Journal of Neurochemistry, 2001, 77, 1531±1541

Cleavage of Bax is mediated by -dependent or -independent activation in dopaminergic neuronal cells: protective role of Bcl-2

Won-Seok Choi,* Eun-Hee Lee,* Chul-Woong Chung,² Yong-Keun Jung,² Byung K. Jin,³ Seung U. Kim,³ Tae H. Oh,§ Takaomi C. Saido¶ and Young J. Oh*

*Department of Biology, Yonsei University College of Science, Seoul, Korea ²Department of Life Science, Kwangju Institute of Science and Technology, Kwangju, Korea ³Brain Research Center, Ajou University School of Medicine, Suwon, Korea §Department of Anatomy and Neurobiology, University of Maryland School of Medicine, Baltimore, Maryland, USA ¶Laboratory for Proteolytic Neuroscience, RIKEN Brain Science Institute, Hirosawa, Wako-shi, Saitama, Japan

Abstract Thus, cotreatment of cells with calpain inhibitor blocked Two cysteine families, caspase and calpain, are both MPP1- and STS-induced Bax cleavage. Intriguingly, known to participate in . We investigated whether a overexpression of baculovirus-derived inhibiting protein of stress-speci®c protease activation pathway exists, and to caspase, p35 or cotreatment of cells with caspase inhibitor what extent Bcl-2 plays a role in preventing drug-induced blocked STS- but not MPP1-induced Bax cleavage. This protease activity and cell death in a dopaminergic neuronal appears to indicate that calpain activation may be either cell line, MN9D. Staurosporine (STS) induced caspase- dependent or independent of caspase activation within the dependent while a dopaminergic neurotoxin, same cells. However, cotreatment with calpain inhibitor MPP1 largely induced caspase-independent necrotic cell rescued cells from MPP1-induced but not from STS-induced death as determined by morphological and biochemical neuronal cell death. In these paradigms of dopaminergic cell criteria including cytochrome c release and ¯uorogenic death, overexpression of Bcl-2 prevented both STS- and caspase cleavage assay. At the late stage of both STS- and MPP1-induced cell death and its associated cleavage of Bax. MPP1-induced cell death, Bax was cleaved into an 18-kDa Thus, our results suggest that Bcl-2 may play a protective role fragment. This 18-kDa fragment appeared only in the by primarily blocking drug-induced caspase or calpain activity mitochondria-enriched heavy membrane fraction of STS- in dopaminergic neuronal cells. treated cells, whereas it was detected exclusively in the Keywords: Bax, Bcl-2, calpain, caspase, MPP1, cytosolic fraction of MPP1-treated cells. This proteolytic staurosporine. cleavage of Bax appeared to be mediated by calpain as J. Neurochem. (2001) 77, 1531±1541. determined by incubation with [35S]methionine-labelled Bax.

Apoptosis is a controlled process to remove unnecessary, aged, or damaged cells in various situations (Thompson 1995). It is well known that apoptosis is characterized by Received September 29, 2000; revised manuscript received March 12, 2001; accepted March 17, 2001. distinct morphological changes such as cellular shrinkage, Address correspondence and reprint requests to Y. J. Oh, Department blebbing and chromatin condensation. Among the cellular of Biology, Yonsei University College of Science, 134 Shinchondong molecules known to regulate apoptosis is a family of Seodaemoongu, Seoul 120±749, Korea. E-mail: [email protected] cysteine ± recently termed (Martin and Abbreviations used: Ac-DEVD-AMC, acetyl-Asp-Glu-Val-Asp-7- Green 1995; Cryns and Yuan 1998). Once activated through amino-4-methylcoumarin; BAF, Boc-aspartyl(OMe)-¯uoromethylketone; CCM, complete culture medium; CPT, calpeptin; MTT, 3-[4,5- a proteolytic cascade, caspases cleave endogenous cellular dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide; PARP, poly- substrates and consequently amplify the death signals (ADP-ribose)polymerase; STS, staurosporine; Z-VAD, N-benzyloxy- (Thornberry and Lazebnik 1998). carbonyl-Val-Ala-Asp-¯uoromethylketone.

q 2001 International Society for Neurochemistry, Journal of Neurochemistry, 77, 1531±1541 1531 1532 W.-S. Choi et al.

