Resveratrol regulates the PTEN/AKT pathway through androgen receptor-dependent and -independent mechanisms in prostate cancer cell lines

Abstract

The tumor suppressor gene PTEN (phosphatase and tensin homolog deleted on chromosome 10) and the androgen receptor (AR) play important roles in tumor development and progression in prostate carcinogenesis. Among many functions, PTEN negatively regulates the cytoplasmic phosphatidylinositol-3-kinase/AKT anti-apoptotic pathway; and nuclear PTEN affects the cell cycle by also negatively regulating the MAPK pathway via cyclin D. Decreased PTEN expression is correlated with prostate cancer progression. Over-expression of AR and upregulation of AR transcriptional activity are often observed in the later stages of prostate cancer. Recent studies indicate that PTEN regulates AR activity and stability. However, the mechanism of how AR regulates PTEN has never been studied. Furthermore, resveratrol, a phytoalexin enriched in red grapes, strawberries and peanuts, has been shown to inhibit AR transcriptional activity in prostate cancer cells. In this study, we use prostate cancer cell lines to test the hypothesis that resveratrol inhibits cellular proliferation in both AR-dependent and -independent mechanisms. We show that resveratrol inhibits AR transcriptional activity in both androgen-dependent and -independent prostate cancer cells. Additionally, resveratrol stimulates PTEN expression through AR inhibition. In contrast, resveratrol directly binds epidermal growth factor receptor (EGFR) rapidly inhibiting EGFR phosphorylation, resulting in decreased AKT phosphorylation, in an AR-independent manner. Thus, resveratrol may act as potential adjunctive treatment for late-stage hormone refractory prostate cancer. More importantly, for the first time, our study demonstrates the mechanism by which AR regulates PTEN expression at the transcription level, indicating the direct link between a nuclear receptor and the PI3K/AKT pathway.

Figure 1.

Resveratrol inhibits prostate cancer cell proliferation. (A) Effect of resveratrol on LNCaP (left panel) and C4-2 (right panel) cell proliferation. Fifteen thousand cells were plated in each of 24-well plates in complete medium (FBS) and treated with DMSO (control) or different concentrations of resveratrol. Growth rates of the cells were assessed by MTT assay over a period of 5 days. Each growth-data point represents a mean value of three experiments and the error bars indicate the standard deviation, unless otherwise indicated. A Jonckheere–Terpstra trend test was performed to evaluate alternatives of ordered class differences by dose concentrations (P < 0.001). All statistical tests were two sided. Both cell lines show growth inhibition in response to increasing resveratrol doses. (B) Effect of Formononetin (P < 0.01) and (C) Biochanin A (P < 0.001) on LNCaP and C4-2 cell growth rates as assessed by the MTT assay. Molecular structures of the three phytoestrogens are shown as inserts on the right of the growth curves. (D) Flow cytometric analysis of the effects of resveratrol on proliferation and apoptosis (sub-G1). LNCaP and C4-2 cells were treated with 25 µm resveratrol for 48 h, ethanol-fixed, stained with propidium iodide and subjected to flow cytometry. The fraction of cells in the different phases of the cell cycle and those undergoing apoptosis was measured as described.

Figure 2.

Resveratrol inhibits AR transcriptional activity. (A) LNCaP and C4-2 cells were co-transfected with plasmids expressing hARE-Luc and Renilla-Luc and treated with DMSO (control) and different concentrations of resveratrol for 48 h before lysis of cells for luciferase assay. Data represent mean luciferase firefly luminescence + SD (n = 3) normalized to Renilla-luminescence. Right panel shows western blots demonstrating increasing concentrations of resveratrol associated with decreasing PSA expression in LNCaP cells. Parenthetically, at the 1 µm concentration, resveratrol may have slightly increased AR transcriptional activity and PSA expression in LNCaP cells. (B) LNCaP and (C) C4-2 cells were transfected with plasmids expressing hARE-Luc and Renilla-Luc and treated as indicated. AR activity was then measured by luciferase assay (*P < 0.005). (D) LNCaP cells were treated with DMSO (control) or different concentrations of resveratrol. Growth rate of the cells was assessed by the MTT assay over a period of 5 days. Note that resveratrol and Casodex induced cell growth inhibition and the two-drug combination had the same effect as resveratrol by itself (*P < 0.01).

