Advertisement

  • Loading metrics

Open Access

Peer-reviewed

Research Article

A point mutation in the FAT domain constitutively increases the kinase activity of Rad3ATR and bypasses the requirement for 9-1–1 phosphorylation to activate the DNA replication checkpoint

  • Kamal Dev,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Writing – review & editing

    Affiliation Department of Pharmacology and Toxicology, Boonshoft School of Medicine, Wright State University, Dayton, Ohio, United States of America

    ⨯
  • S. Dean Rider Jr.,

    Roles Formal analysis

    Affiliation Department of Pharmacology and Toxicology, Boonshoft School of Medicine, Wright State University, Dayton, Ohio, United States of America

    ⨯
  • Balveer Singh,

    Roles Investigation

    Affiliation Department of Pharmacology and Toxicology, Boonshoft School of Medicine, Wright State University, Dayton, Ohio, United States of America

    ⨯
  • Abhinav Saini,

    Roles Investigation

    Affiliation Department of Pharmacology and Toxicology, Boonshoft School of Medicine, Wright State University, Dayton, Ohio, United States of America

    ⨯
  • Yong-jie Xu

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Supervision, Validation, Writing – original draft, Writing – review & editing

    yong-jie.xu@wright.edu

    Affiliation Department of Pharmacology and Toxicology, Boonshoft School of Medicine, Wright State University, Dayton, Ohio, United States of America

    ⨯

Abstract

Ataxia telangiectasia and Rad3-related (ATR) initiates cell cycle checkpoints to maintain genome integrity in the presence of replication stress or various forms of DNA damage. However, how ATR is activated for checkpoint initiation remains incompletely understood. The canonical model suggests that binding of an ATR-activator protein relieves the autoinhibitory PIKK regulatory domain (PRD) within the kinase domain, thereby activating ATR by granting substrate access to the catalytic centre. To better understand the checkpoint initiation mechanism, we conducted a genetic screen in fission yeast that identified a charge-reversal mutation, E1369K, in the conserved FRAP-ATM-TRRAP (FAT) domain of Rad3, the ortholog of ATR. In vitro kinase assays show that the mutation converts Rad3 into a constitutively active form. This allows rescue of the Rad3 kinase signaling defect in cells lacking the phosphorylation of the Rad9-Rad1-Hus1 (9-1-1) complex specifically in the DNA replication checkpoint, not the damage checkpoint pathway. Since the mutation is not in the kinase domain and is away from the PRD, these findings show that, in addition to the canonical mechanism, Rad3 may also be activated allosterically via the FAT domain, a mechanism likely conserved in higher eukaryotes.

Author summary

ATR is the sensor kinase that activates cell cycle checkpoints in response to DNA replication stress or damage. Although essential for safeguarding the genome, how ATR initiates the checkpoints at the stressed fork or site of damage remains incompletely understood. The current model suggests that binding of an activator protein relieves the autoinhibitory function of the PRD domain within the kinase domain, which activates ATR by granting substrate access to the catalytic centre. To investigate the checkpoint initiation mechanism in fission yeast, we carried out a genetic screen and identified an E1369K mutation in the conserved FAT domain of Rad3, the ortholog of ATR. In vitro kinase assays show that the mutation converts Rad3 into a constitutively active form, which suggests that, in addition to the current model, Rad3 may also be activated allosterically via the FAT domain. Interestingly, the mutation can specifically rescue the Rad3 kinase signaling defect caused by phospho-mutations in the Rad9-Rad1-Hus1 complex at the stressed fork, not the DNA damage site. These results provide insights into the checkpoint mechanisms likely conserved in higher eukaryotes.

Citation: Dev K, Rider SD Jr, Singh B, Saini A, Xu Y-j (2026) A point mutation in the FAT domain constitutively increases the kinase activity of Rad3ATR and bypasses the requirement for 9-1–1 phosphorylation to activate the DNA replication checkpoint. PLoS Genet 22(6): e1012213. https://doi.org/10.1371/journal.pgen.1012213

Editor: Bruce E. Clurman, Fred Hutchinson Cancer Research Center, UNITED STATES OF AMERICA

Received: April 7, 2026; Accepted: June 15, 2026; Published: June 22, 2026

Copyright: © 2026 Dev et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are available within the manuscript and its Supporting Information files.

Funding: This work was funded by the NIH R35GM144307 grant to YJX. The funder had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Eukaryotic cells ensure high-fidelity of DNA replication, repair, and cell division through tightly integrated cell signaling networks that guard genome integrity. Replication stress, whether caused by external genotoxins or internal metabolic imbalances or suppression, can stall or collapse replication forks, posing a serious threat to genome stability [1,2]. Eukaryotic cells manage replication stress mainly through a highly conserved mechanism, the DNA replication checkpoint (DRC), which stabilizes stalled forks, prevents premature mitosis, and promotes replication and repair, ensuring DNA replication is complete before cell division [3]. When the DRC fails, stalled forks become unstable and may collapse into double-strand breaks (DSBs), the most lethal form of DNA damage. DNA damage can also activate the DNA damage checkpoint (DDC) to promote repair while suppressing premature mitosis. Therefore, the DRC and DDC are crucial for maintaining genome integrity. Defects in the DRC and DDC cause chromosomal abnormalities and genome instabilities, which drive cancer development and other genetic disorders [4].

Although eukaryotic cells rely on the checkpoints to manage replication stress, the molecular mechanisms, particularly at the initiation stage, remain incompletely understood. The current model suggests that a set of sensor proteins conserved in all eukaryotes assembles at the perturbed fork or site of damage to initiate the checkpoint signaling [3]. Checkpoint signaling is then relayed through mediator proteins to the effector kinases CHK1 and CHK2, which phosphorylate hundreds of cellular proteins to mediate the checkpoint functions described above. In humans, ataxia–telangiectasia mutated (ATM) activates the checkpoint mainly through CHK2 in the presence of DSBs, whereas ATR activates CHK1 in response to replication stress or various other forms of DNA lesions. For ATR-initiated checkpoint signaling, the replication protein A (RPA)-coated single-strand DNA (ssDNA) binds to ATRIP, the cofactor of ATR, which recruits the ATR-ATRIP complex to the fork or damage site to initiate the checkpoints. The RPA-ssDNA platform also promotes the loading of the Rad9-Rad1-Hus1 (9-1-1) clamp at the 5’ end of the ssDNA/dsDNA junction [5–7]. The loaded 9-1-1, upon phosphorylation by ATR, recruits additional checkpoint proteins, such as TopBP1, carrying an ATR activation domain [8]. The recruited TopBP1 activates ATR and amplifies its kinase signaling. Like TopBP1, Ewing Tumor-Associated Antigen 1 (ETAA1) can also activate ATR both in vitro and in vivo [9–11]. In budding yeast, three proteins have been identified that possess a Mec1ATR-activation domain: Ddc1Rad9, the large subunit of the 9-1-1 complex; Dna2, an Okazaki maturation factor on the lagging strand; and Dpb11TopBP1/Rad4, which is recruited by phosphorylated Ddc1 and the major Mec1 activator in budding yeast [12–14]. Although structural evidence is still lacking, it is believed that binding of an activation domain activates ATR by relieving the autoinhibitory PRD, which enables substrate access to the catalytic center [11,15,16]. Once activated, ATR activates CHK1, and similarly, Mec1 phosphorylates Rad53CHK2/Cds1 in budding yeast.

In the fission yeast S. pombe, Rad3ATR/Mec1 activates both the DRC and the DDC pathways, whereas Tel1ATM has a minimal checkpoint role. Unlike the human ATR-Chk1 and budding yeast Mec1-Rad53 signaling cascades, Rad3 activates the effector kinase Cds1CHK2/Rad53 at the fork and Chk1 at the damage site to activate the DRC and DDC separately. Therefore, examining the phosphorylation of Cds1 or Chk1 can reveal the origins of Rad3 kinase signaling. This promotes an unambiguous description of the checkpoint initiation mechanism. By taking this technical advantage in fission yeast, we carried out a genetic screen to better understand DRC mechanisms. Here, we report the discovery of a missense mutation in Rad3 that specifically bypasses Rad9 phosphorylation in the DRC, not the DDC pathway. Examination of Rad3 kinase activity by in vitro and in-cell assays reveals that, in addition to the current PRD relief mechanism, Rad3 may also be activated by an allosteric mechanism through the FAT domain.

Results

Phosphorylation of Rad9 is required for the activation of Cds1Chk2/Rad53, not Mrc1Claspin

Rad3ATR/Mec1, together with its cofactor Rad26ATRIP/Ddc2, is the master sensor kinase of the DRC and DDC pathways in fission yeast (Fig 1A). In the DRC, Rad3 phosphorylates two redundant TQ motifs (T645 and T653) in Mrc1, the mediator of the DRC [17,18]. Phosphorylated Mrc1 recruits the effector kinase Cds1 to be phosphorylated by Rad3 at Cds1-T11. Phosphorylation of Cds1-T11 promotes autophosphorylation of Cds1-T328, which directly activates the effector kinase [19]. Activated Cds1 mediates most biological functions of the DRC. When DNA damage occurs during G2, the longest cell cycle phase in fission yeast, Rad3 phosphorylates Crb253 BP1/Rad9 and Chk1 to activate the DDC [20–22].

thumbnail
Download:
Fig 1. The C-terminus of Rad9 is required for the phosphorylation of Cds1, but not Mrc1.

