Resveratrol metabolites inhibit human metastatic colon cancer cells progression and synergize with chemotherapeutic drugs to induce cell death

2.1 Cell lines

SW480 human colon carcinoma cell line and its derived metastatic cell line SW620 obtained from the American Tissue Culture Collection (ATCC, Rockville, MD, USA) were maintained in RPMI 1640 medium, (Biowhittaker Co, Fontenay sous Bois, France) supplemented with 10% fetal calf serum and 2 mM l-glutamine (Biowhittaker).

2.2 Drugs, chemical reagents, and antibodies

Trans-RSV, SN38, and oxaliplatin were obtained from Sigma-Aldrich (St Quentin Fallavier, France). R3S, R3G, and R4G were obtained from Bertin Pharma (Montigny-le-Bretonneux, France). For all experiments except those of dose–response, the three metabolites were used at 30 μM. For chemosensitization assays, RSV and RSV metabolites were used at 10 and 20 μM alone and the mixture (M) of metabolites was performed by using 10 or 20 μM of each metabolite. Polyclonal antibodies (Abs) against cyclin A, cyclin B1, cyclin E, Cdk1, Cdk2, γH2AX, ATM, ATR, p21, and p53 were purchased from Santa-Cruz Biotechnology (Santa-Cruz, CA, USA), human caspase-3 active fragments and PARP from Cell Signalling (Beverly, MA, USA), and mouse monoclonal Ab against β-actin from Sigma-Aldrich.

2.3 Stability studies

Emission spectra of trans-RSV (30 μM in water with 0.1% ethanol) and RSV metabolites alone (10 μM in water in 0.1% ethanol) were carried out in a 1-cm quartz cell with a spectrofluorimeter. Ultraviolet (UV) irradiations were obtained by exposure of solutions in quartz cells on a 6 × 15 W/312 nm UV transilluminator table. Metabolites (10 μM in water in 0.1% ethanol) were also exposed to air or to room temperature for 72 h before analysis by spectrofluorimetry. Data are expressed as relative fluorescence units.

2.4 Cell viability measurements

SW480 and SW620 cells were seeded 24 h before treatment into 24-well plates in complete growth medium. The next day, cells were challenged for 24, 48, or 72 h with RSV or with various concentrations of its metabolites alone, i.e. R3S, R3G, and R4G or with a mixture of these three metabolites (M). All control and treated cells received the same amount of ethanol (0.1%). After the indicated times, cells were washed with PBS and then stained with crystal violet (0.5% (w/v) for 5 min and then rinsed thrice with water. Absorbance was read at 540 nm after extraction of the dye by 0.1 M sodium citrate in 50% ethanol.

2.5 Cell cycle analysis by flow cytometry

Cell cycle analysis was performed as described previously 7. In brief, SW620 cells were grown in 6-well plates, treated or not with RSV or metabolites alone or with the mixture of the three metabolites (30 μM each). After 48 h of treatment, cells were labeled with 30 μg/mL 5-bromo-2′-deoxyuridine (BrdU, Sigma-Aldrich) for 1 h 30 min at 37°C. Then, detached and adherent cells were collected and pooled, washed with PBS, fixed in cold 80% ethanol and then incubated with a mouse anti-BrdU monoclonal antibody (Dako, Denmark) for 1 h at room temperature. After washing, pellets were subsequently incubated with a FITC-conjugated anti-mouse antibody (Dako) for 40 min and were then counterstained with a 50 μg/mL solution of propidium iodide (PI) containing 200 μg/mL of RNAse A. Stained cells were analyzed with a Galaxy flow cytometer (Partec, Münster, Germany) equipped with a 488 nm blue laser providing acquisition of 20 000 events. Red fluorescence from PI was detected through a 625 nm long pass filter (FL3, DNA content), and green fluorescence from FITC was detected through a 520/10 nm band pass filter (FL1, BrdUrd incorporation). The acquired data were analyzed with FlowMax software (Partec) and displayed as dual-parameter PI versus log-FITC histograms.

