Coupling Endoplasmic Reticulum Stress to the Cell Death Program AN Apaf-1-INDEPENDENT INTRINSIC PATHWAY*

Accumulation of misfolded proteins and alterations in Ca 2 (cid:1) homeostasis in the endoplasmic reticulum (ER) causes ER stress and leads to cell death. However, the signal-transducing events that connect ER stress to cell death pathways are incompletely understood. To dis-cern the pathway by which ER stress-induced cell death proceeds, we performed studies on Apaf-1 (cid:2) / (cid:2) (null) fi broblasts that are known to be relatively resistant to apoptotic insults that induce the intrinsic apoptotic pathway. While these cells were resistant to cell death initiated by proapoptotic stimuli such as tamoxifen, they were susceptible to apoptosis induced by thapsigargin and brefeldin-A, both of which induce ER stress. This pathway was inhibited by catalytic mutants of caspase-12 and caspase-9 and by a peptide inhibitor of caspase-9 but not by caspase-8 inhibitors. Cleavage of caspases and poly(ADP-ribose) polymerase was observed in cell-free extracts lacking cytochrome c that were isolated from thapsigargin or brefeldin-treated cells. To define the molecular re-quirements for this Apaf-1 and cytochrome c -inde-pendent apoptosis pathway further, we developed a cell-free system of ER

The endoplasmic reticulum (ER) 1 is a principal site for protein synthesis and folding and also serves as a cellular storage site for calcium (1). Perturbation of Ca 2ϩ homeostasis, increased production of free radicals, inhibition of protein glycosylation, and accumulation of misfolded proteins in the ER can all elicit cellular stress responses, particularly ER stress signals, to protect cells against changes in Ca 2ϩ levels and toxic buildup of misfolded proteins (1)(2)(3). Prolonged ER stress leads to cell death and is linked to the pathogenesis of some neurodegenerative disorders that feature misfolded proteins, including Alzheimer's disease, Parkinson's disease, and amyotrophic lateral sclerosis (3,4). Activation of caspases, a family of cysteine proteases with aspartate P1 specificity, is a central mechanism in the apoptotic cell death process (5). The extrinsic pathway of caspase activation involves signal transduction through cellular death receptors such as Fas, resulting in caspase-8 activation, which in turn activates downstream effector caspases such as caspase-3 and caspase-7 (6). The intrinsic pathway involves release of the mitochondrial protein cytochrome c, which forms an oligomeric complex with dATP and Apaf-1 (7,8). It is this oligomeric complex that recruits procaspase-9 directly, activates it, and then releases active caspase-9 from the complex to set in motion the caspase-9-dependent activation of effector caspases such as caspase-3, -6, and -7 (6,8,9). Once active, the effector caspases cleave various cellular targets, including poly(ADP-ribose) polymerase (10) and other substrates (6), ultimately leading to cell death. Earlier studies have demonstrated a molecular link between ER stress and caspase-12 activation, resulting in increased cell death (11)(12)(13)(14)(15). However, the downstream targets of caspase-12 yet remain to be identified. Therefore, the question arose as to whether activation of downstream caspases and cell death following ER stress involves the previously described extrinsic pathway, the mitochondria-dependent intrinsic apoptotic pathway, or an alternate pathway (16,17).
In the present study, we investigated the mechanism of ER stress-induced activation of caspases and cell death in Sak2 cells, which are Apaf-1 null cells (18,19). Sak2 cells are generally more resistant to cell death initiated by various apoptotic agents, including those that utilize Fas or ceramide-mediated pathways (18,19). We demonstrate that treatment of Sak2 cells with thapsigargin or brefeldin-A induces ER stress, activation of caspase-12, and cell death. ER stress-induced caspase-12-mediated cell death proceeds in a caspase-9dependent pathway yet by a mechanism that is independent of the previously described intrinsic (mitochondria-dependent) apoptotic pathway (7,8). ER stress-induced cell death was inhibited by catalytic mutants of caspase-12 and caspase-9 and by the peptide inhibitor of caspase-9 but not by inhibition of caspase-8. We also developed a cell-free model of ER stressinduced cell death that involves the addition of microsomes to a 300,000 ϫ g cell-free extract that not only lacks Apaf-1 but also cytochrome c, both of which are required for activating caspase-9 and other downstream caspases through the previously described intrinsic apoptotic pathway. This system was capable of reproducing a key element of apoptosis, namely caspase processing and activation. Caspase-12 was identified as one of the microsomal components required for downstream caspase processing. Thus, ER stress-induced caspase-12 activation defines a novel, ER-based intrinsic pathway for apoptosome-independent effector caspase activation and cell death.