(A) Overview of the Rad3ATR/Mec1 kinase signaling in the DRC (left) and DDC (right) pathways in fission yeast. Numbers represent Rad3-specific phosphorylation sites. In the DRC, Rad3 phosphorylates Mrc1, which recruits Cds1 to be phosphorylated by Rad3. In the DDC, phosphorylated Rad9 recruits Rad4 to promote Rad3 phosphorylation of Crb2 and Chk1. Rad3 also phosphorylates Rad9 of the Rad9-Rad1-Hus1, or 9-1-1 complex, which is necessary for Rad3 phosphorylation of both Cds1 and Chk1. However, the function of Rad9 phosphorylation in Cds1 activation remains unknown, as indicated by the question mark. (B) Rad3 phosphorylation of Cds1 was examined by Western blotting using a phospho-specific antibody in wild-type S. pombe and the mutants with the indicated mutation treated with (+) or without (-) 15 mM HU for 3 h. Cds1-HA was IPed from whole cell lysates and separated by SDS PAGE for the Western analysis using anti-HA antibodies to detect Cds1 (lower panel). The membrane was stripped and re-blotted with the phospho-specific antibody (upper panel). The phosphorylation bands were quantified and shown below in percentages. (C) Rad3 phosphorylation of Mrc1 in wild-type and mutant S. pombe with indicated mutations. The cells were treated with HU as in B and fixed in 15% trichloroacetic acid. Whole cell extracts made by using a mini-bead beater were analysed by Western blotting using the antibody against Mrc1-pT645 (top panel). The membrane was stripped, washed overnight, and reblotted with the antibody against Mrc1 (middle panel). A section of the Ponceau S-stained membrane is shown for loading (bottom panel). The quantitation results are shown at the bottom. (D) Drug sensitivities of S. pombe mutants used in B and C were determined by spot assay. Wild-type and rad3∆ strains were included as controls. A series of five-fold dilutions of logarithmically grown S. pombe were spotted on a YE6S plate or YE6S plates containing HU or MMS at the indicated concentrations. Plates were incubated at 30˚C for 3 days and then photographed.

https://doi.org/10.1371/journal.pgen.1012213.g001

Rad3 also phosphorylates T412 at the C-terminus of Rad9 in the independently loaded 9-1-1 clamp. In the DDC, phosphorylated Rad9 recruits Rad4TopBP1/Dpb11 (also known as Cut5) [23], which collaborates with phosphorylated Crb2 to recruit Chk1 to be phosphorylated by Rad3. Rad9 phosphorylation is also required for Cds1 activation in the DRC, as the phospho-mutations, rad9-T412A and rad9-∆C, lacking the entire C-terminus (411–426 aa), abolished Cds1 phosphorylation in the presence of hydroxyurea (HU), a replication stress inducer (Fig 1B) [18]. The mediator Crb2 in the DDC is essential for Chk1 phosphorylation [24]; however, its deletion has minimal impact on Cds1 phosphorylation (Fig 1B). Previous studies suggest that Rad4 carries a Rad3 activation domain in the C-terminus [25]. Deletion of the entire C-terminus (∆498–648 aa) in Rad4, however, did not affect Cds1 phosphorylation (Fig 1B) [26]. We also examined Mrc1 phosphorylation in HU (Fig 1C). In HU-treated wild-type cells, Rad3 phosphorylation of Mrc1 was significantly increased. Since Mrc1 is expressed during the G1/S phase and the activated DRC promotes Mrc1 expression [27,28], the Mrc1 protein levels were also increased in HU. Under similar conditions, Mrc1 phosphorylation was moderately increased or unaffected in rad9-T412A, rad9-∆C, crb2∆, and rad4-∆C mutants (Fig 1C), confirming the previous results [18,26]. Consistent with the defect in Cds1 phosphorylation, the spot assay showed that the rad9-T412A mutant was sensitive to HU, although the sensitivity was slightly lower than that of rad3∆ and rad9-∆C cells (Fig 1D). Since Rad9 phosphorylation is required for Chk1 phosphorylation, the phospho-mutant was also sensitive to the DNA-damaging agent methyl methane sulfonate (MMS) [18,29]. Both HU and MMS cause replication stress, which is mainly managed by DRC, in collaboration with DDC. The crb2∆ cells were therefore moderately sensitive to both agents [24]. These results show that Rad9 phosphorylation is necessary for Rad3 phosphorylation of both Cds1 and Chk1, but not for Mrc1. Although the role of Rad9 phosphorylation in Chk1 activation is relatively clear [22,30], its function in promoting Cds1 activation remains unknown (Fig 1A, the question mark).

Genetic screen identified suppressors of rad9 phospho-mutants

To identify the missing link between the phosphorylation of Rad9 and Cds1, we mutagenized the genomes of rad9-T412A and rad9-∆C and screened for suppressors that conferred HU resistance using the strategy illustrated in S1 Fig. With the rad9-∆C mutant, we screened a set of K suppressors with increased resistance to HU, MMS, or both (S2 Fig A, top half). Similarly, several L suppressors were screened in rad9-T412A (S2 Fig A, bottom half). When Cds1 phosphorylation was examined in the screened suppressors, only L11 showed robust phosphorylation (S2 Fig B). Since L11 increased the resistance to MMS, we examined Chk1 activation in the DDC. Surprisingly, Chk1 phosphorylation was not restored in L11 (S2 Fig C), suggesting a DRC-specific rescuing effect. Since Cds1 phosphorylation depends on phosphorylated Mrc1 [17,18], we examined Mrc1 phosphorylation in HU and found that, while the phosphorylation was unaffected in rad9-T412A, it was significantly increased in L11 containing the rad9-T412A mutation (S2 Fig D). Furthermore, the basal level of Mrc1 phosphorylation in untreated L11 was also higher (Fig 2D), suggesting heightened Rad3 kinase activity in L11 (see below). To confirm the mutation, we crossed L11 into rad9-∆C. Spot assay showed that L11 also increased drug resistance in rad9-∆C (S3 Fig A). We then extrachromosomally expressed Rad4, Cds1, Rad9, and Suc22, the small subunit of ribonucleotide reductase, under their native promoters in the L11 rad9-∆C cells (S3 Fig B). While expression of Rad9 increased drug resistance in L11 rad9-∆C to wild-type level as expected, Suc22 increased resistance to HU, not MMS, as Suc22 is the target of HU. Under similar conditions, expression of Rad4 or Cds1 did not change drug resistance in L11 rad9-∆C, suggesting that they are not mutated in L11. Since other suppressors did not restore Cds1 phosphorylation (S2 Fig B), the following studies were focused on L11.

thumbnail
Download:
Fig 2. Identification and confirmation of the E1369K mutation in Rad3.

(A) Sanger sequencing confirms the rad3-E1369K mutation identified by whole-genome sequencing in the L11 suppressor. The upper panel shows the domain organization of Rad3. The asterisk indicates the location of the mutation. The bottom panel is the electropherogram showing the G to A nucleotide change. (B) The E1369K mutation was integrated at the rad3 genomic locus using the method illustrated in S6 Fig A. Integrants of wild-type and the mutant rad3 were analysed by Western blotting using anti-myc antibody (top panel). Ponceau S staining serves as the loading control (bottom panel). Asterisk indicates a non-specific band. Quantification results are shown at the bottom. (C) Drug sensitivities of the rad3 integrant in rad9 or rad9-ΔC background were examined by spot assay as in Fig 1D. (D) Acute HU sensitivities of wild type, rad3∆, the primary L11, and the rad3 integrants were determined by spot assay. The cells were incubated in YE6S medium containing 15 mM HU. Every 2 h during the HU treatment, a small aliquot of the culture was removed. The cells were washed once and spotted onto a YE6S plate to recover at 30˚C for 3 days. (E) Colony recovery assay. The primary L11 and the rad3-E1369K integrant in rad9-∆C background were treated with HU, as in D. Every 2 h during the treatment, an equal volume of culture was removed, diluted 1000-fold, and spread onto three YE6S plates. The recovered colonies were counted, and the results are presented in percentages.

https://doi.org/10.1371/journal.pgen.1012213.g002

Genome sequencing of L11 identified a missense mutation in the FAT domain of Rad3

For genome sequencing, we performed random spore analysis after backcrossing L11 with the parental rad9-∆C strain. Multiple HU-resistant colonies with restored Cds1 phosphorylation were pooled for genome sequencing, which revealed several candidate genes. One of them was rad3. Sanger sequencing of the genomic locus identified a single G- > A nucleotide change, causing a charge reversal E1369K mutation in Rad3 (Fig 2A). The mutation is in the conserved N-terminal region of the FAT domain (Fig 2A, top panel, and S4 Fig A). The mutated E1369 residue, however, is not highly conserved in budding yeast Mec1 and human ATR (S4 Fig A). Interestingly, the equivalent residue Y1623 in ATR is close to the catalytic centre and distant from the autoinhibitory PRD in the cryo-EM structure of ATR-ATRIP (S4 Fig B) [31–33].

To confirm that the rad3 mutation restores Cds1 phosphorylation, we tagged Rad3 at the N-terminus with a 10xmyc epitope and expressed wild-type Rad3 and the mutant Rad3 from a vector under the control of its native promoter in the double mutant of rad3∆ rad9-∆C. The spot assay showed that while wild-type Rad3 did not affect the drug sensitivity, Rad3-E1369K significantly increased the drug resistance, like the L11 mutant in rad9-∆C (S5 Fig A), although HU resistance was less increased (see below). Western blotting showed that the protein expression level of Rad3-E1369K was similar to wild-type Rad3 (S5 Fig B). When Cds1 phosphorylation was examined, we found that while wild-type Rad3 did not increase the phosphorylation, Rad3-E1369K significantly increased the phosphorylation to a level even slightly higher than in wild-type cells (S5 Fig C). When Mrc1 phosphorylation was examined in these cells (S5 Fig D), the mutant Rad3 increased Mrc1 phosphorylation in HU significantly higher than in wild-type cells.