In addition to caspases, activation of proteolytic pathways Materials and methods by serine proteases, aspartic proteases and proteosomes is involved in cell death. Moreover, a signi®cant focus has Cell culture and drug treatments been directed towards another , calpain, MN9D cells that stably overexpressed human Bcl-2 (MN9D/Bcl-2), and with a concomitant evaluation of its contribution to both or vector alone (MN9D/Neo) have been previously established, and apoptosis (Chan and Mattson 1999). Two forms characterized, and maintained (Oh et al. 1995). Both MN9D of , m- and m-calpain, are expressed ubiquitously in parental cells and MN9D/Neo cells had no detectable levels of endogenous Bcl-2 as determined by immunoblot analysis (Oh et al. tissues, and activated by Ca21 and autolytic 1995). MN9D cells stably overexpressing baculovirus p35 (MN9D/ processing. Activated calpain induces of speci®c P35) were established by transfecting with a pCDNA3 eukaryotic cellular substrates. These include several cytoskeletal, expression vector containing a full-length p35 cDNA and regulatory and membrane proteins of various cell types subsequently characterized by RT-PCR. Primers used included (Chan and Mattson 1999; Wang 2000). Over the past decade 50-CGACGAACGCAACGACTACT-30 (forward primer) and 50- or so, the possible implication of calpain in neuronal death CTTTTCGGATTTGCCCCAGC-3 0 (reverse primer). Cells from of the central nervous system has been suggested. For each stable cell line were plated at a density of 2 104 cells on  example, calpain expression is elevated in the brains of 25 mg/mL poly-d-lysine-coated 48-well plates (Costar, Corning, patients with multiple sclerosis, amyotrophic lateral sclero- NY, USA). Cells were maintained in Dulbecco's modi®ed Eagle's sis and Alzheimer's disease as well as in experimental medium supplemented with 10% heat-inactivated fetal bovine models of traumatic injury and ischemic neuronal death serum (Life Technologies, Rockville, MD, USA) and 250 mg/mL G418 (Life Technologies; complete culture medium, CCM) for 3 (Saito et al. 1993; Kamp¯ et al. 1997; Springer et al. 1997; days in an incubator with an atmosphere of 10% CO at 378C. Cells Lipton 1999; Shields et al. 1999; Stracher 1999; Nixon 2 were subsequently switched to serum-free N2 medium containing 2000). Recently, putative mechanisms by which calpain the various experimental reagents and further incubated for the mediates neuronal death have also been proposed in certain indicated time periods. Reagents used included STS (Sigma, St neurological disorders (Steiner et al. 1998; Zhang et al. Louis, MO, USA), MPP1 (RBI), Boc-aspartyl(OMe)-¯uoromethyl- 1999). Although calpain is known to be elevated in ketone (BAF; Systems Products, Dublin, CA, USA), patients with Parkinson's disease (Mouatt-Prigent et al. N-benzyloxycarbonyl-Val-Ala-Asp-¯uoromethylketone (Z-VAD; 1996), the molecular mechanisms for this still remain to be Enzyme Systems Products), calpeptin (Calbiochem, San Diego, determined. CA, USA) and MDL 28170 (Calbiochem). Recent reports indicate that caspases and calpains may Transmission electron microscopy work either in concert or independently during cell death MN9D/Neo cells were plated at 2 105 cells on 25 mg/mL poly  (Wang 2000). Furthermore, caspase and calpain sequentially d-lysine-coated 6-well plates (Costar), maintained in CCM for in¯uence cell death. For example, calpain acts as a negative 3 days, switched to N2 medium and treated with 1 mm STS or or positive upstream regulator of caspase processing (Ruiz- 50 mm MPP1 for the indicated time periods. Cells were ®xed in Vela et al. 1999; Wolf et al. 1999; Chua et al. 2000). Karnovsky's ®xative overnight at 48C, and then post®xed in 1% Conversely, caspase-3-mediated degradation of an endo- osmium tetroxide/1.5% ferrocyanide solution for 30 min at room genous calpain inhibitor, calpastatin facilitates calpain temperature (218C). Following dehydration in a series of graded ethanols, cells were embedded in Epon resin, and heat-polymerized. activation that may further the proteolysis initiated by Ultrathin sections were mounted, stained with uranyl acetate and lead caspase (Porn-Ares et al. 1998; Wang et al. 1998). However, citrate, and then examined with a Zeiss EM 902 A transmission the potential cross-talk and the relative contributions of electron microscope (Zeiss, Zena, Germany). these two proteases to neuronal death are relatively unknown. In this study, we speci®cally investigated whether a stress- In vitro ¯uorogenic caspase cleavage assay Caspase activity was measured by a ¯uorometric assay. Brie¯y, speci®c calpain activation pathway exists, and examined its 1 potential role in cleaving endogenous substrate and pro- MN9D/Neo cells treated with 1 mm STS or 50 mm MPP were lysed in a buffer containing 50 mm Tris, pH 7.0/2 mm EDTA/1.0% moting cell death in a murine mesencephalon-derived Triton X-100. Cell lysates (10 mg) recovered after centrifugation at dopaminergic neuronal cell line, MN9D (Choi et al. 1991, 13 000 g were incubated with 25 mm acetyl-Asp-Glu-Val-Asp-7- 1992; Heller et al. 1996). Our results suggest that (i) calpain amino-4-methylcoumarin (Ac-DEVD-AMC, a substrate for cas- is activated and the major protease cleaving Bax protein in pase-3 as well as caspase-6, -7, -8, and -10; Calbiochem) for 1 h at 1 staurosporine (STS)-induced apoptosis and MPP -induced 378C in the presence of a caspase reaction buffer containing necrosis; (ii) calpain activation is either dependent or 100 mm HEPES pH 7.4, 10% sucrose, 5 mm dithiothreitol, 0.1% independent of caspase activity; (iii) relative contribution CHAPS. Proteolysis of the ¯uorescent peptides was measured with of calpain to drug-induced cell death and to Bax cleavage excitation at 380 nm and emission at 460 nm using a ¯uorescence varies between these two pathways; and (iv) over- spectrophotometer (Hitachi, Ibaraki, Japan). expression of Bcl-2 prevents both drug-induced cell death Immunoblot analysis and Bax cleavage by primarily blocking caspase or calpain For immunoblot analysis of Bax, approximately 1 106 cells from  activity. MN9D/Neo, MN9D/Bcl-2, or MN9D/P35 plated on P-100 culture

