If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
* This work was supported by grants from the National Institutes of Health (NIH) and Hereditary Disease Foundation (HiQ) (to A. L. G.) and by a postdoctoral award from the NIH (to X.-B. Q.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The on-line version of this article (available at http://www.jbc.org) contains Supplemental Fig. 1s and Table 1s.
Smac/DIABLO, HtrA2/Omi, and caspase-9 play key roles in the initiation of apoptosis. The inhibitor of apoptosis proteins (IAPs) are believed to bind to the N-terminal IAP binding motifs of the mature (proteolytically processed) forms of Smac, HtrA2, and caspase-9. However, we show here that BRUCE/Apollon, a 528-kDa IAP whose degradation promotes apoptosis, associates with their precursors as well as the mature forms by binding to regions in addition to the IAP binding motif. Through these associations, BRUCE promotes the degradation of Smac and inhibits the activity of caspase-9 but not the effector caspase, caspase-3. In response to apoptotic stimuli, BRUCE is degraded by proteasomes and/or cleaved by caspases and HtrA2 depending on the specific stimulus and the cell type. These results suggest that the ability of BRUCE to antagonize both the precursor and mature forms of Smac and caspase-9 is an important mechanism for the prevention of apoptosis under normal conditions.
In response to apoptotic stimuli, caspases are activated to initiate irreversible proteolytic cascades that result in the destruction of key cellular components and rapid cell death (
). Cytochrome c then binds to Apaf-1 and activates the initiator caspase-9, which cleaves and activates the effector caspases, such as caspases-3 and -7. The activity of caspases can be inhibited by the inhibitor of apoptosis protein (IAP)
). On the other hand, the actions of IAP proteins are opposed by certain pro-apoptotic factors, especially Smac/DIABLO and HtrA2/Omi. These proteins contain the specific IAP binding motifs at their N termini. Smac and HtrA2 are synthesized as larger cytosolic precursors that are proteolytically processed in mitochondria to expose their IAP binding motifs. In response to apoptotic stimuli, the processed Smac and HtrA2 are released into the cytosol where they interact with the IAPs (
). Smac and the Drosophila pro-apoptotic proteins, Hid, Rpr, and Grim, which also contain the IAP binding motifs, inhibit IAPs by tight binding, but HtrA2 is a serine protease that cleaves and inactivates certain IAPs (
In addition to inhibiting caspase activities, several IAPs have been shown to promote proteasomal degradation of proteins by catalyzing ubiquitination. Ubiquitination involves the sequential actions of the ubiquitin (Ub)-activating enzyme (E1), a Ub-carrier protein (E2), and a Ub-protein ligase (E3), which binds the substrate and catalyzes the transfer of the activated Ub from a specific E2 to the substrate (
). In certain cells, these IAPs catalyze their own ubiquitination and degradation as well as degradation of other proteins important for apoptosis, including mature Smac/DIABLO, active caspase-3, and tumor necrosis factor receptor-associated factor 2 (
). BRUCE itself is regulated by ubiquitin-dependent degradation that is mediated by the E2 UbcH5 and the E3 Nrdp1, which together also ubiquitinate the epidermal growth factor receptor family member, ErbB3 (
). These findings suggest that BRUCE, unlike other mammalian IAPs, is essential to inhibit apoptosis in certain cell types. In Drosophila, overexpression of dBRUCE (the Drosophila homolog of BRUCE) suppresses cell death induced by Rpr and Grim (
) reported that BRUCE is essential to inhibit Smac-induced apoptosis by promoting ubiquitination and degradation of the mature Smac. We present complementary evidence that BRUCE promotes degradation of mature Smac. However, we demonstrate that BRUCE, unlike other IAPs, also binds to the precursor of Smac and promotes its degradation. Furthermore, we show that BRUCE binds to pro-caspase-9 and inhibits its cleavage. Pro-caspase-9 in complex with Apaf-1 plays a key role in the initiation of the caspase cascades and is resistant to other IAPs (
). Therefore, the interaction of BRUCE with pro-caspase-9 is likely to be particularly important in inhibiting apoptosis. In addition, we show that BRUCE can inhibit the activity of mature caspase-9 but not caspase-3. BRUCE is the only essential IAP in mammals (
Cell Culture and Transfections—293T adenovirus-transformed human embryo kidney cells, HT1080 human fibrosarcoma cells, and HeLa human epitheloid carcinoma cells were maintained in Dulbecco's modified Eagle's medium (CellGro) supplemented with 10% fetal bovine serum and antibiotics in 5% CO2. All of the cell lines and serum were obtained from the American Type Culture Collection. Transfection was carried out using a Trans-IT transfection kit (Mirus).