Cell-free Extracts, Cell Fractionation, and Western Blotting-Cellfree cytoplasmic extracts were prepared as previously described (14,20). The 16,000 ϫ g cytoplasmic extract lacked whole cells, nuclei, and mitochondria. The 16,000 ϫ g extract was recentrifuged at 300,000 ϫ g for 1 h for the preparation of microsomes. The resulting supernatant contained the soluble cytosolic fraction, and the pellet constituted the microsomal fraction. The 300,000 ϫ g microsomal fraction represents ER membrane, lumen proteins, and Golgi membranes. The microsomal pellet was washed twice with 10 mM Tris-Cl/NaCl, pH 7.4. The final microsomal pellet was reconstituted in the same buffer and briefly sonicated to disperse the pellet fraction. The purity of the microsomal fraction was assessed by the presence of protein-disulfide isomerase, an ER lumen protein. Equal amounts (50 -100 g of protein) of microsomes were added to a 300,000 ϫ g untreated cytosolic extract (150 -200 g of protein) from Sak2 cells and incubated at 36°C for 1 h.
Immunoprecipitation Assay-Microsomes were isolated from 16,000 ϫ g cell-free extracts prepared from thapsigargin-treated cells as described above. The microsomal pellet was washed twice with 10 mM Tris-Cl/NaCl, pH 7.4. The final microsomal pellet was reconstituted in the above buffer and briefly sonicated to disperse the pellet fraction. A total of 200 g of protein was subjected to immunoprecipitation using the anti-caspase-12 antibody and anti-caspase-7 monoclonal antibody (BD Pharmingen). Following an overnight incubation at 4°C with the antibodies, protein A/G-Sepharose (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) was added to the samples and incubated at 4°C for an additional 6 h. A similar aliquot of microsomal preparation was incubated only with protein A/G-Sepharose. Samples were spun briefly to pellet the protein A/G-Sepharose conjugate. The supernatant was collected and incubated with 300,000 ϫ g untreated cytosolic extract (150 -200 g of protein) from Sak2 cells at 36°C for 1 h. Samples were analyzed by Western blot analysis using the respective antibody.
Caspase Activity Assay and Evaluation of Apoptosis-The synthetic tetrapeptide substrates benzyloxycarbonyl-Ile-Glu(OMe)-Thr-Asp-(OMe)-FMK(IETD.fmk)andbenzyloxycarbonyl-Leu-Glu(OMe)-His-Asp-(OMe)-FMK (LEHD.fmk) were purchased from Enzyme Systems Products and dissolved in dimethyl formamide as a 10 mM stock solution. These tetrapeptides function as potent inhibitors of caspase-8 and caspase-9 activity. The fluorogenic substrates benzyloxycarbonyl-Asp-Glu-Val-Asp-7-amino-4-trifluoromethylcoumarin and benzyloxycarbonyl-Leu-Glu-His-Asp-7-amino-4-trifluoromethylcoumarin were purchased from Enzyme Systems Products and dissolved in dimethyl formamide as a 10 mM stock solution. Cell-free extracts (100 g of protein) made from untreated and thapsigargin-treated cells were incubated with 100 M peptide substrate. Caspase activity was determined by measuring the release of amino-4-trifluoromethylcoumarin from the synthetic substrates using continuous recording instruments as described earlier (21). Enzyme activities were analyzed using a SpectraMAX 340 plate reader (Molecular Devices) at excitation and emission wavelengths of 444 and 538 nm, respectively.