q 2001 International Society for Neurochemistry, Journal of Neurochemistry, 77, 1531±1541 Calpain activation in neuronal death 1533 dishes were maintained in CCM for 3 days, switched to N2 medium formazan grains was measured at 540 nm using a microplate reader and treated with drugs in the presence or absence of caspase or (Molecular Devices, Palo Alto). Values from each treatment were calpain inhibitors for the indicated time periods. Cells were then calculated as a percentage relative to the untreated control (de®ned washed with ice-cold PBS and lysed on ice in a buffer containing as 100% survival). 50 mm Tris pH 7.0, 2 mm EDTA, 1.0% Triton X-100, 2 mm Nuclear staining with Hoechst 33258 and propidium iodide PMSF, 10 mg/mL aprotinin and leupeptin for 10 min. Lysates were MN9D/Neo or MN9D/Bcl-2 cells plated on 25 mg/mL poly-d- subsequently centrifuged at 13 000 g for 15 min at 48C. Protein lysine-coated 12-well plates were maintained in CCM for 3 days, content was measured using the Bio-Rad (Hercules, CA, USA) switched to N2 medium, and exposed to 1 mm STS for the indicated protein assay kit. For detecting cytochrome c release into the time periods. Cells were then ®xed with 4% paraformaldehyde, cytosolic fraction and Bax cleaved in the subcellular compartments, permeablized with ice-cold ethanol and stained with 50 ng/mL cells were treated with 1 mm STS or 50 mm MPP1 for the indicated Hoechst 33258 as described previously (Jacobson and Raff 1995). time periods, lysed and subsequently fractionated as previously The nuclei of necrotic cells were stained with propidium iodide as described with minor modi®cations (Gross et al. 1998). Equal described with minor modi®cations (Vermes et al. 1995; van amounts of soluble proteins (approximately 20±50 mg) were separ- Engeland et al. 1998). In brief, MN9D/Neo or MN9D/Bcl-2 cells ated on 10% sodium dodecyl sulfate±polyacrylamide gel electro- were treated with 50 mm MPP1 for the indicated time periods and phoresis (SDS±PAGE) gels and transblotted onto prewetted washed with a buffer containing 10 mm HEPES, 150 mm NaCl, polyvinylidene di¯uoride (PVDF)-nitrocellulose ®lters (Bio-Rad). 5mmKCl, 1 mm MgCl , 1.8 mm CaCl , pH 7.4 for 10±15 min. Primary antibodies used were mouse monoclonal anticytochrome c 2 2 Un®xed cells were then stained with propidium iodide at a ®nal (1 : 3000; Pharmingen, San Diego, CA, USA), mouse monoclonal concentration of 2 mg/mL for 20 min. Following several washes, anti-poly(ADP-ribose) polymerase (PARP; 1 : 5000; Enzyme the staining pattern of nuclei was observed and photographed with Systems Products), rabbit polyclonal anti-Bax raised against a Zeiss Axiovert 100 ¯uorescent microscope. amino acid residues 43±61 of mouse Bax (1 : 3000; Krajewski et al. 1994), and rabbit polyclonal antibodies speci®c for 150-kDa calpain-proteolyzed fodrin fragment (1 : 500), 80-kDa m-calpain Results and 76-kDa postautolytic m-calpain (1 : 600; Shields et al. 1999). Detection of speci®c bands was by enhanced chemiluminescence Characterization of STS- or MPP1-induced cell death in (ECL; Amersham Pharmacia Biotech, Piscataway, NJ, USA). MN9D dopaminergic neuronal cells We previously suggested that, in MN9D/Neo cells, 1 mm In vitro cleavage assay STS induced apoptosis over 24 h while the dopaminergic Bax, interleukin-1-b and PARP were prepared by a coupled in 1 35 neurotoxin, MPP (10±100 mm), largely induces necrosis vitro transcription/translation in the presence of [ S]methionine 1 (Amersham) according to the manufacturer's recommendations. To over 48 h (lower than 10 mm MPP did not induce any sign prepare caspase-1 and -3 extracts, bacterial plasmids expressing the of cell death within 2±3 days (Oh et al. 1995, 1997; Choi corresponding caspases were introduced into BL21(DE3) cells. et al. 1999; Kim et al. 1999). Here, several morphological Exponentially growing cells were incubated with 0.2 mm iso- and biochemical assays were conducted to further determine propyl-1-thio-a-d-galactopyranoside for 2 h, harvested, and lysed the mode of cell death and its association with caspase. As by sonication. The lysates were cleared by centrifugation, and the shown in Fig. 1(b), in MN9D/Neo cells, 1 mm STS induced protein contents were measured using the Bio-Rad protein assay perinuclear chromatin condensation while it did not induce kit. The puri®ed m-calpain and m-calpain were purchased from any discernible mitochondrial swelling. By contrast, 50 mm Calbiochem. In vitro cleavage reactions were performed in a buffer MPP1 induced dramatic mitochondrial swelling and irregu- containing 0.5% Nonidet P-40/20 mm HEPES, pH 7.4/100 mm larly scattered heterochromatin typical of necrosis as early NaCl/20 mm dithiothreitol for 1 h at 378C (for caspase-1 or -3) or in a calpain reaction buffer provided by the manufacturer in a as 20 h after addition and continued to do so thereafter total reaction volume of 30 mL. Approximately 20±25 mg of each (Figs 1c and d). In these cell death paradigms, release of caspase lysate or 10 mL of each puri®ed calpain were included. cytochrome c into the cytosolic fraction was detected as Following incubation, reactions were terminated by the addition of early as 2 h after STS treatment (Fig. 2a). Cytochrome c 2 sample buffer followed by boiling for 5 min. Products of the release was not observed for up to 30±34 h in MPP1- Â cleavage reactions were separated on 10% SDS±PAGE prior to induced cell death. To assess activation of caspase following drying and autoradiography. release of cytochrome c into the cytosol, we ®rst assayed the proteolytic activity of caspase using a ¯uorogenic substrate, MTT reduction assay Ac-DEVD-AMC after either 1 mm STS or 50 mm MPP1 Following incubation with various experimental reagents, the rate treatment. As shown in Fig. 2(b), caspase activity was ele- of cell survival was assessed by colorimetric measurement of 3- vated as early as 3±6 h after STS treatment and continued to [4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) reduction as described previously (Shearman et al. 1994). In brief, increase up to 40±50-fold over the untreated control MN9D/ after the indicated incubation times, the MTT solution was added to Neo cells. This activity remained virtually unaltered follow- 1 a ®nal concentration of 1 mg/mL. The cells were then incubated for ing MPP treatment for up to 48 h, however. Similar patterns 2 h at 378C, followed by lysis in 20% SDS in 50% aqueous were observed using a general caspase substrate, acetyl-Val- dimethylformamide for 24 h. The optical density of the dissolved Ala-Asp-7-amino-tri¯uoromethylcoumarin (Ac-VAD-AFC;

q 2001 International Society for Neurochemistry, Journal of Neurochemistry, 77, 1531±1541 1534 W.-S. Choi et al.

Fig. 1 Ultrastructural changes in MN9D dopaminergic neuronal cells treated with STS or MPP1. MN9D/Neo cells were treated 1 with 1 mM STS or 50 mM MPP , processed for electron microscopy and photographed under a Zeiss EM 902 A transmission elec- tron microscope as described in Materials and methods. (a) Untreated MN9D/Neo cells, (b) STS-treated cells at 12 h, (c) MPP1-treated cells at 20 h, and (d) MPP1- treated cells at 30±36 h. Note perinuclear chromatin condensation and largely intact mitochondria in STS-treated cells. In con- trast, MPP1-treated cells exhibited the swollen mitochondria (arrowhead) and irre- gularly scattered heterochromatin (arrow) typical of necrosis. Bar 0.6 mm. ˆ not shown). Proteolytic cleavage of a well-known endogen- (Fig. 3b). By contrast, the 18-kDa Bax fragment was ous substrate of caspase, PARP into an 85-kDa fragment exclusively detected in the cytosolic fraction of MPP1- was observed as early as 8±12 h following STS treatment treated cells (Fig. 3b). From subsequent immunoblot and continued up to 20±24 h (Fig. 2c). By contrast, this analysis, m-calpain was found to be located both in the 85-kDa fragment of PARP was not detected even at the heavy membrane fraction and in the cytosolic fraction of later stages of MPP1-induced cell death (Fig. 2c). Taken MN9D/Neo cells (Fig. 3c). By contrast, m-calpain was together, these data further support the idea that STS induces exclusively present in the cytosolic fraction of MN9D/Neo a caspase-dependent apoptosis while MPP1 largely induces cells (Fig. 3c). a caspase-independent necrosis in MN9D/Neo cells. Identi®cation of protease that cleaves Bax in STS- or Cleavage of Bax in STS- and MPP1-induced neuronal MPP1-induced neuronal cell death cell death To ®rst characterize the protease responsible for Bax We have previously shown that Bax was cleaved into an cleavage into an 18-kDa fragment, [35S]methionine-labeled 18-kDa form at least a few hours after caspase-mediated Bax was ®rst incubated with several proteases. Reaction PARP cleavage during STS-induced cell death (Kim et al. mixtures were then separated on SDS±PAGE and analyzed 1999). In the present study, this 18-kDa form of the Bax by autoradiography. Neither caspase-1 nor -3 had any protein appeared not only in the STS-induced, caspase- effects on the appearance of 18-kDa Bax while caspase-1 dependent form of cell death but also in the MPP1-induced, and -3 cleaved interleukin-1-b and PARP, respectively caspase-independent cell death paradigm (Fig. 3a). In both (Figs 4a and b). By contrast, both m- and m-calpain were STS- and MPP1-induced cell death, the 18-kDa Bax frag- able to effectively cleave a full-length of 35S-labeled Bax ment appeared at the later stage of cell death. This fragment into an 18-kDa fragment (Fig. 4b). To further con®rm the was detected approximately 15±24 h, or 30±40 h following protease responsible for Bax cleavage in culture, MN9D/ STS or MPP1 treatment, respectively. Intriguingly, this Neo cells were treated with 1 mm STS for 18 h or 50 mm 18-kDa fragment appeared only in the mitochondria-enriched MPP1 for 40 h in the presence or absence of either a pan- heavy membrane fraction of STS-treated cells while a full- caspase (100 mm BAF or Z-VAD) or calpain inhibitor length Bax in the cytosolic fraction decreased over time (25 mm calpeptin; CPT). The proteolytic cleavage of Bax