2.6 Western blotting

Cells washed in PBS were lysed in a RIPA buffer (10 mM Tris-HCl, pH 7.2, 150 mM NaCl, 0.5% Nonidet NP40, 0.5% Na deoxycholate, 0.1% SDS, 2 mM EDTA, and 50 mM NaF) in the presence of 1/25 complete protease inhibitor cocktail tablets (Roche Diagnostics Corporation) for 15 min on ice. Cell lysates were cleared by a 15 min centrifugation at 20 000 × g. Protein concentration was measured in the supernatant using the BCA protein assay kit (Sigma-Aldrich). Thirty micrograms of proteins were diluted in loading buffer (125 mM Tris-HCl, pH 6.8, 10% β-mercaptoethanol, 4.6% SDS, 20% glycerol and 0.003% bromophenol 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 for 2 h with 3% nonfat milk and 1% BSA in TBST (10 mM Tris-HCl, 150 mM NaCl, 0.1% Tween 20, pH8), the membrane was incubated overnight with the primary antibody diluted in TBST, washed twice with TBST and incubated for 45 min at room temperature with horseradish peroxidase-conjugated goat anti-mouse or anti-rabbit antibody (Santa-Cruz Biotechnology). The membrane was washed twice with TBST and revealed using an enhanced chemiluminescence detection kit (Amersham, Les Ulis, France) and autoradiography.

2.7 Apoptosis identification

Apoptosis was identified by staining the nuclear chromatin of trypsinized cells (untreated and RSV metabolites-treated cells) with 1 μg/mL Hoechst 33342 (Sigma-Aldrich) for 15 min at 37°C. The percentage of apoptotic cells was determined by analyzing 300 cells from randomly selected fields for each condition of treatment.

2.8 Chemosensitization assays

SW620 cells were seeded 24 h before treatment into 24-well plates. The next day, cells were left untreated or pretreated in triplicate wells with RSV, RSV metabolites or mixture of metabolites for 24 h and then challenged for additional 24 h with either SN38 (50 nM) or oxaliplatin (500 nM). After treatments, cells were washed with PBS and then stained with crystal violet (0.5% (w/v) as described previously in Section 2.3.

2.9 Statistical analysis

Data are expressed as means ± SD (n = 6) of at least three independent experiments. The significance of differences was established with the Student's paired t-test. Values of p < 0.05 were considered significant.

3.1 RSV metabolites inhibit colon cancer cell proliferation

To determine whether RSV metabolites exert a potential anticancer action, we first compared in human colorectal cancer cells, the antiproliferative capacities of the three main RSV metabolites identified in both healthy volunteers 19 and colon cancer patients 20 who received oral doses of RSV, R3S, R3G, and R4G, to 30 μM of RSV aglycone (R30). At this concentration, we have previously shown that RSV inhibits cancer cells proliferation 26, 27 without cytotoxicity in normal human monocyte and in normal rat intestinal IEC18 cells 8, 29. As various studies have shown that RSV is not stable after UV exposure or to air 28, we have tested the stability of RSV metabolites under UV exposure or air or room temperature (Fig. 1B and C). It appeared that emission spectra of RSV metabolites are modified by UV exposure with a shift to the left (Fig. 2B). We also observed instability of the glucuronide metabolites when they were exposed to sunlight or air for 72 h with a lowering of the spectrum (Fig. 2C).

As shown in Fig. 2A, only R3S was able to inhibit SW480 and its metastatic-derived SW620 cell line in a time and dose-dependent manner. Moreover, we observed that after 72 h of treatment inhibition rate induced by R3S was similar to that induced by R30, with 50% and 35% of inhibition in SW480 and SW620, respectively (Fig. 2A and B). Contrary to R3S, glucuronide metabolites did not significantly affect colorectal cancer cell lines growth (Fig. 2A). R3S, R3G, and R4G are found together in gastrointestinal tract 20, 24, we next tested the efficiency of an equimolar mixture of the three metabolites (M30) to determine a potential synergy. Interestingly, although RSV glucuronides used alone had no effect on cell proliferation, the mixture showed a potent synergic, time, and dose-dependent antiproliferative effect which was even more potent than RSV or R3S alone (Fig. 2B). Indeed, the combination of 30 μM of each metabolite inhibited by 65% and 80% cell proliferation after 48 and 72 h, respectively (Fig. 2B). This is an important point because it has been recently shown that after an oral administration of 1.0 g of RSV for 29 days, the plasma mean concentrations of the three RSV metabolites were around 30 μM 24. At this concentration, which is described in plasma levels of patients 24, the proliferation inhibition induced by the mixture was specific since it did not increased cell permeability after PI labeling (data not shown). Thus, in the subsequent experiments we used the combination of 30 μM of each metabolite.