Control and treated cells were stained with Hoechst 33258 and quantified as previously described (14,20,22). Apoptotic cells were also monitored by Annexin staining using the ApoAlert Annexin-V apoptosis kit (CLONTECH). Annexin-V labeled with enhanced green fluorescent protein was used for staining the Sak2 cells. Apoptotic cells specifically stained by the Annexin-V-enhanced green fluorescent protein were examined using a Nikon Eclipse-800 epifluorescence microscope equipped with appropriate filters.

ER Stress Induces Apoptosis in Apaf-1 Null
Cells-Apaf-1, a mammalian homolog of CED-4 (23), binds to cytochrome c, dATP and caspase-9, leading to the activation of caspase-9 (7). Sak2 cells, which lack Apaf-1 protein expression (Fig. 1a), are known to be less susceptible to various apoptotic stimuli including ultraviolet radiation, etoposide, staurosporine, and cisplatinum (18,19). In order to test whether these cells are susceptible to ER stress-inducing agents, cells were treated with 500 nM thapsigargin or 2 M brefeldin-A for different time periods. As shown in Fig. 1b, while the cells were relatively resistant to 5 M tamoxifen treatment, exposure of cells to thapsigargin or brefeldin led to a time-dependent decrease in cell viability. In contrast, NIH3T3 mouse fibroblasts that express the Apaf-1 protein (Fig. 1a) were highly susceptible not only to the ER stress inducers but also to tamoxifen treatment (Fig. 1c).
Thapsigargin treatment of Sak2 cells resulted in apoptotic cell death as evidenced by membrane blebbing, nuclear condensation and fragmentation, and Annexin-V-positive staining (Fig. 2). Such phenotypic changes were less marked in cells treated with 5 M tamoxifen (Fig. 2, last column). A similar pattern of apoptotic cell death was observed with brefeldin treatment (data not shown). A high level of GRP78 protein expression is indicative of ER stress (24,25), and thapsigargin or brefeldin treatment of Sak2 cells resulted in the induction of GRP78 expression (Fig. 3a). No induction of GRP78 protein expression was observed in tamoxifen treated cells. It was previously demonstrated that caspase-12 is specifically involved in apoptosis that results from stress in the endoplasmic reticulum (11,12,14), and accordingly treatment of Sak2 cells with thapsigargin or brefeldin caused processing of caspase-12 as revealed by the presence of the 42-kDa fragment of caspase-12 (Fig. 3a). The 42-kDa fragment of caspase-12 was not observed in cells treated with tamoxifen (Fig. 3a). This suggests that while Sak2 cells are resistant to some proapoptotic agents, they are susceptible to ER stress-mediated cell death, which has been shown previously to be mediated by caspase-12 (11,12,14).
ER Stress Activates a Subset of Caspases-Since many apoptotic cell deaths proceed either through the receptor-mediated caspase-8 activation (extrinsic pathway) or through the mitochondria-mediated intrinsic pathway involving an Apaf-1, cytochrome c, and caspase-9 complex (the apoptosome), we proceeded to determine whether or not ER stress-mediated cell death involves either of these two pathways. Sak2 cells were treated with brefeldin for 24 h in the presence of specific inhibitors of caspase-8 (IETD.fmk) or caspase-9 (LEHD.fmk). As shown in Fig. 3b, the addition of LEHD.fmk suppressed cell death, whereas the addition of the caspase-8 inhibitor (IETD. fmk) showed no effect on cell death induction by brefeldin.
Similar results were seen following treatment with thapsigargin (data not shown). To complement the studies that utilized peptide inhibitors, Sak2 cells were transfected with dominant negative (catalytic) mutants of caspase-12, -9, and -8 and later exposed to thapsigargin for 24 h. Whereas expression of the dominant negative mutant of caspase-8 failed to inhibit thapsigargin-induced cell death, expression of the catalytic mutant of either caspase-12 or -9 blocked thapsigargin-induced cell death (Fig. 3c). Similar inhibitory profiles were seen following treatment with brefeldin (data not shown). The combined results suggest that ER stress-induced cell death proceeds through a caspase-8-independent, caspase-9-dependent pathway.