Expression Vectors—The C-terminal region of BRUCE (residues 4412–4829, BRUCE-C), full-length Smac (residues 1–239), truncated Smac (Δ2–55), full-length HtrA2 (residues 1–458), and truncated HtrA2 (Δ2–133) with Myc-His6 tags at the C termini were subcloned into pcDNA6 (Invitrogen). The SmacV57A/P58A mutant was made according to standard procedures. The DNA vectors for siRNA expression, BS/U6/BRUCE, and BS/U6/GFP were obtained as described previously (
). The plasmids with T7 promoters for pro-caspase-3 and pro-caspase-9 were kindly provided by Dr. Junying Yuan. The sequences of all of the constructs were verified by DNA sequencing.
Co-immunoprecipitation and Western Blot Analysis—Proteins were extracted and immunoprecipitated in buffer A containing 20 mm HEPES, pH 7.5, 10 mm KCl, 1.5 mm MgCl2, 1 mm EDTA, 1 mm EGTA, 1 mm dithiothreitol, 0.5% CHAPS, 0.1 mm phenylmethylsulfonyl fluoride, and a protease inhibitor mixture (Roche Applied Science). To purify BRUCE or XIAP, immunoprecipitation was carried out in a high stringency buffer containing 20 mm Tris-HCl, pH 8.0, 500 mm NaCl, 0.2% Nonidet P40, 10% glycerol, 1 mm ZnCl2, and the protease inhibitor mixture. Membranes carrying proteins from 14 mouse parenchymal tissues were purchased from RNWAY Laboratories Inc. (Seoul, Korea). Except where noted, the levels of proteins were analyzed by Western blot using horseradish peroxidase-conjugated or alkaline-phosphatase-conjugated secondary antibodies. Antibodies against BRUCE and XIAP were purchased from BD Biosciences. Caspase-7, actin, and FLAG were from Sigma. Caspase-9 and caspase-3 were purchased from Cell Signaling Inc. Myc was from Oncogene, HtrA2 and caspase-9 were from R&D Systems, and Smac was from Zymed Laboratories Inc.
Caspase and HtrA2 Assays—To assay caspase activation, 293T cell extracts were prepared in buffer A (but without CHAPS and the protease inhibitor mixture) essentially as described (
). The activity of caspase-9 was analyzed in the buffer containing 0.1 m Tris, pH 6.8, 10 mm dithiothreitol, 0.1% CHAPS, 10% polyethylene glycol, and 20 μm Ac-LEHD-amino-4-methylcoumarin (AMC). The activity of caspase-3 was assayed using either Z-DEVD-7-AMC or 35S-labeled pro-casapse-9 as the substrate. The release of AMC from the substrates was monitored continuously at 380/460 nm (excitation/emission), and the cleavage of pro-casapse-9 was detected by PhosphorImager following SDS-PAGE. The cleavage of the immunopurified BRUCE by the purified recombinant caspases (obtained from BIOMOL) or HtrA2 (from R&D Systems) was assayed according to the manufacturer's instructions.
