Selectively Targeting the Dimerization Interface of Human Androgen Receptor with Small-Molecules to Treat Castration-Resistant Prostate Cancer
This week we profile a recent publication in Cancer Letters from Drs. Kush Dalal (right),
Paul Rennie (left) and Artem Cherkasov at the Vancouver Prostate Centre.
Can you provide a brief overview of your lab’s current research focus?
The Rennie and Cherkasov labs have a long history of developing therapeutics to target the function of transcription factors that drive advanced and metastatic forms of endocrine cancers. Currently, our main focus is to use Computer-Aided Drug Design (CADD) to create small-molecule inhibitors of the androgen receptor (AR), a hormone inducible transcription factor that is responsible for the disease progression of prostate cancer (PCa). Whereas most conventional drugs used for PCa interfere with hormone recruitment to the AR, mutations in the binding site for the inhibitors eventually render them ineffective, prompting our lab to explore alternative surface exposed sites on the transcription factor including its DNA binding and trans-activation protein domains. By modeling prospective inhibitors into these mostly unexplored regions, we have developed new compounds that can effectively inhibit the transcriptional activity of mutated AR or truncated splice variant isoforms, both which are associated with the development of drug resistance in PCa patients.
What is the significance of the findings in this publication?
Following AR activation by testosterone, the receptor undergoes nuclear translocation, dimerization and binding of the functional dimeric assembly to DNA, upon which co-factors are recruited to drive expression of genes that promote tumor growth and survival. In this study, we analyzed the crystal structure of the AR DNA binding domain (DBD) and used CADD approaches to virtually screen inhibitors against the dimerization interface found within this domain. Promising “anti-dimer” compounds were purchased and experimentally tested, a small subset of which could inhibit the transcriptional activity of the AR as measured by cell-based reporter assays. Importantly, the lead compound was selective for the AR, and did not affect the transcriptional activity of conserved family members such as the glucocorticoid and estrogen receptor. This is significant because sequence conservation between the DBDs of related family members has frustrated prior attempts to selectively target the AR in this way, but is now enabled by advances in CADD algorithms to distinguish subtle differences in similar protein surfaces. The lead compound also blocked dimerization of the AR, and the splice isoform ARV7, in the nucleus of cultured PCa cells, validating the proposed mechanism of action. Finally, we used biochemistry approaches to analyze the ternary complex between the purified AR DBD, oligonucleotide and lead compound to confirm the proposed binding site within the dimerization interface using site-directed mutagenesis. Overall, this study offers the first proof-of-concept for selectively targeting the dimerization function of nuclear receptors as a potential therapy for endocrine cancers.
What are the next steps for this research?
A serious challenge in drug-discovery is to develop inhibitors that have sufficient metabolic stability to accumulate to therapeutic levels in tumors. A major focus for the future will therefore be to improve the potency and pharmacokinetic profile of the lead compound, which displayed poor half-life when incubated with liver microsomes. The appropriate medicinal chemistry efforts are underway to address this problem using iterative CADD approaches to modify the compound-structure and test for improved parameters. In addition, given that a ternary complex is possible in vitro, structural biology approaches such as X-ray crystallography and cryo-EM are planned to directly visualize the lead compound in the dimerization interface. If this plan is successful, an improved version of the compound will enter pre-clinical investigation in mouse models of prostate cancer to determine efficacy against tumor volume and AR transcriptional activity in vivo.
This research was funded by:
This study was supported by grants from CIHR, DoD (US) and TFRI funding agencies.