SMAC/Diablo mediates the proapoptotic function of PUMA by regulating PUMA-induced mitochondrial events

ORIGINAL ARTICLE SMAC/Diablo mediates the proapoptotic function of PUMA by regulating PUMA-induced mitochondrial events J Yu, P Wang, L Ming, MA Wood and L Zhang (2007) 26, 4189 4198 & 2007 Nature Publishing Group All rights reserved 0950-9232/07 $30.00 www.nature.com/onc Departments of Pharmacology and Pathology, University of Pittsburgh Cancer Institute, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA p53-upregulated modulator of apoptosis (PUMA) is a BH3-only Bcl-2 family protein and an essential mediator of DNA damage-induced apoptosis. PUMA is localized in the mitochondria and induces apoptosis through the mitochondrial pathway. However, the mechanisms of PUMA-induced apoptosis remain unclear. In this study, we found that second mitochondria-derived activator of caspase (SMAC)/Diablo, a mitochondrial apoptogenic protein, mediates the proapoptotic function of PUMA by regulating PUMA-induced mitochondrial events. SMAC is consistently released into the cytosol in colon cancer cells undergoing PUMA-induced apoptosis. In SMACdeficient cells, execution of PUMA-induced apoptosis is abrogated, in company with decreases in caspase activation, cytosolic release of cytochrome c and collapse of mitochondrial membrane potential. Reconstituting SMAC expression restored these events in the SMAC-deficient cells. Furthermore, SMAC and agents that mimic the inhibitor of apoptosis proteins (IAPs) inhibition function of SMAC significantly sensitize cells to PUMA-induced apoptosis. These results demonstrate an important role of SMAC in executing DNA damage-induced and PUMAmediated apoptosis and suggest that SMAC participates in a feedback amplification loop to promote cytochrome c release and other mitochondrial events in apoptosis. (2007) 26, 4189 4198; doi:10.1038/sj.onc.1210196; published online 22 January 2007 Keywords: SMAC/Diablo; PUMA; apoptosis; mitochondria; DNA damage Introduction p53-upregulated modulator of apoptosis (PUMA), a BH3-only Bcl-2 family member, plays an essential role in apoptosis induced by a variety of stimuli. In response to DNA-damaging agents, such as chemotherapeutic Correspondence: Dr L Zhang, University of Pittsburgh Cancer Institute, Hillman Cancer Center Research Pavilion, Suite 2.42D, 5117 Centre Ave., Pittsburgh, PA 15213, USA. E-mail: zhanglx@upmc.edu Received 10 April 2006; revised 3 October 2006; accepted 9 November 2006; published online 22 January 2007 drugs and ionizing radiation, p53 tumor suppressor activates the transcription of PUMA to induce apoptosis (Nakano and Vousden, 2001; Yu et al., 2001). Other apoptotic stimuli can activate PUMA independent of p53 (Han et al., 2001). Deletion of PUMA in human cancer cells abrogates apoptosis induced by p53, DNAdamaging agents, hypoxia, endoplasmic reticulum (ER) stress and HIV proteins (Reimertz et al., 2003; Yu et al., 2003; Perfettini et al., 2004). In mice, PUMA is required for p53-dependent apoptosis induced by g-irradiation, DNA-damaging drugs, c-myc and E1A oncogenes, as well as p53-independent apoptosis induced by cytokine withdrawal, glucocorticoids, kinase inhibitors and phorbol esters (Jeffers et al., 2003; Villunger et al., 2003). Deficiency in PUMA promotes oncogenic transformation and accelerates lymphomagenesis induced by the em-myc oncogene, suggesting a role of PUMA in tumor suppression (Hemann et al., 2004). PUMA is localized in the mitochondria and regulates apoptosis through other members of the Bcl-2 family. PUMA interacts with antiapoptotic proteins Bcl-2 and Bcl-X L, and is dependent on Bax to induce apoptosis (Yu et al., 2001, 2003). In response to PUMA expression, Bax translocates from the cytosol into the mitochondria and forms multimers in the mitochondria (Yu et al., 2003). PUMA also promotes collapse of mitochondrial membrane potential (Yu et al., 2003). However, it remains unclear how these mitochondriarelated events are coordinated to initiate PUMAinduced apoptosis. During apoptosis execution in mammalian cells, several apoptogenic proteins are released from the mitochondria into the cytosol to trigger downstream apoptotic events (Green and Reed, 1998; Wang, 2001; Cory et al., 2003; Danial and Korsmeyer, 2004). For example, cytochrome c is released to form, along with Apaf-1 and caspase-9, an apoptosome and initiates caspase activation cascade (Wang, 2001). Release of apoptosis-inducing factor (AIF) leads to chromatin condensation and nuclear fragmentation (Susin et al., 1999). Second mitochondria-derived activator of caspase (SMAC), also called direct inhibitor of apoptosis proteins (IAP)-binding protein with low pi (Diablo), was also found to be released into the cytosol of apoptotic cells (Du et al., 2000; Verhagen et al., 2000). SMAC/Diablo (hereafter referred to as SMAC) interacts with and antagonizes IAPs, such as XIAP, ciap1

