Portal de Periódicos da CAPES

Reaper-induced apoptosis in a vertebrate system

1997; Springer Nature; Volume: 16; Issue: 24 Linguagem: Inglês

10.1093/emboj/16.24.7372

1460-2075

Erica Evans,

interferon and immune responses

Article15 December 1997free access Reaper-induced apoptosis in a vertebrate system Erica K. Evans Erica K. Evans Department of Pharmacology and Cancer Biology, Duke University Medical Center, Box 3686, C366 LSRC, Research Drive, Durham, NC 27710 USAE.K.Evans and T.Kuwana contributed equally to this work Search for more papers by this author Tomomi Kuwana Tomomi Kuwana Division of Cellular Immunology, La Jolla Institute for Allergy and Immunology, 10355 Science Center Drive, San Diego, CA 92121 USAE.K.Evans and T.Kuwana contributed equally to this work Search for more papers by this author Susan L. Strum Susan L. Strum Department of Pharmacology and Cancer Biology, Duke University Medical Center, Box 3686, C366 LSRC, Research Drive, Durham, NC 27710 USA Search for more papers by this author Jesse J. Smith Jesse J. Smith Department of Pharmacology and Cancer Biology, Duke University Medical Center, Box 3686, C366 LSRC, Research Drive, Durham, NC 27710 USA Search for more papers by this author Donald D. Newmeyer Donald D. Newmeyer Division of Cellular Immunology, La Jolla Institute for Allergy and Immunology, 10355 Science Center Drive, San Diego, CA 92121 USA Search for more papers by this author Sally Kornbluth Corresponding Author Sally Kornbluth Department of Pharmacology and Cancer Biology, Duke University Medical Center, Box 3686, C366 LSRC, Research Drive, Durham, NC 27710 USA Search for more papers by this author Erica K. Evans Erica K. Evans Department of Pharmacology and Cancer Biology, Duke University Medical Center, Box 3686, C366 LSRC, Research Drive, Durham, NC 27710 USAE.K.Evans and T.Kuwana contributed equally to this work Search for more papers by this author Tomomi Kuwana Tomomi Kuwana Division of Cellular Immunology, La Jolla Institute for Allergy and Immunology, 10355 Science Center Drive, San Diego, CA 92121 USAE.K.Evans and T.Kuwana contributed equally to this work Search for more papers by this author Susan L. Strum Susan L. Strum Department of Pharmacology and Cancer Biology, Duke University Medical Center, Box 3686, C366 LSRC, Research Drive, Durham, NC 27710 USA Search for more papers by this author Jesse J. Smith Jesse J. Smith Department of Pharmacology and Cancer Biology, Duke University Medical Center, Box 3686, C366 LSRC, Research Drive, Durham, NC 27710 USA Search for more papers by this author Donald D. Newmeyer Donald D. Newmeyer Division of Cellular Immunology, La Jolla Institute for Allergy and Immunology, 10355 Science Center Drive, San Diego, CA 92121 USA Search for more papers by this author Sally Kornbluth Corresponding Author Sally Kornbluth Department of Pharmacology and Cancer Biology, Duke University Medical Center, Box 3686, C366 LSRC, Research Drive, Durham, NC 27710 USA Search for more papers by this author Author Information Erica K. Evans1, Tomomi Kuwana2, Susan L. Strum1, Jesse J. Smith1, Donald D. Newmeyer2 and Sally Kornbluth 1 1Department of Pharmacology and Cancer Biology, Duke University Medical Center, Box 3686, C366 LSRC, Research Drive, Durham, NC 27710 USA 2Division of Cellular Immunology, La Jolla Institute for Allergy and Immunology, 10355 Science Center Drive, San Diego, CA 92121 USA *Corresponding author. E-mail: [email protected] The EMBO Journal (1997)16:7372-7381https://doi.org/10.1093/emboj/16.24.7372 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The reaper protein of Drosophila melanogaster has been shown to be a central regulator of apoptosis in that organism. However, it has not been shown to function in any vertebrate nor have the cellular components required for its action been defined. In this report we show that reaper can induce rapid apoptosis in vitro using an apoptotic reconstitution system derived from Xenopus eggs. Moreover, we show that a subcellular fraction enriched in mitochondria is required for this process and that reaper, acting in conjunction with cytosolic factors, can trigger mitochondrial cytochrome c release. Bcl-2 antagonizes these effects, but high