatory factor-7 (IRF7) atranscription factor known to be regulated by the TNFand NFkB families Interestingly the TNF familymember TRAIL and the NFkB family member RELBlie upstream of IRF7 in our gene network Thecomplete parentndashchild relationships for this genenetwork are given in file 2 in the electronic supple-mentary material Web-based tools such as CellIllustrator (httpwwwcellillustratorcom) and Cytos-cape (httpwwwcytoscapeorg) can be used by thereader to explore this putative network and the gene-to-gene relationships it symbolizes

As a further illustration of gene network analysistechniques we also generated a second gene networkfrom the same apoptosis time-course gene array dataThis network differs from the one shown in figure 7afrom which it was derived by the incorporation ofnew gene array data from eight siRNA disruptantexperiments This new gene array data was used as aBayesian prior and is provided as file 3 in the electronicsupplementary material In these eight disruptantexperiments specific RNAs related to cell cycle control(CDC45L CCNE1 CENPA CENPF CDC2CDC25C MCM6 and CDC6) were reduced inabundance by more than 65 in HUVEC usingsiRNA pools Gene array analysis showed that thesesiRNA treatments caused large numbers of transcriptsto be regulated (presumably as a consequence of thetargeted RNA downregulation or possibly in somecases also as a consequence of unexpected off-targetsiRNA effects) An example of the effects of siRNAtreatment on the HUVEC transcriptome is shown infigure 8ab The putative time-course gene networkmodified by the incorporation of this disruptant


information as a Bayesian prior is shown in figure 8cand its complete parentndashchild relationships are listed infile 4 in the electronic supplementary material Of the488 edges contained in the modified gene networkshown in figure 8c 338 are shared with the unmodifiedgene network shown in figure 7a Of the 150 edges foundin the modified but not the unmodified networks 94have as parents one of the eight RNAs that were targetedin the siRNA disruptant experiments These particularedges appear to have inherited cause and effect infor-mation from the disruptant gene array data since theyall correctly predict the effects on children of disruptingparents by siRNA treatment (data not shown) There-fore the inclusion of disruptant data as a Bayesian priorappears to enhance the ability of dynamic time-coursegene networks to predict specific directional relationshipsbetween transcripts It seems possible that the additionof data from biological database annotations andproteomics experiments could further enhance thepredictive power of this type of gene network We andother researchers are generating larger-scale generegulatory networks using gene array data fromhundreds of disruptant experiments which will becombined with time-course data using Bayesian tech-niques as illustrated above (Imoto et al 2006) Webelieve these large networks will provide valuablehypotheses about functional interactions within ECswhich can then be tested experimentally

Therefore gene regulatory network technologiesalong with other techniques for inferring cause and effectrelationships in cells from transcriptome data promise agreat deal for vascular biology research and foraccelerating the rational design of new cardiovascularmedications They will be especially valuable if theirresults are used to compliment those of traditional cellbiology studies and EC proteomic studies currentlyunderway Cautious integration of the cause and effectrelationships that are predicted by gene networks intosystems biology databases (Aggarwal amp Lee 2003) maybe especially useful A study showing the usefulness ofthis approach modelled a yeast galactose utilizationpathway by analysing experimental perturbations usinginformation from microarrays proteomics and computerdatabases (Ideker et al 2001) Another example is a studythat defined pathways that control endoderm andmesoderm specification during sea urchin developmentIn this study the results of perturbation analyses werecombined with information from computer databasescis-regulatory analysis and molecular embryology(Davidson et al 2002) In Professor Eugene Butcherrsquoslaboratory systems biology approaches using smallamounts of data from cultured ECs have already revealedimportant new information about the role of NFkB andRas EC signalling pathways (Plavec et al 2004) Theaddition of data from gene regulatory networks maydramatically increase the power of this type of systemsbiology analysis

However a critical challenge for technologies suchas gene regulatory networks is that the predictions theymake must be extensively tested using traditional lsquowetrsquolaboratory experiments This will require clear com-munication and extensive cooperation between bioin-formaticians and biologistsclinicians Hopefully theemerging generation of vascular biologists fluent in

(a)

(b) 25

20

15

10

30

20

10

0

RN

A tr

ansc

ript

abu

ndan

ce

perc

enta

ge o

f ac

tive

casp

ase

3po

sitiv

e ce

lls

5

0 10

hours after serum deprivation

20 30LUC

anti-luciferase siRNA anti-GABARAP siRNA

GAB

(d )

(c)

GABARAP

CDC45L

ASNS

RRM1

TCOF1CDC6

FAM46A

C9 or f81

Figure 7 An example of a dynamic time-course gene regulatory network (a) A graph representing a dynamic Bayesian genenetwork generated from the median of triplicated apoptosis time-course data Dots represent transcripts (lsquonodesrsquo) and linesbetween the dots represent potential cause and effect interactions (lsquoedgesrsquo) (b) A graph showing apoptosis time-course genearray transcript abundance data for a parent RNA transcript in the network (CDKN1C black) and its putative children(C1QTNF5 green AKR1C3 blue MLF1 orange SUV39H1 red) (c) Reduction of GABARAP mRNA abundance inhibitsHUVEC apoptosis Incidence of apoptosis induced by 24 h of SFD in anti-GABARAP siRNA-transfected HUVEC (GAB) andanti-luciferae siRNA-transfected HUVEC (LUC control) were assessed by immunocytochemical detection of caspase 3cleavage Horizontal bars indicate median values (d ) Images of HUVEC cultures deprived of serum for 24 h after treatmentwith anti-luciferase siRNA (control) and anti-GABARAP siRNA Fields are 225300 mM in size



(a)

(c)

siRNA againstluciferase

no siRNA treatment

1 10 100

100

10

1

(b)

siRNA againstNFκB p105

siRNA against luciferase

1 10 100

Figure 8 An example of a dynamic time-course gene regulatory network modified by the inclusion of prior information fromsiRNA disruptant experiments (a) A scatterplot comparing transcript abundance in untransfected HUVEC (x -axis) withtranscript abundance in mock HUVEC transfected with control siRNAs directed against luciferase ( y-axis)mdashlittle regulation oftranscript abundance appears to have occurred (b) A scatterplot comparing transcript abundance in HUVEC transfected withsiRNAs directed against luciferase (x -axis) with transcript abundance in HUVEC transfected with siRNAs directed againstNFkB p105 ( y-axis)mdashNFkB p105 (circled) was downregulated more than fourfold and a large number of additional transcriptswere also regulated in abundance (c) A graph representing an apoptosis time-course gene network generated that had beenmodified by incorporating as a Bayesian prior gene array data from eight siRNA disruptant experiments similar to those shownin (ab) Edges in common with the network shown in figure 7a are green and edges present in this network but not in thenetwork shown in figure 7a are pink


both bioinformatics and wet laboratory experi-mentation will be able to meet this challenge

11 CONCLUSIONSEC apoptosis appears to play an essential role in bloodvessel development and remodelling and has beenassociated with cardiovascular pathologies Thereforeunderstanding the regulation of EC apoptosis with thegoal of intervening in this process has become a currentresearch focus Much is known about the protein-basedsignalling and cleavage cascades that regulate ECapoptosis However the potential role of pro-apoptotictranscriptome and glycome changes in the orchestrationof EC death has only recently been recognized For acomplete understanding of EC apoptosis we now needto define the complex relationships between the molecu-lar signals that operate in these cells New technologiessuch as gene regulatory networks that promise to infercause and effect relationships from EC transcriptomedata provide powerful tools especially when integratedwith information from proteomic and cell biology studies


and with computer databases However the predictionsmade by all of these techniques must be extensivelyvalidated by laboratory experiments This will requireclose cooperation between bioinformaticians and experi-mental biologists and is likely to be driven by a newgeneration of cross-disciplinary scientists fluent in thelanguages of both fields

APPENDIX MATERIAL AND METHODSHUVEC were isolated from umbilical cords bycollagenase digestion and cultured at 378C5 CO2 inbasal culture medium supplemented with a proprietarymixture of heparin hydrocortisone epidermal growthfactor FGF and 2 foetal calf serum (FCS EGM-2Cambrex Workingham UK) HUVEC from 10individuals were pooled plated at 48105 cells in25 cm2 flasks and allowed to recover for 24 h at whichtime they were approximately 60 confluent Supple-mented medium was then removed and replaced withbasal culture medium containing only 2 charcoalstripped FCS (Invitrogen Paisley UK) At time points


0 05 15 3 6 9 12 and 24 h total RNA was preparedusing Trizol reagent (Invitrogen) and assessed using anAgilent 2100 bioanalyser Biotin-labelled complexcRNAs were prepared and hybridized to CodeLinkUniSet Human 20K gene chips according to themanufacturerrsquos protocols (GE Healthcare AmershamUK) To generate triplicate data (of biological repli-cates) the protocol was repeated two further times usingHUVEC from 10 different individuals for each poolThe quality of the expression data from all gene chipswas confirmed using CodeLink expression analysisSoftware (v 41) To ensure that expression levels werecomparable between the arrays the datawas normalizedusing intensity-dependent scaling as described inSchoenfeld et al (2004) To knock down the abundanceof RNA transcripts in HUVEC siRNA lsquosmartpoolsrsquofrom Dharmacon Inc (Lafayette Colorado USA) weretransfected into HUVEC using the siFectamine trans-fection reagent (ICVEC London UK) used accordingto the manufacturers instructions The incidence ofHUVEC apoptosis was assessed by cytospinningHUVEC onto glass slides followed by immunohisto-chemical staining using an anti-active caspase 3antibody (rabbit polyclonal Promega SouthamptonUK 1 250 dilution as described in Johnson et al(2004)) For each experimental condition the stainingof more than 1000 cells was counted

