SMAC Mimetics Sensitize Nonsteroidal Anti-inflammatory Drug–Induced Apoptosis by Promoting Caspase-3–Mediated Cytochrome c Release

Introduction

Prevention of human cancers through the use of chemical agents or dietary manipulation has emerged as a promising approach to reduce morbidity and mortality of cancer ( 1). A number of epidemiologic studies, clinical trials, and animal studies have shown that nonsteroidal anti-inflammatory drugs (NSAID), such as sulindac and aspirin, are effective against colorectal cancer, the second leading cause of cancer-related death ( 2, 3). However, the mechanisms underlying the antineoplastic effects of NSAIDs remain unclear. The effects of NSAIDs are incomplete, and resistance to NSAIDs is often developed ( 2, 3). Furthermore, the side effects associated with high dose of NSAIDs, such as the cardiovascular toxicity of the cyclooxygenase 2 (COX2)–specific inhibitors, have presented a significant obstacle for general use of these agents for cancer prevention ( 4).

Among many described activities of NSAIDs, induction of apoptosis secondary to inhibition of COX enzymes seems to play a critical role in NSAID-mediated chemoprevention ( 2, 3, 5). Apoptosis is a major mechanism regulating turnover of intestinal epithelial cells from which colon tumors are derived ( 6). During the formation of a colon tumor, apoptosis is progressively inhibited due to oncogenic mutations ( 7). Deregulation of apoptosis is necessary for oncogenic transformation and drives neoplastic cells to gain additional tumorigenic features ( 8). NSAIDs have been found to induce apoptosis in colon cancer cells in vitro and in vivo ( 2, 3). The apoptotic level can be used as a biomarker for NSAID-mediated chemoprevention ( 9, 10). Furthermore, it has been shown that NSAID treatment can reverse the antiapoptotic effects of COX 2 ( 11).

In mammalian cells, apoptotic cell death is executed through a cascade of events, among which a key event is the translocation of apoptogenic proteins from the mitochondria into the cytosol to trigger caspase activation. Cytochrome c is released from the mitochondria into the cytosol, where it binds to other proteins to form an “apoptosome” ( 12). Second mitochondria-derived activator of caspase (SMAC thereafter), also called direct inhibitor of apoptosis protein (IAP)–binding protein with low isoelectric point (Diablo), is the second mitochondrial protein found to be released into the cytosol during apoptosis ( 13, 14). After its release, SMAC binds to IAPs through its N-terminal AVPI domain and relieves their inhibition of caspases ( 15).

The unfolding of the complex pathways involved in apoptosis signaling has stimulated intensive efforts to restore apoptosis in cancer cells for therapeutic purposes ( 16). It has been shown that SMAC overexpression and SMAC mimetic peptides enhance the proapoptotic effects of several anticancer agents, such as tumor necrosis factor (TNF)–related apoptosis-inducing ligand (TRAIL; refs. 13, 17– 20). Several small molecules mimicking the AVPI domain of SMAC have also been developed ( 21– 23). For example, a compound called C3 was synthesized based on the structure of SMAC ( 21). C3 is a symmetric dimer of an oxazoline-derived compound. It can penetrate cell membranes and bind to the IAPs with high affinity. Nanomolar concentrations of C3 promote both TRAIL-induced and TNF-α–induced apoptosis in human cancer cells ( 21). However, it remains unclear how SMAC and SMAC mimetics potentiate apoptosis induced by anticancer agents. SMAC release occurs after mitochondrial outer membrane permeabilization, a process thought to happen after commitment to cell death ( 24).

Although numerous efforts have been made toward developing therapeutic agents, manipulation of apoptotic pathways for chemoprevention has not been extensively explored. Our previous studies showed that SMAC plays an essential role in NSAID-induced apoptosis in colon cancer cells ( 25). SMAC is consistently released from the mitochondria into the cytosol during NSAID-induced apoptosis. Deletion of SMAC by homologous recombination or knockdown of SMAC by RNA interference abrogates NSAID-induced caspase activation and apoptosis ( 25). Unexpectedly, NSAID-induced cytochrome c release was also found to be compromised in the SMAC-deficient cells ( 25). In this study, we further investigated how SMAC functions in NSAID-induced apoptosis, as well as the effects of small-molecule SMAC mimetics on NSAID-induced apoptosis and NSAID resistance.