BRUCE Associates with Both Precursor and Mature Forms of Smac—Overexpression of dBRUCE has been shown to suppress cell death induced by the Smac functional homologs, Rpr and Grim (
). Because mature Smac, through its IAP binding motif (the AVPI sequence), interacts with other IAPs, we examined whether mammalian BRUCE associates with Smac. BRUCE could be co-immunoprecipitated with an anti-Smac antibody (Fig. 1A). In contrast, the C-terminal region of BRUCE, which lacks the BIR domain, was not brought down by this antibody. When 293T cells were transfected with the full-length Smac containing a Myc tag at its C terminus, both BRUCE and XIAP could be co-immunoprecipitated from the lysates with an anti-Myc antibody (Fig. 1B). However, when these cells were transfected instead with a truncated Smac, which contains a methionine on its N-terminal IAP binding motif in place of its N-terminal sequence (Smac Δ2–55), XIAP was not co-immunoprecipitated with Smac Δ2–55 because this N-terminal methionine blocks the function of the IAP binding motif in Smac Δ2–55 (
). In contrast, some of the BRUCE was reproducibly co-immunoprecipitated with Smac Δ2–55, although the amount of BRUCE precipitated was significantly smaller than when the full-length molecule was transfected (Fig. 1B). These findings suggest that XIAP and BRUCE bind Smac by distinct mechanisms (for further evidence, see below).
When an antibody against BRUCE or XIAP was used for immunoprecipitation, the mature (proteolytically processed) Smac could be co-immunoprecipitated with either antibody (Fig. 1C). However, the full-length Smac precursor was also co-immunoprecipitated with BRUCE but not with XIAP (Fig. 1C). Because Smac can dimerize (
), we tested whether the co-immunoprecipitation of the Smac precursor with BRUCE was due to dimerization of the precursor with the mature Smac. However, when the Smac precursor was generated by in vitro transcription and translation, in the absence of the mature Smac, BRUCE was still immunoprecipitated together with the Smac precursor (Fig. 1D and data not shown). Thus, unlike XIAP, BRUCE binds to not only the mature but also the precursor form of Smac. The N-terminal sequence of the Smac precursor appears to be important for this association with BRUCE, because the truncated Smac, Smac Δ2–55, which lacks the N-terminal sequence, was co-immunoprecipitated with BRUCE to a much lesser extent than the precursor (Fig. 1D).
Region(s) in Mature Smac Besides the AVPI Sequence Can Bind to BRUCE—Because the N-terminal sequence of the precursor blocks the interaction of other IAPs with the AVPI sequence (
), the association of BRUCE with the Smac precursor must involve a distinct mechanism. To further examine whether the interaction of BRUCE with the mature Smac requires the AVPI sequence, we made a mutant Smac (V57A/P58A) that contains an AAAI stretch in place of the AVPI. When this mutant was transfected into 293T cells, both the precursor and the mature form were detectable by Western blot (Fig. 2A and data not shown) and, as expected, these mutations greatly reduced (by over 90%) the association with XIAP (compared with that of Smac containing the AVPI motif). However, an appreciable amount of BRUCE was still co-immunoprecipitated with this mutated mature Smac (in fact only 20% less BRUCE was precipitated than with the wild-type Smac, Fig. 2A). These findings indicate that Smac can associate with BRUCE through non-AVPI region(s).
To confirm this surprising result, we tested whether the AVPI-containing peptide, AVPIA, could inhibit this interaction. When this peptide was present in the immunoprecipitation experiments, it blocked the association of Smac with XIAP but not that with BRUCE (Fig. 2B). In contrast, an irrelevant sequence MALLK (the N-terminal peptide of the Smac precursor) or NFEKL had no effect on the association of Smac with either XIAP or BRUCE. Thus, although the AVPI motif seems to contribute to the association of mature Smac with BRUCE, BRUCE also associates with both the mature Smac and its precursor through an AVPI-independent mechanism.
