Download - Induction of Oligodendrocyte Apoptosis by C 2 -Ceramide

Neurochemical Research, Vol. 22, No. 4, 1997, pp. 529-534

Induction of Oligodendrocyte Apoptosis by C2-Ceramide*

J. N. Larocca,1'2 M. Farooq,1 and W. T. Norton1

(Accepted December 2, 1996)

Tumor necrosis factor-a induces oligodendrocytes apoptosis, and is known to stimulate the hy-drolysis of sphingomyelin to form the lipid mediator, ceramide. These data encouraged us todetermine whether ceramide itself is able to induce apoptosis in oligodendrocytes. For this purposethe cell-permeable ceramide analog, C2-ceramide was used. Treatment of bovine oligodendrocytecell cultures with this compound induced cell death in a time- and concentration-dependent manner.The induction of cell death was specifically associated with the action of C2-ceramide and couldnot be elicited by dioctanoylglycerol (DC8) or phorbol 12-myristate 13-acetate (PMA). Treatmentof the cultures with neutral sphingomyelinase, which increased the hydrolyses of endogenous sphin-gomyelin, resulted in oligodendrocyte death, whereas exposure of the cells to phospholipase C andA2 did not. C2-ceramide treatment caused DNA fragmentation. Morphologic analysis of the cellsshowed that C2-ceramide treatment resulted in a loss of their processes, reduction of cell volume,chromatin condensation, and formation of apoptotic bodies. These results indicate that ceramidecan induce oligodendrocyte apoptosis, and suggest that sphingolipid metabolism plays a key rolein the regulation of this process.

KEY WORDS: Oligodendrocyte; apoptosis; ceramide.

INTRODUCTION

Apoptosis is a term originally proposed by Kerr,Wyllie, and Currie (1) to describe a process of controlledcell deletion. Apoptosis, or programmed cell death, is anactive process of cellular self-destruction with distinctivemorphological and biochemical features (2). Duringapoptosis the cells shrink and their chromatin degradesbefore loss of plasma membrane integrity. The dead cellremnants are recognized and rapidly phagocytosed byneighboring cells in the tissue (3). Cell death is con-spicuous from the very beginning of development, andit is also used for homeostasis in adult organisms (4). In

' Department of Neurology, Albert Einstein College of Medicine,Bronx, New York.

2 Address reprint requests to: Dr. Jorge N. Larocca, Department ofNeurology, Albert Einstein College of Medicine, 1300 Morris ParkAvenue, Bronx, New York 10461. Tel.: (718) 430-2831; fax: (718)430-8790.

* Special issue dedicated to Dr. Eduardo Soto.

5290364-3190/97/0400-0529S12.50/0 C 1997 Plenum Publishing Corporation

the developing nervous system neurons that fail to makecorrect connections die (5). Similarly, it has been pos-tulated that neuroglial apoptosis is involved in establish-ing the normal ratio of glial cells to neurons and inrefining the association between neurons and glia (6).The immune system has mechanisms to protect againstviral infections by targeting infected cells for death (7).Cell death may also be used to minimize the risk fromcells frequently subjected to mutagenic chemicals or ra-diation (8). As protection against malignancy, tumor ne-crosis factor can trigger the apoptotic death oftransformed host cells (9,10).

With cell death being a normal part of so manydifferent systems, the molecular mechanism that carriesit out must be properly regulated; when this regulationis disturbed, diseases can result (11-13). Both inhibitionof cell death and inappropriate cell death may be dele-terious (12,13). In that regard, increased oligodendrocyteapoptosis has been described in different pathologic con-ditions that lead to myelin deficiency; for example in

530 Larocca, Farooq, and Norton

experimental allergic encephalomyelitis (14), and cupri-zone intoxication (15).

