Flavopiridol Induces Cellular FLICE-Inhibitory Protein Degradation by the Proteasome and Promotes TRAIL–Induced Early Signaling and Apoptosis in Breast Tumor Cells

Introduction

Flavopiridol is a semisynthetic flavone with potent inhibitory effects on cell proliferation in cultured tumor cell lines ( 1). This effect has also been observed in vivo, xenografting different human tumor cell lines in nude mice ( 2). The antitumor activity of flavopiridol has been related not only to cell cycle arrest but also to induction of apoptosis ( 3– 6) and antiangiogenic activity ( 7). In clinical trials of flavopiridol as a single agent and in drug combinations, evidence of antitumor activity has been observed at plasma concentrations that inhibit cyclin-dependent kinase (CDK)–related functions ( 8).

Tumor necrosis factor (TNF)–related apoptosis-inducing ligand (TRAIL) is a ligand of the TNF family capable of inducing apoptosis in a wide variety of cancer cells on binding to the proapoptotic receptors TRAIL-R1 and TRAIL-R2 but with no effect in the majority of normal human cells tested. However, despite the ubiquitous expression of TRAIL receptors, some cancer cells show either partial or complete resistance to TRAIL ( 9). Resistance of tumor cells to TRAIL can be overcome by treatment with protein synthesis inhibitors, which suggests the presence of apoptosis inhibitors in those cell lines ( 10).

Activation of TRAIL receptors leads to the formation of the death-inducing signaling complex (DISC), which includes the receptor itself, the adapter molecule Fas-associated death domain (FADD), and procaspase-8 ( 11). Processing and activation of caspase-8 at the DISC leads to a cascade of apoptotic events that results in the death of the cell. Inhibition of this cascade may occur at different stages in tumor cells ( 12). At the DISC level, the apoptotic signal may be inhibited by cellular FLICE-inhibitory proteins (c-FLIP), the homologue of viral FLIP in vertebrate cells ( 13). In most cells, c-FLIP exists as two alternative spliced isoforms: c-FLIP_L, a homologue of caspase-8, which lacks critical amino acids for proteolytic caspase activity, and c-FLIP_S, consisting only of two death effector domains ( 13). A third 23-kDa recently identified form called c-FLIP_R, with properties similar to those of c-FLIP_S, is also expressed in some cells ( 14). Although the role of c-FLIP in apoptotic signaling has been controversial, data obtained from cells stably overexpressing c-FLIP and from mice deficient in c-FLIP clearly support an antiapoptotic function (for review, see ref. 15). c-FLIP expression fluctuates in a cell type–specific manner and in response to various stimuli: transcriptionally through the nuclear factor-κB (NF-κB) pathway ( 16) and at the protein level via altered rates of proteasomal degradation ( 17), which makes it a versatile inhibitor of apoptotic responses mediated by death receptors.

To induce apoptosis in tumor cells through activation of death receptors represents a new therapeutic strategy against different types of cancer. However, whereas agonistic antibodies that activate death receptor CD95/Fas are very toxic for liver cells ( 18), TRAIL and agonistic antibodies that trigger TRAIL receptors may be considered as more suitable tools ( 19). TRAIL does not show toxic effects in preclinical studies with mice and nonhuman primates when given at doses that inhibit the growth of tumor xenografts ( 20). Although it has been reported that a particular form of human recombinant TRAIL may be toxic for human hepatocytes ( 21), more recent data have shown that different recombinant versions of TRAIL vary considerably in toxicity toward normal human cells while maintaining their antitumor properties ( 22).

Recent reports have described that the combination of flavopiridol and TRAIL induces apoptosis in various cancer cells, although the molecular mechanism of this synergistic action remains unclear. Flavopiridol has been reported to sensitize human non–small cell lung carcinoma (NSCLC) and colon cancer cells to TRAIL-induced apoptosis by inhibiting NF-κB ( 23). A reduction in the cellular levels of XIAP has been suggested to be responsible for the antileukemic effects of flavopiridol and TRAIL ( 24). On the other hand, it was reported that flavopiridol down-regulates Mcl-1 expression in human cholangiocarcinoma cells ( 25). Despite the fact that many cancer cells are sensitive to TRAIL-induced apoptosis, a number of them, in particular breast cancer cells, are resistant to this death ligand ( 26). In these cells, resistance to TRAIL seems to reside in the initial activation of caspase-8 ( 27). Taken together, these results prompted us to investigate whether flavopiridol could abrogate the resistance to TRAIL-induced apoptosis observed in breast tumor cells and, if so, elucidate the mechanisms of this sensitization.

In this study, we show that flavopiridol sensitizes all human breast cancer cell lines examined to a mitochondria-operated, TRAIL-induced apoptotic process. In the highly resistant to TRAIL, ErbB2-overexpressing cell line SKBR-3, flavopiridol facilitates DISC formation and caspase-8 activation at the DISC on TRAIL receptor ligation, without altering the levels of TRAIL receptors, FADD or caspase-8. In these cells, flavopiridol markedly down-regulates the levels of c-FLIP_L and c-FLIP_S proteins by a mechanism involving the proteasome. Moreover, we show that knocking down c-FLIP proteins with specific small interfering RNA (siRNA) oligonucleotides is sufficient to sensitize various breast tumor cell lines to TRAIL-induced apoptosis.

