Granzyme B Mimics Apical Caspases
1998; Elsevier BV; Volume: 273; Issue: 51 Linguagem: Inglês
10.1074/jbc.273.51.34278
ISSN1083-351X
AutoresXiaohe Yang, Henning R. Stennicke, Baikun Wang, Douglas R. Green, Reiner U. Jänicke, Anu Srinivasan, Prem Seth, Guy S. Salvesen, Christopher J. Froelich,
Tópico(s)RNA Interference and Gene Delivery
ResumoGranzyme B (GrB) is predicted to trigger apoptosis by activating preferred caspases, but the zymogens that are directly processed by the granzyme and the requirements for these interactions remain unclarified. We examined this dilemma by comparing the kinetics and pattern of GrB-mediated activation of the executioner caspase-7 in vitro and in vivo. GrB rapidly activates procaspase-7 in vitro by cleaving between the large and small subunits leaving the propeptide intact. During GrB-mediated apoptosis, the caspase-7 propeptide is removed and cleavage occurs between the subunits. Strikingly, caspase-7 is unprocessed in caspase-3-deficient MCF-7 cells exposed to GrB but is rapidly activated when the cells are solubilized. Transfection with caspase-3 restores the removal of the caspase-7 propeptide and the capacity of GrB to subsequently activate the caspase. The data suggest that GrB activates caspase-3, which then removes the propeptide of caspase-7 allowing activation by GrB. Thus GrB initiates the death pathway by processing the accessible caspase-3, and the caspase-7 propeptide regulates trans-activation of the zymogen by granzyme. As a consequence, two proteases, caspase-3 and GrB, are required to activate procaspase-7. Granzyme B (GrB) is predicted to trigger apoptosis by activating preferred caspases, but the zymogens that are directly processed by the granzyme and the requirements for these interactions remain unclarified. We examined this dilemma by comparing the kinetics and pattern of GrB-mediated activation of the executioner caspase-7 in vitro and in vivo. GrB rapidly activates procaspase-7 in vitro by cleaving between the large and small subunits leaving the propeptide intact. During GrB-mediated apoptosis, the caspase-7 propeptide is removed and cleavage occurs between the subunits. Strikingly, caspase-7 is unprocessed in caspase-3-deficient MCF-7 cells exposed to GrB but is rapidly activated when the cells are solubilized. Transfection with caspase-3 restores the removal of the caspase-7 propeptide and the capacity of GrB to subsequently activate the caspase. The data suggest that GrB activates caspase-3, which then removes the propeptide of caspase-7 allowing activation by GrB. Thus GrB initiates the death pathway by processing the accessible caspase-3, and the caspase-7 propeptide regulates trans-activation of the zymogen by granzyme. As a consequence, two proteases, caspase-3 and GrB, are required to activate procaspase-7. granzyme B replication-deficient adenovirus type 2 polyacrylamide gel electrophoresis. A family of cytosolic cysteine proteases, the caspases, stored in most cells as zymogens play an essential role in the execution of apoptosis. The caspases involved in cell death are divided into apical (-2, -8, -9, and -10) and executioner subsets (-3, -6, and -7) (1Salvesen G.S. Dixit V.M. Cell. 1997; 91: 443-446Abstract Full Text Full Text PDF PubMed Scopus (1943) Google Scholar, 2Froelich C.J. Dixit V.M. Yang X. Immunol. Today. 1998; 19: 30-36Abstract Full Text PDF PubMed Scopus (116) Google Scholar). The proteolytic signal initiated by the apical caspases is transmitted to the executioners (caspases-3, -6, and -7) (1Salvesen G.S. Dixit V.M. Cell. 1997; 91: 443-446Abstract Full Text Full Text PDF PubMed Scopus (1943) Google Scholar, 2Froelich C.J. Dixit V.M. Yang X. Immunol. Today. 