4190 and ciap2 (Verhagen et al., 2000). SMAC expression and agents that mimic the IAP interacting function of SMAC sensitize human cancer cells to apoptosis induced by several anticancer agents, such as tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) (Fulda et al., 2002; Arnt and Kaufmann, 2003; Li et al., 2004). Although substantial evidence suggests that SMAC plays an important role in execution of apoptosis, the precise function of SMAC remains obscure. In SMAC-deficient mice, cell death induced by a variety of stimuli was not found to be altered (Okada et al., 2002). Furthermore, the mechanisms by which SMAC and other mitochondrial apoptogenic proteins are released into the cytosol during execution of apoptosis are poorly understood. Our previous study in colon cancer cells demonstrated the involvement of SMAC in apoptosis induced by nonsteroidal anti-inflammatory drugs (NSAIDs) (Kohli et al., 2004). Studies by other groups suggest the existence of a caspase-mediated feedback amplification loop in regulating cytochrome c release and other mitochondrial events in apoptosis (Chen et al., 2000; Lakhani et al., 2006). These studies prompted us to investigate the role of SMAC in PUMA-induced apoptosis, as well as its potential involvement in PUMA-induced mitochondrial events during execution of apoptosis. Results PUMA induces cytochrome c, SMAC and AIF release in colon cancer cells Our previous studies demonstrated that PUMA-induced apoptosis in HCT116 colon cancer cells is dependent on Bax and occurs through a mitochondrial pathway (Yu et al., 2001, 2003). To study Bax-mediated downstream events, we infected parental HCT116 cells and BAXknockout (BAX-KO) cells with an adenovirus expressing PUMA (Ad-PUMA), and analysed the cytosolic release of cytochrome c, SMAC and AIF. All three proteins were found to be enriched in the cytosol of HCT116 cells undergoing PUMA-induced apoptosis, but not present in the cytosol of BAX-KO cells resistant to apoptosis (Figure 1a). The Bax-dependent release of cytochrome c, SMAC and AIF was not unique for Ad- PUMA treatment, and also observed in cells undergoing apoptosis induced by the DNA-damaging agent etoposide (Figure 1b). We then examined whether apoptosis mediated by endogenous PUMA also involves the release of mitochondrial apoptogenic proteins. Our previous study showed that Adriamycin, a DNA-damaging agent commonly used in chemotherapy, could induce apoptosis in HCT116 cells in the absence of the cyclindependent kinase (CDK) inhibitor p21, and such apoptosis is strictly PUMA-dependent (Yu et al., 2003). Our recent study showed that PUMA is responsible for Adriamycin-induced Bax activation and subsequent mitochondrial changes (Ming et al., 2006). Therefore, p21-knockout (p21-ko) and p21/ PUMA double knockout (p21-ko/puma-ko) HCT116 cells were compared following Adriamycin treatment. We found that in the p21-ko cells, but not in the p21-ko/puma-ko cells, Adriamycin could induce the release of cytochrome c, SMAC and AIF (Figure 1c), as well as mitochondrial membrane permeabilization (Figure 1d), suggesting that Baxdependent release of cytochrome c, SMAC and AIF plays an important role in the execution of PUMAmediated apoptosis. To further study the role of mitochondrial apoptogenic proteins in PUMA-induced apoptosis, we compared HCT116 and four additional colorectal cancer cell lines. In each case, cytochrome c, SMAC and AIF were found to be released into the cytosol in the cells undergoing apoptosis following Ad-PUMA infection (Figure 1e). However, the release of SMAC appeared to be more significant compared with that of cytochrome c or AIF, with over 90% of mitochondrial SMAC released 48 h after Ad-PUMA infection in each of the five cell lines (Figure 1e). Interestingly, the amount of cytosolic SMAC seemed to correlate best with the extent of apoptosis in these cells (Figure 1e). These results suggest a consistent involvement of mitochondrial apoptogenic proteins and a prominent role of SMAC in execution of PUMA-mediated apoptosis in colon cancer cells. PUMA-induced apoptosis is inhibited in SMAC-deficient cells, but not in AIF-deficient cells To determine whether SMAC release is relevant, or just a coincidental event in PUMA-induced apoptosis, we analysed SMAC-knockout (SMAC-KO) HCT116 cells that we previously created by homologous recombination (Kohli et al., 2004). After Ad-PUMA infection, significantly less apoptosis was detected by nuclear staining in the SMAC-KO cells compared with that in the parental cells 48 h after treatment (18 vs 68%; Figure 2a and b). Analysis of apoptosis by different methods, including cell cycle analysis and Annexin V staining, confirmed that PUMA-induced apoptosis is inhibited in the SMAC-KO cells (Figure 2c and d). If these observations were indeed due to the deficiency in SMAC, we would expect that PUMA-induced apoptosis can be restored by reconstitution of SMAC expression. Therefore, a SMAC-reconstituted cell line (SMAC-R) we previously generated was analysed for its response to PUMA (Kohli et al., 2004). PUMA-induced apoptosis was found to be completely restored by the reconstituted SMAC in the SMAC-KO cells (Figure 2a d). To rule out the possibility that these observations are due to the specific genetic background of HCT116 cells, we analysed SMAC-deficient DLD1 colon cancer cells (SMAC-KD) generated by stable transfection of small interference RNA (sirna) (Kohli et al., 2004). Knockdown of SMAC in DLD1 cells led to a significant decrease in PUMA-induced apoptosis, as determined by nuclear staining, Annexin V staining and cell cycle analysis (Figure 3 and data not shown). To test if the effect of SMAC on PUMA-induced apoptosis is specific, RNA interference (RNAi) was used

