Induction of Apoptosis by Rewiring the Signal Tran
http://www.100md.com
病菌学杂志 2005年第8期
Institute of Immunology, Biomedical Sciences Research Center Al. Fleming, Vari, Greece
ABSTRACT
The Epstein-Barr virus latent membrane protein 1 (LMP1) is an oncoprotein which mimics activated tumor necrosis factor receptor family members. Here we demonstrate the principle that an inducible association of the LMP1 cytoplasmic carboxyl terminus with caspase-8 by a heterodimerizing agent causes apoptosis. This process depends on the catalytic activity of caspase-8 and the ability of LMP1 to oligomerize constitutively at the plasma membrane. Our data indicate that chemical inducers of the association of the LMP1 carboxyl terminus with caspase-8 can kill LMP1-expressing cells selectively. Such compounds could be used as chemotherapeutic agents for LMP1-associated malignancies.
TEXT
The latent membrane protein 1 (LMP1) is an Epstein-Barr virus (EBV)-encoded latent infection antigen which plays a prominent role in the process of EBV-associated oncogenesis (8). It is an integral membrane protein, consisting of a short amino-terminal cytoplasmic region, six transmembrane domains connected by short turns, and a 200-amino-acid-long cytoplasmic carboxyl terminus (CCT). The transmembrane domains mediate constitutive and ligand-independent oligomerization of the protein at the plasma membrane, which is essential for signal transduction. The CCT engages cellular signaling proteins, which include TRAF family members TRADD and JAK3, to induce pathways that activate the transcription factors NF-B, AP1, and STAT1/3 (6). A number of pieces of evidence at the molecular and cellular level support the notion that LMP1 acts as a constitutively active member of the tumor necrosis factor receptor (TNFR) family to induce cell growth and prevent apoptosis. Particular similarities have been noted between the function of LMP1 and the TNFR family member CD40 (9). Although CD40 and several TNFR family members have primarily antiapoptotic functions, other members of this family, such as FAS, have mainly proapoptotic properties. The apoptotic pathway induced by FAS is initiated by the ligand-dependent adoption of the proper trimeric conformation of the receptor and the concomitant dimerization of caspase-8, a cysteine protease bound to the cytoplasmic tail of the receptor (11). Activation of caspase-8 initiates a cascade of activation events of similar proteases that leads to the demise of the cell.
This study was initiated to test the hypothesis that an inducible association of the LMP1 CCT with caspase-8 could mimic an activated FAS receptor and initiate an apoptotic process that could kill LMP1-expressing cells. For this purpose, we have generated chimeric forms of LMP1 (LMP1FKBP2) and caspase-8 (FRBC8CR) that could be heterodimerized only in the presence of the rapamycin derivative AP21967 (ARIAD Pharamceuticals, Inc.) (4). Two copies of the FK506-binding protein FKBP12 were fused to the carboxyl terminus of LMP1 to generate LMP1FKBP2, and an amino-terminally FLAG-tagged and modified rapamycin-binding domain of the protein FRAP (FRB) was fused to the amino terminus of the catalytic region of procaspase-8 to generate FRBC8CR (Fig. 1A (2, 3). The FKBP and modified FRB domains can be heterodimerized only in the presence of AP21967.
The antiapoptotic function of LMP1 is primarily mediated by NF-B activation. To determine whether the addition of the FKBP domains to the carboxyl terminus of LMP1 compromises the ability of the protein to activate NF-B, we compared the NF-B activation properties of LMP1 and LMP1FKBP2. For this purpose, we have used a standard NF-B-dependent luciferase reporter assay, as described previously (12). LMP1FKBP2 activated NF-B similarly to wild-type LMP1 in human embryonic kidney (HEK) 293T cells, indicating that the addition of the FKBP domains does not compromise the signaling properties of LMP1 (Fig. 1B).