1998; 19: 30-36Abstract Full Text PDF PubMed Scopus (116) Google Scholar) whose action on cellular proteins defines apoptosis. The caspases are processed to form active heterodimeric enzymes by cleavage at specific Asp residues. Activation is not thought to require removal of the propeptide; cleavage between the large and small subunits is the activating event. The zymogens of apical caspases are recruited by specific adapter molecules, either at the cytosolic face of death receptors (in the case of caspase-8 and -10) (3Muzio M. Stockwell B.R. Stennicke H.R. Salvesen G.S. Dixit V.M. J. Biol. Chem. 1998; 273: 2926-2930Abstract Full Text Full Text PDF PubMed Scopus (885) Google Scholar), or via a post-mitochondrial route (caspase-9) (4Susin S.A. Zamzami N. Castedo M. Daugas E. Wang H.G. Geley S. Fassy F. Reed J.C. Kroemer G. J. Exp. Med. 1997; 186: 25-37Crossref PubMed Scopus (590) Google Scholar, 5Li P. Nijhawan D. Budihardjo I. Srinivasula S.M. Ahmad M. Alnemri E.S. Wang X.D. Cell. 1997; 91: 479-489Abstract Full Text Full Text PDF PubMed Scopus (6261) Google Scholar). Unlike the apical caspases where the propeptide plays a crucial role in signal transmission, the function of this segment in executioner caspases has not been clarified. The third pathway that activates the caspase cascade is initiated by cytotoxic cells. The immune system utilizes lymphocyte granule-mediated apoptosis to protect the host from intracellular pathogens and tumor cells through perforin/granzyme induced apoptosis where granzyme B (GrB)1 plays a vital role. Our laboratory has proposed that perforin and GrB are internalized to endosomes of the target cell during granule-mediated apoptosis. Perforin then permeabilizes the vesicles delivering the granzyme to the cytosol where GrB rapidly induces cell death by activating a preferred caspase substrate (6Froelich C.J. Orth K. Turbov J. Seth P. Babior B.M. Gottlieb R.A. Shah G.M. Bleackley R.C. Dixit V.M. Hanna W.L. J. Biol. Chem. 1996; 271: 29073-29081Abstract Full Text Full Text PDF PubMed Scopus (315) Google Scholar, 7Stoorvogel W. Strous G.J. Geuze H.J. Oorschot V. Schwartz A.L. Cell. 1991; 65: 417-427Abstract Full Text PDF PubMed Scopus (231) Google Scholar). GrB has been an important tool in analyzing the maturation of caspase-3 (8Darmon A.J. Nicholson D.W. Bleackley R.C. Nature. 1995; 377: 446-448Crossref PubMed Scopus (647) Google Scholar, 9Quan L.T. Tewari M. O'Rourke K. Dixit V.M. Snipas S.J. Poirier G.G. Ray C. Pickup D.J. Salvesen G.S. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 1972-1976Crossref PubMed Scopus (176) Google Scholar, 10Martin S.J. Amarante-Mendes G.P. Shi L.F. Chuang T.H. Casiano C.A. O'Brien G.A. Fitzgerald P. Tan E.M. Bokoch G.M. Greenberg A.H. Green D.R. EMBO J. 1996; 15: 2407-2416Crossref PubMed Scopus (273) Google Scholar), caspase-7 (11Chinnaiyan A.M. Orth K. Hanna W.L. Duan H.J. Poirier G.G. Froelich C.J. Dixit V.M. Curr. Biol. 1996; 6: 897-899Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar, 12Gu Y. Sarnecki C. Fleming M.A. Lippke J.A. Bleackley R.C. Su M.S.S. J. Biol. Chem. 1996; 271: 10816-10820Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar, 13Fernandes-Alnemri T. Takahashi A. Armstrong R. Krebs J. Fritz L. Tomaselli K.J. Wang L. Yu Z. Croce C.M. Salvesen G.S. Earnshaw W.C. Litwack G. Alnemri E.S. Cancer Res. 1995; 55: 6045-6052PubMed Google Scholar), caspase-6 (14Orth K. Chinnaiyan A.M. Garg M. Froelich C.J. Dixit V.M. J. Biol. Chem. 1996; 271: 16443-16446Crossref PubMed Scopus (378) Google Scholar,15Fernandes-Alnemri T. Litwack G. Alnemri E.S. Cancer Res. 1995; 55: 2737-2742PubMed Google Scholar), caspase-8 (16Muzio M. Chinnaiyan A.M. Kischkel F.C. O'Rourke K. Shevchenko A. Scaffidi C. Bretz J.D. Zhang M. Ni J. Gentz R. Mann M. Krammer P.H. Peter M.E. Dixit V.M. Cell. 1996; 86: 817-821Abstract Full Text Full Text PDF Scopus (2743) Google Scholar), caspase-9 (17Duan H.J. Orth K. Chinnaiyan A.M. Poirier G.G. Froelich C.J. He W.