Oligodendrocytes undergo apoptosis in response toa death-inducing stimulus or because of the loss of astimulus whose function is to suppress the execution ofa death program. Recent reports show that oligodendro-cytes undergo apoptosis in vitro if insufficient amountsof serum or trophic factors are provided (16,17). Theexistence of factors capable of inducing oligodendrocyteapoptosis was demonstrated by in vitro studies (18-20).Different stimuli, as well as viral infection, induces ne-onatal rat astrocytes to produce a soluble cytotoxic fac-tor, identified as tumor necrosis factor alpha (TNFa),that causes oligodendrocyte apoptosis (21). It was alsoestablished that tumor necrosis factor beta (TNFB) in-duces cell death in Oligodendrocytes (18,22).

Although it is well known that TNF induces oli-godendrocyte apoptosis, the intracellular mediator of oli-godendrocyte apoptosis it is unknown. The resultspresented here clearly indicate that ceramide is a biolog-ically active molecule able to induce oligodendrocyteapoptosis.

EXPERIMENTAL PROCEDURE

was synthesized by acetylation of sphingosine with (CH3CO)20 andpurified by thin layer chromatography (24).

In Situ DNA Fragmentation Detection. These experiments werecarried out using an in situ apoptosis detection kit, "ApopTag Plus"(ONCOR). This method is based on the specific staining of the veryhigh concentration of 3'-OH ends produced by the DNA fragmentationin apoptotic cells. The procedure was carried out according to theinstructions of the manufacturer. Briefly, after treatment of the cultureswith C2-ceramide (20 uM) for 4 hours, the cells were washed withcold phosphate buffered saline (PBS) and then fixed in 4% neutralbuffered formalin for 15 min at room temperature. Digoxigenin-nu-cleotides were catalytically added to the DNA by terminal deoxynu-cleotidyl transferase and identified by specific anti-digoxigeninantibody conjugated to peroxidase. The staining was performed byincubating the preparation with diaminobenzidene (DAB). Positivecontrols were obtained by incubating the cells with DNAse I afterfixation. When negative-stained controls were desired, terminal deox-ynucleotidyl transferase was omitted in the staining procedure. Thenumber of positive nuclei was determined in at least 20 random fieldsusing a X 40 microscope objective on a light microscope. The resultswere expressed as % of dead cells relative to the total number of cells.

Morphologic Analysis. The morphologic changes occurring in Oli-godendrocytes exposed to C2-ceramide were assessed by electron mi-croscopy. Oligodendrocyte cultures were exposed to C2-ceramide (20uM) for 6 and 24 hours. After treatment the cells were fixed in 2.5%glutaraldehyde and embedded in Epon 812 according to standard pro-tocols. One-micron sections were stained with toluidine blue and an-alyzed by light microscopy. Representative fields were then preparedfor electron microscopy and observed in a Siemens 101.

Materials. Sphingosine, acetic anhydride, and all buffers and saltswere purchased from Sigma Chemical Co. (St. Louis, MO.). The InSitu apoptosis detection kit "ApopTag Plus" was purchased from ON-COR (Gaithersburg, MD.). The Live/Dead cell assay kit was pur-chased from Molecular Probes, Inc. (Eugene, OR.). The culture mediawere purchased from Life Technologies, Inc. (Gaithersburg, MD.).

Cell Culture. Mature, differentiated Oligodendrocytes from bovinebrain were isolated and established in culture according to Norton andFarooq (23). The cells were plated in poly-1-lysine-coated 55 mmdishes at a concentration of 5 X 106 cells per dish and cultured inMEM medium with Earle's salts, 15% heat inactivated calf serum,penicillin-streptomycin, fungizone, glutamine and 0.5% glucose. After5 days the culture medium was replaced with MEM medium contain-ing Earle's salts, transferrin, 50 ug/ml; biotin, 10 ng/ml; sodium sel-enite, 30 nM; insulin, 5 ug/ml and triiodothyronine, 15 nM; andcultured for 24 hours before experimentation.