Integration of the rad3-E1369K mutation at the genomic locus rescues rad9-∆C in HU

We then integrated the mutation at the rad3 genomic locus in wild-type S. pombe using the strategy illustrated in S6 Fig A. As a control, wild-type rad3 was integrated by the same method. Western blotting showed that Rad3-E1369K was expressed in rad9-∆C at the wild-type level (Fig 2B). Spot assay showed that while integration of wild-type rad3 did not rescue rad9-∆C as expected, the integration of rad3-E1369K increased resistance to HU, but not MMS, in rad9-∆C (Fig 2C), which is different from what was observed in L11 (compare Fig 2C and S2 Fig A). We then compared the rad3-E1369K integrant with L11 on the same drug plate. The integrant was indeed less resistant to HU than L11, which showed a higher resistance to both HU and MMS (S6 Fig B).

We have previously reported that chronic HU treatment, such as the spot assay, generates replication stress as well as oxidative stress [34–36]. Acute HU treatment in liquid culture, however, mainly causes replication stress. To investigate the differences between the rad3-E1369K integrant and L11 mutant, we examined acute HU sensitivity. The spot assay showed that when treated with HU for several hours in liquid culture, the rad3-E1369K integrant significantly increased HU resistance in rad9-∆C, similar to L11 (Fig 2D). We also performed the colony recovery assay; the rad3-E1369K integrant also behaved similarly to the L11 mutant (Fig 2E). These results suggest that L11 carries a secondary mutation with a further enhanced HU resistance. To investigate, we performed tetrad dissection of L11 after backcrossing with rad9-∆C and found that while some of the tetrads had two HU-resistant and two HU-sensitive spores, as expected, three tetrads showed three HU-resistant spores (S6 Fig C, green dashed squares) and the levels of HU resistance varied among the spores. This result confirms that L11 carries a secondary mutation that promotes cell survival, likely in HU-induced oxidative stress, such as the metabolic mutants we have discovered earlier [35,36]. To confirm, we added the antioxidant N-acetyl cysteine to the spot assay and found that it significantly increased the HU resistance of rad3-E1369K integrant in rad9-∆C (S6 Fig D). Together, these results show that the rad3-E1369K mutation rescues rad9-∆C from HU-induced replication stress.

Mrc1 phosphorylation is necessary for Cds1 phosphorylation in rad3-E1369K

Using the rad3-E1369K integrant, we re-examined Cds1 phosphorylation in rad9-T412A or rad9-∆C cells. As shown in Fig 3A, HU treatment significantly increased Cds1 phosphorylation in both rad3 and rad3-E1369K integrants in rad9 cells, as in wild-type cells. However, unlike the rad3 integrant, which did not increase Cds1 phosphorylation in HU-treated rad9-∆C, the rad3-E1369K integrant significantly increased Cds1 phosphorylation in rad9-∆C, although the level was slightly lower than in wild-type cells. We also re-examined the phosphorylation of Chk1 and Mrc1. In the presence of MMS, Chk1 phosphorylation was not restored in rad3-E1369K rad9-∆C cells (Fig 3B), similar to L11 (S2 Fig C). When Mrc1 phosphorylation was re-examined, the phosphorylation was increased in rad3-E1369K rad9-∆C more than twofold than in wild-type cells. These results are consistent with the results described for L11 and confirm that the rad3-E1369K mutation bypasses Rad9 phosphorylation to specifically activate Cds1 at the HU-treated fork.

thumbnail
Download:
Fig 3. The rad3-E1369K mutation restores the phosphorylation of Cds1, not Chk1, in rad9-∆C.

(A) Rad3 phosphorylation of Cds1 in wild-type S. pombe and the two rad3 integrants in rad9 or rad9-ΔC background was examined by Western blotting as in Fig 1B. Quantification results are shown at the bottom. (B) Chk1 phosphorylation in wild-type and the indicated rad3 integrant cells was examined by mobility-shift assay. The cells were treated with 0.01% MMS for 90 min. Whole cell lysates were analysed by Western blotting using anti-HA antibody. Chk1 phosphorylation in the upper shifted band was quantified and shown at the bottom as the ratios of phosphorylated Chk1 over total Chk1. (C) Rad3 phosphorylation of Mrc1 in wild-type S. pombe, the checkpoint mutants mrc1∆ and rad3∆, and the two rad3 integrants was examined by Western blotting as in Fig 1C. (D) The rescuing effect of rad3-E1369K in rad9-∆C relies on Mrc1 phosphorylation. The HU sensitivities of wild-type and the mutants with the indicated mutations were examined by spot assay. The mrc1-T645A-T653A mutation eliminates Rad3-specific phosphorylation on Mrc1. (E) Mrc1 phosphorylation was examined by Western blotting in the strains used in D. (F) Cds1 phosphorylation was examined in the strains used in D.

https://doi.org/10.1371/journal.pgen.1012213.g003

As described above, Mrc1 phosphorylation recruits Cds1 to be phosphorylated by Rad3. Since the rad3-E1369K mutation bypasses Rad9 phosphorylation, we then investigated whether it bypasses Mrc1 phosphorylation. We crossed mrc1∆ or mrc1-T645A-T653A, lacking Rad3-specific phosphorylation sites [17,18], into rad3-E1369K rad9-∆C. Spot assay showed that rad3-E1369K did not rescue rad9-∆C in the presence of either mrc1-T645A-T653A or mrc1∆ (Fig 3D). Western blotting for Mrc1 phosphorylation confirmed the strains (Fig 3E). Consistent with the spot assay, phosphorylation of Cds1 was not increased in the mrc1 mutants (Fig 3F), showing that, unlike Rad9 phosphorylation, the mutant Rad3 still relies on Mrc1 phosphorylation to activate Cds1. We then examined whether the mutation bypasses rad9∆ and rad1∆ of the 9-1-1 complex, and the loader rad17∆ mutant (S7 Fig). The results clearly showed that although it can bypass Rad9 phosphorylation, Rad3-E1369K still needs the loaded 9-1-1 for Cds1 activation.

Rad3-E1369K overcomes Rad26ATRIP/Ddc2 mutations with defects in Rad3 recruitment

The main function of Rad26 is to recruit Rad3 to the fork or damage site for checkpoint initiation. Rad26 carries an RPA-binding domain (RBD) in the N-terminus. The RBD cooperates with the KKRK motif in Rad26 for Rad3 recruitment. Eliminating the RBD and the KKRK motif significantly reduced the Rad3 kinase signaling at the fork [37]. To see whether the E1369K mutation bypasses the Rad26 mutants, we deleted rad26 in rad3 or rad3-E1369K integrants and expressed Rad26 or Rad26 mutants on a vector under the control of its native promoter. Spot assay showed that expression of Rad26 mutants of ∆30, lacking the RBD, or the KKRK mutation, sensitized rad26∆ rad3 cells to HU as previously reported [37], whereas the same rad26 mutations did not sensitize rad26∆ rad3-E1369 significantly (Fig 5A). The F18A mutation, known to eliminate the RPA-binding activity [37], sensitized ∆rad26 rad3, but not much in ∆rad26 rad3-E1369K cells. Expressing Rad26 with the combined mutation ∆30 + KKRK significantly sensitized both ∆rad26 rad3 and ∆rad26 rad3-E1369K cells, although ∆rad26 rad3-E1369K cells also showed enhanced cell survival (Fig 4A).

thumbnail
Download:
Fig 4. The rad3-E1369K mutation partially rescues rad26 mutants with defects in Rad3 recruitment.

(A) Wild-type Rad26 and mutant Rad26 with the indicated mutations were expressed on a vector under its native promoter in the Δrad26 S. pombe expressing wild-type Rad3 or Rad3-E1369K from the genomic locus. The HU sensitivities of the indicated strains were determined by spot assay. (B) Mrc1 phosphorylation in the strains used in (A) was examined by Western blotting. The middle portion of the same membrane was blotted with anti-HA antibody to detect the N-terminally tagged Rad26. The asterisk indicates a cross-reaction material. A section of the Ponceau S-stained membrane is shown as the loading control. Quantification results are shown at the bottom.

https://doi.org/10.1371/journal.pgen.1012213.g004

thumbnail
Download:
Fig 5. Rad3-E1369K is constitutively active.