Loss of BRUCE Reduces Degradation of Smac—In response to TRAIL (TNF-related apoptosis-inducing ligand), Smac in MCF-7 cells is released from mitochondria and its level falls due to degradation by the proteasome (
). In HT1080 cells, the DNA-damaging agent, camptothecin (CPT), also reduced the levels of Smac (data not shown). To test whether BRUCE mediates this degradation of Smac, we attempted to decrease the levels of BRUCE in HT1080 cells. Transfection of the E3 Nrdp1, which catalyzes the degradation of BRUCE especially during the initiation of apoptosis (
), led to a marked reduction in the levels of BRUCE and a clear increase in the levels of Smac in the presence of CPT (Fig. 3A). This increase in Smac is probably due to the reduction in BRUCE, because the knockdown of BRUCE by RNAi also led to a marked increase in the levels of endogenous Smac (Fig. 3B). Similarly, the levels of HtrA2, another protein containing an IAP binding motif, also increased following the knockdown of BRUCE. In contrast, the levels of pro-caspase-9 did not increase under this condition (Fig. 3B).
To confirm that BRUCE decreases the levels of Smac, 293T cells were transfected with the full-length Smac and with BRUCE-siRNA. As noted above, decreasing the levels of BRUCE by RNAi caused an increase in the levels of mature Smac in these cells (Fig. 3C). No Smac precursor was detected in this experiment, probably because only small amounts of the vector for Smac were used for transfection. Overexpression of Nrdp1, which reduced the levels of BRUCE, also increased the levels of mature Smac in the 293T cells (Fig. 3D). It is noteworthy that the overexpression of Nrdp1 did elevate the levels of the Smac precursor, perhaps because Nrdp1 reduced BRUCE content more than BRUCE-siRNA (Fig. 3D).
To determine whether BRUCE reduces Smac by accelerating its degradation, we measured the stability of mature Smac by pulse-chase analysis. Both BRUCE-siRNA and Nrdp1 reduced the degradation of Smac (Fig. 3E). The half-life of Smac was less than 1 h in control cells, but it increased to more than 2 h in the cells transfected with BRUCE-siRNA or Nrdp1, indicating that BRUCE promotes Smac degradation.
Because depletion of BRUCE by Nrdp1 caused accumulation of both the mature Smac and its precursor, it is possible that BRUCE controls the levels of Smac in normal cells. To investigate this possibility, we measured the relative levels of these proteins in various mouse tissues by Western blot. The relative amounts of BRUCE were found to be inversely correlated with the levels of Smac in the different tissues (n = 14, p < 0.05) (Fig. 3F). This inverse correlation implies that when the levels of BRUCE are high, the precursor (and probably also the mature form) of Smac is rapidly degraded. In contrast, although the knockdown of BRUCE also led to an increase in the levels of HtrA2 in HT1080 cells (Fig. 3B), no inverse correlation was found between the levels of HtrA2 and BRUCE in these mouse tissues (Fig. 3F and data not shown).
BRUCE Associates with Both Pro-caspase-9 and Caspase-9 and Antagonizes Caspase-9 —Because caspase-9 also contains an IAP binding motif, we examined whether pro-caspase-9 may bind to BRUCE in a similar manner as the Smac precursor. Unexpectedly, no association between the endogenous pro-caspase-9 and full-length BRUCE was detected by co-immunoprecipitation. However, a 90-kDa fragment that reacted with the anti-BRUCE antibody was co-precipitated with pro-caspase-9 (Fig. 4A). Therefore, this failure to find full-length BRUCE in the immunoprecipitates may be due to its cleavage by pro-caspase-9 (see below). Accordingly, the addition of the caspase inhibitor Z-VAD-fmk to the cell lysates led to the detection of the proteins larger than 90 kDa that reacted with the anti-BRUCE antibody (Fig. 4A). To test more rigorously whether BRUCE associates with pro-caspase-9, we incubated the immunopurified BRUCE or XIAP with pro-caspase-9 synthesized by in vitro transcription and translation and found that pro-caspase-9 could be pulled down with an antibody against BRUCE but not an anti-XIAP antibody (Fig. 4B). Whereas the antibody against XIAP pulled down a small amount of pro-caspase-3, the anti-BRUCE antibody was much more effective (Fig. 4B). When mature caspase-9 or caspase-3 was incubated with immunopurified BRUCE or XIAP, an antibody against BRUCE pulled down caspase-9 but not caspase-3. In contrast, an antibody against XIAP co-immunoprecipitated with both caspase-3 and caspase-9 (Fig. 4C). Therefore, BRUCE binds in vitro to both pro-caspase-9 and caspase-9 but not caspase-3.
