Resistance to apoptosis is a hallmark of pancreatic cancer, a leading cause of cancer deaths. Therefore, novel strategies are required to target apoptosis resistance. Here, we report that the combination of X-linked inhibitor of apoptosis (XIAP) inhibition and tumor necrosis factor–related apoptosis-inducing ligand (TRAIL) is an effective approach to trigger apoptosis despite Bcl-2 overexpression and to suppress pancreatic cancer growth in vitro and in vivo. Knockdown of XIAP by RNA interference cooperates with TRAIL to induce caspase activation, loss of mitochondrial membrane potential, cytochrome c release, and apoptosis in pancreatic carcinoma cells. Loss of mitochondrial membrane potential and cytochrome c release are extensively inhibited by a broad range or caspase-3 selective caspase inhibitor and by RNAi-mediated silencing of caspase-3, indicating that XIAP inhibition enhances TRAIL-induced mitochondrial damage in a caspase-3–dependent manner. XIAP inhibition combined with TRAIL even breaks Bcl-2–imposed resistance by converting type II cells that depend on the mitochondrial contribution to the death receptor pathway to type I cells in which TRAIL-induced activation of caspase-3 and caspase-9 and apoptosis proceeds irrespective of high Bcl-2 levels. Most importantly, XIAP inhibition potentiates TRAIL-induced antitumor activity in two preclinical models of pancreatic cancer in vivo. In the chicken chorioallantoic membrane model, XIAP inhibition significantly enhances TRAIL-mediated apoptosis and suppression of tumor growth. In a tumor regression model in xenograft-bearing mice, XIAP inhibition acts in concert with TRAIL to cause even regression of established pancreatic carcinoma. Thus, this combination of XIAP inhibition plus TRAIL is a promising strategy to overcome apoptosis resistance of pancreatic cancer that warrants further investigation. [Cancer Res 2008;68(19):7956–65]
Despite intensive protocols, treatment resistance of pancreatic cancer considerably contributes to the poor prognosis of this disease (1, 2). Evasion of apoptosis (programmed cell death) can lead to treatment failure in pancreatic cancer because intact apoptosis pathways are crucial for therapy-induced cytotoxicity (3, 4). Thus, new strategies that target apoptosis resistance are required to improve the dismal prognosis of pancreatic cancer patients.
Two principal apoptosis pathways eventually lead to activation of caspases, i.e., the death receptor (extrinsic) pathway and the mitochondrial (intrinsic) pathway (4, 5). Ligation of death receptors, e.g., agonistic tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) receptors by TRAIL, results in caspase-8 activation that cleaves effector caspase-3 either directly or indirectly via the mitochondrial route (6). In the mitochondrial pathway, effector caspases are activated upon the release of apoptogenic factors such as cytochrome c or second mitochondria-derived activator of caspase (Smac)/direct inhibitor of apoptosis protein (IAP) binding protein with low PI (DIABLO) from mitochondria into the cytosol (7). Cytochrome c triggers caspase-3 activation via formation of the cytochrome c/caspase-9/Apaf-1 apoptosome complex, whereas Smac/DIABLO promotes apoptosis by neutralizing IAPs (7). The mitochondrial contribution to death receptor signaling is of special relevance in type II cells, for example pancreatic carcinoma, which rely on the apoptotic function of mitochondria for full activation of caspase-3 and apoptosis (8–10). In type I cells, caspase-8 is activated upon death receptor ligation at the death-inducing signaling complex (DISC) in quantities sufficient to directly activate caspase-3 (8). Accordingly, blocking the mitochondrial pathway, e.g., by overexpression of Bcl-2, inhibits death receptor-induced apoptosis in type II but not in type I cells (8).
Among the death receptor ligands, TRAIL is considered as the most promising candidate for medical application because it triggers apoptosis preferentially in cancer cells with no or minimal toxicity to nonmalignant human cells (11). However, many human cancers including pancreatic carcinoma display primary or acquired resistance to TRAIL (9, 12). X-linked inhibitor of apoptosis (XIAP) may contribute to resistance to TRAIL because XIAP blocks apoptosis at a key node of the apoptotic machinery through inhibition of caspase-3 and caspase-9 (13, 14). Searching for strategies to restore sensitivity to TRAIL-induced apoptosis in pancreatic cancer, we investigated XIAP as a molecular target in vitro and in vivo.