4191 Figure 1 PUMA induces the release of mitochondrial apoptogenic proteins in colon cancer cells. (a) Bax-dependent release of mitochondrial apoptogenic proteins induced by PUMA. HCT116 and BAX-KO cells were infected with Ad-PUMA for 48 h. Mitochondrial and cytosolic fractions were isolated and examined for SMAC, cytochrome c and AIF expression by Western blotting. (b) Bax-dependent release of mitochondrial apoptogenic proteins induced by etoposide. HCT116 and BAX-KO cells were treated with etoposide (200 mm) for 48 h. Mitochondrial and cytosolic fractions were isolated and examined for SMAC, cytochrome c and AIF expression by Western blotting. (c) PUMA-mediated SMAC release during Adriamycin-induced apoptosis. p21-ko and p21-ko/ PUMA-KO HCT116 cells were treated with Adriamycin (0.2 mg/ml) for 48 h. SMAC expression was examined by Western blotting as in (b). (d) PUMA-dependent mitochondrial membrane permeabilization. p21-ko and p21-ko/puma-ko cells were harvested 48 h after Adriamycin treatment, stained by MitoTracker Red (CMXROS) and analysed by flow cytometry as described in the Materials and methods. (e) PUMA-induced SMAC release in colon cancer cells. Indicated colon cancer cell lines were infected with Ad-PUMA for 48 h to induce apoptosis. Left panel: cytosolic and mitochondrial fractions were isolated and examined for SMAC expression by Western blotting. Right panel: apoptosis was analysed by nuclear staining. The results are the average of three independent experiments with one standard deviation shown. Cox IV and a-tubulin, which are exclusively expressed in the mitochondria and cytosol, respectively, were used as controls for loading and fractionation. to knock down the expression of AIF in HCT116 cells as described in the Materials and methods. Two stable cell lines with markedly decreased expression of AIF were isolated after puromycin selection (Figure 4a). Both cell lines were found to be as sensitive as the parental HCT116 cells to PUMA-induced apoptosis (Figure 4b), suggesting that the role of AIF may not be as important as that of SMAC in execution of PUMA-induced apoptosis. Caspase activation, cytochrome c release and mitochondrial membrane potential collapse are inhibited in the SMAC-deficient cells It was shown that following its release from the mitochondria into the cytosol, SMAC promotes caspase activation (Du et al., 2000; Verhagen et al., 2000). We therefore examined whether caspases, including caspases-3 and caspases-9, can still be cleaved in the SMAC-deficient cells. Upon Ad-PUMA infection,

4192 Figure 2 PUMA-induced apoptosis is inhibited in SMAC-knockout HCT116 cells. Parental HCT116, BAX-knockout (BAX-KO), SMAC-knockout (SMAC-KO) and SMAC-KO reconstituted with SMAC (SMAC-R) cells were infected with Ad-PUMA and a control adenovirus expressing BH3 domain-deleted PUMA (Ad-DBH3) for 48 h. (a) Apoptosis quantified by nuclear staining. Cells were fixed and analysed by fluorescence microscopy after PI staining. The results are the average of three independent experiments with error bars representing one standard deviation. (b) Cell and nuclear morphology after treatment. Cells were visualized with phase contrast (phase) and fluorescence microscopy following PI staining. Apoptotic cells contain fragmented and condensed nuclei, with examples indicated by arrows. (c) Cell cycle analysis. Cells stained by PI were analysed by flow cytometry for cell cycle. The percentages of sub-g 1 (apoptotic) cells were indicated. (d) Apoptosis analysed by Annexin V staining. Cells were stained by Annexin V, counterstained by DAPI and then analysed by flow cytometry. The percentages of Annexin V-positive (apoptotic) cells within two right quadrants were indicated. cleavage of both caspases was found to be significantly decreased in the SMAC-KO and SMAC-KD cells, compared with that in the parental HCT116 and DLD1 cells, respectively (Figure 5a and data not shown). Reconstitution of SMAC expression completely restored caspase activation in the SMAC-KO cells (Figure 5a). Caspase activation was not detectable in the BAX-KO cells that are deficient in SMAC release and resistant to PUMA-induced apoptosis, suggesting that SMAC regulates a large extent of PUMA-induced and Bax-dependent caspase activation (Figures 1 and 5a). Activation of caspase-3 and caspase-9 was also decreased in SMAC-KO cells undergoing apoptosis induced by the DNA-damaging agent etoposide (Figure 5b). In contrast, treating the cells with