Materials and Methods

Reagents and antibodies. Flavopiridol was kindly provided by Dr. José Ramón Suarez (Aventis Pharmaceuticals, Bridgewater, NJ). MG-132, streptavidin-agarose beads, and mouse anti-α-tubulin antibody were obtained from Sigma Chemical Corp. (St. Louis, MO). Recombinant human TRAIL (residues 95-281) was produced as described previously ( 28). Caspases inhibitor benzyloxy-carbonyl-Val-Ala-Asp-(OMe) fluoromethyl ketone (Z-VAD-FMK) was from Enzyme System, Inc. (Dublin, CA). Anti-human caspase-8 monoclonal antibody was purchased from Cell Diagnostica (Münster, Germany). Anti-FADD monoclonal antibody was from BD Transduction Laboratories (Heidelberg, Germany). Anti-poly(ADP-ribose) polymerase (PARP) monoclonal antibody was from Roche Molecular Biochemicals (Monza, Italy). Anti-c-FLIP monoclonal antibody (NF6) and anti-TRAIL-R1, anti-TRAIL-R2, anti-TRAIL-R3, and anti-TRAIL-R4 antibodies were from Alexis Corp. (Lausen, Switzerland). Anti-cytochrome c, Bax, and XIAP monoclonal antibodies were from Biosciences PharMingen (San Jose, CA). Anti-cyclooxygenase IV (COX IV) rabbit polyclonal antibody was from Abcam (Cambridge, United Kingdom). Horseradish peroxidase– or FITC-conjugated secondary antibodies, goat anti-mouse and goat anti-rabbit, were obtained from DAKO (Cambridge, United Kingdom). Rabbit anti-Bid polyclonal antibody was generously provided by Dr. X. Wang (Howard Hughes Medical Institute, Dallas, TX).

Cell lines. Stable MCF-7 cell line overexpressing human Bcl-2 protein has been described previously ( 27). The cell lines were either maintained in RPMI 1640 (MCF-7, MDA-MB-231, EVSA-T, and Jurkat) or in DMEM (MDA-MB-468, SKBR-3, and BT474) containing 10% fetal bovine serum, 1 mmol l-glutamine, and gentamicin in a 5% CO₂ humidified atmosphere. Culture medium of MDA-MB-435S cells was also supplemented with insulin (10 μg/mL).

Determination of cell viability and apoptosis. Cell viability was determined by the crystal violet method as described ( 27). Hypodiploid apoptotic cells were assessed by flow cytometry according to published procedures ( 29).

Immunoblot analysis of proteins. The assay for measurements of cytochrome c and Bax was done as described ( 29). Proteins were resolved on SDS-polyacrylamide minigels and detected as described previously ( 27).

Isolation of the TRAIL DISC. DISC precipitation was done using biotin-tagged recombinant TRAIL (bio-TRAIL; ref. 30). SKBR-3 cells were incubated for 16 hours in the presence or absence of 100 nmol/L flavopiridol. After this incubation, cells were treated with bio-TRAIL for the times indicated in the figure legends. DISC formation was determined as reported ( 29).

Analysis of TRAIL receptors by flow cytometry. SKBR-3 cells were detached with RPMI 1640/3 mmol/L EDTA, washed in ice-cold PBS, and resuspended in PBS. Cells were then labeled with anti-TRAIL receptor antibodies (5 μg/mL) or no antibody (negative control) and then incubated with goat anti-mouse FITC-conjugated antibody F(ab′)₂ fragment. Labeled cells were analyzed by flow cytometry using the CellQuest software.

siRNA. siRNAs against c-FLIP (5′-GGGACCUUCUGGAUAUUUUtt-3′ and 5′-AAAAUAUCCAGAAGGUCCCtg-3′) and XIAP (5′-GCUGUAGAUAGAUGGCAAUtt-3′ and 5′-AUUGCCAUCUAUCUACAGCtg-3′) were synthesized by Ambion, Inc. (Austin, TX). Nontargeting scrambled siRNAs were synthesized by Proligo LLC (Boulder, CO). Cells were transfected with either c-FLIP, XIAP, or scrambled siRNAs (50 nmol/L for SKBR-3 cells or 10 nmol/L for MDA-MB-231, BT474, and MDA-MB-435S cells) using DharmaFECT 1 (Dharmacon, Lafayette, CO) as described by the manufacturer. After 48 (SKBR-3) or 24 (MDA-MB-231, BT474, and MDA-MB-435S) hours of transfection, medium was replaced with regular medium before further analysis.

Cell transfections and luciferase assays. Cells were cotransfected with 0.7 μg of a NF-κB luciferase reporter plasmid (pKBF-Luc; Dr. J.M. Redondo, Centro de Biologia Molecular, Madrid, Spain) and 0.25 μg β-galactosidase expression plasmid using Fugene reagent (Roche Applied Science, Mannheim, Germany). After transfection, cells were harvested, and luciferase activity was measured in an FB12 luminometer (Berthold Detection Systems, Pforzheim, Germany) according to the instructions of the Luciferase system kit (Promega, Madison, WI). Transfection experiments were done in duplicate. NF-κB activity was always normalized by measuring β-galactosidase activity and expressed as relative luciferase units.

Reverse transcription-PCR. Total RNA was isolated from cells with Trizol reagent (Life Technologies, Inc., Grand Island, NY) as recommended by the supplier. cDNA was synthesized from 2 μg total RNA using a RNA PCR kit (Perkin-Elmer, Indianapolis, IN) with the supplied random hexamers under conditions described by the manufacturer. PCRs were done using specific primers for c-FLIP_L and c-FLIP_S as described previously ( 31).

Statistical analysis. All data are presented as mean ± SE of at least three independent experiments. The differences among different groups were determined by the Student's t test. P < 0.05 was considered significant.

