Tumor Necrosis Factor–Related Apoptosis-Inducing Ligand–Induced Apoptosis Is Inhibited by Bcl-2 but Restored by the Small Molecule Bcl-2 Inhibitor, HA 14-1, in Human Colon Cancer Cells

Tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL; also known as Apo2L) is a transmembrane protein and cytotoxic ligand belonging to the TNF superfamily that includes TNFα and FasL. TRAIL engages the extrinsic apoptotic pathway by binding to its membrane-bound death receptors (DR4 and DR5), which transmit an apoptotic signal via their intracellular death domains (reviewed in ref. 1). The adaptor protein Fas-associated death domain (FADD) is then recruited and engages initiator caspase-8 at the receptor site. TRAIL also binds to decoy receptors DcR1 and DcR2 that sequester the ligand but are unable to initiate an apoptotic signal. TRAIL is constitutively expressed in several cells and tissues, however, few normal cell populations are sensitive to this ligand, whereas multiple malignant cell types are highly sensitive (2, 3, 4). Preclinical data in mice and nonhuman primates have shown that TRAIL can suppress tumor growth without systemic toxicity (4, 5). Recent data also indicate that TRAIL can augment the antitumor efficacy of anticancer drugs in vivo, including agents used in colorectal cancer prevention (6) and treatment (4, 7). A second pathway “intrinsic” apoptotic pathway is initiated at the mitochondria (8) by cellular stresses, including antineoplastic drugs and radiation. Engagement of this pathway results in altered mitochondrial membrane permeability and the release of pro-apoptotic factors including cytochrome c, caspase-9, and second mitochondria-derived activator of caspases (Smac)/DIABLO into the cytosol (9, 10). Cytochrome c binds to apoptosis-inducing factor-1 (Apaf1) and procaspase-9 to form the “apoptosome,” which leads to activation of caspase-9 and subsequently caspase-3, resulting in apoptosis (8). Smac/DIABLO neutralizes the inhibitor of apoptosis (IAP) proteins and thereby allows caspase activation to proceed (9, 10). IAPs are a family of apoptosis-suppressing proteins commonly overexpressed in human cancers (11). XIAP, the most potent IAP, is an endogenous inhibitor of caspases-9, -7, and -3 and is overexpressed in many cancers. Evidence indicates that signals originating from the CD95 (Apo-1/Fas) or TRAIL death receptors may be linked to mitochondria by caspase-8 cleavage of the pro-apoptotic Bcl-2 family member, Bid (refs. 12, 13, 14; Fig. 1). The resultant truncated Bid translocates to mitochondria where it can trigger cytochrome c release. In this regard, TRAIL-induced apoptosis has been shown to involve the mitochondrial pathway in certain tumor cell types (15, 16). Such cross-talk suggests that an apoptotic signal originating at the cell membrane could be modulated by regulators of mitochondrial apoptotic signaling such as Bcl-2. Interestingly, two classes of cells have been described that are distinguished based on their response to a death receptor stimulus (17). In type I cells, a death receptor ligand activates caspase-8 to a level sufficient for activation of a downstream caspase cascade. In type II cells, the level of activated caspase-8 is believed to be too low to induce an apoptotic response in the absence of a mitochondrial amplification step. Hence, type II cells are mitochondria dependent, whereas type I cells are independent (17).

Mitochondria-mediated apoptosis is regulated by the Bcl-2 family of proteins, which can promote (i.e., Bax and Bid) or inhibit (i.e., Bcl-2 and Bcl-X_L) apoptosis (18). Bcl-2 preserves the integrity of the outer mitochondrial membrane and thereby prevents the release of pro-apoptotic factors from mitochondria including cytochrome c (19, 20). Although the effects of Bcl-2 on cytochrome c release are well described, other functions of Bcl-2 in the upstream and downstream regulation of apoptosis remain poorly understood. Overexpression of Bcl-2 and Bcl-X_L seem to inhibit apoptosis induced by both FasL and TNF (17, 21). However, the role of Bcl-2 in inhibiting TRAIL-mediated apoptosis is less clear because several prior reports, mostly in lymphoma cell lines, have failed to demonstrate a role for either Bcl-2 or Bcl-X_L proteins in protecting cells from TRAIL-induced apoptosis (22, 23, 24, 25, 26). These data suggested that TRAIL-induced apoptosis was independent of mitochondria and that this drug could be used in Bcl-2–overexpressing tumors. However, recent reports indicate that Bcl-2 may confer protection against TRAIL-induced apoptosis at least in certain cell types, suggesting its importance in TRAIL resistance (16, 27, 28, 29).

