of March 27, 2018.This information is current as

of Caspase-3 ActivationSynthesis of Bcl-2 and Mediated by InhibitionSublytic C5b-9 Is Associated with Enhanced Inhibition of Oligodendrocyte Apoptosis by

ShinLucian Soane, Horea Rus, Florin Niculescu and Moon L.

http://www.jimmunol.org/content/163/11/61321999; 163:6132-6138; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/163/11/6132.full#ref-list-1

, 33 of which you can access for free at: cites 65 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 1999 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on March 27, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on March 27, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Inhibition of Oligodendrocyte Apoptosis by Sublytic C5b-9 IsAssociated with Enhanced Synthesis of Bcl-2 and Mediated byInhibition of Caspase-3 Activation1

Lucian Soane, Horea Rus, Florin Niculescu, and Moon L. Shin2

We have previously shown that generation of sublytic C5b-9, the membrane attack complex of complement, induces oligoden-drocytes to enter cell cycle and reduces apoptotic cell death in vitro. In the present study, the cellular factors involved in apoptosisof oligodendrocyte progenitor cells and oligodendrocytes, and the inhibitory effect of C5b-9 on apoptotic process were investigated.Oligodendrocyte progenitor cells identified by mAb A2B5 that were isolated from neonatal rat brains were differentiated intooligodendrocytes in serum-free defined medium. The differentiation, which occurs simultaneously with apoptotic cell death, wasassociated with a rapid loss ofbcl-2 mRNA and increased expression of caspase-3 mRNA. Activation of caspase-3 in differentiatingcells was demonstrated by the generation of 17- and 12-kDa fragments of caspase-3 proenzyme and by cleavage of poly(ADP-ribose) polymerase, a specific caspase-3 substrate. Cell death associated with differentiation was inhibited by the caspase-3 in-hibitor DEVD-CHO in a dose-dependent manner. Assembly of sublytic C5b-9 resulted in inhibition of caspase-3 activation. Inaddition, synthesis of BCL-2 protein in oligodendrocytes was significantly increased by C5b-9. The TNF-a-induced apoptosis ofoligodendrocytes was also inhibited by C5b-9. These results indicate that up-regulation of BCL-2 protein and inhibition ofcaspase-3 activation are potential mechanisms by which C5b-9 increases survival of oligodendrocyte in vitro and possibly in vivoduring inflammation and immune-mediated demyelination affecting the CNS. The Journal of Immunology,1999, 163: 6132–6138.

D emyelination in multiple sclerosis (MS)3 and its animalmodel experimental allergic encephalitis (EAE) iscaused by damage to myelin and myelin-producing ol-

igodendrocyte (OLG) by activated immune effectors. These effec-tors include macrophages, T cells, proinflammatory cytokinesTNF-a and monocyte chemoattractant protein-1, and C5b-9 com-plexes generated during complement activation (1–3). Sequentialinteraction of C5b6, C7, C8, and C9 is associated with amphipathicconformational changes of C7, C8a, C8b, and C9, resulting inassembly of membrane-inserted C5b-7, C5b-8, and C5b-9 com-plexes, collectively referred to as the terminal complement com-plexes (TCC) (4). Sublytic C5b-9 stimulates target cells and in-duces a variety of cellular activities in the absence of cell death (4,5). One of the activities induced by C5b-9 is cell cycle induction(6–9), which is mediated by Gi-dependent activation of Ras,Raf-1, and ERK1, and associated with expression of protoonco-genes c-fosand c-jun, and increased DNA synthesis (8–10). Cell

cycle activation by C5b-9 also occurs through release of fibroblastgrowth factor (FGF) and platelet-derived growth factor (PDGF) (6,7). Increase in cytosolic Ca21 and protein kinase C activation areresponsible for some of the TCC activities, such as platelet acti-vation and generation of arachidonic acid and metabolites, and areelicited by the pore-forming C5b-8 and C5b-9 complexes (4, 5,11–14). Additionally, membrane-inserted TCC, including C5b-7,are able to generate diacylglycerol and stimulate the Ras/Raf-1/ERK1 pathway via Gbg effectors of the G protein (10, 15, 16).However, C5b-9 is most effective in inducing DNA synthesis andcell cycle, in a Gi-ERK1-dependent manner (9, 10, 15, 16).

OLG that myelinate the central nerve axons differentiate fromthe O-2A progenitors, and this process requires axonal contact andsoluble growth factors (17–20). Survival of differentiated OLGalso requires factors such as PDGF and basic FGF (17–19). Indeveloping rat optic nerve, more than 50% of newly differentiatedOLG undergo apoptotic death, which is an essential process forbrain tissue modeling during development (17). In serum-free me-dium, O-2A cells differentiate into OLG concomitantly with apo-ptosis, as in vivo. Apoptosis in vitro is also inhibited by PDGF,insulin-like growth factor, ciliary neurotrophic factor, and leuke-mia-inhibitory factor (18–22).

A critical role of complement in EAE is supported by experi-ments in which abrogation of systemic complement activity bycobra venom factor or by soluble CR1 inhibited demyelinationinduced by encephalitogenic Ag or Ag-specific T cells (23, 24).Deposition of C5b-9 in MS and EAE brains and increased levels ofsoluble C5b-9 in MS spinal fluids indicated in situ activation andassembly of C5b-9 (25–27). The central nerve myelin, but not theperipheral nerve myelin, directly activates the classical pathway ofcomplement (28, 29). Activation and assembly of C5b-9 on myelincause hydrolysis of myelin basic protein (MBP) and extensivevesiculation with eventual loss of myelin membrane (30, 31). Inaddition, the complement-inhibitory proteins CD55 and CD46 areabsent in myelin, causing the myelin membrane to be susceptible

Department of Pathology, University of Maryland, School of Medicine, Baltimore,MD 21201

Received for publication June 15, 1999. Accepted for publication September 8, 1999.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby markedadvertisementin accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by National Institutes of Health Grants NS36231 andNS15662.2 Address correspondence and reprint requests to Dr. Moon L. Shin, University ofMaryland School of Medicine, Department of Pathology, 10 South Pine Street, MSTF600-E, Baltimore, MD 21201. E-mail address: mshin@umaryland.edu3 Abbreviations used in this paper: MS, multiple sclerosis; C7D, normal human serumimmunochemically depleted of C7; DEVD-CHO, Asp-Glu-Val-Asp-Cho; EAE, ex-perimental allergic encephalomyelitis; ERK, extracellular signal-related kinase; FGF,fibroblast growth factor; GC, galactocerebroside; MBP, myelin basic protein; MTS,methyl tetrazolium salt; NHS, normal human serum; O-2A, OLG progenitor cellsidentified by mAb A2B5; OLG, oligodendrocyte; PARP, poly(ADP-ribose) polymer-ase; PDGF, platelet-derived growth factor; PI-3, phosphatidylinositol-3; PLP, prote-olipid protein; RT, room temperature; TCC, terminal complement complexes repre-senting C5b-7, C5b-8, and C5b-9.

