Mechanistic Dissection of PARP1 Trapping and the Impact on in Vivo Tolerability and Efficacy of PARP Inhibitors.

Abstract Poly(ADP-ribose) polymerases (PARP1, -2, and -3) play important roles in DNA damage repair. As such, a number of PARP inhibitors are undergoing clinical development as anticancer therapies, particularly in tumors with DNA repair deficits and in combination with DNA-damaging agents. Preclinical evidence indicates that PARP inhibitors potentiate the cytotoxicity of DNA alkylating agents. It has been proposed that a major mechanism underlying this activity is the allosteric trapping of PARP1 at DNA single-strand breaks during base excision repair; however, direct evidence of allostery has not been reported. Here the data reveal that veliparib, olaparib, niraparib, and talazoparib (BMN-673) potentiate the cytotoxicity of alkylating agents. Consistent with this, all four drugs possess PARP1 trapping activity. Using biochemical and cellular approaches, we directly probe the trapping mechanism for an allosteric component. These studies indicate that trapping is due to catalytic inhibition and not allostery. The potency of PARP inhibitors with respect to trapping and catalytic inhibition is linearly correlated in biochemical systems but is nonlinear in cells. High-content imaging of γH2Ax levels suggests that this is attributable to differential potentiation of DNA damage in cells. Trapping potency is inversely correlated with tolerability when PARP inhibitors are combined with temozolomide in mouse xenograft studies. As a result, PARP inhibitors with dramatically different trapping potencies elicit comparable in vivo efficacy at maximum tolerated doses. Finally, the impact of trapping on tolerability and efficacy is likely to be context specific. Implications: Understanding the context-specific relationships of trapping and catalytic inhibition with both tolerability and efficacy will aid in determining the suitability of a PARP inhibitor for inclusion in a particular clinical regimen. Mol Cancer Res; 13(11); 1465–77. ©2015 AACR.