Deregulated apoptosis is a major mechanism of drug resistance and may contribute to tumor progression (8). Molecular defects in apoptotic signaling, such as the frequent up-regulation of Bcl-2 in human tumors, will require novel strategies for their circumvention. A strategy to inhibit Bcl-2 function involves the use of small molecules that bind in the hydrophobic groove of Bcl-2 and potentially displace BH3-only proteins. Using a computer screening strategy based on the predicted structure of the Bcl-2 protein, Wang et al. (30) identified a small molecule, ethyl 2-amino-6-bromo-4-(1-cyano-2-ethoxy-2-oxoethyl)-4H-chromene-3-carboxylate (HA14-1), that interacts at the BH3-binding site of Bcl-2 shown by in vitro binding studies. HA14-1 (M_r 409,000) induced apoptosis of human acute myeloid leukemia (HL-60) cells overexpressing Bcl-2 (31) that was associated with the decrease in mitochondrial membrane potential and activation of caspase-9 followed by caspase-3 (30). Furthermore, HA14-1 strongly induced the death of NIH 3T3 [Apaf-1(+/+)] cells but had minimal apoptotic effect on Apaf-1-deficient [Apaf-1(−/−)] mouse embryonic fibroblast cells (30).

We determined whether Bcl-2 overexpression can inhibit TRAIL-induced mitochondrial apoptotic signaling and downstream cleavage of XIAP in TRAIL-sensitive SW480 (32) human colon cancer cells. To address this question, we used isogenic SW480 colon cancer cells with and without ectopic Bcl-2 overexpression. Attempted reversal of Bcl-2–mediated resistance to TRAIL was performed using the small molecule Bcl-2 inhibitor, HA14-1.

The SW480 human colon cancer cell line was obtained from the American Type Culture Collection (Manassas, VA) and grown in RPMI 1640 supplemented with 5% fetal bovine serum, 2 mmol/L l-glutamine, and antibiotics. Cells were grown in a monolayer and maintained at 37°C in a humidified atmosphere including 5% CO₂. SW480 cells or stable Bcl-2 transfectants, described below, were seeded at a density of 3 × 10⁶ cells per 100-mm dish in medium. After 48 hours, medium was aspirated and replaced with medium containing drug. Recombinant human TRAIL ligand (rhTRAIL) comprising residues 114 to 281 was obtained from Calbiochem (San Diego, CA), and a preparation containing a 6× histidine-tag at the NH₂-terminal end of the extracellular domain of human TRAIL (Thr-95 to Gly-281) was obtained from R&D Systems (Minneapolis, MN). TRAIL was dissolved in 1× PBS and then diluted in medium for experiments. Other drugs used included HA14-1 (Calbiochem) and the caspase-9 inhibitor Z-LEHD-FMK (R&D Systems). After drug treatment, floating and attached cells were harvested for subsequent analysis.

SW480 cells (3 × 10⁵ in 2 mL of Leibovitz’s L-15) were plated in six-well Costar tissue culture plates. Twenty-four hours later, cells were stably transfected with vector alone (PC3-neo) or with the PC3-Bcl-2 plasmid [a gift of Dr. Timothy J. McDonnell (University of Texas M.D. Anderson Cancer Center, Houston, TX)]. Transfection was performed with LipofectAMINE reagent (Life Technologies, Inc., Carlsbad, CA), according to the manufacturer’s instructions. Positive transfectants were selected in medium containing 500 μg/mL gentamicin (Life Technologies, Inc.). Cell lines were established from individual colonies using cloning cylinders.

