The role of reactive oxygen species and subsequent DNA-damage response in the emergence of resistance towards resveratrol in colon cancer models

Abstract

In spite of the novel strategies to treat colon cancer, mortality rates associated with this disease remain consistently high. Tumour recurrence has been linked to the induction of resistance towards chemotherapy that involves cellular events that enable cancer cells to escape cell death. Treatment of colon cancer mainly implicates direct or indirect DNA-damaging agents and increased repair or tolerances towards subsequent lesions contribute to generate resistant populations. Resveratrol (RSV), a potent chemosensitising polyphenol, might share common properties with chemotherapeutic drugs through its indirect DNA-damaging effects reported in vitro. In this study, we investigated how RSV exerts its anticancer effects in models of colon cancer with a particular emphasis on the DNA-damage response (DDR; PIKKs-Chks-p53 signalling cascade) and its cellular consequences. We showed in vitro and in vivo that colon cancer models could progressively escape the repeated pharmacological treatments with RSV. We observed for the first time that this response was correlated with transient activation of the DDR, of apoptosis and senescence. In vitro, a single treatment with RSV induced a DDR correlated with S-phase delay and apoptosis, but prolonged treatments led to transient micronucleations and senescence phenotypes associated with polyploidisation. Ultimately, stable resistant populations towards RSV displaying higher degrees of ploidy and macronucleation as compared to parental cells emerged. We linked these transient effects and resistance emergence to the abilities of these cells to progressively escape RSV-induced DNA damage. Finally, we demonstrated that this DNA damage was triggered by an overproduction of reactive oxygen species (ROS) against which cancer cells could adapt under prolonged exposure to RSV. This study provides a pre-clinical analysis of the long-term effects of RSV and highlights ROS as main agents in RSV's indirect DNA-damaging properties and consequences in terms of anticancer response and potent resistance emergence.

Figure 1

Resveratrol promotes transient accelerated senescence and apoptosis in vivo. (a) Effects of daily oral administrations of 40 and 200 mg/kg of RSV on the growth of PROb tumours in vivo. PROb cells were subcutaneously injected at day 0 to BD-IX rats and the treatments with RSV or with the vehicle (Ctl) began at day 2 or 14. Data are medians of tumour volumes of nine rats per group±the first and third quartile of one representative experiment among three. No statistically significant differences were found by the Mann–Whitney's test. Inserts show the median±the first and third quartile of plasma concentrations of free and metabolised RSV measured 1 h after the last administration of RSV to the rats of the experiment shown. (b) Microscopic evaluation of DNA fragmentation by TUNEL staining in PROb tumours. Pixels intensities were measured on 10 random microscopic fields of five tumours per group and is plotted as mean intensity±S.D. and compared with control tumours (Ctl). (c) Microscopic evaluation of SA-β-galactosidase activity in PROb tumours resected after 7 and 10 days of daily treatments with 200 mg/kg of RSV (R200). Negative pixels intensities measurement related to SA-β-galactosidase staining were performed as described above. Typical microscopic fields are shown on the left; bar, 100 μm. On both panels, statistical significance was determined by the Student's t-test with P<0.05 (*), P<0.01 (**) and P<0.001 (***)

Figure 2

Resveratrol induces a transient activation of the DNA-damage response in vivo. Levels of expression and phosphorylation of the components of the DNA-damage response were analysed by immunoblotting using the specified antibodies. Analyses were performed on PROb tumours extracts collected from five rats per group after 7 (a) and 10 days (c) of daily treatments with 200 mg/kg of RSV (R200) and compared to control tumours (Ctl). Quantifications by densitometry (b, d) are plotted as medians±the first and third quartile. Statistical analyses were performed by the Mann–Whitney's test and are shown as significant with P<0.05 (*) and P<0.01 (**)

Figure 3

Resveratrol induces a transient activation of the DNA-damage response in vivo. (a) Effects of daily oral administrations of 200 mg/kg of RSV on the growth of SW620 tumours in vivo. SW620 cells were subcutaneously injected at day 0 to nude mice and the treatments with RSV or with the vehicle (Ctl) began at day 2. Data are medians of tumour volumes of 9–12 mice per group±the first and third quartile. Statistical analyses were performed by the Mann–Whitney's test and are shown as significant with P<0.05 (*) and P<0.01 (**). (b, c) Levels of expression and phosphorylation of the components of the DNA-damage response were analysed by immunoblotting using the specified antibodies. Analyses were performed on SW620 tumours extracts collected from five mice per group after 11 (b) and 15 days (c) of daily treatments with 200 mg/kg of RSV (R200) and compared to control tumours (Ctl). Quantifications by densitometry are plotted as medians±the first and third quartile. Statistical analyses were performed by the Mann–Whitney's test and are shown as significant with P<0.05 (*) and P<0.01 (**)