Results

Flavopiridol sensitizes human breast tumor cells to TRAIL-induced apoptosis. To investigate the mechanism of flavopiridol-induced sensitization to TRAIL, we initially used the human breast carcinoma cell line SKBR-3 that express elevated levels of ErbB2 and is markedly resistant to TRAIL-induced apoptosis ( 32). Cells were first incubated with different flavopiridol doses and subsequently treated with TRAIL, and apoptosis was determined as percentage of sub-G₁ population. Results shown in Fig. 1A show that these cells are very resistant to TRAIL-induced apoptosis and a marked sensitization to apoptosis by TRAIL can be observed in the range between 50 and 500 nmol/L flavopiridol. In subsequent experiments, we normally used 100 to 200 nmol/L flavopiridol because these concentrations produced almost maximal sensitization and were not toxic for the cells.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Flavopiridol sensitizes human breast tumor cell lines to TRAIL-induced apoptosis. A, SKBR-3 cells were incubated with the indicated flavopiridol concentrations for 7 hours before the addition of soluble recombinant TRAIL. Apoptosis was measured 17 hours after the addition of TRAIL as percentage of cells with sub-G₁ DNA content as described in Materials and Methods. B, SKBR-3 cells were treated with flavopiridol for 7 hours before the addition of soluble TRAIL at the indicated concentrations. Apoptosis was measured 17 hours after the addition of TRAIL as in (A). C, cells were incubated with the indicated flavopiridol concentrations for 7 hours before the addition of soluble TRAIL at the concentrations shown in the different panels. Apoptosis was assessed 17 hours after the addition of TRAIL. Columns, average of three different experiments; bars, SE. *, P < 0.005; **, P < 0.0001.

We next examined the effect of different TRAIL concentrations in SKBR-3 cells preincubated with 200 nmol/L flavopiridol. As can be seen in Fig. 1B, flavopiridol promoted TRAIL-induced apoptotic cell death at the lowest concentration of TRAIL used (50 ng/mL) and a maximal effect was observed at doses between 250 and 500 ng/mL TRAIL. The later TRAIL concentration was subsequently used in the experiments done in SKBR-3 cells.

Sensitization to TRAIL-induced apoptosis by flavopiridol was a general phenomenon of breast tumor cells. In a panel of breast tumor cell lines (BT474, MDA-MB-231, MCF-7, MDA-MB-468, and EVSA-T) differing in variables, such as expression of ErbB2, presence of estrogen receptors, p53 status, or sensitivity to TRAIL, flavopiridol markedly sensitized all of them to TRAIL-induced apoptosis ( Fig. 1C; data not shown).

Down-regulation of XIAP or abrogation of NF-κB activity by flavopiridol is not sufficient to promote TRAIL-induced apoptosis in breast tumor cells. Down-regulation of XIAP by flavopiridol has been reported to be responsible for the antileukemic effects of flavopiridol and TRAIL ( 24). We have examined the role of XIAP in the resistance of breast tumor cells to TRAIL-induced cell death and the sensitization to TRAIL observed on flavopiridol treatment. Flavopiridol caused a marked decrease in XIAP protein levels in SKBR-3 cells in a dose-dependent manner ( Fig. 2A ), with maximal effects being observed at doses that induced a clear sensitization to TRAIL ( Fig. 1A). We next determined whether specific down-regulation of XIAP by siRNA methodology could be sufficient to abrogate resistance to TRAIL in SKBR-3 cells. Results shown in Fig. 2B show that a siRNA targeted against XIAP efficiently down-regulated XIAP protein to levels similar to those achieved on flavopiridol treatment ( Fig. 2A). However, in contrast to the sensitization observed with flavopiridol, down-regulation of XIAP with siRNA failed to promote sensitization to TRAIL-induced apoptosis ( Fig. 2C).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Down-regulation of endogenous XIAP and abrogation of NF-κB activity by flavopiridol are not sufficient to promote apoptosis induced by TRAIL. A, SKBR-3 cells were not treated or treated with the indicated concentrations of flavopiridol for 24 hours. XIAP levels were analyzed by Western blotting. Tubulin was used as a protein loading control. Results are representative of two independent experiments. B, SKBR-3 cells were transfected either with siRNA oligonucleotide targeting XIAP or with a scrambled RNA oligonucleotide as described in Materials and Methods. After 72 hours, cells were harvested for immunoblot analysis to verify protein knockdown. Tubulin was used as a protein loading control. Results are representative of three independent experiments. C, SKBR-3 cells were transfected as in (B), and after 72 hours, TRAIL was added to the cultures. Apoptosis was measured 24 hours after the addition of TRAIL as described in Fig. 1. Columns, average of two different experiments; bars, range. D, SKBR-3 cells were transfected with a NF-κB reporter plasmid, and 6 hours later, some of the transfected cultures were treated with flavopiridol for 17 hours. Next day, another set of transfected cells were treated with Bay 11-7085 (Bay) 1 hour before the addition of TRAIL. Luciferase activity was determined 6 hours after the addition of TRAIL and normalized to the β-galactosidase activity. Columns, average of three different experiments; bars, SE. *, P < 0.005. E, SKBR-3 cells were incubated either with flavopiridol for 7 hours or with Bay 11-7085 for 1 hour before the addition of soluble recombinant TRAIL. Apoptosis was measured 17 (flavopiridol pretreatment) or 24 (Bay 11-7085 pretreatment) hours after the addition of TRAIL as described in Fig. 1. Columns, average of three different experiments; bars, SE. **, P < 0.001.

Flavopiridol inhibits NF-κB activity in tumor cells and sensitizes NSCLC and colon cancer cells to TRAIL-induced apoptosis ( 23). We have studied the role of NF-κB in the regulation of TRAIL-induced apoptosis by flavopiridol in breast tumor cells. In SKBR-3 cells, flavopiridol caused an important inhibition of NF-κB activity as measured with a NF-κB promoter-driven luciferase reporter plasmid ( Fig. 2D). In these experiments, the specific inhibitor of NF-κB activation Bay 11-7085 decreased NF-κB promoter activity to a similar extent as observed for flavopiridol. Although it has been reported that TRAIL can stimulate the NF-κB pathway in various human tumor cell lines ( 33), in SKBR-3 cells, TRAIL failed to induce NF-κB promoter activity ( Fig. 2D). Similar results were obtained in other breast tumor cell lines (data not shown). Furthermore, in the presence of TRAIL, both flavopiridol and Bay 11-7085 reduced NF-κB activity to a similar extent ( Fig. 2D). However, in contrast to the sensitization observed with flavopiridol, Bay 11-7085 was not able to promote TRAIL-induced apoptosis in SKBR-3 breast tumor cells ( Fig. 2E), which suggests that inhibition of NF-κB by flavopiridol is not sufficient to cause sensitization to TRAIL-induced apoptosis.