The effect of TRAIL on clonogenic survival was determined. In brief, PC3-neo or PC3-Bcl-2 cells were seeded in a six-well plate, and 24 hours later, TRAIL (100 ng/mL) was added and incubated with cells for 7 days. A known number of cells was then replated in 100-mm culture dishes and returned to the incubator to allow macroscopic colony development. Plates were fixed in 10% formaldehyde, stained with crystal violet, and visualized under a scanner. Total cell clone (more than 10 cells) numbers in each well were counted in an area of 200 mm² in the center of the wells under an inverted microscope.

Western analyses were performed as described previously (6). The protein concentration in the samples was measured using a Bradford protein assay kit (Bio-Rad, Richmond, CA). Protein concentrations were equalized and were then added to a 1× SDS-PAGE loading buffer. Samples were maintained at 4°C, boiled for 5 minutes, and then separated on 10 or 12% polyacrylamide gels. Proteins were then transferred electrophoretically to Immobilon-P PVDF membranes (Millipore, Billerica, MA). Membranes were blocked with 5% nonfat dry milk in 1× PBS overnight at 4°C and then incubated at room temperature with primary antibodies. For analysis of the cytosolic release of mitochondrial pro-apoptotic proteins, cytoplasmic-enriched cell fractions were used, and their separation is described under Cellular Fractionation. Blots were incubated with polyclonal antibodies against Bid (R&D Systems), caspase-3 active fragment (BD PharMingen, San Diego, CA) and XIAP (Cell Signaling, Beverly, MA). Additionally, monoclonal antibodies were used against caspase-8 and cytochrome c (BD PharMingen), Bcl-2 and Smac (Upstate Biotechnology, Lake Placid, NY), β-actin and Bax (Sigma, St. Louis, MO), full-length caspase-3 and DNA fragmentation factor 45 (DFF45; BD Transduction, San Diego, CA), tyrosine tubulin (ICN Biomedicals, Aurora, OH), and COX IV (Molecular Probes, Eugene, OR). Blots were washed with PBS containing 0.1% Tween 20 for 15 minutes each, and incubated with a secondary antibody (Sigma). The signal was detected by a chemiluminescent detection kit using alkaline phosphatase detection (Applied Biosystems, Foster City, CA).

SW480/neo and SW4890/Bcl-2 cells were plated at a density of 4 × 10⁶ in 100-mm³ dishes in RPMI containing 5% fetal bovine serum and allowed to proliferate for 48 hours before drug treatment. After treatment for 15 hours, cells were collected in 1× PBS and lysed in a permeabilization buffer (210 mmol/L d-mannitol, 70 mmol/L sucrose, 10 mmol/L HEPES, 5 mmol/L sodium succinate, 200 μmol/L EGTA, 0.15% bovine serum albumin, and 80 μg/mL digitonin) on ice, as described previously (33). The lysate was centrifuged at 13,000 × g, and the supernatant (cytoplasm-enriched fraction) was collected. Protein from the subcellular fractionation was quantified, separated on SDS-PAGE, and immunoblotted. For analysis of release of mitochondrial apoptotic proteins, SW480/Bcl-2 cells were treated for 15 hours and then lysed in mitochondrial lysis buffer with a Dounce homogenizer and subjected to centrifugation at 1,000 × g to pellet nuclei, as described previously (34). The supernatant was centrifuged at 10,000 × g to pellet the mitochondria-enriched heavy membrane fraction. The resulting supernatant was further centrifuged at 100,000 × g to obtain the cytosolic fraction. Protein from subcellular fractionation was quantified, separated on SDS-PAGE, and immunoblotted.

SW480 cells (wt and rhBcl-2 transfectants) were plated at a density of 2 × 10⁵ cells per well of a six-well plate 2 days before experimentation. Cells were washed with 1× PBS and treated as noted for 6 or 72 hours. After treatment, adherent and floating cells were collected. Pelleted cells were washed twice with 1× PBS and resuspended in 1× Annexin Binding Buffer (BD PharMingen) to a concentration of 1 × 10⁶ cells per milliliter. A 200-μL aliquot was taken and mixed with Annexin V-fluorescein isothiocyanate (FITC; BD PharMingen) and propidium iodide for 20 minutes at room temperature. After incubation, the cell solution was augmented to 500 μL with Annexin Binding Buffer, and the results were read using a Becton Dickinson fluorescence-activated cell sorting (FACS) Caliber bench-top flow cytometer. All samples were run in triplicate, and average binding was determined and graphed.