Figure 3.

Resveratrol-induced PTEN promoter activity is mediated by AR. (A) C4-2 cells were co-transfected with plasmids expressing PTEN-Luc/Renilla-Luc and treated as indicated for 48 h. PTEN promoter activity was measured by dual luciferase assay. AR-ligand DHT inhibited PTEN promoter activity in a dose-dependent manner and resveratrol stimulated PTEN promoter activity. In contrast, AR-antagonist Casodex was shown to stimulate PTEN promoter activity, with resveratrol only adding slightly to this effect (*P < 0.005). (B) Knockdown of AR results in increased PTEN promoter activity. C4-2 cells were co-transfected with control siRNA or anti-AR siRNA and PTEN-luc/Renilla-Luc plasmids. After 48 h of treatment, PTEN promoter activity was measured by luciferase assay. The insert shows the knockdown of AR protein expression (*P < 0.005). (C) DU145 and C4-2 cells were transfected with PTEN-Luc/Renilla-Luc plasmids. Cells were treated with DMSO, or 10 µm resveratrol for 48 h. PTEN promoter activity was measured by luciferase assay. The structures of the −1134 to −1 and truncated −1134 to −1001 PTEN promoters are shown on the left panel. (D) AR-negative DU145 and AR-positive CWR22rv1 cells were treated with DMSO, 10, 20 or 50 µm resveratrol for 48 h. Western blots show PTEN, phospho-AKT, total-AKT and Actin levels. The quantification of phospho-AKT is shown in Supplementary Material, Fig. S3. Note that resveratrol did not change PTEN levels in DU145 cells.

Figure 4.

Resveratrol inhibits cell proliferation in AR-negative prostate caner cells. (A) Resveratrol inhibits AR-negative DU145 cell proliferation. DU145 cells were treated with DMSO (control) or increasing concentrations of resveratrol. Growth rate of the cells was assessed by MTT assay over a period of 1–5 days. (B) DU145 cells were treated with DMSO, and 10, 20 or 50 µm resveratrol. Western blots show resveratrol decreased phosphorylation of EGFR-Tyr1068 (upper panel), HER2-Tyr1221/1222 (second panel). (C) LNCaP and C4-2 cells were treated with DMSO, and 1, 10, 20 or 50 µm resveratrol. Resveratrol decreased EGFR (Tyr1068), HER2 (Tyr1221/1222) and AKT phosphorylation. The quantifications of protein phosphorylation are shown in Supplementary Material, Fig. S4.

Figure 5.

Resveratrol inhibits EGFR/HER2 signaling pathway. (A) Time course of resveratrol treatment of CWR22rv1 cells. Cells were treated with 10 µm resveratrol and cell lysates were harvested at each time point indicated. EGFR, HER2 and AKT phosphorylation are significantly decreased after the treatment. Resveratrol also stimulates PTEN protein expression. (B) C4-2 cells were cultured with DMSO, and 1, 10, 20 or 50 µm resveratrol for 48 h. After 3 min of DMSO or EGF (1 ng/ml) treatment, cells were harvested and subjected to western blot. EGF abrogated resveratrol's effect on EGFR, HER2 and AKT phosphorylation. (C) Emission fluorescence of resveratrol (5 µm) in PBS in the context of water as control, 1 µm BSA, 1 µm EGFR or 1 µm EGFR (672-1210) at pH 7.0. Excitation wavelength is 334 nm and resveratrol's emission fluorescence peaked at 392 nm. (D) Schematic model depicting the mechanism by which resveratrol inhibits proliferation in both AR-dependent and -independent mechanisms. Resveratrol inhibits AR transcriptional activity and PSA expression. In contrast, it inhibits phosphorylation of EGFR and HER2, rapidly suppressing downstream AKT pathways. Resveratrol lifts AR repression of PTEN promoter resulting in PTEN transcription and a later-effect suppression of AKT signaling. The quantifications of protein phosphorylation are shown in Supplementary Material, Fig. S5.