-W. Dixit V.M. J. Biol. Chem. 1996; 271: 16720-16724Abstract Full Text Full Text PDF PubMed Scopus (297) Google Scholar), and caspase-10a/b (18Fernandes-Alnemri T. Armstrong R.C. Krebs J. Srinivasula S.M. Wang L. Bullrich F. Fritz L.C. Trapani J.A. Tomaselli K.J. Litwack G. Alnemri E.S. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 7464-7469Crossref PubMed Scopus (694) Google Scholar, 19Vincenz C. Dixit V.M. J. Biol. Chem. 1997; 272: 6578-6583Abstract Full Text Full Text PDF PubMed Scopus (240) Google Scholar)in vitro. Although caspases -2, -3, -6, -7, and -8 are reported to undergo processing in targets during GrB-mediated apoptosis (6Froelich C.J. Orth K. Turbov J. Seth P. Babior B.M. Gottlieb R.A. Shah G.M. Bleackley R.C. Dixit V.M. Hanna W.L. J. Biol. Chem. 1996; 271: 29073-29081Abstract Full Text Full Text PDF PubMed Scopus (315) Google Scholar, 8Darmon A.J. Nicholson D.W. Bleackley R.C. Nature. 1995; 377: 446-448Crossref PubMed Scopus (647) Google Scholar, 11Chinnaiyan A.M. Orth K. Hanna W.L. Duan H.J. Poirier G.G. Froelich C.J. Dixit V.M. Curr. Biol. 1996; 6: 897-899Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar, 20Shi L.F. Chen G. MacDonald G. Bergeron L. Li H.L. Miura M. Rotello R.J. Miller D.K. Li P. Seshadri T. Yuan J.Y. Greenberg A.H. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 11002-11007Crossref PubMed Scopus (77) Google Scholar), the apparent polyspecificity of GrB for the caspases has made it difficult to establish which member is directly activated by granzyme to initiate the death pathway in vivo. Given this dilemma, it was argued that the caspase that is processed most rapidly by the granzyme in vitro would be a prime target for initial activation in vivo. Testing the human caspases, GrB was found to have the greatest preference for caspase-7 (21Talanian R.V. Yang X. Turbov J. Seth P. Ghayur T. Casiano C.A. Froelich C.J. J. Exp. Med. 1997; 186: 1323-1331Crossref PubMed Scopus (165) Google Scholar, 22Zhou Q. Salvesen G.S. Biochem. J. 1997; 324: 361-364Crossref PubMed Scopus (123) Google Scholar). During GrB-mediated apoptosis, however, the propeptide of caspase-7 is removed first. This observation indicated that GrB is unable to interact with procaspase-7 and that an upstream activator, possibly the less preferred caspase-3 or -10, removes the propeptide. We hypothesized that the caspase (-3 or -10) that removes the propeptide of caspase-7 would also be the member activated by GrB to initiate the death pathway (6Froelich C.J. Orth K. Turbov J. Seth P. Babior B.M. Gottlieb R.A. Shah G.M. Bleackley R.C. Dixit V.M. Hanna W.L. J. Biol. Chem. 1996; 271: 29073-29081Abstract Full Text Full Text PDF PubMed Scopus (315) Google Scholar, 21Talanian R.V. Yang X. Turbov J. Seth P. Ghayur T. Casiano C.A. Froelich C.J. J. Exp. Med. 1997; 186: 1323-1331Crossref PubMed Scopus (165) Google Scholar). Using caspase-3/10a-deficient MCF-7 cells (21Talanian R.V. Yang X. Turbov J. Seth P. Ghayur T. Casiano C.A. Froelich C.J. J. Exp. Med. 1997; 186: 1323-1331Crossref PubMed Scopus (165) Google Scholar, 23Jänicke R.U. Sprengart M.L. Wati M.R. Porter A.G. J. Biol. Chem. 1998; 273: 9357-9360Abstract Full Text Full Text PDF PubMed Scopus (1727) Google Scholar, 24Scaffidi C. Fulda S. Srinivasan A. Friesen C. Li F. Tomaselli K.J. Debatin K.M. Krammer P.H. Peter M.E. EMBO