The effect of LMP1FKBP2 association with FRBC8CR on cell viability was tested by coexpressing the two proteins in HEK293T cells along with green fluorescent protein (GFP) to monitor the condition of the transfected cells. In the absence of the heterodimerizing agent AP21967, coexpression of LMP1FKBP2 and FRBC8CR was well tolerated by the cells, with minimal evidence for apoptotic cell death (Fig. 1C). However, following treatment of the transfected cells with AP21967, used at 200 nm throughout the study), a significant number of cells showed evidence of membrane blebbing, followed by cell disintegration into apoptotic bodies (Fig. 1C). To determine whether cell death was mediated by activation of the caspase pathway, we monitored the activity of caspase-3, which acts at a late stage in the pathway. For this purpose, cell lysates were prepared and incubated with a synthetic caspase-3 substrate (DEVD-pNA) at 37°C for 4 h (CaspACE colorimetric assay system; Promega). Upon cleavage of DEVD-pNA by caspase-3, free pNA is generated and produces a yellow color, which is monitored with a spectrophotometer at 405 nm. The amount of yellow color produced is proportional to the caspase-3 activity present in the cell lysate. Whereas coexpression of LMP1FKBP2 and FRBC8CR resulted in caspase-3 activity at or near background levels, the addition of AP21967 caused a dramatic increase in caspase-3 activity, consistent with the induction of caspase-dependent apoptosis (Fig. 1D). In four independent experiments, the average increase in caspase 3 activities by AP21967 in cells that coexpress LMP1FKBP2 and FRBC8CR was 4.1-fold ± 1.65-fold. The apoptotic mechanism of cell death was further confirmed by demonstrating AP21967-dependent cleavage of the 112-kDa poly(ADP-ribose) polymerase 1 (PARP) into a characteristic 85-kDa fragment in cells that had been cotransfected with LMP1FKBP2 and FRBC8CR expression vectors (Fig. 1E). LMP1FKBP2 or FRBC8CR was not able to induce apoptosis by itself in the presence of AP21967 without a functional FPBC8CR or LMP1FKBP2, respectively (see below).
To investigate whether AP21967 induces an association between LMP1FKBP2 and FRBC8CR as expected, FRBC8CR was immunoprecipitated from cells that had been cotransfected with LMP1FKBP2 and FRBC8CR expression vectors in the presence or absence of AP21967. LMP1FKBP2 was coimmunoprecipitated with FRBC8CR only in the presence of AP21967 (Fig. 2). Furthermore, FRBC8CR was able to interact with a nonoligomerizing LMP1FKBP2 mutant (D1LMP1FKBP2; described below) in an AP21967-dependent manner. This result indicates that the drug-dependent interaction of the two proteins does not depend on the ability of LMP1 to oligomerize at the plasma membrane.
To analyze the mechanism of AP21967-dependent apoptosis of cells that coexpress LMP1FKBP2 and FRBC8CR, we tested the effect of the caspase inhibitor CrmA in the process, and we used mutated forms of LMP1FKBP2 and FRBC8CR. Expression of CrmA inhibited AP21967-dependent caspase-3 activation and PARP cleavage in cells cotransfected with LMP1FKBP2 and FRBC8CR (Fig. 3). Because CrmA is a specific inhibitor of caspases 1 and 8 but not caspase-3, our data suggest that activation of the caspase-8 derivative FRBC8CR is essential for AP21967-dependent apoptosis (13). In addition, cotransfection of LMP1FKBP2 with a catalytically inactive mutant of FRBC8CR (FRBC8CRC360S) did not lead to caspase-3 activation or PARP cleavage in the presence of AP21967, thus confirming further the requirement for a functional caspase-8 moiety in FRBC8CR to induce apoptosis (Fig. 3). FRBC8CRC360S was expressed at a higher level than wild-type FRBC8CR (Fig. 3A, inset, upper panel). Finally, cotransfection of FRBC8CR with an LMP1FKBP2 mutant (D1LMP1FKBP2) that lacks the first four transmembrane domains of LMP1 and cannot oligomerize at the plasma membrane did not induce caspase-3 activation or PARP cleavage (Fig. 3). D1LMP1FKBP2 was expressed at a level similar to that of wild-type LMP1FKBP2 (Fig. 3A, inset, lower panel). This result indicates a requirement for LMP1 oligomerization to induce AP21967-mediated activation of caspase-8 and apoptosis, because D1LMP1FKBP2 was able to interact with FRBC8CR in the presence of AP21967 (Fig. 2).