Wild-type Rad3, Rad3-E1369K, and kinase-inactive Rad3 were IPed from rad9 or rad9-∆C cells treated with (+) or without (-) HU using anti-myc antibody bound to the magnetic Dynabeads. After a brief wash, the IPed Rad3 was incubated with bacterially purified kinase-dead Cds1(D312E) as the substrate to examine the kinase activities as described in Materials and Methods. The samples were incubated for 60 min at 25°C and separated by SDS-PAGE followed by Western blotting using the phosphor-specific antibody against Cds1-pT11 in the bottom part of the blot (top two panels with short or long exposures). The top part of the blot was incubated with anti-myc antibody to detect Rad3 (3rd panel from the top). The Cds1 substrate and the IgG used for the IP were shown by Ponceau S staining (4th panel from the top). The Cds1 phosphorylation bands were quantified and shown in percentages at the bottom. The lower two panels are 2.0% inputs of the IP experiment. Rad3 was detected by Western blotting using an anti-myc antibody. Asterisk indicates a cross-reacting material. A portion of the Ponceau S-stained membrane is shown for the loading (bottom panel).

https://doi.org/10.1371/journal.pgen.1012213.g005

Mrc1 is a nonessential replisome protein. Once Rad3 is recruited to the fork, it phosphorylates Mrc1 without the need for other known checkpoint proteins (Fig 1) [18]. To investigate how rad3-E1369K rescues the rad26 mutants, we examined Mrc1 phosphorylation (Fig 4B). As previously reported, all rad26 mutants were expressed at or near the wild-type levels (Fig 4B, 3rd panel from the top) [37]. In rad26∆ rad3 cells, expression of Rad26 with KKRK or ∆30 mutations mildly or moderately reduced Mrc1 phosphorylation, whereas the combined ∆30 + KKRK mutation almost eliminated Mrc1 phosphorylation in HU. The same Rad26 mutations, when expressed in rad26∆ rad3-E1369K cells, did not reduce Mrc1 phosphorylation significantly as in rad26∆ rad3 cells. Together, these results show that the rad3-E1369K mutation significantly rescues rad26 recruitment mutants by increasing Mrc1 phosphorylation, although it still relies on Rad3 recruitment for the phosphorylation.

The mutated residue E1369 resides in a short SRESS motif of Rad3. AlphaFold modelling of the Rad3-Rad26 heterodimer showed that the SRESS motif forms a small loop between two α helices, and the charge-reversal mutation converts the loop into an α helix (S8 Fig A). The conformational change impacts Rad26, enabling the N-terminus of Rad26 with multiple acidic residues to bind to the mutated region of Rad3 (S8 Fig A, right panels). This raises a possibility that the mutation may enhance Rad26 binding to Rad3 and thus increase Rad3 recruitment, leading to more robust signaling. To investigate this possibility, we performed a co-IP experiment. The results showed that the mutation did not increase the co-IP of Rad3 with Rad26 (S8 Fig B). Since Rad26 binds to Ssb1 [37], the large subunit of RPA, Ssb1 was also co-IPed with Rad26 in rad3-E1369K at a level similar to that in wild-type cells. We then performed the DNA pull-down assay (S8 Fig C) [38]. The results showed that Rad3, Rad26, and the Rad26 mutants were pulled down similarly in rad3-E1369K as in wild-type rad3 cells. These results ruled out the possibility that the mutation increases Rad3 recruitment, leading to the increased phosphorylation of Mrc1 and Cds1 in HU-treated rad9-∆C cells.

The E1369K mutation constitutively increases Rad3 kinase activity

The higher level of Mrc1 phosphorylation observed in Fig 3C suggests that the mutation may increase Rad3 kinase activity. To investigate, we IPed Rad3 from the strains used in S8 Fig C for an in vitro kinase assay using purified kinase-inactive Cds1(D312E) as substrate (S9 Fig) [26]. After the kinase reaction, the samples were analysed by Western blotting to detect Rad3 (3rd panel from the top), the co-IPed Rad26 (2nd panel), and phosphorylated Cds1 (top panel). The results clearly showed that Rad3-E1369K had a significantly higher kinase activity than wild-type Rad3 in cells expressing Rad26 or the Rad26 mutants. The kinase activity of Rad3-E1369K is dependent on Rad26, as only background kinase activity was detected in the vector control.

To further investigate the increased kinase activity of Rad3-E1369K, we IPed Rad3 and Rad3-E1369K from rad9 or rad9-∆C cells treated with or without HU. The kinase-dead Rad3(D2249E) was IPed in parallel, which showed no detectable kinase activity on Cds1 substrate (Fig 5, lane 9). In the absence of HU, wild-type Rad3 showed a basal kinase activity in rad9 and rad9-∆C cells (Fig 5, lanes 3 and 7), while Rad3-E1369K significantly increased kinase activity several-fold in rad9-∆C and rad9 cells (compare lane 1 with lane 3, and lane 5 with lane 7). Interestingly, the kinase activity of Rad3-E1369K was significantly lower in rad9 cells (compare lane 1 with lane 5). When treated with HU, the kinase activity in wild-type Rad3 did not increase in rad9-∆C and rad9 cells (lanes 4 and 8), suggesting that the HU-activated Rad3 inside the cells, once IPed, returned to the inactive state. However, Rad3-E1369K further increased the kinase activity in rad9 cells (lanes 5 and 6), but not in rad9-∆C cells (lanes 1 and 2), suggesting that the Rad9 C-terminus may suppress the kinase activity of Rad3-E1369K (see Discussion). Together, the results in S9 Fig and Fig 5 clearly showed that the E1369K mutation converts Rad3 into a constitutively active form, and the Rad9 C-terminus may modulate Rad3 kinase activity in both positive and negative manners.

The Mec1-F2244L equivalent mutation F2250L in Rad3 does not rescue rad9-∆C in fission yeast

A recent study in budding yeast identified an F2244L mutation in the DFD motif of the kinase domain in Mec1ATR/Rad3 that increases basal kinase activity 10–20-fold in vitro and promotes survival of cells lacking all three activators in HU and the genotoxin 4-nitroquinolin 1-oxide [16]. The F2244 residue in Mec1 is conserved in the DFN motif of both Rad3 and ATR (S4 Fig A). We made the equivalent mutation, F2250L, in Rad3 and integrated it at the genomic locus using the same method as for rad3-E1369K. After confirming the integration by colony PCR, Western blotting, and sequencing, we crossed the rad3-F2250L mutation into rad9-∆C to investigate whether it could rescue the rad9 phospho-mutant, as rad3-E1369 K does. The drug sensitivity assay clearly showed that the mutation did not rescue rad9-∆C, and the mutation alone minimally sensitized S. pombe to HU and MMS (Fig 6A). Western blotting showed that the rad3-F2250L mutation did not increase Cds1 phosphorylation in HU-treated rad9-∆C cells (Fig 6B). Instead, it moderately reduced Cds1 phosphorylation in rad9 and eliminated the phosphorylation in rad9-∆C cells. The mutation did not affect Mrc1 phosphorylation in HU-treated rad9 cells, but reduced the phosphorylation in HU-treated rad9-∆C cells (Fig 6C). The rad3-F2250L minimally affected Chk1 phosphorylation in MMS-treated rad9 and deletion of the Rad9 C-terminus eliminated the phosphorylation as in rad3-E1369K cells (Fig 6D). These results show that the rad3-F2250L mutation did not increase the kinase activity of Rad3 as in Mec1. Interestingly, the C-terminus of Rad9 can significantly stimulate the kinase activity of Rad3-F2250L in Mrc1 phosphorylation to nearly the wild-type level in the presence of HU. We then examined the kinase activity of Rad3-F2250L in vitro. Western blotting showed that the protein level of Rad3-F2250L was similar to that of Rad3 or Rad3-E1369K (Fig 6E, top panel). Like Fig 5, Rad3-E1369K showed a higher kinase activity than Rad3 in both rad9 and rad9-∆C cells in the absence or presence of HU. Under similar conditions, Rad3-F2250L behaved like wild-type Rad3, although the basal kinase activity was slightly lower, which is consistent with the reduced Cds1 phosphorylation in HU-treated rad9 and rad9-∆C cells (Fig 6B).

thumbnail
Download:
Fig 6. The Mec1-F2244L equivalent mutation F2250L in Rad3 mildly reduces Rad3 kinase activity in vitro and in HU-treated cells.

(A) Drug sensitivities of wild-type and the rad3-F2250L integrant in rad9 or rad9-∆C background were examined by spot assay. (B) Rad3 phosphorylation of Cds1 in the strains used in A was examined by Western blotting as in Fig 1B. (C) Rad3 phosphorylation of Mrc1 was examined by Western blotting using the phospho-specific antibody (top panel) and anti-Mrc1 antibody (middle panel) as in Fig 1C. A section of the Ponceau S-stained membrane is shown for loading. (D) Chk1 phosphorylation in MMS-treated rad3-F2250L in rad9 or rad9-∆C background was examined as in Fig 3B. (E) Wild-type Rad3, Rad3-E1369K, and Rad3-F2250L were IPed from rad9 or rad9-ΔC cells treated with (+) or without (-) HU. The kinase activity of the IPed Rad3 was examined using the Cds1(KD) substrate as in Fig 5.

https://doi.org/10.1371/journal.pgen.1012213.g006

Discussion

The cryo-EM structures of human ATR-ATRIP and budding yeast Mec1-Ddc2 suggest a common feature: the kinase domain is autoinhibited by the PRD via steric hindrance of substrate access [15,32,33]. Although structural evidence is still lacking, it has been proposed that binding of an activation domain to the PRD relieves the autoinhibition, allowing substrate access and realignment of catalytic residues [3]. This activates ATR and Mec1 to initiate the kinase signaling cascades, which, depending on the severity of stress or damage, can be further amplified or suppressed to fine-tune the checkpoint signaling. Consistent with this model, three proteins have been identified in budding yeast that carry the Mec1-activation domain: Ddc1RAD9, Dna2, and Dpb11TopBP1/Rad4 [12,13,39]. Among them, Dpb11 is the main Mec1 activator in response to replication stress or DNA damage and is recruited by phosphorylated Ddc1Rad9 of the 9-1-1 complex. In Xenopus and humans, TopBP1 and ETAA1 have been discovered to carry the ATR-activation domain [8,9,11,31].

