Resveratrol-induced Apoptosis Is Associated with Fas Redistribution in the Rafts and the Formation of a Death-inducing Signaling Complex in Colon Cancer Cells*

Resveratrol, a polyphenol found in grape skin and various other food products, may function as a cancer chemopreventive agent for colon and other malignant tumors and possesses a chemotherapeutic potential through its ability to trigger apoptosis in tumor cells. The present study analyses the molecular mechanisms of resveratrol-induced apoptosis in colon cancer cells, with special attention to the role of the death receptor Fas in this pathway. We show that, in the 10–100 μm range of concentrations, resveratrol activates various caspases and triggers apoptosis in SW480 human colon cancer cells. Caspase activation is associated with accumulation of the pro-apoptotic proteins Bax and Bak that undergo conformational changes and relocalization to the mitochondria. Resveratrol does not modulate the expression of Fas and Fas-ligand (FasL) at the surface of cancer cells, and inhibition of the Fas/FasL interaction does not influence the apoptotic response to the molecule. Resveratrol induces the clustering of Fas and its redistribution in cholesterol and sphingolipid-rich fractions of SW480 cells, together with FADD and procaspase-8. This redistribution is associated with the formation of a death-inducing signaling complex (DISC). Transient transfection of either a dominant-negative mutant of FADD, E8, or MC159 viral proteins that interfere with the DISC function, decreases the apoptotic response of SW480 cells to resveratrol and partially prevents resveratrol-induced Bax and Bak conformational changes. Altogether, these results indicate that the ability of resveratrol to induce the redistribution of Fas receptor in membrane rafts may contribute to the molecule's ability to trigger apoptosis in colon cancer cells.

with particularly high levels in grape skin (50 -100 g/g) and red wine (1,5-20 mg/liter). Because of their spectrum of biological activities, especially as antioxidants, polyphenolics are suspected to afford the cardioprotective effects of red wine identified by epidemiological studies (1)(2)(3)(4)(5). The discovery of resveratrol as a chemopreventive agent in colon and other carcinomas has offered renewed interest in grape products and dietary supplements based on resveratrol. In various in vitro and in vivo models, this polyphenolic compound has proved to be capable of retarding or preventing the stages of carcinogenesis (5). This protective effect could be related to the ability of resveratrol to arrest cell cycle progression (6 -9) or to trigger tumor cell death by apoptosis (10 -13).
A number of pathways leading to apoptosis converge on the mitochondria to induce the release of soluble molecules from their intermembrane space. One of these molecules is cytochrome c, which, in the cytosol, induces oligomerization of the adapter molecule Apaf-1 to generate a complex in which caspase-9 is activated. Active caspase-9 then triggers the catalytic maturation of caspase-3 and other downstream caspases, thus leading to cell death. The release of cytochrome c can be prevented by anti-apoptotic proteins of the Bcl-2 family, which are supposed to preserve the mitochondrial membrane integrity. The pro-apoptotic members of the Bcl-2 family known as BH3-only proteins behave as sensors of cellular damage and initiate the death process by either inhibiting anti-apoptotic members of the family or activating another group of proapoptotic, Bcl-2-related proteins, the Bax/Bak proteins, through induction of conformational changes (14). Various reports have shown that resveratrol could induce apoptosis of tumor cells through a mitochondrial-dependent pathway that partially depended on Bax conformational changes and cellular redistribution (15)(16)(17)(18).
Another signaling pathway to apoptosis implicates the death receptors at the plasma membrane level. One of the most studied death receptors is Fas (APO-1/CD95). In response to engagement by its specific ligand (FasL) or an agonistic anti-Fas Ab, the cytoplasmic domain of the trimerized receptor recruits the adaptor molecule FADD (Fas-associating protein with death domain) (19 -20). In turn, FADD recruits procaspase-8 to form the death-inducing signaling complex (DISC) in which caspase-8 is activated, thus leading to apoptosis (21)(22)(23). We and others (24,25) have shown that the Fas-mediated pathway may contribute, in a ligand-dependent or -independent manner, to the death of cells exposed to various stimuli, including anticancer drugs and immunomodulatory molecules (26). Whether resveratrol-induced apoptosis also implicates death receptors remains a controversial issue (10 -13).
