Regulation of androgen-responsive transcription by the chromatin remodeling factor CHD8

Abstract

The androgen receptor (AR) mediates the effect of androgens through its transcriptional function during both normal prostate development and in the emergence and progression of prostate cancer. AR is known to assemble coactivator complexes at target promoters to facilitate transcriptional activation in response to androgens. Here we identify the ATP-dependent chromatin remodeling factor chromodomain helicase DNA-binding protein 8 (CHD8) as a novel coregulator of androgen-responsive transcription. We demonstrate that CHD8 directly associates with AR and that CHD8 and AR simultaneously localize to the TMPRSS2 enhancer after androgen treatment. In the LNCaP cell line, reduction of CHD8 levels by small interfering RNA treatment severely diminishes androgen-dependent activation of the TMPRSS2 gene. We demonstrate that the recruitment of AR to the TMPRSS2 promoter in response to androgen treatment requires CHD8. Finally, CHD8 facilitates androgen-stimulated proliferation of LNCaP cells, emphasizing the physiological importance of CHD8. Taken together, we present evidence of a functional role for CHD8 in AR-mediated transcriptional regulation of target genes.

Figure 1

CHD8 interacts with AR. A, Nuclear extracts were prepared from the indicated cell lines and immunoprecipitated with CHD8 antibodies. After washing, the input and immunoprecipitated samples (IP) were subjected to Western blot analysis using the indicated antibodies. B, Cellular extracts were prepared from SF9 cells after coinfection with the indicated viruses. Immunoprecipitations were performed with anti-Flag antibody-linked M2 agarose beads. Immunoprecipitated samples were subjected to Western blot analysis using the indicated antibodies.

Figure 2

CHD8 and AR colocalize to the TMPRSS2 ARE. A, LNCaP cells were treated with ethanol (EtOH) or 10 nm DHT for 6 h. Chromatin was cross-linked in vivo with formaldehyde. Cells were lysed, and ChIP was performed with the indicated antibodies (Ab). Bound DNA was detected by quantitative PCR using primers to the ARE of TMPRSS2 (ARE) or a control TMPRSS2 promoter region (−7 kb). Control IgG-precipitated samples were less than 0.005% of input and therefore are not shown. Data are representative of multiple experiments. B, LNCaP cells were treated as in A. Re-ChIP experiments were performed by successively immunoprecipitating the cross-linked chromatin with the indicated antibodies. Bound DNA was detected by quantitative PCR using primers to the TMPRSS2 ARE. Data are representative of multiple experiments.

Figure 3

CHD8 activates AR-mediated transcription of the TMPRSS2 gene. LNCaP cells were transfected with the indicated siRNA constructs. After selection of the transfected cells, cultures were treated with ethanol (EtOH) or 10 nm DHT for 6 h. A and B, Total RNA was isolated, and TMPRSS2 expression (A) or PSA expression (B) was analyzed by quantitative RT-PCR. *, P < 0.05; **, P < 0.01 by Student's t test. C, Efficiency of CHD8 knockdown was determined by Western blot analysis with CHD8 antibodies. Actin is blotted as a loading control.

Figure 4

CHD8 activates TMPRSS2 in an androgen-dependent context. The indicated androgen-independent cell lines were transfected with the specified siRNA constructs. After selection of the transfected cells, cultures were treated with ethanol (EtOH) or 10 nm DHT for 6 h. Total RNA was isolated, and TMPRSS2 expression was analyzed by quantitative RT-PCR.

Figure 5

CHD8 is required for optimal androgen-responsive binding of AR to the TMPRSS2 ARE. LNCaP cells were transfected with the indicated siRNA constructs. After selection of the transfected cells, cultures were treated with ethanol or 10 nm DHT for 6 h. Chromatin was cross-linked in vivo with formaldehyde. Cells were lysed, and ChIP was performed with the indicated antibodies. Bound DNA was detected by quantitative PCR using primers to the TMPRSS2 ARE. Control IgG-precipitated samples were less than 0.005% of input and therefore are not shown. Shown is a typical result from multiple experiments.

Figure 6

CHD8 regulates androgen-dependent cell proliferation. LNCaP cells were transfected with the indicated siRNA constructs. After selection of the transfected cells, cultures were treated with ethanol (EtOH) or 4 nm R1881. At the indicated time points, proliferation was determined using a luminescent-based assay of metabolically active cells. A, Data are expressed as relative proliferation, which is calculated as the luminescent signal for each condition normalized to the uninduced sample at d 1. B, Data are expressed as fold R1881-induced proliferation, which is calculated as the ratio of the luminescent signals from the induced to the uninduced cells for each siRNA treatment at each indicated time point. *, P < 0.02 by Student's t test.