N-acetylleu-leu-norleucinal (ALLN), a proteasome inhibitor, which does not rely on SMAC to induce apoptosis (Kohli et al., 2004), led to similar extent of caspase-3 and caspase- 9 activation in the parental HCT116 and SMAC-KO cells (Figure 5b), suggesting that the role of SMAC in apoptosis is limited to certain stimuli. To understand the mechanisms by which SMAC mediates PUMA-induced caspase activation, we analysed the release of cytochrome c and mitochondrial membrane potential change in the SMAC-KO cells. We found that cytochrome c release was significantly decreased in the SMAC-KO cells compared with that in the parental cells, which correlated with changes in apoptosis and caspase activation (Figure 5c). Time course experiments revealed that the release of cytochrome c and AIF were both impaired and delayed in the SMAC-KO cells compared with that in the parental

4193 Figure 3 PUMA-induced apoptosis is inhibited in SMAC-knockdown DLD1 cells. Parental and SMAC-knockdown (SMAC-KD) DLD1 cells were infected with Ad-PUMA and Ad-DBH3 for 48 h. (a) Apoptosis analysed by nuclear staining. Upper panel: SMAC expression in DLD1 and SMAC-KD cells analysed by Western blotting. Lower panel: the percentage of apoptotic cells determined by PI staining. Results are the average of three independent experiments with one standard deviation shown. (b) Cell and nuclear morphology after treatment. Cells were visualized with phase contrast (phase) and fluorescence microscopy following PI staining. An example apoptotic cell is indicated by an arrow. (c) Cell cycle analysis. Cells stained by PI were analysed for cell cycle by flow cytometry. The percentages of sub-g 1 (apoptotic) cells were indicated. cells (Figure 5d). Furthermore, mitochondrial membrane permeabilization induced by PUMA is much reduced in the SMAC-deficient cells (Figure 5e). Mitochondrial membrane potential change was completely restored by reconstitution of SMAC expression (Figure 5e). These data suggest that SMAC mediates PUMA-induced apoptosis and Caspase activation by promoting mitochondrial membrane potential collapse and cytochrome c release. SMAC and SMAC mimetics sensitize cells to PUMA-induced apoptosis The above data suggest that activation of SMAC may enhance PUMA-mediated apoptosis. To test this possibility, we examined whether overexpression of SMAC has an effect on apoptosis induced by Ad- PUMA in HCT116 cells. Cells were transfected with SMAC or the control empty vector before Ad-PUMA infection. Indeed, transfection with SMAC, but not the control empty vector, significantly enhanced PUMAinduced apoptosis in HCT116 cells (Figure 6a). Our data suggest that SMAC is involved in a feedback amplification loop to promote cytochrome c release and other mitochondrial events in execution of apoptosis. If this is the case, we would expect that the SMAC-mediated IAP inhibition is also involved in PUMA-mediated apoptosis. To test this possibility, we examined whether pharmacological agents that mimic the IAP inhibition function of SMAC have similar effects on PUMA-induced apoptosis. A small molecule mimics the AVPI amino-acid residues of SMAC that are responsible for IAP binding (C3) and a control compound (C4), which differs from C3 by a single methyl group, were used to treat the cells (Li et al., 2004). It was previously shown that C3, but not C4, at nanomolar concentrations can bind to IAPs and sensitize human cancer cells to TRAIL and TNF-ainduced apoptosis (Li et al., 2004). When used alone at 100 nm, both compounds have no effect on cell growth and do not induce significant apoptosis (Figure 6b and data not shown). However, 100 nm of C3 significantly enhanced apoptosis induced by Ad-PUMA in HCT116 cells (Figure 6b). In contrast, PUMA-induced apoptosis was not affected by the control compound C4 (Figure 6b). Taken together, our data indicate that SMAC plays an important role in execution of DNA damage-induced and PUMA-mediated apoptosis. Our data also suggest that SMAC does so by participating in a feedback amplification loop to promote cytochrome c release and other mitochondrial events.

4194 Figure 4 PUMA-induced apoptosis in AIF-knockdown cells. (a) RNAi knockdown of AIF. HCT116 cells were transfected with an AIF RNAi construct. Puromycin-resistant clones with knockdown of AIF (AIF-KD1 and AIF-KD2) were identified by Western blotting. a-tubulin was used as a loading control. (b) PUMAinduced apoptosis in the AIF-knockdown cells. Parental, AIF-KD1 and AIF-KD2 HCT116 cells were infected with Ad-PUMA for 48 h to induce apoptosis. Apoptosis was analysed by nuclear staining. The results are the average of three independent experiments with error bars indicating one standard deviation. Discussion Understanding the mechanisms of PUMA-mediated apoptosis has important implications for anticancer therapy. PUMA is a p53 target and an essential mediator of DNA damage-induced apoptosis (Yu and Zhang, 2003). Chemotherapy and radiation often cause DNA damage, which induces cell cycle arrest and apoptosis in cancer cells. Deficiencies in apoptosis