q 2001 International Society for Neurochemistry, Journal of Neurochemistry, 77, 1531±1541 Calpain activation in neuronal death 1535

Fig. 3 Time course of Bax cleavage in drug-treated cells. (a) 1 MN9D/Neo cells were treated with 1 mM STS or 50 mM MPP for the indicated time periods. Subsequently, cells were harvested, lysed and processed for immunoblot analysis. Speci®c bands of Bax were Fig. 2 Time course of cytochrome c release and caspase activation detected using rabbit polyclonal antibodies raised against amino acid in drug-treated cells. MN9D/Neo cells were treated with 1 mM STS or residues 43±61 of mouse Bax (1 : 3000) in whole cell lysates. (b) 1 50 mM MPP for the indicated time periods and lysed. Subsequently, Immunoblot analysis for Bax was performed in the cytosolic, soluble lysates were processed for (a) immunoblot analysis of cytochrome c (S) or the mitochondria-enriched heavy membrane (HM) extracts 1 (Cyt c) released into the cytosolic fraction, (b) in vitro ¯uorogenic from MN9D/Neo cells treated with 1 mM STS (left) or 50 mM MPP caspase cleavage assay, and (c) immunoblot analysis of PARP clea- (right) for the indicated time periods. Untreated control cells were vage as described in Materials and methods. Primary antibodies maintained in drug-free N2 medium. (c) Immunoblot analysis was used were mouse monoclonal anti-cytochrome c (1 : 3000) and performed in the cytosolic (S) and the heavy membrane (HM) frac- mouse monoclonal anti-PARP (1 : 5000). For ¯uorogenic caspase tions from MN9D/Neo cells for (left) m-calpain or (right) m-calpain. cleavage assay, equal amounts of cell lysates harvested from (W) STS-or (X) MPP1-treated cells for the indicated time periods were 1 incubated with the ¯uorogenic peptide substrate, Ac-DEVD-AMC treated with 1 mm STS for 20±24 h or 50 mm MPP for (25 mM) for 1 h at 378C. Data were expressed as a fold increase 40 h in the presence or absence of the indicated protease relative to untreated control. Each point represents mean from tripli- inhibitor. As expected, cotreatment with a caspase inhibitor cate wells from one experiment representative of three similar (100 mm BAF or Z-VAD) or overexpression of p35 greatly repeats. attenuated STS-induced cell death (Fig. 5a). However, cotreatment with a wide concentration ranges of calpain produced by treatment with STS was completely prevented inhibitors, CPT or MDL 28170 (1±100 mm), did not have in the presence of either calpain or caspase inhibitors any protective effect on STS-induced cell death. In contrast, (Fig. 4c). However, MPP1-induced Bax cleavage was cotreatment with either 25 mm CPT or MDL 28170 attenuated MPP1-induced cell death while inhibition of completely blocked in the presence of CPT but not by 1 BAF or Z-VAD (Fig. 4d). Consequently, overexpression caspase activity failed to prevent MPP -induced cell death of the baculovirus-derived caspase inhibitor protein, p35 (Fig. 5b). (Bump et al. 1995) in MN9D cells blocked the appearance of the 18-kDa Bax fragment in response to STS but not Role of Bcl-2 in STS- and MPP1-induced cell death MPP1 treatment (Fig. 4e). These data suggest that calpain is and its associated Bax cleavage the major protease cleaving Bax protein in MN9D/Neo cells Bcl-2 is known to block apoptosis induced by a wide variety following treatment with either STS or MPP1, and that of stresses and necrosis in certain cases (Kane et al. 1995; calpain activation may be either dependent or independent Thompson 1995; Shimizu et al. 1996; Tsujimoto et al. of caspase activity. 1997). To further evaluate the protective role of Bcl-2 in both STS- and MPP1-induced cell death, we examined the Role of calpain activation in neuronal cell death appearance of apoptotic nuclei with Hoechst dye and To investigate whether caspase-dependent or -independent necrotic nuclei with propidium iodide in both MN9D/Neo calpain activation is associated with cell death, cells were and MN9D/Bcl-2 cells. Hoechst dye staining revealed

q 2001 International Society for Neurochemistry, Journal of Neurochemistry, 77, 1531±1541 1536 W.-S. Choi et al.

