Protection of nascent DNA at stalled replication forks is mediated by phosphorylation of RIF1 intrinsically disordered region
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
RIF1 is a protein of the DNA damage response with key roles on genome integrity: it prevents DSB resection and hence accurate HR repair, whilst protecting stalled forks from degradation under replication stress conditions. The authors' main finding is the identification of 3 residues in RIF1 protein, that can be phosphorylated in ATM/ATR-dependent manner. However, this phosphorylation is dispensable for the ability of RIF1 to limit double-strand break resection, but is required to counteract the degradation of stalled replication intermediates mediated by the DNA2 nuclease. Therefore, the manuscript suggests that the three sites can provide a potential switch between the two functions of RIF1. These findings will spark the interest of readers working in the DNA replication and repair fields. However, the actual mechanism by which blocking RIF1 phosphorylation prevents RIF1 function at replication forks still needs to be determined.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Molecular Biology (eLife)
Abstract
RIF1 is a multifunctional protein that plays key roles in the regulation of DNA processing. During repair of DNA double-strand breaks (DSBs), RIF1 functions in the 53BP1-Shieldin pathway that inhibits resection of DNA ends to modulate the cellular decision on which repair pathway to engage. Under conditions of replication stress, RIF1 protects nascent DNA at stalled replication forks from degradation by the DNA2 nuclease. How these RIF1 activities are regulated at the post-translational level has not yet been elucidated. Here, we identified a cluster of conserved ATM/ATR consensus SQ motifs within the intrinsically disordered region (IDR) of mouse RIF1 that are phosphorylated in proliferating B lymphocytes. We found that phosphorylation of the conserved IDR SQ cluster is dispensable for the inhibition of DSB resection by RIF1, but is essential to counteract DNA2-dependent degradation of nascent DNA at stalled replication forks. Therefore, our study identifies a key molecular feature that enables the genome-protective function of RIF1 during DNA replication stress.
Article activity feed
-
-
Author Response:
Reviewer #1 (Public Review):
The authors previously performed an I-DIRT experiment in irradiated and activated splenocytes to define the RIF1 interactome (Delgado-Benito et al., 2018). Here data from this assay are used to extract RIF1 post-translational modifications, focusing on SQ/TQ consensus sites for ATM/ATR-dependent phosphorylation. Some of these RIF1 sites seem to be conserved not only between mouse and human, but also from yeast to humans (Table S2).
Figure 1 describes the RIF1 interactome from the previously reported I-DIRT assay performed in the irradiated, activated B cells. It would have been interesting if some of the interactions relevant to replication/repair could be validated in this setting, as B cells represent a special cellular model, being activated to undergo CSR and showing an interval of …
Author Response:
Reviewer #1 (Public Review):
The authors previously performed an I-DIRT experiment in irradiated and activated splenocytes to define the RIF1 interactome (Delgado-Benito et al., 2018). Here data from this assay are used to extract RIF1 post-translational modifications, focusing on SQ/TQ consensus sites for ATM/ATR-dependent phosphorylation. Some of these RIF1 sites seem to be conserved not only between mouse and human, but also from yeast to humans (Table S2).
Figure 1 describes the RIF1 interactome from the previously reported I-DIRT assay performed in the irradiated, activated B cells. It would have been interesting if some of the interactions relevant to replication/repair could be validated in this setting, as B cells represent a special cellular model, being activated to undergo CSR and showing an interval of high proliferation that could make them intrinsically susceptible to replication stress.
We agree with the Reviewer that activated primary B cells could provide a powerful model system to probe for the molecular mechanisms underlying replication stress. However, we believe that assessing the specific biological function of the various interactions identified via RIF1 I-DIRT is out of the scope of our current manuscript. Nonetheless, we hope that this data will provide novel insights and prompt new lines of investigation into the regulation of the several roles played by RIF1 in DNA replication and repair.
The manuscript is focused on three SQ/TQ sites (S1387Q, S1416Q and S1528Q) localized within RIF1 intrinsically disordered region (IDR). These should be highlighted in Table S2. The reason for choosing these sites is not entirely clear (other than their proximity, meaning they could form a cluster), as most of the sites listed in Table S2 are conserved between mouse and human. It is not clear either whether these sites correspond to 7 clustered SQ/TQ sites identified in yeast (referred to in Discussion).