regulation contribute to acquired resistance to anticancer drugs (Johnstone et al., 2002). A number of studies have demonstrated that SMAC and SMAC mimetics potentiate apoptosis induced by anticancer agents, including chemotherapeutic drugs and irradiation, in human cancer cells (Fulda et al., 2002; Arnt and Kaufmann, 2003; Giagkousiklidis et al., 2005; Zhao et al., 2006). Efforts have also been made to develop improved SMAC mimetic agents (Li et al., 2004; Sun et al., 2004). However, the precise function of SMAC in promoting apoptosis induced by anticancer agents was unclear. Our previous studies demonstrated that in response to DNA damage, PUMA dissociates Bax from the antiapoptotic protein Bcl-X L, leading to Bax mitochondrial translocation and multimerization, which in turn triggers mitochondrial membrane permeabilization and subsequent release of cytochrome c, SMAC and AIF (Yu et al., 2003; Ming et al., 2006). The current study suggests that SMAC plays an important role in executing DNA damage-induced and PUMA-mediated apoptosis by participating in a feedback amplification loop and other mitochondrial events, which implies that SMAC enhances the antitumor activities of chemotherapeutic drugs and irradiation by promoting the proapoptotic function of PUMA. Although SMAC release is a general response to different apoptotic stimuli (Deng et al., 2002; Arnt and Kaufmann, 2003; Rashmi et al., 2005), the requirement of SMAC in apoptosis appears to be somewhat specific. Among a number of stimuli analysed, only a few, including Ad-PUMA, NSAIDs and certain DNA-damaging agents, were found to induce SMAC-dependent apoptosis and caspase activation (Kohli et al., 2004). Mitochondrial apoptogenic proteins were initially identified as important regulators of apoptosis by biochemical studies. Recently, studies carried out using genetically engineered animals indicate that these proteins are required for apoptosis under certain conditions. For example, knock-in of a cytochrome c mutant that fails to activate Apaf-1 in mice abrogates caspase activation and apoptosis in fibroblasts (Hao et al., 2005). Knockout of AIF in mice inhibited neuronal cell death (Yu et al., 2002). However, targeted deletion of SMAC in mice was not found to influence murine development and apoptosis in several tissues, raising the question whether SMAC is required for apoptosis in mammalian cells (Okada et al., 2002). Our data suggest that the role of SMAC in apoptosis varies depending on cell types and apoptotic stimuli. In colon cancer cells, SMAC plays an important role in executing apoptosis induced by PUMA and NSAIDs (Kohli et al., 2004). The partial phenotypes observed in the SMAC- KO cells in both cases, as well as the observation that SMAC deficiency does not rescue cells long-term viability in clonogenic assays (data not shown), suggest that the function of SMAC in apoptosis is limited to the execution, rather than the commitment stage of apoptosis. As a result, apoptotic cell death cannot be fully executed without SMAC, but cells may eventually die through other mechanisms such as necrosis and autophagy (Okada and Mak, 2004). However, the function of SMAC in the execution of apoptosis warrants further examination using different experimental systems and additional apoptotic stimuli. Although the release of mitochondrial apoptogenic proteins is known to be critical for apoptosis execution and caspase activation, the mechanism of this process is poorly understood. Our data obtained from two different models, that is, PUMA- and NSAID-induced apoptosis, suggest that the release of SMAC and that of cytochrome c or AIF may play a different role in apoptosis. These events also seem to be related to each other. It appears that SMAC itself is involved in promoting mitochondrial membrane potential change, thereby mediating cytochrome c release and caspase activation. These results are consistent with a recent study showing that the release of different mitochondrial

4195 Figure 5 Deficiencies in caspase activation and mitochondrial apoptotic pathway in SMAC-deficient cells. (a) Caspase cleavage following Ad-PUMA infection. Lysates from HCT116 and SMAC-KO cells collected at the indicated time points following Ad-PUMA infection were analysed for caspase-3 and caspase-9 by Western blotting. Arrows indicate the cleavage fragments of caspases. (b) Caspase cleavage in cells treated by etoposide or ALLN. HCT116 and SMAC-KO cells were treated with the DNA-damaging agent etoposide (200 mm) or the proteasome inhibitor ALLN (50 mm) for 48 h. Activation of caspase-3 and caspase-9 was analysed by Western blotting. Arrows indicate the cleavage fragments of caspases. (c) Cytochrome c release induced by Ad-PUMA. Mitochondrial and cytosolic fractions isolated from HCT116, BAX-KO and SMAC-KO cells after Ad-PUMA infection for 48 h were subject to Western analysis. (d) Time course of cytochrome c and AIF release. Cytosolic cellular fractions isolated from HCT116 and SMAC-KO cells at the indicated time points after Ad-PUMA infection