J. 1998; 17: 1675-1687Crossref PubMed Scopus (2633) Google Scholar) and stable transfectants expressing the deficient proteases, we have decisively proven that GrB triggers the caspase pathway by activating caspase-3, which then cleaves the propeptide of caspase-7. Remarkably, the removal of propeptide is crucial for the release of caspase-7 from a sequester site so it may then interact with the granzyme. MCF-7 cells were maintained RPMI 1640, 10% heat-inactivated fetal calf serum supplemented with 2 mml-glutamine, 100 units/ml penicillin and 50 μg/ml streptomycin. The caspase-3/-10a-deficient MCF-7 line was kindly provided by J. Boothman (University of Wisconsin). The absence of detectable caspase-3 and -10 was verified by immunoblotting with anti-caspase-3 and -10a/b rabbit polyclonal antibodies. The absence of caspase-3 is because of a 47-base pair deletion within exon 3 of the Casp-3 gene. This deletion results in the skipping of exon 3 during pre-mRNA splicing thereby abrogating translation of the caspase-3 mRNA (23Jänicke R.U. Sprengart M.L. Wati M.R. Porter A.G. J. Biol. Chem. 1998; 273: 9357-9360Abstract Full Text Full Text PDF PubMed Scopus (1727) Google Scholar). The caspase-deficient MCF-7 cells were transfected with the following constructs to generate stable lines: caspase-3 in a retroviral vector (pBabe-puro) (provided by Dr. T. Sladek) and caspase-10a in pCl-neo vector. After the selection with puromycin and geneticin, respectively, stable expression of the particular caspase was verified by immunoblotting. MCF-7 control cell lines were generated by transfecting empty pBabe-puro and pCl-neo vectors. Human GrB was isolated to homogeneity from a human natural killer cell line (YT) (25Hanna W.L. Zhang X. Turbov J. Winkler U. Hudig D. Froelich C.J. Protein Purif. Exp. 1993; 4: 398-402Crossref PubMed Scopus (66) Google Scholar) and a nonreplicating strain of adenovirus type 2 was cultured and isolated as described (26Seth P. Biochem. Biophys. Res. Commun. 1994; 203: 582-587Crossref PubMed Scopus (16) Google Scholar). Active caspase-3, -7, and -10a were expressed in Escherichia coli and isolated as described previously (27Roy N. Deveraux Q.L. Takahashi R. Salvesen G.S. Reed J.C. EMBO J. 1997; 16: 6914-6925Crossref PubMed Scopus (1140) Google Scholar). The expression constructs for caspase-3 and -7 contained a His6 tag at the C terminus of the full-length protein. Full-length procaspase 7 was obtained by reducing the previously described expression times in the presence of 0.2 mm isopropyl-1-thio-β-d-galactopyranoside to 30 min. This protocol has the advantage because it yields the full-length zymogens of both caspases and not the truncated form of procaspase 7 previously described by Zhou and Salvesen (22Zhou Q. Salvesen G.S. Biochem. J. 1997; 324: 361-364Crossref PubMed Scopus (123) Google Scholar). The concentrations of the purified enzymes were determined from the absorbance at 280 nm based on the molar absorption coefficients for the caspases calculated from the Edelhoch relationship: Caspase-3 (e280 = 26,000 m−1 cm−1), caspase-7 (e280 = 24,510 m−1cm−1) and caspase-10 (e280 = 29,140m−1 cm−1). The processing pattern of procaspase 7 by caspases 3 and 7 were determined by incubating 2.65 mm procaspase 7 with caspase 3 at concentration ranging from ∼0.2 to 400 nm or with caspase 7 at concentration ranging from ∼0.2 to 450 nm. The reactions were incubated for 30 min at 37 °C, stopped by boiling in SDS sample buffer for 5 min, and the reaction products separated by SDS-PAGE on an 8–18% gradient gel and visualized by Coomassie staining. Target cells were treated with isolated GrB and AD as described