Androgens suppress EZH2 expression via retinoblastoma (RB) and p130-dependent pathways: a potential mechanism of androgen-refractory progression of prostate cancer

Abstract

Androgens and the androgen receptor are important for both normal prostate development and progression of prostate cancer (PCa). However, the underlying mechanisms are not fully understood. The Polycomb protein enhancer of zeste homolog 2 (EZH2) functions as an epigenetic gene silencer and plays a role in oncogenesis by promoting cell proliferation and invasion. EZH2 has been implicated in human PCa progression, because its expression is often elevated in hormone-refractory PCa. Here, we demonstrated that expression of EZH2 is lower in androgen-sensitive LNCaP PCa cells compared with Rf and C4-2 cells, two androgen-refractory sublines that are derived from LNCaP cells. Androgen ablation by castration increased the level of EZH2 proteins in LNCaP xenografts in mice. In contrast, treatment of LNCaP cells in culture with the synthetic androgen methyltrieolone (R1881) at doses of 1 nm or higher suppressed EZH2 expression. Moreover, our data suggest that androgen repression of EZH2 requires a functional androgen receptor and this effect is mediated through the retinoblastoma protein and its related protein p130. We further showed that androgen treatment not only increases expression of EZH2 target genes DAB2IP and E-cadherin but also affects LNCaP cell migration. Our results reveal that androgens function as an epigenetic regulator in prostatic cells by repression of EZH2 expression through the retinoblastoma protein and p130-dependent pathways. Our findings also suggest that blockade of EZH2 derepression during androgen deprivation therapy may represent an effective tactic for the treatment of androgen-refractory PCa.

Figure 1

EZH2 is overexpressed in PCa cells depleted of androgens. A, Cells were cultured in media as described in Materials and Methods. At the indicated time points after subculture, protein expression levels of EZH2 and Erk2 (loading control) were analyzed by immunoblotting (IB). B, Relative EZH2 and PSA mRNA levels were determined using real-time PCR. Error bars represent the sd of three experiments. C, Mice with LNCaP xenografts were either castrated or left intact. One week after castration, tumors from intact (n = 3) and castrated (n = 3) mice were harvested. Immunohistochemistry was performed with tumor tissues using the EZH2 antibody. D, Tumor tissues were subjected to homogenization, and 50 μg of total cell lysate were analyzed by Western blotting using the EZH2, Bcl-2, and Erk2 antibodies. The average level of EZH2 protein among the tumors from either intact or castrated mice was determined by the quantified value of EZH2 in each tumor normalized to Erk2 levels.

Figure 2

The effect of androgen treatment on EZH2 expression. A, LNCaP cells were grown in CSS for 24 h and then incubated for an additional 48 or 72 h with the given concentrations of R1881. Western blottings were performed with antibodies to EZH2 and Erk2 for loading control. The level of EZH2 was quantified, normalized to Erk2 levels, and given relative to mock-treated cells for each time point. B, EZH2 and PSA mRNA levels after androgen treatment for 72 h were determined by real-time PCR. C and D, LAPC-4 (C) or C4-2 (D) cells were grown in CSS for 24 h and then incubated for an additional 48 h with the given concentrations of R1881. Western blottings were performed with antibodies to EZH2 and Erk2. The level of EZH2 was quantified as described above. E, Time-course study of LNCaP cells treated with 100 nm R1881. Protein expression of EZH2 and Erk2 was analyzed by Western blotting (upper panel). mRNA levels of EZH2 at 0 and 48 h were determined by real-time PCR (lower panel). Error bars represent the sd of three experiments. IB, Immunoblotting.

Figure 3

Role of AR in androgen regulation of EZH2. A, LNCaP cells were transfected with NS control (NS) or AR siRNA for 24 h and then treated with or without R1881 (10 nm) for an additional 48 h. EZH2, AR, and Erk2 protein expression was analyzed by Western blotting, and the level of EZH2 was quantified, normalized to Erk2 levels, and given relative to mock-treated cells. B, mRNA from LNCaP cells treated with NS or AR siRNA was analyzed by PCR to determine expression levels of OAS2 and GAPDH. IB, Immunoblotting.

Figure 4

RB and its related proteins, p130, are required for androgen regulation of EZH2. A, LNCaP cells were transfected with the empty vector pCI, wild-type (WT) E1A, or two mutants of E1A, RG2, and YH47/CG134. At 24 h after transfection, the media were changed to CSS and then R1881 or vehicle was added. After 48 h of androgen treatment, cells were harvested, and protein expression of EZH2, E1A, and Erk2 was determined by Western blotting. B, LNCaP cells were treated with the indicated siRNA pools. At 48 h after transfection, 10 nm R1881 or vehicle was added and cells were incubated for an additional 48 h. Western blottings were performed to analyze protein expression of EZH2, RB, p107, p130, and Erk2. C, upper panel, LNCaP or RB^−/− PrE cells were cultured in CSS for 24 h and then incubated for an additional 48 h with the given concentrations of R1881. Western blottings were performed with antibodies to EZH2, AR, and RB. The membrane was stained with Ponceau S for loading control. Lower panel, LNCaP cells were grown in CSS for 24 h and then incubated for an additional 72 h with the given concentrations of R1881. Western blottings were performed with antibodies to p130–S672-p and Erk2. D, DU145 cells were transfected with the given plasmids and harvested after 48 h. Protein levels for EZH2 and Erk2 were analyzed by Western blotting. IB, Immunoblotting.

Figure 5

Binding of p130 and E2F4 to the EZH2 promoter. A, Diagram of primers in the promoter of EZH2 used for ChIP assay. The black box between the primer set a and the primer set b represents four putative E2F1 binding sites. LNCaP cells were treated with 10 nm R1881 or vehicle for 48 h and then harvested. Chromatin proteins cross-linked to DNA were immunoprecipitated with p130, E2F4, or NS IgG. DNA that was pulled down was analyzed by PCR using primers that amplified the E2F binding region in the EZH2 promoter (B), p130/E2F binding site in E2F1 promoter (C), and a NS region of the EZH2 promoter (D).

Figure 6

Effect of androgens on expression and biological consequences of EZH2 repression genes. A, LNCaP cells were grown in CSS for 24 h and then incubated for an additional 72 h with or without the addition of 10 nm R1881. mRNA levels of EZH2 target genes ADRB2, DAB2IP, E-Cad, and RUNX3 were determined by real-time PCR. Inset, Protein level of EZH2 and Erk2. B, LNCaP cells were transfected with NS or DAB2IP and E-Cad siRNA. On d 1, cells were treated with or without 10 nm R1881. Cells were plated in culture inserts on d 2, which were removed on d 3 to create a gap. The area of the gaps and percent wound area filled were determined at 0 and 60 h. A representative figure from two experiments is shown. C, mRNA levels of DAB2IP and E-Cad were determined by real time at 48 and 96 h after transfection. IB, Immunoblotting.