In the present study, we addressed the role of the Fas-dependent pathway in resveratrol-induced apoptosis of colon carcinoma cells. We show that resveratrol induces conformational changes of Bax, its redistribution to the mitochondria and caspase-dependent apoptosis in the human colon adenocarcinoma cell line SW480. Resveratrol does not increase the expression of Fas nor does it induce the appearance of Fas-ligand at the surface of tumor cells, but the compound triggers the redistribution of the transmembrane Fas receptor in plasma membrane rafts. Then, the DISC is formed and actively contributes to Bax redistribution and cell death. These results suggest that redistribution of Fas in the plasma membrane rafts induced by resveratrol in colon carcinoma cells could account for the ability of this polyphenol to trigger apoptosis in these cells.

MATERIALS AND METHODS
Cell Lines, Vectors, and Transient Transfection-SW480, SW620, and HCT116 human colon carcinoma cell lines were obtained from the American Tissue Culture Collection (ATCC). SW480 cells were maintained in RPMI 1640 and HCT116 and SW620 cells in Eagle's minimum essential medium, (Biowhittaker Co, Fontenay sous Bois, France). Both media were complemented with 10% fetal calf serum and 2 mM Lglutamine (Biowhittaker). The vectors used for transient transfections include pCI-neo (Promega, Madison, WI), pCI-MC159, and pCI-E8 (kindly provided by Dr. J. I. Cohen, National Institutes of Health, Bethesda, MD) and pCI-FADD-DN (a kind gift from Dr. C. M. Zacharchuk). Exponentially growing cells were incubated with a mixture of 9 l of FuGENE 6 (Roche Applied Science) with 1 g of the aboveindicated plasmid together with 1 g of an EGFP-encoding plasmid (pEGFP-C1, Clontech, Palo Alto, CA). Transfected cells were treated 24 h later for indicated times before measuring the number of EGFPpositive, apoptotic cells.
Apoptosis Identification-Apoptosis was identified by staining the nuclear chromatin of trypsinized cells with 1 g/ml Hoechst 33342 (Sigma-Aldrich) for 15 min at 37°C. The percentage of apoptotic cells was determined by analyzing 300 cells. When indicated, cells were incubated for 1 h with 2 g/ml antagonistic anti-Fas ZB4 Ab or 5 g/ml NOK-1 anti-FasL Ab prior exposure to resveratrol (usually 30 M) or for 72 h with the anti-Fas CH11 agonistic Ab (100 ng/ml in the presence of 0.8 g/ml cycloheximide).
Flow Cytometry Analyses-Mitochondrial membrane depolarization was measured by using the DepSipher kit (R&D Systems) according to the manufacturer's instructions. The probe forms aggregates in normal polarized mitochondria that result in a green-orange emission of 590 nm after excitation at 490 nm whereas the monomeric form present in cells with depolarized mitochondrial membranes emits only green fluorescence at 527 nm. Briefly, 10 6 cells were incubated for 20 min at 37°C in the presence of 5 g/ml DepSipher solution, then washed twice in DPBS (Dulbecco's phosphate-buffered saline) (BioWhittaker) before analysis. As a control, cells were treated for 30 min with the uncoupling agent carbonyl cyanide m-chlorophenylhydrazone (CCCP, 100 M; Sigma-Aldrich). Analysis of Fas and FasL expression on the plasma membrane was performed by incubating the cells for 45 min at 4°C with either clone DX2 mouse anti-Fas Ab or clone H11 rat anti-FasL Ab or isotype-matched controls (Jackson ImmunoResearch Laboratories, West Grove, PA). Abs were diluted in DPBS containing 0.5% bovine serum albumin and 0.1% NaN 3 . After two washes in DPBS, cells were incubated for 45 min with a fluorescein isothiocyanate-labeled donkey anti-mouse or anti-rat IgG (Jackson ImmunoResearch Laboratories), and analysis was performed using a FACscan cytometer (BD Biosciences).
