ATR, activated by replication stress, protects replication forks locally and suppresses origin firing globally. Here, we show that these functions of ATR are mechanistically coupled. Although initially stable, stalled forks in ATR-deficient cells undergo nucleus-wide breakage after unscheduled origin firing generates an excess of single-stranded DNA that exhausts the nuclear pool of RPA. Partial reduction of RPA accelerated fork breakage, and forced elevation of RPA was sufficient to delay such "replication catastrophe" even in the absence of ATR activity. Conversely, unscheduled origin firing induced breakage of stalled forks even in cells with active ATR. Thus, ATR-mediated suppression of dormant origins shields active forks against irreversible breakage via preventing exhaustion of nuclear RPA. This study elucidates how replicating genomes avoid destabilizing DNA damage. Because cancer cells commonly feature intrinsically high replication stress, this study also provides a molecular rationale for their hypersensitivity to ATR inhibitors.

Intrinsically disordered proteins can phase separate from the soluble intracellular space, and tend to aggregate under pathological conditions. The physiological functions and molecular triggers of liquid demixing by phase separation are not well understood. Here we show in vitro and in vivo that the nucleic acid-mimicking biopolymer poly(ADP-ribose) (PAR) nucleates intracellular liquid demixing. PAR levels are markedly induced at sites of DNA damage, and we provide evidence that PAR-seeded liquid demixing results in rapid, yet transient and fully reversible assembly of various intrinsically disordered proteins at DNA break sites. Demixing, which relies on electrostatic interactions between positively charged RGG repeats and negatively charged PAR, is amplified by aggregation-prone prion-like domains, and orchestrates the earliest cellular responses to DNA breakage. We propose that PAR-seeded liquid demixing is a general mechanism to dynamically reorganize the soluble nuclear space with implications for pathological protein aggregation caused by derailed phase separation.

Our paper identified nuclear proteins likely harboring disordered low-complexity sequences via precipitation by b-isox microcrystals. In Table S2, we ranked 580 nuclear proteins isolated in this manner and indicated that they were ordered according to the density of spectral counts. It has come to our attention that the proteins in this table are ordered by the relative density of [G/S]Y[G/S] triplet repeats rather than by spectral counts. This error affects the following sentence in the text of the Results section: ''Among the 580 mammalian proteins selectively precipitated by b-isox microcrystals, TAF15 registered the second highest number of spectral counts, and the largest subunit of RNA polymerase II registered the third highest number of spectral counts (Table S2).'' This is because the named positions had been based on ranking by triplet repeat density. We now provide with the article online the correctly ordered Table S2 (by spectral counts instead of triplet repeat density), and the affected sentence has now been changed to indicate the ranking positions of these proteins when ordered by spectral counts, such that TAF15 is 23 rd on the list and the largest subunit of RNA polymerase is 46 th. All proteins on the list are well above the false discovery rate, and the fact that both TAF15 and the largest subunit of RNA polymerase II are close to the very top of the list means that these adjustments do not alter any results or conclusions presented in the paper. We note that Table S3, which presents the yeast nuclear proteins precipitated by b-isox microcrystals, was correctly ordered by density of spectral counts as indicated. We wish to thank David Trudgian, a computational scientist in our Mass Sepctrometry Shared Resource Core, for pointing out the inconsistency in the organization and annotation of the original Table S2.