Requirement of Downstream Caspases in ER Stress-induced Cell Death-In order to determine which caspases are activated during ER stress-induced apoptosis, cell-free extracts from thapsigargin-treated Sak2 cells that lacked cytochrome c (Fig. 4a) were analyzed by Western blotting. As shown in the time course in Fig. 4a, thapsigargin treatment resulted in the processing of caspase-12, -9, -7, and -3. Processing of caspase-12 was seen by 24 h of thapsigargin treatment, with complete disappearance of the active caspase-12 bands at 36 h of treatment. Procaspase-9 is processed into a large active subunit and a small subunit by autocatalysis at Asp 315 (7,8,26). The antibody we employed recognized full-length (50 kDa) and amino-terminal cleaved product (39 kDa; prodomain ϩ large subunit) of caspase-9. As shown in Fig. 5b, complete processing of the pro form to the 39-kDa product occurred by 36 h of treatment, by which time caspase-12 was completely processed. Procaspase-7 (35 kDa) is first converted to a 32-kDa intermediate (prodomain ϩ large subunit), which is further processed into active subunits consisting of 20and 11-kDa forms (27,28). Activation of caspase-7 was very similar to caspase-12, with more of the active intermediate form present at 24 h of treatment and complete disappearance of the product at 36 h of treatment. Similarly, processing of caspase-3 occurs in a two-step manner, with the initial appearance of p28/ p12 active products, followed by accumulation of the mature p17/p12 subunits of the enzyme (29,30). Processing of the pro form of caspase-3 (32 kDa) to the p28 form was observed by 36 h of thapsigargin treatment. ER stress also resulted in cleavage of the known caspase substrates PARP and the inhibitor of caspaseactivated DNase (ICAD) (Fig. 4a and data not shown), both of which represent nuclear proteins. Although PARP cleavage seems to occur prior to caspase-3 processing (Fig. 4a), studies have demonstrated that PARP is also cleaved by caspase-7 and that caspase-7 activation occurs concomitantly with PARP cleavage (31). Similar results were seen following treatment with brefeldin (data not shown). Caspase activity measurements on the cell-free extracts were performed using Asp-Glu-Val-Asp-7amino-4-trifluoromethylcoumarin and Leu-Glu-His-Asp-7-ami-no-4-trifluoromethylcoumarin as substrates. As shown in Fig 4b, our results demonstrated that Sak2 cells, treated with thapsigargin, activated caspases and displayed both LEHD-ase and DEVD-ase activities. Thus, the above results suggest that activation of caspases in Sak2 cells following ER stress does not require release of cytochrome c into the cytosol. It is however possible that prolonged ER stress leading to an irreversible cell damage and demise may involve the concerted action of ER and mitochondria and the proapoptotic molecules associated with them (although in that case, it is not clear how the presence of cytochrome c could lead to caspase activation in the absence of Apaf-1).
Activation of Cell-free Apoptosis with Microsomes from ER Stress-induced Cell Extracts-In order to dissect the ER stressactivated apoptotic pathway, a cell-free system was developed for the study of cellular events involving ER stress. The system involved the addition of microsomes derived from untreated or ER stressed Sak2 cells to cytosolic (300,000 ϫ g supernatant) extracts derived from untreated Sak2 cells. The purity of the microsomes was assessed by staining for cytochrome c and protein-disulfide isomerase, an ER lumen protein (Fig. 5a). As shown in Fig. 5b, the addition of microsomes isolated from untreated Sak2 cells to the cytosolic extracts did not result in cleavage of caspase-9 or -3; however, the addition of microsomes isolated from thapsigargin-treated cell extracts to cytosolic extracts resulted in the processing of caspase-9 and -3. Whereas a reduction in the pro form of caspase-9 and -3 was observed in extracts incubated with microsomes isolated from Sak2 cells that had been treated with thapsigargin for 12 h, complete processing of the pro form of caspase-9 (from 50 kDa to the 39-kDa product) and caspase-3 (from 35 kDa to the 28-kDa product) occurred in extracts incubated with microsomes isolated from Sak2 cells that had been treated with thapsigargin for 24 h. Similar results were obtained following treatment with brefeldin (data not shown). These studies dem- onstrate that ER stress-induced microsomes activate cell extracts independent of Apaf-1 and mitochondria.
