Investigation of the role of poly(ADP-ribose)polymerase inhibition in topoisomerase I poison-induced cytotoxicity

Abstract

The topoisomerase I (TopoI) poisons (e.g. camptothecin, CPT) cause stabilisation of the TopoI-DNA cleavable complex resulting in DNA single strand breaks (SSBs). DNA breakage and cytotoxicity are related to TopoI activity and TopoI poisons are thought to be primarily cytotoxic in S phase due to conversion of SSB to double strand breaks (DSB) through replication fork collision. DNA SSB are repaired by the base excision repair/SSB repair (BER/SSBR) pathway and DSB are predominantly repaired by homologous recombination repair (HRR) or non-homologous end joining (NHEJ). Poly (ADP-ribose) polymerase-1 (PARP-1) recognises SSB and DSB and promotes their repair. Previous studies in this laboratory demonstrate that PARP inhibitors enhance TopoI poison-induced cytotoxicity by a mechanism that involves elevation of DNA breaks level or retarding the repair of TopoI poison-induced DNA strand breaks. The role of BER/SSBR in this process was verified previously however, there is no data linking sensitisation of TopoI poison-induced cytotoxicity by PARP inhibitors to a specific cell cycle phase. The aim of this project was to test the hypothesis that a) PARP inhibition inhibits the repair of CPT-induced SSB, resulting in more replication-associated DSB in S phase cells, b) PARP inhibition further retards DSB repair and c) consequently the greatest sensitisation is seen in S phase cells. To elucidate the mechanisms DNA breakage, and cytotoxicity were determined in asynchronous cells and those separated into G1, S and G2/M phases by centrifugal elutriation, in cells exposed to the clinically active PARP-1 inhibitor, AG014699, and CPT. Cell cycle-specific variations in TopoI and PARP expression and activity were also investigated as possible contributors to ultimate cytotoxicity. Survival of asynchronous Lovo colorectal cancer cells exposed to CPT was substantially reduced by co-incubation with AG014699. CPT cytotoxicity was greater in a 60% pure S phase population than in 90% pure G1 population. AG014699 increased the cytotoxicity of CPT in the S phase cells but not cells in G1. K562 leukaemia cells were used to confirm findings obtained with Lovo cells. Highly purified (>90% pure) cell cycle specific fraction of K562 cells were used to demonstrate that CPT was most cytotoxic to S phase cells and that potentiation of CPT-induced cytotoxicity by AG014699 is predominantly related to S phase cells whose survival was reduced by 2-fold. TopoI activity was greatest in S phase cells, corresponding to the greater cytotoxicity of CPT in this phase. However, the magnitude of the effect of AG014699 on survival did not correspond with the level of PARP-1 activity, which was highest in G2 cells. The potentiation of CPT-induced cytotoxicity by AG014699 in S phase was shown to correlate with the level of SSB and DSB, which was highest in S phase. Repair of CPT-induced SSB and DSB was most rapid during S phase and AG014699 hindered the repair of both SSB and DSB to the greatest extent in S phase compared with other phases of the cell cycle. This suggests that collision with the replication fork is a major mechanism for the induction of DSB and resultant cytotoxicity. and that PARP is involved in the repair of CPT-induced DSB as well as SSB. BER/SSBR requires XRCC1 as well as PARP, furthermore, PARP-1 is proposed to act in a backup NHEJ pathway that also involves XRCC1. To test these hypotheses the effect of AG014699 on the DNA repair and cytotoxicity of CPT, temozolomide (a DNA methylating agent that induces SSB) and neocarzinostatin (an inducer of DSB) was investigated in XRCC1 wild type (AA8) and mutant (EM9) cells. AG014699 enhanced the cytotoxicity of these agents in parallel with increasing the number of DSB and inhibition of their repair in EM9 as well as AA8 cells. Therefore PARP has an additional role in the protection of cells from DNA SSB and DSB independent of XRCC1. Currently there are 9 PARP inhibitors undergoing clinical evaluation, including combinations with TopoI poisons. The data presented here demonstrate that PARP inhibitors reduce the repair of TopoI poison-induced DNA DSB as well as SSB and consequent cell survival primarily in S phase. Furthermore, we demonstrate that the activity of PARP inhibition is independent of XRCC1 indicating that tumours associated with XRCC1 polymorphisms will be similarly sensitive. These findings may have implications for the rational design of clinical trials involving PARP inhibitor – TopoI poison combinations.