Analysis of the Androgen Receptor Transcriptional complexes in Prostate Cancer

ROGGERO, CARLOS M.; RIZO, JOSEP

Kyoto, Japon

Congreso; ICMRBS XXVIIth International Conference on Magnetic Resonance in Biological Systems; 2016

Prostate cancer (PCa) is the most common cancer affecting men in the developed world. Signalling through the androgen receptor (AR) is critical for survival and proliferation of prostate cancer cells. The AR is a transcription factor that drives the differentiation of prostate epithelium by regulating the expression of several hundred genes. In fact, androgen deprivation therapy remains the most widely used treatment for patients with advanced prostate cancer. However, although androgen deprivation initially results in prostate tumor regression, patients develop resistance to androgen depletion, leading to the invariably fatal castration-resistant PCa (CRPC). The mechanisms responsible for the transition from PCa to CRPC are not well understood and include AR amplification, AR co-regulator protein alterations, AR promiscuity and AR splice variant or mutations. Therefore there is an urgent need to find alternative therapies that do not depend solely on hormone deprivation. There is strong evidence that a variety of epigenetic alterations, including histone modifications, play a role in the development of CRPC. AR function is regulated by diverse epigenetic enzymes. Among them, histone lysine demethylases (KDMs) including lysine specific demethylase 1 (LSD1) and Jumonji C-domain containing demethylases (JMJDs) play a pivotal role. KDMs are overexpressed in many human tumors including PCa and are found in AR transcriptional complexes as co-activators. The molecular and structural details of how KDMs interact with AR are not well understood. In this work we perform a systematic analysis to map the interactions between AR and several KDMs. To address this, we took advantage of the high sensitivity of TROSY-HSQC NMR spectroscopy. Other approaches including pull-down, crosslinking and gel filtration were used in addition to NMR. We first generated constructs of the different domains of AR and KDMs and expressed the 15N labelled proteins. Using TROSY-HSQC, we found that the N-terminal catalytic domains of JMJD2A do not bind to the ligand binding domain (LBD) of AR or a fragment spanning the LDB and the DNA binding domain (DBD) of AR, in contrast to previous findings by other groups. Surprisingly, a weak interaction was observed between the catalytic domains of JMJD2s (isoforms A,B,C) and the N-terminal domain (NTD) or the DBD of AR. More importantly, a fragment comprising the NTD and the DBD of AR showed a strong interaction to the catalytic domains of JMJD2A and C, but not JMJD2B. In addition, we found that the central domain of JMJD2A interact with AR-LBD, but the C-terminal domain of JMJD2A (PHD+Tudor domains) do not bind. These results show that the AR NTD and DBD domains have a cooperative effect in binding to JMJD2s catalytic domains, with affinities varying in different JMJD2 isoforms. The AR-NTD-DBD/JMJD2s interactions may be crucial to recruit JMJD2s to histone methylated lysines to remove the repressive marks and activate AR-dependent gene transcription. Designing small compounds that disrupt the AR/KDMs interactions may have a promising potential therapeutic use.