SW480 cells (2 × 10⁷; PC-3 neo or PC-3-h Bcl-2 transfectants) were grown in 10-cm dishes and incubated with TRAIL (25 and 100 ng/mL) for 24 hours. Cells were then harvested by scraping in lysis buffer as described previously (6). The cell lysates were placed on ice for 10 minutes and centrifuged at 10,000 × g for 10 minutes. The protein concentration in supernatants was determined using the Bradford assay (Bio-Rad). Caspase-8 activity was determined with the Caspase-8 Assay kit (Calbiochem) using Ac-IETD-pNA as a substrate with incubation at 37°C for 2 hours. Caspase-3 activity was determined using the Caspase-3 Cellular Activity Assay kit (Calbiochem) using Ac-DEVD-NA as a substrate and a reaction time of 2 hours at 37°C. Caspase-9 activity was determined using the caspase-9/Mch6 Colorimetric Protease Assay kit (MBL, Nagoya, Japan) using LEHD-NA as a substrate with incubation at 37°C for 2 hours. For all caspase activation assays, 100 μg of extracted protein were mixed with the reaction buffer and substrate (included in kits) in 96-well plates and incubated for 5 hours at 37°C. Samples were then read at a wavelength of 405 nm, and absorbances at 3 hours of reaction were recorded. Caspase activity was evaluated by the absolute absorbance of TRAIL-treated cells subtracted from the absolute absorbance of untreated control cells.

To determine the role of Bcl-2 in TRAIL-induced apoptosis, SW480 cells were stably transfected with the human Bcl-2 cDNA (SW480/Bcl-2) or vector alone (SW480/neo). G418-resistant clones found to overexpress Bcl-2 proteins were selected and used for subsequent experiments (Fig. 2,A). SW480 cells were incubated with increasing concentrations of TRAIL (0–200 ng/mL) or medium alone for 6 hours, and the extent of apoptosis was analyzed by annexin V-FITC labeling. In SW480/neo cells, TRAIL treatment induced apoptosis in a dose-dependent manner (Fig. 2,B) and also induced DNA fragmentation laddering (data not shown). Bcl-2 overexpression was found to significantly protect cells from TRAIL-induced apoptosis at all doses evaluated relative to SW480/neo cells (Fig. 2,B). Bcl-2 suppressed apoptosis to control levels except for the highest dosage (200 ng/mL) of TRAIL evaluated. To determine whether Bcl-2 exerts an effect on long-term cell survival, we performed a clonogenic assay to analyze reproductive cell death. SW480/Bcl-2 and SW480/neo cells were incubated in the presence or absence of TRAIL (100 ng/mL) for 7 days. Treatment of SW480/neo cells with TRAIL almost completely abrogated colony formation compared with untreated vector control cells, whereas SW480/Bcl-2 cells were protected from TRAIL-induced cell death (Fig. 3). Specifically, SW480/Bcl-2 cells showed an approximately 52% reduction in clone number after TRAIL treatment compared with a 95% reduction found in SW480/neo cells. These findings indicate that Bcl-2 overexpression not only delays cell death but significantly extends the survival of TRAIL-treated SW480 cells.