Pro-caspase-9 is active in cleaving pro-caspase-3 when it forms a complex with Apaf-1 (
). Therefore, we tested whether the association of BRUCE with pro-caspase-9 also inhibits its activity in vivo. Unlike most of the other cell lines we tested (data not shown), HeLa cells have some detectable processed caspase-9 in the absence of apoptotic stimuli, probably due to auto-cleavage of pro-caspase-9 in these cells. Decreasing the content of BRUCE by RNAi and/or treatment with CPT reduced the levels of pro-caspase-9, probably because the lack of BRUCE led to greater auto-cleavage activity of pro-caspase-9 (Fig. 4D). Accordingly, with the RNAi for BRUCE, there was additionally an accumulation of the processed caspase-9 (Fig. 4D). This accumulation could also be due to the reduced degradation of caspase-9. Taken together, these findings suggest that BRUCE normally inhibits the cleavage of pro-caspase-9 by inhibiting its auto-cleaving activity but it may also retard the breakdown of mature caspase-9.
To test whether BRUCE inhibits the activity of caspase-9, BRUCE was incubated with the recombinant caspase-9 in the presence of LEHD-7-AMC. Full-length BRUCE, but not its C-terminal region, markedly inhibited caspase-9 activity (Fig. 4E). In contrast, unlike XIAP, BRUCE did not inhibit hydrolyses of DEVD-AMC and pro-caspase-9 by caspase-3 (Fig. 4, F and G). Therefore, BRUCE appears to be a potent inhibitor of the initiator caspase but not effector caspases.
HtrA2 Associates with and Cleaves BRUCE—HtrA2 is another pro-apoptotic factor containing an IAP binding motif. In 293T cells transfected with the Myc-tagged C-terminal region of BRUCE, endogenous mature HtrA2 was coimmunoprecipitated with BRUCE and XIAP but not with the Myc-tagged C-terminal region of BRUCE (Fig. 5A). Because the Smac precursor interacted with BRUCE, we tested whether the precursor of HtrA2 did also. When the 35S-labeled precursor of HtrA2 (synthesized by in vitro transcription and translation) was incubated with BRUCE or XIAP, it could be pulled down with an antibody against BRUCE but not an anti-XIAP antibody (Fig. 5B). When the N-terminal sequence was replaced by Met, this form of HtrA2 (HtrA2-(Δ2–133)) no longer associated with BRUCE (Fig. 5B). Thus, BRUCE associates with both the precursor and the mature form of HtrA2.
HtrA2 is a serine protease that cleaves certain IAPs (
). Therefore, we examined whether BRUCE is also a substrate for HtrA2. Overexpression of HtrA2 reduced the levels of BRUCE in 293T cells (Fig. 5C). HtrA2-(Δ2–133) also reduced the levels of BRUCE, although it did not form a stable complex with BRUCE. To further determine whether HtrA2 cleaves BRUCE, immunopurified BRUCE was incubated with the purified mature HtrA2. BRUCE was cleaved by HtrA2, and two fragments of 180 and 150 kDa could be detected with the anti-BRUCE antibody in Western blots (Fig. 5D). Thus, similar to other IAPs, BRUCE is a substrate for HtrA2.