Cell Death. Cell death following treatment with C2-ceramide orother agents was determined with the Live/Dead Cell assay (MolecularProbes), which employs calcein AM and ethidium homodimer. Calceinis cell permeable, retained in live cells and converted to a fluorescent(530) product by esterase activity in live cells. Ethidium homodimerenters cells with damaged membranes and undergoes a 40-fold en-hancement of fluorescence (> 600 nM) on binding to nucleic acids.After treatment calcein (0.5 uM, final concentration) and ethidiumhomodimer (2 uM, final concentration) were added to each well andthe plates were incubated at room temperature for 30 min. The numberof dead and live cells was determined in at least 20 random fieldsusing a X 20 microscope objective on a fluorescence microscope fittedwith appropriate filters. The results were expressed as % of dead cellsrelative to the total number of cells. The C2-ceramide used in this work

RESULTS

To define whether ceramide is able to mediate oli-godendrocyte death, the action of the cell-permeable cer-amide analog, C2-ceramide was investigated. Culturedbovine Oligodendrocytes were incubated for 24 hours insynthetic medium prior to experimentation. In the firstset of experiments the cells were treated with 10 uM C2-ceramide delivered in an ethanol vehicle, (ethanol 0.1%v/v final concentration), for 24 hours. After treatmentthe number of dead cells was determined by theLive/Dead Cell assay. Figure 1 shows representative pic-tures of oligodendrocyte cultures after C2-ceramide treat-ment. We found that treatment of cultures with 10 uMCj-ceramide in ethanol induced a marked increase in thenumber of dead cells (Fig. IF). Approximately 75% ofthe Oligodendrocytes died after 24 hours of treatmentwith 10 uM ceramide. Because treatment with ethanolalone did not increase the number of dead cells (Fig.1D), we concluded that the death of Oligodendrocyteswas induced by C2-ceramide and not by the vehicle.

The effect of C2-ceramide was time-dependent. Ad-dition of 5 uM C2-ceramide to the oligodendrocyte cul-tures resulted in the death of 25%, 75% and 95% of the

Oligodendrocyte Apoptosis

Fig. 1. Live/Dead cell assay on oligodendrocyte cultures. Oligoden-drocyte cultures were untreated (A and B), or treated with 0.1% v/v(final concentration) ethanol (C and D), or 10 uM C2-ceramide inethanol (E and F) for 24 hours. Live and dead cells were detected byfluorescent microscopy using the Live/Dead Cell assay (MolecularProbes). Live cells A, B and E, and dead cells B, D, and F.

Fig. 2. Time dependence of C2-ceraraide-induced oligodendrocytedeath. Oligodendrocyte cultures were treated with 0.1% v/v (final con-centration ethanol or 5 uM C2-ceramide in ethanol. At the indicatedtimes the percentage of dead cells relative to the total number of cellswas determined as described in "Experimental Procedure'', o-o eth-anol, •-• C2-ceramide. The results were obtained from three indepen-dent determinations; bars represent standard error of the mean. * p <0.05 and ** < 0.01 (Student's two-tailed t test).

cells after 24, 48, and 72 hours of treatment, respectively(Fig. 2).

To determine the dependence of oligodendrocytedeath on C2-ceramide concentration, oligodendrocytecultures were treated with C2-ceramide at increasingconcentrations and the number of dead cells was deter-mined after 24 hours of treatment (Fig. 3). C2-ceramidecaused dose-dependent death of the oligodendrocytes.The maximum effect was observed at a concentration of20 uM. However, the minimum concentration used, 2.5uM, caused a statistically significant increase in thenumber of dead oligodendrocytes.

To determine the specificity of C2-ceramide in theinduction of oligodendrocyte death, the effects of otheramphiphilic lipid analogs, dioctanoylglycerol (DC8), acell-permeable analog of diacylglycerol, and phorbol 12-myristate 13-acetate (PMA), a pharmacologic activatorof protein kinase C, were determined. Bovine oligoden-drocyte cultures were treated with different concentra-tions of DC8, PMA and C2-ceramide (0.1, 1 and 10 uM),delivered in an ethanol vehicle or ethanol (0.1% v/v finalconcentration), for 48 hours. Treatment with ethanol didnot increase the number of dead cells (results notshown). As shown in Fig. 4, DC8 did not cause cell deathat any of the concentrations used. PMA at 0.1 uM, aconcentration previously shown to activate protein ki-

nase C, did not induce oligodendrocyte death. Moreover,10 uM PMA only increased the number of dead cellsby approximately 10%. On the contrary, treatment of thecultures with 10 uM C2-ceramide resulted in the deathof 95% of the cells. These results supported the role ofceramide as a selective mediator of oligodendrocytedeath.