levels of reaper can overcome the Bcl-2 block. These results demonstrate that reaper can function in a vertebrate context, suggesting that reaper-responsive factors are conserved elements of the apoptotic program. Introduction Apoptosis is a dramatic form of cell death used by multicellular organisms to rid themselves of superfluous or potentially harmful cells. While apoptosis plays a critical role in development, tissue homeostasis and immune regulation, dysregulation of this important cellular process may contribute to the development of numerous pathologies, including cancer, neurodegeneration and diabetes. Although the morphological hallmarks of apoptosis have been extensively documented (plasma membrane blebbing, DNA condensation, cleavage and nuclear fragmentation), the molecular participants in the process are only beginning to be defined (Bellamy et al., 1995). Proteins belonging to the interleukin 1β-converting enzyme (ICE) family of proteases (also known as caspases) have been widely implicated in execution of the apoptotic program in response to a diversity of stimuli (reviewed in Yuan, 1995; Chinnaiyan and Dixit, 1996). Modulators of the apoptotic process, believed to act upstream of the caspases, include pro- and anti-apoptotic members of the Bcl-2 family of proteins (e.g. bad, bax, bak, bcl-Xs and bcl-xl) (Nunez and Clarke, 1994; Reed, 1994) and ‘adaptor’ proteins such as FADD, TRADD and RAIDD, which appear to couple cell surface receptors to downstream caspase activation (Chinnaiyan et al., 1995; Hsu et al., 1995; Duan and Dixit, 1997). In a screen to identify novel regulators of the apoptotic process in Drosophila melanogaster, White and co-workers identified a small open reading frame encoding a 65 amino acid protein, which they named reaper (White et al., 1994). During fly development induction of reaper mRNA consistently preceded the onset of morphological apoptosis by 1–2 h and ectopic induction of reaper caused massive cell death both in cultured cells and in whole flies (Pronk et al., 1996; White et al., 1996). A chromosomal deletion that includes the reaper gene prevented almost all programed cell deaths during fly development and suppressed apoptosis in response to a number of different external stimuli. These findings suggest that reaper might be a key regulator of cell death in Drosophila. Although reaper participates in diverse apoptotic events, it does not appear to be part of the execution machinery, because apoptosis can occur in reaper mutant embryos when they are irradiated with very high levels of X-rays. Reaper-induced apoptosis is, however, prevented by caspase inhibitors (including peptide inhibitors and the viral inhibitor p35), consistent with the notion that reaper acts upstream of apoptotic proteases in cell death pathways (Pronk et al., 1996; White et al., 1996). Apoptotic regulators such as caspases and Bcl-2 have been well conserved from Caenorhabditis elegans to man, suggesting that the underlying mechanism of apoptotic cell death has been conserved through evolution. However, a vertebrate reaper homolog has not yet been found. Reaper has not been shown to be relevant to apoptosis in higher eukaryotes and has only been demonstrated to induce apoptosis upon ectopic expression in insect cells (Pronk et al., 1996; White et al., 1996; Vucic et al., 1997). It therefore remains an open question as to whether reaper represents an insect-specific inducer of the apoptotic program or whether similar pathways exist in other eukaryotes. Recently a cell-free extract of Xenopus eggs which can support apoptosis was described (Newmeyer et al., 1994). To induce egg laying, female frogs are normally primed with pregnant mare serum gonadotropin (PMSG) and then injected several days later with human chorionic gonadotropin (hCG) to induce egg laying. This regimen allows the preparation of egg extracts that form functional nuclei around added chromatin. Extending the interval between PMSG and hCG administration to several weeks produces extracts with apoptotic activities, most likely reflecting the in vivo process of oocyte atresia, wherein oocytes which have matured but have not been laid are apoptotically resorbed. Prolonged incubation (>4 h) of these