Data were log2-transformed and log ratios betweeneach time point and the 0 h control calculated Inaddition the ratio between each time point and 24 hwas also calculated to capture regulation at both ends ofthe time course For each transcript at each time pointZ-scores were calculated Log2 ratios of individualtranscripts were subtracted from the mean of log2 ratiosfor that time point and then divided by the standarddeviation of log2 ratios for that time point For each time-course replicate transcripts that had lsquogoodrsquo- or lsquolowrsquo-quality flags (based on CodeLink expression analysissoftware results) and aZ-score ofK2ZC2 at three ormore adjacent time pointswere selected for further studyTranscripts that contained lsquogood flags onlyrsquo and had aZ-score of K15ZC15 at three or more adjacenttime points were also selected Those regulated tran-scripts concordant between the triplicate time-coursedata were selected for the network generation Inaddition transcripts that were known to be involved inapoptosis and were significantly regulated in a previousAffymetrix apoptosis study (Johnson et al 2004) thatwere also regulated by greater than 15-fold in thetriplicate CodeLink datawere included This gave a list of276 transcripts for apoptosis network building

Dynamic Bayesian apoptosis time-course networkswere generated from the list of 276 apoptosis regulatedtranscripts as described by Kim et al (2003) Datagenerated from eight siRNA knock down arrays wereused to make an array prior to the apoptosis time-coursenetwork The eight knockdown arrays were selected froma list of 270 different knockdown arrays based on thefollowing criteria (i) the transcript knocked down is inthe list of 276 transcripts regulated in the apoptosis time-course triplicate data (as explained above) and (ii) a largenumber of the regulated genes resulting from the knockdown were also contained in the 276 list of apoptosis-regulated transcripts Gene networks both with and


without this prior information were inferred from theapoptosis data as described (Kim et al 2003) Compari-sons of the gene networks generated with and withoutknockdown prior information and their visualizationand data mining were carried out using the softwareiNet (now known as Cell Illustrator) which can bedownloaded from httpwwwcellillustratorcom

We wish to thank Mrs Deborah Sanders and Mrs SallyHumphreys for assistance with HUVEC culture as well asProf Satoru Kuhara for discussions about gene regulatorynetworks Prof Stephen Smith and Dr Shiladitya Senguptaco-supervised some the microarray and glycomic studiesdescribed in this review which were funded by TheBiotechnology and Biological Sciences Research CouncilThe Wellcome Trust and by a research collaborationagreement between Cambridge University and the bio-technology company GNI Ltd Computation time wasprovided by the Super Computer System Human GenomeCentre Institute of Medical Science University of Tokyo

REFERENCESAburatani S et al 2003 Discovery of novel transcription

control relationships with gene regulatory networks gener-ated from multiple-disruption full genome expressionlibraries DNA Res 10 1ndash8 (doi101093dnares1011)

Ackermann E J Taylor J K Narayana R amp BennettC F 1999 The role of antiapoptotic Bcl-2 familymembers in endothelial apoptosis elucidated with anti-sense oligonucleotides J Biol Chem 274 11 245ndash11 252(doi101074jbc2741611245)

Adams J M et al 1999 Control of apoptosis in hemato-poietic cells by the Bcl-2 family of proteins Cold SpringHarb Symp Quant Biol 64 351ndash358 (doi101101sqb199964351)

Aggarwal K amp Lee K H 2003 Functional genomics andproteomics as a foundation for systems biology BriefFunct Genomics Proteomics 2 175ndash184 (doi101093bfgp23175)

Akerman G S et al 2005 Alterations in gene expressionprofiles and the DNA-damage response in ionizingradiation-exposed TK6 cells Environ Mol Mutagen 45188ndash205 (doi101002em20091)

Akutsu T Kuhara S Maruyama O amp Miyano S 1998 Asystem for identifying genetic networks from geneexpression patterns produced by gene disruptions andoverexpressions Genome Inform Ser Workshop GenomeInform 9 151ndash160

Aoudjit F amp Vuori K 2001 Matrix attachment regulatesFas-induced apoptosis in endothelial cells a role for c-flipand implications for anoikis J Cell Biol 152 633ndash643(doi101083jcb1523633)

Arfuso F amp Meyer G T 2003 Apoptosis does not affect thevasculature of the corpus luteum of pregnancy in the ratApoptosis 8 665ndash671 (doi101023A1026283414726)

Bach F H Hancock W W amp Ferran C 1997 Protectivegenes expressed in endothelial cells a regulatory responseto injury Immunol Today 18 483ndash486 (doi101016S0167-5699(97)01129-8)

Baffert F Le T Sennino B Thurston G Kuo C JHu-Lowe D amp McDonald D M 2006 Cellular changes innormal blood capillaries undergoing regression after inhi-bition of VEGF signalingAm J PhysiolHeart Circ Physiol290 H547ndashH559 (doi101152ajpheart006162005)

Bannerman D D Tupper J C Ricketts W A BennettC F Winn R K amp Harlan J M 2001 A constitutivecytoprotective pathway protects endothelial cells fromlipopolysaccharide-induced apoptosis J Biol Chem 27614 924ndash14 932 (doi101074jbcM100819200)


Baumer S et al 2006 Vascular endothelial cell-specific

phosphotyrosine phosphatase (VE-PTP) activity is

required for blood vessel development Blood 107

4754ndash4762 (doi101182blood-2006-01-0141)

Benjamin L E Hemo I amp Keshet E 1998 A plasticity

window for blood vessel remodelling is defined by pericyte

coverage of the preformed endothelial network and is

regulated by PDGF- B and VEGF Development 125

1591ndash1598

Biancone L Martino A D Orlandi V Conaldi P G

Toniolo A amp Camussi G 1997 Development of inflam-

matory angiogenesis by local stimulation of Fas in vivoJ Exp Med 186 147ndash152 (doi101084jem1861147)

Bishop E T Bell G T Bloor S Broom I J Hendry

N F K amp Wheatley D N 1999 An in vitro model of

angiogenesis basic features Angiogenesis 3 335ndash344

(doi101023A1026546219962)

Blagosklonny M V 2003 Apoptosis proliferation differen-

tiation in search of the order Semin Cancer Biol 13

97ndash105 (doi101016S1044-579X(02)00127-X)

Bombeli T Karsan A Tait J F amp Harlan J M 1997

Apoptotic vascular endothelial cells become procoagulant

Blood 89 2429ndash2442

Brazhnik P de la Fuente A amp Mendes P 2002 Gene

networks how to put the function in genomics Trends

Biotechnol 20 467ndash472 (doi101016S0167-7799(02)

02053-X)

Bruneel A Labas V Mailloux A Sharma S Vinh J

Vaubourdolle M amp Baudin B 2003 Proteomic study of

human umbilical vein endothelial cells in culture

Proteomics 3 714ndash723 (doi101002pmic200300409)

Bruneel A Labas V Mailloux A Sharma S Royer N

Vinh J Pernet P Vaubourdolle M amp Baudin B 2005

Proteomics of human umbilical vein endothelial cells

applied to etoposide-induced apoptosis Proteomics 5

3876ndash3884 (doi101002pmic200401239)

Cardier J E amp Erickson-Miller C L 2002 Fas (CD95)-

and tumor necrosis factor-mediated apoptosis in liver

endothelial cells role of caspase-3 and the p38 MAPK

Microvasc Res 63 10ndash18 (doi101006mvre20012360)

Carmeliet P 2000 Mechanisms of angiogenesis and arter-

iogenesis Nat Med 6 389ndash395 (doi10103874651)

Carmeliet P amp Jain R K 2000 Angiogenesis in cancer and

other diseases Nature 407 249ndash257 (doi101038

35025220)

Carmeliet P et al 1999 Targeted deficiency or cytosolic

truncation of the VE-cadherin gene in mice impairs

VEGF-mediated endothelial survival and angiogenesis

Cell 98 147ndash157 (doi101016S0092-8674(00)81010-7)

Chen T He H L amp Church G M 1999 Modeling gene

expression with differential equations Pac Symp Biocom-

put 99 29ndash40

Chen J Somanath P R Razorenova O Chen W S Hay

N Bornstein P amp Byzova T V 2005 Akt1 regulates

pathological angiogenesis vascular maturation and per-

meability in vivo Nat Med 11 1188ndash1196 (doi101038

nm1307)

Chiang L W Grenier J M Ettwiller L Jenkins L P

Ficenec D Martin J Jin F DiStefano P S amp Wood

A 2001 An orchestrated gene expression component of

neuronal programmed cell death revealed by cDNA array

analysis Proc Natl Acad Sci USA 98 2814ndash2819

(doi101073pnas051630598)

Cho A Courtman D W amp Langille B L 1995 Apoptosis

(programmed cell death) in arteries of the neonatal lamb

Circ Res 76 168ndash175

Davidson E H et al 2002 A genomic regulatory network for

development Science 295 1669ndash1678 (doi101126

science1069883)


de Hoon M J Imoto S Kobayashi K Ogasawara N ampMiyano S 2003 Inferring gene regulatory networks fromtime-ordered gene expression data of Bacillus subtilis usingdifferential equations Pac Symp Biocomput 8 17ndash28

Desmouliere A Redard M Darby I amp Gabbiani G 1995Apoptosis mediates the decrease in cellularity during thetransition between granulation tissue and scar AmJ Pathol 146 56ndash66

