by Xiaotun Zhang, Colm Morrissey, Shihua Sun, Melanie Ketchandji, Peter S. Nelson, Lawrence D. True, Funda Vakar-Lopez, Robert L. Vessella, Stephen R. Plymate
Although androgens are depleted in castration resistant prostate cancer (CRPC), metastases still express nuclear androgen receptor (AR) and androgen regulated genes. We recently reported that C-terminal truncated constitutively active AR splice variants contribute to CRPC development. Since specific antibodies detecting all C-terminal truncated AR variants are not available, our aim was to develop an approach to assess the prevalence and function of AR variants in prostate cancer (PCa). Methodology/Principal Findings
Using 2 antibodies against different regions of AR protein (N- or C-terminus), we successfully showed the existence of AR variant in the LuCaP 86.2 xenograft. To evaluate the prevalence of AR variants in human PCa tissue, we used this method on tissue microarrays including 50 primary PCa and 162 metastatic CRPC tissues. RT-PCR was used to confirm AR variants. We observed a significant decrease in nuclear C-terminal AR staining in CRPC but no difference between N- and C-terminal AR nuclear staining in primary PCa. The expression of the AR regulated proteins PSA and PSMA were marginally affected by the decrease in C-terminal staining in CRPC samples. These data suggest that there is an increase in the prevalence of AR variants in CRPC based on our ability to differentiate nuclear AR expression using N- and C-terminal AR antibodies. These findings were validated using RT-PCR. Importantly, the loss of C-terminal immunoreactivity and the identification of AR variants were different depending on the site of metastasis in the same patient. Conclusions
We successfully developed a novel immunohistochemical approach which was used to ascertain the prevalence of AR variants in a large number of primary PCa and metastatic CRPC. Our results showed a snapshot of overall high frequency of C-terminal truncated AR splice variants and site specific AR loss in CRPC, which could have utility in stratifying patients for AR targeted therapeutics.