extracts results in apoptotic nuclear fragmentaton and caspase activation. The development of apoptotic events in Xenopus egg extracts requires the presence of a heavy membrane (HM) fraction enriched in mitochondria. In response to factors present in Xenopus egg cytosol, mitochondria release a pro-apoptotic protein, cytochrome c (Kluck et al., 1997a,b). Normally, in non-apoptotic cells cytochrome c resides in the intermembrane space of mitochondria, where it participates in the electron transport chain. However, after the cell receives an appropriate apoptotic stimulus cytochrome c is released from the mitochondria. Once this protein is present in the cytosol it induces activation of proteases such as caspase-3 (CPP32), which in turn effect downstream apoptotic events, such as internucleosomal DNA cleavage (Liu et al., 1997). The Bcl–2 protein, through binding to the outer mitochondrial membrane, inhibits release of cytochrome c from mitochondria, thereby preventing caspase activation and downstream apoptotic events (Kluck et al., 1997a; Yang et al., 1997). In Xenopus extracts a lag period occurs prior to release of cytochrome c and the manifestation of overt apoptotic events. This suggests that important signaling events occur upstream of cytochrome c release. Recent studies, in fact, have shown that interaction of tyrosine-phosphorylated proteins and SH2 domains are important for upstream signaling in Xenopus extracts (Farschon et al., 1997). In particular, the adaptor protein Crk, which contains SH2 and SH3 domains, is required for apoptosis in this system (Evans et al., 1997). However, the full sequence of signaling events leading to the release of cytochrome c from mitochondria is still unknown. In this report we demonstrate that the Drosophila reaper protein can induce or accelerate apoptosis in Xenopus egg extracts. In addition, we show that reaper acts in conjunction with cytoplasmic factors to trigger mitochondrial cytochrome c release. These results provide the first demonstration that reaper can function in a vertebrate context and strongly indicate that reaper acts upstream of mitochondria. These data suggest that reaper-responsive factors are conserved elements of the apoptotic program and that the advantages of the biochemically tractable Xenopus system can be applied to elucidate the molecular mechanism of reaper-induced cell death. Results In order to determine whether Drosophila reaper could induce or accelerate apoptosis in Xenopus egg extracts, we produced recombinant reaper protein fused to the C–terminus of glutathione S-transferase (GST–reaper). GST–reaper protein (final concentration 600 ng/μl) was added to egg extracts along with sperm chromatin (for formation of synthetic nuclei). Although there was some extract-to-extract variability in the timing of apoptotic events, between 70 and 120 min after reaper addition, membrane blebs formed on the surface of the nuclei, followed by visible fragmentation of the chromatin and packaging of the DNA into small membrane-enclosed vesicles. Complete fragmentation of the nuclei was observed within 90–140 min in extracts supplemented with reaper protein, while nuclei formed in extracts lacking exogenously added reaper were stable for >4 h (Figure 1A and B). Nuclear fragmentation in the extract was highly synchronous, with all nuclei beginning fragmentation within 10 min of each other (even at nuclear concentrations >1000 per μl extract). The nuclear fragmentation induced by reaper was strongly reminiscent of apoptotic nuclear fragmentation seen in vivo and was morphologically identical to that previously described in egg extracts. This nuclear fragmentation occurred in cycloheximide-treated extracts, arguing that translation was not required for reaper action. Figure 1.Reaper-induced apoptosis in Xenopus egg extracts. GST–reaper protein (final concentration 600 ng/μl), or an equivalent concentration of GST protein alone, was added to extracts of Xenopus eggs in the presence of sperm chromatin (1000 nuclei/μl) and an ATP regenerating system. (A) Photomicrograph of a representative nucleus induced to enter apoptosis by addition of recombinant GST–reaper protein. All nuclei in the extract behaved similarly, entering