Treatment with flavopiridol enhances TRAIL-induced activation of caspase-8 and the mitochondrial apoptotic pathway. To further establish the mechanism of flavopiridol-promoted, TRAIL-induced cell death, we examined the caspase dependency of this cell death process. We found that the generation of sub-G₁ cells induced by the combination of flavopiridol and TRAIL was dependent on caspase activation as it was completely prevented by the general caspase inhibitor Z-VAD-FMK ( Fig. 3A ).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Apoptosis induced by the combination of flavopiridol and TRAIL requires caspase activation and involves activation of caspase-8 and the mitochondrial pathway of apoptosis. A, SKBR-3 cells were pretreated with flavopiridol (200 nmol/L) for 7 hours before the addition of soluble TRAIL (250 ng/mL) for 17 hours; Z-VAD-FMK (50 μmol/L) was added 2 hours before TRAIL. Apoptosis was measured as percentage of sub-G₁ cells. Columns, average of three different experiments; bars, SE.**, P < 0.001. B, SKBR-3 cells were treated with 100 nmol/L flavopiridol for 16 hours before 8 hours of incubation with 500 ng/mL TRAIL. Activation of caspase-8 was assessed by Western blotting as described in Materials and Methods. Results are representative of three independent experiments. C, SKBR-3 cells were treated as in (B). Loss of Bid was assessed by Western blotting. Results are representative of two independent experiments. D, SKBR-3 cells were treated as in (B). Translocation of cytosolic Bax and release of cytochrome c from mitochondria were assessed by Western blotting. Results are representative of two independent experiments. COX IV was used as a mitochondrial loading control. E, SKBR-3 cells were treated as in (B). PARP degradation was assessed by Western blotting. Results are representative of three independent experiments. In all the previous experiments, tubulin was used as a protein loading control. F, MCF-7^pc and MCF-7^Bcl-2 were pretreated or not with 100 nmol/L flavopiridol during 6 hours before the addition of 50 ng/mL TRAIL. After 18 hours, cellular viability was measured by crystal violet staining. Columns, average of three different experiments; bars, SE. **, P < 0.001.

During TRAIL-induced apoptosis, procaspase-8 is recruited and processed at the DISC in a FADD- and TRAIL receptor–dependent manner ( 34). Procaspase-8 is first cleaved to the p43/p41 intermediate fragments releasing the small subunit p12 and then subsequently processed to generate the large catalytically active p18 subunit ( 11). To test if caspase-8 activation by TRAIL was blocked in the TRAIL-resistant cell line SKBR-3, processing of procaspase-8 was determined by Western blotting in cells treated with TRAIL following incubation in the absence or presence of flavopiridol. As shown in Fig. 3B, TRAIL-induced processing of procaspase-8 to its 43- to 41-kDa intermediate fragments was observed only in the cells previously treated with flavopiridol.

Activation of caspase-8 leads to the processing of its substrate Bid generating a 15-kDa fragment, which translocates to mitochondria ( 35). Moreover, on death receptor activation, the cytoplasmic protein Bax migrates to the mitochondria where it cooperates with truncated Bid in the release of cytochrome c ( 36). Thus, we examined the activation of the mitochondria-controlled apoptotic pathway by TRAIL in flavopiridol-treated SKBR-3 cells by determining the loss of intact Bid, Bax translocation from citosol to mitochondria, and the release of cytochrome c from mitochondria. Results shown in Fig. 3C indicate that the amount of intact Bid was clearly diminished in the cells treated with flavopiridol and TRAIL in comparison with the cultures treated only with TRAIL. We next analyzed the loss of Bax and the presence of cytochrome c in cytosolic extracts from SKBR-3 cells on TRAIL receptor ligation. We observed that pretreatment with flavopiridol highly enhanced both events ( Fig. 3D). These effects were associated with the increase of Bax and the loss of cytochrome c levels in the mitochondria-containing membrane fraction of lysed SKBR-3 cells.

To confirm that the apoptosis cascade was fully active in SKBR-3 cells treated with both flavopiridol and TRAIL and that caspase activation was involved in the process, we analyzed the proteolytic degradation of the nuclear protein PARP, a substrate of effector caspases. As shown in Fig. 3E, PARP cleavage was only clearly induced in cells pretreated with flavopiridol and subsequently treated with TRAIL.

To get further insight into the importance of mitochondria-mediated events in flavopiridol-induced sensitization to TRAIL in breast cancer and because it is difficult to obtain stable transfectants of SKBR-3 cells overexpressing Bcl-2 protein, we studied whether TRAIL-induced apoptosis was facilitated by flavopiridol in a clone of MCF-7^Bcl-2 cells, which we have described previously as resistant to TRAIL ( 37). As shown in Fig. 3F, flavopiridol failed to promote TRAIL-induced loss of cell viability in MCF-7^Bcl-2 cells, in contrast to what was observed in cultures of cells transfected with an empty vector (MCF-7^pc). In these experiments, Bcl-2 overexpression did not affect caspase-8 activation by the combination of TRAIL and flavopiridol (data not shown).