We examined caspase activation by performing immunoblot analysis of cytoplasm-enriched fractions of SW480 cells. Bcl-2 overexpression was found to inhibit the TRAIL-induced (50 ng/mL; 15 hours) cleavage of initiator caspase-8, Bid, and effector caspase-3 into its active M_r 18,000 fragment (Fig. 4). Activation of capase-8 requires a proteolytic cleavage of procaspase-8 (M_r 55,000) to a caspase-8 intermediate that is further cleaved. Inhibition of caspase-8 cleavage by Bcl-2 suggests that caspase-8 is activated both upstream and downstream of mitochondria by way of a feedback amplification loop (17). TRAIL treatment induced the loss of full-length Bid consistent with caspase-8-dependent Bid cleavage (13). Translocation of Bid to the mitochondria has been shown to engage the intrinsic apoptotic pathway after a death receptor stimulus such as TRAIL (12, 14). Furthermore, the TRAIL-induced cleavage of DFF45, a substrate of caspase-3, was completely blocked in SW480/Bcl-2 cells in contrast to SW480/neo cells (Fig. 4). The active form of caspase-3 cleaves DFF that functions downstream of caspase-3 to effect apoptosis (35). Using a colorimetric assay, TRAIL treatment was found to increase caspase-9 activity, which was reduced in Bcl-2–overexpressing cells (Fig. 5,A). We then determined whether the specific caspase-9 inhibitor Z-LEHD-FMK could block TRAIL-induced apoptosis in SW480/neo cells. Z-LEHD-FMK was found to block TRAIL-induced apoptosis, as measured by Annexin V-FITC binding, yet suppression was partially overcome at the higher dose of TRAIL evaluated (Fig. 5 B). Taken together, these data indicate that Bcl-2 inhibits initiator and effector caspase activation and cleavage of caspase substrates, e.g., DFF45, required for apoptosis induction. In addition, results for Z-LEHD-FMK further demonstrate the importance of a mitochondrial amplification step in TRAIL-induced apoptosis in SW480 cells.

Fractionation of SW480/neo and SW480-/Bcl-2 cells was performed with analysis of protein expression in the cytosolic fraction by immunoblotting. We found that ectopic Bcl-2 overexpression completely inhibited the TRAIL-induced cytosolic release of the mitochondrial pro-apoptotic proteins cytochrome c and Smac/DIABLO relative to TRAIL-treated SW480/neo cells (Fig. 6). Upon treatment with TRAIL, cytosolic Bax expression was lost in SW480/neo cells consistent with its translocation to the mitochondria (Fig. 6; ref. 36). In contrast, Bcl-2 overexpression blocked Bax translocation, trapping it in the cytosolic fraction. XIAP is the most potent IAP and is known to inhibit proteolytically processed caspase-9, caspase-7, and caspase-3 to exert its anti-apoptotic function (11). XIAP has also been shown to function as a ubiquitin ligase for activated caspases, thereby promoting their degradation (11). Upon TRAIL treatment, XIAP was cleaved in SW480/neo cells whereas cleavage was blocked in Bcl-2–overexpressing cells (Fig. 6). These data indicate that Bcl-2 inhibits TRAIL-induced Bax redistribution and the release of mitochondrial pro-apoptotic proteins, in addition to blocking downstream XIAP cleavage and inactivation. Accordingly, these results and suppression of apoptosis by a caspase-9 inhibitor indicate that a mitochondrial amplification step is required for apoptosis induction by TRAIL in SW480 cells.

We determined the ability of HA14-1, a small molecule Bcl-2 inhibitor, to reverse Bcl-2–mediated resistance to TRAIL-mediated apoptosis in Bcl-2–overexpressing SW480 cells. HA14-1 alone did not induce apoptosis in these cells or in SW480/neo cells, except at the highest dose (40 μmol/L; 72 hours) evaluated (Fig. 7). Consistent with these data, dosages of HA14-1 exceeding 25 μmol/L have been shown to exhibit a loss of selectivity for Bcl-2 (31). When SW480/Bcl-2 cells were incubated with HA14-1 plus TRAIL (50 ng/mL), enhancement of apoptosis was observed with a maximal augmentation seen at 20 μmol/L HA14-1 (Fig. 8,A). These data indicate that HA14-1 can augment TRAIL-induced apoptosis in Bcl-2–overexpressing cells. We then determined whether HA14-1 can reverse Bcl-2–mediated suppression of apoptotic signaling. SW480/Bcl-2 cells were treated with TRAIL (50 ng/mL), HA14-1 (20 μmol/L), or their combination; and cell fractionation and immunoblotting were performed. Neither TRAIL nor HA14-1 treatment alone was found to trigger the release of mitochondrial proapoptotic proteins in SW480/Bcl-2 cells (Fig. 8,B). However, coadministration of TRAIL and HA14-1 resulted in a reduction of cytosolic Bax expression, consistent with its redistribution to mitochondria, and a marked increase in cytochrome c release compared with cells treated with either drug alone. HA14-1 did not restore TRAIL-induced Smac/DIABLO release or XIAP cleavage (Fig. 8 B). These data demonstrate that HA14-1 can reverse Bcl-2–mediated resistance to TRAIL-induced apoptosis. Specifically, HA14-1 restored TRAIL-induced Bax translocation to mitochondria, which enabled the release of cytochrome c.