Immunofluorescence Studies-Tumor cells were seeded into tissue culture chambers at 20,000 per well (Chamber Slide, Invitrogen) for 24 h, then treated and subsequently fixed in 2% paraformaldehyde (Sigma-Aldrich, Chemical Co.) for 10 min at 4°C, washed twice with DPBS for 10 min, preincubated with 1% bovine serum albumin for 15 min at room temperature, and incubated with the primary Ab (DPBS, 0,1% saponin, 0.5% bovine serum albumin) for 2 h at room temperature. After washing, cells were incubated for 30 min with 488-Alexa goat anti-rabbit and/or 568-Alexa goat anti-mouse Ab (Molecular Probes, Eugene, OR). Nuclei were stained with Hoechst 33342. Analysis was made by using a fluorescence microscope (Nikon, Champigny, France) or a confocal laser-scanning microscope (28). A non-relevant isotypematching Ab was used as negative control (not shown).
Western Blotting-Cells washed in DPBS were lysed in boiling buffer (1% SDS, 1 mM sodium vanadate 10 mM, Tris, pH 7.4) in the presence of 1:50 Complete protease inhibitor mixture tablets (Roche Applied Science) for 10 min at 4°C. The viscosity of the samples was reduced by several passages through a 26-gauge needle. Mitochondrial (M) and cytosolic (C) fractions were prepared as previously described (29). Protein concentrations were measured using the Bio-Rad DC protein assay kit (Ivry-sur Seine, France). 30 g of protein were incubated in loading buffer (125 mM Tris-HCl, pH 6.8, 10% ␤-mercaptoethanol, 4.6% SDS, 20% glycerol, and 0.003% bromphenol blue), boiled for 5 min, separated on a polyacrylamide SDS-containing gel and transferred onto a polyvinylidene difluoride membrane (Bio-Rad). After blocking nonspecific binding sites overnight by 5% nonfat milk in TPBS (DPBS with 0.1% Tween 20), the membrane was incubated for 3 h at room temperature with the primary Ab, washed twice with TPBS and incubated for 30 min at room temperature with horseradish peroxidase-conjugated goat antimouse or anti-rabbit Ab (Jackson ImmunoResearch Laboratories). The membrane was washed twice with TPBS and revealed using an enhanced chemiluminescence detection kit (Amersham Biosciences) and autoradiography.
Immunoprecipitation-Cells (15 ϫ 10 6 ) were detached, washed in DPBS, and incubated in lysis buffer (30 mM Tris, pH 7.5, 150 mM NaCl, 1% Nonidet P-40, 10% glycerol, 1 mM phenylmethylsulfonyl fluoride, 4 g/ml aprotinin) for 15 min on ice. After centrifugation at 14,000 rpm at 4°C for 15 min, 0.1 g/ml of an anti-human CD95 monoclonal Ab (APO1-3, Alexis Co.) was added at 4°C overnight. Immune complexes were precipitated using protein G-Sepharose (Amersham Biosciences) during 2 h at 4°C and washed three times in lysis buffer. The precipitate was resuspended in a loading buffer and boiled for 5 min. Samples were resolved on SDS-polyacrylamide gel electrophoresis and transferred onto a polyvinylidene difluoride membrane (Bio-Rad) for Western blot analysis.

Raft Isolation and Biochemical
Characterization-Cells (100 ϫ 10 6 ) were washed with ice-cold PBS and lysed in 1 ml of buffer MES (Sigma-Aldrich) (25 mM MES, pH 6.5, 150 mM NaCl, complete protease inhibitor mixture (Roche Applied Science) containing 1% Triton X-100 for 30 min at 4°C, before passing them through an ice-cold cylinder cell homogenizer. Then, lysate was diluted with 2 ml of buffer MES containing 80% sucrose (w/v) and placed at the bottom of a linear sucrose gradient. The sample was centrifuged at 39,000 rpm for 20 h at 4°C, and 1-ml fractions were collected from the top of the gradient. Then, 60 l of each fraction were subjected to SDS-polyacrylamide gel electrophoresis and immunoblotted.