In a separate group of experiments the putative roleof sphingomyelin hydrolysis in the induction of oligo-dendrocyte death was analyzed using an exogenous neu-tral sphingomyelinase. The oligodendrocyte cultureswere treated with sphingomyelin phosphodiesterasefrom Staphylococcus aureus (10~5 to 10-2 U/ml) for 48hours. As is shown in Fig. 5, the number of dead cellsincreased with increasing sphingomyelinase concentra-tions. Treatment with a similar range of concentrationsof phospholipase C from Bacillus cereus, and phospho-lipase A2 from porcine pancreas, did not result in oli-godendrocyte death (results not shown).

The C2-ceramide-treatment of oligodendrocyte cul-tures caused DNA fragmentation. In this group of ex-periments the oligodendrocyte cultures were treated with20 uM C2-ceramide for 4 hours. After treatment the cellswere washed with cold PBS and then fixed in 4% neutralbuffered formalin and DNA fragmentation was deter-minated using an in situ apoptosis detection kit,


Fig. 3. Concentration dependence of C2-ceramide-induced oligoden-drocyte death. Oligodendrocyte cultures were treated with increasingconcentrations of C2-ceramide (2.5 to 20 )uM) for 24 hours. Cell deathwas assessed as described in "Experimental Procedure". The resultswere obtained from three independent determinations; bars representstandard error of the mean. * p < 0.05 and ** < 0.01 (Student's two-tailed t test).

CONCENTRATION (uM)

Fig. 4. Effect of lipid analogs on oligodendrocyte cultures. BovineOligodendrocyte cultures were treated with DC8, PMA and C2-ceram-ide (0.1, 1, and 10 uM), delivered in and ethanol vehicle for 48 hours.Cell death was assessed as described in "Experimental Procedure".The results were obtained from three independent determinations; barsrepresent standard error of the mean. *p < 0.05 and ** < 0.01 (Stu-dent's two-tailed t test).

"ApopTag Plus" (ONCOR). Figure 6 shows represen-tative pictures of oligodendrocyte cultures after treat-ment. We found that C2-ceramide treatment induced

Fig. 5. Effect of exogenous sphingomyelinase on oligodendrocytedeath. Bovine oligodendrocyte cultures were treated with sphingo-myelin phosphodiesterase from Staphylococcus aureus (10-5 to 10-2

U/ml) for 48 hours. Cell death was assessed as described in "Exper-imental Procedure". The results were obtained from three independentdeterminations; bars represent standard error of the mean. * p < 0.05and ** < 0.01 (Student's two-tailed t test).

Fig. 6. C2-ceramide-induction of DNA fragmentation. Oligodendrocytecultures were treated with C2-ceramide (20 uM final concentration) for4 hours. A and B untreated cultures. C and D treated cultures. Bpositive control, after fixation the cells were incubated with DNAse I.D negative control, terminal deoxynucleotidyl transferase was omittedin the staining procedure. Cell death was assessed as described in"Experimental Procedure" X400.

DNA fragmentation. Approximately 20 to 30% of thecell nuclei showed a positive reaction (Fig. 6C). In con-trast, less than 1% of the nuclei showed a positive re-action in the untreated cultures (Fig. 6A). Fig. 6B showsa positive control. In this case, after fixation untreatedcultures were incubated with a buffer containing DNAse

Oligodendrocyte Apoptosis 533

DISCUSSION

Fig. 7. One micron toluidine blue stained epoxy sections of oligoden-drocyte culture. A, cultures nontreated; B, oligodendrocytes treatedwith 20 uM C2-ceramide for 6 hours; and C, oligodendrocytes treatedwith C2-ceramide for 24 hours. X600.

Fig. 8. Electron micrographs of oligodendrocyte cultures. A, controlcells; B, oligodendrocytes treated with 20 uM C2-ceramide for 6 hours;and C, oligodendrocytes treated with C2-ceramide for 24 hours.X9,000 A and B; X I5,000 C.