apoptosis within 10 min of each other. (B) Photomicrograph of a representative stable nucleus formed in vitro by incubation of sperm chromatin in egg extract. This morphology was not altered by the addition of GST protein alone. Download figure Download PowerPoint Nuclear fragmentation induced in response to reaper was accompanied by cleavage of the DNA into fragments, producing a laddering pattern characteristic of that seen in apoptotic cells when resolved on agarose gels. Furthermore, these nuclear events were inhibited by addition of 10 μM Ac-YVAD-cmk or zVAD-FMK, peptide inhibitors of caspases (data not shown). Reaper induces caspase activation in vitro To directly demonstrate reaper-induced activation of caspases in the extract, we added trace amounts of various 35S-labeled substrates of these proteases to the extract. The substrates used were poly(ADP) ribose polymerase (PARP) (Lazebnik et al., 1994), the zymogens pro-caspase 3 (Yama/CPP32) (Nicholson et al., 1995; Tewari et al., 1995), pro-caspase 1 (ICE), and pro-caspase 7 (ICE-LAP3) (Duan et al., 1996) (members of the caspase family which are themselves cleaved to form active proteases from inactive precursors) and baculovirus p35, which acts as a competitive inhibitor of caspases (at higher concentrations than used here) while being cleaved by them (Bump et al., 1995; Xue and Horvitz, 1995). Aliquots of extract containing these radiolabeled substrates were withdrawn every 10 min during a 3 h room temperature incubation and were then resolved by SDS–PAGE and processed for autoradiography. With the sole exception of pro-caspase 1, all of these substrates were cleaved to their characteristic apoptotic fragments in extracts to which reaper had been added, but not in control extracts (Figure 2). Routinely, cleavage of all these proteins preceded the initial stages of nuclear fragmentation by ∼10 min. It is unclear whether the failure to cleave pro-caspase 1 reflects the fact that caspase 1 is not activated during reaper-induced apoptosis or was due to some incompatibility between the heterologous components in this assay (Xenopus extract, Drosophila reaper protein and human caspase 1). However, it is striking that reaper can engage the Xenopus apoptotic machinery, triggering endogenous protease activation. Figure 2.Reaper induces proteolytic cleavage of apoptotic substrates. 35S-labeled pro-caspases 1, 3 or 7 (ICE, YAMA and ICE-LAP-3 respectively) or caspase substrates (baculovirus p35 and PARP) were added to extracts in the presence (right hand panel in every pair, lanes 9–16, A–C, and 10–18, D and E) or absence (left hand panel in every pair, lanes 1–8, A–C, and 1–9, D and E) of recombinant GST–reaper. Beginning 30 min after the start of room temperature incubation (lanes 1 and 9, A–C, and 1 and 10, D and E), samples were withdrawn at 10 min intervals, resolved by SDS–PAGE and processed for autoradiography. The position of signature apoptotic cleavage fragments are denoted by arrows. (A) Pro-caspase 7; (B) pro-caspase 3; (C) baculovirus p35; (D) PARP; (E) pro-caspase 1. Download figure Download PowerPoint Mitochondria are required for reaper-induced apoptosis Ultracentrifugation of crude egg extracts at 250 000 g produces cytosolic and microsomal membrane fractions which are cleanly separated from mitochondria, ribosomes and other dense organelles. As noted previously, naturally occurring in vitro apoptosis in these extracts requires a subcellular fraction containing mitochondria. Therefore, although extracts reconstituted by mixing the light membrane and cytosolic fractions are fully competent to form functional nuclei around added sperm chromatin templates, they never enter apoptosis, even upon prolonged room temperature (e.g. >8 h) incubation. Similarly, we found that reconstituted extracts lacking mitochondria did not respond to reaper. These extracts failed to undergo microscopically visible nuclear fragmentation and did not initiate proteolytic cleavage of caspase substrates after reaper addition, as exemplified using pro-caspase 3 as substrate (Figure 3A). However, when the reconstituted cytosolic and membrane fractions were supplemented with a purified subcellular fraction containing mitochondria the ability