Materials and Methods
Cell culture and reagents. Pancreatic carcinoma cell lines PancTu1 and PaTuII were established from distinct human poorly differentiated pancreatic adenocarcinoma specimens (15, 16) and cultured in RPMI1640 (Life Technologies, Inc.) as described (12). TRAIL was purchased from R&D Systems, Inc., and all chemicals from Sigma unless indicated otherwise.
Transduction and transfection. To knockdown XIAP, shRNA targeting XIAP sequence 241 (gtggtagtcctgtttcagc; ref. 17) corresponding to mRNA sequence were cloned into pRETRO-SUPER as previously described (12). A sequence with no corresponding part in the human genome (gatcatgtagatacgctca) was used as control. Stable bulk cultures were generated by selection with 2 μg/mL puromycin and clones by limited dilutions. For RNAi-mediated knockdown of caspase-3, cells were transiently transfected twice with 0.06 nmol caspase-3 Stealth or control RNAi (Invitrogen) using TransMessenger (Qiagen). For Bcl-2 overexpression, cells were transduced with pMSCV vector containing mouse Bcl-2 or empty vector using the packaging cell line PT67 (BD Biosciences). Stable cell lines were selected by 10 μg/mL blasticidin (Invitrogen).
Western blot analysis. Western blot analysis was performed as described previously (18) using the following antibodies: mouse anti–caspase-8 (ApoTech Corporation), rabbit anti–caspase-3 (Cell Signaling), rabbit anti–caspase-9, mouse anti–caspase-2, mouse anti–caspase-7, mouse anti-XIAP from BD Biosciences, mouse anti-mouse Bcl-2 (Zymed), or mouse anti–β-actin (Sigma) followed by goat-anti-mouse IgG or goat-anti-rabbit IgG conjugated to horseradish peroxidase (Santa Cruz Biotechnology). Enhanced chemiluminescence was used for detection (Amersham Bioscience).
Bax immunoprecipitation. Cells were lysed for 25 min in CHAPS lysis buffer [10 mmol/L HEPES (pH 7.4), 150 mmol/L NaCl, 1% CHAPS]. Five hundred micrograms of protein were incubated with 4 μg mouse anti-Bax antibody (6A7; Sigma) overnight at 4°C on rotator followed by addition of 10 μL Dynabeads Pan Mouse IgG (Dako) and incubated for 2 h at 4°C on rotator. Beads were washed thrice with CHAPS lysis buffer. Immune complexes were analyzed by Western blotting using rabbit anti-BaxNT antibody (Upstate Biotechnology).
Determination of apoptosis and cell viability. Apoptosis was determined by Annexin-V staining or by fluorescence-activated cell sorting (FACS) analysis of DNA fragmentation of propidium iodide–stained nuclei as described (12). Cell viability was assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay according to the manufacturer's instructions (Roche Diagnostics).
Determination of mitochondrial membrane potential and cytochrome c release. CMXRos (1 μmol/L; Molecular Probes) was used to measure the mitochondrial transmembrane potential. Cells were incubated for 30 min at 37°C in the presence of the fluorochrome and immediately analyzed by flow cytometry. Cytochrome c release was determined in permeabilized cells using mouse anti–cytochrome c monoclonal antibody (BD Biosciences) as described (19).
Determination of caspase activity. Caspase activity was determined in living nonfixed, nonlysed cells as described (20) using caspase substrates conjugated to rhodamine R110: zDEVD-R110 for caspase-3/7, zVDVAD-R110 for caspase-2, zIETD-R110 for caspase-8, and zLEHD-R110 for caspase-9, all from Molecular Probes. Caspase inhibitors (zVAD.fmk, zDEVD.fmk) were purchased from Bachem.
Chorioallantoic membrane assay. Chorioallantoic membrane (CAM) assay was done as described previously (21). Briefly, 1 × 106 tumor cells were resuspended in 25 μL serum-free medium and 25 μL Matrigel Matrix (BD Biosciences) and implanted on fertilized chicken eggs on day 8 of incubation. Tumors were topically treated with 0.7 ng TRAIL in 15 μL PBS daily for 3 d, sampled with the surrounding CAM 4 d after seeding, fixed in 4% paraformaldehyde, paraffin embedded, cut in 5-μm sections, and analyzed by immunohistochemistry.