Taken together, these results demonstrate that chemically induced association between the CCT of LMP1 and caspase-8 can initiate an apoptotic cascade of events which overcomes the antiapoptotic function of LMP1 and leads to cell death. The activation of caspase-8 is apparently mediated by its oligomerization in an LMP1-dependent manner. Our results suggest that oligomerized LMP1 adopts a conformation which permits LMP1-associated caspase-8 molecules to approach each other in a way that leads to their mutual activation. This study demonstrates the potential for the development of chemical compounds that could link the LMP1 CCT with the death effector domain of caspase-8 and induce cell death of LMP1-expressing tumors selectively. Such chemical compounds could be developed in a combinatorial manner by linking small molecules that interact with the LMP1 CCT or caspase-8, using, for example, a polyethylene glycol linker (1). Alternatively, cell-permeable or virally delivered peptides, consisting of fused LMP1 CCT- and caspase-8-interacting moieties, could be used as apoptosis-inducing agents for the treatment of LMP1-expressing tumors. Consistent with the latter principle is the demonstration by a recent report of cell death induction by polypeptide adaptors that link activated receptor tyrosine kinases with caspase-8 (7).
ACKNOWLEDGMENTS
We thank Theologos Loukidis for technical assistance.
This work was supported by an International Scholarship from the Howard Hughes Medical Institute and an EMBO Young Investigator award (to G.M.). G.M. is a Leukemia & Lymphoma Society of America Scholar.
REFERENCES
Becker, F., K. Murthi, C. Smith, J. Come, N. Costa-Roldan, C. Kaufmann, U. Hanke, C. Degenhart, S. Baumann, W. Wallner, A. Huber, S. Dedier, S. Dill, D. Kinsman, M. Hediger, N. Bockovich, S. Meier-Ewert, A. Kluge, and N. Kley. 2004. A three-hybrid approach to scanning the proteome for targets of small molecule kinase inhibitors. Chem. Biol. 11:211-223.
Chen, J., X. F. Zheng, E. J. Brown, and S. L. Schreiber. 1995. Identification of an 11-kDa FKBP12-rapamycin-binding domain within the 289-kDa FKBP12-rapamycin-associated protein and characterization of a critical serine residue. Proc. Natl. Acad. Sci. USA 92:4947-4951.
Choi, J., J. Chen, S. L. Schreiber, and J. Clardy. 1996. Structure of the FKBP12-rapamycin complex interacting with the binding domain of human FRAP. Science 273:239-242.
Chong, H., A. Ruchatz, T. Clackson, V. M. Rivera, and R. G. Vile. 2002. A system for small-molecule control of conditionally replication-competent adenoviral vectors. Mol. Ther. 5:195-203.
Hatzivassiliou, E., P. Cardot, V. I. Zannis, and S. A. Mitsialis. 1997. Ultraspiracle, a Drosophila retinoic X receptor alpha homologue, can mobilize the human thyroid hormone receptor to transactivate a human promoter. Biochemistry 36:9221-9231.
Hatzivassiliou, E., and G. Mosialos. 2002. Cellular signaling pathways engaged by the Epstein-Barr virus transforming protein LMP1. Front Biosci. 7:d319—d329.
Howard, P. L., M. C. Chia, S. Del Rizzo, F. F. Liu, and T. Pawson. 2003. Redirecting tyrosine kinase signaling to an apoptotic caspase pathway through chimeric adaptor proteins. Proc. Natl. Acad. Sci. USA 100:11267-11272.
Kieff, E., and A. Rickinson. 2001. Epstein-Barr virus and its replication, p. 2511-2573. In D. M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed. Lippincott, Williams & Wilkins, Philadelphia, Pa.
Lam, N., and B. Sugden. 2003. CD40 and its viral mimic, LMP1: similar means to different ends. Cell Signal. 15:9-16.
Mitchell, T., and B. Sugden. 1995. Stimulation of NF-B-mediated transcription by mutant derivatives of the latent membrane protein of Epstein-Barr virus. J. Virol. 69:2968-2976.
Thorburn, A. 2004. Death receptor-induced cell killing. Cell Signal. 16:139-144.
Trompouki, E., E. Hatzivassiliou, T. Tsichritzis, H. Farmer, A. Ashworth, and G. Mosialos. 2003. CYLD is a deubiquitinating enzyme that negatively regulates NF-kappaB activation by TNFR family members. Nature 424:793-796.