Our genetic screen has identified an E1369K mutation in Rad3, the S. pombe ortholog of ATR and Mec1. The mutation restored Cds1 phosphorylation in a Mrc1 phosphorylation-dependent manner and cell survival of rad9 phospho-mutants in HU (Figs 2 and 3). The in vitro kinase assay, together with the in cell data, shows that the mutation converts Rad3 into a constitutively active form, which bypasses the requirement for Rad9 phosphorylation and partially rescues Rad26 mutations defective in Rad3 recruitment. This Rad3 mutation, therefore, raises several interesting questions that require further investigation:

The activation mechanism of Rad3ATR/Mec1

Rad3 robustly phosphorylates Mrc1 in HU without the need for Rad9Ddc1 phosphorylation, the C-terminus of Rad4TopBP1/Dpb11 (Fig 1C), and even the 9-1-1 complex [18], suggesting that this kinase activity of Rad3 does not require an activator in fission yeast. The increased phosphorylation of Mrc1 and Cds1 by Rad3-E1369K in HU-treated rad9-∆C cells (Fig 3A and C) suggests that the charge reversal mutation may increase the binding or proper repositioning of the substrates Mrc1 and Cds1 for more efficient phosphorylation by Rad3, similar to that in mTORC1 and mTORC2 [40,41]. A recent study in budding yeast showed that the F2244L mutation in the DFD motif converts Mec1 into a constitutively active form [16]. However, the equivalent mutation in Rad3 did not rescue rad9-∆C cells and mildly reduced Rad3 kinase activity in vitro and in HU-treated cells (Fig 6), suggesting that the function of the F2244 residue in the Mec1’s DFD motif may not be conserved in Rad3. The first Asp residue of the DFD motif in the activation loop of Mec1 is absolutely conserved for coordinating with Mg2+, whereas the FD residues are less conserved as LG, FN, or FG in PIKKs. The F2244L mutation in Mec1 increases its ATPase activity. However, this increased activity is less efficiently coupled with the phosphor-transfer to the substrate, or kinase activity of the mutant Mec1 [16]. Whether the equivalent mutation F2250L in the DFN motif of Rad3 further decouples the two catalytic processes is unclear and needs further examination. Nonetheless, several lines of evidence support an allosteric activation mechanism in Rad3. First, the E1369K mutation is located in the FAT, not the kinase domain of Rad3. Although not conserved, the equivalent residue Y1623 in ATR is located near the catalytic center, but away from the PRD (S4 Fig B). ATR also autophosphorylates the T1989 residue in the FAT domain to regulate its kinase activity [42,43]. Moreover, autophosphorylation in or near the FAT domain of ATM and DNA-PKcs, two other PIKKs essential for checkpoint signaling and repair of DSBs, respectively, can also regulate their functions [44–47]. Second, the E1369K mutation creates a constitutively active form of Rad3 as demonstrated by the in vitro kinase assay (Figs 5, 6E, and S9). Third, as mentioned above, the robust phosphorylation of Mrc1 by Rad3 occurs in HU-treated rad9-∆C and rad4-∆C cells (Fig 1C), lacking the potential activators Rad9 and Rad4. Rad3-E1369K also hyperphosphorylates Mrc1 in HU-treated rad9-∆C cells (Fig 3C). Similarly, colocalization of Mrc1 and Ddc2 in budding yeast by tethering them at an undamaged chromosomal locus allows Mec1 phosphorylation of Rad53 in the absence of all three activators [48]. Fourth, the activated Rad3 in HU-treated cells, once IPed, returned to inactive status in vitro (Fig 5). Finally, the constitutively active Rad3-E1369K phosphorylates Cds1, not Chk1, in rad9 phospho-mutants (Fig 3A and B), suggesting that proper positioning of Chk1 is necessary for the phosphorylation by Rad3. It is likely that Rad3 can be activated by two mechanisms, one through binding to a Rad3-activator protein that remains to be identified in S. pombe and the other through allosteric regulation via the FAT domain or the HEAT repeats [49], allowing fine-tuning of its kinase signaling in the context of stress management or damage being actively repaired. It would be interesting to investigate whether the allosteric mechanism revealed by the E1369K mutation occurs under physiological conditions. Nevertheless, further biochemical and structural studies may provide more insights into the activation mechanisms of Rad3 and, potentially, for other PIKKs.

The positive and negative regulatory roles of the Rad9Ddc1 C-terminus

That the constitutively active Rad3-E1369K bypasses rad9-∆C to phosphorylate Cds1 suggests that the main function of the Rad9 C-terminus is to promote Rad3 kinase activity. Since Mrc1 is hyper-phosphorylated in rad3-E1369K in rad9-∆C (Fig 3C), the hyperphosphorylation may increase the recruitment of Cds1, leading to the increased Cds1 phosphorylation in HU (Fig 3A). The dependency of Mrc1 phosphorylation on the increased Cds1 phosphorylation in rad3-E1369K supports this possibility (Fig 3D-F). Alternatively, defective Cds1 phosphorylation in rad9-∆C promotes Mrc1 phosphorylation, leading to its hyperphosphorylation in rad3-E1369K, but not in rad3 cells. This notion is consistent with the canonical role of Rad9 C-terminus in stimulating Rad3 kinase signaling. In budding yeast, the C-terminus of Ddc1Rad9 carries a Mec1-activation domain [13], and the phosphorylated Ddc1 C-terminus can further recruit the major Mec1 activator protein Dpb11 [14,39]. Similarly, the phosphorylated Rad9 C-terminus recruits Rad4 in fission yeast and TopBP1 in mammalian cells [23,29]. Consistent with this positive role of the Rad9 C-terminus, less Cds1 phosphorylation was observed in the double mutant of rad9-∆C rad3-E1369K than in rad9 rad3-E1369K cells in HU (Fig 3A). In HU-treated rad3-F2250L cells with lower Rad3 kinase activity, Rad9 C-terminus significantly increases Mrc1 phosphorylation (Fig 6C). Although the function of the recruited Rad4 in promoting Chk1 to be phosphorylated by Rad3 is relatively clear in fission yeast, whether and how Rad4TopBP1/Dpb11 is recruited at the fork for Cds1 phosphorylation remains to be investigated. Surprisingly, in the absence of HU, Rad3-E1369K IPed from rad9-∆C showed a significantly higher in vitro kinase activity than that from wild-type rad9 cells (Fig 5 and Fig 6E), suggesting that the Rad9 C-terminus may induce a long-lasting conformational change in Rad3-E1369K to suppress the constitutively active kinase, which may be harmful to the cell under normal conditions. It would be interesting to examine whether the negative regulatory function of the Rad9 C-terminus on Rad3-E1369K is a direct or an indirect effect via other factors. Nevertheless, although further studies are needed, the regulatory effects of the Rad9 C-terminus may play an important role in fine-tuning Rad3 kinase signaling at the fork.

Different initiation mechanisms of the DRC and the DDC pathways

Although the constitutively active Rad3-E1369K significantly promotes Cds1 phosphorylation in rad9-∆C, it does not stimulate Chk1 phosphorylation in the same rad9 phospho-mutant (Fig 3B), showing that at the DNA damage site, Rad3 kinase signaling is regulated by a different, not a universal, checkpoint initiation mechanism. In support of this notion, we have previously reported that while the cooperation of the RPA-binding domain and the KKRK motif in Rad26 is crucial for the Rad3 signaling at the fork, it is less important at the DNA damage site, particularly at the site of strand breaks [37]. As shown in Fig 1, while Crb253 BP1/Rad9 is required for Chk1 activation, it plays a minimal role in the phosphorylation of Cds1 and Mrc1, suggesting that even in the presence of a constitutively active Rad3, Chk1 has to be properly positioned by the collaboration of the recruited Rad4, Crb2, or other proteins for its phosphorylation. In support of this possibility, previous studies have identified several repair proteins that contribute to checkpoint initiation at the DNA damage site in Xenopus and mammalian cells [50–54]. Whether a repair protein modulates the checkpoint sensor kinase activity in a genetically tractable system such as S. pombe remains to be investigated.

Materials and methods

Yeast strains and plasmids

The S. pombe strains used in the study were cultured in YE6S liquid media containing 0.5% yeast extract, 3% dextrose, with six supplements or in EMM6S media. The yeast strains, plasmids, and PCR primers used in the study are listed in S1-S3 Tables. The mutations were identified by whole genome sequencing (Innomics, Inc.) and confirmed by Sanger sequencing (Retrogen).

Drug sensitivity assays

Sensitivities of wild-type and the mutant S. pombe to HU and MMS were measured using the spot assay or colony recovery as previously described [34,55].

Suppressor screen

rad9 phospho-mutants were cultured in EMM6S liquid media to log phase, harvested, and washed with 50 mM Tris-Maleate buffer, pH 6.0, and resuspended in the same buffer at a cell density of 35.0 OD/ml. A 250 µl aliquot of the cell suspension was incubated with 0.2 mg/ml MNNG (N-methyl-N′-nitro-N-nitrosoguanidine) at 25°C for 90 min [23]. The mutagenized cells were washed twice and incubated in 5 mL of EMM6S medium at 25°C for 3 h to recover. The cells were then spread on YE6S media containing 5 mM HU and phloxine B. The plates were incubated at 30°C for 4–5 days to allow colony formation.

AlphaFold2 modeling

The protein sequences of S. pombe Rad3, Rad3-E1369K, and Rad26 were submitted to the AlphaFold2 server (https://alphafoldserver.com/). The predicted models of Rad3-Rad26 and Rad3-E1369K-Rad26 were analyzed in PyMOL.