We found that Bcl-2 overexpression confers protection against TRAIL-induced apoptosis in both short-term cell culture experiments and a long-term clonogenic survival assay in SW480 cells. This result and the ability of a caspase-9 inhibitor to attenuate TRAIL-induced apoptosis indicates that TRAIL also engages the intrinsic apoptotic pathway, constituting a mitochondrial amplification step. Cross-talk between the extrinsic and intrinsic pathways is mediated by the Bcl-2 family member Bid that is cleaved by initiator caspase-8 activation in response to TRAIL exposure, as shown here (refs. 12, 13, 14; Fig. 1). We found that Bcl-2 overexpression blocked TRAIL-induced caspase-8 and Bid cleavage, suggesting that caspase-8 is activated upstream and downstream of mitochondria, as shown for TRAIL (29, 37) and CD95(APO-1/Fas) (17) in type II cells. In support of this finding, effector caspase-3 has been shown to serve as a feedback loop by cleaving caspase-8 to amplify the apoptotic signal (38, 39). We found that ectopic Bcl-2 inhibited caspase-3 cleavage to its M_r 18,000 active fragment as well as cleavage of the caspase-3 substrate DFF45 that is essential for DNA fragmentation during apoptosis (35). Bcl-2 overexpression also attenuated caspase-9 activation in response to TRAIL. Caspase-9 has been shown to activate caspase-8 in a caspase-3–dependent manner, generating an amplification loop (38, 40). Furthermore, an inhibitor of caspase-9 reduced caspase-8 activation and vice versa in ovarian cancer cells (41). Therefore, our findings suggest that attenuation of TRAIL-induced caspase-9 activation by Bcl-2, as shown here, may contribute to a reduction in caspase-8 activation. As shown here, Bcl-2 overexpression can decrease caspase activation upstream and downstream of mitochondria to inhibit TRAIL-mediated apoptosis.

Bcl-2 overexpression blocked the TRAIL-induced redistribution of Bax from the cytosol to the mitochondria that is necessary for cytochrome c release. A requirement for Bax in TRAIL-induced apoptosis was shown in Bax-deficient human colon carcinoma cells that were resistant to death receptor ligands, whereas Bax-expressing sister clones were sensitive (42).

Ectopic Bcl-2 inhibited the TRAIL-induced cytosolic release of cytochrome c and Smac/DIABLO. Whereas Bcl-2 is known to inhibit cytochrome c release (20), a report (43) in human tumor cell lines has shown that Bcl-2 can inhibit Smac/DIABLO release. Upon commitment to cell death, Smac/DIABLO is released from mitochondria and binds to IAP family members to neutralize their anti-apoptotic activity (9). Inhibition of Smac/DIABLO release by Bcl-2 can destabilize the XIAP caspase complex, leading to a reduced caspase-inhibitory function (43). Importantly, we found that Bcl-2 blocked TRAIL-induced cleavage and inactivation of XIAP, indicating that Bcl-2 also serves a regulatory function downstream of mitochondria. XIAP is known to bind to and inhibit caspase-9, caspase-7, and caspase-3 and can also promote caspase degradation (11). TRAIL-induced cleavage and inactivation of XIAP seems to enable full activation of the caspase cascade. Furthermore, overexpression of XIAP has been shown to confer resistance to TRAIL (29) and to other cytotoxic drugs (44). Other determinants of TRAIL sensitivity in colon carcinoma cells may include c-FLIP and Bcl-X_L (45). In this regard, the long isoform of c-FLIP was shown to be recruited to the DISC in response to TRAIL receptor engagement in HT29 cells (46). Together, our findings indicate that a mitochondrial amplification step is required for complete activation of effector caspases and apoptosis induction by TRAIL in SW480 colon cancer cells. We also provide evidence that an additional level of control of TRAIL signaling occurs downstream of mitochondria by XIAP that is negatively regulated by Bcl-2.