Lipids were extracted from these fractions by the method of Folch et al. (30). Phospholipid concentration was determined by evaporating an aliquot of the chloroformic phase, then adding 100 l of chloroform/ methanol 1/2 (v/v) for quantitative LC/MS (Liquid chromatography/ mass spectrometry). Phospholipid analysis was performed on a Hypersil Si 2*200 mm column (Agilent Technologies) with a binary gradient of solvent A (5 mM ammonium acetate in chloroform/methanol, 4:1) and solvent B (5 mM ammonium acetate in chloroform/methanol/water, 6:3.4:0.5). Positive ESI-MS (electrospray ionization-mass spectrometry) was performed on a MSD 1100 mass spectrometer (Agilent Technologies). To analyze cholesterol content, another aliquot of the chloroformic phase was evaporated before adding 60 l of KOH (10 M) and 1 ml of methanol and incubating at 56°C for 45 min. Then, 2 ml of chloroform and 1 ml of water were added before shaking, then evaporating the chloroformic phase. 100 l of a mixture of bis(trimethylsilyl)trifluoroacetamited/trimethylchlorosilane (4/1 v/v) (Acros Organics) were added to each samples before incubation for 1 h at 80°C, evaporation and addition of 100 l of hexane. Trimethylsilylethers of sterols analysis was performed by GCMS (Gas Liquid Mass Spectrometry) in a 6890 gas chromatograph coupled with a 7673 Mass Detector (Agilent Technologies). The column was an HP-5MS 30m*.25 mm (Agilent Technologies) and helium was used as the carrier gas. Concentrations of phospholipids and cholesterol were determined from the ratio of the peak area of a given molecule to the peak area of the internal standard. Levels were determined by comparison of this ratio with a standard curve of known amounts of cholesterol or of different species of phospholipids. Concentrations were expressed in g/mg total proteins.

Resveratrol Induces Caspase-mediated Apoptosis in SW480
Cells-When cultured in the presence of resveratrol, SW480 colon carcinoma cells underwent apoptosis in a dose-and timedependent manner. Cell staining with Hoechst 33342 demonstrated that resveratrol induced an increase in the nucleus size that preceded the appearance of characteristic apoptotic changes, i.e. the condensation and fragmentation of the nuclear chromatin (Fig. 1, A and B). These nuclear changes were inhibited by co-treatment with the cell permeant caspase inhibitor z-VAD-fmk (50 M), suggesting caspase involvement in the death process (Fig. 1B). The role of these proteases in resveratrol-induced apoptosis was further suggested by the observation that the nuclear DNA repair enzyme poly(ADP-ribose)polymerase (PARP) was cleaved into an N-terminal 89-kDa fragment in cells treated with 30 M resveratrol for 48 h (Fig. 1C).
Several experiments confirmed caspase involvement in resveratrol-induced SW480 apoptosis. First, immunoblots demonstrated that the M r 32,000 proform of caspase-3, the M r 55,000 proform of caspase-8, and the M r 47,000 proform of caspase-9 were cleaved into their active fragments during the death process (Fig. 1C). Secondly, active caspase-3 was detected by immunofluorescence in SW480 apoptotic cells by using an Ab that specifically detects the active fragments of the protease (Fig.  1D). Third, a time-dependent increase in the cleavage of fluorogenic peptide substrates that mimic the target cleavage sites of different caspases was identified in cell lysates (Fig.  1E). Lastly, various cell-permeant peptides designed to inhibit caspases with various specificities were observed to negatively interfere with resveratrol-induced cell death. Altogether, these results suggested that resveratrol induced SW480 cell apoptosis through one or several caspase-dependent pathways. Similar observations were made in SW620 and HCT116 human colon carcinoma cell lines when treated in the same conditions (not shown).