We share the Reviewer’s interest in assessing whether phosphorylation of other conserved SQ/TQ motifs contributes to regulate the (many) molecular functions of RIF1. In this study, we have however focused on S1387, S1416 and S1528 since these sites were the only SQ/TQ motifs to be found reproducibly phosphorylated in RIF1 I-DIRT pull-downs (Fig 1,E and F). In addition, they all reside within RIF1 IDR, and phosphorylation of residues within disordered regions has been reported to affect protein functions in a variety of cellular contexts. Finally, and as the Reviewer her/him-self hints at, the motifs’ proximity solidified our interest and the resolution to investigate these phosphosites as a cluster. We have now emphasized these points in the revised manuscript, and highlighted the three sites in Figure 1 – source data 2 (previously Table S2) as suggested.
Because of the high sequence divergency of RIF1 intrinsically disordered regions, where these sites reside, it is not possible to unambiguously assign residue equivalence between the mammalian and yeast proteins. However, as we point out in the manuscript discussion, orthologous IDRs exhibit molecular features, such as length, complexity and net charge, that are crucial for function but do not necessarily translate into any noticeable similarity at the level of primary amino acid sequences (Zarin et al. PNAS 2017; Zarin et al. eLife 2019). The identification of a cluster of SQ/TQ motifs whose phosphorylation influences fork protection in both mouse and S. cerevisiae RIF1 (our study; Monerawela et al., bioRxiv 2020) hints at such a molecular feature. This evolutionary signature likely involves changes in net charge due to a combination of phosphorylation events within the SQ cluster. The idea is also supported by the new finding in the revised manuscript that no single SQ mutant can fully recapitulate the fork degradation phenotype of RIF1S->A, thus indicating that multiple phosphorylation events within the IDR-CII SQ cluster are necessary to support RIF1 fork protection function (Fig. 3, new panels E and F, and 3 – figure supplement 1, new panels F and G).
Reviewer #2 (Public Review):
This manuscript uncovers a novel regulatory mechanism that modulates RIF1 function during the DNA replication stress response. The authors identify a cluster of three phosphorylation sites within the intrinsically disordered region of mouse RIF1 using a mass spectrometry-based approach. They show that phosphorylation of these three sites is dispensable for the ability of RIF1 to limit double-strand break resection, but is required to counteract the degradation of stalled replication intermediates mediated by the DNA2 nuclease. Collectively, the authors' findings would be of interest for the DNA replication and repair fields. However, the study is very preliminary and the authors need to include new experiments to strengthen their conclusions and support their model. Specifically, additional data are necessary to define mechanism by which blocking RIF1 phosphorylation regulates DNA2-dependent degradation of stalled replication intermediates. Moreover, the model that RIF1 phosphorylation is dispensable for the ability of RIF1 to inhibit DSB resection is not fully supported by the data.
We thank the Reviewer for recognizing that our findings would be of interest for the DNA replication and repair fields.
We have now performed additional experiments to mechanistically dissect how phosphorylation of the conserved IDR cluster contributes to RIF1 role in the protection of nascent DNA at stalled forks. We first assessed the integrity of RIF1-PP1 interaction, and found that abrogation of phosphorylation events in the conserved cluster does not have a major impact on RIF1-PP1 association (new Fig. 4A). Next, we investigated HU-induced localization of RIF1 to newly-replicated DNA, and showed that the interaction of RIF1SA mutant to stalled replication forks is not as efficient as for the wild-type protein counterpart (new Fig. 4B). Altogether, these findings support a model where phosphorylation of RIF1 IDR-CII SQ enables the efficient recruitment of RIF1 to DNA replication forks under conditions of replication stress.
To strengthen our conclusion that phosphorylation of RIF1 IDR-CII SQ cluster is dispensable for its ability to inhibit DSB resection, we have complemented the previous data on PARPi-induced genome instability in BRCA1-deficient cells and CSR efficiency in B cells with the assessment of IR-induced DSB processing. To this end, we compared RPA (RPA32 S4/S8) phosphorylation levels in RIF1-proficient, -deficient, and RIF1S->A-expressing cells, on both WT and Brca1 mut genetic backgrounds. As expected, IR induced a marked phosphorylation of RPA in the absence of RIF1. In contrast, Rif1S->A cells were as proficient as controls to counteract RPA phosphorylation following IR-induced DSBs. This new data is presented in Figures 2 – figure supplement 2H and 3 – figure supplement 1E, and it provides a more direct evidence that phosphorylation of the IDR-CII SQ cluster does not contribute to RIF1 ability to inhibit DSB resection.