Dickson S E Bicknell R amp Fraser H M 2001 Mid-lutealangiogenesis and function in the primate is dependent onvascular endothelial growth factor J Endocrinol 168409ndash416 (doi101677joe01680409)

Djonov V Andres A C amp Ziemiecki A 2001 Vascularremodelling during the normal and malignant lifecycle of the mammary gland Microsc Res Tech 52182ndash189 (doi1010021097-0029(20010115)522182AID-JEMT1004O30CO2-M)

Dorstyn L Colussi P A Quinn L M Richardson H ampKumar S 1999 DRONC an ecdysone-inducibleDrosophila caspase Proc Natl Acad Sci USA 964307ndash4312 (doi101073pnas9684307)

Durand E et al 2004 Invivo induction of endothelial apoptosisleads tovessel thrombosis and endothelial denudation a clueto the understanding of the mechanisms of thromboticplaque erosion Circulation 109 2503ndash2506 (doi10116101CIR00001301726248190)

Duriez P J Wong F Dorovini-Zis K Shahidi R ampKarsan A 2000 A1 functions at the mitochondria to delayendothelial apoptosis in response to tumor necrosis factorJ Biol Chem 275 18 099ndash18 107 (doi101074jbcM908925199)

Durr E Yu J Krasinska K M Carver L A Yates J RTesta J E Oh P amp Schnitzer J E 2004 Directproteomic mapping of the lung microvascular endothelialcell surface in vivo and in cell culture Nat Biotechnol 22985ndash992 (doi101038nbt993)

Duval H Harris M Li J Johnson N amp Print C 2003New insights into the function and regulation ofendothelial cell apoptosis Angiogenesis 6 171ndash183(doi101023BAGEN000002139009275bc)

Eisen M B Spellman P T Brown P O amp Botstein D1998 Cluster analysis and display of genome-wideexpression patterns Proc Natl Acad Sci USA 9514 863ndash14 868 (doi101073pnas952514863)

Fierlbeck W Liu A Coyle R amp Ballermann B J 2003Endothelial cell apoptosis during glomerular capillarylumen formation in vivo J Am Soc Nephrol 141349ndash1354 (doi10109701ASN00000617797053006)

Filippatos G Ang E Gidea C Dincer E Wang R ampUhal B D 2004 Fas induces apoptosis in humancoronary artery endothelial cells in vitro BMC Cell Biol5 6 (doi1011861471-2121-5-6)

Fisher S A Langille B L amp Srivastava D 2000 Apoptosisduring cardiovascular development Circ Res 87856ndash864

Friedman N Linial M Nachman I amp Persquoer D 2000Using Bayesian networks to analyze expression dataJ Comput Biol 7 601ndash620 (doi101089106652700750050961)

Friis T Hansen A B Houen G amp Engel A M 2006Influence of angiogenesis inhibitors on endothelial cellmorphology in vitro APMIS 114 211ndash224 (doi101111j1600-04632006apm_189x)

Frisch S M Vuori K Ruoslahti E amp Chan-Hui P Y1996 Control of adhesion-dependent cell survival by focaladhesion kinase J Cell Biol 134 793ndash799 (doi101083jcb1343793)

Fuks Z Persaud R S Alfieri A McLoughlin MEhleiter D Schwartz J L Seddon A P Cordon-Cardo C amp Haimovitz-Friedman A 1994 Basic


fibroblast growth factor protects endothelial cells against

radiation-induced programmed cell death in vitro and

in vivo Cancer Res 54 2582ndash2590

Galli C Meucci O Scorziello A Werge T M Calissano

P amp Schettini G 1995 Apoptosis in cerebellar granule

cells is blocked by high KCl forskolin and IGF-1 through

distinct mechanisms of action the involvement of

intracellular calcium and RNA synthesis J Neurosci 15

1172ndash1179

Gao C Sun W Christofidou-Solomidou M Sawada M

Newman D K Bergom C Albelda S M Matsuyama

S amp Newman P J 2003 PECAM-1 functions as a specific

and potent inhibitor of mitochondrial-dependent apopto-

sis Blood 102 169ndash179 (doi101182blood-2003-01-

0003)

Gaytan F Morales C Bellido C amp Sanchez-Criado J E

2002 Selective apoptosis of luteal endothelial cells in

dexamethasone-treated rats leads to ischemic necrosis of

luteal tissue Biol Reprod 66 232ndash240 (doi101095

biolreprod661232)

Gerber H P Vu T H Ryan A M Kowalski J Werb Z

amp Ferrara N 1999 VEGF couples hypertrophic cartilage

remodeling ossification and angiogenesis during endo-

chondral bone formation Nat Med 5 623ndash628 (doi10

10389467)

Gleichmann M Buchheim G El-Bizri H Yokota Y

Klockgether T Kugler S Bahr M Weller M amp Schulz

J B 2002 Identification of inhibitor-of-differentiation 2

(Id2) as a modulator of neuronal apoptosis J Neurochem

80 755ndash762 (doi101046j0022-3042200200760x)

Golpon H A Fadok V A Taraseviciene-Stewart L

Scerbavicius R Sauer C Welte T Henson P M amp

Voelkel N F 2004 Life after corpse engulfment

phagocytosis of apoptotic cells leads to VEGF secretion

and cell growth FASEB J 18 1716ndash1718

Granville D J Shaw J R Leong S Carthy C M

Margaron P Hunt D W amp McManus B M 1999

Release of cytochrome c Bax migration Bid cleavage and

activation of caspases 2 3 6 7 8 and 9 during endothelial

cell apoptosis Am J Pathol 155 1021ndash1025

Greenhalgh D G 1998 The role of apoptosis in wound

healing Int J Biochem Cell Biol 30 1019ndash1030 (doi10

1016S1357-2725(98)00058-2)

Gygi S P Rochon Y Franza B R amp Aebersold R 1999

Correlation between protein and mRNA abundance in

yeast Mol Cell Biol 19 1720ndash1730

Hall A V amp Jevnikar A M 2003 Significance of endothelial

cell survival programs for renal transplantation Am

J Kidney Dis 41 1140ndash1154 (doi101016S0272-

6386(03)00345-7)

Hristov M Erl W Linder S amp Weber P C 2004

Apoptotic bodies from endothelial cells enhance the

number and initiate the differentiation of human endo-

thelial progenitor cells in vitro Blood 104 2761ndash2766

(doi101182blood-2003-10-3614)

Hughes S amp Chang-Ling T 2000 Roles of endothelial cell

migration and apoptosis in vascular remodeling during

development of the central nervous systemMicrocirculation

7 317ndash333 (doi101038sjmn7300119)

Ideker T et al 2001 Integrated genomic and proteomic

analyses of a systematically perturbed metabolic

network Science 292 929ndash934 (doi101126science

2925518929)

Imoto S Goto T amp Miyano S 2002 Estimation of genetic

networks and functional structures between genes by using

Bayesian networks and nonparametric regression Pac

Symp Biocomput 7 175ndash186

Imoto S Kim S Goto T Miyano S Aburatani S

Tashiro K amp Kuhara S 2003a Bayesian network and


nonparametric heteroscedastic regression for nonlinearmodeling of genetic network J Bioinform Comput Biol 1231ndash252 (doi101142S0219720003000071)

Imoto S Savoie C J Aburatani S Kim S Tashiro KKuhara S amp Miyano S 2003b Use of gene networks foridentifying and validating drug targets J BioinformComput Biol 1 459ndash474 (doi101142S0219720003000290)

Imoto S 2006 Computational strategy for discoveringdruggable gene networks from genome-wide RNAexpression profiles Pac Symp Biocomput 11 559ndash571

Johnson N A et al 2004 Endothelial cells preparing to dieby apoptosis initiate a program of transcriptome andglycome regulation FASEB J 18 188ndash190

Jordan B R 2004 How consistent are expression chipplatforms Bioessays 26 1236ndash1242 (doi101002bies20128)

Kamino H Hiratsuka M Toda T Nishigaki R OsakiM Ito H Inoue T amp Oshimura M 2003 Searching forgenes involved in arteriosclerosis proteomic analysis ofcultured human umbilical vein endothelial cells under-going replicative senescence Cell Struct Funct 28495ndash503 (doi101247csf28495)

Kerr J F Wyllie A H amp Currie A R 1972 Apoptosis abasic biological phenomenon with wide-ranging impli-cations in tissue kinetics Br J Cancer 26 239ndash257

Khodarev N N Kataoka Y Murley J S WeichselbaumR R amp Grdina D J 2004 Interaction of amifostine andionizing radiation on transcriptional patterns of apoptoticgenes expressed in human microvascular endothelial cells(HMEC) Int J Radiat Oncol Biol Phys 60 553ndash563(doi101016jijrobp200404060)

Kilpatrick Z E Cakouros D amp Kumar S 2005 Ecdysone-mediated upregulation of the effector caspase DRICE isrequired for hormone-dependent apoptosis in Drosophilacells J Biol Chem 25 11 981ndash11 986 (doi101074jbcM413971200)

Kim H S Hwang K K Seo J W Kim S Y Oh B HLee M M amp Park Y B 2000 Apoptosis and regulation ofBax and Bcl-X proteins during human neonatal vascularremodeling Arterioscler Thromb Vasc Biol 20 957ndash963

Kim S Y Imoto S amp Miyano S 2003 Inferring genenetworks from time series microarray data using dynamicBayesian networks Brief Bioinform 4 228ndash235 (doi101093bib43228)