Resveratrol Induces Cytochrome c Release from the Mitochondria-One of the main pathways leading to caspase activation involves the release of cytochrome c from the mitochondria to the cytosol. To identify involvement of the mitochondria in resveratrol-induced apoptosis of SW480 cells, we measured first mitochondrial potential by using the DepSipher probe. SW480 cells were left untreated or treated with 30 M resveratrol for 48 h, stained with DepSipher, then analyzed by flow cytometry. Resveratrol induced a weak but reproducible increase in the percentage of cells that emitted only green fluorescence, indicating a decrease in mitochondrial ⌬⌿ m . This decrease was higher in SW480 cells treated for 48 h with FasL, which correlates with a higher level of apoptosis ( Fig. 2A). These changes in the mitochondrial membrane potential were associated with the release of cytochrome c from the mitochondria to the cytosol, as identified by immunoblot showing a decrease of cytochrome c expression in the mitochondrial fraction and its accumulation in the cytosolic fraction under resveratrol exposure (Fig. 2B). This apoptosis-associated redistribution of cytochrome c was confirmed by immunofluorescence studies that identified a punctuate staining pattern in the cytoplasm of untreated cells (Fig. 2C), similar to the pattern produced by the mitochondrial-specific Mitotracker Red dye (not shown), and a more diffuse cytosolic staining in resveratrol-treated cells exhibiting chromatin condensation and nuclear fragmentation (Fig. 2C).
Resveratrol Induces Bax Redistribution to the Mitochondria-Apoptosis-associated mitochondrial events have been shown to be modulated by Bcl-2 and related proteins. Western blot analyses indicated that exposure of SW480 cells to 30 M resveratrol for 48 h induced a decrease in the expression of anti-apoptotic Bcl-2 and Mcl-1 proteins without modifying Bcl-X L protein level (Fig. 3A). This treatment also induced an increase in Bax and Bak protein expression (Fig. 3A). Bax and Bak protein level decreased in the cytosol of resveratrol-treated cells and accumulated in the mitochondria (Fig. 3B). Bax and Bak involvement in apoptotic pathways was demonstrated to be associated with exposure of their N terminus (31)(32)(33). Using two Abs that recognize distinct epitopes in the N terminus of Bax that are exposed under activation (Fig. 3, C and D) and a monoclonal Ab that specifically recognizes the N terminus of activated Bak, resveratrol (30 M) was observed to trigger exposure of Bax and Bak N terminus epitopes in SW480 cells. In addition, both proteins colocalized with the mitochondrial HSP70 in resveratrol-treated cells (Fig. 3C). Changes in Bax immunoreactivity were observed as soon as 24 h of treatment, thus preceding mitochondrial changes and apoptosis (data not shown). Controls with irrelevant and secondary antibodies did not show any labeling (not shown). Treatment of SW480 cells for 1 h with the large spectrum caspase inhibitor, z-VAD-fmk (50 M) before exposure to resveratrol (30 M) inhibited apoptosis-associated nuclear changes but only partially decreased Bax and Bak N terminus exposure (Fig. 3D). The caspase inhibitory peptide z-IETD-fmk, designed to inhibit caspase-8, also partially decreased Bax and Bak N terminus exposure (Fig. 3D). This inhibition was associated with a partial inhibition of apoptosis, suggesting that an IETD-sensitive caspase may contribute to both Bax and Bak structural changes and cell death.

Resveratrol-induced Apoptosis Does Not Involve a Fas/FasL
Interaction-The observation that caspase-8 was activated in resveratrol-treated colon cancer cells undergoing apoptosis (Fig. 1) and the partial inhibitory effects of z-IETD-fmk on the death pathway (Fig. 3D) led us to explore whether Fas receptor could play a role in resveratrol-induced colon cancer cell death. We first analyzed whether resveratrol-induced apoptosis could involve an interaction between Fas receptor and its natural ligand. Flow cytometry analysis detected the expression of Fas at the surface of SW480 cells and this expression was not significantly influenced by a 48-h exposure to 30 M resveratrol (Fig. 4A). These colon cancer cells were observed not to express FasL at their surface and exposure to 30 M resveratrol for 48 h did not induce the expression of the ligand (Fig. 4A). This latter observation suggested that an autocrine or paracrine Fas/FasL interaction could hardly be responsible for resveratrol-induced apoptosis in the studied colon cancer cells.