Copyright © 1999 by The American Association of Immunologists 0022-1767/99/$02.00

by guest on March 27, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

to C5b-9 (32). Therefore, C5b-9 can contribute to demyelinationby directly damaging the myelin, even in the absence of myelin-specific Abs. In OLG, C5b-9 at a sublytic concentration inducescell cycle, as shown by activation of ERK1 and c-Jun N-terminalkinase 1, protooncogenes, and G1 progression to S phase (8, 33).Sublytic C5b-9 also induces phenotype changes in OLG by accel-erating the decay of mRNA encoding myelin-specific genes (8,34). While activating cell cycle, C5b-9 was also found to inhibitapoptosis of OLG associated with differentiation (8).

In this study, we have examined the differentiation-associatedapoptosis of OLG in vitro by investigating involvements ofcaspase-3 and Bcl-2 as possible target sites regulated by C5b-9.We also tested the ability of C5b-9 to protect OLG from TNF-a-induced apoptosis.

Materials and MethodsDifferentiation of OLG from O-2A progenitor cells in culture

Primary O-2A progenitor cells were prepared according to Saneto and deVellis (35). Glial cells were isolated from neonatal Sprague Dawley ratbrains, as described in detail (35, 34). Dispersed glia cells are grown for 10days as stratified mixed glial cultures. O-2A progenitors growing on sur-face of the mixed culture were isolated by a series of differential shaking.Cells were placed in OLG defined medium consisting of serum-freeDMEM/Ham’s F-12 containing 500 ng/ml transferrin (Sigma, St. Louis,MO), 75 ng/ml insulin (Sigma), 75mg/ml basic FGF (Collaborative Re-search, Lexington, MA), and 1 mM sodium pyruvate. O-2A cells isolatedby serial shaking at the time of plating showed 2–3% cell death, as deter-mined by trypan blue dye exclusion. Differentiation was stepwise, asshown by the expression of MBP and proteolipid protein (PLP) mRNAbefore the expression of galactocerebroside (GC) (17, 35). After 56 h inOLG defined medium, more than 85% of cells expressed GC, MBP, andPLP. Less than 5% of the MBP-negative cells were astrocytes and micro-glia, and the remaining cells were O-2A cells in different stages of differ-entiation. O-2A cells grown in a defined medium for 3 days are designatedas OLG.

Determination of cell viability

Viability of O-2A cells during differentiation and the effect of C5b-9 on cellviability were determined by using CellTiter 96 Aqueous cell proliferationassay, according to the instruction supplied by Promega (Madison, WI).Cells were seeded on poly(D-lysine)-coated 96-well plates at 105 cells/wellin 200 ml of OLG defined medium and cultured at 37°C. At the indicatedtime points, 40ml methyl tetrazolium salt (MTS) solution was added toeach well. Plates were kept at 37°C for additional 2 h, followed by deter-mination of OD at 540 nm under a condition in which absorbance was inlinear range. The results are expressed as percentage of dead cells6 SD,relative to the initial cell number.

Analysis of apoptosis

DNA strand break was detected in cells by TdT-dependent incorporation ofdUTP (Apoptag, Oncor, Gaithersburg, MD). O-2A cells were cultured onplastic slide chambers for the indicated time period. Cells were fixed inbuffered Formalin at room temperature (RT), then treated with TdT in thepresence of digoxigenin-dUTP for 1 h at37°C. After washing, cells weretreated with peroxidase-conjugated anti-digoxigenin IgG F(ab9)2 fragmentsfor 1 h; then color was developed using diaminobenzidine as a substrate.Approximately 600 cells with clearly defined nucleus were examined ineach sample by TUNEL staining. The number of cells showing apoptosiswas counted by identifying TUNEL-positive nuclei. The percentage ofapoptotic cells was then calculated using the following formula: (numberof cells with TUNEL-positive nuclei/total number of cells examined)3100. Results are expressed as mean percentage of cells with TUNEL-pos-itive nuclei 6 SD.

Activation of serum complement and C5b-9 assembly

Normal human serum (NHS) pooled from several healthy donors was usedas a source of serum complement. Rabbit antiserum to GC was used tosensitize rat OLG. The specific anti-GC activity was assayed by treatingGC-expressing liposomes with trapped86Rb aqueous marker with anti-serum, then measuring the released marker (31). Because anti-GC Abs aremostly IgM isotype, IgM fraction of the antisera was used in most exper-iments. A sublytic dose of Ab was predetermined by titrating anti-GC Ab

using an excess of NHS (8, 34). To evaluate the effect of serum C5b-9,OLG sensitized with a dose of anti-GC Ab for 30 min at RT were incubatedwith a 1/20 dilution of NHS depleted of C7 (C7D) reconstituted with C7(10 mg/ml). Alternatively, sensitized cells were treated with NHS (1/10)and NHS treated with K76 (Otsuka Pharmaceutical, New York, NY)(NHS-K76) as a control (8, 34). K76 prevents C5b-9 assembly in serum bybinding to C5 (36). Therefore, C7D and NHS-K76 allow complement ac-tivation to proceed up to C6 and C3, respectively. Purified human com-plement proteins C5-C9 were purchased from Quidel (San Diego, CA), andC5b6 complex was prepared from C5 and C6, as described (37). To as-semble sublytic C5b-9 by using purified proteins, cells were incubated withC5b6 (30mg) for 15 min, then with C7 (10mg) for 5 min at RT, followedby addition of C8 (10mg) and C9 (10mg) in a final volume of 1 ml (8, 10).Cells were then incubated at 37°C for the indicated time periods.