to respond to reaper was restored, resulting in caspase activation (Figure 3B). Thus, reaper-induced apoptosis in these extracts requires the mitochondrial fraction. Figure 3.Reaper requires a mitochondrial component for caspase activation. (A) 35S-labeled pro-caspase 3 (YAMA) and recombinant GST–reaper protein were added to fractionated egg extracts lacking mitochondria. Samples were withdrawn at 0, 0.5, 1, 1.5, 2 and 2.5 h (lanes 1–6 respectively) and processed for SDS–PAGE and autoradiography. Pro-caspase 3 remains uncleaved during this 2.5 h incubation. (B) Addition of mitochondria to fractionated extract results in cleavage of 35S-labeled pro-caspase 3. Radiolabeled pro-caspase 3 and GST–reaper protein were added to extracts supplemented with purified mitochondria. Samples were withdrawn at 0, 0.5, 1, 1.5, 2 and 2.5 h (lanes 1–6) and processed as in (A). Specific caspase 3 cleavage products appear at 1.5 h of room temperature incubation and are denoted by arrows. Download figure Download PowerPoint Reaper accelerates development of latent apoptotic activities Although reconstituted egg extracts lacking mitochondria cannot sustain full-blown apoptosis, we recently reported that incubation of these extracts for 2.5 h at room temperature allows development of latent apoptotic activities which are manifested upon transfer of a 1/10 volume of the preincubated reconstituted extract into crude extract containing nuclei and mitochondria (schematically depicted in Figure 4A; Evans et al., 1997). This results in a marked acceleration of caspase activation and apoptotic nuclear fragmentation (which occurs within 60–90 min, instead of the typical 4 h), suggesting that the mitochondrial fraction of the crude extract allows execution of the apoptotic program in response to latent activities developed during the 2.5 h preincubation of membrane and cytosol. We therefore refer to the reconstituted light membrane and cytosolic fractions, where latent apoptotic activities develop, as the ‘latent extract’ and the crude extract containing mitochondria and nuclei as the ‘execution extract’. Preincubation of the latent extract for <2.5 h does not result in acceleration of apoptosis upon dilution into the execution extract. Figure 4.Reaper accelerates the development of latent apoptotic activities. (A) Schematic diagram illustrating the assay for apoptosis. The cytoplasm and membrane components of the egg extract were incubated together for 2.5 h at room temperature. After this incubation period the cytoplasm and membrane were transferred at 1/10 volume into a fresh crude extract (containing mitochondria). Upon addition of an ATP regenerating mix and sperm chromatin, nuclei formed and underwent apoptosis after 60–90 min at room temperature. (B) Reaper does not accelerate the onset of apoptosis when the latent extract (cytoplasm plus membrane) is preincubated for 2.5 h. Recombinant GST–reaper protein or ELB was added to the latent extract at the start of the 2.5 h incubation. This extract was transferred into an execution extract as described above. Samples were withdrawn at 10 min intervals and viewed under fluorescence microscopy. While nuclei appeared stable at 70 min in extracts supplemented with GST–reaper or buffer alone, both extracts displayed the apoptotic phenotype at 80 min. (C) Schematic diagram demonstrating the experimental design. Latent extract was incubated for 0.5, 1 or 1.5 h in the presence of recombinant reaper protein or buffer alone and transferred at 1/10 volume into an execution extract. (D) Reaper reduces the incubation time for development of latent apoptotic activities. (Top) Apoptotic nuclei in crude execution extract after addition of latent extract (cytoplasm plus membrane) supplemented with reaper and incubated for 0.5 or 1.5 h before transfer. (Bottom) Stable nuclei formed in crude execution extracts after addition of latent extract which had been incubated for 0.5 or 1.5 h. Latent extract incubated for 1 h with reaper but not with buffer alone induced apoptosis in the same manner. Download figure Download PowerPoint Although we found that reaper was unable to induce apoptosis in the absence of mitochondria, we postulated that reaper might accelerate the