To better assess this conclusion, SW480 cells were treated for 72 h with 30 M resveratrol or 100 ng/ml CH11 anti-Fas agonistic Ab (in the presence of 0.8 g/ml cycloheximide) for 24 h or soluble recombinant FasL (5 AU/ml) for 72 h. These treatments were performed in the absence or in the presence of either 2 g/ml ZB4, an antagonistic anti-Fas Ab (Fig. 4B), or 5 g/ml NOK-1, an anti-Fas-ligand Ab (Fig. 4B), added to the culture medium 1 h before cell treatment. Under these conditions, both ZB4 and NOK-1 Abs inhibited colon cancer cell apoptosis induced by CH11 Ab and soluble FasL. By contrast, neither ZB4 nor NOK-1 Abs had any significant influence on resveratrol-induced cell death (Fig. 4B). Similar observations were made in three other colon carcinoma cells, namely SW620, HT29, and HCT116 (not shown). SW480 cell death induced by CH11 agonistic anti-Fas antibody and resveratrol was observed to be additive, which further strengthened the demonstration that no interaction between FasL and Fas was involved in resveratrol-induced cell death (Fig. 4C).
Resveratrol Induces Fas Receptor Redistribution in the Plasma Membrane Rafts-To further analyze the potential role of Fas signaling in resveratrol-induced cell death, we then examined whether resveratrol could influence Fas receptor distribution in the plasma membrane. Using confocal laserscanning microscopy and ZB4 anti-Fas Ab, we observed that exposure of SW480 cells to resveratrol for 48 h (Fig. 5A) or 72 h (not shown) induced the clustering of the receptor at their surface, as previously observed in these cells when exposed to recombinant soluble FasL (24). Cell exposure to 10 ng/ml TNF-␣, which does not modify the pattern of Fas expression, was used as a negative control (Fig. 5A). A similar observation was made in three other human colon cancer cell lines (not shown). Fas colocalized with the raft-associated protein caveolin-2 at the surface of resveratrol-treated cells (Fig. 5B). To determine the mechanisms of this redistribution, cell lysates were fractionated on a sucrose gradient and the lipid content of these fractions was determined by HPLC-coupled mass spectrometry (Fig. 5C, upper panels). Immunoblot analysis of Fas expression indicated that Fas receptor could not be detected in the fractions enriched in cholesterol and sphingomyelin of untreated cells whereas the receptor was easily detected in these fractions when studied in resveratrol-treated cells (Fig. 5C). This observation suggested redistribution of Fas in the plasma membrane rafts, which were further identified by expression of caveolin-2, under resveratrol exposure. Interestingly, immunoblot analysis of these fractions also identified a redistribution of FADD and procaspase-8 but not procaspase-3 in the rafts (Fig. 5C). Cell treatment with nystatin (20 ng/ml) for 12 h did not modify the distribution of Fas, FADD, and procaspase-8 in SW480 cells. Nystatin prevented redistribution of these proteins in the rafts when added in the last 12 h of a 48-h treatment with resveratrol (Fig. 5D).
Fas Signaling Pathway Is Involved in Resveratrol-induced Apoptosis-We then analyzed whether the specific redistribution of FADD and procaspase-8, together with Fas, in the raft fraction of plasma membrane was associated with resveratrolinduced formation of a Fas-including DISC. Co-immunoprecipitation experiments using Apo 1-3 anti-Fas Ab demonstrated the time-dependent recruitment of FADD, procaspase-8, and, to a lesser extent, procaspase-10 to Fas in SW480 cells exposed to 30 M resveratrol (Fig. 6A). To determine whether resveratrol-induced formation of the DISC could contribute to apoptosis induction by this compound, we transiently transfected SW480 cells with constructs encoding either a dominant-negative mutant of FADD (F/DN) (34) or the viral proteins MC159 and E8 that were shown to inhibit apoptosis at the level of FADD and procaspase-8, respectively (35). A EGFP-encoding vector was co-transfected with each of this construct and apoptosis was studied in EGFP-positive cells (Fig. 6, B-D). Transient expression of either F/DN, MC159, or E8 was able to protect SW480 cells from resveratrol-induced apoptosis (Fig.  6B). In addition, transient expression of any of these proteins delayed caspase-3 activation triggered by resveratrol in SW480 cells (Fig. 6C) and partially prevented Bax and Bak conformational changes induced by the studied compound in these cells (Fig. 6, E and F). As a control, we observed that transient transfection of F/DN did not demonstrate any significant effect on staurosporine-induced apoptosis, as well as Bax and Bak conformational changes (Fig. 6, E and F).  