Northern blot analysis

RNA was isolated from cells lysed with buffer containing guanidine iso-thiocyanate and 2-ME, and total RNA was purified by ultracentrifugationon 5.7 M CsCl (as described in Ref. 38). Poly(A)1 RNA was prepared fromtotal RNA using Dynabead mRNA purification system (Dynal, Great Neck,NY). Poly(A)1 RNA was denatured and electrophoresed on 0.8% agarose-formaldehyde gels, then transferred to a nitrocellulose membrane. Afterbaking for 2 h at80°C, the membrane was hybridized with32P-labeledcDNA probes. The probe binding was quantitated by measuring band den-sities of autoradiogram using Computing Densitometer (Molecular Dy-namics, Sunnyvalle, CA). Integrated volume of each band was calculatedusing the ImageQuant software (Molecular Dynamics), and the results areexpressed by density ratio to actin. Caspase-3 cDNA probe was obtainedby RT-PCR cloning of rat caspase-3 cDNA with the forward (59-GCGAAGCTTAAGTGACCATGGACAACCAAC) and reverse (59-GCGTCTAGACCCAGTCATTCCTTTAGTGA) primers designed according to ratCPP32 cDNA (39). The ratbcl-2 andbaxcDNA were gifts from Dr. E. Podack(University of Miami) and Dr. S. Korsmeyer (Washington University,St. Louis, MO), respectively. The cDNA was labeled with [a-32P]dCTP(New England Nuclear, Boston, MA) using reagents for DNA labeling fromPharmacia (Piscataway, NJ).

Western blot analysis of caspase-3, PARP protein, and BCL-2

The levels of caspase-3, PARP, and their cleavage products were deter-mined by Western and immunoblot. Cells were lysed with RIPA buffer (30mM Tris-HCl, pH 7.4, 0.15 M NaCl (NaCl), 1% Nonidet P-40, 0.1% SDS,0.5% sodium deoxycholate, 1 mM EDTA, 1 mM DTT, 2 mM MgCl2, 1mM NaVO4, 0.5 mM PMSF, 100mg/ml aprotinin, and leupeptin), as de-scribed (16). An equal amount of protein from each cell lysate was useddirectly for SDS-PAGE and Western blot, a method sufficient to detectcaspase-3 proenzyme and high levels of the cleavage fragment. To detectthe cleavage fragments, cell lysates (100mg protein) were immunoprecipi-tated with rabbit anti-caspase-3 IgG (Santa Cruz Biotechnology, SantaCruz, CA) in the presence of protein A/G agarose at 4°C overnight. Celllysates or immunoprecipitates were analyzed on 10% SDS-PAGE, then byWestern blotting using the same rabbit anti-caspase-3 IgG. For PARP,immunoprecipitates using polyclonal anti-PARP IgG (Boehringer Mann-heim, Indianapolis, IN) were analyzed by 7% SDS-PAGE, and monoclonalanti-PARP IgG1 (Zymed, San Francisco, CA) was used for immunoblot-ting. This was followed by reaction with peroxidase-conjugated goat anti-rabbit or anti-mouse IgG (Santa Cruz Biotechnology), then by enhancedchemiluminescence (ECL; Pierce, Rockford, IL). BCL-2 protein was de-termined similarly by immunoprecipitation of cell lysates. The BCL-2Western blot reagents were from Oncogene (Cambridge, MA).

Effects of caspase-3 inhibitor on OLG viability

To test whether caspase-3 activity is required for differentiation-inducedapoptosis, the cell-permeable caspase-3 inhibitor DEVD-CHO (Calbio-chem, San Diego, CA) was used. O-2A cells were seeded in 96-well platesat 105 cells/well in 200ml of OLG defined medium and cultured for 24 h.Cells were further incubated for 48 h in the presence of 10–100mM ofDEVD-CHO. Cell viability was then determined, as described earlier.

Effect of C5b-9 on OLG apoptosis induced by TNF-a.

To test whether sublytic C5b-9 also protects OLG from apoptotic cell deathinduced by TNF-a (21, 22, 40), O-2A cells were differentiated in 96-wellplates, then cells were exposed to sublytic NHS or NHS-K76 for 1 h. Afteraddition of 100 ng/ml of human rTNF-a (R&D Systems, Minneapolis,MN), cells were incubated for 18 h at 37°C, and viability was determined.

6133The Journal of Immunology

by guest on March 27, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

ResultsApoptotic cell death of O-2A progenitor cells during in vitrodifferentiation

Differentiation of OLG is associated with cell death in developingbrains and during in vitro differentiation (17–19). As shown in Fig.1A, cell death reached 36% at 48 h, and was increased further to70% at 96 h. Many of these cells showed the characteristic featuresof apoptosis, including cell process retraction, chromatin conden-sation, and DNA cleavage with ladder formation (data not shown).By TUNEL stain, 31.36 6.5% of cells were apoptotic after 48 hin OLG defined medium (Fig. 1B).

Bcl-2, Bax, and caspase-3 expression during OLG differentiation

The expression of caspase-3,bcl-2, andbax during OLG differ-entiation was examined. Northern blot analysis of poly(A)1 RNAshowed a gradual increase in caspase-3 mRNA expression and arapid decline ofbcl-2 mRNA as early as 2 h (Fig. 2A–C). Thechanges inbcl-2 and caspase-3 mRNA level are associated withdifferentiation, as shown by the robust expression of PLP mRNA(Fig. 2A). ThebaxmRNA expression was reduced by 30% at 6 h,as shown by density ratios to actin, obtained by quantitative den-sitometry (Fig. 2C). Thebax mRNA level after 3 days in differ-entiation medium, determined in a separate experiment, was sim-ilar to the initial level (data not shown). Western blot analysis ofcell lysates showed an increase in 32-kDa caspase-3 proenzymeafter 24 h (1 day) (Fig. 3,A and B), which correlated with in-creased expression of caspase-3 mRNA. The caspase-3 proenzymebegan to decrease on day 3, as shown by the density ratio tob-ac-tin on the same blot. Because caspase-3 activity requires the proen-zyme cleavage to active subunits (41), the appearance of 17- and12-kDa subunits was evaluated by anti-caspase-3 immunoprecipi-tation, followed by Western blot. The appearance of anti-caspase-3-reactive 17- and 12-kDa bands was identified on day 3 (Fig. 3C).Decreased caspase-3 proenzyme with increased cleavage productson day 3 (Fig. 3C) is clearly evident in experiments when celllysates were immunoprecipitated, then examined by Westernblotting.

To determine whether caspase-3 activation is required for celldeath, O-2A cells in medium for 24 h were cultured for additional48 h with the cell-permeable caspase-3 inhibitor DEVD-CHO. Celldeath was inhibited in a dose-dependent manner, with 50% pro-tection at 25mM and 100% at 100mM of DEVD-CHO (Fig. 4).DMSO at concentrations used to resuspend DEVD-CHO was nottoxic to the cell.