previously (6Froelich C.J. Orth K. Turbov J. Seth P. Babior B.M. Gottlieb R.A. Shah G.M. Bleackley R.C. Dixit V.M. Hanna W.L. J. Biol. Chem. 1996; 271: 29073-29081Abstract Full Text Full Text PDF PubMed Scopus (315) Google Scholar); unless indicated cells (1 × 106/ml) were mixed with GrB (1 μg/ml–30 nm) and AD (100 pfu) in 1-ml microcentrifuge tubes containing RPMI 1640 supplemented with 0.5% bovine serum albumin. To minimize in vitro processing of caspases in cell lysates before immunoblotting, targets were washed with phosphate-buffered saline 5 min before the end of the assay to remove excess GrB, transferred to a fresh tube, and lysed in buffer containing the GrB-specific anti-protease, anti-GraB (6Froelich C.J. Orth K. Turbov J. Seth P. Babior B.M. Gottlieb R.A. Shah G.M. Bleackley R.C. Dixit V.M. Hanna W.L. J. Biol. Chem. 1996; 271: 29073-29081Abstract Full Text Full Text PDF PubMed Scopus (315) Google Scholar). Detection of processed caspase -3, -7, -8, -9, and -10 was performed as described previously (6Froelich C.J. Orth K. Turbov J. Seth P. Babior B.M. Gottlieb R.A. Shah G.M. Bleackley R.C. Dixit V.M. Hanna W.L. J. Biol. Chem. 1996; 271: 29073-29081Abstract Full Text Full Text PDF PubMed Scopus (315) Google Scholar). Anti-caspase-3 and -7 were supplied by V. Dixit. Anti-caspase-8, -9, and -10 were supplied by Pharmingen, Inc., D. Green, and A. Srinivasan, respectively. Treated cells (106/ml) were lysed, resolved by SDS-PAGE (10–15%), and transferred to nitrocellulose. Rabbit antisera were used at a dilution of 1:1,000 followed by incubation with anti-rabbit Ig-horseradish peroxidase (Amersham Pharmacia Biotech) at a dilution of 1:10,000. Signal was visualized with the ECL kit (Amersham). During apoptotic responses in whole cells, caspase-7 is first processed by removal of the propeptide then full maturation occurs by a cut between the large and small subunits (interdomain linker site) (6Froelich C.J. Orth K. Turbov J. Seth P. Babior B.M. Gottlieb R.A. Shah G.M. Bleackley R.C. Dixit V.M. Hanna W.L. J. Biol. Chem. 1996; 271: 29073-29081Abstract Full Text Full Text PDF PubMed Scopus (315) Google Scholar, 28Orth K. O'Rourke K. Salvesen G.S. Dixit V.M. J. Biol. Chem. 1996; 271: 20977-20980Abstract Full Text Full Text PDF PubMed Scopus (187) Google Scholar). However, in MCF-7 cells lacking caspase-3 and -10a, processing of caspase-7 does not follow this pattern. Therefore, we predicted that either one or both these caspases are necessary to mature caspase-7 in vivo. Accordingly, recombinant caspase-3 and -10a were tested to determine whether either preferentially removed the propeptide of caspase-7. Caspase-3 was found to efficiently clip the propeptide of procaspase-7, but importantly it was unable to process procaspase-7 to its mature active form (Fig.1 a). The specificity of cleavage mediated by caspase-3 was verified by N-terminal analysis showing the cut was between the Asp and Ala residues in the sequence Asp-Ser-Val-Asp23-Ala-Val (data not shown). Because removal of the propeptide has been postulated to endow caspases with autocatalytic activity, we also examined the pattern of processing that caspase-7 manifested against its own precursor. Active caspase-7 was found to exert only minimal proteolytic activity on procaspase-7 in accordance with previous observations that caspase 7 does not activate the truncated form of procaspase-7 and, thus, does not undergo extensive autocatalytic activation (22Zhou Q. Salvesen G.S. Biochem. J. 1997; 324: 361-364Crossref PubMed Scopus (123) Google Scholar) (Fig. 1 b). This suggested that removal of the prodomain does not endow caspase-7 with autocatalytic activity