Research Papers:
Dual effects of constitutively active androgen receptor and full-length androgen receptor for N-cadherin regulation in prostate cancer
PDF | HTML | Supplementary Files | How to cite
Metrics: PDF 2425 views | HTML 3197 views | ?
Abstract
Félicie Cottard1, Pauline Ould Madi-Berthélémy1, Eva Erdmann1, Frédérique Schaff-Wendling1,3, Céline Keime2, Tao Ye2, Jean-Emmanuel Kurtz1,3 and Jocelyn Céraline1,3
1Université de Strasbourg, INSERM, FMTS, Strasbourg, France
2Université de Strasbourg, CNRS, INSERM, Institut de Génétique et de Biologie Moléculaire et Cellulaire, Illkirch-Graffenstaden, France
3Service d’Onco-Hématologie, Hôpitaux Universitaires de Strasbourg, Strasbourg, France
Correspondence to:
Félicie Cottard, email: [email protected]
Keywords: androgen receptor, constitutively active androgen receptor variants, N-cadherin, EMT, prostate cancer
Received: August 29, 2016 Accepted: May 12, 2017 Published: May 29, 2017
ABSTRACT
Constitutively active androgen receptor (AR) variants have been involved in the expression of mesenchymal markers such as N-cadherin in prostate cancer (PCa). However, the underlying molecular mechanisms remain elusive. It remains unclear, whether N-cadherin gene (CDH2) is a direct transcriptional target of AR variants or whether the observed upregulation is due to indirect effects through additional regulatory factors. Moreover, the specific contribution of full-length AR and AR variants in N-cadherin regulation in PCa has never been explored deeply. To investigate this, we artificially mimicked the co-expression of AR variants together with a full-length AR and performed miRNA-seq, RNA-seq and ChIP assays. Our results were in favor of a direct AR variants action on CDH2. Our data also revealed a distinctive mode of action between full-length AR and AR variants to regulate N-cadherin expression. Both wild type AR and AR variants could interact with a regulatory element in intron 1 of CDH2. However, a higher histone H4 acetylation in this genomic region was only observed with AR variants. This suggests that full-length AR may play an occluding function to impede CDH2 upregulation. Our data further highlighted a negative effect of AR variants on the expression of the endogenous full-length AR in LNCaP. These differences in the mode of action of AR variants and full-length AR for the control of one key gene for prostate cancer progression could be worth considering for targeting AR variants in PCa.

PII: 18270