Inhibition of caspase-3 activation by C5b-9

To evaluate the antiapoptotic activity of C5b-9 previously shown(8), OLG exposed to serum C5b-9 were examined for caspase-3cleavage. As shown in Fig. 5A, a prominent 17-kDa cleavage prod-uct was seen in control cells treated with C7D for 18 h, which wasinhibited by addition of C7 to C7D. Because the data were ob-tained by direct analysis of the cell lysates by SDS-PAGE/Westernblotting, presence of the cleavage fragment in unstimulated cellswas not detected. When cell lysates were immunoprecipitated first,then examined by Western/immunoblot (Fig. 5B), the caspase-3cleavage fragment increased with time in cells exposed toNHS-K76. This increase was inhibited when C5b-9 assembly wasallowed in NHS, in contrast to NHS-K76. To exclude a possibilitythat C5b-9 may have enhanced the serum effect on caspase-3, iden-tical experiments were performed by treating cells with C5b-9 as-sembled using purified proteins (Fig. 5C). A cleavage fragment ofcaspase-3 was detected in unstimulated OLG. This 17-kDa bandincreased with time in control cells exposed to C5b6, C8, and C9without C7. However, the cleavage product was barely detected incells exposed to C5b-9. We have also examined the effect of C5b-9on cleavage of PARP, a specific substrate for caspase-3 (41, 42).

FIGURE 1. O-2A progenitor cells undergo cell death by apoptosis inserum-free defined medium.A, O-2A cells were placed in 96-well plates at105 cells/well in 200ml of OLG defined medium to allow differentiation.Cell viability was assessed by a method using MTS, as described inMa-terials and Methods. Background cell death at the time of plating was2–3%. The percentage of cell death was expressed relative to the initialviable cell number, and the results are shown as mean6 SD of threeexperiments performed in triplicate.B, O-2A cells were differentiated inchamber slides as above. Apoptosis was determined by TUNEL method, asdescribed inMaterials and Methods. Approximately 600 cells with clearlydefined nucleus were examined in each sample, and apoptotic cells wereidentified by the presence of TUNEL-positive nuclei. Results are expressedas mean percentage of cells with TUNEL-positive nuclei relative to thetotal number of cells examined6 SD. Data were derived from two separateexperiments performed in triplicate.

FIGURE 2. Expression of caspase-3 andbcl-2 andbax mRNA duringOLG differentiation. O-2A cells cultured in OLG defined medium for theindicated time periods were examined by Northern blot for mRNA encod-ing caspase-3 and PLP (A), and forbcl-2 andbax(B) using 1mg poly(A)1

RNA/lane. The data shown represent one of two identical experiments. Theresults are also expressed by density ratios tob-actin (C).

6134 ANTI-APOPTOTIC ACTIVITY OF C5b-9

by guest on March 27, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

On Western blotting, the 89-kDa fragment of PARP protein wasdetected in unstimulated OLG. The PARP cleavage was signifi-cantly reduced in cells treated with serum C5b-9, compared withthe level of NHS-K76 (Fig. 6).

Expression of BCL-2 in OLG exposed to C5b-9

In view of the ability of C5b-9 to inhibit caspase-3 activation, thesteps upstream to caspase-3 activation that may be affected byC5b-9 were explored. Caspase-3 can be activated by caspase-9through mitochondrial pathway or by caspase-8 in death receptor-dependent pathway (43). We have analyzed the effect of C5b-9 onBcl-2, a potent antiapoptotic factor, which inhibits caspase-3 ac-tivation by regulation of mitochondrial pathway (44). The expres-sion of bcl-2 mRNA was not affected by C5b-9, as shown in Fig.7B. However, C5b-9 significantly increased the level of BCL-2protein within 4 h, and to the maximum level at 8 h (Fig. 7A).BCL-2 protein was not detected in unstimulated OLG and in OLGtreated with control C5b6.

FIGURE 3. Caspase-3 proenzyme and its activation during OLG dif-ferentiation.A, Cell lysates (50mg protein) of O-2A cells cultured for 1, 3,and 4 days were examined for caspase-3 proenzyme by 10% SDS-PAGEand Western blot. The blot was immunostained with anti-caspase-3 IgG,then with anti-b-actin, as loading control.B, Densitometric analysis of theradiographic bands is shown as density ratios tob-actin. Representativedata from three experiments are shown.C, Caspase-3 activation duringOLG differentiation was determined by detection of the 17- and 12-kDacaspase-3 fragments. O-2A cells in culture for 1, 2, and 3 days were lysed,and the lysates, 100mg protein, were immunoprecipitated with rabbit anti-caspase-3 IgG. The immunoprecipitates were then analyzed for the cleav-age products, as inA.

FIGURE 4. Inhibition of caspase-3 activity prevented cell death asso-ciated with differentiation. O-2A cells were cultured for 24 h in 96-wellplates at 105 cells/well in OLG defined medium. Cells were incubated foradditional 48 h with 10–100mM of DEVD-CHO. Cell viability was de-termined by a method using MTS, as described inMaterials and Methods,and the percentage of cell death was assessed at 24 h in culture and also at72 h in which cells were incubated with or without the inhibitor for the last48 h. Results are expressed as mean percentage of cell death6 SD of threeexperiments performed in triplicate.

FIGURE 5. Inhibition of caspase-3 activation in OLG by C5b-9.A, Dif-ferentiated OLG, 53 106 cells/flask, were sensitized with anti-GC Ab,then exposed to C7D6 C7, as described inMaterials and Methods. At theindicated time points, cell lysates were examined by 10% SDS-PAGE andWestern blot.B, Cells were exposed to TCC as above, except that theywere treated with Ab and NHS or NHS-K76 (K76). Cell lysates, 100mgprotein, were immunoprecipitated with rabbit anti-caspase-3 IgG in thepresence of a mixture of protein A- and G-coated agarose. The Ag-Abcomplexes were eluted from the beads; then the eluates were examined by10% SDS-PAGE and Western blotting. Generation of a 17-kDa fragmentof caspase-3 was detected as inA using the same anti-caspase-3 IgG. Den-sitometric scan of the band is also shown.C, Similar experiment as inBwas conducted by treating cells with purified proteins to assemble C5b-9.Control cells were treated with C5b6, C8, and C9 without C7.