Materials and Methods

Cell culture and drug treatment. The colorectal cancer cell lines used in the study, including HCT116, HT29, and DLD1, were obtained from American Type Culture Collection. SMAC-knockout (SMAC-KO) and BAX-knockout (BAX-KO) HCT116 cell lines were previously described ( 25, 26). All cell lines were cultured in McCoy's 5A media (Invitrogen) supplemented with 10% defined fetal bovine serum (Hyclone), 100 units/mL penicillin, and 100 mg/mL streptomycin (Invitrogen) and were maintained at 37°C in 5% CO₂. SMAC-reconstituted SMAC-KO cell lines were generated by transfecting SMAC-KO cells with wild-type (AVPI) or mutant SMAC (ΔA; ref. 27), followed by selection in 0.4 mg/mL G-418 (Invitrogen). To induce apoptosis, cells were plated at 20% to 30% density in 12-well plates and allowed to attach overnight. DMSO stock solutions of sulindac sulfide (Merck) at 40 mmol/L and indomethacin (Sigma) at 100 mmol/L were freshly prepared and diluted into appropriate concentrations by cell culture medium. Different concentrations of sulindac and indomethacin were used for apoptosis induction in different colon cancer cell lines and were determined based on their dose responses, as previously described ( 25). The SMAC mimetic and control compounds, C3 and C4, respectively ( 21), were provided by Dr. Xiaodong Wang and Dr. Patrick G. Harran at University of Texas Southwestern Medical Center. An independent set of SMAC mimetic testing (GT-T) and control (GT-C) compounds were supplied by TetraLogic Pharmaceuticals. All compounds were prepared as 100 μmol/L stock solutions in DMSO. In some cases, caspase-3 inhibitor Z-DEVD-fmk (20 μmol/L; R & D Systems) was used to treat cells in combination with the agents described above.

Western blotting. The antibodies used for Western blotting included those against caspase-9, Bid (Cell Signaling Technology), cytochrome c, α-tubulin (BD Biosciences), caspase-3 (Stressgen Bioreagents), cytochrome oxidase subunit IV (Cox IV; Invitrogen), Bim, and SMAC (EMD Biosciences). Western blotting analysis was performed as previously described ( 28).

Apoptosis assays. After treatment, attached and floating cells were harvested at various time points. The fractions of apoptotic cells were evaluated by nuclear staining and Annexin V/propidium iodide staining. For nuclear staining, cells were fixed in a solution containing 3.7% formaldehyde, 0.5% Nonidet P40, and 10 μg/mL 4′,6-diamidino-2-phenylindole, dihydrochloride in PBS and assessed for apoptosis through microscopic visualization of condensed chromatin and micronucleation. For each measurement, at least three independent experiments and a minimum of 300 cells were analyzed. For Annexin V/propidium iodide staining, cells were stained by propidium iodide and then Annexin V-Alexa 594 (Invitrogen) according to the manufacturer's instructions. Flow cytometry was used to quantitate Annexin V–positive and propidium iodide–positive cells. Long-term cell survival was evaluated by colony formation assays. In brief, harvested cells were plated in 12-well plates at appropriate dilutions. Cells were allowed to grow for 10 to 14 days before staining with Crystal Violet (Sigma).

Analysis of cytochrome c release. Mitochondrial and cytosolic fractions were isolated from treated cells by differential centrifugation as previously described ( 29). Concentrations of cytosolic fractions obtained from different samples were normalized using protein assay dye reagent from Bio-Rad. All fractions were mixed with equal volumes of 2× Laemmli sample buffer for Western blotting analysis.

Caspase-3 RNA interference. Two caspase-3 small interfering RNA (siRNA) duplexes (Cas-3 986, 5′-UGA GGU AGC UUC AUA GUG Gtt-3′; Cas-3 182, 5′-TGA CAT CTC GGT CTG GTA Ctt-3′) were purchased from Dharmacon, Inc., as 20 μmol/L stock solutions. Cells were transfected with siRNA at 100 nmol/L final concentration by Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. Down-regulation of caspase-3 expression was verified by Western blotting.

Statistical analysis. Statistical analyses were performed using GraphPad Prism IV software. The averages ± 1 SD were displayed in the figures.