development of apoptotic activities in the latent extract. To determine if this was the case, we added reaper (600 ng/μl) to latent extracts and incubated them for various times at room temperature. Interestingly, we found that reaper substantially reduced the time required to develop latent apoptotic activities, as evidenced by its ability to shorten the required preincubation time of the latent extract from 2.5 h to 30 min. (Figure 4C and D). However, latent extracts incubated for the full 2.5 h induced apoptosis 70–80 min after transfer into crude extracts in the presence or absence of added reaper, suggesting that maximal levels of latent apoptotic activity develop in 2.5 h and that downstream components in the execution extract then become rate limiting (Figure 4B). It should be noted that the amounts of reaper present in the execution extract after transfer of latent extracts containing reaper (equivalent to 1/10 the amount used in the experiments shown in Figures 1) did not accelerate apoptosis when added directly to the execution extract. These experiments suggest that reaper participates in events upstream of the step requiring mitochondria, during the ‘latent’ period of the apoptotic process in vitro. Reaper circumvents the requirement for crk in apoptosis We recently reported that the SH2 and SH3 domain-containing protein crk also acts during this latent stage of apoptosis in the egg extract (Evans et al., 1997). Specifically, addition of dominant–negative variants of crk or immunodepletion of crk from latent extracts prevents acceleration of apoptosis upon dilution into execution extracts. Moreover, direct addition of anti-crk sera to latent extracts also impedes subsequent apoptosis, which can be reversed by addition of recombinant Xenopus crk protein. Since we found that reaper protein could accelerate events occurring in the latent extract, we wished to determine whether this acceleration relied upon the presence of crk. As reported previously, when we immunodepleted crk from the latent extract subsequent apoptosis in the execution extract was not accelerated (Figure 5C). However, when we added reaper to latent extracts depleted of crk apoptosis occurred on schedule 60 min after dilution of the preincubated latent extract into the execution extract, demonstrating that reaper-responsive factors promoted acceleration of apoptosis even in the absence of crk (Figure 5D). Figure 5.Reaper circumvents the requirement for crk in induction of apoptosis. (A) The latent extract was immunodepleted with non-immune serum. This depleted extract was recombined with membrane, incubated for 2.5 h at room temperature and transferred into crude execution extract. Apoptotic nuclei were observed at 75 min after transfer. (B) Latent extract immunodepleted with non-immune serum was supplemented with GST–reaper protein and membrane and incubated for 2.5 h. Apoptosis was observed 75 min after transfer into the crude extract. (C) Latent extract immunodepleted with anti-crk serum was reconstituted with membrane and incubated as described in (A). Apoptosis was not acclerated upon transfer into the crude extract and nuclei remained stable. (D) GST–reaper protein was added to latent extract immunodepleted with anti-crk serum. This extract was reconstituted with membrane and incubated for 2.5 h. After 75 min in the crude extract apoptotic nuclei were observed. Download figure Download PowerPoint Interestingly, we found that GST–reaper protein was able to induce apoptosis in otherwise non-apoptotic extracts (i.e. those which did not spontaneously initiate the apoptotic process even after 6 h). This suggests that reaper engages the apoptotic machinery downstream of the endogenous signaling events responsible for activating the apoptotic process in Xenopus egg extracts, consistent with the ability of reaper to circumvent the requirement for crk protein. Reaper promotes cytochrome c release from mitochondria The fact that reaper could acclerate events occurring in the latent extract suggested that reaper could modulate factors acting upstream of the requirement for mitochondria. Recently Liu et al. (1996) reported that cytochrome c is required for in vitro apoptosis initiated by addition of dATP