Treatment with flavopiridol does not affect the expression levels of TRAIL receptors but increases DISC formation on TRAIL treatment in SKBR-3 cells. To investigate the possibility that flavopiridol treatment might change TRAIL receptor levels at the cell surface, SKBR-3 cells were cultured in the presence or absence of 100 nmol/L flavopiridol for 16 hours, and the expression of both proapoptotic and antiapoptotic receptors was analyzed by flow cytometry. Results in Fig. 4A show that SKBR-3 cells express the proapoptotic receptors TRAIL-R1 and TRAIL-R2 and the antiapoptotic receptor TRAIL-R4. Expression of the antiapoptotic receptor TRAIL-R3 was not detected in these cells. Results shown in this figure show that flavopiridol treatment did not change the expression of either proapoptotic or antiapoptotic TRAIL receptors in SKBR-3 cells.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

Treatment with flavopiridol does not affect the expression levels of TRAIL receptors in SKBR-3 cells but enhances DISC formation and induces down-regulation of c-FLIP_L and c-FLIP_S proteins in several breast tumor cell lines. A, SKBR-3 cells cultured in six-well plates were incubated either with or without flavopiridol (100 nmol/L) for 16 hours. Cells were then harvested, and cell surface receptor expression was assessed by flow cytometry using monoclonal antibodies to TRAIL-R1, TRAIL-R2, TRAIL-R3, or TRAIL-R4 as described in Materials and Methods. Discontinuous line, cells labeled with secondary antibody alone were used as a control for background fluorescence. Data are representative of two independent experiments. B, SKBR-3 cells were either treated or not with 100 nmol/L flavopiridol for 16 hours before the incubation with bio-TRAIL (1 μg/mL) for the indicated times. Unstimulated receptor controls (u/s) are the addition of bio-TRAIL to an equivalent volume of lysate isolated from unstimulated cells. DISC was isolated as described in Materials and Methods, and its components caspase-8 and FADD were analyzed by Western blotting. Lysates are included as a positive control for the expression of these proteins in SKBR-3 cells. Two different exposures of the caspase-8 DISC Western blot are presented to clearly show the differences in procaspase-8 recruitment and processing in the DISC between untreated or flavopiridol-treated cells. Data are representative of four independent experiments. C, analysis by Western blotting of c-FLIP_L, c-FLIP_S, caspase-8, and FADD in SKBR-3 cells treated or not with 100 nmol/L flavopiridol for the indicated times. α-Tubulin was blotted to show equal loading. Data are representative of three independent experiments. D, SKBR-3 cells were treated or not with the indicated concentrations of flavopiridol for 24 hours. c-FLIP_L and c-FLIP_S levels were analyzed by Western blotting. Tubulin was used as a protein loading control. Results are representative of two independent experiments. E, analysis by Western blotting of c-FLIP_L and c-FLIP_S in the indicated cells treated or not with 100 nmol/L flavopiridol during 17 hours. α-Tubulin was blotted to show equal loading. Data are representative of three independent experiments.

As the resistance to TRAIL occurred at a step before caspase-8 activation in SKBR-3 cells, it was possible that these cells were not able to form a functional DISC. In SKBR-3 cells incubated in the absence of flavopiridol, TRAIL induced the formation of a DISC with maximal recruitment of procaspase-8 and FADD in 30 minutes ( Fig. 4B). Within the DISC, procaspase-8 was processed to its p43 and p41 forms at the different times tested, with maximal procaspase-8 processing at 30 minutes after TRAIL addition. Interestingly, pretreatment with flavopiridol clearly promoted an increase in TRAIL DISC formation and procaspase-8 processing when compared with cells cultured in the absence of flavopiridol as indicated by the higher amounts of procaspase-8 and FADD in the DISC. Most importantly, in flavopiridol-treated cells but not in untreated cells, activation of procaspase-8 processing at the DISC following TRAIL addition proceeded to completion as determined by the appearance of the p18 subunit of mature caspase-8 at 30 and 90 minutes on TRAIL stimulation.

Flavopiridol induces down-regulation of c-FLIP proteins in breast tumor cells. Because flavopiridol did not alter the levels of TRAIL receptors ( Fig. 4A) but promoted TRAIL DISC formation and caspase-8 activation in SKBR-3 cells ( Fig. 4B), we hypothesized that the mechanism of action of this CDK inhibitor could involve either the up-regulation of other proapoptotic DISC components or the down-regulation of intracellular inhibitory proteins. In this respect, we observed that FADD and procaspase-8 levels were not altered by flavopiridol treatment ( Fig. 4C). One of the antiapoptotic proteins acting at the level of death receptors is c-FLIP, a short-lived protein with homology to caspase-8 that lacks caspase activity and inhibits death receptor–mediated activation of caspase-8 ( 13). Two alternatively spliced forms of c-FLIP, c-FLIP_L and c-FLIP_S, exist in different cell types ( 13). To investigate whether these short-lived proteins could be involved in the sensitization process induced by flavopiridol in SKBR-3 cells, we examined the effect of flavopiridol in c-FLIP protein levels in SKBR-3 cells. Cells were treated with 100 nmol/L flavopiridol for different periods and the levels of c-FLIP isoforms in cell lysates were analyzed by Western blotting. As shown in Fig. 4C, treatment with flavopiridol caused a marked decrease in the levels of both c-FLIP isoforms at 8 hours, and down-regulation of the short form of c-FLIP was complete after 16 hours of treatment. Down-regulation of c-FLIP isoforms by flavopiridol was dose dependent ( Fig. 4D) and correlated well with the flavopiridol concentrations that promoted TRAIL-induced apoptosis in SKBR-3 cells ( Fig. 1A). Moreover, down-regulation of c-FLIP isoforms by flavopiridol was observed in all the human breast cancer cell lines tested ( Fig. 4E).

Studies with metabolic inhibitors acting on either transcription or translation have indicated that these drugs sensitize resistant cells to death receptor–mediated apoptosis by down-regulating c-FLIP protein expression ( 38). We have examined in SKBR-3 cells whether the protein synthesis inhibitor cycloheximide could render these cells sensitive to TRAIL. Data not shown indicate that treatment of SKBR-3 cells with TRAIL in the presence of cycloheximide resulted in a marked induction of apoptosis. Both c-FLIP isoforms were lost on incubation of the cells with cycloheximide (data not shown).