I. Finally Fig. 6D shows a negative control. In this in-stance, C2-ceramide treated cultures were processed asin Fig. 6C, except that terminal deoxynucleotidyl trans-ferase was omitted in the staining procedure. These re-sults strongly suggested that the staining observed in thenuclei of the oligodendrocyte cultures treated with C2-ceramide is the result of DNA fragmentation.

The morphologic changes induced by the C2-cer-amide treatment of the oligodendrocyte cultures were as-sessed by analysis of one-micron sections stained withtoluidine blue by light microscopy. Fig. 7B, shows thatafter 6 hours of treatment with 20 uM C2-ceramide thesecells lost their fine processes and suffered a reduction ofcytoplasm, suggesting that the cytotoxic mechanism didnot involve lytic activity. These changes was more evi-dent at the electron microscope level (Fig. 8B). It is alsoevident that the C2-ceramide treatment resulted in an in-crease in the condensation of chromatin around the nu-clear membrane (Fig. 8B). When the cells wereexamined 24 hours after C2-ceramide treatment, thechromatin in the nucleus was completely condensed, andthe disintegration of the cell into apoptotic bodies wasevident (Fig. 8C).

The results presented here strongly suggest that cer-amide is a biologically active molecule capable of in-ducing oligodendrocyte apoptosis.

Sphingolipids are known to play a important rolein the regulation of cell growth and differentiation, cell-cell contact, and oncogenesis (25,26). The discovery ofinhibition of protein kinase C by sphingosine suggestedthat sphingolipid-derived molecules are biologically ac-tive substances that can play the role of second messen-gers (27). The fact that sphingomyelin hydrolysis isactivated by several extracellular signals including TNF,1,25-dihidroxyvitamin D3 and -y-interferon, strongly in-dicated that ceramide, the product of sphingomyelin hy-drolysis is the intracellular mediator of the action ofthese substances (28-30). These observations, togetherwith the fact that TNF induces apoptosis of oligoden-drocytes, persuaded us to determine if ceramide is ableto induce oligodendrocyte apoptosis. For that purposethe ability of the cell-permeable ceramide analog, C2-ceramide, to induce oligodendrocyte apoptosis was in-vestigated. Our results show that C2-ceramide inducedoligodendrocyte death in a time and concentration-de-pendent manner. The range of C2-ceramide concentra-tions used in our experiments was similar to that usedin previous studies. Other investigators have shown thatC2-ceramide, in concentrations ranging from 1 to 10 uM,was able to induce apoptosis of Jurkat T cells (24), in-hibit cell proliferation, induce monocytic differentiationof HL-60 cells (31), down-regulate c-myc levels in HL-60 cells (32) and cause DNA fragmentation (24). Ourresults showed that C2-ceramide-treatment of oligoden-drocyte cultures caused DNA fragmentation as early asfour hours after treatment. Further indication that C2-ceramide treatment resulted in apoptosis of oligodendro-cytes was provided by the morphologic analysis of thecell cultures; treatment of the cells resulted in a loss oftheir fine processes, reduction of cell volume, chromatincondensation, and, finally, formation of apoptotic bodies.

The induction of cell death was specifically asso-ciated with the action of ceramide and could not be elic-ited by other lipids, including DC8 and PMA. Theseobservation are in agreement with the fact that hydrol-ysis of endogenous sphingomyelin by the action ofsphingomyelinase from Staphylococcus aureus inducedoligodendrocyte death, while exposure of the cultures tophospholipases C and A2 did not produce cell death.These observations clearly indicated the specific involve-ment of the sphingolipid pathway in the regulation of oli-godendrocyte apoptosis. In addition, preliminaryexperiments indicated that treatment of oligodendrocyte


cultures with TNFa resulted in increased levels of en-dogenous ceramide (results not shown). The importanceof knowing the pathway responsible for the regulation ofoligodendrocyte apoptosis is emphasized by the findingof TNFa and TNFB(3 in CNS tissue samples from multiplesclerosis patients (33), a finding which strongly suggeststhat these cytokines are involved in the disease process.