were probed for cytochrome c and AIF by Western blotting. (e) Mitochondrial membrane potential collapse. Indicated cell lines were harvested 24 h after Ad-PUMA or Ad-DBH3 infection, stained by MitoTracker Red (CMXROS) and analysed by flow cytometry. Cox IV and a-tubulin were used as controls for fractionation and loading. apoptogenic proteins is coordinated and occurs through different time courses (Munoz-Pinedo et al., 2006). It has been shown that caspase activation can occur before mitochondrial permeabilization (Lassus et al., 2002). A feedback amplification loop linking caspase activation to mitochondrial dysfunction in genotoxic stress-induced apoptosis has been proposed (Chen et al., 2000). Recent studies also suggest that caspase-3 and caspase-7 are involved in a feedback amplification loop in triggering cytochrome c release and other mitochondrial events in apoptosis (Lakhani et al., 2006). It is possible that SMAC participates in this feedback

4196 Cruz Biotechnology, Santa Cruz, CA, USA) and SMAC (a generous gift from Dr Eileen White at Rutgers University, Piscataway, NJ, USA). Western blotting analysis was performed as previously described (Kohli et al., 2004). Figure 6 SMAC and SMAC mimetics enhance PUMA-induced apoptosis. (a) SMAC transfection enhances PUMA-mediated apoptosis. HCT116 cells were transfected with a SMAC expression vector (SMAC) and a control empty vector (empty vector). Twenty-four hours after transfection, cells were infected with Ad- PUMA. Apoptosis was analysed by nuclear staining 48 h after treatment. (b) A SMAC mimetic compound sensitizes colon cancer cells to PUMA-induced apoptosis. HCT116 cells were treated with 100 nm of the SMAC mimetic compound (C3) or the control compound (C4), alone or in combination with Ad-PUMA infection. Apoptosis was determined by nuclear staining 48 h after treatment. The results are the average of three independent experiments with one standard deviation shown. amplification loop by providing the initial caspase activity to unleash further mitochondrial changes that are necessary for full execution of apoptotic events. However, other mechanisms have also been proposed to explain how SMAC and cytochrome c can affect each other during apoptosis initiation. A recent study showed that SMAC release is suppressed in cytochrome c- deficient cells (Hansen et al., 2006). It has been shown that SMAC and cytochrome c are released together from the mitochondria during UV-induced apoptosis (Zhou et al., 2005). Therefore, questions remain as to how SMAC release affects other mitochondria-related events during apoptosis, and whether its function in regulating mitochondrial events is related to IAP inhibition. Materials and methods Western blotting The antibodies used for Western blotting included those against caspase-9 (Cell Signaling Technology, Danvers, MA, USA), cytochrome c, a-tubulin (BD Biosciences, San Jose, CA, USA), caspase-3 (Assay Designs & Stressgen Bioreagents, Ann Arbor, MI, USA), cytochrome oxidase subunit IV (Cox IV) (Invitrogen, Carlsbad, CA, USA), Bax (N-20), AIF (Santa Cell culture and transfection The cell lines used in this study, including HCT116 colorectal cancer cell derivatives SMAC-knockout (SMAC-KO), SMAC- KO reconstituted with SMAC (SMAC-R), BAX-knockout (BAX-KO), p21-knockout (p21-ko), p21/puma double knockout (p21-ko/puma-ko) and SMAC-knockdown DLD1 cells (SMAC-KD) were previously described (Waldman et al., 1995; Zhang et al., 2000; Yu et al., 2003; Kohli et al., 2004). All cell lines were cultured in McCoy s 5A media (Invitrogen) supplemented with 10% defined fetal bovine serum (Hyclone, Logan, UT, USA), 100 U/ml penicillin, 100 mg/ml streptomycin (Invitrogen), and were maintained at 371Cin5%CO 2. Cells were transfected by Lipofectamine 2000 (Invitrogen) according to the manufacturer s instructions. To induce apoptosis, cells were plated at B20% density and allowed to attach for 12 h. Ad-PUMA and mutant PUMA with a BH3 domain deletion (Ad-DBH3) were used to infect cells (Yu et al., 2003). Adriamycin (0.2 mg/ml) was used to treat p21-ko and p21-ko/puma-ko cells. Etoposide and ALLN were used at 200 and 50 mm, respectively. The SMAC mimetic and control compounds were provided by Dr Xiaodong Wang and Dr Patrick G Harran at the University of Texas Southwestern Medical Center (Li et al., 2004). Apoptosis assays After treatment, floating and adherent cells were collected and analysed for apoptosis by nuclear staining. Briefly, cells were fixed in 70% ethanol and stained with phosphate-buffered saline solution containing 0.1% Triton X-100, 50 mg/ml RNaseA and 5 mg/ml propidium iodide (PI). Apoptotic cells, which contain condensed chromatin and fragmented nuclei, were counted after microscopic visualization. For each treatment, cells were counted three times and a minimum of 300 cells were analysed. The PI-stained cells were also subjected to flow cytometry to determine the fraction of sub- G 1 population. For analysis of apoptosis by Annexin V staining, cells were stained by Annexin V-Alexa 594 (Molecular Probes) following the manufacturer s instructions, counterstained by 4 0,6-diamidino-2-phenylindole, dihydrochloride (DAPI) and then analysed by flow cytometry. For detection of mitochondrial membrane potential change, cells were harvested and stained by Mitotracker Red CMXROS (Molecular Probes) for 15 min at room temperature and then analysed by flow cytometry on a Beckman Coulter (Fullerton, CA, USA) EPICS XL using the FL3 channel according to the manufacturer s instructions. Analysis of cytochrome c, SMAC and AIF release Mitochondrial and cytosolic fractions were isolated from treated cells by differential centrifugation. Briefly, cells were washed, resuspended in homogenization buffer (0.25 M sucrose, 10 mm HEPES, ph 7.4 and 1 mm ethylene glycol bis(b-aminoethylether)-n,n,n 0,N 0,-tetraacetic acid) and subjected to 40 strokes of homogenization in a Dounce homogenizer. The homogenates were centrifuged at 1000 g for 15 min at 41C to pellet nuclei and unbroken cells. The supernatant was subsequently centrifuged at 10 000 g for 15 min at 41C to obtain cytosolic fraction (supernatant) and mitochondrial fraction (pellet). The mitochondrial fraction was resuspended in homogenization buffer following one

wash. Both fractions were mixed with equal volumes of 2 Laemmli sample buffer for Western blotting analysis. Knockdown of AIF by RNA interference Nucleotides 481 499 (GATCCTGAGCTGCCGTACA) within the AIF coding sequence was selected as the target for RNAi. Two 64-base oligonucleotides containing this sequence were designed based on previously described criteria (Elbashir et al., 2002). The oligonucleotides were annealed and the duplex was cloned into the psuper vector (Oligoengine, Seattle, WA, USA), which can direct the synthesis of sirna (Brummelkamp et al., 2002). Forty-eight hours following transfection with the RNAi constructs, HCT116 cells were plated in 96-well plates in the presence of Puromycin (2 mg/ml, Invitrogen). Puromycin-resistant clones were isolated and expanded. Clones with significant downregulation of AIF were identified by Western blotting. References Arnt CR, Kaufmann SH. (2003). The saintly side of Smac/ DIABLO: giving anticancer drug-induced apoptosis a boost. Cell Death Differ 10: 1118 1120. Brummelkamp TR, Bernards R, Agami R. (2002). A system for stable expression of short interfering RNAs in mammalian cells. Science 296: 550 553. Chen Q, Gong B, Almasan A. (2000). Distinct stages of cytochrome c release from mitochondria: evidence for a feedback amplification loop linking caspase activation to mitochondrial dysfunction in genotoxic stress induced apoptosis. Cell Death Differ 7: 227 233. Cory S, Huang DC, Adams JM. (2003). The Bcl-2 family: roles in cell survival and oncogenesis. 22: 8590 8607. Danial NN, Korsmeyer SJ. (2004). Cell death. Critical control points. Cell 116: 205 219. Deng Y, Lin Y, Wu X. (2002). TRAIL-induced apoptosis requires Bax-dependent mitochondrial release of Smac/ DIABLO. Genes Dev 16: 33 45. Du C, Fang M, Li Y, Li L, Wang X. (2000). Smac, a mitochondrial protein that promotes cytochrome c-dependent caspase activation by eliminating IAP inhibition. Cell 102: 33 42. Elbashir SM, Harborth J, Weber K, Tuschl T. (2002). Analysis of gene function in somatic mammalian cells using small interfering RNAs. Methods 26: 199 213. Fulda S, Wick W, Weller M, Debatin KM. (2002). Smac agonists sensitize for Apo2L/TRAIL- or anticancer druginduced apoptosis and induce regression of malignant glioma in vivo. Nat Med 8: 808 815. Giagkousiklidis S, Vogler M, Westhoff MA, Kasperczyk H, Debatin KM, Fulda S. (2005). Sensitization for gammairradiation-induced apoptosis by second mitochondriaderived activator of caspase. Cancer Res 65: 10502 10513. Green DR, Reed JC. (1998). Mitochondria and apoptosis. Science 281: 1309 1312. Han J, Flemington C, Houghton AB, Gu Z, Zambetti GP, Lutz RJ et al. (2001). Expression of bbc3, a pro-apoptotic BH3-only gene, is regulated by diverse cell death and survival signals. Proc Natl Acad Sci USA 98: 11318 11323. Hansen TM, Smith DJ, Nagley P. (2006). Smac/DIABLO is not released from mitochondria during apoptotic signalling in cells deficient in cytochrome c. Cell Death Differ 13: 1181 1190. Hao Z, Duncan GS, Chang CC, Elia A, Fang M, Wakeham A et al. (2005). Specific ablation of the apoptotic functions of Acknowledgements We thank Dr Xiaodong Wang and Dr Patrick G Harran at the University of Texas Southwestern Medical Center for providing the SMAC mimetic and control compounds, Dr Eileen White for providing the SMAC antibody, Dr Chunying Du at the Stowers Institute for Medical Research for providing the SMAC expression plasmid, Ms Hongtao Liu for technical assistance and members of our laboratories for helpful discussions. This work is supported by NIH Grant CA106348, grants from the Edward Mallinckrodt Jr Foundation and Elsa U Pardee foundation (LZ) and grants from the Flight Attendant Medical Research Institute (FAMRI), the Alliance for Cancer Gene Therapy (ACGT) and the Hillman Foundation (to JY). LZ is a scholar