During Apoptosis, BRUCE Can Be Degraded by Proteasomes or Cleaved by HtrA2 or Caspases—In response to apoptotic stimuli, most IAPs can be destroyed by the ubiquitin-proteasome pathway or by caspases (
). In fact, the sequence of BRUCE was found to contain multiple putative cleavage sites for caspases-1, -3, -6, -7, and -8 (Supplemental Table 1s). Therefore, we measured the levels of BRUCE in the lysates of 293T cells following in vitro caspase activation by cytochrome c and dATP. As expected, the levels of BRUCE decreased markedly upon caspase activation and this decrease could be blocked by two general caspase inhibitors, Z-VAD-fmk or BD-fmk (Fig. 6A). Under these cell-free conditions, the proteasome inhibitor, lactacystin, at concentrations that block proteasomal activity (
) had no effects (Fig. 6A). Thus, BRUCE can be cleaved by caspases under conditions where the ubiquitin-proteasome pathway is inactive. To determine which caspases were responsible for this cleavage, we incubated the immunopurified BRUCE with purified caspases. All of the four caspases tested, especially caspases-3 and -9, appeared to cleave BRUCE (Fig. 6B).
During apoptosis, BRUCE can also be ubiquitinated by Nrdp1 and degraded by the proteasome (
). In HeLa cells, both Z-VAD-fmk and lactacystin were found to partially reduce the loss of BRUCE induced by etoposide (a topoisomerase inhibitor) (Fig. 6C). Therefore, both caspases and proteasomes contribute to the destruction of BRUCE induced by etoposide. It is noteworthy that, in the absence of etoposide, the treatment with Z-VAD-fmk or lactacystin for 12 h also increased the levels of BRUCE, suggesting that caspases and the proteasome also contribute to the destruction of BRUCE under normal conditions. In 293T cells, which possessed no detectable active caspases-7 and -9 (Fig. 6D and data not shown), lactacystin still inhibited the etoposide-induced degradation of BRUCE (Fig. 6D). Thus, in these cells, proteasomal degradation of BRUCE does not require caspase activation. However, the pathway primarily responsible for the destruction of BRUCE seems to vary with apoptotic stimuli. As shown in Fig. 6E, in HeLa cells, Z-VAD-fmk could completely block the TNFα-induced loss of BRUCE, suggesting that caspase cleavage is the dominant mechanism for TNFα-mediated cleavage of BRUCE. Thus, in addition to being cleaved by HtrA2, BRUCE can also be degraded by the proteasome and/or cleaved by caspases, depending on the nature of apoptotic stimulus and cell.
Unlike other mammalian IAPs, BRUCE/Apollon is essential for viability of various cell lines (
) showed that BRUCE promotes degradation of mature Smac. These results complement our finding that reducing the cellular content of BRUCE by RNAi or overexpression of Nrdp1 leads to reduced degradation of both the endogenous and transfected Smac. However, this study has uncovered some additional surprising features of BRUCE action that differ from those found in these related studies. Specifically, we have found that: (A) BRUCE, unlike other IAPs, can bind to the precursor of Smac and promote its degradation; (B) BRUCE, unlike other IAPs, can bind to pro-caspase-9 and inhibit its cleavage to the mature form; (C) BRUCE can also inhibit the activity of mature caspase-9 but not the effector caspase, caspase-3; and (D) multiple caspases and HtrA2 (in addition to the proteasome) can degrade BRUCE during apoptosis. This exceptional ability of BRUCE to antagonize the precursors of Smac and pro-caspase-9 (in addition to their mature forms) might explain why BRUCE is the only essential IAP in mammals.