and that another protease beside caspase-3 is required to fully mature caspase-7. Caspase-10 is postulated to be upstream of caspase-3 and -7 (21Talanian R.V. Yang X. Turbov J. Seth P. Ghayur T. Casiano C.A. Froelich C.J. J. Exp. Med. 1997; 186: 1323-1331Crossref PubMed Scopus (165) Google Scholar). We then examined the rate and pattern of cleavage that caspase-10a generated against the latter family members. Unlike caspase-3, caspase-10 preferred to cleave the zymogen of caspase-7 between the large and small subunits and not release the propeptide (Fig.2 a). N-terminal sequencing verified that the cut made by caspase-10a was between the Asp and Ser residues in the sequence Ile-Gln-Ala-Asp198-Ser-Gly to separate the large and small subunits (data not shown). Consistent with a preference for cleavage after the Ile-X-X-Asp sequence (29Thornberry N.A. Rano T.A. Peterson E.P. Rasper D.M. Timkey T. Garcia-Calvo M. Houtzager V.M. Nordstrom P.A. Roy S. Vaillancourt J.P. Chapman K.T. Nicholson D.W. J. Biol. Chem. 1997; 272: 17907-17911Abstract Full Text Full Text PDF PubMed Scopus (1852) Google Scholar), caspase-10a also processed caspase-3 at Ile-Glu-Thr-Asp174-Ser-Gly (Fig. 2 b). Importantly, caspase-10a appeared to more efficiently cleave caspase-3 than caspase-7 at this Ile-X-X-Asp motif (data not shown). Combined with evidence that GrB activates caspase-10a, a pattern emerges in vitro where an active, apical caspase (caspase-10a for example) matures caspase-3. Subsequently, active caspase-3 removes the propeptide of caspase-7 but then caspase-10a (or GrB) is required to complete the maturation of caspase-7. On the basis of this hypothesis, we investigated the processing of caspase-7 in specific MCF-7 cell lines. Using cleavage of the propeptide of caspase-7 as an index of caspase-3 activation, the pattern and rate of processing of caspase-7 was evaluated in a caspase-3/-10a negative MCF-7 line (MCFnull) and in cells with stable expression of caspase-3. When MCFnull cells were induced to undergo GrB-mediated apoptosis, surprisingly little cleavage of procaspase-7 was detected (Fig.3 a) indicating the granzyme was unable to efficiently process this caspase. To ensure that caspase-7 was expressed and could undergo processing in the caspase-3 negative line, proteolysis was examined in a detergent-generated cytosolic extract of the MCFnull. The predicted cleavage between the large and small catalytic subunits was observed indicating that the caspase was available (Fig. 3 b). In contrast to the MCFnull cells, GrB-induced apoptosis of the caspase-3+ MCF-7 cells was associated with removal of the caspase-7 propeptide (Fig.3 c) and followed by substantial cleavage between the large and small catalytic subunits. The onset of processing of caspase-7 coincided with the appearance of cleaved caspase-3 in the transfected line (Fig. 3 d) and in the cytosolic extract (Fig.3 e). Taken together, in targets expressing caspase-3 but lacking caspase-10a, GrB appears to have the capacity to directly activate caspase-3. Caspase-7, although highly preferred by GrBin vitro, is sequestered and cannot be processed directly by the granzyme unless the propeptide is first cleaved by caspase-3. After removal of the propeptide, caspase-7 becomes susceptible to activation by GrB suggesting that a novel two-step mechanism may exist to induce this executioner caspase. The data suggest that GrB directly activates caspase-3 delivered into whole cells deficient in caspase-10a. Nevertheless, based on rates of proteolysis in vitro, GrB might initiate the caspase cascade by activating caspase-10a, which then induces caspase-3 activity. As shown for