Xenograft mouse model. PancTu1 cells (1 × 106) transduced with control or XIAP shRNA were s.c. inoculated at left and right dorsal sides, respectively, of 6-wk-old female NMRI/nu-nu mice (Janvier Laboratories). Treatment started 1 wk after tumor cell inoculation by daily i.p. injection of isoleucine zipper–TRAIL (22) for 10 consecutive d, TRAIL maintenance buffer served as control. Tumor size was assessed twice weekly by caliper; the tumor volume was calculated according to the formula 0.5 × L × W × T (L, length; W, width; T, thickness). Relative tumor growth ratio (treated versus control, T/C) was calculated by tumor volume in treated group divided by tumor volume in control group. At the end of experiments, tumors were recovered for histologic and pathologic analysis. Additional 3 mice were sacrificed 24 h after the second injection of TRAIL on day 10 of the experiment, and tumor tissue was analyzed by immunohistochemistry. Animal experiments were performed in accordance with relevant institutional and national regulations; research protocols were approved by relevant authorities.
Immunohistochemistry. Immunohistochemical staining was performed with 1:1 hematoxyline and 0.5% eosin, anti–Ki-67 (Dako) or anticleaved caspase-3 antibody (Biozol) as previously described (23). Apoptosis was assessed by terminal deoxynucleotidyltransferase–mediated dUTP nick end labeling (TUNEL; Roche Diagnostics) according to the manufacturer's instructions. Negative controls were performed by omitting the first step with primary antibody and yielded negative results. The percentage of positive cells was independently determined by two examiners.
Statistical analysis. Statistical significance was assessed by Student's t test or Mann-Whitney U test, where appropriate, using Winstat (R. Fitch Software) or SPSS (SPSS GmbH Software) software.
XIAP knockdown sensitizes pancreatic carcinoma cells for TRAIL-induced apoptosis. To antagonize XIAP in pancreatic cancer, we generated clones of PancTu1 and PaTuII pancreatic carcinoma cells in which XIAP was substantially down-regulated by retroviral short-hairpin RNA (shRNA) vectors (Fig. 1A). Knockdown of XIAP significantly enhanced TRAIL-induced apoptosis in a concentration- and time-dependent manner (Fig. 1B and C; XIAP shRNA clone 10 for PancTu1 cells, XIAP shRNA clone 4 for PaTuII cells). Similar results were obtained in distinct clones with XIAP knockdown (Supplementary Fig. S1A) and also in another pancreatic carcinoma cell line, i.e., Colo357 (12). For subsequent experiments, XIAP shRNA clone 10 were used for PancTu1 cells and XIAP shRNA clone 4 for PaTuII cells. Apoptotic cell death was confirmed by analysis of DNA fragmentation and by typical morphologic features, e.g., nuclear condensation and fragmentation (data not shown). Furthermore, XIAP knockdown cooperated with TRAIL to suppress clonogenic survival (Supplementary Fig. S1B). By comparison, XIAP knockdown did not significantly alter chemotherapeutic drug–induced cytotoxicity in PancTu1 and PaTuII cells (Supplementary Fig. S1C). We also extended our study to another IAP protein, i.e., survivin (24). Knockdown of survivin significantly increased apoptosis without the addition of TRAIL and also acted in concert with TRAIL to induce apoptosis (Supplementary Fig. S2). Together, these findings show that inhibition of XIAP or survivin sensitizes pancreatic carcinoma cells to TRAIL-induced apoptosis.
XIAP knockdown enhances TRAIL-induced caspase activation and mitochondrial perturbations. Next, we explored molecular mechanisms responsible for the cooperative action of XIAP inhibition and TRAIL in pancreatic carcinoma cells. Analysis of caspase activity by enzymatic caspase assay revealed that XIAP knockdown significantly increased TRAIL-induced activation of caspase-3, caspase-9, caspase-8, and caspase-2 (Fig. 2A). Activation of caspases was also confirmed by detection of active caspase cleavage fragments by Western blotting (data not shown). Addition of the broad range caspase inhibitor zVAD.fmk extensively inhibited TRAIL-induced apoptosis in cells with XIAP knockdown (Supplementary Fig. S3), demonstrating that apoptosis occurs in a caspase-dependent manner.