Overexpression of the Smac precursor has been shown to promote the apoptosis induced by certain apoptotic stimuli (
), although this precursor by itself is probably inactive in promoting apoptosis. Presumably, this enhancement of apoptosis occurred because higher amounts of this precursor in the cytosol led to increased levels of Smac in the mitochondria; therefore, the cellular sensitivity to apoptotic stimuli rises. Thus, BRUCE-mediated degradation of the Smac precursor can be important in making cells more resistant to apoptotic stimuli. It is noteworthy in this regard that, under normal conditions, the levels of Smac in various tissues are inversely correlated with the content of BRUCE, presumably because of BRUCE-mediated degradation of the Smac precursor. The additional capacity of BRUCE to bind and destroy mature Smac after its release from mitochondria (
), it should have the capacity to conjugate ubiquitin to these pro-apoptotic proteins (i.e. to act as both an E3 and E2). Indeed, BRUCE (together with E1) has been shown to catalyze ubiquitination of mature Smac (
). Presumably, BRUCE also catalyzes the ubiquitination and degradation of the Smac precursor by a similar mechanism.
Our finding that BRUCE associates with the Smac precursor as well as the mature form was unexpected. This ability to act on the Smac precursor not only has important regulatory consequences but also indicates that BRUCE functions by distinct mechanisms from other IAPs. Because the N-terminal sequence of the precursor prevents the interaction of the IAP binding motif of Smac (AVPI) with other IAPs (
), Smac must associate with BRUCE through a novel mechanism. Accordingly, the mutation of this AVPI sequence to AAAI blocked the association of mature Smac with XIAP but only slightly reduced the extent of its association with BRUCE. Because the BRUCE BIR domain by itself could not bind to Smac (Supplemental Fig. 1s), some additional region appears to be required for these associations. In fact, the closely related BIR domain of Survivin is also not sufficient by itself to bind to mature Smac (
) failed to detect an association of BRUCE with the precursor of Smac, perhaps because they used an N-terminal glutathione S-transferase fusion of the Smac precursor. As we showed here (Fig. 1D), the N-terminal sequence of the Smac precursor is important for its association with BRUCE and the presence of the large glutathione S-transferase moiety might well interfere with the binding of the Smac precursor to BRUCE.
Pro-caspase-9 in complex with Apaf-1, similar to the processed caspase-9, can activate pro-caspase-3, and the ectopic expression of pro-caspase-9 promotes apoptosis (
). Because pro-caspase-9 is resistant to other IAPs, the ability of BRUCE to interact with pro-caspase-9 is likely to be especially important in the inhibition of apoptosis in normal cells.
BRUCE also associated with and inhibited the processed caspase-9 but not the active caspase-3. In contrast, several IAPs inhibit both caspase-9 and caspase-3. BRUCE thus appears to differ from other IAPs by only inhibiting the initial steps of the caspase cascade. Although BRUCE was not found to directly inhibit caspase-3 (as shown here by direct enzymatic assays), Bartke et al. (
) recently reported that BRUCE inhibits caspase-3 activation in 293T cell lysates. Therefore, this inhibition of caspase-3 activation in the cell lysates by BRUCE is probably an indirect downstream consequence of the inhibition of pro-caspase-9 or caspase-9 activity. In addition, Hao et al. (
) recently reported that transfected BRUCE promotes degradation of the processed caspase-9, which is consistent with our finding that RNAi for BRUCE led to an accumulation of the processed endogenous caspase-9 in HeLa cells.
Once the normal balance between pro-apoptotic and anti-apoptotic factors starts to tip toward apoptosis, cell death quickly develops due to the release and activation of pro-apoptotic factors as well as the degradation and inhibition of anti-apoptotic factors. For example, in response to apoptotic stimuli, certain IAPs can be removed by the ubiquitin-proteasome pathway, HtrA2, and/or caspases (
). This study shows that, depending on the nature of apoptotic stimulus and cell, BRUCE can also be efficiently degraded by HtrA2 and multiple caspases. Because BRUCE is important in inhibiting apoptosis under normal conditions, its degradation must be a key irreversible step in the initiation of the apoptotic process; therefore, it is not surprising that mammalian cells contain multiple enzymatic and signaling systems to trigger its destruction.
We thank Dr. Junying Yuan for advice and reagents and our labmates for their valuable discussions.