6135The Journal of Immunology

by guest on March 27, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Protection of TNF-a-induced cell death by C5b-9

We have examined whether C5b-9 also protects OLG from apo-ptosis induced by other factors. TNF-a was tested, since TNF-ainduces apoptotic cell death in OLG both in vivo and in vitro (21,22, 45). In our system, 100 ng/ml of TNF-a induced 50% celldeath after 18 h (Fig. 8A). Pretreatment with NHS, but not withNHS-K76, protected OLG from cell death (Fig. 8A). The cleavageproduct of caspase-3 proenzyme, which was increased by TNF-a,was abolished in OLG treated with NHS (Fig. 8B).

DiscussionThe C5b-9 complex is a pleiotropic effector generated during in-flammation and immune response. When inserted into the targetcell membrane, C5b-9, depending on doses, causes cell death orcell activation (4, 5). We have previously shown that at a sublyticconcentration, C5b-9 enhances OLG survival in vitro through in-hibition of apoptosis (8). C5a and C5b-9 have been implicated incell injury associated with apoptosis seen in a rat ischemia/reper-fusion model (46). However, regulation of apoptosis by sublyticC5b-9 has not been reported. In addition, C5a receptor is not ex-pressed by OLG (47). Therefore, the antiapoptotic activity ofC5b-9 may be a new biological function with a potentialsignificance.

Although apoptosis is induced by a variety of stimuli, executionof the apoptotic program involves a common mechanism, whichrelies on the activation of caspases, cysteine proteases belonging tothe IL-1-converting enzyme/CED-3 family. The role of individualcaspases and their relative importance in apoptosis have been re-cently clarified (43). Caspase-8 and caspase-10 are activated earlyin apoptotic process and are considered initiators, while caspase-3and caspase-7 activated at a later phase of apoptosis are effectorsacting on a large number of substrates. Death receptor-inducedpathways of apoptosis require activation of the caspase-8 andcaspase-3, whereas apoptosis following growth factor deprivationand stress-induced cell injury appears to be through mitochondrialdysfunction by releasing cytochromec, which triggers activationof caspase-9, then caspase-3 (43–45, 48–51). In addition, apopto-sis may also be mediated by a poorly understood caspase-indepen-dent pathway (52). Disruption of caspase-8 gene produces fetaldeath without anomalies of the nervous system. In contrast, dis-ruption of caspase-3 or caspase-9 genes results in abnormal neuraldevelopment, in addition to fetal death (48–50). Despite the find-ing that caspase-1 and caspase-3 are expressed in OLG and in-volved in TNF-a-induced apoptosis (53), the specific caspases re-sponsible for differentiation-induced OLG apoptosis have not beenclearly defined.

In this study, we have investigated caspase-3 and Bcl-2 in OLGapoptosis and the effect of sublytic C5b-9 in this process. In viewof the potent antiapoptotic activity of BCL-2 (54) and the key roleof caspase-3 as an apoptosis effector (43, 48), rapid loss ofbcl-2mRNA expression concomitant with increasing caspase-3 mRNAand protein at the onset of cell differentiation suggested a role forthese two proteins in differentiation-induced apoptosis. Proteolyticactivation of caspase-3, as indicated by the generation of 17- and12-kDa subunits and the 89-kDa cleavage fragment of PARP, wasdetected during OLG differentiation. Furthermore, caspase-3 in-hibitor DEVD-CHO effectively protected OLG from cell death.Together, these findings indicated that caspase-3 activation is es-sential for differentiation-induced apoptosis. Caspase-3 activation

FIGURE 6. Inhibition of PARP cleavage by C5b-9. OLG treated withAb and NHS or NHS-K76 for 3 and 6 h were lysed. Cell lysates, 100mgprotein, were immunoprecipitated with anti-PARP IgG, as in caspase-3;then the precipitates were analyzed by 7.5% SDS-PAGE and Western blot.A representative data of three experiments is shown.

FIGURE 7. Effects of C5b-9 on expression ofbcl-2 mRNA and proteinin OLG. A, O-2A cells differentiated in vitro for 3 days (OLG) were ex-posed to C5b-9 or C5b6 assembled with purified components. After 4, 8,and 18 h, cells were lysed. Cell lysates were immunoprecipitated withanti-BCL-2 IgG, and the protein was analyzed by 10% SDS-PAGE andWestern blot, using anti-BCL-2 IgG (Oncogene, Cambridge, MA). PurifiedBCL-2 protein (Pierce, Rockford, IL) was used as a positive control.B,OLG were exposed to C5b-9 assembled with purified components as inA.Northern blot was performed at 3 and 6 h, using 1mg of poly(A)1 RNA/lane, forbcl-2 andb-actin mRNA expression.

FIGURE 8. Effect of C5b-9 on cell death induced by TNF-a. A, Anti-GC-sensitized OLG were incubated with NHS or NHS-K76 for 1 h at37°C. Then 100 ng/ml of human rTNF-a was added, and cells were incu-bated for 18 h before determining the cell viability.B, An identical exper-iment as inA was performed, except that cell lysates were prepared and theprotein (100mg) was immunoprecipitated with anti-caspase-3 IgG, thenexamined for the 17-kDa cleavage fragment.