Zhou, Q., S. Snipas, K. Orth, M. Muzio, V. M. Dixit, and G. S. Salvesen. 1997. Target protease specificity of the viral serpin CrmA. Analysis of five caspases. J. Biol. Chem. 272:7797-7800.(Eudoxia G. Hatzivassiliou)
ABSTRACT
The Epstein-Barr virus latent membrane protein 1 (LMP1) is an oncoprotein which mimics activated tumor necrosis factor receptor family members. Here we demonstrate the principle that an inducible association of the LMP1 cytoplasmic carboxyl terminus with caspase-8 by a heterodimerizing agent causes apoptosis. This process depends on the catalytic activity of caspase-8 and the ability of LMP1 to oligomerize constitutively at the plasma membrane. Our data indicate that chemical inducers of the association of the LMP1 carboxyl terminus with caspase-8 can kill LMP1-expressing cells selectively. Such compounds could be used as chemotherapeutic agents for LMP1-associated malignancies.
TEXT
The latent membrane protein 1 (LMP1) is an Epstein-Barr virus (EBV)-encoded latent infection antigen which plays a prominent role in the process of EBV-associated oncogenesis (8). It is an integral membrane protein, consisting of a short amino-terminal cytoplasmic region, six transmembrane domains connected by short turns, and a 200-amino-acid-long cytoplasmic carboxyl terminus (CCT). The transmembrane domains mediate constitutive and ligand-independent oligomerization of the protein at the plasma membrane, which is essential for signal transduction. The CCT engages cellular signaling proteins, which include TRAF family members TRADD and JAK3, to induce pathways that activate the transcription factors NF-B, AP1, and STAT1/3 (6). A number of pieces of evidence at the molecular and cellular level support the notion that LMP1 acts as a constitutively active member of the tumor necrosis factor receptor (TNFR) family to induce cell growth and prevent apoptosis. Particular similarities have been noted between the function of LMP1 and the TNFR family member CD40 (9). Although CD40 and several TNFR family members have primarily antiapoptotic functions, other members of this family, such as FAS, have mainly proapoptotic properties. The apoptotic pathway induced by FAS is initiated by the ligand-dependent adoption of the proper trimeric conformation of the receptor and the concomitant dimerization of caspase-8, a cysteine protease bound to the cytoplasmic tail of the receptor (11). Activation of caspase-8 initiates a cascade of activation events of similar proteases that leads to the demise of the cell.
This study was initiated to test the hypothesis that an inducible association of the LMP1 CCT with caspase-8 could mimic an activated FAS receptor and initiate an apoptotic process that could kill LMP1-expressing cells. For this purpose, we have generated chimeric forms of LMP1 (LMP1FKBP2) and caspase-8 (FRBC8CR) that could be heterodimerized only in the presence of the rapamycin derivative AP21967 (ARIAD Pharamceuticals, Inc.) (4). Two copies of the FK506-binding protein FKBP12 were fused to the carboxyl terminus of LMP1 to generate LMP1FKBP2, and an amino-terminally FLAG-tagged and modified rapamycin-binding domain of the protein FRAP (FRB) was fused to the amino terminus of the catalytic region of procaspase-8 to generate FRBC8CR (Fig. 1A (2, 3). The FKBP and modified FRB domains can be heterodimerized only in the presence of AP21967.
The antiapoptotic function of LMP1 is primarily mediated by NF-B activation. To determine whether the addition of the FKBP domains to the carboxyl terminus of LMP1 compromises the ability of the protein to activate NF-B, we compared the NF-B activation properties of LMP1 and LMP1FKBP2. For this purpose, we have used a standard NF-B-dependent luciferase reporter assay, as described previously (12). LMP1FKBP2 activated NF-B similarly to wild-type LMP1 in human embryonic kidney (HEK) 293T cells, indicating that the addition of the FKBP domains does not compromise the signaling properties of LMP1 (Fig. 1B).