IP and co-IP

5 OD logarithmically growing cells were harvested and lysed using mini-bead beater in a buffer containing 25 mM HEPES/NaOH (pH 7.5), 50 mM NaF, 1 mM NaVO4, 10 mM NaP2O7, 40 mM ß-glycerophosphate, 0.1% Tween 20, 0.5% NP-40, and protease inhibitors. The lysates were centrifuged at 16,000 g at 4˚C for 10 min. The resulting supernatant was used as a whole-cell extract. Agarose resin cross-linked to anti-HA or anti-myc antibodies was washed three times with Tris-buffered saline containing 0.05% Tween 20 (TBS-T) before being incubated with the whole-cell extract by rotating at 4˚C for 2 h. The samples were washed three times with TBS-T at 4˚C for 10 min. The IP samples were separated by 8% SDS-PAGE followed by Western blotting.

Western blotting

Rad3-dependent phosphorylation of Cds1-T11 and Mrc1-T645 was analyzed by Western blotting using the phospho-specific antibodies described in previous studies [18]. Mrc1 was detected by a polyclonal antibody on the same membrane [55]. Rad3 was tagged with a 10xMyc epitope at the N-terminus, while Cds1 and Rad26 were tagged with an HA epitope at the C- and N-terminus, respectively, and detected by Western blotting using anti-myc or anti-HA antibodies. Phosphorylation of Chk1 by Rad3 was examined by a mobility shift assay [21]. The large subunit of RPA Ssb1 was detected using an anti-Ssb1 antibody [55]. The blotting signal was detected by electrochemiluminescence using the ChemiDoc XRS Imaging system (BioRad). Signal intensities of the bands were quantified and analyzed by ImageLab (BioRad).

In vitro Rad3 kinase assay

Whole cell extracts were prepared as described above for IP, except that 15 OD cells were harvested. Pre-washed Protein G Dynabeads were incubated with anti-myc antibodies at 25°C for 60 min. The whole-cell extracts were then incubated with the antibodies bound to Dynabeads by rotating at 4˚C for 5 h. After a quick wash in TBS-T containing protease inhibitors, the beads were collected in a magnetic rack, and the IPed Rad3 was examined by the kinase assay using bacterially purified kinase-inactive Cds1(D312E) as the substrate [19,26]. Briefly, a 20 µl kinase reaction mixture of 50 mM Tris-HCl, pH 7.5, 10 mM MgCl2, 2 mM DTT, 0.1 mM EDTA, 0.01% NP-40, 200 µM ATP, and 0.48 µM of Cds1(D312E) was added to the beads. The reactions were incubated at 25°C for 60 min and then stopped by adding SDS gel loading buffer. After heating at 92˚C for 10 min, the sample was analyzed by SDS PAGE. After transferring to a nitrocellulose membrane, Cds1(D312E) was revealed by Ponceau S staining followed by Western blotting using phosphor-specific antibody to detect Cds1-pT11. The upper part of the membrane was used to detect Rad3 or Rad26 using anti-myc or anti-HA antibody, respectively.

DNA pull-down assay

The assay was conducted using the previously described method [38]. Briefly, whole-cell extract was prepared from 15 OD cells using a mini-bead beater and acid-washed glass beads in 2X buffer containing 100 mM HEPES, pH 7.4, 400 mM NaOAc, 100 mM MgOAc, 1 mM EDTA, 1 mM sodium orthovanadate, 0.1% 2-mercaptoethanol, 1% (w/v) Triton X-100, and protease inhibitors. The lysates were clarified at 16,000g, 4˚C, for 5 min. A 72 base pair dsDNA (10 µg) created by annealing P1-Biotin/P2 oligonucleotides was incubated with 4 mg streptavidin-coated magnetic Dynabeads pre-equilibrated in 50 mM HEPES, pH 7.4, containing 1 M NaCl, washed 4x with 50 mM HEPES, pH 7.4, containing 0.15 M NaCl, and finally resuspended in the same buffer at 25°C for 30 min in 400 µl. 10 µl of this suspension was incubated with the whole cell extract at 4°C for 60 min on a shaking platform. The beads were collected in a magnetic rack and washed three times in buffer containing 50 mM HEPES (pH 7.4), 0.1 M NaCl, 1% Triton X-100, 0.1% 2-mercaptothanol, and protease inhibitors. The beads were resuspended in SDS-loading buffer and subjected to 8% SDS-PAGE followed by Western blotting to detect Rad26 and Rad3.

Supporting information

S1 Fig. Schematic for the screening of suppressors of rad9 phospho-mutants.

The HU-sensitive rad9-T412A and rad9-ΔC strains were exposed to the mutagen MNNG (N-methyl-N′-nitro-N-nitrosoguanidine) to achieve approximately 90% killing [23]. The cells were allowed to recover in rich medium for 2–3 h before spreading onto plates containing 5 mM HU and phloxine B, a lethality dye. The HU-resistant colonies were selected, streaked out into single colonies, and tested by spot assays for HU resistance. After backcrossing with the parental rad9-T412A or rad9-ΔC, the suppressors were examined for Cds1 phosphorylation, which identified the L11 suppressor that restored Cds1 phosphorylation and HU-resistance in both rad9-T412A and rad9-∆C mutants. After backcrossing L11 with rad9-∆C, the HU-resistant colonies with restored Cds1 phosphorylation were pooled for purification of genomic DNA and subsequent genome sequencing. The HU-sensitive colonies were similarly pooled for genome sequencing as the reference. The rad3-E1369K mutation identified in L11 by the genome sequencing was then confirmed by Sanger sequencing.

https://doi.org/10.1371/journal.pgen.1012213.s001

(TIF)

S2 Fig. Screening of the L11 suppressor of rad9-T412A with increased Rad3 phosphorylation of Mrc1 and Cds1.

(A) Drug sensitivities of the screened suppressors of the K series isolated in rad9-∆C and the L series in rad9-T412A were determined by spot assay. Wild-type, rad3∆, rad9-ΔC, and rad9-T412A strains were included as controls. (B) Phosphorylation of Cds1 in the screened suppressors was examined by Western blotting using the phospho-specific antibody against Cds1-pT11. Among the suppressors, only L11 rescued the Cds1 phosphorylation in rad9-T412A or rad9-∆C. (C) Chk1 phosphorylation was examined by mobility shift assay. Wild-type S. pombe, rad9-T412A, and the L11 suppressor were treated with (+) or without (-) 0.01% MMS for 90 min. Whole-cell lysates were analysed by SDS-PAGE followed by Western blotting with anti-HA antibodies to detect the C-terminally tagged Chk1 (top panel). A section of the Ponceau S-stained membrane is shown as the loading control. The upper-shifted phosphorylation band was quantified and shown at the bottom as the ratio of phosphorylated Chk1 vs total Chk1. (D) Mrc1 phosphorylation in the L11 suppressor was examined by Western blotting using the phospho-specific antibody against Mrc1-pT645 as in Fig 1C.

https://doi.org/10.1371/journal.pgen.1012213.s002

(TIF)

S3 Fig. The L11 suppressor also rescues rad9-∆C.

(A) The L11 suppressor rescues both rad9-T412A and rad9-∆C in HU and MMS. L11 was crossed into rad9-ΔC, and the drug sensitivities were determined by spot assay, in which a series of 5-fold dilutions of the cells was spotted on plates containing HU or MMS at the indicated concentrations. (B) The L11 rad9-∆C cells were transformed with plasmids expressing Rad4 (also known as Cut5), Suc22, the small subunit of ribonucleotide reductase, Cds1, and Rad9 under their native promoters. The drug sensitivities were examined by the three-spot assay as in (A), except the cells were diluted in a 10-fold series.

https://doi.org/10.1371/journal.pgen.1012213.s003

(TIF)

S4 Fig. Alignment of fission yeast Rad3, budding yeast Mec1, and human ATR, and the locations of the E1369K and F2250L mutations in Rad3.

(A) The primary amino acid sequences of S. pombe Rad3, S. cerevisiae Mec1, and human ATR were aligned by CLUSTALW using MacVector. The less conserved N-terminal HEAT repeat region is not shown. The N-FAT, M-FAT, C-FAT, kinase domain, PRD, and FATC are highlighted by blue, green, purple, and brown lines, respectively [16]. The activation loop, the KαC helix, and the PRD-I in the PRD of the kinase domain are marked by red squares and arrows, respectively. The E1369K mutation in N-FAT and the F2250L mutation in the activation loop of Rad3 are marked in red. (B) Y1623 in ATR, the equivalent residue of Rad3-E1369, is located close to the catalytic centre but distant from the PRD in the cryo-EM structure of ATR-ATRIP [32]. Also highlighted in the ATR-ATRIP structure are the activation loop, PRD-I, and the KαC helix.

https://doi.org/10.1371/journal.pgen.1012213.s004

(TIF)

S5 Fig. Extrachromosomal expression of Rad3-E1369K rescues the drug sensitivities of rad9-∆C by promoting Rad3 phosphorylation of Cds1 and Mrc1.

(A) The drug sensitivities of the double mutant rad9-ΔC Δrad3 expressing wild-type Rad3 or Rad3-E1369K on a vector under the control of the rad3 promoter were examined by spot assay. The double mutant with an empty vector was used as a control. (B) Western blotting confirms similar expression levels of Rad3 and Rad3-E1369K using the anti-myc antibody to detect the N-terminal epitope tag. Asterisk indicates a cross-reacting material. (C) Cds1 phosphorylation was examined in the strains used in (A) by Western blotting before (-) or after (+) HU treatment. (D) Mrc1 phosphorylation was examined by Western blotting in the strains used in (A).

https://doi.org/10.1371/journal.pgen.1012213.s005

(TIF)

S6 Fig. The primary L11 mutant carries a secondary uncharacterized mutation that promotes cell survival in oxidative stress caused by chronic HU exposure.