6136 ANTI-APOPTOTIC ACTIVITY OF C5b-9

by guest on March 27, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

was abrogated by C5b-9, as shown by inhibition of caspase-3proenzyme cleavage into its active subunits. This finding was con-sistent with the inhibition of PARP cleavage, a substrate for acti-vated caspase-3. Regulation of BCL-2 expression was examined asa possible step upstream to the caspase-3 affected by C5b-9. InOLG, bcl-2 mRNA was expressed at a very low level withoutdetectable protein, as examined by sensitive methods, such as theuse of poly(A)1 RNA for Northern blot and analysis of cell lysatesby immunoprecipitation and Western immunoblot. Interestingly,C5b-9 was able to increase BCL-2 protein without significantlyaffecting the mRNA level, suggesting a possible role of C5b-9 inposttranscriptional regulation of Bcl-2. Detection ofbcl-2 mRNAin the absence of BCL-2 protein in germinal center B cells (55) andin a trophoblastic tumor cell line when induced to differentiate (56)also suggested a step of translational regulation of Bcl-2. A spe-cific cis element within the promoter has been identified as a reg-ulatory site involved in the translational control ofbcl-2 gene (57).How BCL-2 synthesis is regulated by C5b-9 remains unclear. Wehave shown that sublytic C5b-9 induces ERK1 pathway, and thisis through activation of phosphatidylinositol-3 (PI-3) kinase (9, 10,33). In OLG, ERK1 activated by C5b-9 are responsible for en-hanced DNA synthesis (33), and C5b-9 increased the p70 S6 ki-nase activity (33), a ribosomal kinase responsible for protein syn-thesis (58). PI-3 kinase has been shown to inhibit apoptosis, andthis is thought to be through activation of Akt kinase and by in-creasing BCL-2 (59, 60). In postmitotic cells such as OLG, C5b-9,instead of inducing proliferation, may enhance cell survival. Theputative antiapoptotic signaling generated by C5b-9 may includePI-3 kinase. C5b-9, by increasing BCL-2, may stabilize mitochon-drial inner membrane permeability, or inhibit the interaction ofBAX with outer membrane proteins (61). BCL-2 prevents cyto-chromec release and inhibits activation of caspase-9 and caspase-3(54). Therefore, up-regulation of BCL-2 protein by C5b-9 in OLGmay precede the inhibition of caspase-3 activation. C5b-9 alsoinhibited cell death and caspase-3 activation induced by TNF-a.TNF-a induces apoptosis via caspase-8 through the recruitment ofTRADD/FADD (TNFR-associated death domain/Fas-associateddeath domain) proteins to the TNFR1 (62). However, TNF-a alsogenerates ceramide, which induces caspase-9 activation and apo-ptosis, in a caspase-8-independent manner (49, 63). We can spec-ulate that C5b-9 inhibits OLG apoptosis induced by differentiationand by TNF-a, and this is mediated through up-regulation ofBCL-2 protein and inhibition of caspase-9 and caspase-3.

Our finding that C5b-9 rescues OLG from differentiation-in-duced apoptosis and apoptosis caused by TNF-a may have a bi-ological significance in inflammatory and immune-mediated de-myelination. Apoptosis of OLG has been observed in EAE and MS(64, 65). IFN-g, cuprizone, and HTLV-1, known to induce demy-elination in vivo, also induce OLG apoptosis (66–68). Therefore,an understanding of mechanisms leading to and preventing apo-ptosis of OLG and its progenitor cells is critically important todevelop rational approaches to enhance OLG survival andremyelination.

AcknowledgmentsWe thank Drs. E. Podack and S. Korsemyer who generously providedbcl-2and bax cDNA, respectively. We also appreciate the excellent technicalhelp of Dr. M. Chi in preparing primary OLG and O-2A progenitors andthe preparation of the manuscript by N. Dehghan.

References1. Benveniste, E. N. 1997. Role of macrophages/microglia in multiple sclerosis and

experimental allergic encephalomyelitis.J. Mol. Med. 75:165.

2. Martin, R., H. F. McFarland, and D. E. MCFarlin. 1992. Immunologic aspects ofdemyelinating diseases.Annu. Rev. Immunol. 10:153.

3. Shin, M. L., H. Rus, and F. Niculescu. 1998. Complement system in centralnervous system disorders. InThe Human Complement System in Health andDisease. J. E. Volanakis and M. M. Frank, eds. Marcel Dekker, New York,p. 499.

4. Shin, M. L., H. G. Rus, and F. I. Niculescu. 1996. Membrane attack by comple-ment: assembly and biology of the terminal complement complexes. InBiomem-branes. A. G. Lee, ed. JAI Press, Greenwich, p. 123.

5. Morgan, B. P. 1989. Complement membrane attack on nucleated cells: resistance,recovery and non-lethal effects.Biochem. J. 264:1.

6. Halperin, J. A., A. Taratuska, and A. Nicholson Weller. 1993. Terminal comple-ment complex C5b-9 stimulates mitogenesis in 3T3 cells.J. Clin. Invest. 91:1974.

7. Benzaquen, L. R., A. Nicholson-Weller, and J. A. Halperin. 1994. Terminal com-plement proteins C5b-9 release basic fibroblast growth factor and platelet-derivedgrowth factor from endothelial cells.J. Exp. Med. 179:985.

8. Rus, H. G., F. Niculescu, and M. L. Shin. 1996. Sublytic complement attackinduces cell cycle in oligodendrocytes.J. Immunol. 156:4892.

9. Niculescu, F., T. Badea, and H. Rus. 1999. Sublytic C5b-9 induces proliferationof aortic smooth muscle cells: role of mitogen activated protein kinase and phos-phatidylinositol 3-kinase. Atherosclerosis 142:47.

10. Niculescu, F., H. Rus, T. van Biesen, and M. L. Shin. 1997. Activation of Ras andmitogen-activated protein kinase pathway by terminal complement complexes isG protein dependent.J. Immunol. 158:4405.

11. Carney, D. F., T. J. Lang, and M. L. Shin. 1990. Multiple signal messengersgenerated by terminal complement complexes and their role in terminal comple-ment complex elimination.J. Immunol. 145:623.

12. Wiedmer, T., B. Ando, and P. J. Sims. 1987. Complement C5b-9-stimulatedplatelet secretion is associated with a Ca21-initiated activation of cellular proteinkinases.J. Biol. Chem. 262:13674.

13. Seeger, W., N. Suttorp, A. Hellwig, and S. Bhakdi. 1986. Noncytolytic terminalcomplement complexes may serve as calcium gates to elicit leukotriene B4, gen-eration in human polymorphonuclear leukocytes.J. Immunol. 137:1286.

14. Cybulsky, A. V., J. Papillon, and A. J. McTavish. 1998. Complement activatesphospholipases and protein kinases in glomerular epithelial cells.Kidney Int.54:360.

15. Niculescu, F., H. Rus, S. Shin, T. Lang, and M. L. Shin. 1993. Generation ofdiacylglycerol and ceramide during homologous complement activation.J. Im-munol. 150:214.

16. Niculescu, F., H. Rus, and M. L. Shin. 1994. Receptor-independent activation ofguanine nucleotide-binding regulatory proteins by terminal complement com-plexes.J. Biol. Chem. 269:4417.

17. Barres, B. A., I. K. Hart, H. S. Coles, J. F. Burne, J. T. Voyvodic,W. D. Richardson, and M. C. Raff. 1992. Cell death and control of cell survivalin the oligodendrocyte lineage.Cell 70:31.

18. Barres, B. A., R. Schmid, M. Sendnter, and M. C. Raff. 1993. Multiple extra-cellular signals are required for long-term oligodendrocyte survival.Development118:283.