The effect of LMP1FKBP2 association with FRBC8CR on cell viability was tested by coexpressing the two proteins in HEK293T cells along with green fluorescent protein (GFP) to monitor the condition of the transfected cells. In the absence of the heterodimerizing agent AP21967, coexpression of LMP1FKBP2 and FRBC8CR was well tolerated by the cells, with minimal evidence for apoptotic cell death (Fig. 1C). However, following treatment of the transfected cells with AP21967, used at 200 nm throughout the study), a significant number of cells showed evidence of membrane blebbing, followed by cell disintegration into apoptotic bodies (Fig. 1C). To determine whether cell death was mediated by activation of the caspase pathway, we monitored the activity of caspase-3, which acts at a late stage in the pathway. For this purpose, cell lysates were prepared and incubated with a synthetic caspase-3 substrate (DEVD-pNA) at 37°C for 4 h (CaspACE colorimetric assay system; Promega). Upon cleavage of DEVD-pNA by caspase-3, free pNA is generated and produces a yellow color, which is monitored with a spectrophotometer at 405 nm. The amount of yellow color produced is proportional to the caspase-3 activity present in the cell lysate. Whereas coexpression of LMP1FKBP2 and FRBC8CR resulted in caspase-3 activity at or near background levels, the addition of AP21967 caused a dramatic increase in caspase-3 activity, consistent with the induction of caspase-dependent apoptosis (Fig. 1D). In four independent experiments, the average increase in caspase 3 activities by AP21967 in cells that coexpress LMP1FKBP2 and FRBC8CR was 4.1-fold ± 1.65-fold. The apoptotic mechanism of cell death was further confirmed by demonstrating AP21967-dependent cleavage of the 112-kDa poly(ADP-ribose) polymerase 1 (PARP) into a characteristic 85-kDa fragment in cells that had been cotransfected with LMP1FKBP2 and FRBC8CR expression vectors (Fig. 1E). LMP1FKBP2 or FRBC8CR was not able to induce apoptosis by itself in the presence of AP21967 without a functional FPBC8CR or LMP1FKBP2, respectively (see below).
To investigate whether AP21967 induces an association between LMP1FKBP2 and FRBC8CR as expected, FRBC8CR was immunoprecipitated from cells that had been cotransfected with LMP1FKBP2 and FRBC8CR expression vectors in the presence or absence of AP21967. LMP1FKBP2 was coimmunoprecipitated with FRBC8CR only in the presence of AP21967 (Fig. 2). Furthermore, FRBC8CR was able to interact with a nonoligomerizing LMP1FKBP2 mutant (D1LMP1FKBP2; described below) in an AP21967-dependent manner. This result indicates that the drug-dependent interaction of the two proteins does not depend on the ability of LMP1 to oligomerize at the plasma membrane.
To analyze the mechanism of AP21967-dependent apoptosis of cells that coexpress LMP1FKBP2 and FRBC8CR, we tested the effect of the caspase inhibitor CrmA in the process, and we used mutated forms of LMP1FKBP2 and FRBC8CR. Expression of CrmA inhibited AP21967-dependent caspase-3 activation and PARP cleavage in cells cotransfected with LMP1FKBP2 and FRBC8CR (Fig. 3). Because CrmA is a specific inhibitor of caspases 1 and 8 but not caspase-3, our data suggest that activation of the caspase-8 derivative FRBC8CR is essential for AP21967-dependent apoptosis (13). In addition, cotransfection of LMP1FKBP2 with a catalytically inactive mutant of FRBC8CR (FRBC8CRC360S) did not lead to caspase-3 activation or PARP cleavage in the presence of AP21967, thus confirming further the requirement for a functional caspase-8 moiety in FRBC8CR to induce apoptosis (Fig. 3). FRBC8CRC360S was expressed at a higher level than wild-type FRBC8CR (Fig. 3A, inset, upper panel). Finally, cotransfection of FRBC8CR with an LMP1FKBP2 mutant (D1LMP1FKBP2) that lacks the first four transmembrane domains of LMP1 and cannot oligomerize at the plasma membrane did not induce caspase-3 activation or PARP cleavage (Fig. 3). D1LMP1FKBP2 was expressed at a level similar to that of wild-type LMP1FKBP2 (Fig. 3A, inset, lower panel). This result indicates a requirement for LMP1 oligomerization to induce AP21967-mediated activation of caspase-8 and apoptosis, because D1LMP1FKBP2 was able to interact with FRBC8CR in the presence of AP21967 (Fig. 2).