(A) A schematic of the integration of the rad3-E1369K mutation at the genomic locus. rad3 was tagged with a 10myc epitope at the N-terminus and the nmt1 terminator (nmtT) to replace its own terminator, followed by a kanR marker. The DNA fragment was released from the vector by digestion with SalI and SbfI, gel-purified, and transformed into wild-type S. pombe. G418-resistant colonies were screened by colony PCR and subsequent Western blotting to detect the N-terminal myc tag. The integrated mutation was confirmed by Sanger sequencing. (B) Drug sensitivities of the primary L11 suppressor carrying the rad9-T412A or rad9-ΔC mutation were compared with the rad3-E1369K integrant carrying the rad9-∆C mutation. (C) Tetrad dissection analysis of the crosses between L11 rad9-T412A and rad9-ΔC. Colonies formed on a YE6S plate were replicated onto a YE6S plate lacking adenine to show a 2:2 ratio of the two ade6 alleles, which confirmed the dissection. The colonies were also replicated onto a YE6S plate containing 5 mM HU and phloxine B. Tetrads with three HU-resistant spores were marked by green dashed squares. Dashed lines indicate discontinuity. (D) Antioxidant N-acetyl cysteine significantly increased the HU-resistance in rad3-E1369K integrant carrying the rad9-∆C mutation.

https://doi.org/10.1371/journal.pgen.1012213.s006

(TIF)

S7 Fig. The rad3-E1369K mutation does not rescue rad9∆ and rad1∆ mutants of the 9-1-1 complex, and the 9-1-1 loader rad17∆ mutant in HU.

The integrated rad3-E1369K mutation was crossed into rad9∆, rad1∆, and rad17∆ strains. Four individual colonies from each cross were selected for the HU sensitivity assay. Wild-type, rad3∆, and the parental strains were used as the controls. Rad9 and Rad1 bind to Hus1 to form the 9-1-1 clamp complex, while Rad17 is the loader of the 9-1-1 complex.

https://doi.org/10.1371/journal.pgen.1012213.s007

(TIF)

S8 Fig. The charge reversal E1369K mutation does not affect Rad3 binding to Rad26 and the recruitment of the Rad3-Rad26 complex to DNA.

(A) AlphaFold2 structures of the Rad3-Rad26 heterodimer containing Rad3 (left) or Rad3-E1369K (right). Rad3 is shown in green, while Rad26 is in orange. E1369 resides in a SRESS motif, forming a small loop connecting the α-helices. The mutation alters the local structure, enabling binding of the N-terminus of Rad26 to the Rad3 kinase domain. (B) The E1369K mutation did not affect the interactions between Rad3 and Rad26, and Rad3-Rad26 with Ssb1, the large subunit of RPA. Rad26 was IPed using an anti-HA antibody from whole-cell extracts expressing Rad26 (top panel on the right). The co-IPed Rad3 and Ssb1 were analysed by Western blotting using anti-myc and anti-Ssb1 antibodies, respectively (middle and bottom panels on the right). A 2% portion of the whole cell extract was analysed as input in the left three panels. Asterisk indicates a non-specific band. (C) DNA pull-down assay for Rad3 and Rad3-E1369K in complex with wild-type Rad26 or Rad26 with the indicated mutations. Whole cell extracts were incubated with a 72 bp dsDNA bound to magnetic beads. The beads were washed three times and analysed by Western blotting using anti-HA antibody to detect Rad26 (top panel) and anti-myc antibody to detect Rad3 (middle panel). A portion of the Ponceau S-stained membrane is shown for the loading (bottom).

https://doi.org/10.1371/journal.pgen.1012213.s008

(TIF)

S9 Fig. The IPed Rad3-E1369K showed a constitutively increased kinase activity in vitro regardless of the Rad26 N-terminal mutations.

Rad3 and Rad3-E1369K were IPed from the rad26Δ rad9-ΔC double mutant expressing Rad26 or Rad26 mutants with the indicated mutations. The in vitro Rad3 kinase assay was conducted using the kinase-dead Cds1(D312E) substrate as described in Materials and Methods. Phosphorylated Cds1 was detected using a phospho-specific antibody against Cds1-pT11, quantified, and the results are shown at the bottom in percentages. A portion of the Ponceau S-stained membrane containing Cds1 substrate and the IgG used for IP is shown in the bottom panel.

https://doi.org/10.1371/journal.pgen.1012213.s009

(TIF)

S1 Table. List of S. pombe strains used in this study.

https://doi.org/10.1371/journal.pgen.1012213.s010

(PDF)

S2 Table. List of plasmids used in this study.

https://doi.org/10.1371/journal.pgen.1012213.s011

(PDF)

S3 Table. List of primers used in this study.

https://doi.org/10.1371/journal.pgen.1012213.s012

(PDF)

Acknowledgments

We are thankful to NBRP/YGRC in Japan for S. pombe strains and other members of the Xu lab for their help and support.