1 Fig. 4 Characterization of caspase-dependent or -independent cal- MPP for 40 h in the presence or absence of 100 mM BAF (caspase pain activation that induces Bax cleavage. (a) Interleukin-1-b (IL-1b) inhibitor), 100 mM Z-VAD (caspase inhibitor), or 25 mM calpeptin and PARP were translated in vitro in the presence of [35S]methio- (CPT, calpain inhibitor). Speci®c bands of Bax were detected in nine and incubated at 378C for 1 h with bacterially expressed whole cell lysates. (e) Both MN9D/Neo and MN9D cells overexpres- caspase-1 and 23 as described in Materials and methods. (b) sing baculovirus p35 (MN9D/P35) characterized by RT-PCR (left) 35 1 [ S]methionine-labeled Bax was also incubated with one of the indi- were treated with 1 mM STS for 20 h or 50 mM MPP for 40 h. Sub- cated proteases at 378C for 1 h. Following incubation, reaction mix- sequently, whole cell lysates were processed for immunoblot analy- tures were separated on 10% SDS±PAGE prior to autoradiography. sis to detect Bax (right). Untreated control cells were maintained in Arrows indicate intact proteins and cleaved forms in each case. (c, d) serum-free N2 medium. MN9D/Neo cells were treated with 1 mM STS for 18 h or 50 mM condensation and fragmentation of chromatin in MN9D/Neo Bcl-2 in MN9D cells prevented both STS-and MPP1- cells 12 h after 1 mm STS treatment (Fig. 6d). However, induced Bax cleavage (Figs 8a and b). As shown in Fig. 8(c these changes were not detected in STS-treated MN9D/Bcl- and d), overexpression of Bcl-2 also inhibited drug-induced 2 cells or in untreated control cells (Figs 6b, f and h). These generation of 150-kDa calpain-proteolyzed fodrin frag- nuclear changes detected by Hoechst dye staining were not ments, suggesting that Bcl-2 in MN9D cells may block observed in MPP1-treated MN9D/Neo cells (Oh et al. calpain activity following drug treatment. 1995). By contrast, the majority of MN9D/Neo cells bearing a distruption in their membrane integrity were stained red by Discussion propidium iodide (PI) 30±32 h after 50 mm MPP1 treat- ment; no signs of PI-positive nuclei were detected in MN9D/ Our ®ndings suggest that the activation of calpain occurs in Bcl-2 cells (Figs 7d and h). PI-positive nuclei were not both STS-induced apoptosis and MPP1-induced necrosis in detected in STS-treated MN9D/Neo cells (not shown). In MN9D dopaminergic neuronal cells. This activation then accord with these morphological data, overexpression of seems to result in a speci®c cleavage of Bax localized within Bcl-2 attenuated STS-induced cell death (25.9% survival the heavy membrane fraction enriched with mitochondria or in MN9D/Neo cells versus 73.9% in MN9D/Bcl-2 cells within the cytosolic fraction, respectively. Furthermore, following 1 mm STS treatment, n 3) and, to a lesser calpain-mediated cleavage of Bax into the 18-kDa fragment ˆ extent, MPP1-induced cell death (30.5% survival in MN9D/ can be either caspase-dependent or caspase-independent Neo cells versus 59.5% in MN9D/Bcl-2 cells following according to the type of stress applied. In the caspase- 50 mm MPP1 treatment, n 3) as determined by the MTT dependent paradigm of calpain activation induced by STS, ˆ reduction assay. Co-treatment of MN9D/Bcl-2 cells with inhibition of calpain activity blocks Bax cleavage but not 25 mm CPT did not signi®cantly alter the cell death kinetics drug-induced cell death, indicating that calpain activity following MPP1 treatment (60.6% survival in MN9D/Bcl-2 responsible for Bax cleavage may not be a major factor cells, n 3), suggesting that protective role of Bcl-2 in contributing to the STS-induced apoptosis. By contrast, ˆ MPP1-induced cell death primarily reside in its anti-calpain cotreatment with calpain inhibitors not only prevents Bax activity. In these cell death paradigms, overexpression of cleavage, but also rescues cells from MPP1-induced death.

q 2001 International Society for Neurochemistry, Journal of Neurochemistry, 77, 1531±1541 Calpain activation in neuronal death 1537

following STS treatment, it seems likely that a full-length Bax in the cytosol is ®rst translocated into the mitochondria before it is cleaved there. Thus, the data presented here seem to provide a new evidence that cleavage of Bax following translocation may comprise an alternative mode of Bax processing in certain cases of apoptotic death. Like STS- induced Bax cleavage shown here, secondly, other caspase- dependent cell death paradigms evoked by such drugs as etoposide (a topoisomerase II inhibitor) and nigericin (a potassium-proton ionophore) also induce Bax cleavage in MN9D/Neo cells (Kim et al. unpublished data). This is not universal, however, since the caspase-dependent cell death induced by another dopaminergic neurotoxin, 6-hydroxy- dopamine in MN9D/Neo cells (Oh et al. 1998; Choi et al. 1999) does not lead to calpain-mediated Bax cleavage (not shown). Taken together, these ®ndings raise the possibility that not all caspase-dependent apoptosis is coupled to Bax cleavage, and that the pattern of Bax processing during cell death is both cell- and stress-type speci®c. Although a mode of MPP1-induced cell death still remains to be unambiguously resolved, recent evidence, including data from our lab, indicates that MPP1 does not seem to induce apoptosis at least in dopaminergic neurons derived from the mesencephalon in culture (Choi et al. 1999; Lotharius et al. 1999). Here, we have further demonstrated Fig. 5 A distinct role of caspase or calpain activation during drug- that MPP1 in MN9D/Neo cells does not induce any signs of induced cell death. Both MN9D/Neo and MN9D/P35 cells were 1 caspase-dependent apoptotic pathway as determined by mM mM treated with (a) 1 STS for 20±24 h or (b) 50 MPP for 40 h several biochemical criteria including cytochrome c release in the presence or absence of several protease inhibitors as into the cytosol, and cleavage of ¯uorogenic caspase described in Fig. 4. Another calpain inhibitor, MDL 28170 (MDL, substrates and an endogenous cellular substrate, PARP. In 25 mM) was also included. Viability was assessed by the MTT reduc- tion assay. Values from each treatment group were expressed as a this model, cleavage of Bax into the 18-kDa form is detected 1 percentage relative to untreated control (100%). Data from three in the cytosolic fraction 30±40 h after MPP treatment independent cultures in triplicate were expressed as the mean without any discernible translocation of the cytosolic Bax ^ SEM. Signi®cance of difference was determined with a t-test into the mitochondria (see Fig. 3b). Appearance of this (*p , 0.02, **p , 0.002). fragment can be completely blocked by an inhibition of calpain activity but not by an inhibition of caspase activity. Thus, the relative contribution of calpain to drug-induced Thus, our present study provides an evidence for the ®rst cell death and to Bax cleavage may vary between these two time that calpain-mediated Bax cleavage can be triggered pathways. In these paradigms of cell death, overexpression not only in some cases of a caspase-dependent apoptosis but of Bcl-2 in MN9D cells prevents both cell death and Bax also in a certain case of a caspase-independent necrosis. cleavage induced by STS or MPP1 treatment. Therefore, it We have demonstrated that Bax cleavage by activated is likely that Bcl-2 may play a protective role by primarily calpain occurs in the heavy membrane fraction enriched blocking caspase or calpain activity in STS- or MPP1- with mitochondria from STS-treated cells. By contrast, in induced paradigm of cell death, respectively. MPP1-treated cells, this phenomenon was observed exclu- Previous ®ndings by others suggest that translocation of sively in the cytosolic fraction. It is intriguing to postulate Bax from the cytosol to the mitochondria is a critical event how calpain, activated in response to drugs, ®rst recognizes in apoptosis of various cell types including neurons (Wolter and then exclusively cleaves Bax present within a particular et al. 1997; Gross et al. 1998; Putcha et al. 1999). On the compartment. Others have demonstrated that m- and other hand, recent evidence also indicates that Bax protein m-calpain differ in their binding af®nity for calcium and is cleaved in virus-infected ®broblasts or in apoptosis of in their subcellular localization within the brain (Siman et al. HL-60 cells treated with topoisomerase I inhibitor (Grand- 1983; Baudry et al. 1987; Sorimachi et al. 1997). According girard et al. 1998; Wood et al. 1998; Wood and Newcomb to these and other reports, with increased levels of intra- 1999). Since the intensity of a full-length Bax in the cellular free calcium, these proteases appear to be trans- cytosolic fraction decreased over time in MN9D/Neo cells located to cellular membranes where they may degrade