In summary, pharmacological manipulation that in-creased the level of free ceramide selectively induced apop-tosis of oligodendrocytes. This suggests that activation ofceramide-sensitive enzymes constitutes a key step in theregulation of programmed death of oligodendrocytes.

ACKNOWLEDGMENTS

To Eduardo: friend, mentor and colleague; with much gratitudefor fostering my career (JNL). Best wishes, good luck and good health,and special thanks for the great visit to Buenos Aires (WTN).

This work was supported by National Multiple Sclerosis Societygrant RG1941 (JNL), and U.S.P.H.S. grant NS02476 (W.T.N.).

REFERENCES

1. Kerr, J. F. R., Wyllie, A. H., and Currie, A. R. 1972. Apoptosis:A basic biological phenomenon with wider ranging implicationsin tissue kinetics. Br. J. Cancer. 24:239-257.

2. Kerr, J. F. R., Serol, J., Harmon, B. V., and Bishop, C. J. 1987.Apoptosis. Pages 93-128, in C. S. Potten, (ed.) Perspectives onMammalian Cell Death, Chap. 5 Oxford Science Publications, Ox-ford and New York.

3. Wyllie, A. H. 1981. Cell death: a new classification separatingapoptosis from necrosis. Pages 3-34, in I. D. Bowen and R. A.Lockshin, (eds.) Cell Death in Biology and Pathology, Chap. 1Chapman & Hall, London and New York.

4. Vaux, D. L. 1993. Toward an understanding of the molecularmechanisms of physiological cell death. Proc. Natl. Acad. Sci.USA. 90:786-789.

5. Kimura, K. I., and Truman, J. W. 1990. Postmetamorphic celldeath in the nervous and muscular systems of Drosophila melan-ogaster. J. Neurosci. 10:403-411.

6. Knapp, P. E., Skoff, R. P., and Redstone, D. W. 1986. Oligoden-droglial cell death in jimpy mice: An explanation for the myelindeficit. J. Neurosci. 6:2813-2822.

7. Cohen, J. J., Duke, R. C., Fado, V. A., and Sellins, K. S. 1992.Apoptosis and programmed cell death in immunity. Ann. Rev.Immunol. 10:267-293.

8. Hickman, J. A. 1992. Apoptosis induced by anticancer drugs. Can-cer Metastisis Rev. 11:121-139.

9. Laster, S. M., Wood, J. G., and Gooding, L. R. 1988. Target-induced changes in macrophage migration may explain differencesin lytic sensitivity among simian virus 40-transformed fibroblasts.J. Immunol. 141:2629-2634.

10. Heller, R. A., Song, K., Fan, N, and Chang, D. J. 1992. The p70tumor necrosis factor receptor mediates cytotoxicity. Cell. 70:47-56.

11. Stewart, B. W. 1994. Mechanisms of apoptosis: Integration ofgenetic, biochemical, and cellular indicators. J. Nat. Cancer Inst.86:1287-1296.

12. Cohen, J. J. 1993. Apoptosis. Immunol. Today. 14:126-130.13. Wyllie, A. H. 1995. The genetic regulation of apoptosis. Current

Opinion in Genetics and Development. 5:97-104.14. Pender, M. P., Nguyen, K. B., McCombe, P. A., and Kerr, J. F.

R. 1991. Apoptosis in the nervous system in experimental allergicencephalomyelitis. J. Neurol. Sci. 104:81-87.

15. Blakemore, W. F. 1972. Observations on oligodendrocyte degen-eration, the resolution of statuspongiosus and remyelination in cu-prizone intoxication in mice. J. Neurocytol. 1:413-426.

16. Barres, B. A., Hart, I. K., Coles, H. S. R., Burne, J. F., Voyvodic,J. T., Richardson, W. D., and Raff, M. C. 1992. Cell death andcontrol of cell survival in the oligodendrocyte lineage. Cell. 70:31-46.