Results

NSAID-induced cytochrome c release is delayed and attenuated in SMAC-KO cells. Our previous study showed that SMAC deficiency perturbed NSAID-induced release of cytochrome c and apoptosis-inducing factor in HCT116 colon cancer cells ( 25). To further characterize how SMAC mediates NSAID-induced apoptosis, we analyzed the time courses of apoptotic events in HCT116 cells treated with sulindac at 120 μmol/L, which is the minimum concentration that is necessary to induce a high level (>70%) of apoptosis in these cells ( 25). A burst of SMAC release was detected as early as 8 h after the treatment ( Fig. 1A ), long before cells started to show morphologic signs of apoptosis. Caspase-3 was dissociated from IAP family members cIAP-1 and cIAP-2, but not XIAP, as early as 8 h after sulindac treatment ( Fig. 1B and data not shown). At 12 to 16 h, caspase-3 was found to be activated ( Fig. 1C), but a very low level (<5%) of nuclear fragmentation could be detected (data not shown). In contrast, the majority of cytochrome c release occurred at 24 and 32 h after the treatment ( Fig. 1D) when high levels (>50%) of nuclear fragmentation could be detected ( 25, 26). However, sulindac-induced cytochrome c release and nuclear fragmentation were significantly delayed and attenuated in the SMAC-KO cells compared with the parental HCT116 cells ( Fig. 1D and data not shown; ref. 25). These results showed that SMAC release and cytochrome c–independent caspase-3 activation are early events in NSAID-induced apoptosis and regulate later apoptotic events, such as cytochrome c release and nuclear fragmentation.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

SMAC and cytochrome c release during NSAID-induced apoptosis. A, time course of SMAC release. HCT116 cells were treated with 120 μmol/L sulindac. Cytosolic fractions at the indicated time points were isolated, and the expression of SMAC was analyzed by Western blotting. B, after HCT116 cells were treated with 120 μmol/L sulindac, the association between caspase-3 and cIAP-1 or cIAP-2 at indicated time points was analyzed by immunoprecipitation (IP) with cIAP-1 or cIAP-2 antibodies, respectively. C, time course of caspase-3 activation. Caspase-3 activation was analyzed by Western blotting at the indicated time points after sulindac treatment. D, cytochrome c release in HCT116 and SMAC-KO cells. Cytochrome c release in HCT116 and SMAC-KO cells treated with 120 μmol/L sulindac was analyzed by Western blotting. Cox IV and α-tubulin, which are exclusively expressed in the mitochondria and cytosol, respectively, were used as controls for loading and fractionation.

IAP binding is required for SMAC-mediated caspase-3 activation and cytochrome c release. To further study how SMAC mediates cytochrome c release, we tested whether the IAP inhibition and/or caspase-3 activation functions of SMAC, which are mediated by the N-terminal AVPI domain of cytosolic SMAC ( 30), are necessary for NSAID-induced apoptosis and cytochrome c release. Wild-type (AVPI) or mutant SMAC containing a deletion of alanine in the AVPI domain (ΔA), which abolishes the interactions between SMAC and IAPs ( 30), was transfected into the SMAC-KO cells ( Fig. 2A ). Two independent stable cell lines expressing the wild-type or mutant SMAC were analyzed for their responses to 120 μmol/L sulindac. The deficiency in NSAID-induced apoptosis in the SMAC-KO cells was completely rescued by the expression of the wild-type, but not the mutant, SMAC ( Fig. 2B). Expression of the wild-type SMAC also rescued caspase-3 activation in the SMAC-KO cells ( Fig. 2C). Importantly, wild-type, but not the mutant, SMAC restored sulindac-induced cytochrome c release to a level similar to that in the parental HCT116 cells ( Fig. 2D). These results suggest that inhibition of IAPs by cytosolic SMAC through their direct interactions promotes cytochrome c–independent caspase-3 activation at early stage and cytochrome c release at later stages in NSAID-induced apoptosis.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

The IAP interacting function of SMAC is necessary for its activity in NSAID-induced apoptosis. A, SMAC-KO cells were transfected with wild-type (AVPI) or mutant SMAC containing a deletion of alanine in the AVPI domain (ΔA). Two pairs of clones with stable expression of each SMAC construct were isolated, and SMAC expression was verified by Western blotting. B, the indicated cell lines were treated with 120 μmol/L sulindac for 48 h. Apoptosis was analyzed by nuclear staining. C, effects of wild-type and mutant SMAC on NSAID-induced caspase activation were analyzed by Western blotting. D, effects of wild-type and mutant SMAC on NSAID-induced cytochrome c release were analyzed by Western blotting.