3.2 RSV metabolites accumulate colon cancer cells in S phase of the cell cycle

We have previously shown that RSV affects cell proliferation by disturbing cell cycle progression 7. In order to determine whether metabolites, in particular R3S and M30, have the same effect we analyzed by flow cytometry the distribution of cells in cell cycle 7. Colon cancer cells were exposed for 48 h to the different compounds alone or with the mixture before addition of BrdUrd in the last hour of treatment. Cells with high fluorescence were S phase cells that incorporated BrdUrd. Those with background fluorescence (Fig. 3A) represented G₁ or G₂/M phase cells, depending on DNA content (2C or 4C, respectively).

The results reveal that at a time of treatment inducing an arrest of cell proliferation such as 48 h, all RSV metabolites induce marked modifications of the cell cycle at 30 μM (Fig. 3A). We particularly noticed that R3S increased by +65% the number of cells in the S phase which was comparable to the accumulation in this cell cycle phase induced by RSV that was used as a positive control. Interestingly, contrary to the action on cell proliferation, RSV glucuronides, R3G and R4G, induced an accumulation almost equivalent to that observed with R3S. In a same manner, the metabolites mixture induced a strong accumulation of cancer cells in this DNA replication phase with 75% of BrdUrd positive cells (Fig. 3B). These effects induced by RSV metabolites were associated with a decrease in the percentage of tumoral cells in G1 and in G2/M phases, respectively (Fig. 3B).

Furthermore, bivariate BrdUrd/DNA histograms shows that R3S and M30 treatments shifted BrdUrd positive cells to higher DNA content which suggest the induction of polyploidy processes (Fig. 3A).

3.3 RSV metabolites modulate key regulators of cell cycle

Cancer cells accumulation in S phase prompted us to investigate whether the key regulators of the cell cycle progression, cyclins and their partners, the cyclin-dependent kinases (Cdks), could be affected upon exposure to the studied compounds. As we have previously published 7, RSV (R30) increases cyclins A and B expression as well as Cdk 2 protein levels which have been shown to be required for DNA synthesis 30 (Fig. 4A). In this study, immunoblotting demonstrate that R3S and more strongly the mixture (M30) dramatically increased cyclin A, cyclin E, Cdk1, and Cdk2 expressions (Fig. 4). A concomitant cyclin B expression was also observed. Indeed, quantitative analysis show that after 48 h of treatment, R3S increases by 2.43-fold cyclin A, by 3.51-fold cyclin E, by 1.56-fold Cdk1, and by 1.67-fold Cdk2 protein expression levels as compared to control (Co) (Fig. 4B). In agreement with the previous results obtained on cell proliferation, the mixture (Mix 30) more potently enhanced the expression levels of key cell cycle (+3.49-fold for cyclin A, +11.26-fold for cyclin E, 2.72-fold for Cdk2) (Fig. 4B). In these experiments, RSV glucuronides slightly increased cyclins and Cdks expression after 48h of treatment (Fig. 4). These results underline the importance of cylin E and Cdk1 in the disruption of cell cycle progression induced by RSV metabolites, which we have previously demonstrated in colon cancer cells after RSV treatment 3, 7, 27.

3.4 RSV metabolites induce p53 and H2AX phosphorylation

Recent reports have correlated the delay through S→G2/M progression in the cell cycle to the increase in p53 phosphorylation and to the induction of DNA damages. Both processes are reported to depend on the DNA damage checkpoint kinases ataxia telangiectasia mutated (ATM)/ataxia telangiectasia-Rad3-related (ATR) 31 and Chk (checkpoint kinase) 32. ATR is a protein that is well known as a central mediator of responses to DNA double strand breaks and subsequent replication stress. Some studies have reported that RSV induces S phase arrest via an activation of DNA checkpoint pathways including ATM/ATR kinases 33.