17. Barres, B. A., Schmid, R., Sendnter, M., and Raff, M. C. 1993.Multiple extracellular signals are required for long-term oligoden-drocyte survival. Development. 118:283-295.

18. Selmaj, K.,Raine, C. S., Farooq, M., Norton, W. T., and Brosnan,C. F. 1991. Cytokine cytotoxicity against oligodendrocytes. J. Im-munol. 147:1522-1529.

19. Robbins, D. S., Shirazi, Y., Drysdale, B. E., Lieberman, A., Shin,H.S., and Shin, M. L. 1987. Production of cytotoxic factor foroligodendrocytes by stimulated astrocytes. J. Immunol. 139:2593-2597.

20. Ruddle, N. H., Bergman, C. M., McGrath, K.M., Lingenheld, E.G., Grunnet, M. L., Padula, J. S., and Clark, R. B. 1990. Anantibody to lymphotoxin and tumor necrosis factor prevents trans-fer of experimental allergic encephalomyelitis. J. Exp. Med. 172:1193-1200.

21. Lieberman, A. P., Pitha, P. H., Shin, H. S., and Shin, M. L. 1989.Production of tumor necrosis factor and other cytokines by astro-cytes stimulated with lipopolysaccharide or neurotrophic virus.Proc. Natl. Acad. Sci. USA. 86:6348-6352.

22. Louis, J. C., Magal, E., Takayama, S., and Varon, S. 1993. CNTFprotection of oligodendrocytes against natural and tumor necrosisfactor-induced death. Science. 259:689-692.

23. Norton, W. T., and Farooq, M. 1990. Bulk isolation and cultureof oligodendroglia from mature brain. Pages 171-178, in Conn,P. M. (ed.), Methods in Neurosci. Vol. 2, Cell culture, AcademicPress, Inc., New York, New York.

24. Obeid, L. M., Linardic, C. M., Karolak, L. A., and Hannun, Y.A. 1993. Programmed cell death induced by ceramide. Science.259:1769-1771.

25. Hakomori, S. 1981. Glycosphingolipids in cellular interaction dif-ferentiation, and oncogenesis. Annual Review of Biochemistry.50:733-764.

26. Hannun, Y. A., and Bell, R. M. 1989. Functions of sphingolipidsand sphingolipid breakdown products in cellular regulation. Sci-ence. 243:500-507.

27. Hannun, Y. A., Merrill, A. H. Jr., and Bell, R. M. 1991. Use ofsphingosine as inhibitor of protein kinase C. Methods in Enzy-mology. 201:316-328.

28. Mathias, S., Younes, A., Kan, C. C., Orlow, I., Joseph, C., andKolesnick, R. M. 1993. Activation of the sphingomyelin signalingpathway in intact EL4 cells and in a cell-free system by L-lB.Science. 259:519-522.

29. Okazaki, T., Bielawska, A., Domae, N., Bell, R. M., and Hannun,Y. A. 1994. Characteristics and partial purification of a novelcy-tosolic, magnesium-independent, neutral sphingomyelinase activatedin early signal transduction of la,25-dihydroxyvitamin D3-inducedHL-60 cell differentiation. J. Biol. Chem. 269:4070-4077.

30. Kim, M. Y., Linardic, C., Obeid, L., and Hannun, Y. A. 1991.Identification of sphingomyelin turnover as an effector mechanismfor the action of tumor necrosis factor a. and -y-Interferon. J. Biol.Chem. 266:484-489.

31. Okazaki, T., Bielawska, A., Bell, R. M., and Hannun, Y. A. 1990.Role of ceramide as a lipid mediator of la,25-dihydroxyvitamin D3-induced HL-60 cell differentiation. J.Biol. Chem. 265:15823-31.

32. Dbaibo, G. S., Obeid, L. M., and Hannun, Y. A. 1993. Tumornecrosis factor-a (TNF-a) signal transduction through ceramide.J. Biol. Chem. 268:17762-17766.

33. Herman, F. M., Hilton, D. R., Johnson, K., and Merrill, J. E. 1989.Tumor necrosis factor identified in multiple sclerosis brain. J. Exp.Med. 170:607-612.