q 2001 International Society for Neurochemistry, Journal of Neurochemistry, 77, 1531±1541 1538 W.-S. Choi et al.

Fig. 6 Protective role of Bcl-2 in STS-induced apoptotic cell photomicrographs were obtained from (a and e) untreated and death. (a±d) MN9D/Neo cells and (e±h) MN9D cells overexpressing (c and g) STS-treated cells. Photomicrographs for nuclear staining Bcl-2 (MN9D/Bcl-2) were treated with 1 mM STS for 12 h, stained of the same microscopic ®elds were obtained from (b and f ) with 50 ng/mL Hoechst 33258 as described in Materials and untreated and (d and h) STS-treated cells. Arrowheads indicate the methods, and subsequently photographed under a Zeiss Axiovert condensed or fragmented chromatin in STS-treated MN9D/Neo 100 microscope equipped with epi¯uorescence. Untreated control cells. Bar 50 mm. ˆ cells were maintained in serum-free N2 medium. Phase-contrast membrane proteins or may be freed to the cytosol where local concentration of the intracellular calcium surge alone they cleave cytosolic substrates. Although it is highly or in combination may contribute to the activation of a speculative at present, therefore it appears likely that such speci®c pool or type of calpain that eventually leads to factors as timing, the cellular site of calcium release, and the cleavage of Bax present only in a particular compartment.

Fig. 7 Protective role of Bcl-2 in MPP1-induced necrotic cell death. (a and e) untreated and (c and g) MPP1-treated cells. Photomicro- (a±d) MN9D/Neo cells and (e±h) MN9D/Bcl-2 cells were treated graphs for nuclear staining of necrotic cells in the same microscopic 1 with 50 mM MPP for 30±32 h. Un®xed cells were then incubated ®elds were obtained from (b and f ) untreated cells maintained in with 2 mg/mL propidium iodide as described in Materials and serum-free N2 medium and (d and h) MPP1-treated cells. Note that methods. After several washes, cells were examined and photo- propidium iodide-positive nuclei appear in almost all of the MPP1- graphed under a Zeiss Axiovert microscope equipped with epi¯uor- treated MN9D/Neo cells. Bar 50 mm. ˆ escence. Phase-contrast photomicrographs were obtained from

q 2001 International Society for Neurochemistry, Journal of Neurochemistry, 77, 1531±1541 Calpain activation in neuronal death 1539

cytotoxicity in human embryonic 293T cells (Wood and Newcomb 2000), another possibility could be that cleav- age of Bax somehow alters the channel-forming activity of Bax and its interactive capacity with other cell death- regulating proteins (Antonsson et al. 1997; Schlesinger et al. 1997; Jurgensmeier et al. 1998; Shimizu et al. 1999). Although any de®nitive proof of this hypothesis and a direct connection between calpain-mediated Bax cleavage and cell death in certain cases remains to be determined, this might be ascribed to an active process adapted by dying cells in favor of the completion of cell death by serving to seal the ultimate fate of those cells. A number of previous studies have demonstrated that caspase and calpain alone or in combination contribute to many of the in vivo and in vitro models of neuronal cell death (Chan and Mattson 1999; Wang 2000). Much work remains in order to explore fully the molecular mechanisms governed by these proteases during neuronal apoptosis and necrosis. However, future progress in this ®eld may further strengthen the rationale for using inhibitors of these Fig. 8 Protective role of Bcl-2 in drug-induced Bax cleavage. Both proteases as potential treatment modalities of use in various MN9D/Neo and MN9D/Bcl-2 cells were treated with (a and c) 1 mM neurodegenerative diseases, including Parkinson's disease. 1 STS or (b and d) 50 mM MPP for the indicated time periods. Equal amounts of whole cell lysates for each case were subjected to immunoblot analysis for (a and b) Bax or (c and d) calpain-proteo- Acknowledgements lyzed fodrin fragments (t-Fodrin). Note that Bcl-2 blocks both STS- The authors gratefully acknowledge Dr A. Heller for providing 1 and MPP -induced Bax cleavage into an 18-kDa fragment. Bcl-2 us with the MN9D cell line. We thank Drs J.C. Reed and also completely inhibits appearance of 150-kDa calpain-proteolyzed S. Krajewski for providing polyclonal anti-mouse Bax anti- fodrin fragments following drug treatment. bodies. We are also indebted to Dr G.J. Markelonis for critical reading of this manuscript. This work was supported by the KOSEF through the Brain Disease Research Center at Ajou Clearly, the distinct mechanisms whereby calpain is acti- University, KOSEF 1999±2-207±007±3, HMP-00-CH-13±0012, vated in each of these paradigms of cell death remain to be FG-4-04 to YJO, and by the Brain Research Program, MOST. thoroughly determined in the future. In addition to the transcriptional control of expres- sion, such post-translational modi®cations as phosphoryla- References tion and proteolysis of members of the Bcl-2 family have Antonsson B., Conti F., Ciavatta A., Montessuit S., Lewis S., Martinou emerged as important mechanisms for the regulation of cell I., Bernasconi L., Bernard A., Mermod J. J., Mazzei G., Maundrell death induced by a wide variety of stresses (Fadeel et al. K., Gambale F., Sadoul R. and Martinou J. C. (1997) Inhibition of 1999). However, it is not known whether and/or how the Bax channel-forming activity by Bcl-2. Science 277, 370±372. proteolytic cleavage of Bax into the 18-kDa protein by Baudry M., DuBrin R. and Lynch G. (1987) Subcellular compartmen- talization of calcium-dependent and calcium- independent neutral calpain affects its known cell death-promoting function in proteases in brain. Synapse 1, 506±511. neuronal cells. Interestingly, unlike many of the post- Bump N. J., Hackett M., Hugunin M., Seshagiri S., Brady K., Chen P., translational modi®cations of Bcl-2 family members, the Ferenz C., Franklin S., Ghayur T. and Li P. (1995) Inhibition of proteolytic cleavage of Bax in MN9D/Neo cells occurs at ICE family proteases by baculovirus antiapoptotic protein p35. the later stage of cell death. The simplest assumption Science 269, 1885±1888. Chan S. L. and Mattson M. P. (1999) Caspase and calpain substrates: therefore is that the appearance of cleaved Bax at the later roles in synaptic plasticity and cell death. J. Neurosci. Res. 58, stage of cell death could be a secondary event having no 167±190. functional signi®cance in neuronal cells. On the other hand, Choi H. K., Won L. A., Kontur P. J., Hammond D. N., Fox A. P., Bax cleavage does not seem to be the results of bulk Wainer B. H., Hoffmann P. C. and Heller A. (1991) Immortaliza- degradation of mitochondrial or cytosolic proteins during tion of embryonic mesencephalic dopaminergic neurons by drug-induced cell death as cytochrome oxidase subunit IV, somatic cell fusion. Brain Res. 552, 67±76. Choi H. K., Won L., Roback J. D., Wainer B. H. and Heller A. (1992) microtubule-associated protein 2 and p53 are not cleaved Speci®c modulation of dopamine expression in neuronal hybrid (not shown). Considering the recent suggestion that transient cells by primary cells from different brain regions. Proc. Natl expression of the Bax cleavage product exhibits enhanced Acad. Sci. USA 89, 8943±8947.