19. Raff, M. C., B. A. Barres, J. F. Burne, H. S. Coles, Y. Ishizaki, andM. D. Jacobson. 1993. Programmed cell death and the control of cell survival:lessons from the nervous system.Science 262:695.

20. Ludwin, S. K. 1997. The pathobiology of the oligodendrocyte.J. Neuropathol.Exp. Neurol. 56:111.

21. Louis, J. C., E. Magal, S. Takayama, and S. Varon. 1993. CNTF protection ofoligodendrocytes against natural and tumor necrosis factor-induced death.Sci-ence 259:689.

22. D’Souza, S. D., K. A. Alinauskas, and J. P. Antel. 1996. Ciliary neurotrophicfactor selectively protects human oligodendrocytes from tumor necrosis factor-mediated injury.J. Neurosci. Res. 43:289.

23. Linington, C., B. P. Morgan, N. J. Scolding, P. Wilkins, S. Piddlesden, andD. A. Compston. 1989. The role of complement in the pathogenesis of experi-mental allergic encephalomyelitis.Brain 112:895.

24. Piddlesden, S. J., M. K. Storch, M. Hibbs, A. M. Freeman, H. Lassmann, andB. P. Morgan. 1994. Soluble recombinant complement receptor 1 inhibits in-flammation and demyelination in antibody-mediated demyelinating experimentalallergic encephalomyelitis.J. Immunol. 152:5477.

25. Linington, C., H. Lassmann, B. P. Morgan, and D. A. Compston. 1989. Immu-nohistochemical localization of terminal complement component C9 in experi-mental allergic encephalomyelitis.Acta Neuropathol. 79:78.

26. Compston, D. A., B. P. Morgan, A. K. Campbell, P. Wilkins, G. Cole,N. D. Thomas, and B. Jasani. 1989. Immunocytochemical localization of terminalcomplement complex in multiple sclerosis.Neuropathol. Appl. Neurobiol. 15:307.

27. Sanders, M. E., C. L. Koski, D. Robbins, M. L. Shin, M. M. Frank, andK. A. Joiner. 1986. Activated terminal complement in cerebrospinal fluid in Guil-lain-Barre syndrome and multiple sclerosis.J. Immunol. 136:4456.

28. Vanguri, P., B. A. Silverman, C. L. Koski, and M. L. Shin. 1982. Complementactivation by isolated myelin: activation of the classical pathway in the absenceof myelin-specific antibodies.Proc. Natl. Acad. Sci. USA 79:3290.

29. Cyong, J. C., S. S. Witkin, B. Rieger, E. Barbarese, R. A. Good, and N. K. Day.1982. Antibody-independent complement activation by myelin via the classicalcomplement pathway.J. Exp. Med. 155:587.

30. Vanguri, P., and M. L. Shin. 1988. Hydrolysis of myelin basic protein in humanmyelin by terminal complement complexes.J. Biol. Chem. 263:7228.

6137The Journal of Immunology

by guest on March 27, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

31. Liu, W. T., P. Vanguri, and M. L. Shin. 1983. Studies on demyelination in vitro:the requirement of membrane attack components of the complement system.J. Immunol. 131:778.

32. Koski, C. L., A. E. Estep, S. Sawant-Mane, M. L. Shin, L. Highbarger, andG. M. Hansch. 1996. Complement regulatory molecules on human myelin andglial cells: differential expression affects the deposition of activated complementproteins.J. Neurochem. 66:303.

33. Rus, H., F. Niculescu, T. Badea, and M. L. Shin. 1997. Terminal complementcomplexes induce cell cycle entry in oligodendrocytes through mitogen activatedprotein kinase pathway.Immunopharmacology 38:177.

34. Shirazi, Y., H. G. Rus, W. B. Macklin, and M. L. Shin. 1993. Enhanced degra-dation of messenger RNA encoding myelin proteins by terminal complementcomplexes in oligodendrocytes.J. Immunol. 150:4581.

35. Saneto, R. P., and J. de Vellis. 1985. Characterization of cultured rat oligoden-drocytes proliferating in a serum-free, chemically defined medium.Proc. Natl.Acad. Sci. USA 82:3509.

36. Hong, K., T. Kinoshita, W. Miyazaki, T. Izawa, and K. Inoue. 1979. An anti-complementary agent, K-76 monocarboxylic acid: its site and mechanism of in-hibition of the complement activation cascade.J. Immunol. 122:2418.

37. DiScipio, R. G., C. A. Smith, H. J. Muller-Eberhard, and T. E. Hugli. 1983. Theactivation of human complement component C5 by a fluid phase C5 convertase.J. Biol. Chem. 258:10629.

38. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Analysis of RNA: Northernhybridization. InMolecular Cloning, Vol. 1. N. Ford, C. Nolan, and M. Ferguson,eds. Cold Spring Harbor Laboratory Press, Plainview, p. 7.39.

39. Ni, B., X. Wu, Y. Du, Y. Su, E. Hamilton Byrd, P. K. Rockey, P. Rosteck Jr.,G. G. Poirier, and S. M. Paul. 1997. Cloning and expression of a rat brain in-terleukin-1b-converting enzyme (ICE)-related protease (IRP) and its possiblerole in apoptosis of cultured cerebellar granule neurons.J. Neurosci. 17:1561.

40. Robbins, D. S., Y. Shirazi, E. Drysdale, A. Lieverman, H. S. Shin, andM. L. Shin. 1987. Production of cytotoxic factor for oligodendrocytes by stim-ulated astrocytes.J. Immunol. 139:2593.

41. Nicholson, D. W., A. Ali, N. A. Thornberry, J. P. Vaillancourt, C. K. Ding, MGallant, Y. Gareau, P. R. Griffin, M. Labelle, Y. A. Lazebnik, et al. 1995. Iden-tification and inhibition of the ICE/CED-3 protease necessary for mammalianapoptosis.Nature 376:37.

42. Lazebnik, Y. A., S. H. Kaufmann, S. Desnoyers, G. C. Poirier, andW. C. Earnshaw. 1994. Cleavage of poly(ADP-ribose) polymerase by a protein-ase with properties like ICE. Nature 371:346.