Furthermore, we investigated whether XIAP inhibition has an effect on the mitochondrial pathway during TRAIL-induced apoptosis. XIAP knockdown significantly enhanced TRAIL-induced loss of mitochondrial membrane potential and cytochrome c release from mitochondria in a time-dependent manner (Fig. 2B and C). We also assessed Bid cleavage and Bax conformational change as possible links between caspase activation and mitochondrial injury. XIAP knockdown enhanced TRAIL-induced Bid cleavage as well as Bax activation as indicated by higher levels of conformationally active Bax in TRAIL-treated cells with XIAP knockdown compared with control cells, whereas total levels of Bax protein were not up-regulated in cells with XIAP knockdown (Fig. 2D). Together, this shows that XIAP inhibition cooperates with TRAIL to trigger caspase activation and mitochondrial perturbations in pancreatic carcinoma cells.
XIAP knockdown promotes TRAIL-induced mitochondrial perturbations in a caspase-dependent manner. Because we observed an increase in caspase activity before mitochondrial perturbations in TRAIL-treated cells with XIAP knockdown (Fig. 2A–C), we asked whether caspases are involved in causing mitochondrial damage. To address this issue, we examined the effect of the broad range caspase inhibitor zVAD.fmk and the relatively specific caspase-3 inhibitor zDEVD.fmk on mitochondrial apoptogenic functions upon treatment with TRAIL. zVAD.fmk as well as zDEVD.fmk extensively blocked TRAIL-induced loss of mitochondrial membrane potential and cytochrome c release (Fig. 3A and B). In a second approach, we specifically knocked down caspase-3 by RNA interference because chemical inhibitors of caspases lack absolute substrate specificity (25). Control experiments showed that caspase-3 siRNA caused down-regulation of caspase-3 with no detectable effects on expression levels of caspase-7, caspase-9, caspase-8, and caspase-2 (Fig. 3C). Interestingly, silencing of caspase-3 significantly reduced TRAIL-induced loss of mitochondrial membrane potential and cytochrome c release (Fig. 3D). This indicates that XIAP inhibition promotes TRAIL-induced mitochondrial damage in a caspase-3–dependent manner.
XIAP knockdown overcomes Bcl-2–mediated resistance to TRAIL. Because pancreatic carcinoma cells have been reported to classify as type II cells that express antiapoptotic Bcl-2 family proteins at high levels and depend on the mitochondrial contribution to death receptor–induced apoptosis (9), we next asked whether blockade of the mitochondrial pathway impairs apoptosis induced by the combination of XIAP knockdown and TRAIL. To address this issue, we stably overexpressed Bcl-2 in cells with XIAP knockdown (Fig. 4A). Interestingly, XIAP knockdown enhanced TRAIL-induced cytotoxicity irrespective of Bcl-2 overexpression, whereas Bcl-2 overexpression conferred resistance to TRAIL in control cells with high XIAP expression (Fig. 4B). Also, XIAP knockdown increased TRAIL-induced cleavage of caspase-3 into the active p17/p12 cleavage fragments and cleavage of caspase-9 regardless of high Bcl-2 levels (Fig. 4C).
To elucidate whether XIAP knockdown bypasses the requirement of the mitochondrial pathway for TRAIL-induced apoptosis in Bcl-2–overexpressing cells or whether it alternatively breaks mitochondrial resistance, we examined variables of mitochondrial apoptosis that are controlled by Bcl-2 (26). Bcl-2 overexpression failed to inhibit loss of mitochondrial membrane potential and cytochrome c release in TRAIL-treated cells with XIAP knockdown, whereas it significantly reduced these mitochondrial perturbations in control cells with high XIAP expression (Fig. 4D). This set of experiments shows that XIAP knockdown converts type II pancreatic carcinoma cells, in which Bcl-2 confers protection against TRAIL, to type I cells in which TRAIL triggers activation of caspase-3 and caspase-9 and apoptosis regardless of high Bcl-2 levels.