q 2001 International Society for Neurochemistry, Journal of Neurochemistry, 77, 1531±1541 1540 W.-S. Choi et al.

Choi W. S., Yoon S. Y., Oh T. H., Choi E. J., O'Malley K. L. and Oh Nixon R. A. (2000) A `protease activation cascade' in the pathogenesis Y. J. (1999) Two distinct mechanisms are involved in of Alzheimer's disease. Ann. NY Acad. Sci. 924, 117±131. 6-hydroxydopamine- and MPP1-induced dopaminergic neuronal Oh J. H., Choi W. S., Kim J. E., Seo J. W., O'Malley K. L. and Oh Y. J.

cell death: role of caspases, ROS, and JNK. J. Neurosci. Res. 57, (1998) Overexpression of HA-Bax but not Bcl-2 or Bcl-XL 86±94. attenuates 6-hydroxydopamine-induced neuronal apoptosis. Exp. Chua B. T., Guo K. and Li P. (2000) Direct cleavage by the calcium- Neurol. 154, 193±198. activated protease calpain can lead to inactivation of caspases. Oh Y. J., Wong S. C., Moffat M. and O'Malley K. L. (1995) Over- J. Biol. Chem. 275, 5131±5135. expression of Bcl-2 attenuates MPP1, but not 6-ODHA, induced Cryns V. and Yuan J. (1998) Proteases to die for. Dev. 12, cell death in a dopaminergic neuronal cell line. Neurobiol. Dis. 2, 1551±1570. 157±167. van Engeland M., Nieland L. J., Ramaekers F. C., Schutte B. and Oh Y. J., Uhland-Smith A., Kim J. E. and O'Malley K. L. (1997) Reutelingsperger C. P. (1998) Annexin V-af®nity assay: a review Regions outside of the Bcl-2 homology domains, BH1 and BH2 on an apoptosis detection system based on phosphatidylserine protect a dopaminergic neuronal cell line from staurosporine- exposure. Cytometry 31, 1±9. induced cell death. Brain Res. Mol. Brain Res. 51, 133±142. Fadeel B., Zhivotovsky B. and Orrenius S. (1999) All along the Porn-Ares M. I., Samali A. and Orrenius S. (1998) Cleavage of the watchtower: on the regulation of apoptosis regulators. FASEB J. calpain inhibitor, calpastatin, during apoptosis. Cell Death. Differ. 13, 1647±1657. 5, 1028±1033. Grandgirard D., Studer E., Monney L., Belser T., Fellay I., Borner C. Putcha G. V., Deshmukh M. and Johnson E. M. Jr (1999) Bax and Michel M. R. (1998) Alphaviruses induce apoptosis in Bcl-2- translocation is a critical event in neuronal apoptosis: regulation overexpressing cells: evidence for a caspase-mediated, proteolytic by neuroprotectants, Bcl-2, and caspases. J. Neurosci. 19, inactivation of Bcl-2. EMBO J. 17, 1268±1278. 7476±7485. Gross A., Jockel J., Wei M. C. and Korsmeyer S. J. (1998) Enforced Ruiz-Vela A., Gonzalez D. B. and Martinez A. (1999) Implication of dimerization of BAX results in its translocation, mitochondrial calpain in caspase activation during B cell clonal deletion. EMBO dysfunction and apoptosis. EMBO J. 17, 3878±3885. J. 18, 4988±4998. Heller A., Price S. and Won L. (1996) Glial-derived neurotrophic factor Saito K., Elce J. S., Hamos J. E. and Nixon R. A. (1993) Widespread (GDNF) induced morphological differentiation of an immortal- activation of calcium-activated neutral proteinase (calpain) in the ized monoclonal hybrid dopaminergic cell line of mesencephalic brain in Alzheimer disease: a potential molecular basis for neuronal origin. Brain Res. 725, 132±136. neuronal degeneration. Proc. Natl Acad. Sci. USA 90, 2628±2632. Jacobson M. D. and Raff M. C. (1995) and Schlesinger P. H., Gross A., Yin X. M., Yamamoto K., Saito M., Bcl-2 protection in very low oxygen. Nature 374, 814±816. Waksman G. and Korsmeyer S. J. (1997) Comparison of the ion Jurgensmeier J. M., Xie Z., Deveraux Q., Ellerby L., Bredesen D. and channel characteristics of proapoptotic BAX and antiapoptotic Reed J. C. (1998) Bax directly induces release of cytochrome c BCL-2. Proc. Natl Acad. Sci. USA 94, 11357±11362. from isolated mitochondria. Proc. Natl Acad. Sci. USA 95, Shearman M. S., Ragan C. I. and Iversen L. L. (1994) Inhibition of 4997±5002. PC12 cell redox activity is a speci®c, early indicator of the Kamp¯ A., Posmantur R. M., Zhao X., Schmutzhard E., Clifton G. L. mechanism of beta-amyloid-mediated cell death. Proc. Natl Acad. and Hayes R. L. (1997) Mechanisms of calpain proteolysis Sci. USA 91, 1470±1474. following traumatic brain injury: implications for pathology and therapy: implications for pathology and therapy: a review and Shields D. C., Schaecher K. E., Saido T. C. and Banik N. L. (1999) A update. J. Neurotrauma 14, 121±134. putative mechanism of demyelination in multiple sclerosis by a Kane D. J., Ord T., Anton R. and Bredesen D. E. (1995) Expression of proteolytic enzyme, calpain. Proc. Natl Acad. Sci. USA 96, Bcl-2 inhibits necrotic neural cell death. J. Neurosci. Res. 40, 11486±11491. 269±275. Shimizu S., Eguchi Y., Kamiike W., Waguri S., Uchiyama Y., Matsuda Kim J. E., Oh J. H., Choi W. S., Chang I. I., Sohn S., Krajewski S., Reed H. and Tsujimoto Y. (1996) Retardation of chemical hypoxia- J. C., O'Malley K. L. and Oh Y. J. (1999) Sequential cleavage of induced necrotic cell death by Bcl-2 and ICE inhibitors: possible poly (ADP-ribose) polymerase and appearance of a small Bax- involvement of common mediators in apoptotic and necrotic signal transductions. Oncogene 12, 2045±2050. immunoreactive protein are blocked by Bcl-XL and caspase inhibitors during staurosporine-induced dopaminergic neuronal Shimizu S., Narita M. and Tsujimoto Y. (1999) Bcl-2 family proteins apoptosis. J. Neurochem. 72, 2456±2463. regulate the release of apoptogenic cytochrome c by the mito- Krajewski S., Krajewska M., Shabaik A., Miyashita T., Wang H. G. and chondrial channel VDAC. Nature 399, 483±487. Reed J. C. (1994) Immunohistochemical determination of in vivo Siman R., Baudry M. and Lynch G. (1983) Puri®cation from synapto- distribution of Bax, a dominant inhibitor of Bcl-2. Am. J. Pathol. somal plasma membranes of calpain I, a thiol protease activated 145, 1323±1336. by micromolar calcium concentrations. J. Neurochem. 41, Lipton P. (1999) Ischemic cell death in brain neurons. Physiol. Rev. 79, 950±956. 1431±1568. Sorimachi H., Ishiura S. and Suzuki K. (1997) Structure and physio- Lotharius J., Dugan L. L. and O'Malley K. L. (1999) Distinct mech- logical function of calpains. Biochem. J. 328, 721±732. anisms underlie neurotoxin-mediated cell death in cultured Springer J. E., Azbill R. D., Kennedy S. E., George J. and Geddes J. W. dopaminergic neurons. J. Neurosci. 19, 1284±1293. (1997) Rapid calpain I activation and cytoskeletal protein Martin S. J. and Green D. R. (1995) Protease activation during degradation following traumatic spinal cord injury: Attenuation apoptosis: death by a thousand cuts? Cell 82, 349±352. with riluzole pretreatment. J. Neurochem. 69, 1592±1600. Mouatt-Prigent A., Karlsson J. O., Agid Y. and Hirsch E. C. (1996) Steiner H., Capell A., Pesold B., Citron M., Kloetzel P. M., Selkoe D. J., Increased m-calpain expression in the mesencephalon of patients Romig H., Mendla K. and Haass C. (1998) Expression of with Parkinson's disease but not in other neurodegenerative Alzheimer's disease-associated presenilin-1 is controlled by disorders involving the mesencephalon: a role in nerve cell death? proteolytic degradation and complex formation. J. Biol. Chem. Neuroscience 73, 979±987. 273, 32322±32331.