caspase-7, however, caspase-10a also appears to be inaccessible to the granzyme after delivery into the target cell (Fig.4). Using MCF-7 cells transfected with caspase-10a, we examined the rate of processing during GrB-induced apoptosis and in cytosolic extracts. When GrB was delivered to the caspase-10a + transfectant, no processing was detected (Fig.4 a). For comparison, cleavage of caspase-10a could be readily detected in detergent-generated cytosolic extracts (Fig.4 b). Although caspase-10a is a preferred substrate for GrBin vitro, because of the inability of GrB to access this zymogen in vivo, the granzyme presumably cannot initiate the cascade by activating this protease. It has recently been suggested, based on experiments performed with CTL and purified GrB, that the granzyme could directly activate caspase-8 to trigger the death pathway (30Medema J.P. Toes R.E.M. Scaffidi C. Zheng T.S. Flavell R.A. Melief C.J.M. Peter M.E. Offringa R. Krammer P.H. Eur. J. Immunol. 1997; 27: 3492-3498Crossref PubMed Scopus (118) Google Scholar). Although caspase-8 is not a preferred substrate processed by GrB in vitro, the zymogen might be accessible to GrB in vivo. This possibility was examined by evaluating proteolysis of caspase-8 in MCFnull and MCF caspase 3+ cells. Caspase-8 was not processed in MCFnull cells during the time-course studied (Fig.5 a). However, cleavage of caspase-8 was detectable at 2 h in the MCF caspase-3+ line (Fig.5 b). Correspondingly, the onset of processing of caspase-8 in cytosolic extract of MCFnull was apparent at 1 h by a decrease in the intensity of the zymogen band (Fig. 5 c). These results demonstrate that caspase-8 indeed is processed in cells undergoing GrB-mediated apoptosis, but proteolysis occurs only in the presence of activated caspase-3. We also considered the possibility that GrB might activate the cascade through caspase-9 by a direct action on mitochondria, which appears to be an important initiating event in apoptosis triggered, for example, by genotoxic damage (31Reed J.C. Cell. 1997; 91: 559-562Abstract Full Text Full Text PDF PubMed Scopus (702) Google Scholar). Using the strategy described above, processing of caspase-9 in the presence and absence of caspase-3 was examined after intracellular delivery of GrB. Although no processing was observed in MCFnull cells during the time-course (Fig.6 a), for cells expressing caspase-3, the zymogen of caspase-9 (48 kDa) was converted to a 36-kDa form at 30 min with the large subunit faintly visible at 1 h (Fig.6 b). The 36-kDa product is consistent with cleavage at Asp330 (D330, caspase-9 numbering) (32Srinivasula S.M. Fernandes-Alnemri T. Zangrillia J. Robertson N. Armstrong R.C. Wang L. Trapani J.A. Tomaselli K.J. Litwack G. Alnemri E.S. J. Biol. Chem. 1996; 271: 27099-27106Abstract Full Text Full Text PDF PubMed Scopus (281) Google Scholar). Because caspase-9 is a relatively poor substrate for GrB, the caspase-9 zymogen was not cleaved in cytosolic extract derived from MCF-7null cells (Fig.6 c). Therefore, similar to caspase-8, caspase-9 is processed in GrB-treated cells, but proteolysis is dependent on the presence of active caspase-3. Because GrB shares a primary specificity for Asp with caspases, the serine protease might initiate the caspase pathway by signaling activation of an apical caspase, by imitating an apical caspase or by posing as an executioner. We focused on caspase-7 as a point of reference in the cascade, because unlike most family members, the propeptide of the zymogen is removed first in vivo. This signature is readily identified in immunoblotted cells undergoing GrB-induced, as well as Fas receptor-dependent, and