Fig. 4) for 48 h before measuring apoptosis as in Fig. 1A (mean Ϯ S.D. of two independent experiments). C, caspase-3 expression was studied by using an Ab that specifically recognizes the active form of the enzyme and analyzed by flow cytometry (gray, control, black line, treated cells) and immunofluorescence (green, active caspase-3; blue, Hoechst-stained nuclei) in cells treated as in B. One representative of two experiments is shown. D, apoptosis induced by resveratrol was studied in EGFP-positive cells whose nucleus was stained with Hoechst 33342 (Ho). E and F, the expression of Bax (E) and Bak (F) was studied by immunofluorescence using Abs raised against their conformation-sensitive N terminus and the percentage of

DISCUSSION
The present study demonstrates that resveratrol induces the redistribution of Fas receptor in membrane rafts of colon carcinoma cells. This redistribution is associated with the formation of a death-inducing signaling complex that involves Fas, FADD, procaspase-8 and, to a lesser extent, procaspase-10, in the absence of any interaction with Fas-ligand. This signaling pathway contributes to Bax conformational changes, caspase activation, and apoptosis of resveratrol-treated colon cancer cells.
Resveratrol was proposed to function as a cancer chemopreventive agent through inhibition of promutagen bioactivation and stimulation of carcinogen detoxification, i.e. the polyphenolic compound inhibits cyclooxygenases COX1 and COX2 as well as some cytochrome P450 isoenzymes such as CYP 1A1 (36 -38). Resveratrol also demonstrated a protective effect toward reactive oxygen species and nitric oxide synthesis (39 -41). In addition to these chemopreventive effects, resveratrol possesses a chemotherapeutic potential through the inhibition of cell cycle, stimulation of differentiation, and induction of apoptosis. Its antiproliferative properties were associated with inhibition of ribonucleotide reductase, DNA polymerase, and ornithine decarboxylase whereas its cytotoxic effects were related to apoptosis induction. Resveratrol has demonstrated specific cytotoxic effects toward tumor cells when compared with normal lung (42) and blood cells (13). In accordance with previous observations in a variety of tumor cell lines, we observed that exposure of SW480 colon carcinoma cells to 10 -100 M concentrations of resveratrol for 48 h induced death by apoptosis.
Previous studies have documented that resveratrol-induced apoptosis involved caspase activation, both in vitro (10 -13) and in vivo (43,44), which was further confirmed in the cell lines tested in the present study. The caspases involved in the death process and the pathways leading to their activation could differ from one cell type to another. For example, resveratrol has been documented to activate the transcription factor p53, which was proposed to contribute to death (45,46), but the polyphenol also induces apoptosis in p53-deficient cells (47,48), indicating that p53 is not an absolute requirement for the cytotoxic effect of the molecule. According to caspases, resveratrol-induced apoptosis appears to converge on caspase-9 and downstream caspase-3 activation, suggesting a mitochondrialand cytochrome c-mediated mechanism. The decrease in the mitochondrial membrane potential ⌬⌿ m and the release of cytochrome c from the mitochondria to the cytosol indicate that the mitochondrial pathway to cell death is activated in resveratrol-treated colon cancer cells.
These mitochondrial events may be facilitated by the downmodulation of anti-apoptotic proteins such as Bcl-2 and Mcl-1 and the up-regulation of Bax and Bak. Resveratrol has been shown previously to trigger an increase in Bax expression, the exposure of its occluded N terminus and its translocation to the mitochondria in HCT116 colon cancer cells (49). In addition, Bax has been involved in the chemopreventive effect of resveratrol toward an animal model of colon carcinogenesis (44). However, a bax-independent pathway to cell death has been identified in an HCT116 colon cancer cell clone in which both bax alleles had been inactivated (49). The ability of resveratrol to trigger colon cancer cell apoptosis in the absence of Bax could be explained by the redundancy of Bax functions with those of Bak. Cells from mice deficient for both Bax and Bak, but not cells deficient for one or the other only, are almost completely resistant to mitochondria-mediated apoptosis (50). Exposure of SW480 cells to resveratrol induced conformational changes and mitochondrial redistribution of both Bax and Bak, suggesting that the two proteins are involved in resveratrol-induced cell death.