Kiyono M amp Shibuya M 2006 Inhibitory Smad transcrip-tion factors protect arterial endothelial cells fromapoptosis induced by BMP4 Oncogene 25 7131ndash7137(doi101038sjonc1209700)

Klebig C Seitz S Arnold W Deutschmann N Pacyna-Gengelbach M Scherneck S amp Petersen I 2005Characterization of g-aminobutyric acid type A receptor-associated protein a novel tumor suppressor showingreduced expression in breast cancer Cancer Res 65394ndash400

Krupnick A S et al 2002 Mechanism of T cell-mediatedendothelial apoptosis Transplantation 74 871ndash876(doi10109700007890-200209270-00022)

Lang R A amp Bishop J M 1993 Macrophages are requiredfor cell death and tissue remodeling in the developingmouse eye Cell 74 453ndash462 (doi1010160092-8674(93)80047-I)

Lang R Lustig M Francois F Sellinger M amp PleskenH 1994 Apoptosis during macrophage-dependent oculartissue remodelling Development 120 3395ndash3403

Lee J H Rho S B amp Chun T 2005 GABAA receptor-associated protein (GABARAP) induces apoptosis byinteracting with DEAD (Asp-Glu-Ala-AspHis) boxpolypeptide 47 (DDX 47) Biotechnol Lett 27 623ndash628(doi101007s10529-005-3628-2)


Letarte M Voulgaraki D Hatherley D Foster-Cuevas

M Saunders N J amp Barclay N 2005 Analysis of

leukocyte membrane protein interactions using protein

microarrays BMC Biochem 6 2 (doi1011861471-

2091-6-2)

Li A Dubey S Varney M L Dave B J amp Singh R K

2003 IL-8 directly enhanced endothelial cell survival

proliferation and matrix metalloproteinases production

and regulated angiogenesis J Immunol 170 3369ndash3376

Lian Z Kluger Y Greenbaum D S Tuck D

Gerstein M Berliner N Weissman S M amp

Newburger P E 2002 Genomic and proteomic analysis

of the myeloid differentiation program global analysis of

gene expression during induced differentiation in the

MPRO cell line Blood 100 3209ndash3220 (doi101182

blood-2002-03-0850)

Limaye V S Li X Hahn C Xia P Berndt M C

Vadas M A amp Gamble J R 2005 Sphingosine kinase-1

enhances endothelial cell survival through a PECAM-1-

dependent activation of PI-3KAkt and regulation of Bcl-2

family members Blood 105 3169ndash3177 (doi101182

blood-2004-02-0452)

Lobov I B Brooks P C amp Lang R A 2002 Angiopoietin-

2 displays VEGF-dependent modulation of capillary

structure and endothelial cell survival in vivo Proc Natl

Acad Sci USA 99 11 205ndash11 210 (doi101073pnas

172161899)

Lobov I B et al 2005 WNT7b mediates macrophage-

induced programmed cell death in patterning of the

vasculature Nature 437 417ndash421 (doi101038

nature03928)

Madge L A Li J H Choi J amp Pober J S 2003 Inhibition

of phosphatidylinositol 3-kinase sensitizes vascular endo-

thelial cells to cytokine-initiated cathepsin-dependent

apoptosis J Biol Chem 278 21 295ndash21 306 (doi10

1074jbcM212837200)

Matsumoto M Nishinakagawa H Kurohmaru M

Hayashi Y amp Otsuka J 1992 Pregnancy and lactation

affect the microvasculature of the mammary gland in mice

J Vet Med Sci 54 937ndash943

Mecham B H et al 2004 Sequence-matched probes

produce increased cross-platform consistency and more

reproducible biological results in microarray-based gene

expression measurements Nucleic Acids Res 32 e74

(doi101093nargnh071)

Medh R D Webb M S Miller A L Johnson B H

Fofanov Y Li T Wood T G Luxon B A amp

Thompson E B 2003 Gene expression profile of

human lymphoid CEM cells sensitive and resistant to

glucocorticoid-evoked apoptosis Genomics 81 543ndash555

(doi101016S0888-7543(03)00045-4)

Meeson A Palmer M Calfon M amp Lang R 1996 A

relationship between apoptosis and flow during pro-

grammed capillary regression is revealed by vital analysis

Development 122 3929ndash3938

Meeson A P Argilla M Ko K Witte L amp Lang R A

1999 VEGF deprivation-induced apoptosis is a com-

ponent of programmed capillary regression Development

126 1407ndash1415

Mehta U Kang B P Bansal G amp Bansal M P 2002

Studies of apoptosis and bcl-2 in experimental athero-

sclerosis in rabbit and influence of selenium supple-

mentation Gen Physiol Biophys 21 15ndash29

Meier P Finch A amp Evan G 2000 Apoptosis in

development Nature 407 796ndash801 (doi101038

35037734)

Meredith Jr J E Fazeli B amp Schwartz M A 1993 The

extracellular matrix as a cell survival factor Mol Biol Cell

4 953ndash961


Mineo C Deguchi H Griffin J H amp Shaul P W 2006

Endothelial and antithrombotic actions of HDL Circ

Res 98 1352ndash1364 (doi10116101RES0000225982

0198893)

Mitchell C A Risau W amp Drexler H C 1998 Regression of

vessels in the tunicavasculosa lentis is initiatedbycoordinated

endothelial apoptosis a role for vascular endothelial growth

factor as a survival factor for endothelium Dev Dyn 213

322ndash333 (doi101002(SICI)1097-0177(199811)2133322AID-AJA8O30CO2-E)

Modlich U Kaup F J amp Augustin H G 1996 Cyclic

angiogenesis and blood vessel regression in the ovary

blood vessel regression during luteolysis involves endo-

thelial cell detachment and vessel occlusion Lab Invest74 771ndash780

Morris P N Dunmore B J amp Brindle N P 2006 Mutant

Tie2 causing venous malformation signals through Shc

Biochem Biophys Res Commun 346 335ndash338 (doi10

1016jbbrc200605128)

Nakagawa I Nakata M Kawabata S amp Hamada S 2004

Transcriptome analysis and gene expression profiles of

early apoptosis-related genes in Streptococcus pyogenes-

infected epithelial cells Cell Microbiol 6 939ndash952

(doi101111j1462-5822200400412x)

Niswender G D Juengel J L Silva P J Rollyson M K

amp McIntush E W 2000 Mechanisms controlling the

function and life span of the corpus luteum Physiol Rev

80 1ndash29

Norata G D Tonti L Roma P amp Catapano A L 2002

Apoptosis and proliferation of endothelial cells in early

atherosclerotic lesions possible role of oxidised LDL

Nutr Metab Cardiovasc Dis 12 297ndash305

Nylund R amp Leszczynski D 2004 Proteomics analysis of

human endothelial cell line EAhy926 after exposure to

GSM 900 radiation Proteomics 4 1359ndash1365 (doi10

1002pmic200300773)

Oh P Li Y Yu J Durr E Krasinska K M Carver

L A Testa J E amp Schnitzer J E 2004 Subtractive

proteomic mapping of the endothelial surface in lung and

solid tumours for tissue-specific therapy Nature 429

629ndash635 (doi101038nature02580)

Park J E Keller G A amp Ferrara N 1993 The vascular

endothelial growth factor (VEGF) isoforms differential

deposition into the subepithelial extracellular matrix and

bioactivity of extracellular matrix-bound VEGF Mol Biol

Cell 4 1317ndash1326

Plavec I et al 2004 Method for analyzing signaling networks

in complex cellular systems Proc Natl Acad Sci USA

101 1223ndash1228 (doi101073pnas0308221100)

Pollman M J Naumovski L amp Gibbons G H 1999

Endothelial cell apoptosis in capillary network remodel-

ing J Cell Physiol 178 359ndash370 (doi101002

(SICI)1097-4652(199903)1783359AID-JCP10O3

0CO2-O)

Porn-Ares M I Saido T C Andersson T amp Ares M P

2003 Oxidized low-density lipoprotein induces calpain-

dependent cell death and ubiquitination of caspase 3 in

HMEC-1 endothelial cells Biochem J 374 403ndash411

(doi101042BJ20021955)

Pru J K Lynch M P Davis J S amp Rueda B R 2003

Signaling mechanisms in tumor necrosis factor alpha-

induced death of microvascular endothelial cells of the

corpus luteum Reprod Biol Endocrinol 1 17 (doi10

11861477-7827-1-17)

Rangel C Angus J Ghahramani Z Lioumi M

Sotheran E Gaiba A Wild D L amp Falciani F 2004

Modeling T-cell activation using gene expression profiling

and state-space models Bioinformatics 20 1361ndash1372

(doi101093bioinformaticsbth093)


Rath P C amp Aggarwal B B 1999 TNF-induced signaling inapoptosis J Clin Immunol 19 350ndash364 (doi101023A1020546615229)

Raymond M A Desormeaux A Laplante P VigneaultN Filep J G Landry K Pshezhetsky A V amp HebertM J 2004 Apoptosis of endothelial cells triggers acaspase-dependent anti-apoptotic paracrine loop activeon VSMC FASEB J 18 705ndash707

Re F Zanetti A Sironi M Polentarutti N LanfranconeL Dejana E amp Colotta F 1994 Inhibition of anchorage-dependent cell spreading triggers apoptosis in culturedhuman endothelial cells J Cell Biol 127 537ndash546(doi101083jcb1272537)

Richardson B C Lalwani N D Johnson K J amp MarksR M 1994 Fas ligation triggers apoptosis in macrophagesbut not endothelial cells Eur J Immunol 24 2640ndash2645(doi101002eji1830241111)