of the General Motors (GM) Cancer Research Foundation and the V Foundation for Cancer Research. cytochrome c reveals a differential requirement for cytochrome c and Apaf-1 in apoptosis. Cell 121: 579 591. Hemann MT, Zilfou JT, Zhao Z, Burgess DJ, Hannon GJ, Lowe SW. (2004). Suppression of tumorigenesis by the p53 target PUMA. Proc Natl Acad Sci USA 101: 9333 9338. Jeffers JR, Parganas E, Lee Y, Yang C, Wang J, Brennan J et al. (2003). Puma is an essential mediator of p53-dependent and -independent apoptotic pathways. Cancer Cell 4: 321 328. Johnstone RW, Ruefli AA, Lowe SW. (2002). Apoptosis: a link between cancer genetics and chemotherapy. Cell 108: 153 164. Kohli M, Yu J, Seaman C, Bardelli A, Kinzler KW, Vogelstein B et al. (2004). SMAC/Diablo-dependent apoptosis induced by nonsteroidal antiinflammatory drugs (NSAIDs) in colon cancer cells. Proc Natl Acad Sci USA 101: 16897 16902. Lakhani SA, Masud A, Kuida K, Porter Jr GA, Booth CJ, Mehal WZ et al. (2006). Caspases 3 and 7: key mediators of mitochondrial events of apoptosis. Science 311: 847 851. Lassus P, Opitz-Araya X, Lazebnik Y. (2002). Requirement for caspase-2 in stress-induced apoptosis before mitochondrial permeabilization. Science 297: 1352 1354. Li L, Thomas RM, Suzuki H, De Brabander JK, Wang X, Harran PG. (2004). A small molecule Smac mimic potentiates TRAIL- and TNFalpha-mediated cell death. Science 305: 1471 1474. Ming L, Wang P, Bank A, Yu J, Zhang L. (2006). PUMA dissociates Bax and BCL-XL to induce apoptosis in colon cancer cells. J Biol Chem 281: 16034 16042. Munoz-Pinedo C, Guio-Carrion A, Goldstein JC, Fitzgerald P, Newmeyer DD, Green DR. (2006). Different mitochondrial intermembrane space proteins are released during apoptosis in a manner that is coordinately initiated but can vary in duration. Proc Natl Acad Sci USA 103: 11573 11578. Nakano K, Vousden KH. (2001). PUMA, a novel proapoptotic gene, is induced by p53. Mol Cell 7: 683 694. Okada H, Mak TW. (2004). Pathways of apoptotic and nonapoptotic death in tumour cells. Nat Rev Cancer 4: 592 603. Okada H, Suh WK, Jin J, Woo M, Du C, Elia A et al. (2002). Generation and characterization of Smac/DIABLOdeficient mice. Mol Cell Biol 22: 3509 3517. Perfettini JL, Roumier T, Castedo M, Larochette N, Boya P, Raynal B et al. (2004). NF-kappaB and p53 are the dominant apoptosis-inducing transcription factors elicited by the HIV-1 envelope. J Exp Med 199: 629 640. 4197

4198 Rashmi R, Kumar S, Karunagaran D. (2005). Human colon cancer cells lacking Bax resist curcumin-induced apoptosis and Bax requirement is dispensable with ectopic expression of Smac or downregulation of Bcl-XL. Carcinogenesis 26: 713 723. Reimertz C, Kogel D, Rami A, Chittenden T, Prehn JH. (2003). Gene expression during ER stress-induced apoptosis in neurons: induction of the BH3-only protein Bbc3/PUMA and activation of the mitochondrial apoptosis pathway. J Cell Biol 162: 587 597. Sun H, Nikolovska-Coleska Z, Yang CY, Xu L, Liu M, Tomita Y et al. (2004). Structure-based design of potent conformationally constrained Smac mimetics. J Am Chem Soc 126: 16686 16687. Susin SA, Lorenzo HK, Zamzami N, Marzo I, Snow BE, Brothers GM et al. (1999). Molecular characterization of mitochondrial apoptosis-inducing factor. Nature 397: 441 446. Verhagen AM, Ekert PG, Pakusch M, Silke J, Connolly LM, Reid GE et al. (2000). Identification of DIABLO, a mammalian protein that promotes apoptosis by binding to and antagonizing IAP proteins. Cell 102: 43 53. Villunger A, Michalak EM, Coultas L, Mullauer F, Bock G, Ausserlechner MJ et al. (2003). p53- and drug-induced apoptotic responses mediated by BH3-only proteins Puma and Noxa. Science 302: 1036 1038. Waldman T, Kinzler KW, Vogelstein B. (1995). p21 is necessary for the p53-mediated G(1) arrest in human cancer cells. Cancer Res 55: 5187 5190. Wang X. (2001). The expanding role of mitochondria in apoptosis. Genes Dev 15: 2922 2933. Yu J, Wang Z, Kinzler KW, Vogelstein B, Zhang L. (2003). PUMA mediates the apoptotic response to p53 in colorectal cancer cells. Proc Natl Acad Sci USA 100: 1931 1936. Yu J, Zhang L. (2003). No PUMA, no death: implications for p53-dependent apoptosis. Cancer Cell 4: 248 249. Yu J, Zhang L, Hwang PM, Kinzler KW, Vogelstein B. (2001). PUMA induces the rapid apoptosis of colorectal cancer cells. Mol Cell 7: 673 682. Yu SW, Wang H, Poitras MF, Coombs C, Bowers WJ, Federoff HJ et al. (2002). Mediation of poly(adp-ribose) polymerase-1-dependent cell death by apoptosis-inducing factor. Science 297: 259 263. Zhang L, Yu J, Park BH, Kinzler KW, Vogelstein B. (2000). Role of BAX in the apoptotic response to anticancer agents. Science 290: 989 992. Zhao J, Jin J, Zhang X, Shi M, Dai J, Wu M et al. (2006). Transfection of Smac sensitizes tumor cells to etoposide-induced apoptosis and eradicates established human hepatoma in vivo. Cancer Gene Ther 13: 420 427. Zhou LL, Zhou LY, Luo KQ, Chang DC. (2005). Smac/ DIABLO and cytochrome c are released from mitochondria through a similar mechanism during UV-induced apoptosis. Apoptosis 10: 289 299.