RSV metabolites, which are able to induce a polyploidization (Fig. 3A), especially R3S and M30, induced a strong increase in phosphorylated-histone H2AX (γH2AX) which reflects DNA damage levels. Indeed, as well as RSV, R3S, and M30 strongly enhanced H2AX serine-139 phosphorylation after 48 h of treatment (Fig. 5) which represent +1.66- and +2.45-fold increases, respectively. In various models, RSV has been shown to induce DNA damages through the activation phosphoinositide kinase homologs such as ATR 31, 33, 34. Our results show that RSV metabolites alone or together increased ATR expression as compared to nontreated cells (Fig. 5). These activations lead to p53 activation, which has been shown to respond to DNA damage by an increase in its phosphorylation at serine-20 (Fig. 5). In agreement with the activation of p53, we observed that R3S and M30 induced p21 expression, which has been described as one of the transcriptional targets of tumor suppressor p53 protein (Fig. 5).

3.5 RSV metabolites induce apoptosis

Following DNA damages, cells either repair the damages or undergo apoptotic death mostly in a p53-dependent manner. To precise the relationships between tumoral cells accumulation in the S phase, the increase in polyploidy cells and the induction of ATR/p53 pathway signaling, we investigated the cell death induced by RSV metabolites. As we have previously shown 9, when cultured in the presence of RSV (R30), colon cancer cells underwent apoptosis (Fig. 6). This process was also observed with R3S and M30, as evidenced by tumor cells staining with Hoechst 33342, which showed characteristic apoptotic changes, i.e. condensation and fragmentation of nuclear chromatin (Fig. 6A). This apoptotic induction was not observed with glucuronide metabolites (Fig. 6A and B). R3S and M30-mediated apoptosis and DNA damages are confirmed with the cleavage of the nuclear DNA repair enzyme poly(ADP-ribose)-polymerase (PARP) into an N-terminal 89 kDa fragment and by the cleavage of the Mr 32 000 proform of caspase-3 into its active fragment (Fig. 6C).

3.6 Synergic effects of RSV metabolites with SN38 or oxaliplatin on colon cancer cells

We and others have shown that RSV could be used as a chemosensitizing agent or a chemotherapeutic adjuvant particularly due to, in some cancer models, cell cycle arrest, cyclins and Cdks modulations, and apoptosis induction 7, 10, 11.

The ability of RSV metabolites to induce (i) cell cycle arrest in DNA replication phase, (ii) apoptotic cell death, (iii) important DNA damages shown by γH2AX induction, seems to suggest that RSV metabolites could act as chemosensitizing agents. In order to evaluate this potential, we looked for a synergistic effect of RSV metabolites with SN38 (irinotecan's active metabolite), a classical anticancer agent widely used in the treatment of metastatic colon cancer 35. Recent studies have shown that SN38 resistant colon cancer cells present a huge reduction of DNA double strand breaks formation compared to sensitive cancer cells 36. In our previous experiment (Fig. 4), we observed a γH2AX increase in SW620 metastatic colon cells induced by R3S and the mixture, reflecting a DNA double strand break formation. Pretreatment with RSV, R3S or the mixture at 10 or 20 μM for 24 h dramatically enhanced SN38-induced growth inhibition of SW620 cells (Fig. 7A and B). Moreover, it is noted that R3G and R4G, that also induced ATR pathway (Fig. 5), cooperated with SN38 to kill tumor cells (Fig. 7). Similar results were obtained with another anticancer agent acting on DNA repair processes, oxaliplatin, which is a third generation platinum derivative (Fig. 7). These results are very important since at the concentrations used (10 and 20 μM) in these chemosensitizing experiments, RSV metabolites concentrations are compatible with plasmatic concentrations detected in patients receiving an oral dose of RSV 24.