Investigating the molecular determinants of PARG inhibitor sensitivity

Abstract

The treatment of cancer is moving towards personalised medicine, exploiting therapeutic strategies that selectively target cancer cells based their underlying genetic characteristics. One such strategy is synthetic lethality which describes a combination of genes which are lethal when disrupted together. Based on this conceptual framework, inhibitors of poly(ADP-ribose) polymerase (PARP) are having a major impact in the clinic treating women with BRCA1/2- mutant ovarian cancer. However, because only ~20% of ovarian cancers harbour mutations in BRCA1/2, there is a pressing need to develop other novel treatment strategies. Recently, a first-in-class inhibitor of poly(ADP-ribose) glycohydrolase (PARG), the enzyme that counterbalances PARP activity, has been developed. However, it remains to be established which cancer genotypes are sensitive to PARG inhibitors (PARGi), and whether PARGi could offer a distinct therapeutic strategy to benefit this remaining subset of patients. To address these questions, this thesis explores the PARGi in the context of ovarian cancer. To determine the effect of PARG inhibition, previous work screened a panel of ovarian cancer cell lines and identified a subset intrinsically sensitive to PARGi. It follows, this work set out to address what the unique genetic determinants in this subset of PARGi-sensitive cell lines may be. To define the underlying vulnerability causing PARGi sensitivity, a synthetic lethal siRNA screen was carried out in a resistant cell line yielding several genes involved in DNA replication including TIMELESS, HUS1, RFC2, and CHK1. In turn, expression profiling of DNA replication genes allowed the identification of additional cell lines intrinsically sensitive to PARGi, and to sensitise otherwise resistant models using a CHK1 inhibitor. Characterisation of PARGisensitive models found sensitivity is dependent on entry into S-phase and is accompanied by replication fork asymmetry, RPA accumulation and activation of DNA damage signalling. PARGi-sensitive cells also display a pre-mitotic cell cycle block and distinctive pan-nuclear γH2AX staining, indicating replication catastrophe. Furthermore, these phenotypes are reproduced in PARGi-sensitive models by siRNA knockdown of RECQ1, a helicase involved in replication fork restart. These observations support a model whereby cells with defects in DNA replication are dependent on PARG activity for replication fork progression. This could be based on PARG’s role in counterbalancing PARP-mediated inhibition of RECQ1, enabling the restart of replication forks following stalling. Consistently, cells display differential sensitivity to PARG versus PARP inhibition, likely reflecting the antagonistic nature of these enzymes at the replication fork. Indeed, PARP inhibition protects cells from the anti-proliferative effects of PARGi. Taken together, these results demonstrate PARG inhibitors have therapeutic potential to complement PARP inhibitors in the treatment of ovarian cancer and has implications on the design of early phase trials testing clinical candidates when they become available.

Date of Award	1 Aug 2020
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Stephen Taylor (Supervisor) & Martin Lowe (Supervisor)