A proteasome inhibitor blocks down-regulation of c-FLIP by flavopiridol and apoptosis induced by the combination of flavopiridol and TRAIL in SKBR-3 cells. It has been shown that flavopiridol down-regulates gene expression broadly ( 39), possibly through its inhibitory action on the positive transcription elongation factor b (P-TEFb; cyclin T1/CDK9) complex ( 40). On the other hand, it has been described previously that c-FLIP levels can be modulated through a ubiquitin-proteasome pathway ( 17). We first examined by reverse transcription-PCR (RT-PCR) whether SKBR-3 cells express mRNA for both c-FLIP_L and c-FLIP_S isoforms using the human leukemic T-cell line Jurkat as a positive control and specific oligonucleotide primers for c-FLIP_L and c-FLIP_S as described in Materials and Methods. Results are shown in Fig. 5A , where it can be seen that breast tumor SKBR-3 cells only have mRNA for c-FLIP_L, whereas Jurkat cells express both mRNAs. In agreement with these data, we have observed that the short c-FLIP isoform present in SKBR-3 cells is slightly smaller than c-FLIP_S ( Fig. 5A). In this respect, a third splice isoform of c-FLIP named c-FLIP_R has been described recently in certain cell types ( 14). Whether the short c-FLIP isoform present in SKBR-3 cells corresponds to c-FLIP_R is an issue that requires further investigation. To test whether modulation of c-FLIP expression by flavopiridol in SKBR-3 cells occurred at the mRNA level, we did RT-PCR analysis of c-FLIP_L mRNA in cells treated for different times with 100 nmol/L flavopiridol. Results shown in Fig. 5B indicate that, in contrast to the observed down-regulation by flavopiridol of c-FLIP_L protein levels, treatment with flavopiridol did not change the levels of c-FLIP_L mRNA at any of the time points examined. Higher doses of flavopiridol (1 μmol/L) strongly reduced c-FLIP_L mRNA expression (data not shown) as reported previously ( 39).

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

A proteasome inhibitor, but not Z-VAD-FMK, prevents down-regulation of c-FLIP_L and c-FLIP_S by flavopiridol. A, top, c-FLIP mRNA was analyzed by RT-PCR (40 cycles) in Jurkat and SKBR-3 cells using specific primers for c-FLIP_L and c-FLIP_S. β-Actin was included as a control for mRNA input. Data are representative of two independent experiments. Bottom, c-FLIP_L and c-FLIP_S isoforms were analyzed in cell extracts from Jurkat and SKBR-3 cells. B, c-FLIP_L protein and mRNA levels were analyzed by Western blotting (WB) and RT-PCR in SKBR-3 cells after incubation with or without 100 nmol/L flavopiridol for the indicated times. RT-PCR was done using c-FLIP_L primers for 27 cycles. RT-PCR product of β-actin was used as a control for mRNA input. Data are representative of two independent experiments. C, SKBR-3 cells were pretreated or not with 25 μmol/L MG-132 for 30 minutes before culturing them in the presence or absence of 100 nmol/L flavopiridol for the indicated periods. c-FLIP_L and c-FLIP_S levels were analyzed by Western blotting. Arrow, c-FLIP_S band; asterisks, two protein bands that are up-regulated by MG-132. Tubulin was used as a protein loading control. Results are representative of two independent experiments. D, SKBR-3 cells were pretreated or not with either 25 μmol/L MG-132 or 50 μmol/L Z-VAD-FMK for 30 minutes before culturing them in the presence or absence of 100 nmol/L flavopiridol during 17 hours. c-FLIP_L and c-FLIP_S levels were analyzed by Western blotting. Arrow, c-FLIP_S band; asterisks, two protein bands up-regulated by MG-132. Tubulin was used as a protein loading control. Results are representative of two independent experiments. E, SKBR-3 cells were treated or not with 25 μmol/L MG-132 for 30 minutes before culturing them with or without flavopiridol for 7 hours. TRAIL was then added to the cultures and apoptosis was measured 16 hours after the addition of TRAIL as described in Fig. 1. Columns, average of three different experiments; bars, SE. **, P < 0.001.

To assess the role of the proteasome in the down-regulation of c-FLIP protein levels on flavopiridol treatment, SKBR-3 cells were treated with the proteasome inhibitor MG-132 before the addition of 100 nmol/L flavopiridol. After different periods, the levels of c-FLIP in cell lysates were analyzed by Western blotting. As shown in Fig. 5C, treatment with MG-132 blocked flavopiridol-induced down-regulation of both c-FLIP isoforms at all the times tested. In contrast, a general caspase inhibitor (Z-VAD-FMK) did not prevent flavopiridol-induced loss of c-FLIP expression ( Fig. 5D). Then, we determined whether the proteasome inhibitor could abrogate flavopiridol-induced sensitization to TRAIL-induced apoptosis. MG-132 markedly inhibited the apoptosis induced by the combination of flavopiridol and TRAIL in SKBR-3 cells ( Fig. 5E), under the same conditions that blocked c-FLIP degradation by flavopiridol. These data indicate that c-FLIP protein is down-regulated in SKBR-3 cells treated with flavopiridol through a proteasome-mediated pathway and that this decrease in c-FLIP protein levels could be responsible for the apoptosis induced by flavopiridol and TRAIL in these breast tumor cells.