to HeLa cell extract. Moreover, regulated release of cytochrome c from mitochondria accompanies activation of the apoptotic program, both in mammalian cells and in the Xenopus cell-free system (Kluck et al., 1997a; Yang et al., 1997). Bcl-2, a potent inhibitor of apoptosis, acts, at least in part, by inhibiting cytochrome c release (Kluck et al., 1997a; Yang et al., 1997). We surmised that the absence of cytochrome c in our reconstituted extracts (lacking mitochondria) might account for the inability of reaper to induce apoptosis. Therefore, we added bovine or equine heart cytochrome c to the reconstituted extracts in the presence and absence of reaper and observed nuclei in these extracts for the appearance of apoptotic changes. Between 70 and 85 min after initiation of room temperature incubation the nuclei in these extracts began to enter apoptosis, as monitored by fluorescence microscopy, regardless of whether reaper was present (data not shown). This is in marked contrast to extracts lacking cytochrome c, which showed no signs of apoptotic nuclear fragmentation, even 6 h after reaper addition. These results were confirmed at a biochemical level by analysis of baculovirus p35 degradation (Figure 6). Addition of cytochrome c to extracts lacking mitochondria promoted rapid apoptotic cleavage of p35 regardless of whether or not reaper was present (Figure 6). Even after careful titration we were not able to identify a concentration of cytochrome c which could accelerate apoptosis only in the presence of reaper (data not shown). Interestingly, addition of free cytochrome c to extracts containing mitochondria also greatly accelerated apoptosis, consistent with the idea that release of cytochrome c from mitochondria is a rate limiting step in this process (E.K.Evans et al., unpublished results). Taken together these data suggest that reaper might induce downstream release of cytochrome c from mitochondria, thereby triggering caspase activation. Figure 6.Reaper does not accelerate cytochrome c-induced caspase activation. 35S-labeled baculovirus p35 protein was added to extracts lacking a subcellular fraction enriched in mitochondria without (lanes 1–9, A and B) or with (lanes 10–18, A and B) added reaper protein. In (B) extracts were supplemented with bovine heart cytochrome c. Starting 20 min after the beginning of room temperature incubation (lanes 1 and 10) samples were withdrawn at 20 min intervals and processed for SDS–PAGE and autoradiography. Although some undegraded p35 can still be seen at the first time point (lanes 1 and 10), a specific p35 cleavage product is present by 20 min in the presence of exogenously added cytochrome c. Arrows indicate the uncleaved protein and a signature apoptotic cleavage product. On several occasions we saw p35 cleavage products appear as early as 5 min after cytochrome c addition. Download figure Download PowerPoint To determine whether reaper could indeed induce mitochondrial cytochrome c release, we incubated GST–reaper in unfractionated Xenopus egg extracts. Aliquots were assayed at various times for cytochrome c release and caspase activity. Although the exact timing of apoptotic events varied between extracts (Farschon et al., 1997), we consistently observed that reaper accelerated both the release of cytochrome c from mitochondria and the consequent activation of DEVD-cleaving caspases, as measured by cleavage of a fluorimetric substrate, DEVD-pNA. In this experiment, release of cytochrome c and DEVDase activation occurred at ∼6–7 h incubation in the absence of reaper, but at ∼1–2 h in the presence of reaper (Figure 7). In another experiment (not shown) apoptotic events occurred earlier in the control extract, at ∼4 h incubation, but reaper nevertheless caused precocious release of cytochrome c, at 1–1.5 h incubation. We conclude that cytochrome c release is a downstream consequence of reaper action. Figure 7.Reaper accelerates cytochrome c release from mitochondria and DEVDase activation. Recombinant reaper protein or buffer were added to the crude extract at a 1:10 (v/v) dilution (600 ng/μl final concentration). At the indicated times 15 μl samples were removed for measurement of cytochrome c release by Western blotting (A) and 2