Caspase-3 activation is necessary for SMAC-mediated cytochrome c release. Next, we determined whether caspase-3 activation is critical for NSAID-induced and SMAC-mediated apoptosis and cytochrome c release. Cells were transiently transfected with two caspase-3 siRNA (Cas-3 986 and Cas-3 182) before sulindac treatment. Knockdown of caspase-3 expression resulted in a significant reduction of sulindac-induced apoptosis in HCT116 cells ( Fig. 3A and B ). However, no further decrease in apoptosis was observed after knockdown of caspase-3 in the SMAC-KO cells ( Fig. 3B), suggesting that SMAC and caspase-3 function in a linear pathway to promote NSAID-induced apoptosis. Caspase-3 knockdown also led to significant inhibition of cytochrome c release ( Fig. 3C). The critical role of caspase-3 in promoting sulindac-induced cytochrome c release was confirmed by treating HCT116 and another colon cancer cell line HT29 using the caspase-3 inhibitor Z-DEVD-fmk ( Fig. 3C).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Caspase-3 is necessary for NSAID-induced cytochrome c release. A, siRNA knockdown of caspase-3. Caspase-3 expression was probed by Western blotting 24 h after HCT116 cells were transfected with indicated siRNA. Ctrl, control scrambled siRNA. B, effects of caspase-3 knockdown on sulindac-induced apoptosis. Parental and SMAC-KO HCT116 cells were transfected with caspase-3 or control siRNA for 24 h followed by sulindac (120 μmol/L) treatment. Apoptosis was analyzed by nuclear staining 48 h after sulindac treatment. C, effects of caspase-3 inhibition on NSAID-induced cytochrome c release. HCT116 and HT29 cells were transfected with control scrambled or caspase-3 siRNA or treated with caspase-3 inhibitor Z-DEVD-fmk for 24 h followed by sulindac (120 μmol/L for HCT116; 180 μmol/L for HT29) treatment. Cytochrome c release 24 h after the treatment was analyzed. D, expression of Bid and Bim in sulindac-treated cells. After treatment of the parental, BAX-KO and SMAC-KO HCT116 cells with 120 μmol/L sulindac for 24 h, the cytosolic and mitochondrial fractions were analyzed for Bid and Bim expression by Western blotting. Arrows, Bid and Bim cleavage fragments.

The dependence of mitochondrial cytochrome c release on the cytosolic actions of SMAC and caspase-3 suggested a protein, presumably a caspase or a caspase substrate, could translocate from the cytosol to the mitochondria to induce cytochrome c release. We first considered caspase-3 itself, but did not detect any mitochondrial accumulation of caspase-3 after sulindac treatment (data not shown). Another candidate is the BH3-only Bcl-2 family member Bid, which is cleaved by caspases in the cytoplasm ( 31). The truncated tBid can translocate to the mitochondria to induce cytochrome c release ( 31). We found that Bid was indeed cleaved and tBid was enriched in the mitochondria after sulindac treatment ( Fig. 3D). However, the cleavage and translocation of Bid were also detected at the same level in the BAX-KO and SMAC-KO cells ( Fig. 3D), which are deficient in NSAID-induced apoptosis and cytochrome c release ( 25, 26), excluding Bid as the candidate for regulating SMAC-mediated cytochrome c release. We also found that Bim, another BH3-only protein and a caspase substrate that can translocate to the mitochondria ( 32), underwent similar proteolytic cleavage ( Fig. 3D). Although the shorter forms of Bim were enriched in the mitochondria of sulindac-treated HCT116 cells, no difference was observed in the BAX-KO and SMAC-KO cells ( Fig. 3D). Therefore, Bid and Bim did not seem to play a role in NSAID-induced and SMAC-mediated cytochrome c release.