Earlier studies suggested that caspase-12 exists on the cytoplasmic side of the ER, and ER stress involves movement of caspase-7 to the ER surface (11,12,14). To determine whether caspase-12 or caspase-7 was required for processing of downstream caspases, we immunodepleted caspase-12 and caspase-7 from fragmented (sonicated) microsomes isolated from 24-h thapsigargin-treated cell extracts. In our studies, about 50% of caspase-12 still exists as a pro form (60 kDa) at 24 h of thapsigargin/brefeldin treatment of cells. Immunodepletion of caspase-12 and -7 from the extracts was achieved with the respective antibodies (data not shown). The immunodepleted microsomal fraction was incubated with untreated cytosolic extracts. As shown in Fig. 5c, whereas immunodepletion of caspase-7 failed to inhibit the cleavage of procaspase-9 and -3 by the microsomal fraction, immunodepletion of caspase-12 inhibited the cleavage of procaspase-9 and -3 by the microsomal fraction. These results suggest that caspase-12 present in the microsomal fraction is required for downstream caspase acti- FIG. 4. ER stress induces activation of caspases and PARP. Sak2 cells were treated with 0.5 M thapsigargin for the indicated times. Cell-free cytosolic extracts (150 g of protein) from thapsigargintreated cells were prepared as described under "Experimental Procedures" and analyzed by Western blot analysis. Membranes were probed with anti-caspase-12, anti-caspase-9, anti-caspase-7, anticaspase-3, anti-PARP, and anti-cytochrome c antibodies. All caspase antibodies were capable of detecting the pro and active forms. The cell-free cytosolic extracts lacked any detectable cytochrome c as shown. Lane 1 in the cytochrome c panel shows a whole cell extract to indicate the position of cytochrome c. Each Western blot is representative of three independent experiments. b, cell-free extracts (100 g of protein) made from untreated and thapsigargin-treated cells were assayed with the fluorogenic substrates benzyloxycarbonyl-Asp-Glu-Val-Asp-7-amino-4-trifluoromethylcoumarin (Z-DEVD-AFC) (measures caspase-3 activity) and benzyloxycarbonyl-Leu-Glu-His-Asp-7-amino-4-trifluoromethylcoumarin (Z-LEHD-AFC) (measures caspase-9 activity) and analyzed as described under "Experimental Procedures." FIG. 5. Activation of cell-free apoptosis with microsomes from ER stress-induced cell extracts. Sak2 cells were treated with 0.5 M thapsigargin for 12 or 24 h. Cell-free cytosolic extracts from untreated or thapsigargin-treated cells were prepared as described under "Experimental Procedures." Microsomes were prepared from cell-free cytosolic extracts as described under "Experimental Procedures." a, microsomes isolated from untreated and 0.5 M thapsigargin (12 and 24 h)-treated cell extracts lacked any detectable cytochrome c. Lane 1 is a whole cell extract to indicate the position of cytochrome c. The purity of the microsomal fraction was also assessed by the presence of protein-disulfide isomerase (PDI). b, microsomes (50 g of protein) from untreated or 0.5 M thapsigargin (thaps)-treated cell extracts were incubated with normal 300,000 ϫ g extracts (150 g of protein) from Sak2 cells at 36°C for 1 h. The 300,000 ϫ g cell-free extract lacks whole cells, nuclei, mitochondria, and microsomes. At the end of the reaction, samples were detected by Western blotting. Membranes were probed with anti-caspase-9 and anti-caspase-3 antibodies. c, microsomes isolated from 24-h thapsigargin-treated Sak2 cell extracts were immunodepleted of caspase-12 (C12ab) and caspase-7 (C7ab) as described under "Experimental Procedures." The immunodepleted microsomal fraction was later incubated with a normal 300,000 ϫ g extract at 36°C for 1 h. At the end of the reaction, samples were detected by Western blotting. Membranes were probed with anti-caspase-9 and anti-caspase-3 antibodies. vation (at least in this model system).