Robinson M Jiang P Cui J Li J Wang Y SwaroopM Madore S Lawrence T S amp Sun Y 2003 Globalgenechip profiling to identify genes responsive to p53-induced growth arrest and apoptosis in human lungcarcinoma cells Cancer Biol Ther 2 406ndash415

Saidi S A Holland C M Kreil D P MacKay D JCharnock-Jones D S Print C G amp Smith S K 2004Independent component analysis of microarray data in thestudy of endometrial cancer Oncogene 23 6677ndash6683(doi101038sjonc1207562)

Sata M amp Walsh K 1998 Endothelial cell apoptosis inducedby oxidized LDL is associated with the down-regulation ofthe cellular caspase inhibitor FLIP J Biol Chem 27333 103ndash33 106 (doi101074jbc2735033103)

Sata M Hirata Y amp Nagai R 2002 Role of FasFas ligandinteraction in ischemia-induced collateral vessel growthHypertens Res 25 577ndash582 (doi101291hypres25577)

Satake S Kuzuya M Ramos M A Kanda S amp IguchiA 1998 Angiogenic stimuli are essential for survival ofvascular endothelial cells in three-dimensional collagenlattice Biochem Biophys Res Commun 244 642ndash646(doi101006bbrc19988313)

Savoie C J Aburatani S Watanabe S Eguchi Y MutaS Imoto S Miyano S Kuhara S amp Tashiro K 2003Use of gene networks from full genome microarraylibraries to identify functionally relevant drug-affectedgenes and gene regulation cascades DNA Res 10 19ndash25(doi101093dnares10119)

Schechner J S et al 2000 In vivo formation of complexmicrovessels lined by human endothelial cells in animmunodeficient mouse Proc Natl Acad Sci USA 979191ndash9196 (doi101073pnas150242297)

Schlitt T amp Brazma A 2005 Modelling gene networks atdifferent organisational levels FEBS Lett 5791859ndash1866 (doi101016jfebslet200501073)

Schoenfeld J et al 2004 Bioinformatic analysis of primaryendothelial cell gene array data illustrated by the analysisof transcriptome changes in endothelial cells exposed toVEGF-A and PlGF Angiogenesis 7 143ndash156 (doi101007s10456-004-1677-0)

Schulz J B Weller M amp Klockgether T 1996 Potassiumdeprivation-induced apoptosis of cerebellar granuleneurons a sequential requirement for new mRNA andprotein synthesis ICE-like protease activity and reactiveoxygen species J Neurosci 16 4696ndash4706

Segura I et al 2002 Inhibition of programmed cell deathimpairs in vitro vascular-like structure formation andreduces in vivo angiogenesis FASEB J 16 833ndash841(doi101096fj01-0819com)

Shmulevich I Dougherty E R Kim S amp Zhang W 2002Probabilistic Boolean networks a rule-based uncertaintymodel for gene regulatory networks Bioinformatics 18261ndash274 (doi101093bioinformatics182261)


Slowik M R Min W Ardito T Karsan A KashgarianM amp Pober J S 1997 Evidence that tumor necrosisfactor triggers apoptosis in human endothelial cellsby interleukin-1-converting enzyme-like protease- depen-dent and -independent pathways Lab Invest 77257ndash267

Sprenger R R Speijer D Back J W De Koster C GPannekoek H amp Horrevoets A J 2004 Comparativeproteomics of human endothelial cell caveolae and raftsusing two-dimensional gel electrophoresis and massspectrometry Electrophoresis 25 156ndash172 (doi101002elps200305675)

Stehlik C de Martin R Kumabashiri I Schmid J ABinder B R amp Lipp J 1998 Nuclear factor (NF)-kappaB-regulated X-chromosome-linked iap geneexpression protects endothelial cells from tumor necrosisfactor alpha- induced apoptosis J Exp Med 188211ndash216 (doi101084jem1881211)

Suri C Jones P F Patan S Bartunkova SMaisonpierre P C Davis S Sato T N ampYancopoulos G D 1996 Requisite role of angiopoietin-1 a ligand for the TIE2 receptor during embryonicangiogenesis Cell 87 1171ndash1180 (doi101016S0092-8674(00)81813-9)

Tamayo P Slonim D Mesirov J Zhu Q Kitareewan SDmitrovsky E Lander E S amp Golub T R 1999Interpreting patterns of gene expression with self-organiz-ing maps methods and application to hematopoieticdifferentiation Proc Natl Acad Sci USA 962907ndash2912 (doi101073pnas9662907)

Tan P K Downey T J Spitznagel Jr E L Xu P Fu DDimitrov D S Lempicki R A Raaka B M amp CamM C 2003 Evaluation of gene expression measurementsfrom commercial microarray platforms Nucleic Acids Res31 5676ndash5684 (doi101093nargkg763)

Tedgui A amp Mallat Z 2003 Apoptosis a majordeterminant of atherothrombosis Arch Mal CoeurVaiss 96 671ndash675

Tertemiz F Kayisli U A Arici A amp Demir R 2005Apoptosis contributes to vascular lumen formation andvascular branching in human placental vasculogenesisBiol Reprod 72 727ndash735 (doi101095biolreprod104034975)

Valgimigli M Merli E Malagutti P Soukhomovskaia OCicchitelli G Macri G amp Ferrari R 2003 Endothelialdysfunction in acute and chronic coronary syndromesevidence for a pathogenetic role of oxidative stress ArchBiochem Biophys 420 255ndash261 (doi101016jabb200307006)

Venkataraman G Shriver Z Davis J C amp SasisekharanR 1999 Fibroblast growth factors 1 and 2 are distinct inoligomerization in the presence of heparin-like glycosami-noglycans Proc Natl Acad Sci USA 96 1892ndash1897(doi101073pnas9651892)

Walker D L Vacha S J Kirby M L amp Lo C W 2005Connexin43 deficiency causes dysregulation of coronaryvasculogenesis Dev Biol 284 479ndash498

Wang S Sorenson C M amp Sheibani N 2005 Attenuationof retinal vascular development and neovascularizationduring oxygen-induced ischemic retinopathy in Bcl-2--mice Dev Biol 279 205ndash219 (doi101016jydbio200412017)

Weihua Z Tsan R Schroit A J amp Fidler I J 2005Apoptotic cells initiate endothelial cell sprouting viaelectrostatic signaling Cancer Res 65 11 529ndash11 535(doi1011580008-5472CAN-05-2718)

Winn R K amp Harlan J M 2005 The role of endothelial cellapoptosis in inflammatory and immune diseasesJ Thromb Haemost 3 1815ndash1824 (doi101111j1538-7836200501378x)


Woywodt A Streiber F de Groot K Regelsberger HHaller H amp Haubitz M 2003 Circulating endothelialcells as markers for ANCA-associated small-vessel vascu-litis Lancet 361 206ndash210 (doi101016S0140-6736(03)12269-6)

Wullner U Weller M Schulz J B Krajewski S ReedJ C amp Klockgether T 1998 Bcl-2 Bax and Bcl-xexpression in neuronal apoptosis a study of mutantweaver and lurcher mice Acta Neuropathol (Berl) 96233ndash238 (doi101007s004010050889)

Wyllie A H amp Arends M J 2003 Apoptosis in health anddisease In Oxford textbook of medicine (eds D A WarrellT M Cox J D Firth amp E J Benz) pp 119ndash130 4th ednOxford UK Oxford University Press

Yasumoto K Okamoto S Mukaida N Murakami SMai M amp Matsushima K 1992 Tumor necrosis factoralpha and interferon gamma synergistically induce inter-leukin 8 production in a human gastric cancer cell linethrough acting concurrently on AP-1 and NF-kB-likebinding sites of the interleukin 8 gene J Biol Chem 26722 506ndash22 511

Yu J Tian S Metheny-Barlow L Chew L J HayesA J Pan H Yu G L amp Li L Y 2001 Modulationof endothelial cell growth arrest and apoptosis byvascular endothelial growth inhibitor Circ Res 891161ndash1167


Yu M Chen D M Hu G amp Wang H 2004 Proteomic

response analysis of endothelial cells of human coronary

artery to stimulation with carbachol Acta Pharmacol Sin

25 1124ndash1130

Zaina G Morassutti C De Amicis F Fogher C amp

Marchetti S 2003 Endonuclease genes up-regulated in

tissues undergoing programmed cell death are expressed

during male gametogenesis in barley Gene 315 43ndash50

(doi101016S0378-1119(03)00820-5)

Zhang W Ramdas L Shen W Song S W Hu L amp

Hamilton S R 2003 Apoptotic response to 5-fluorouracil

treatment is mediated by reduced polyamines non-

autocrine Fas ligand and induced tumor necrosis factor

receptor 2 Cancer Biol Ther 2 572ndash578

Zhang W Li D amp Mehta J L 2004 Role of AIF in human

coronary artery endothelial cell apoptosis Am J Physiol

Heart Circ Physiol 286 H354ndashH358 (doi101152

ajpheart005792003)