43. Thornberry, N. A., and Y. Lazebnik. 1998. Caspases: enemies within.Science281:1313.

44. Green, D. R., and J. C. Reed. 1998. Mitochondria and apoptosis.Science 281:1309.

45. Akassoglou, K., J. Bauer, G. Kassiotis, M. Pasparakis, H. Lassmann, G. Kollias,and L. Probert. 1998. Oligodendrocyte apoptosis and primary demyelination in-duced by local TNF/p55TNF receptor signaling in the central nervous system oftransgenic mice: models for multiple sclerosis with primary oligodendrogliopa-thy. Am. J. Pathol. 153:801.

46. Vakeva, A. P., A. Agah, S. A. Rollins, L. A. Matis, L. Li, and G. L. Stahl. 1998.Myocardial infarction and apoptosis after myocardial ischemia and reperfusion:role of terminal complement components and inhibition by anti-C5 therapy.Cir-culation 97:2259.

47. Gasque, P., S. K. Singhrao, J. W. Neal, O. Gotze, and B. P. Morgan. 1997.Expression of the receptor for complement C5a (CD88) is up-regulated on reac-tive astrocytes, microglia, and endothelial cells in the inflamed human centralnervous system.Am. J. Pathol. 150:31.

48. Kuida, K., T. S. Zheng, S. Na, C. Kuan, D. Yang, H. Karasuyama, P. Rakic, andR. A. Flavell. 1996. Decreased apoptosis in the brain and premature lethality inCPP32-deficient mice.Nature 384:368.

49. Varfolomeev, E. E., M. Schuchmann, V. Luria, N. Chiannilkulchai,J. S. Beckmann, I. L. Mett, D. Rebrikov, V. M. Brodianski, O. C. Kemper,

O. Kollet, et al. 1998. Targeted disruption of the mouse caspase 8 gene ablatescell death induction by the TNF receptors, Fas/Apo1, and DR3 and is lethalprenatally.Immunity 9:267.

50. Kuida, K., T. F. Haydar, C. Y. Kuan, Y. Gu, C. Taya, H. Karasuyama, M. S. Su,P. Rakic, and R. A. Flavell. 1998. Reduced apoptosis and cytochromec-mediatedcaspase activation in mice lacking caspase 9.Cell 94:325.

51. Hakem, R., A. Hakem, G. S. Duncan, J. T. Henderson, M. Woo, M. S. Soengas,A. Elia, J. L. de la Pompa, D. Kagi, W. Khoo, et al. 1998. Differential requirementfor caspase 9 in apoptotic pathways in vivo.Cell 94:339.

52. Xiang, J., D. T. Chao, and S. J. Korsmeyer. 1996. BAX-induced cell death maynot require interleukin 1b-converting enzyme-like proteases.Proc. Natl. Acad.Sci. USA 93:14559.

53. Hisahara, S., S. Shoji, H. Okano, and M. Miura. 1997. ICE/CED-3 family exe-cutes oligodendrocyte apoptosis by tumor necrosis factor.J. Neurochem. 69:10.

54. Adams, J. M., and S. Cory. 1998. The Bcl-2 protein family: arbiters of cellsurvival.Science 281:1322.

55. Chleq-Deschamps, C. M., D. P. LeBrun, P. Huie, D. P. Besnier, R. A. Wamke,R. K. Sibley, and M. L. Cleary. 1993. Topographical dissociation of BCL-2messenger RNA and protein expression in human lymphoid tissues.Blood 81:293.

56. Sakuragi, N., H. Matsuo, G. Coukos, E. E. Furth, M. P. Bronner,C. M. VanArsdale, S. Krajewski, J. C. Reed, and J. F. Struss III. 1994. Differ-entiation-dependent expression of the BCL-2 proto-oncogene in the human tro-phoblast lineage.J. Soc. Gynecol. Invest. 1:164.

57. Harigai, M., T. Miyashita, M. Hanada, and J. C. Reed. 1996. Acis-acting elementin theBCL-2gene controls expression through translational mechanisms.Onco-gene 12:1369.

58. Evans, S. W., and W. L. Farrar. 1987. Interleukin 2 and diacylglycerol stimulatephosphorylation of 40 S ribosomal S6 protein: correlation with increased proteinsynthesis and S6 kinase activation.J. Biol. Chem. 262:4624.

59. Ahmed, N. N., H. L. Grimes, A. Bellacosa, T. O. Chan, and P. N. Tsichlis. 1997.Transduction of interleukin-2 antiapoptotic and proliferative signals via Akt pro-tein kinase.Proc. Natl. Acad. Sci. USA 94:3627.

60. Kennedy, S. G., A. J. Wagner, S. D. Conzen, J. Jordan, A. Bellacosa,P. N. Tsichlis, and N. Hay. 1997. The PI 3 kinase/Akt signaling pathway deliversan anti-apoptotic signal.Genes Dev. 11:701.

61. Shimizu, S., M. Narita, and Y. Tsujimoto. 1999. Bcl-2 family proteins regulatethe release of apoptogenic cytochromec by the mitochondrial channel VDAC.Nature 399:483.

62. Ashkenazi, A., and V. M. Dixit. 1998. Death receptors: signaling and modulation.Science 281:1305.

63. Larocca, J. N., M. Farooq, and W. T. Norton. 1997. Induction of oligodendrocyteapoptosis by C2-ceramide.Neurochem. Res. 22:529.

64. Pender, M. P., K. B. Nguyen, P. A. McCombe, and J. F. R. Kerr. 1991. Apoptosisin the nervous system in experiental allergic encephalomyelitis.J. Neurol. Sci.104:81.

65. Ozawa, K., G. Suchanek, H. Breitschopf, W. Bruck, W. Budka, H. Budka,K. Jellinger, and H. Lassmann. 1994. Patterns of oligodendroglial pathology inmultiple sclerosis.Brain 117:1311.

66. Vartanian, T., Y. Li, M. Zhao, and K. Stefansson. 1995. Interferon-g-inducedoligodendrocyte cell death: implications for the pathogenesis of multiple sclero-sis.Mol. Med. 1:732.

67. Blakemore, W. F. 1972. Observations on oligodendrocyte degeneration, the res-olution of status spongiosus and remyelination in cuprizone intoxication in mice.J. Neurocytol. 1:413.

68. Seto, K., M. Abe, O. Ohya, O. Itakura, N. Ishiguro, H. Ikeda, A. Wakisaka, andT. Yoshiki. 1995. A rat model of HTLV-I infection: development of chronicprogressive myeloneuropathy in seropositive WKAH rats and related apoptosis.Acta Neuropathol. 89:483.

6138 ANTI-APOPTOTIC ACTIVITY OF C5b-9

by guest on March 27, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from