Receptor-mediated apoptosis was shown to depend upon upstream activation of caspase-8, which was capable of activating the downstream caspases resulting in apoptosis. Activation of caspase-8 identified in resveratrol-treated SW480 cells could have occurred upstream at the level of death receptors or downstream in the caspase cascade to amplify the apoptotic pathway. Initial description of the death pathway triggered by resveratrol in tumor cells involved up-regulation of FasL mRNA and FasL/Fas interaction in an autocrine or paracrine manner (13) but these results were subsequently challenged by several groups in various tumor models (10 -12), based on the observations that (i) FasL mRNA up-regulation was not confirmed, (ii) Abs that prevent the FasL/Fas interaction did not prevent resveratrol-induced apoptosis, and (iii) cell lines resistant to Fas-mediated apoptosis, e.g. leukemic cell lines, still underwent resveratrol-induced cell death. These arguments do not rule out the possibility that Fas plays a role in resveratrol induced cell death. We and others (24) have shown that the death receptor is involved in cytotoxic drug-induced tumor cell apoptosis through a Fas-L-independent, FADD-dependent pathway. In addition Fas pathway may not be an absolute requirement for resveratrol induced apoptosis, although it contributes to cell death when functional.
The present study provides a potential explanation to these controversies by showing that resveratrol does not increase the expression of Fas and FasL at the surface of tumor cells but induces a redistribution of Fas in the raft domains of the plasma membrane. These lipid microdomains result from preferential packing of complex sphingolipids and cholesterol in ordered plasma membrane structures and contain a variety of lipid-anchored and transmembrane proteins. Rafts play an important role in clustering or aggregating surface receptors, signaling enzymes and adaptor molecules into membrane complexes at specific sites and were shown to be essential for initiating signaling from a number of receptors. Accordingly, several recent studies have demonstrated an essential role for membrane rafts in the initiation of Fas-mediated apoptosis (51,52). In the present study, fluorescent microscopy analysis demonstrated a homogeneous distribution of the protein in untreated and a more clustered distribution in resveratrol-treated SW480 cells. In addition, resveratrol induced a redistribution of Fas, together with FADD and procaspase-8, in the fractions enriched in cholesterol and sphingolipids.
The mechanisms trapping receptor molecules in membrane rafts are yet to be characterized. Selective clustering of Fas was proposed to involve acid sphingomyelinase-induced release of ceramide in lymphocytes or fibroblasts (53). Whether the decrease content in cholesterol and sphingomyelin observed in the Triton X-100-insoluble fractions of resveratrol-treated compared with untreated SW480 cell lysates plays a role in Fas clustering in the rafts remains to be determined. Other hypotheses include hydrophobic modifications of the receptor, interaction with a binding partner that itself associates with raft lipids, or increased affinity induced by initial clustering of Fas. Whatever these mechanisms, resveratrol-induced redistribution of Fas in the rafts could contribute to the formation of the DISC observed in colon cancer cells treated with the polyphenol. This signaling complex is shown to contribute to Bax and Bak conformational changes, caspase-3 activation, and apoptosis observed in resveratrol-treated colon cancer cells. Whether other molecules involved in cell death, including other death receptors, are redistributed in the rafts under resveratrol exposure and contribute to the apoptotic process remains to be determined.
Altogether, our results suggest that Fas contributes to resveratrol-induced cell death in colon carcinoma cells through translocation and capping into membrane rafts. A similar observation was made recently in human leukemic cells exposed to a lipid ether (54). Additional studies are in progress to determine whether resveratrol induces a similar redistribution of other death receptors such as DR4 and DR5 into rafts and analyze if death receptor redistribution induces a synergy between resveratrol and death receptor ligands to trigger cancer cell death.