Zheng L Dengler T J Kluger M S Madge L A

Schechner J S Maher S E Pober J S amp Bothwell

A L 2000 Cytoprotection of human umbilical vein

endothelial cells against apoptosis and CTL-mediated

lysis provided by caspase-resistant Bcl-2 without altera-

tions in growth or activation responses J Immunol 164

4665ndash4671

0 min 3 6 9 12 15

18 21 24 27 30 33

Figure 1 HUVEC were cultured to 70 confluence in their optimal medium washed and their media replaced with a low serum(2 charcoal-stripped FCS) medium Their morphology was followed over time in the presence of Hoechst 33342 and theirdeath imaged by time-lapse epifluorescence microscopy The following series of events was observed (i) loss of contact withmatrix and adjacent cells (ii) retraction of cytoplasm (iii) membrane blebbing and (iv) nuclear condensation (seen as bluecoloration representing the density of Hoechst fluorescence surpassing a pre-set threshold for display) Times are in minutes(adapted from Johnson et al 2004)


relationships between the molecular signals thatcontrol this process In sect10 of this review we willassess whether a deeper analysis of the EC transcrip-tome using emerging technologies such as generegulatory network analysis may help us understandthe complex signals that control EC fate Ourconclusion is that these emerging technologies havegreat potential for EC research especially when theirresults are integrated with those of cell biology studiesproteomics and in silico databases However it is criticalthat the cause and effect relationships predicted usingtechnologies such as gene regulatory networks arethoroughly validated by laboratory experiments

2 THE FUNCTION OF EC APOPTOSIS IN NORMALPHYSIOLOGY(a) EC apoptosis may contribute to blood vessel

remodelling

Several lines of evidence support the view that ECapoptosis plays an important role in vascular biologyand pathology Firstly apoptosis can be detectedduring developmental vessel regression in the aorticarches (Fisher et al 2000) the abdominal aorta (Choet al 1995) and the ductus arteriosus (Kim et al 2000)EC apoptosis may play an especially important role inthe eye in development of the tunica vasculosa lentis ofthe ocular lens (Mitchell et al 1998) and the pupillarymembrane of the anterior chamber of the eye (Langet al 1994) Macrophages may promote this ECapoptosis (Lang et al 1994) possibly through Wntsignalling (Lobov et al 2005)

Apoptosis in these regressing vessels is thought topromote their remodelling to meet the changing require-ments of the tissues they supply (Carmeliet 2000) Vesselregression may involve a classical series of events Forexample in one scenario following the inhibition ofVEGF-A signalling blood flow ceases followed by ECapoptosis and loss of pericytes leaving empty basementmembrane sheaves that can act as scaffolds for futurecapillary re-growth (Baffert et al 2006)

EC apoptosis appears to be less frequent in adultanimals than during development For example ECapoptosis appears to be much less common in adultrat kidney than in the developing rat kidney (Fierlbeck


et al 2003) Nevertheless adult ECs retain alatent capacity for apoptosis and EC apoptosis isdetected within specific adult tissues where substantialvessel remodelling occurs such as healing wounds(Desmouliere et al 1995 Greenhalgh 1998) regressingalveoli in the mammary gland after weaning (Matsumotoet al 1992 Djonov et al 2001) the developing placenta(Tertemiz et al 2005) and during the cyclical remodellingof female reproductive tissues (Modlich et al 1996Niswender et al 2000 Dickson et al 2001 Gaytan et al2002) Further support for a role of EC apoptosis invessel remodelling comes from in vitro studies Forexample incubation with pro-survival growth factors(Satake et al 1998) or retroviral-mediated anti-apoptoticBcl-2 expression (Pollman et al 1999) reduces theincidence of EC apoptosis and inhibits the regression ofvessel-like structures

It is important to note that two studies have beenunable to detect significant EC apoptosis withinregressing tissues where blood vessel remodellingwould be expected to occur (Hughes amp Chang-Ling2000 Arfuso amp Meyer 2003) In addition apoptosiswas not observed in one in vitro study where vesselswere disrupted by angiogenic inhibitors (Friis et al2006) However the majority of studies suggest thatEC apoptosis is frequently seen in remodelling vesselsimplying that EC apoptosis plays a role in theremodelling process

(b) EC apoptosis may contribute to blood vessel

morphogenesis and lumen formation

In addition to participating in vessel regression ECapoptosis may also contribute to angiogenesis andlumen formation Several EC types spontaneouslyform vessel-like structures in vitro (vascular morpho-genesis) when cultured on an extracellular matrix orwith matrix-producing cells Incubation of EC withanti-apoptotic caspase inhibitors or overexpression ofthe anti-apoptotic protein Bcl-2 appears to inhibitvascular morphogenesis in vitro (Segura et al 2002 andM Harris amp C Print 2003 unpublished results) Itinitially seems counterintuitive that EC apoptosis maycontribute to vascular morphogenesis However it ispossible that EC apoptosis contributes to this process


apoptosis




signallingcascades


transcription

RNAdegradation


uarrdarr















ECM components

ECMS

S

SSS

endothelial cell































et al 2004)















(a)

(b)

(c)

30

20

10

10 15min

10 15min

healthcells

(mole )


12

11

1

1

2

3

5

532

di-Smono-S

tri-S

8

7

76

10


1 (DU2SHNS6S) 458

102

101

0

148

25

75

92

173

284

780

0

365

00

101

0

4 (DU2SHNAc6S)

7 (DUHNAc6S)


6 (DU2SHNAc)


5 (DUHNS)

mA

Um

AU


































































(a)

100

10

1

(b)

100

10

1

15 h 6 h

(c)

100

100

10

10

1

(d )

100

10

1

1

9 h 24 h

0 h 0 h

0 h

100101

0 h




































05 h 15 h 3 h 6 h 9 h 12 h 24 h





10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01









6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9norm

aliz

ed in

tens

ity(l

og s

cale

)no

rmal

ized

inte

nsity

(log

sca

le)

norm

aliz

ed in

tens

ity(l

og s

cale

)








































(a)

(b) 25

20

15

10

30

20

10

0

RN

A tr

ansc

ript

abu

ndan

ce

perc

enta

ge o

f ac

tive

casp

ase

3po

sitiv

e ce

lls

5

0 10


20 30LUC


GAB

(d )

(c)

GABARAP

CDC45L

ASNS

RRM1

TCOF1CDC6

FAM46A

C9 or f81




(a)

(c)


no siRNA treatment

1 10 100

100

10

1

(b)



1 10 100































4754ndash4762 (doi101182blood-2006-01-0141)





1591ndash1598







(doi101023A1026546219962)



97ndash105 (doi101016S1044-579X(02)00127-X)







02053-X)









3876ndash3884 (doi101002pmic200401239)









35025220)




Cell 98 147ndash157 (doi101016S0092-8674(00)81010-7)



put 99 29ndash40





nm1307)






(doi101073pnas051630598)






science1069883)




























1172ndash1179






0003)





biolreprod661232)





10389467)





80 755ndash762 (doi101046j0022-3042200200760x)













1016S1357-2725(98)00058-2)







6386(03)00345-7)





(doi101182blood-2003-10-3614)




7 317ndash333 (doi101038sjmn7300119)




2925518929)






























2091-6-2)











blood-2002-03-0850)






blood-2004-02-0452)





172161899)




nature03928)





1074jbcM212837200)















(doi101016S0888-7543(03)00045-4)








126 1407ndash1415







35037734)



4 953ndash961




Res 98 1352ndash1364 (doi10116101RES0000225982

0198893)













1016jbbrc200605128)





(doi101111j1462-5822200400412x)




80 1ndash29








1002pmic200300773)













101 1223ndash1228 (doi101073pnas0308221100)




(SICI)1097-4652(199903)1783359AID-JCP10O3

0CO2-O)





(doi101042BJ20021955)





11861477-7827-1-17)
















































25 1124ndash1130





(doi101016S0378-1119(03)00820-5)









ajpheart005792003)







4665ndash4671


apoptosis




signallingcascades


transcription

RNAdegradation


uarrdarr















ECM components

ECMS

S

SSS

endothelial cell































et al 2004)















(a)

(b)

(c)

30

20

10

10 15min

10 15min

healthcells

(mole )


12

11

1

1

2

3

5

532

di-Smono-S

tri-S

8

7

76

10


1 (DU2SHNS6S) 458

102

101

0

148

25

75

92

173

284

780

0

365

00

101

0

4 (DU2SHNAc6S)

7 (DUHNAc6S)


6 (DU2SHNAc)


5 (DUHNS)

mA

Um

AU


































































(a)

100

10

1

(b)

100

10

1

15 h 6 h

(c)

100

100

10

10

1

(d )

100

10

1

1

9 h 24 h

0 h 0 h

0 h

100101

0 h




































05 h 15 h 3 h 6 h 9 h 12 h 24 h





10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01









6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9norm

aliz

ed in

tens

ity(l

og s

cale

)no

rmal

ized

inte

nsity

(log

sca

le)

norm

aliz

ed in

tens

ity(l

og s

cale

)








































(a)

(b) 25

20

15

10

30

20

10

0

RN

A tr

ansc

ript

abu

ndan

ce

perc

enta

ge o

f ac

tive

casp

ase

3po

sitiv

e ce

lls

5

0 10


20 30LUC


GAB

(d )

(c)

GABARAP

CDC45L

ASNS

RRM1

TCOF1CDC6

FAM46A

C9 or f81




(a)

(c)


no siRNA treatment

1 10 100

100

10

1

(b)



1 10 100































4754ndash4762 (doi101182blood-2006-01-0141)





1591ndash1598







(doi101023A1026546219962)



97ndash105 (doi101016S1044-579X(02)00127-X)







02053-X)









3876ndash3884 (doi101002pmic200401239)









35025220)




Cell 98 147ndash157 (doi101016S0092-8674(00)81010-7)



put 99 29ndash40





nm1307)






(doi101073pnas051630598)






science1069883)




























1172ndash1179






0003)





biolreprod661232)





10389467)





80 755ndash762 (doi101046j0022-3042200200760x)