XIAP knockdown enhances TRAIL-induced antitumor activity against pancreatic carcinoma in vivo. We then evaluated the antitumor activity of concomitant XIAP inhibition and TRAIL in vivo using the CAM model, an established in vivo tumor model (21, 27). To this end, PancTu1 cells transduced with XIAP or control shRNA were seeded on the CAM of chicken embryos, allowed to form tumors, and then treated with TRAIL. Importantly, XIAP knockdown together with TRAIL, but neither approach alone, suppressed the growth of pancreatic carcinoma in vivo (Fig. 5A). Immunohistochemical analysis of tumor tissue revealed that XIAP knockdown significantly increased TRAIL-induced caspase-3 activation and apoptosis and also acted in concert with TRAIL to suppress proliferation (Fig. 5B–D). This shows that XIAP inhibition cooperates with TRAIL to induce apoptosis and to inhibit growth of pancreatic cancer in vivo.
XIAP knockdown cooperates with TRAIL to cause regression of established pancreatic cancer in a xenograft mouse model. Finally, we assessed the antitumor activity of XIAP inhibition and TRAIL in a tumor regression model of pancreatic cancer in xenograft-bearing mice. In this model, PancTu1 cells transduced with control or XIAP shRNA were s.c. injected into the left and right dorsal sides of nude mice, respectively, and TRAIL treatment was initiated 1 week afterwards when pancreatic carcinoma had been established. Strikingly, XIAP knockdown cooperated with TRAIL to cause regression of established pancreatic carcinoma in a dose-dependent manner, an effect that was sustained also after discontinuation of TRAIL treatment (Fig. 6A and B). By comparison, TRAIL treatment alone only slowed down growth of control tumors (Fig. 6A and B). XIAP knockdown substantially enhanced the TRAIL-induced reduction in tumor volume even at the lowest dose of TRAIL, with T/C ratios of 4%, 0.3%, or 0.3% upon treatment with 20, 50, or 125 μg TRAIL in tumors with XIAP knockdown compared with T/C ratios of 63%, 44%, or 42% upon treatment with 20, 50, or 125 μg TRAIL in control tumors (Fig. 6A). Similarly, analysis of tumor weight showed that TRAIL treatment was more effective in pancreatic carcinoma with XIAP knockdown compared with control tumors (Fig. 6C). The suppressive effect of XIAP knockdown alone on tumor growth in xenografted tumors suggests that the XIAP-mediated survival effect may become relevant under cellular stress conditions in vivo, i.e., under hypoxia, as XIAP is translationally up-regulated via an internal ribosome entry site during hypoxia (28).
In parallel, we monitored apoptosis in pancreatic carcinoma tissue in situ during the course of TRAIL treatment. To this end, three additional mice of each treatment group were sacrificed 24 h after the second injection of TRAIL on day 10 of the experiment and tumor tissue was analyzed by immunohistochemistry. TRAIL-treated pancreatic carcinoma with XIAP knockdown displayed a significant increase in cells with cleaved caspase-3 and TUNEL positivity and a reduction of Ki-67–positive cells (Fig. 6D). These findings reflect at the cellular level in tumor tissue in vivo that XIAP inhibition acts in concert with TRAIL to trigger apoptosis and to suppress tumor growth. Notably, no signs of acute or delayed toxicity were observed during treatment or at the end of the experiment as assessed by clinical observation, body weight, blood cell counts, and liver enzymes (Supplementary Fig. S4). Together, this set of experiments in xenograft-bearing mice shows that XIAP inhibition strongly augments the antitumor activity of TRAIL, causing regression of established pancreatic carcinoma in vivo without detectable toxicities to normal tissues.
Evasion of apoptosis is a key factor preventing treatment responses in pancreatic cancer, a leading cause of cancer deaths (3). This highlights the need for novel strategies to overcome apoptosis resistance to improve the outcome of this lethal disease.