q 2001 International Society for Neurochemistry, Journal of Neurochemistry, 77, 1531±1541 Calpain activation in neuronal death 1541

Stracher A. (1999) Calpain inhibitors as therapeutic agents in nerve and Wolf B. B., Goldstein J. C., Stennicke H. R., Beere H., Amarante- muscle degeneration. Ann. NY Acad. Sci. 884, 52±59. Mendes G. P., Salvesen G. S. and Green D. R. (1999) Calpain Thompson C. B. (1995) Apoptosis in the pathogenesis and treatment of functions in a caspase-independent manner to promote apoptosis- disease. Science 267, 1456±1462. like events during platelet activation. Blood 94, 1683±1692. Thornberry N. A. and Lazebnik Y. (1998) Caspases: enemies within. Wolter K. G., Hsu Y. T., Smith C. L. A., Xi X. G. and Youle R. J. Science 281, 1312±1316. (1997) Movement of Bax from the cytosol to mitochondria during Tsujimoto Y., Shimizu S., Eguchi Y., Kamiike W. and Matsuda H. apoptosis. J. Cell Biol. 139, 1281±1292. (1997) Bcl-2 and Bcl-xL block apoptosis as well as necrosis: Wood D. E. and Newcomb E. W. (1999) Caspase-dependent activation possible involvement of common mediators in apoptotic and of calpain during drug-induced apoptosis. J. Biol. Chem. 274, necrotic signal transduction pathways. Leukemia 11, 380±382. 8309±8315. Vermes I., Haanen C., Steffens-Nakken H. and Reutelingsperger C. Wood D. E. and Newcomb E. W. (2000) Cleavage of Bax enhances its (1995) A novel assay for apoptosis. Flow cytometric detection of cell death function. Exp. Cell Res. 256, 375±382. phosphatidylserine expression on early apoptotic cells using Wood D. E., Thomas A., Devi L. A., Berman Y., Beavis R. C., Reed ¯uorescein-labelled Annexin V. J. Immunol. Methods 184, 39±51. J. C. and Newcomb E. W. (1998) Bax cleavage is mediated by Wang K. K. (2000) Calpain and caspase: can you tell the difference? calpain during drug-induced apoptosis. Oncogene 17, 1069±1078. Trends Neurosci. 23, 20±26. Zhang L., Song L. and Parker E. M. (1999) Calpain inhibitor I increases Wang K. K., Posmantur R., Nadimpalli R., Nath R., Mohan P., Nixon beta-amyloid peptide production by inhibiting the degradation of R. A., Talanian R. V., Keegan M., Herzog L. and Allen H. (1998) the substrate of gamma-secretase. Evidence that substrate avail- Caspase-mediated fragmentation of calpain inhibitor protein calpas- ability limits beta-amyloid peptide production. J. Biol. Chem. 274, tatin during apoptosis. Arch. Biochem. Biophys. 356, 187±196. 8966±8972.

q 2001 International Society for Neurochemistry, Journal of Neurochemistry, 77, 1531±1541