1016S1357-2725(98)00058-2)







6386(03)00345-7)





(doi101182blood-2003-10-3614)




7 317ndash333 (doi101038sjmn7300119)




2925518929)






























2091-6-2)











blood-2002-03-0850)






blood-2004-02-0452)





172161899)




nature03928)





1074jbcM212837200)















(doi101016S0888-7543(03)00045-4)








126 1407ndash1415







35037734)



4 953ndash961




Res 98 1352ndash1364 (doi10116101RES0000225982

0198893)













1016jbbrc200605128)





(doi101111j1462-5822200400412x)




80 1ndash29








1002pmic200300773)













101 1223ndash1228 (doi101073pnas0308221100)




(SICI)1097-4652(199903)1783359AID-JCP10O3

0CO2-O)





(doi101042BJ20021955)





11861477-7827-1-17)
















































25 1124ndash1130





(doi101016S0378-1119(03)00820-5)









ajpheart005792003)







4665ndash4671





ECM components

ECMS

S

SSS

endothelial cell































et al 2004)















(a)

(b)

(c)

30

20

10

10 15min

10 15min

healthcells

(mole )


12

11

1

1

2

3

5

532

di-Smono-S

tri-S

8

7

76

10


1 (DU2SHNS6S) 458

102

101

0

148

25

75

92

173

284

780

0

365

00

101

0

4 (DU2SHNAc6S)

7 (DUHNAc6S)


6 (DU2SHNAc)


5 (DUHNS)

mA

Um

AU


































































(a)

100

10

1

(b)

100

10

1

15 h 6 h

(c)

100

100

10

10

1

(d )

100

10

1

1

9 h 24 h

0 h 0 h

0 h

100101

0 h




































05 h 15 h 3 h 6 h 9 h 12 h 24 h





10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01









6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9norm

aliz

ed in

tens

ity(l

og s

cale

)no

rmal

ized

inte

nsity

(log

sca

le)

norm

aliz

ed in

tens

ity(l

og s

cale

)








































(a)

(b) 25

20

15

10

30

20

10

0

RN

A tr

ansc

ript

abu

ndan

ce

perc

enta

ge o

f ac

tive

casp

ase

3po

sitiv

e ce

lls

5

0 10


20 30LUC


GAB

(d )

(c)

GABARAP

CDC45L

ASNS

RRM1

TCOF1CDC6

FAM46A

C9 or f81




(a)

(c)


no siRNA treatment

1 10 100

100

10

1

(b)



1 10 100































4754ndash4762 (doi101182blood-2006-01-0141)





1591ndash1598







(doi101023A1026546219962)



97ndash105 (doi101016S1044-579X(02)00127-X)







02053-X)









3876ndash3884 (doi101002pmic200401239)









35025220)




Cell 98 147ndash157 (doi101016S0092-8674(00)81010-7)



put 99 29ndash40





nm1307)






(doi101073pnas051630598)






science1069883)




























1172ndash1179






0003)





biolreprod661232)





10389467)





80 755ndash762 (doi101046j0022-3042200200760x)













1016S1357-2725(98)00058-2)







6386(03)00345-7)





(doi101182blood-2003-10-3614)




7 317ndash333 (doi101038sjmn7300119)




2925518929)






























2091-6-2)











blood-2002-03-0850)






blood-2004-02-0452)





172161899)




nature03928)





1074jbcM212837200)















(doi101016S0888-7543(03)00045-4)








126 1407ndash1415







35037734)



4 953ndash961




Res 98 1352ndash1364 (doi10116101RES0000225982

0198893)













1016jbbrc200605128)





(doi101111j1462-5822200400412x)




80 1ndash29








1002pmic200300773)













101 1223ndash1228 (doi101073pnas0308221100)




(SICI)1097-4652(199903)1783359AID-JCP10O3

0CO2-O)





(doi101042BJ20021955)





11861477-7827-1-17)
















































25 1124ndash1130





(doi101016S0378-1119(03)00820-5)









ajpheart005792003)







4665ndash4671

(a)

(b)

(c)

30

20

10

10 15min

10 15min

healthcells

(mole )


12

11

1

1

2

3

5

532

di-Smono-S

tri-S

8

7

76

10


1 (DU2SHNS6S) 458

102

101

0

148

25

75

92

173

284

780

0

365

00

101

0

4 (DU2SHNAc6S)

7 (DUHNAc6S)


6 (DU2SHNAc)


5 (DUHNS)

mA

Um

AU


































































(a)

100

10

1

(b)

100

10

1

15 h 6 h

(c)

100

100

10

10

1

(d )

100

10

1

1

9 h 24 h

0 h 0 h

0 h

100101

0 h




































05 h 15 h 3 h 6 h 9 h 12 h 24 h





10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01









6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9norm

aliz

ed in

tens

ity(l

og s

cale

)no

rmal

ized

inte

nsity

(log

sca

le)

norm

aliz

ed in

tens

ity(l

og s

cale

)








































(a)

(b) 25

20

15

10

30

20

10

0

RN

A tr

ansc

ript

abu

ndan

ce

perc

enta

ge o

f ac

tive

casp

ase

3po

sitiv

e ce

lls

5

0 10


20 30LUC


GAB

(d )

(c)

GABARAP

CDC45L

ASNS

RRM1

TCOF1CDC6

FAM46A

C9 or f81




(a)

(c)


no siRNA treatment

1 10 100

100

10

1

(b)



1 10 100































4754ndash4762 (doi101182blood-2006-01-0141)





1591ndash1598







(doi101023A1026546219962)



97ndash105 (doi101016S1044-579X(02)00127-X)







02053-X)









3876ndash3884 (doi101002pmic200401239)









35025220)




Cell 98 147ndash157 (doi101016S0092-8674(00)81010-7)



put 99 29ndash40





nm1307)






(doi101073pnas051630598)






science1069883)




























1172ndash1179






0003)





biolreprod661232)





10389467)





80 755ndash762 (doi101046j0022-3042200200760x)













1016S1357-2725(98)00058-2)







6386(03)00345-7)





(doi101182blood-2003-10-3614)




7 317ndash333 (doi101038sjmn7300119)




2925518929)






























2091-6-2)











blood-2002-03-0850)






blood-2004-02-0452)





172161899)




nature03928)





1074jbcM212837200)















(doi101016S0888-7543(03)00045-4)








126 1407ndash1415







35037734)



4 953ndash961




Res 98 1352ndash1364 (doi10116101RES0000225982

0198893)













1016jbbrc200605128)





(doi101111j1462-5822200400412x)




80 1ndash29








1002pmic200300773)













101 1223ndash1228 (doi101073pnas0308221100)




(SICI)1097-4652(199903)1783359AID-JCP10O3

0CO2-O)





(doi101042BJ20021955)





11861477-7827-1-17)
















































25 1124ndash1130





(doi101016S0378-1119(03)00820-5)









ajpheart005792003)







4665ndash4671


















































(a)

100

10

1

(b)

100

10

1

15 h 6 h

(c)

100

100

10

10

1

(d )

100

10

1

1

9 h 24 h

0 h 0 h

0 h

100101

0 h




































05 h 15 h 3 h 6 h 9 h 12 h 24 h





10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01









6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9norm

aliz

ed in

tens

ity(l

og s

cale

)no

rmal

ized

inte

nsity

(log

sca

le)

norm

aliz

ed in

tens

ity(l

og s

cale

)








































(a)

(b) 25

20

15

10

30

20

10

0

RN

A tr

ansc

ript

abu

ndan

ce

perc

enta

ge o

f ac

tive

casp

ase

3po

sitiv

e ce

lls

5

0 10


20 30LUC


GAB

(d )

(c)

GABARAP

CDC45L

ASNS

RRM1

TCOF1CDC6

FAM46A

C9 or f81




(a)

(c)


no siRNA treatment

1 10 100

100

10

1

(b)



1 10 100































4754ndash4762 (doi101182blood-2006-01-0141)





1591ndash1598







(doi101023A1026546219962)



97ndash105 (doi101016S1044-579X(02)00127-X)







02053-X)









3876ndash3884 (doi101002pmic200401239)









35025220)




Cell 98 147ndash157 (doi101016S0092-8674(00)81010-7)



put 99 29ndash40





nm1307)






(doi101073pnas051630598)






science1069883)




























1172ndash1179






0003)





biolreprod661232)





10389467)





80 755ndash762 (doi101046j0022-3042200200760x)













1016S1357-2725(98)00058-2)







6386(03)00345-7)





(doi101182blood-2003-10-3614)




7 317ndash333 (doi101038sjmn7300119)




2925518929)






























2091-6-2)











blood-2002-03-0850)






blood-2004-02-0452)





172161899)




nature03928)





1074jbcM212837200)















(doi101016S0888-7543(03)00045-4)








126 1407ndash1415







35037734)



4 953ndash961




Res 98 1352ndash1364 (doi10116101RES0000225982

0198893)













1016jbbrc200605128)





(doi101111j1462-5822200400412x)




80 1ndash29








1002pmic200300773)













101 1223ndash1228 (doi101073pnas0308221100)




(SICI)1097-4652(199903)1783359AID-JCP10O3

0CO2-O)





(doi101042BJ20021955)





11861477-7827-1-17)
















































25 1124ndash1130





(doi101016S0378-1119(03)00820-5)









ajpheart005792003)