Knockdown of endogenous c-FLIP_L and c-FLIP_S with siRNA oligonucleotides promotes apoptosis induced by TRAIL in several breast cancer cell lines. Our data indicate that flavopiridol decreases c-FLIP expression and sensitizes breast tumor cells to TRAIL-induced apoptosis. Next, we tested the possibility that specific ablation of c-FLIP expression by siRNA has a similar effect on TRAIL-induced apoptosis. In SKBR-3 cells, the levels of both c-FLIP_L and c-FLIP_S were substantially reduced with a specific siRNA ( Fig. 6A ), whereas a similar concentration of scrambled siRNA did not modify c-FLIP protein expression. Transfection with c-FLIP siRNA had no effect on the expression of procaspase-8, a structurally related proapoptotic protease ( Fig. 6A). Most importantly, silencing of c-FLIP expression resulted in a clear sensitization to TRAIL-induced apoptosis ( Fig. 6B). These data correlate well with the effect of flavopiridol and support the hypothesis that down-regulation of c-FLIP expression by this CDK inhibitor is critical for the induced sensitization to TRAIL in breast tumor cells. To further substantiate the role of c-FLIP in the resistance of breast tumor cells to TRAIL, we conducted siRNA experiments in other breast tumor cell lines. Results shown in Fig. 6C show that c-FLIP siRNA specifically knocked down the expression of both c-FLIP isoforms in different breast tumor cell lines. Furthermore, silencing of c-FLIP proteins caused a marked sensitization to TRAIL-induced apoptosis in the various cell lines analyzed ( Fig. 6D), a further indication of the relevance of c-FLIP proteins in contributing to the resistance of breast tumor cells to TRAIL. These results also point to the importance of c-FLIP down-regulation by flavopiridol in the mechanism of sensitization to TRAIL-induced apoptosis by this CDK inhibitor.

• Download figure
• Open in new tab
• Download powerpoint

Figure 6.

Knockdown of endogenous c-FLIP_L and c-FLIP_S in several breast cancer cell lines enhances apoptosis induced by TRAIL. A, SKBR-3 cells were transfected with either a siRNA oligonucleotide targeting both c-FLIP isoforms or scrambled RNA oligonucleotide as described in Materials and Methods. After 72 hours, cells were harvested for immunoblot analysis to verify protein knockdown. Tubulin was used as a protein loading control. Results are representative of two independent experiments. B, SKBR-3 cells were transfected as in (A), and after 48 hours, TRAIL was added to the cultures. Apoptosis was measured 24 hours after the addition of TRAIL as described in Fig. 1. Columns, average of two independent experiments; bars, range. C, BT474, MDA-MB-435S, and MDA-MB-231 cells were transfected with either a siRNA oligonucleotide targeting both c-FLIP isoforms or a scrambled RNA oligonucleotide as described in Materials and Methods. After 24 hours, cells were harvested for immunoblot analysis to verify protein knockdown. Tubulin was used as a protein loading control. Results are representative of two independent experiments. D, BT474, MDA-MB-435S, and MDA-MB-231 cells were transfected as in (C), and after 24 hours, TRAIL was added to the cultures. Apoptosis was assessed 24 hours after the addition of TRAIL as described in Fig. 1. Columns, average of two different experiments; bars, range.

Discussion

TRAIL has emerged recently as a potential anticancer agent due to its ability to induce apoptosis in numerous tumor cells, although normal cells are not killed by this death ligand ( 9, 22). However, despite the fact that many cancer cells are sensitive to TRAIL-induced apoptosis, a number of them, in particular breast cancer cells, are resistant to TRAIL ( 26). In these cases, a combination therapy using chemotherapeutic agents and TRAIL may be more suitable than using TRAIL as a single agent, and in fact, several studies in this direction have been conducted (reviewed in ref. 41).

In our study, we show that pretreatment of human breast cancer cells with flavopiridol, a potent inhibitor of CDKs undergoing clinical trials ( 42, 43), promotes the activation by TRAIL of a caspase-dependent mechanism of apoptosis in these cells. We show that sensitization to TRAIL-induced cell death by flavopiridol is a general phenomenon occurring in all human breast cancer cell lines examined. Our results also show for the first time that, in breast tumor SKBR-3 cells treated with TRAIL, full activation of apical caspase-8 at the DISC is blocked. When resistant cells were treated with flavopiridol, we detected an increased recruitment of both procaspase-8 and FADD to the TRAIL DISC and complete activation of caspase-8 as the earliest event occurring in the sensitization process. Because there are not changes in the levels of TRAIL receptors, FADD or procaspase-8, on flavopiridol treatment of SKBR-3 cells, one possibility to explain the increased recruitment of FADD to the DISC could be that post-translational modifications of either FADD or TRAIL receptors that reduce the binding affinity of the DISC components ( 30) are affected by flavopiridol treatment. Alternatively, competition between FADD and c-FLIP_L for the death domain of TRAIL receptor DR5 ( 44) could also modulate DISC formation. A reduction in c-FLIP_L levels on flavopiridol treatment, as observed in our work, would favor binding of FADD and procaspase-8 to the TRAIL DISC.

The hypothesis that short-lived inhibitory protein(s) may be involved in the resistance of SKBR-3 cells to TRAIL was confirmed by our data using the protein synthesis inhibitor cycloheximide, which rendered SKBR-3 cells sensitive to TRAIL, as has been shown in other systems ( 10). In this regard, a potential mechanism involved in the regulation of TRAIL sensitivity is the modulation of the expression of the death receptor inhibitors c-FLIP_L and c-FLIP_S ( 13), proteins with a very short half-lives ( 17) that are expressed at high levels in breast tumors ( 45). Metabolic inhibitors, such as cycloheximide or actinomycin D, sensitize cells to death receptor–induced apoptosis by strongly down-regulating c-FLIP expression ( 38). Although enhancement of TRAIL-induced caspase-8 activation and apoptosis does not always correlate with c-FLIP levels ( 29), it has been shown that different treatments can induce down-regulation of c-FLIP and subsequent sensitization to TRAIL-induced apoptosis ( 46, 47). In our study, treatment of different breast tumor cells with flavopiridol induced a marked down-regulation of both c-FLIP_L and c-FLIP_S, which suggests that loss of c-FLIP expression may underlie the observed sensitization to TRAIL. It has been described that flavopiridol could inhibit NF-κB activation ( 23), and it is well documented that TRAIL can also activate this pathway ( 48), modulating the sensitivity to TRAIL. On the other hand, because c-FLIP can be transcriptionally regulated through the NF-κB pathway ( 16), it is plausible to think that flavopiridol could inhibit c-FLIP expression through NF-κB inhibition. However, we did not observe a decrease in c-FLIP mRNA in SKBR-3 cells treated with flavopiridol concentrations that caused maximal sensitization to TRAIL, therefore excluding the transcriptional regulation of c-FLIP as the target for the inhibitory action of flavopiridol. Furthermore, the NF-κB inhibitor Bay 11-708 at a dose that inhibits NF-κB activation failed to down-regulate c-FLIP expression (data not shown).