Similarity of Caspase-12-mediated Apoptosis to ER Stressinduced Apoptosis-If caspase-12 indeed mediates ER stressinduced apoptosis, then the biochemical profiles for apoptosis induced by caspase-12 and for ER stress should be similar (e.g. caspase-9 should be required, but not Apaf-1 or caspase-8). However, simple vectorial expression of full-length caspase-12 did not lead to apoptosis (Fig. 6 and Ref. 14). Expression of caspase-12 lacking its N-terminal prodomain p(⌬N)C12, however, readily induced apoptosis in Sak2 cells (Fig. 6). This effect was blocked by its catalytic mutant p(⌬N)C12DN and by a peptide inhibitor of caspase-9 but not by a peptide inhibitor of caspase-8. These results support the notion that caspase-12mediated apoptosis involves caspase-9 activation by a mechanism that does not require Apaf-1 (i.e. the same profile obtained for ER stress-induced apoptosis). We speculate that ER stress may induce processing and activation of caspase-12 either through calpain cleavage of caspase-12 (11) or via a GRP78/caspase-7 complex mediated mechanism (14,15). To determine whether caspase-12 could directly activate caspase-9, we established an in vitro assay involving the addition of in vitro translated p(⌬N)C12 and caspase-9. The combination of the two in vitro-translated proteins was not sufficient to induce caspase-9 cleavage (Fig. 7, lane 2). Whereas caspase-9 cleavage by in vitro translated p(⌬N)C12 was not observed in the presence of untreated 16,000 ϫ g extract, the addition of the two in vitro translated proteins to a 16,000 ϫ g extract isolated from Sak2 cells treated with thapsigargin for 12 h showed prominent cleavage of caspase-9 (Fig. 7, lane 5). These results suggest that active caspase-12 in combination with another ER stress-induced molecule(s) is required for caspase-9 processing and activation during ER stress-induced cell death. DISCUSSION Apoptosis typically proceeds through one of two general signaling pathways, namely the extrinsic apoptotic pathway or the intrinsic apoptotic pathway. In the former case, binding of specific death ligands to their receptors causes oligomerization of death receptors, resulting in recruitment of adaptor molecules involved in activation of caspase-8. In the latter case, when the mitochondrion receives appropriate apoptotic signals, cytochrome c is released into the cytosol (6,32). Earlier studies have shown that together with dATP and cytochrome c, Apaf-1 forms a multimeric complex that activates procaspase-9 (8,9). The formation of this complex occurs through a multistep process and serves as a key commitment step for activation of caspase-9 and downstream caspases. Caspase-9 activation and its release from the multimeric complex requires a fully functional Apaf-1 protein (8,9). Activated caspase-8 and caspase-9 in turn activate executioner caspases, including caspase-3. Cell death is thought to result from the proteolysis of cellular substrates by active caspase-3 and -7 (6,33,34). There has been some controversy over the role of mitochondria in apoptosis; it is believed by some that the primary apoptotic signals communicate directly with the cytosolic caspases, with the mitochondria contributing only a secondary role in the apoptotic process (35). Therefore, in addition to Apaf-1 and cytochrome c, alternate mechanisms may be involved in the activation of procaspase-9.