4665ndash4671


05 h 15 h 3 h 6 h 9 h 12 h 24 h





10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01 2 3

6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9

10

01









6 12 24 3 9 24 24time time time

3 9

1

10

1

10

1

01 2 3

6 12 24 3 9 24 243 9 01 2 3

6 12 24 3 9 24 243 9norm

aliz

ed in

tens

ity(l

og s

cale

)no

rmal

ized

inte

nsity

(log

sca

le)

norm

aliz

ed in

tens

ity(l

og s

cale

)








































(a)

(b) 25

20

15

10

30

20

10

0

RN

A tr

ansc

ript

abu

ndan

ce

perc

enta

ge o

f ac

tive

casp

ase

3po

sitiv

e ce

lls

5

0 10


20 30LUC


GAB

(d )

(c)

GABARAP

CDC45L

ASNS

RRM1

TCOF1CDC6

FAM46A

C9 or f81




(a)

(c)


no siRNA treatment

1 10 100

100

10

1

(b)



1 10 100































4754ndash4762 (doi101182blood-2006-01-0141)





1591ndash1598







(doi101023A1026546219962)



97ndash105 (doi101016S1044-579X(02)00127-X)







02053-X)









3876ndash3884 (doi101002pmic200401239)









35025220)




Cell 98 147ndash157 (doi101016S0092-8674(00)81010-7)



put 99 29ndash40





nm1307)






(doi101073pnas051630598)






science1069883)




























1172ndash1179






0003)





biolreprod661232)





10389467)





80 755ndash762 (doi101046j0022-3042200200760x)













1016S1357-2725(98)00058-2)







6386(03)00345-7)





(doi101182blood-2003-10-3614)




7 317ndash333 (doi101038sjmn7300119)




2925518929)






























2091-6-2)











blood-2002-03-0850)






blood-2004-02-0452)





172161899)




nature03928)





1074jbcM212837200)















(doi101016S0888-7543(03)00045-4)








126 1407ndash1415







35037734)



4 953ndash961




Res 98 1352ndash1364 (doi10116101RES0000225982

0198893)













1016jbbrc200605128)





(doi101111j1462-5822200400412x)




80 1ndash29








1002pmic200300773)













101 1223ndash1228 (doi101073pnas0308221100)




(SICI)1097-4652(199903)1783359AID-JCP10O3

0CO2-O)





(doi101042BJ20021955)





11861477-7827-1-17)
















































25 1124ndash1130





(doi101016S0378-1119(03)00820-5)









ajpheart005792003)







4665ndash4671


















(a)

(b) 25

20

15

10

30

20

10

0

RN

A tr

ansc

ript

abu

ndan

ce

perc

enta

ge o

f ac

tive

casp

ase

3po

sitiv

e ce

lls

5

0 10


20 30LUC


GAB

(d )

(c)

GABARAP

CDC45L

ASNS

RRM1

TCOF1CDC6

FAM46A

C9 or f81




(a)

(c)


no siRNA treatment

1 10 100

100

10

1

(b)



1 10 100































4754ndash4762 (doi101182blood-2006-01-0141)





1591ndash1598







(doi101023A1026546219962)



97ndash105 (doi101016S1044-579X(02)00127-X)







02053-X)









3876ndash3884 (doi101002pmic200401239)









35025220)




Cell 98 147ndash157 (doi101016S0092-8674(00)81010-7)



put 99 29ndash40





nm1307)






(doi101073pnas051630598)






science1069883)




























1172ndash1179






0003)





biolreprod661232)





10389467)





80 755ndash762 (doi101046j0022-3042200200760x)













1016S1357-2725(98)00058-2)







6386(03)00345-7)





(doi101182blood-2003-10-3614)




7 317ndash333 (doi101038sjmn7300119)




2925518929)






























2091-6-2)











blood-2002-03-0850)






blood-2004-02-0452)





172161899)




nature03928)





1074jbcM212837200)















(doi101016S0888-7543(03)00045-4)








126 1407ndash1415







35037734)



4 953ndash961




Res 98 1352ndash1364 (doi10116101RES0000225982

0198893)













1016jbbrc200605128)





(doi101111j1462-5822200400412x)




80 1ndash29








1002pmic200300773)













101 1223ndash1228 (doi101073pnas0308221100)




(SICI)1097-4652(199903)1783359AID-JCP10O3

0CO2-O)





(doi101042BJ20021955)





11861477-7827-1-17)
















































25 1124ndash1130





(doi101016S0378-1119(03)00820-5)









ajpheart005792003)







4665ndash4671

(a)

(c)


no siRNA treatment

1 10 100

100

10

1

(b)



1 10 100































4754ndash4762 (doi101182blood-2006-01-0141)





1591ndash1598







(doi101023A1026546219962)



97ndash105 (doi101016S1044-579X(02)00127-X)







02053-X)









3876ndash3884 (doi101002pmic200401239)









35025220)




Cell 98 147ndash157 (doi101016S0092-8674(00)81010-7)



put 99 29ndash40





nm1307)






(doi101073pnas051630598)






science1069883)




























1172ndash1179






0003)





biolreprod661232)





10389467)





80 755ndash762 (doi101046j0022-3042200200760x)













1016S1357-2725(98)00058-2)







6386(03)00345-7)





(doi101182blood-2003-10-3614)




7 317ndash333 (doi101038sjmn7300119)




2925518929)






























2091-6-2)











blood-2002-03-0850)






blood-2004-02-0452)





172161899)




nature03928)





1074jbcM212837200)















(doi101016S0888-7543(03)00045-4)








126 1407ndash1415







35037734)



4 953ndash961




Res 98 1352ndash1364 (doi10116101RES0000225982

0198893)













1016jbbrc200605128)





(doi101111j1462-5822200400412x)




80 1ndash29








1002pmic200300773)













101 1223ndash1228 (doi101073pnas0308221100)




(SICI)1097-4652(199903)1783359AID-JCP10O3

0CO2-O)





(doi101042BJ20021955)





11861477-7827-1-17)
















































25 1124ndash1130





(doi101016S0378-1119(03)00820-5)









ajpheart005792003)







4665ndash4671
























4754ndash4762 (doi101182blood-2006-01-0141)





1591ndash1598







(doi101023A1026546219962)



97ndash105 (doi101016S1044-579X(02)00127-X)







02053-X)









3876ndash3884 (doi101002pmic200401239)









35025220)




Cell 98 147ndash157 (doi101016S0092-8674(00)81010-7)



put 99 29ndash40





nm1307)






(doi101073pnas051630598)






science1069883)




























1172ndash1179






0003)





biolreprod661232)





10389467)





80 755ndash762 (doi101046j0022-3042200200760x)













1016S1357-2725(98)00058-2)







6386(03)00345-7)





(doi101182blood-2003-10-3614)




7 317ndash333 (doi101038sjmn7300119)




2925518929)






























2091-6-2)











blood-2002-03-0850)






blood-2004-02-0452)





172161899)




nature03928)





1074jbcM212837200)















(doi101016S0888-7543(03)00045-4)








126 1407ndash1415







35037734)



4 953ndash961




Res 98 1352ndash1364 (doi10116101RES0000225982

0198893)













1016jbbrc200605128)





(doi101111j1462-5822200400412x)




80 1ndash29








1002pmic200300773)













101 1223ndash1228 (doi101073pnas0308221100)




(SICI)1097-4652(199903)1783359AID-JCP10O3

0CO2-O)





(doi101042BJ20021955)





11861477-7827-1-17)
















































25 1124ndash1130





(doi101016S0378-1119(03)00820-5)









ajpheart005792003)







4665ndash4671










1172ndash1179






0003)





biolreprod661232)





10389467)





80 755ndash762 (doi101046j0022-3042200200760x)













1016S1357-2725(98)00058-2)







6386(03)00345-7)





(doi101182blood-2003-10-3614)




7 317ndash333 (doi101038sjmn7300119)




2925518929)






























2091-6-2)











blood-2002-03-0850)






blood-2004-02-0452)





172161899)




nature03928)





1074jbcM212837200)















(doi101016S0888-7543(03)00045-4)








126 1407ndash1415







35037734)



4 953ndash961




Res 98 1352ndash1364 (doi10116101RES0000225982

0198893)













1016jbbrc200605128)





(doi101111j1462-5822200400412x)




80 1ndash29








1002pmic200300773)













101 1223ndash1228 (doi101073pnas0308221100)




(SICI)1097-4652(199903)1783359AID-JCP10O3

0CO2-O)





(doi101042BJ20021955)





11861477-7827-1-17)
















































25 1124ndash1130





(doi101016S0378-1119(03)00820-5)









ajpheart005792003)







4665ndash4671






2091-6-2)











blood-2002-03-0850)






blood-2004-02-0452)





172161899)




nature03928)





1074jbcM212837200)















(doi101016S0888-7543(03)00045-4)








126 1407ndash1415







35037734)



4 953ndash961




Res 98 1352ndash1364 (doi10116101RES0000225982

0198893)













1016jbbrc200605128)





(doi101111j1462-5822200400412x)




80 1ndash29








1002pmic200300773)













101 1223ndash1228 (doi101073pnas0308221100)




(SICI)1097-4652(199903)1783359AID-JCP10O3

0CO2-O)





(doi101042BJ20021955)





11861477-7827-1-17)
















































25 1124ndash1130





(doi101016S0378-1119(03)00820-5)









ajpheart005792003)







4665ndash4671











































25 1124ndash1130





(doi101016S0378-1119(03)00820-5)









ajpheart005792003)







4665ndash4671

Understanding endothelial cell apoptosis: what can the transcriptome, glycome and proteome reveal

Documents

Transcript of Understanding endothelial cell apoptosis: what can the transcriptome, glycome and proteome reveal