SMAC mimetics potentiate NSAID-induced apoptosis in colon cancer cells. The critical role of SMAC in NSAID-induced apoptosis suggests that manipulation of the SMAC pathway may improve the effects of NSAIDs. The requirement of the AVPI domain for the SMAC function prompted us to test whether pharmacologic agents that mimic this domain can enhance NSAID-induced apoptosis. We tested C3, a small-molecule mimic of the AVPI residues of SMAC, and its control compound C4, which differs from C3 by a single methyl group ( 21). It was previously shown that C3, but not C4, at nanomolar concentrations can sensitize human cancer cells to TRAIL or TNF-α–induced apoptosis ( 21). To observe potential sensitizing effects, NSAIDs were used at a lower concentration (100 μmol/L) for HCT116 cells. We found that these compounds alone, even at micromolar concentrations, did not have any growth inhibitory or apoptotic effects on colon cancer cells ( Fig. 4A and data not shown). However, only 100 nmol/L of C3 was sufficient to markedly enhance NSAID-induced apoptosis in HCT116 cells, with the percentage of apoptotic cells increasing from 32% to 61% and from 22% to 59% after treatment for 48 h with 100 μmol/L sulindac and 250 μmol/L indomethacin, respectively ( Fig. 4A). In contrast, the control C4 compound did not have any effect on apoptosis ( Fig. 4A). Similar observations were also made in HT29 and DLD1 colon cancer cells ( Fig. 4A). Combinations of C3 with sulindac (50–100 μmol/L) or indomethacin (100–250 μmol/L) at different concentrations all led to 80% to 120% increase in apoptosis in HCT116 cells (data not shown). To verify these results, we also analyzed an independent set of active and control SMAC mimetic compounds, GT-T and GT-C, and observed similar results ( Fig. 4A). Analysis of apoptosis using Annexin V/propidium iodide staining confirmed that NSAID-induced apoptosis is enhanced by the active C3 or GT-T compound but not by the control C4 or GT-C compound (Supplementary Fig. S1). Furthermore, colony formation assay results showed that combinations of NSAIDs with the active compounds were much more effective in inhibiting long-term survival of HCT116 and HT29 cells compared with NSAIDs alone or their combinations with the control compounds ( Fig. 4B). Over 50% of the cells that could not be killed by sulindac or indomethacin alone were eliminated by the combination with C3 or GT-T ( Fig. 4B). These results showed strong cooperative effects of NSAIDs and SMAC mimetics in killing colon cancer cells.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

SMAC mimetics potentiate NSAID-induced apoptosis. A, HCT116, HT29, and DLD1 colon cancer cells were treated with NSAIDs with or without 100 nmol/L control (C4 or GT-C) or test (C3 or GT-T) SMAC mimetic compounds for 48 h. Apoptosis induction was determined by nuclear staining. B, long-term cell viability was assessed by colony formation assay as described in Materials and Methods. Left, representative pictures of cell colonies. Right, quantitation of colony numbers.

SMAC mimetics overcome NSAID resistance in the SMAC-KO and BAX-KO cells. The majority of HCT116 cells surviving from NSAID treatment contain BAX mutations and are defective in SMAC release ( 25, 26). Acquired NSAID resistance was found to be associated with overexpression of antiapoptotic Bcl-2 family proteins ( 33, 34). Because SMAC functions downstream of Bax and other Bcl-2 family members, we asked whether manipulating SMAC can restore NSAID sensitivity in NSAID-resistant colon cancer cells that are deficient in Bax or SMAC. It was shown that cytosolic, but not mitochondrial, SMAC can bypass the requirement of activating Bcl-2 family proteins to activate caspases ( 17, 19). We therefore transiently transfected a cytosolic SMAC expression construct into the BAX-KO cells ( Fig. 5A ), which are completely resistant to NSAIDs even at very high concentrations ( 26). To maximize the apoptotic effect so that molecular markers of apoptosis could be analyzed, we used a higher concentration of sulindac (200 μmol/L) or indomethacin (800 μmol/L) than those used for other cell lines to treat the BAX-KO cells. We found that sulindac-induced apoptosis was indeed restored by the cytosolic SMAC in the BAX-KO cells ( Fig. 5A). Remarkably, the active SMAC mimetic compounds C3 and GT-T, but not the control compounds C4 and GT-C, also restored apoptosis induced by sulindac or indomethacin in both the BAX-KO and SMAC-KO cells ( Fig. 5B). These results were confirmed by analysis of apoptosis using Annexin V/propidium iodide staining (Supplementary Fig. S2). They suggest that activation of the apoptotic pathway by SMAC or SMAC mimetics can overcome NSAID resistance in colon cancer cells.