Figure 4

Repeated treatments with resveratrol lead to a polyploidisation and the emergence of a resistance. (a) Growth curves of colon cancer parental cells PROb and SW620 treated with 30 μM of RSV (R30) at days 0 and 4 or left untreated (Ctl); proliferation of resistant cells R²PROb and R²SW620 obtained after 4-days pulses of R30 during two months is also presented. (b) Representative cell cycle analyses by PI/BrdU stainings of PROb and SW620 cells mock-treated (Ctl) or treated with R30 for 1 day (d1) and 5 days (d5). (c) Cumulative histograms (numbers shown are means) showing cell distribution in the different phases of their cell cycle after indicated times of R30 treatments in days (d). Flow cytometry gating procedure is depicted in B. Data are means±S.D. of three independent experiments

Figure 5

Resveratrol transiently induces cell death and senescence in colon cancer cells. (a,b) Analyses of the phenotypic responses of PROb and SW620 cells and of the isolated resistant towards RSV populations R²PROb and R²SW620. Cells were treated with 30 μM of RSV (R30) for indicated times and analysed by Hoechst 33342 and TUNEL co-stainings. On the left, representative fields showing typical features observed by microscopic analyses of cytospun cells, cells with normal nuclei (N), macronucleated (M), multinucleated (MN) and apoptotic ones; bar, 20 μm. On the right, cumulative histograms (numbers shown are means) showing the percentages of the different morphologies observed after indicated times of treatments in days (d) with R30 or mock-treated (Ctl). (c) Microscopic evaluation of SA-β-galactosidase activity in cells treated with R30. Representative microscopic observations are shown on the left after 5 days of R30 (d5); bar, 40 μm. Percentages of SA-β-galactosidase positive cells after indicated times of R30 treatments in days are shown on the histograms on the right. Data in both panels are means±S.D. determined by counting 300 cells in three independent experiments. Statistical significance was determined by the Student's t-test with P<0.001 (***)

Figure 6

Resveratrol promotes the phosphorylation of the DNA-damage marker histone H2AX. (a) Flow cytometric analyses of γH2AX/PI staining on PROb and SW620 cells after 1 day of treatment (d1) with 30 μM of RSV (R30) or mock-treated (Ctl); the percentages represent the amounts of γH2AX positive cells. (b) Microscopic analysis of γH2AX immunostainings of PROb cells grown on coverslips treated with R30 during 1 (d1) or 3 days (d3). Arrows show macronucleated cells (M) and apoptotic ones (A); bars, on the left 100 μm, on the right 20 μm. (c, d) Time course of expression and phosphorylation of the components of the ATR, ATM and DNA-PK pathways were analysed by immunoblotting using the specified antibodies. PROb and SW620 cells as well as their resistant counterpart, R²PROb and R²SW620 populations, were treated with R30 for indicated times in days (d) and compared with the untreated cells (Ctl)

Figure 7

Resveratrol effects are linked to the activation of the DNA-damage response. (a) Sensitising effects of specific inhibitors of major kinases involved in the DNA-damage response on the growth of parental and resistant colon cancer cells treated with 30 μm RSV (R30) for 24 h. Cells were pre-treated for 2 h with the PIKKs inhibitor caffeine (Caff), Chk1 inhibitor SB218078 (Chk1_i), ATM inhibitor KU55933 (ATM_i) or DNA-PK inhibitor NU7026 (DNA-PK_i). (b) Cell cycle analysis by PI staining of cells treated for 72 h with R30 and with the different inhibitors as described before. Flow cytometry analysis was performed by gating SubG1, diploid (2N) and polyploid (>4N) cells as shown on the 1D plots. Data shown in A and B are means±S.D. of three independent experiments. (c) Levels of expression and phosphorylation of components of the DNA-damage response analysed by immunoblotting using the specified antibodies. PROb and SW620 cells were treated with the specified inhibitors as described above

Figure 8

Reactive oxygen species mediate resveratrol-induced DNA damage and subsequent effects. (a) Levels of reactive oxygen species (ROS) in PROb and SW620 cells as well as on the resistant populations R²PROb and R²SW620 analysed by flow cytometry after 24 h of treatment with 30 μM of resveratrol (R30). The probes used were H₂DCFDA and MitoSOX which respectively detect the global level of ROS and the O₂⁻ produced by mitochondria. Cells were also grown in the presence of the antioxidant-reduced glutathione (GSH), N-Acetylcystein (NAC) and Trolox. (b) Effects of the supplementation of media with the antioxidants described above on the viability of cells treated with R30 for 24 h. (c) Effects of co-treatments of PROb cells with R30 and the antioxidants described above on the activation of H2AX and caspase-3 and on the expression of p21^Cip1 assayed by immunoblotting. (d) Antiproliferative activity of H₂O₂ on colon cancer cells was determined by treatments with the indicated concentrations of H₂O₂ for 18 h and after release in fresh media for 72 h. Data shown in A, B and D are means±S.D. of three independent experiments