tumor necrosis factor-mediated apoptosis. Using recombinant procaspase-7 to examine the role of proteases in the processing of this caspasein vitro and our newly described MCF cell lines to analyze the pattern of cleavage events in vivo, we have been able to test whether caspase-3 and/or caspase-10 were proximal elements of the cascade initiated by the granzyme and to assess the accessibility of procaspase-7 to processing by GrB. The granzyme first activates caspase-3 even though this substrate is not the most preferred member of the death protease family. Caspase-3 then cleaves the N-peptide of procaspase-7 making it accessible to activation by the granzyme. Because GrB processes many different caspases in vitro, the granzyme has been regarded as a polyspecific activator of the caspase cascade in vivo. The inability of the granzyme to directly process caspase-7, -8, -9, and -10, as well as caspase-6 (data not shown) in the absence of caspase-3, was an unexpected finding indicating that a single GrB-induced pathway may exist for cytotoxic cell-mediated activation of this death pathway. We have established that GrB functions like an apical caspase in an undescribed two-step process where accessibility to proteolysis contributes to the sequence in which the two executioner caspases are matured by the granzyme. Our unanticipated observation raises an intriguing question: why is caspase-7 compartmentalized to prevent activation by GrB but is freely activated by caspase-3? Because procaspase-7 is only processed by GrB after detergent extraction of the MCF-7null cells (see Fig. 3 b), the zymogen appears to be located in an inaccessible vesicular compartment. Others have observed, based on cell fractionation studies, that procaspase-7 is compartmentalized to the endoplasmic reticulum and mitochondria (33Chandler J.M. Cohen G.M. MacFarlane M. J. Biol. Chem. 1998; 273: 10815-10818Abstract Full Text Full Text PDF PubMed Scopus (232) Google Scholar) further supporting a physical separation of apical and executioner caspases. Although it is unclear whether the compartmentalization of caspase-7 is physical and/or biochemical, the sequestration of the executioner caspase-7 provides a previously undefined level of regulation in this enzymatic cascade. As a consequence of this sequestration, two distinct proteases are required for proper maturation of caspase-7. Using the MCF-7 sublines treated with tumor necrosis factor, caspase-8 appears to function similarly to the granzyme (34Stennicke H.R. Jurngensmeier J.M. Shin H. Wolf B. Yang X. Zhou Q. Ellerby H.M. Ellerby L.M. Bredesen D.E. Green D.R. Reed J.C. Froelich C.J. Salvesen G.S. J. Biol. Chem. 1998; 273: 27084-27090Abstract Full Text Full Text PDF PubMed Scopus (647) Google Scholar). These results may define a common mechanism shared by the granzyme and apical caspases (-2, -8, -9, and -10) to process the executioners (Fig. 7). Subsequently, uninvolved apical caspases may be activated in a process that serves to amplify the portion of the cascade that converges at caspase-3 underscoring the central role of this protease in producing cell death. Further studies of the proteolytic signal delivered from the apical to executioner caspases must take into account the different locations and intracellular trafficking patterns of the various family members in the cell as well as the efficiency that a particular caspase trans-activates another. We thank Dr. Vishva Dixit for kindly providing the anti-caspase-3 and -7 antibodies plus the construct for caspase-3. We also especially appreciate receipt of the caspase-3/-10-deficient MCF-7 line from Dr. David Boothman.
Referência(s)