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

SMAC mimetics restore NSAID sensitivity in the SMAC-KO and BAX-KO cells. A, BAX-KO cells were transfected with the cytosolic SMAC expression construct (p-cSMAC) or control empty vector. Left, the expression of Flag-tagged cytosolic SMAC 24 h after transfection was confirmed by Western blotting. Right, cells were treated with 200 μmol/L sulindac for 48 h. Apoptosis was analyzed by nuclear staining. B, SMAC-KO and BAX-KO HCT116 cells were treated with 100 nmol/L SMAC mimetic compounds alone or in combination with sulindac or indomethacin at indicated concentrations for 48 h. Apoptosis was analyzed by nuclear staining.

The effects of SMAC mimetics depend on caspase-3 and involve cytochrome c release. We then investigated the mechanisms by which SMAC mimetics potentiate NSAID-induced apoptosis. We found that 100 μmol/L sulindac failed to induce robust caspase-3 and caspase-9 activation in HCT116 cells ( Fig. 6A ), consistent with the low level of apoptosis induced by sulindac at this concentration ( Fig. 4A). However, in combination with SMAC mimetic compounds, 100 μmol/L sulindac strongly induced accumulation of active caspase-3 and caspase-9 ( Fig. 6A). Although sulindac even at 200 μmol/L could not induce caspase processing in the BAX-KO cells, SMAC mimetics rescued caspase-3 and caspase-9 activation in these cells ( Fig. 6B). In agreement with the regulation of cytochrome c release by SMAC, treatment of HCT116 cells with sulindac combined with the active SMAC mimetics markedly enhanced cytochrome c release in HCT116 cells and also restored cytochrome c release in the BAX-KO cells ( Fig. 6A and B). To test whether these effects of SMAC mimetics require caspase-3 activation, caspase-3 siRNA was transfected into cells before the combination treatment. Caspase-3 knockdown resulted in a significant reduction of apoptosis in the cells treated with the combinations of sulindac and SMAC mimetics, which was much more pronounced than the reduction of apoptosis seen in the cells treated with sulindac alone ( Fig. 6C), indicating that the sensitizing effects of the SMAC mimetics are at least in part mediated by caspase-3. Furthermore, caspase-3 knockdown also led to attenuation of cytochrome c release induced by the combination of sulindac and SMAC mimetics ( Fig. 6D). These results suggested that SMAC mimetics potentiate NSAID-induced apoptosis and overcome NSAID resistance by promoting caspase-3 activation, thereby enhancing cytochrome c release.

• Download figure
• Open in new tab
• Download powerpoint

Figure 6.

SMAC mimetics enhance caspase activation and cytochrome c release. A, HCT116 cells were treated with 100 μmol/L sulindac with or without a combination with 100 nmol/L of SMAC mimetic compounds. Top, caspase-3 and caspase-9 activation was analyzed by Western blotting; bottom, cytochrome c release was determined by Western blotting. B, BAX-KO cells were treated with 200 μmol/L sulindac with or without a combination with 100 nmol/L of SMAC mimetic compounds for 36 h. Top, caspase-3 and caspase-9 activation was analyzed by Western blotting; bottom, cytochrome c release was determined by Western blotting. C, effects of caspase-3 knockdown on the sensitizing effects of SMAC mimetics. HCT116 cells were transfected with control scrambled or caspase-3 siRNA 24 h before the indicated treatments. Apoptosis induction was determined by nuclear staining 48 h after sulindac treatment. D, effects of caspase-3 inhibition on cytochrome c release 24 h after sulindac treatment.