In response to this demand, we develop in the present study the combination of XIAP inhibition plus TRAIL as an effective approach to trigger apoptosis in pancreatic carcinoma, which overcomes Bcl-2–mediated resistance and even causes regression of established tumors in vivo. This conclusion is supported by the following pieces of evidence. First, neutralizing XIAP by RNA interference acts in concert with TRAIL to trigger apoptosis in pancreatic carcinoma cells in vitro. Second, the combination of XIAP inhibition and TRAIL breaks Bcl-2–imposed resistance by converting type II pancreatic carcinoma cells to type I cells. Third, inhibition of XIAP substantially enhances the antitumor activity of TRAIL against pancreatic cancer in two preclinical models in vivo. Importantly, the combination of XIAP inhibition and TRAIL causes regression of established pancreatic carcinoma in xenograft-bearing mice. Together, these results show that XIAP inhibition is a powerful strategy to potentiate the antitumor activity of TRAIL against pancreatic cancer in vitro and in vivo.
Conceptually, we show for the first time that down-regulation of XIAP breaks Bcl-2–mediated resistance to TRAIL, switching type II pancreatic carcinoma cells that depend on the mitochondrial contribution to the death receptor pathway to type I cells in which TRAIL-induced apoptosis proceeds irrespective of high Bcl-2 levels. Neutralizing XIAP enhances TRAIL-induced apoptosis at the postmitochondrial level by relieving the XIAP-mediated inhibition of caspase-3 and caspase-9. Thus, XIAP inhibition elicits apoptosis also in cells in which the mitochondrial pathway is blocked, e.g., by antiapoptotic Bcl-2 family proteins. It is interesting to note that there is recent evidence that caspase-9 activation can occur in certain cell types in the absence of a functional apoptosome as long as XIAP is kept in check (29). Also, our data indicate that inhibition of XIAP enhances TRAIL-induced mitochondrial breakdown by promoting caspase-3 activation. Caspase-3 may act back on mitochondria by cleaving Bid, antiapoptotic Bcl-2 family proteins, and/or components of the mitochondrial respiratory chain (26, 30–33). By comparison, γ-irradiation and DNA-damaging anticancer drugs were shown to cooperate with TRAIL in Bcl-2 overexpressing cells by bypassing the mitochondrial pathway via enhanced TRAIL DISC formation without causing mitochondrial disassembling (34, 35). Because antiapoptotic Bcl-2 family proteins have been reported to be expressed at high levels in pancreatic cancer and to contribute to apoptosis resistance (2, 9), bypassing the requirement for mitochondrial apoptogenic events in the course of TRAIL-induced apoposis, e.g., by inhibition of XIAP, may have important implications to restore sensitivity to TRAIL in pancreatic cancer. Furthermore, it will be interesting to explore whether the combination of TRAIL and XIAP inhibition may bypass c-FLIP–imposed apoptosis resistance because a novel caspase-8 activation pathway has very recently been identified in the course of TNFα plus Smac mimetic–triggered apoptosis that is less sensitive to c-FLIP inhibition (36).
Our study also underscores the relevance of TRAIL-based combination protocols for the treatment of pancreatic cancer. The antitumor activity of TRAIL receptor agonists that are currently under evaluation in early clinical trials in various human malignancies (37) may be impaired by resistance of pancreatic carcinoma to TRAIL. This calls for rational combinations to target defective apoptosis programs in pancreatic cancer. Our data provide, for the first time, evidence that the combination of XIAP inhibition plus TRAIL is a potent strategy to trigger regression of established pancreatic carcinoma in xenograft-bearing mice without detectable toxicities to normal tissues. Thus, our findings provide the rationale to further develop the combination of XIAP inhibition and TRAIL to overcome apoptosis resistance of pancreatic cancer.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).
M. Vogler and H. Walczak contributed equally to this work.
Current address for H. Walczak: Division of Medicine, Imperial College, London, United Kingdom.
Grant support: Deutsche Forschungsgemeinschaft, the Deutsche Krebshilfe, IAP6/18, and the European Community (S. Fulda and K.M. Debatin).
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.
We thank R. Agami (The Netherlands Cancer Institute, Amsterdam, Netherlands) for providing pRETRO-SUPER vector, C.A. Schmitt (Berlin, Germany) for providing mouse Bcl-2 vector, K. Hohl (Department of Biostatistics, Ulm, Germany) for help with statistical analysis, and H.J. Gross (Department of Clinical Chemistry, Ulm, Germany) for help with peripheral blood analysis.