It has been reported that c-FLIP protein levels can be down-regulated by proteasomal degradation ( 17) and this modulation may lead to TRAIL sensitization ( 49). In this respect, our results show that, in breast tumor cells, flavopiridol treatment leads to the proteasome-mediated degradation of both c-FLIP isoforms, and a proteasome inhibitor abolishes the sensitization by flavopiridol to TRAIL-induced apoptosis. At present, we do not know the mechanism by which flavopiridol induces the proteasomal degradation of c-FLIP proteins. Flavopiridol has been reported to activate the c-Jun NH2-terminal kinase (JNK) pathway ( 50). On the other hand, JNK activation by TNF-α reduces c-FLIP_L stability by a mechanism involving JNK-mediated phosphorylation and activation of the E3 ubiquitin ligase Itch, which ubiquitinates c-FLIP_L and induces its proteasomal degradation ( 51). Whether a similar mechanism is responsible for the flavopiridol-induced proteasomal degradation of c-FLIP in breast tumor cells is an issue that requires further investigation.

It has been shown recently that flavopiridol induces apoptosis in combination with TNF-α or TRAIL in some human cancer cell lines ( 23). It was proposed that inhibition by flavopiridol of TNF-α-induced NF-κB activation was responsible for the observed synergism on apoptosis. However, the mechanism involved in the sensitization to TRAIL was not elucidated. More recently, a reduction of XIAP or Mcl-1 levels by flavopiridol has been also suggested to be involved in the sensitization to TRAIL-induced apoptosis ( 24, 25). In our work, we have assessed the role of NF-κB and XIAP on flavopiridol-promoted, TRAIL-induced apoptosis in breast tumor cells. Our data clearly show that, in these cells, TRAIL does not stimulate NF-κB activity. Furthermore, although flavopiridol reduced basal NF-κB activity, this effect was not sufficient to explain the sensitization of breast tumor cells to TRAIL-induced apoptosis because a more specific NF-κB inhibitor did not exhibit a similar sensitizing effect. On the other hand, we have evaluated the role of XIAP in TRAIL resistance in breast tumor cells. Experiments with siRNA oligonucleotides clearly showed that elimination of XIAP protein did not abrogate resistance to TRAIL. Therefore, the down-regulation of XIAP levels observed in cells treated with flavopiridol is not sufficient to explain the sensitization observed. Besides its effects on the cell cycle ( 52), flavopiridol regulates transcription due to the potent inhibition of P-TEFb, composed of CDK9 and cyclin T1, which controls the elongation phase of transcription by RNA polymerase II ( 40). In this respect, it has been reported that flavopiridol decreases the mRNA levels of the antiapoptotic proteins XIAP, cIAP-2, Mcl-1, Bcl-x(L), and survivin in breast cancer cells ( 5). However, because all these proteins block TRAIL-induced apoptosis downstream caspase-8 activation, their down-regulation by flavopiridol does not explain the early sensitization induced by this flavonoid in SKBR-3 cells.

The mechanism underlying prevention of initiator procaspase-8 activation by c-FLIP within the DISC is still unclear. Although c-FLIP_S seems to inhibit the activation of this initiator caspase, the role of c-FLIP_L as an antiapoptotic protein is more controversial. c-FLIP_L is processed within the DISC to a 43-kDa subunit that remains at the DISC and a 12-kDa fragment that is released ( 53). In the presence of c-FLIP_L, procaspase-8 is partially cleaved to its p43/p41 fragments within the DISC, and the cleavage intermediates of both caspase-8 and c-FLIP_L remain bound to the receptor, preventing further recruitment of procaspase-8 to replace the processed caspase-8 at the DISC. On the contrary, recent data have indicated that heterodimerization of caspase-8 with c-FLIP_L leads to increased proteolytic activity of caspase-8 ( 54). Our results are compatible with the presence of c-FLIP_L at the DISC of SKBR-3 cells treated with TRAIL, where we detect the initial processing of procaspase-8. In cells pretreated with flavopiridol, there is full activation of caspase-8, which could correspond to a decreased c-FLIP/caspase-8 ratio within the DISC. In this respect, a high c-FLIP/caspase-8 ratio has been observed in B-cell chronic lymphocytic leukemia, which is highly resistant to TRAIL ( 55). In our studies with siRNA methodology, we have shown clearly the important role of c-FLIP levels in the resistant of different breast tumor cells to TRAIL. These results, in conjunction with the data indicating that flavopiridol markedly reduces c-FLIP expression and potently sensitizes breast tumor cells to TRAIL, further support the hypothesis that activation of the proteasome-mediated degradation of c-FLIP by flavopiridol is responsible for the elimination of resistance to TRAIL. Flavopiridol can be a useful approach in tumor therapy because it is able to modulate the sensitivity of tumor cells to other therapeutic strategies. In this respect, our data indicate that the sensitizing effects of flavopiridol to nontoxic death receptor ligands, such as TRAIL, could have therapeutic potential in the treatment of human breast cancer.