Taken together, these results demonstrate that chemically induced association between the CCT of LMP1 and caspase-8 can initiate an apoptotic cascade of events which overcomes the antiapoptotic function of LMP1 and leads to cell death. The activation of caspase-8 is apparently mediated by its oligomerization in an LMP1-dependent manner. Our results suggest that oligomerized LMP1 adopts a conformation which permits LMP1-associated caspase-8 molecules to approach each other in a way that leads to their mutual activation. This study demonstrates the potential for the development of chemical compounds that could link the LMP1 CCT with the death effector domain of caspase-8 and induce cell death of LMP1-expressing tumors selectively. Such chemical compounds could be developed in a combinatorial manner by linking small molecules that interact with the LMP1 CCT or caspase-8, using, for example, a polyethylene glycol linker (1). Alternatively, cell-permeable or virally delivered peptides, consisting of fused LMP1 CCT- and caspase-8-interacting moieties, could be used as apoptosis-inducing agents for the treatment of LMP1-expressing tumors. Consistent with the latter principle is the demonstration by a recent report of cell death induction by polypeptide adaptors that link activated receptor tyrosine kinases with caspase-8 (7).
ACKNOWLEDGMENTS
We thank Theologos Loukidis for technical assistance.
This work was supported by an International Scholarship from the Howard Hughes Medical Institute and an EMBO Young Investigator award (to G.M.). G.M. is a Leukemia & Lymphoma Society of America Scholar.
REFERENCES
Becker, F., K. Murthi, C. Smith, J. Come, N. Costa-Roldan, C. Kaufmann, U. Hanke, C. Degenhart, S. Baumann, W. Wallner, A. Huber, S. Dedier, S. Dill, D. Kinsman, M. Hediger, N. Bockovich, S. Meier-Ewert, A. Kluge, and N. Kley. 2004. A three-hybrid approach to scanning the proteome for targets of small molecule kinase inhibitors. Chem. Biol. 11:211-223.
Chen, J., X. F. Zheng, E. J. Brown, and S. L. Schreiber. 1995. Identification of an 11-kDa FKBP12-rapamycin-binding domain within the 289-kDa FKBP12-rapamycin-associated protein and characterization of a critical serine residue. Proc. Natl. Acad. Sci. USA 92:4947-4951.
Choi, J., J. Chen, S. L. Schreiber, and J. Clardy. 1996. Structure of the FKBP12-rapamycin complex interacting with the binding domain of human FRAP. Science 273:239-242.
Chong, H., A. Ruchatz, T. Clackson, V. M. Rivera, and R. G. Vile. 2002. A system for small-molecule control of conditionally replication-competent adenoviral vectors. Mol. Ther. 5:195-203.
Hatzivassiliou, E., P. Cardot, V. I. Zannis, and S. A. Mitsialis. 1997. Ultraspiracle, a Drosophila retinoic X receptor alpha homologue, can mobilize the human thyroid hormone receptor to transactivate a human promoter. Biochemistry 36:9221-9231.
Hatzivassiliou, E., and G. Mosialos. 2002. Cellular signaling pathways engaged by the Epstein-Barr virus transforming protein LMP1. Front Biosci. 7:d319—d329.
Howard, P. L., M. C. Chia, S. Del Rizzo, F. F. Liu, and T. Pawson. 2003. Redirecting tyrosine kinase signaling to an apoptotic caspase pathway through chimeric adaptor proteins. Proc. Natl. Acad. Sci. USA 100:11267-11272.
Kieff, E., and A. Rickinson. 2001. Epstein-Barr virus and its replication, p. 2511-2573. In D. M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed. Lippincott, Williams & Wilkins, Philadelphia, Pa.
Lam, N., and B. Sugden. 2003. CD40 and its viral mimic, LMP1: similar means to different ends. Cell Signal. 15:9-16.
Mitchell, T., and B. Sugden. 1995. Stimulation of NF-B-mediated transcription by mutant derivatives of the latent membrane protein of Epstein-Barr virus. J. Virol. 69:2968-2976.
Thorburn, A. 2004. Death receptor-induced cell killing. Cell Signal. 16:139-144.
Trompouki, E., E. Hatzivassiliou, T. Tsichritzis, H. Farmer, A. Ashworth, and G. Mosialos. 2003. CYLD is a deubiquitinating enzyme that negatively regulates NF-kappaB activation by TNFR family members. Nature 424:793-796.
Zhou, Q., S. Snipas, K. Orth, M. Muzio, V. M. Dixit, and G. S. Salvesen. 1997. Target protease specificity of the viral serpin CrmA. Analysis of five caspases. J. Biol. Chem. 272:7797-7800.(Eudoxia G. Hatzivassiliou)