Discussion

In this study, we found that SMAC release and cytochrome c release proceed through different time courses during NSAID-induced apoptosis, with bulk of the SMAC release occurring much earlier compared with that of cytochrome c release. The majority of cytochrome c release was abolished in the SMAC-KO cells, suggesting that SMAC plays an important role in promoting cytochrome c release and full execution of NSAID-induced apoptosis. Studies by other groups also showed that the release of different mitochondrial apoptogenic proteins is coordinated and occurs through different time courses ( 35). The IAP-interacting activity of SMAC, which is mediated by the N-terminal AVPI domain, is critical for its ability to promote caspase activation and cytochrome c release during NSAID-induced apoptosis. Because IAP inhibition occurs in the cytoplasm whereas cytochrome c is located in the mitochondria before its release, a feedback loop involving protein translocation seems to drive the events leading to cytochrome c release. Our previous study has described a similar function of SMAC in DNA damage–induced apoptosis mediated by the BH3-only protein PUMA ( 28). However, the requirement of SMAC for full execution of apoptosis does not seem to be general, as only few classes of agents among a number of stimuli tested, including NSAIDs, PUMA, and TRAIL, were found to be dependent on SMAC to induce apoptosis ( 25). Under different conditions, it is possible that cytochrome c plays a major role in driving this feedback amplification loop ( 36).

Using siRNA and inhibitor approaches, we showed that caspase-3 activation is a critical event in this feedback loop. Several previous studies have also shown the functions of other caspases, including caspase-2 and caspase-7, in promoting mitochondrial events during apoptosis ( 37– 39). However, the link between caspase-3 activation and cytochrome c release remains to be identified. We did not detect any mitochondrial accumulation of caspase-3 in HCT116 cells undergoing sulindac-induced apoptosis (data not shown), excluding the possibility that caspase-3 itself directly induces cytochrome c release. Analysis of proapoptotic caspase substrates ruled out the BH3-only proteins Bid and Bim as the potential candidates, because their cleavage and translocation were intact in the BAX-KO and SMAC-KO cells that are deficient in NSAID-induced apoptosis. Several other proteins that can translocate from the cytosol to the mitochondria and promote cytochrome c release remain to be examined. It is also possible that more than one protein mediate the intermediate steps between caspase-3 activation and cytochrome c release. To identify the mediators of this feedback loop, it might be useful to compare the mitochondrial components in the wild-type and SMAC-KO cells using an unbiased proteomic approach.

Our results obtained from two independent sets of compounds showed for the first time that SMAC mimetics can greatly potentiate the anticancer effects of NSAIDs in both short-term apoptotic and long-term clonogenic survival assays. Other studies have described similar effects of SMAC and SMAC mimetics on chemotherapeutic drugs and irradiation ( 17, 40– 42). SMAC mimetics not only greatly enhanced the apoptotic effects of sulindac and indomethacin in colon cancer cells with different genetic backgrounds but also restored apoptosis in the NSAID-resistant SMAC-KO and BAX-KO cells. Our data on BAX-KO cells were obtained using higher concentrations of NSAIDs (200 μmol/L sulindac and 800 μmol/L indomethacin) than those used for other cell lines. When SMAC mimetics were combined with NSAIDs at lower concentrations (120 μmol/L sulindac and 500 μmol/L indomethacin), we also detected ∼10% apoptosis in BAX-KO cells (data not shown). The mechanistic studies showed that SMAC mimetics enhanced NSAID-induced caspase-3 activation and cytochrome c release, thereby promoting the feedback loop for full execution of NSAID-induced apoptosis. Importantly, the effects of SMAC mimetics are caspase-3–dependent, suggesting that caspases play a major role in mediating this feedback loop, also reinforcing that the sensitizing effects are not due to off-target activities of SMAC mimetics ( 21). It will be of interest to test whether previously reported activities of SMAC mimetics also operate through similar feedback mechanisms.

Extensive efforts have been made toward developing therapeutic agents through manipulation of apoptotic pathways ( 43). For example, ABT-737, a compound that mimics the BH3 domains of Bcl-2 family proteins, has shown great promise in early preclinical studies ( 44). However, manipulation of apoptotic pathways for chemoprevention has not been extensively explored. Our results suggest that SMAC mimetics are potentially useful as sensitizers for NSAIDs and possibly for other chemopreventive agents by boosting their proapoptotic activities and also by restoring apoptosis in otherwise resistant neoplastic cells. It is worth noting that the concentrations of SMAC mimetics were less than one of 1,000 of those of NSAIDs used in our study. The side effects of NSAIDs, presumably caused by high drug doses used in prevention studies, have presented a significant challenge for using these agents for cancer prevention ( 4). It is possible that combinations of NSAIDs with SMAC mimetics will help to enhance chemopreventive efficacy, while reducing dose and decreasing toxicity. This possibility can be tested using animal models, such as Apc^Min/+ mice.