Androgen-dependent Regulation of PARP7 Protein Stability and PARP7-mediated ADP-ribosylation of the Androgen Receptor in Prostate Cancer

Among US men, prostate cancer (PCa) is the most common cancer diagnosed and is the second leading cause of cancer deaths, representing a substantial healthcare burden. A key biological feature of PCa is that androgen receptor (AR) signaling is the driver pathway for this disease. AR is a nuclear receptor that binds to androgen and functions as a ligand-activated transcription factor. Androgen binding to AR triggers a conformational change that leads to import into the nucleus where AR regulates gene expression. Because PCa growth is highly reliant on AR signaling, targeting the pathway by androgen deprivation therapy (ADT) is a major therapeutic strategy for treating PCa. The two main approaches for ADT are inhibition of androgen synthesis and AR signaling blockade using receptor antagonists. ADT is highly effective, and regression of PCa is seen in most patients. However, resistance to ADT almost invariably develops, and patients are said to have castration-resistant PCa (CRPC). The prognosis for CRPC is poor, and no therapeutics exist that are curative for this disease. Thus, there is a continuing need for better understanding of AR signaling and how it is regulated in order to uncover new drug targets to treat CRPC.

Our lab recently discovered that AR is ADP-ribosylated by poly(ADP-ribose) polymerase 7 (PARP7) in response to androgen. PARP7 is one of seventeen PARP genes expressed in humans. Using NAD+ as the donor molecule, PARPs catalyze ADP-ribosylation where ADP-ribose is transferred onto acceptor amino acid in substrates. ADP-ribosylation is a major type of post-translational modification involved in many biological processes including DNA damage repair, transcription, immune signaling, protein degradation, and cellular stress response. Previous studies have shown that PARP7 plays a regulatory role in anti-viral response and in liver X receptor, aryl hydrocarbon receptor, and estrogen receptor signaling pathways. In the context of AR signaling, PARP7 also plays a transcription regulatory role. PARP7 is induced by androgen, and accumulation of PARP7 leads to ADP-ribosylation of AR. Subsequently, PARP9/DTX3L, a transcription regulatory complex is recruited to ADP-ribosylated AR, resulting in changes in AR-dependent gene transcription. PARP7 gene expression is altered in multiple types of cancers, and our analysis of PARP7 in existing clinical datasets for PCa showed that low PARP7 gene expression is associated with metastatic tissue samples and earlier disease relapse. Therefore, PARP7-mediated ADP-ribosylation of AR is a novel signaling axis that controls AR activity with potential clinical implications in PCa patients.

In this dissertation, we explored PARP7/AR signaling axis further by investigating AR regulation of PARP7 as well as how PARP7 recognizes AR for ADP-ribosylation. In Chapter 1, the background for AR signaling in PCa, ADP-ribosylation, and PARPs (including PARP7), along with a brief overview of the PARP7/AR signaling axis is presented. In Chapter 2, we found that PARP7 is a rapidly degraded protein that can be stabilized by AR signaling, resulting in PARP7 accumulation in PCa cells. Thus, PARP7 protein level regulation by AR occurs both by induction of PARP7 mRNA and PARP7 protein stabilization. In Chapter 3, we delved deeper into how PARP7 ADP-ribosylates AR, defining features from both substrate and enzyme standpoints that are required for the reaction to take place. We found that AR in an agonist conformation is recognized by PARP7 as substrate. Using structure-function approaches, we found that the PARP7 zinc finger domain is required for AR ADP-ribosylation, while the PARP7 tryptophan-tryptophan-glutamate (WWE) domain appears to make a smaller contribution to this process. The almost complete loss of AR ADP-ribosylation displayed by the PARP7 zinc finger mutants was not explained by loss of enzyme activity, loss of binding to AR, or disruption of nuclear localization of PARP7. These results suggested that the PARP7 zinc finger may be playing a scaffolding role in the PARP7/AR complex to bring together the ADP-ribosylation acceptor sites in AR and the PARP7 active site. Over the course of our investigation, we discovered that the zinc finger domain is required for formation of PARP7 nuclear foci, raising the possibility that foci formation is linked to AR ADP-ribosylation. In Chapter 4, we discuss some of the broader implications of our findings and propose approaches for future investigation into the following: 1) PARP7 turnover and stabilization mechanisms, 2) how PARP9/DTX3L assembles on ADP-ribosylated AR (i.e. configuration of the complex), and 3) biological role of PARP7 in PCa.

Overall, the work presented in this dissertation contributes to our understanding of the PARP7/AR signaling axis by establishing a second layer of AR regulation of PARP7 protein level and highlighting the key role that the PARP7 zinc finger plays in recognizing the agonist conformation of AR as substrate.