Senescence is a tumor suppression mechanism defined by stable proliferation arrest. Here we demonstrate that the known synthetic lethal interaction between poly(ADP-ribose) polymerase 1 inhibitors (PARPi) and DNA repair triggers p53-independent ovarian cancer cell senescence defined by senescence-associated phenotypic hallmarks including DNA-SCARS, inflammatory secretome, Bcl-XL-mediated apoptosis resistance, and proliferation restriction via Chk2 and p21 (CDKN1A). The concept of senescence as irreversible remains controversial and here we show that PARPi-senescent cells re-initiate proliferation upon drug withdrawal, potentially explaining the requirement for sustained PARPi therapy in the clinic. Importantly, PARPi-induced senescence renders ovarian and breast cancer cells transiently susceptible to second-phase synthetic lethal approaches targeting the senescence state using senolytic drugs. The combination of PARPi and a senolytic is effective in preclinical models of ovarian and breast cancer suggesting that coupling these synthetic lethalities provides a rational approach to their clinical use and may together be more effective in limiting resistance.

No abstract

Cellular senescence is a natural tumor suppression mechanism defined by a stable proliferation arrest. In the context of cancer treatment, cancer cell therapy-induced senescence (TIS) is emerging as an omnipresent cell fate decision that can be pharmacologically targeted at the molecular level to enhance the beneficial aspects of senescence. In prostate cancer (PCa), TIS has been reported using multiple different model systems, and a more systematic analysis would be useful to identify relevant senescence manipulation molecular targets. Here we show that a spectrum of PCa senescence phenotypes can be induced by clinically relevant therapies. We found that DNA damage inducers like irradiation and poly (ADP-ribose) polymerase1 (PARP) inhibitors triggered a stable PCa-TIS independent of the p53 status. On the other hand, enzalutamide triggered a reversible senescence-like state that lacked evidence of cell death or DNA damage. Using a small senolytic drug panel, we found that senescence inducers dictated senolytic sensitivity. While Bcl-2 family anti-apoptotic inhibitor were lethal for PCa-TIS cells harboring evidence of DNA damage, they were ineffective against enzalutamide-TIS cells. Interestingly, piperlongumine, which was described as a senolytic, acted as a senomorphic to enhance enzalutamide-TIS proliferation arrest without promoting cell death. Overall, our results suggest that TIS phenotypic hallmarks need to be evaluated in a context-dependent manner because they can vary with senescence inducers, even within identical cancer cell populations. Defining this context-dependent spectrum of senescence phenotypes is key to determining subsequent molecular strategies that target senescent cancer cells.

Senescent cells display senescence-associated (SA) phenotypic programs such as stable proliferation arrest (SAPA) and a secretory phenotype (SASP). Senescence-inducing persistent DNA doublestrand breaks (pDSBs) cause an immediate DNA damage response (DDR) and SAPA, but the SASP requires days to develop. Here, we show that following the immediate canonical DDR, a delayed chromatin accumulation of the ATM and MRN complexes coincides with the expression of SASP factors. Importantly, histone deacetylase inhibitors (HDACi) trigger SAPA and SASP in the absence of DNA damage. However, HDACi-induced SASP also requires ATM/ MRN activities and causes their accumulation on chromatin, revealing a DNA damage-independent, non-canonical DDR activity that underlies SASP maturation. This non-canonical DDR is required for the recruitment of the transcription factor NF-jB on chromatin but not for its nuclear translocation. Non-canonical DDR further does not require ATM kinase activity, suggesting structural ATM functions. We propose that delayed chromatin recruitment of SASP modulators is the result of non-canonical DDR signaling that ensures SASP activation only in the context of senescence and not in response to transient DNA damage-induced proliferation arrest.

Loss of telomeric DNA leads to telomere uncapping, which triggers a persistent, p53-centric DNA damage response that sustains a stable senescence-associated proliferation arrest. Here, we show that in normal cells telomere uncapping triggers a focal telomeric DNA damage response accompanied by a transient cell cycle arrest. Subsequent cell division with dysfunctional telomeres resulted in sporadic telomeric sister chromatid fusions that gave rise to next-mitosis genome instability, including non-telomeric DNA lesions responsible for a stable, p53-mediated, senescence-associated proliferation arrest. Unexpectedly, the blocking of Rad51/RPA-mediated homologous recombination, but not non-homologous end joining (NHEJ), prevented senescence despite multiple dysfunctional telomeres. When cells approached natural replicative senescence, interphase senescent cells displayed genome instability, whereas near-senescent cells that underwent mitosis despite the presence of uncapped telomeres did not. This suggests that these near-senescent cells had not yet acquired irreversible telomeric fusions. We propose a new model for telomere-initiated senescence where tolerance of telomere uncapping eventually results in irreversible non-telomeric DNA lesions leading to stable senescence. Paradoxically, our work reveals that senescence-associated tumor suppression from telomere shortening requires irreversible genome instability at the single-cell level, which suggests that interventions to repair telomeres in the pre-senescent state could prevent senescence and genome instability.