Strategies to overcome Bcl-2–mediated resistance to apoptosis have the potential to greatly increase treatment efficacy. One approach involves the use of small molecule inhibitors of Bcl-2 that bind to and inhibit Bcl-2 function. One such inhibitor, i.e., HA14-1, has recently been developed using molecular modeling techniques, yet very limited data are available as to its efficacy in reversing Bcl-2–mediated resistance. We determined whether HA14-1 could reverse the anti-apoptotic effect of Bcl-2 and potentially enhance TRAIL-mediated apoptosis. HA14-1 alone did not induce apoptosis except at the highest dose (40 μmol/L) evaluated. A prior study in hematopoetic cell lines found that a concentration of HA14-1 exceeding 25 μmol/L resulted in a loss of selectivity for Bcl-2 and marked cytotoxicity in all cell lines tested (31). Although treatment of SW480/Bcl-2 cells with HA14-1 (20 μmol/L) alone failed to overcome the Bcl-2–mediated suppression of mitochondrial apoptotic signaling, coadministration of HA14-1 and TRAIL restored apoptotic susceptibility. This occurred in association with Bax redistribution to the mitochondria, enabling cytochrome c release to occur. Failure of this combination to restore Smac/DIABLO release in SW480/Bcl-2 cells may potentially be related to differences in the mechanism or kinetics of Smac release. In this regard, Adrian et al. (43) found that Smac/DIABLO release was blocked by a broad-spectrum caspase inhibitor, whereas cytochrome c release was largely caspase independent, suggesting that Smac/DIABLO release occurs downstream of cytochrome c.

The ability of Bcl-2 to confer resistance to TRAIL-induced apoptosis and its reversal by HA14-1 have important therapeutic implications. Bcl-2 is expressed in multiple human tumor cell types, including colorectal cancers in the absence of gene rearrangements (47, 48, 49). Because human cancer cells are type II and most chemotherapeutic drugs mediate apoptosis by inducing mitochondrial dysfunction, Bcl-2 is an important contributor to multiple drug resistance. Analysis of Bcl-2 expression may, therefore, identify patients less likely to benefit from TRAIL treatment who are candidates for small molecule inhibitors of Bcl-2, Bcl-2 antisense, or small interfering RNA approaches. Using Bcl-2 small interfering RNA, a critical role for Bcl-2 in suppressing p53-dependent apoptosis was shown in colon cancer cells in which massive apoptosis was seen even in the absence of cytotoxic drugs (50). Furthermore, the requirement for mitochondrial amplification of a death receptor stimulus, as shown in this report, implies that targeting both the extrinsic and intrinsic apoptotic pathways may overcome Bcl-2–mediated resistance and increase therapeutic efficacy. In this regard, TRAIL cooperated synergistically with the chemotherapeutic agents CPT-11 and 5-fluorouracil to produce substantial tumor regression in athymic mice bearing solid tumors derived from cell lines (4) or from human colon cancers (4, 7). In human prostate cancer cells, Munshi et al. (27) demonstrated that TRAIL can enhance the efficacy of chemotherapeutic agents and that their combination partially abrogates the anti-apoptotic effect of Bcl-2 overexpression (46, 51). Furthermore, we found that the combination of TRAIL and nonsteroidal anti-inflammatory drugs, i.e., sulindac sulfide or NS 398, reduced HCT-15 colon cancer cell viability to a greater extent than did either drug along (6).

In summary, we demonstrate that TRAIL treatment requires a mitochondrial amplification step for full activation of the caspase cascade and apoptosis induction in SW480 human colon cancer cells. Bcl-2 overexpression inhibits TRAIL-induced apoptosis by exerting effects at the level of the mitochondria, as well as downstream to it, by regulating the cleavage and inactivation of XIAP. The small molecule inhibitor of Bcl-2, HA14-1, restores Bax redistribution, thereby regulating an important upstream signal of cytochrome c release. These data provide important mechanistic insights related to Bcl-2–mediated TRAIL resistance and suggest the efficacy of novel strategy for reversing TRAIL resistance in Bcl-2–overexpressing tumors.

We thank Luanne Wussow for excellent secretarial assistance in the preparation of the manuscript.