Abstract 2936: Increased PARP1-DNA binding due to autoPARylation inhibition of PARP1 on DNA rather than PARP1-DNA trapping is correlated with PARP1 inhibitor's cytotoxicity

• Published in: Cancer research (Chicago, Ill.) Ill.), 2019-07, Vol.79 (13_Supplement), p.2936-2936
• Main Authors: Chen, Hua-Dong, Chen, Chuan-Huizi, Wang, Yu-Ting, Guo, Ne, Tian, Yu-Nan, Huan, Xia-Juan, Song, Shan-Shan, He, Jin-Xue, Miao, Ze-Hong
• Format: Article
• Language: English
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.AM2019-2936

PARP1 inhibitors (PARPis) have been successfully used in the clinic for cancer therapy. The mechanisms by which they elicit cytotoxicity in cancer cells remain to be clarified. Initially, inhibition of cellular polymerase activity of PARP1 (reflected as inhibition of cellular poly (ADP-ribose) (PAR) formation or PARylation) was proposed as their primary mechanism. However, either cell-free histone-based or cellular PARylation inhibition displays very low correlation with PARPis’ cytotoxicity or clinical anticancer effects. Subsequently, the Pommier’s group proposed the mechanism of PARP1-DNA trapping and demonstrated that this mechanism showed much higher correlation with PARPis’ cytotoxicity. Nevertheless, contradictory results have been reported since then. Here, we showed no significant correlation between PARP1-DNA trapping and cytotoxicity induced by PARPis. We complemented PARP1-knockout sublines with wild-type PARP1 and 11 mutants with different point mutations that affect the polymerase activity. When examining the PARPi talazoparib, the induced cytotoxicity was highly significantly correlated with cellular PARP1 polymerase activity, but not with its PARP1-DNA trapping or polymerase inhibition. Similarly, talazoparib’s PARP1-DNA trapping revealed significant correlation with the polymerase activity rather than its inhibition. Differently, however, when evaluating purified wild-type and mutated PARP1, we identified an almost linear relationship between PARPis’ inhibiting PARP1 dissociation from DNA and their cytotoxicity in 17 cancer cell lines. In contrast, no significant correlation existed between PARP1 polymerase inhibition in the histone-based systems and the cytotoxicity. After careful comparisons on different methods and detection targets, we conclude that the PARPi-mediated increase in PARP1-DNA binding by inhibiting autoPARylation of PARP1 on DNA rather than in PARP1-DNA trapping is correlated with PARPi’s cytotoxicity. Based on this conclusion, we established a new PARPi screening model that more closely predicts cytotoxicity (r=0.9984; p