Combining CRISPR-Cas13 and proximity biotinylation approaches for discovery of novel androgen receptor variant splicing factors in castrate resistant prostate cancer

Abstract

The relevance of androgens to prostate cancer (PCa) has long been appreciated. The cellular effects of androgens are exerted by their interaction with the ligand-binding domain (LBD) of the androgen receptor (AR). Targeting AR signalling through suppression of circulating testosterone with androgen deprivation therapy (ADT) or direct receptor inhibition using antiandrogens is the clinical mainstay for treatment of metastatic PCa. Whilst these can elicit substantial clinical responses, a significant number of patients will relapse despite castrate levels of serum testosterone. It is now recognised that this disease state, castrate resistant prostate cancer (CRPC), continually activates AR through a number of mechanisms including alternative mRNA splicing of androgen receptor-variants (AR-Vs). AR-Vs splice cryptic exons (CEs) in place of the LBD, obviating the need for androgen binding and rendering them immune to targeting with antiandrogens. AR-Vs, in particular AR-V7 which splices cryptic exon 3 (CE3), are linked with therapeutic resistance and poor clinical outcomes. There is currently a lack of knowledge regarding factors that mediate splicing of AR-V7. Here, two contemporary biotechnologies have been employed to identify novel splicing factors. This work has combined RNA-targeting CRISPR protein Cas13 with proximity biotinylation enzyme APEX2 to biotinylate, enrich, and identify local protein interactors of CE3 mRNA in CRPC cell line CWR22Rv1. This has created a rich source of potential AR-V7 splicing regulators, of which TRA2B has shown particular promise. Subsequent validation has confirmed TRA2 proteins as regulators controlling alternative splicing decisions of the AR gene, and their clinical relevance is vindicated by analysis of patient datasets. Therefore, development of these methodologies represents a powerful proof of concept that could in theory be applied to any mRNA region of interest.