Like the mitochondria, the endoplasmic reticulum is a repository for both proapoptotic and antiapoptotic molecules. The known proapoptotic molecules include caspase-12 (11,12,14), p28Bap31 (36), and GADD153 (2), whereas the antiapoptotic molecules identified to date include the ER chaperone proteins GRP78 (2), calreticulin (37), protein-disulfide isomerase, and ORP-150 (2,25,38) as well as DAD1 (39,40). Despite the identification of these apoptotic regulators, the pathways that connect ER stress to apoptotic cell death remain unclear. Earlier reports indicated that ER stress induces the formation of a GRP78⅐procaspase-12⅐procaspase-7 complex (15). Prolonged stress can result in the disruption of this multimeric complex and the release of active caspase-12 that may activate caspase-9 and lead to apoptosis (14,15). Other activators of caspase-12 include the IRE1-TRAF2 complex (13) and calpain (11), both in response to ER stress. However, earlier studies have also suggested that calpains act as negative regulators of caspase processing by inactivating caspase-9 and -3 (41, 42). FIG. 6. Prodomain-deleted caspase-12, like ER stress, induces apoptosis via a caspase-9-dependent pathway. Both p(⌬N)C12 and p(⌬N)C12DN encompass the start site methionine, a His tag sequence, followed by Gly 95 -Asn 419 residues and thus lack all of the residues constituting the N-terminal prodomain. The catalytic mutant p(⌬N)C12DN has a Cys to Ala mutation at the active site's cysteine residue. Sak2 cells were transfected with p(⌬N)C12 and its catalytic mutant p(⌬N)C12DN. Six hours after transfection, cells received 25 M LEHD.fmk (caspase-9 inhibitor) or 25 M IETD.fmk (caspase-8 inhibitor). Surviving versus apoptotic cells were quantified as described under "Experimental Procedures." Thus, while calpains may be required for caspase-12 activation, they may not have a role in the caspase cascade leading to cell death. Activation of caspase-12, by any of the above mechanisms, may therefore initiate downstream caspase processing, activation, and cell death. It is thus important to understand how the caspase-7/caspase-12 pathway differs from the calpain/ caspase-12 pathway as well as the relevance of each of these pathways in ER stress-induced cell death. Studies employing calpain and caspase site-specific antibodies to caspase-12 may prove useful to elucidate these specific pathways.
The present study was undertaken to characterize the biochemical pathway by which ER stress leads to apoptosis. The studies were carried out in Apaf-1 Ϫ/Ϫ cells that are known to be relatively resistant to apoptotic insults that induce the intrinsic pathway. The cell-free extracts from these cells lacked mitochondria and cytochrome c and thus provided a system to assess whether ER stress-mediated apoptosis is triggered by the activation of caspases without the involvement of the apoptosome. Our studies indicate that Apaf-1 Ϫ/Ϫ cells are sensitive to ER stress inducers, undergo classical apoptosis, and possess features typical of caspase activation. Studies on whole cells, cell-free cytosolic extracts, and cell-free extracts containing primed microsomes suggest a role for caspase-12 in caspase-9 activation that is independent of Apaf-1 and mitochondria. Studies are in progress to identify other microsomal components that act in concert with caspase-12 to mediate ER stress-induced cell death.
Whereas our studies rule out a requirement for cytochrome c in the activation of caspases by ER stress, it is possible that prolonged ER stress may involve the concerted action of mitochondria and cytochrome c in caspase activation and cell death. In this context, several reports have demonstrated that thapsigargin induces cytochrome c release from mitochondria in murine and human cells (43)(44)(45). However, in these studies, the activation of downstream caspases proceeded with the active involvement of cytochrome c and Apaf-1 (apoptosome). In the absence of Apaf-1, other modes of caspase activation may exist. The current studies involving caspase-12 highlight one such mechanism. In addition to our present work, other studies have also demonstrated alternative mechanisms for caspase activation in the absence of Apaf-1 and cytochrome c (16, 46 -48). These studies demonstrated activation of caspase-9 by a mitochondrial component (47) or by a proapoptotic receptor complex (16,48).
Whereas the mechanism of ER stress-mediated activation of caspase-12 and other downstream caspases may be relevant in understanding neurodegenerative disorders that feature misfolded proteins, any potential advances in understanding these phenomena may seem irrelevant given the fact that to date there has been no report describing the sequence of human caspase-12. However, earlier observations in HeLa cells (12), A549 human lung carcinoma cells (49), and 293T cells (14) and studies in progress indicate that there exists a "human caspase-12-like protein" that is recognized by mouse caspase-12 antibodies. The human caspase-12 like protein has a similar molecular mass as the mouse caspase-12 and exists as a phosphorylated protein. 2 Studies are in progress to further characterize and identify this protein, which may have a key role in ER stress and neurodegeneration.