Genome-wide impact of androgen receptor trapped clone-27 loss on androgen-regulated transcription in prostate cancer cells

Abstract

The androgen receptor (AR) directs diverse biological processes through interaction with coregulators such as AR trapped clone-27 (ART-27). Our results show that ART-27 is recruited to AR-binding sites by chromatin immunoprecipitation analysis. In addition, the effect of ART-27 on genome-wide transcription was examined. The studies indicate that loss of ART-27 enhances expression of many androgen-regulated genes, suggesting that ART-27 inhibits gene expression. Surprisingly, classes of genes that are up-regulated upon ART-27 depletion include regulators of DNA damage checkpoint and cell cycle progression, suggesting that ART-27 functions to keep expression levels of these genes low. Consistent with this idea, stable reduction of ART-27 by short-hairpin RNA enhances LNCaP cell proliferation compared with control cells. The effect of ART-27 loss was also examined in response to the antiandrogen bicalutamide. Unexpectedly, cells treated with ART-27 siRNA no longer exhibited gene repression in response to bicalutamide. To examine ART-27 loss in prostate cancer progression, immunohistochemistry was conducted on a tissue array containing samples from primary tumors of individuals who were clinically followed and later shown to have either recurrent or nonrecurrent disease. Comparison of ART-27 and AR staining indicated that nuclear ART-27 expression was lost in the majority of AR-positive recurrent prostate cancers. Our studies show that reduction of ART-27 protein levels in prostate cancer may facilitate antiandrogen-resistant disease.

Figure 1. Effect of ART-27 depletion on expression of androgen-regulated genes

(A) Outline of the procedure used to examine androgen-response in control and ART-27 depleted LNCaP cells. Briefly, LNCaP cells were steroid-deprived for 72 h. During this period, the cells were transfected with control siRNA (siControl) or ART-27 siRNA (siART-27), and allowed to recover in steroid-deprived media. The cells were then stimulated with ethanol vehicle (Veh) or R1881 for 18 h. Total RNA was isolated and analyzed by real-time, quantitative PCR (Q-PCR). For genome-wide studies, total RNA was hybridized to Affymetrix Gene Chips and processed. (B) AR and ART-27 protein expression in control and ART-27-depleted cells. Western blot showing AR and ART-27 protein levels in whole extracts obtained from siRNA transfectants treated with ethanol or 10 nM R1881 for 16 h. Anti-ERK antibody was used as a loading control. (C) Q-PCR results showing the relative levels of PSA, NKX3-1 and ART-27 mRNA in LNCaP cells treated as described in Fig. 1A with ethanol vehicle (veh) - also shown in insert, or the indicated concentrations of R1881 (D) Microarray heat map comparing the expression of transcripts encoding the 9 DNA-integrity and cell cycle checkpoint regulators in steroid-deprived control (siART-27-) and ART-27-depleted (siART-27+) LNCaP cells.

Figure 2. ART-27 inhibits expression of androgen-regulated genes

(A) LNCaP cells transfected with control siRNA (siControl) or ART-27 siRNA (siART-27) were steroid-deprived for 72 h. Relative mRNA levels of the indicated androgen-induced, androgen-repressed and androgen-insensitive genes were determined by Q-PCR (Gene descriptions are provided in supplementary Fig. S-1) (B) Expression of the indicated transcripts was examined by Q-PCR in LNCaP cells treated as described in Fig. 1A with ethanol vehicle (Veh) or 10 nM R1881. (C) Expression of checkpoint proteins in whole cell extracts obtained from LNCaP cells treated as described in Fig. 1A was examined by Western blot.

Figure 3. AR mediates the increase in gene expression observed upon ART-27 depletion

(A) LNCaP cells were transfected with control siRNA or ART-27 siRNA (siART-27) and/or AR siRNA (siAR) as indicated and steroid-deprived for 72 h. Total RNA was isolated and analyzed by Q-PCR for the indicated genes. The mRNA levels are shown relative to the naive treatment, which were arbitrarily set to 1 for each gene. Data shown is the average of 3 independent experiments. (B) LNCaP cells were transfected with control siRNA (-), siART-27 and/ or siAR as indicated and steroid-deprived for 72 h. Whole cell extracts were analyzed by western blot using anti-AR, and anti-ART-27 antibodies. Anti-ERK antibody was used as a loading control.

Figure 4. Recruitment of ART-27 to AR-target genes

(A) Schematic illustration showing the positions of the well-characterized AREs of PSA and NKX3-1, the putative AREs of ATR and GTSE1, and the negative control upstream region (UPS), relative to transcription-start sites (+1 bp). (B) LNCaP cells were steroid-starved for 72 h and stimulated with ethanol vehicle (Veh) or 0.1 μM R1881 for 17 h. ChIP assay was performed using anti-AR or anti-ART-27 antibodies, or pre-immune sera. Recruitment of AR and ART-27 to the indicated regions is shown as a percentage of the input. Data shown is the average of three independent experiments and error bars represent standard errors from the mean.

Figure 5. ART-27 loss facilitates resistance to the AR-antagonist bicalutamide

(A) ART-27 depletion inhibits gene repression by bicalutamide. Q-PCR analysis showing relative mRNA levels of the indicated genes in LNCaP cells transfected with control siRNA (siControl) or ART-27 siRNA (siART-27) and steroid-deprived in the presence of ethanol vehicle (Veh) or 0.1 μM bicalutamide (BIC) for 72 h. The mRNA levels were normalized to ethanol-treated samples, which were arbitrarily set to 1. (B) Bicalutamide-resistant cell proliferation in ART-27-depleted cells. LNCaP cells transfected in duplicate with control or ART-27 siRNA (siART-27) were steroid-deprived in the presence of ethanol vehicle or 0.1 μM BIC for 72 h. Cell proliferation rates were then estimated by [³H]-thymidine incorporation assay and shown as average counts per minute (CPM). Errors bars indicate standard errors from the mean (*p = 0.006).

Figure 6. ART-27 expression and clinical outcome in AR-positive Prostate cancer

(A) ART-27 exhibits differential sub-cellular expression profiles in AR-positive prostate cancer. Prostate cancer tissue microarray containing 58 AR-positive cases (9 high-grade prostatic intraepithelial neoplasia (PIN); 25 non-recurrent (NR) i.e. did not relapse after radical prostatectomy; and 24 recurrent (REC) i.e. relapsed after radical prostatectomy) were stained by immunohistochemistry using anti-AR (top row) and anti-ART-27 (bottom row) antibodies. Representative samples from each group are shown in columns. Red arrows indicate cells showing nuclear ART-27 expression. Black arrows indicate cells showing largely cytoplasmic or peri-nuclear ART-27 expression. Magnification: 40× (B) Summary of prostate cancer tissue microarray analysis. Nuclear ART-27 expression status is shown as a percentage of the total number of cases in each group. The p-values for the indicated comparisons are also indicated (*p = 0.058; **p = 0.054). (C) Average PSA-doubling times of nuclear ART-27-positive and -negative REC cases (*p = 0.035).