-
Evaluation Summary:
RIF1 is a protein of the DNA damage response with key roles on genome integrity: it prevents DSB resection and hence accurate HR repair, whilst protecting stalled forks from degradation under replication stress conditions. The authors' main finding is the identification of 3 residues in RIF1 protein, that can be phosphorylated in ATM/ATR-dependent manner. However, this phosphorylation is dispensable for the ability of RIF1 to limit double-strand break resection, but is required to counteract the degradation of stalled replication intermediates mediated by the DNA2 nuclease. Therefore, the manuscript suggests that the three sites can provide a potential switch between the two functions of RIF1. These findings will spark the interest of readers working in the DNA replication and repair fields. However, the actual …
Evaluation Summary:
RIF1 is a protein of the DNA damage response with key roles on genome integrity: it prevents DSB resection and hence accurate HR repair, whilst protecting stalled forks from degradation under replication stress conditions. The authors' main finding is the identification of 3 residues in RIF1 protein, that can be phosphorylated in ATM/ATR-dependent manner. However, this phosphorylation is dispensable for the ability of RIF1 to limit double-strand break resection, but is required to counteract the degradation of stalled replication intermediates mediated by the DNA2 nuclease. Therefore, the manuscript suggests that the three sites can provide a potential switch between the two functions of RIF1. These findings will spark the interest of readers working in the DNA replication and repair fields. However, the actual mechanism by which blocking RIF1 phosphorylation prevents RIF1 function at replication forks still needs to be determined.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)
-
Reviewer #1 (Public Review):
The authors previously performed an I-DIRT experiment in irradiated and activated splenocytes to define the RIF1 interactome (Delgado-Benito et al., 2018). Here data from this assay are used to extract RIF1 post-translational modifications, focusing on SQ/TQ consensus sites for ATM/ATR-dependent phosphorylation. Some of these RIF1 sites seem to be conserved not only between mouse and human, but also from yeast to humans (Table S2).
Figure 1 describes the RIF1 interactome from the previously reported I-DIRT assay performed in the irradiated, activated B cells. It would have been interesting if some of the interactions relevant to replication/repair could be validated in this setting, as B cells represent a special cellular model, being activated to undergo CSR and showing an interval of high proliferation …
Reviewer #1 (Public Review):
The authors previously performed an I-DIRT experiment in irradiated and activated splenocytes to define the RIF1 interactome (Delgado-Benito et al., 2018). Here data from this assay are used to extract RIF1 post-translational modifications, focusing on SQ/TQ consensus sites for ATM/ATR-dependent phosphorylation. Some of these RIF1 sites seem to be conserved not only between mouse and human, but also from yeast to humans (Table S2).
Figure 1 describes the RIF1 interactome from the previously reported I-DIRT assay performed in the irradiated, activated B cells. It would have been interesting if some of the interactions relevant to replication/repair could be validated in this setting, as B cells represent a special cellular model, being activated to undergo CSR and showing an interval of high proliferation that could make them intrinsically susceptible to replication stress.
The manuscript is focused on three SQ/TQ sites (S1387Q, S1416Q and S1528Q) localized within RIF1 intrinsically disordered region (IDR). These should be highlighted in Table S2. The reason for choosing these sites is not entirely clear (other than their proximity, meaning they could form a cluster), as most of the sites listed in Table S2 are conserved between mouse and human. It is not clear either whether these sites correspond to 7 clustered SQ/TQ sites identified in yeast (referred to in Discussion).
-
Reviewer #2 (Public Review):
This manuscript uncovers a novel regulatory mechanism that modulates RIF1 function during the DNA replication stress response. The authors identify a cluster of three phosphorylation sites within the intrinsically disordered region of mouse RIF1 using a mass spectrometry-based approach. They show that phosphorylation of these three sites is dispensable for the ability of RIF1 to limit double-strand break resection, but is required to counteract the degradation of stalled replication intermediates mediated by the DNA2 nuclease. Collectively, the authors' findings would be of interest for the DNA replication and repair fields. However, the study is very preliminary and the authors need to include new experiments to strengthen their conclusions and support their model. Specifically, additional data are …
Reviewer #2 (Public Review):
This manuscript uncovers a novel regulatory mechanism that modulates RIF1 function during the DNA replication stress response. The authors identify a cluster of three phosphorylation sites within the intrinsically disordered region of mouse RIF1 using a mass spectrometry-based approach. They show that phosphorylation of these three sites is dispensable for the ability of RIF1 to limit double-strand break resection, but is required to counteract the degradation of stalled replication intermediates mediated by the DNA2 nuclease. Collectively, the authors' findings would be of interest for the DNA replication and repair fields. However, the study is very preliminary and the authors need to include new experiments to strengthen their conclusions and support their model. Specifically, additional data are necessary to define mechanism by which blocking RIF1 phosphorylation regulates DNA2-dependent degradation of stalled replication intermediates. Moreover, the model that RIF1 phosphorylation is dispensable for the ability of RIF1 to inhibit DSB resection is not fully supported by the data.
-
