Nucleoside Analogs Radiosensitize G0 Cells by Activating DNA End Resection and Alternative End-Joining

Talazoparib enhances resection at DSBs and renders HR-proficient cancer cells susceptible to Polθ inhibition

BACKGROUND AND PURPOSE

The PARP inhibitor (PARPi), Talazoparib (BMN673), effectively and specifically radiosensitizes cancer cells. Radiosensitization is mediated by a shift in the repair of ionizing radiation (IR)-induced DNA double-strand breaks (DSBs) toward PARP1-independent, alternative end-joining (alt-EJ). DNA polymerase theta (Polθ) is a key component of this PARP1-independent alt-EJ pathway and we show here that its inhibition can further radiosensitize talazoparib-treated cells. The purpose of the present work is to explore mechanisms and dynamics underpinning enhanced talazoparib radiosensitization by Polθ inhibitors in HR-proficient cancer cells.

METHODS AND MATERIALS

Radiosensitization to PARPis, talazoparib, olaparib, rucaparib and veliparib was assessed by clonogenic survival. Polθ-proficient and -deficient cells were treated with PARPis and/or with the Polθ inhibitors ART558 or novobiocin. The role of DNA end-resection was studied by down-regulating CtIP and MRE11 expression using siRNAs. DSB repair was assessed by scoring γH2AX foci. The formation of chromosomal abnormalities was assessed as evidence of alt-EJ function using G2-specific cytogenetic analysis.

RESULTS

Talazoparib exerted pronounced radiosensitization that varied among the tested cancer cell lines; however, radiosensitization was undetectable in normal cells. Other commonly used PARPis, olaparib, veliparib, or rucaparib were ineffective radiosensitizers under our experimental conditions. Although genetic ablation or pharmacological inhibition of Polθ only mildly radiosensitized cancer cells, talazoparib-treated cells were markedly further radiosensitized. Mechanistically, talazoparib shunted DSBs to Polθ-dependent alt-EJ by enhancing DNA end-resection in a CtIP- and MRE11-dependent manner - an effect detectable at low, but not high IR doses. Chromosomal translocation analysis in talazoparib-treated cells exposed to Polθ inhibitors suggested that PARP1- and Polθ-dependent alt-EJ pathways may complement, but also back up each other.

CONCLUSION

We propose that talazoparib promotes low-dose, CtIP/MRE11-dependent resection and increases the reliance of irradiated HR-proficient cancer cells, on Polθ-mediated alt-EJ. The combination of Polθ inhibitors with talazoparib suppresses this option and causes further radiosensitization. The results suggest that Polθ inhibition may be exploited to maximize talazoparib radiosensitization of HR-proficient tumors in the clinic.

Shift in G1-Checkpoint from ATM-Alone to a Cooperative ATM Plus ATR Regulation with Increasing Dose of Radiation

The current view of the involvement of PI3-kinases in checkpoint responses after DNA damage is that ATM is the key regulator of G₁-, S- or G₂-phase checkpoints, that ATR is only partly involved in the regulation of S- and G₂-phase checkpoints and that DNA-PKcs is not involved in checkpoint regulation. However, further analysis of the contributions of these kinases to checkpoint responses in cells exposed to ionizing radiation (IR) recently uncovered striking integrations and interplays among ATM, ATR and DNA-PKcs that adapt not only to the phase of the cell cycle in which cells are irradiated, but also to the load of DNA double-strand breaks (DSBs), presumably to optimize their processing. Specifically, we found that low IR doses in G₂-phase cells activate a G₂-checkpoint that is regulated by epistatically coupled ATM and ATR. Thus, inhibition of either kinase suppresses almost fully its activation. At high IR doses, the epistatic ATM/ATR coupling relaxes, yielding to a cooperative regulation. Thus, single-kinase inhibition suppresses partly, and only combined inhibition suppresses fully G₂-checkpoint activation. Interestingly, DNA-PKcs integrates with ATM/ATR in G₂-checkpoint control, but functions in its recovery in a dose-independent manner. Strikingly, irradiation during S-phase activates, independently of dose, an exclusively ATR-dependent G₂ checkpoint. Here, ATM couples with DNA-PKcs to regulate checkpoint recovery. In the present work, we extend these studies and investigate organization and functions of these PI3-kinases in the activation of the G₁ checkpoint in cells irradiated either in the G₀ or G₁ phase. We report that ATM is the sole regulator of the G₁ checkpoint after exposure to low IR doses. At high IR doses, ATM remains dominant, but contributions from ATR also become detectable and are associated with limited ATM/ATR-dependent end resection at DSBs. Under these conditions, only combined ATM + ATR inhibition fully abrogates checkpoint and resection. Contributions of DNA-PKcs and CHK2 to the regulation of the G₁ checkpoint are not obvious in these experiments and may be masked by the endpoint employed for checkpoint analysis and perturbations in normal progression through the cell cycle of cells exposed to DNA-PKcs inhibitors. The results broaden our understanding of organization throughout the cell cycle and adaptation with increasing IR dose of the ATM/ATR/DNA-PKcs module to regulate checkpoint responses. They emphasize notable similarities and distinct differences between G₁-, G₂- and S-phase checkpoint regulation that may guide DSB processing decisions.

Disruption of Chromatin Dynamics by Hypotonic Stress Suppresses HR and Shifts DSB Processing to Error-Prone SSA

The processing of DNA double-strand breaks (DSBs) depends on the dynamic characteristics of chromatin. To investigate how abrupt changes in chromatin compaction alter these dynamics and affect DSB processing and repair, we exposed irradiated cells to hypotonic stress (HypoS). Densitometric and chromosome-length analyses show that HypoS transiently decompacts chromatin without inducing histone modifications known from regulated local chromatin decondensation, or changes in Micrococcal Nuclease (MNase) sensitivity. HypoS leaves undisturbed initial stages of DNA-damage-response (DDR), such as radiation-induced ATM activation and H2AX-phosphorylation. However, detection of ATM-pS1981, γ-H2AX and 53BP1 foci is reduced in a protein, cell cycle phase and cell line dependent manner; likely secondary to chromatin decompaction that disrupts the focal organization of DDR proteins. While HypoS only exerts small effects on classical nonhomologous end-joining (c-NHEJ) and alternative end-joining (alt-EJ), it markedly suppresses homologous recombination (HR) without affecting DNA end-resection at DSBs, and clearly enhances single-strand annealing (SSA). These shifts in pathway engagement are accompanied by decreases in HR-dependent chromatid-break repair in the G₂-phase, and by increases in alt-EJ and SSA-dependent chromosomal translocations. Consequently, HypoS sensitizes cells to ionizing radiation (IR)-induced killing. We conclude that HypoS-induced global chromatin decompaction compromises regulated chromatin dynamics and genomic stability by suppressing DSB-processing by HR, and allowing error-prone processing by alt-EJ and SSA.