References

  1. 1. Hartwell LH, Weinert TA. Checkpoints: controls that ensure the order of cell cycle events. Science. 1989;246(4930):629–34. pmid:2683079
  2. 2. Ciccia A, Elledge SJ. The DNA damage response: making it safe to play with knives. Mol Cell. 2010;40(2):179–204. pmid:20965415
  3. 3. Yates LA, Zhang X, Burgers PM. DNA Damage and Replication Stress Checkpoints. Annu Rev Biochem. 2025;94(1):195–221. pmid:40540755
  4. 4. Yazinski SA, Zou L. Functions, Regulation, and Therapeutic Implications of the ATR Checkpoint Pathway. Annual Review of Genetics. 2016;50:155–73. pmid:27617969
  5. 5. Ellison V, Stillman B. Biochemical characterization of DNA damage checkpoint complexes: clamp loader and clamp complexes with specificity for 5’ recessed DNA. PLoS Biol. 2003;1(2):E33. pmid:14624239
  6. 6. Majka J, Binz SK, Wold MS, Burgers PMJ. Replication protein A directs loading of the DNA damage checkpoint clamp to 5’-DNA junctions. J Biol Chem. 2006;281(38):27855–61. pmid:16864589
  7. 7. Zou L, Liu D, Elledge SJ. Replication protein A-mediated recruitment and activation of Rad17 complexes. Proc Natl Acad Sci U S A. 2003;100(24):13827–32. pmid:14605214
  8. 8. Kumagai A, Lee J, Yoo HY, Dunphy WG. TopBP1 activates the ATR-ATRIP complex. Cell. 2006;124(5):943–55. pmid:16530042
  9. 9. Bass TE, Luzwick JW, Kavanaugh G, Carroll C, Dungrawala H, Glick GG, et al. ETAA1 acts at stalled replication forks to maintain genome integrity. Nat Cell Biol. 2016;18(11):1185–95. pmid:27723720
  10. 10. Lee Y-C, Zhou Q, Chen J, Yuan J. RPA-Binding Protein ETAA1 Is an ATR Activator Involved in DNA Replication Stress Response. Curr Biol. 2016;26(24):3257–68. pmid:27818175
  11. 11. Thada V, Cortez D. Common motifs in ETAA1 and TOPBP1 required for ATR kinase activation. J Biol Chem. 2019;294(21):8395–402. pmid:30940728
  12. 12. Kumar S, Burgers PM. Lagging strand maturation factor Dna2 is a component of the replication checkpoint initiation machinery. Genes Dev. 2013;27(3):313–21. pmid:23355394
  13. 13. Navadgi-Patil VM, Burgers PM. The unstructured C-terminal tail of the 9-1-1 clamp subunit Ddc1 activates Mec1/ATR via two distinct mechanisms. Mol Cell. 2009;36(5):743–53. pmid:20005839
  14. 14. Navadgi-Patil VM, Kumar S, Burgers PM. The unstructured C-terminal tail of yeast Dpb11 (human TopBP1) protein is dispensable for DNA replication and the S phase checkpoint but required for the G2/M checkpoint. J Biol Chem. 2011;286(47):40999–1007. pmid:21956112
  15. 15. Wang X, Ran T, Zhang X, Xin J, Zhang Z, Wu T, et al. 3.9 A structure of the yeast Mec1-Ddc2 complex, a homolog of human ATR-ATRIP. Science. 2017;358(6367):1206–9. pmid:29191911
  16. 16. Tannous EA, Yates LA, Zhang X, Burgers PM. Mechanism of auto-inhibition and activation of Mec1ATR checkpoint kinase. Nat Struct Mol Biol. 2021;28(1):50–61. pmid:33169019
  17. 17. Xu Y, Davenport M, Kelly TJ. Two-stage mechanism for activation of the DNA replication checkpoint kinase Cds1 in fission yeast. Genes Dev. 2006;20(8):990–1003. pmid:16618806
  18. 18. Yue M, Singh A, Wang Z, Xu Y. The phosphorylation network for efficient activation of the DNA replication checkpoint in fission yeast. J Biol Chem. 2011;286(26):22864–74. pmid:21561865
  19. 19. Xu Y-J, Kelly TJ. Autoinhibition and autoactivation of the DNA replication checkpoint kinase Cds1. J Biol Chem. 2009;284(23):16016–27. pmid:19357077
  20. 20. Capasso H, Palermo C, Wan S, Rao H, John UP, O’Connell MJ, et al. Phosphorylation activates Chk1 and is required for checkpoint-mediated cell cycle arrest. J Cell Sci. 2002;115(Pt 23):4555–64. pmid:12415000
  21. 21. Lopez-Girona A, Tanaka K, Chen XB, Baber BA, McGowan CH, Russell P. Serine-345 is required for Rad3-dependent phosphorylation and function of checkpoint kinase Chk1 in fission yeast. Proc Natl Acad Sci U S A. 2001;98(20):11289–94. pmid:11553781
  22. 22. Qu M, Yang B, Tao L, Yates JR 3rd, Russell P, Dong M-Q, et al. Phosphorylation-dependent interactions between Crb2 and Chk1 are essential for DNA damage checkpoint. PLoS Genet. 2012;8(7):e1002817. pmid:22792081
  23. 23. Saka Y, Yanagida M. Fission yeast cut5+, required for S phase onset and M phase restraint, is identical to the radiation-damage repair gene rad4+. Cell. 1993;74(2):383–93. pmid:8343962
  24. 24. Saka Y, Esashi F, Matsusaka T, Mochida S, Yanagida M. Damage and replication checkpoint control in fission yeast is ensured by interactions of Crb2, a protein with BRCT motif, with Cut5 and Chk1. Genes Dev. 1997;11(24):3387–400. pmid:9407031
  25. 25. Lin S-J, Wardlaw CP, Morishita T, Miyabe I, Chahwan C, Caspari T, et al. The Rad4(TopBP1) ATR-activation domain functions in G1/S phase in a chromatin-dependent manner. PLoS Genet. 2012;8(6):e1002801. pmid:22761595
  26. 26. Yue M, Zeng L, Singh A, Xu Y. Rad4 mainly functions in Chk1-mediated DNA damage checkpoint pathway as a scaffold protein in the fission yeast Schizosaccharomyces pombe. PLoS One. 2014;9(3):e92936. pmid:24663817
  27. 27. de Bruin RAM, Kalashnikova TI, Chahwan C, McDonald WH, Wohlschlegel J, Yates J 3rd, et al. Constraining G1-specific transcription to late G1 phase: the MBF-associated corepressor Nrm1 acts via negative feedback. Mol Cell. 2006;23(4):483–96. pmid:16916637
  28. 28. Ivanova T, Alves-Rodrigues I, Gómez-Escoda B, Dutta C, DeCaprio JA, Rhind N, et al. The DNA damage and the DNA replication checkpoints converge at the MBF transcription factor. Mol Biol Cell. 2013;24(21):3350–7. pmid:24006488
  29. 29. Furuya K, Poitelea M, Guo L, Caspari T, Carr AM. Chk1 activation requires Rad9 S/TQ-site phosphorylation to promote association with C-terminal BRCT domains of Rad4TOPBP1. Genes Dev. 2004;18(10):1154–64. pmid:15155581
  30. 30. Qu M, Rappas M, Wardlaw CP, Garcia V, Ren J-Y, Day M, et al. Phosphorylation-dependent assembly and coordination of the DNA damage checkpoint apparatus by Rad4(TopBP1). Mol Cell. 2013;51(6):723–36. pmid:24074952
  31. 31. Mordes DA, Glick GG, Zhao R, Cortez D. TopBP1 activates ATR through ATRIP and a PIKK regulatory domain. Genes Dev. 2008;22(11):1478–89. pmid:18519640
  32. 32. Wang G, Wang P, Zheng Z, Zhang Q, Xu C, Xu X, et al. Molecular architecture and inhibition mechanism of human ATR-ATRIP. Sci Bull (Beijing). 2025;70(13):2137–46. pmid:40379520
  33. 33. Rao Q, Liu M, Tian Y, Wu Z, Hao Y, Song L, et al. Cryo-EM structure of human ATR-ATRIP complex. Cell Res. 2018;28(2):143–56. pmid:29271416
  34. 34. Dev K, Yurtsever I, Bhadra S, Guduri YA, Davi K, Xu Y-J. Dissecting the Cell-Killing Mechanisms of Hydroxyurea Using Spot Assays. Methods Mol Biol. 2025;2862:267–76. pmid:39527207
  35. 35. Singh A, Xu Y-J. Heme deficiency sensitizes yeast cells to oxidative stress induced by hydroxyurea. J Biol Chem. 2017;292(22):9088–103. pmid:28377506
  36. 36. Xu Y-J, Singh A, Alter GM. Hydroxyurea Induces Cytokinesis Arrest in Cells Expressing a Mutated Sterol-14α-Demethylase in the Ergosterol Biosynthesis Pathway. Genetics. 2016;204(3):959–73. pmid:27585850
  37. 37. Xu Y-J, Gao A, Dev K, Zheng Y, Alyahya MY, Pasam S, et al. The RPA-binding domain and the KKRK motif in Rad26ATRIP cooperate at the perturbed DNA replication fork for initiating checkpoint signalling. PLoS Genet. 2026;22(2):e1012052. pmid:41729970
  38. 38. Rouse J, Jackson SP. Lcd1p recruits Mec1p to DNA lesions in vitro and in vivo. Mol Cell. 2002;9(4):857–69. pmid:11983176
  39. 39. Navadgi-Patil VM, Burgers PM. Yeast DNA Replication Protein Dpb11 Activates the Mec1/ATR Checkpoint Kinase. J Biol Chem. 2008;283(51):35853–9.
  40. 40. Yang H, Jiang X, Li B, Yang HJ, Miller M, Yang A, et al. Mechanisms of mTORC1 activation by RHEB and inhibition by PRAS40. Nature. 2017;552(7685):368–73. pmid:29236692
  41. 41. Taylor MS, Chen M, Hancock M, Wranik M, Miller BD, O’Meara TR, et al. Structural basis for the recruitment and selective phosphorylation of Akt by mTORC2. Science. 2026;391(6789):eadv7111. pmid:41308123
  42. 42. Liu S, Shiotani B, Lahiri M, Maréchal A, Tse A, Leung CCY, et al. ATR autophosphorylation as a molecular switch for checkpoint activation. Mol Cell. 2011;43(2):192–202. pmid:21777809
  43. 43. Nam EA, Zhao R, Glick GG, Bansbach CE, Friedman DB, Cortez D. Thr-1989 phosphorylation is a marker of active ataxia telangiectasia-mutated and Rad3-related (ATR) kinase. J Biol Chem. 2011;286(33):28707–14. pmid:21705319
  44. 44. Bakkenist CJ, Kastan MB. DNA damage activates ATM through intermolecular autophosphorylation and dimer dissociation. Nature. 2003;421(6922):499–506. pmid:12556884
  45. 45. Chen X, Xu X, Chen Y, Cheung JC, Wang H, Jiang J, et al. Structure of an activated DNA-PK and its implications for NHEJ. Mol Cell. 2021;81(4):801-810.e3. pmid:33385326
  46. 46. Chan DW, Chen BP-C, Prithivirajsingh S, Kurimasa A, Story MD, Qin J, et al. Autophosphorylation of the DNA-dependent protein kinase catalytic subunit is required for rejoining of DNA double-strand breaks. Genes Dev. 2002;16(18):2333–8. pmid:12231622
  47. 47. Cui X, Yu Y, Gupta S, Cho Y-M, Lees-Miller SP, Meek K. Autophosphorylation of DNA-dependent protein kinase regulates DNA end processing and may also alter double-strand break repair pathway choice. Mol Cell Biol. 2005;25(24):10842–52. pmid:16314509
  48. 48. Berens TJ, Toczyski DP. Colocalization of Mec1 and Mrc1 is sufficient for Rad53 phosphorylation in vivo. Mol Biol Cell. 2012;23(6):1058–67. pmid:22298423
  49. 49. Luzwick JW, Nam EA, Zhao R, Cortez D. Mutation of serine 1333 in the ATR HEAT repeats creates a hyperactive kinase. PLoS One. 2014;9(6):e99397. pmid:24901225
  50. 50. Shell SM, Li Z, Shkriabai N, Kvaratskhelia M, Brosey C, Serrano MA, et al. Checkpoint kinase ATR promotes nucleotide excision repair of UV-induced DNA damage via physical interaction with xeroderma pigmentosum group A. J Biol Chem. 2009;284(36):24213–22. pmid:19586908
  51. 51. Wang Y, Qin J. MSH2 and ATR form a signaling module and regulate two branches of the damage response to DNA methylation. Proc Natl Acad Sci U S A. 2003;100(26):15387–92. pmid:14657349
  52. 52. Yoshioka K, Yoshioka Y, Hsieh P. ATR kinase activation mediated by MutSalpha and MutLalpha in response to cytotoxic O6-methylguanine adducts. Mol Cell. 2006;22(4):501–10. pmid:16713580
  53. 53. Liu Y, Fang Y, Shao H, Lindsey-Boltz L, Sancar A, Modrich P. Interactions of human mismatch repair proteins MutSalpha and MutLalpha with proteins of the ATR-Chk1 pathway. J Biol Chem. 2010;285(8):5974–82. pmid:20029092
  54. 54. Lin Y, Li J, Zhao H, McMahon A, McGhee K, Yan S. APE1 recruits ATRIP to ssDNA in an RPA-dependent and -independent manner to promote the ATR DNA damage response. Elife. 2023;12:e82324. pmid:37216274
  55. 55. Xu Y-J, Bhadra S, Mahdi ATA, Dev K, Yurtsever I, Nakamura TM. Comprehensive mutational analysis of the checkpoint signaling function of Rpa1/Ssb1 in fission yeast. PLoS Genet. 2023;19(5):e1010691. pmid:37200372