Escape of Sgs1 from Rad9 inhibition reduces the requirement for Sae2 and functional MRX in DNA end resection (original) (raw)
Sgs1‐ss suppresses the sensitivity to DNA damaging agents of sae2∆ and mre11‐nd mutants
SAE2 deletion causes hypersensitivity to CPT_,_ which creates replication‐associated DSBs. The lack of Ku suppresses CPT hypersensitivity of sae2∆ mutants, and this rescue requires Exo1 18, 19, indicating that Ku prevents Exo1 from initiating DSB resection. To identify other possible pathways bypassing Sae2 function in DSB resection, we searched for extragenic mutations that suppress the CPT sensitivity of sae2∆ cells. CPT‐resistant sae2∆ candidates were crossed to each other and to the wild‐type strain to identify, by tetrad analysis, 15 single‐gene suppressor mutants that fell into 11 distinct allelism groups. Genome sequencing of the five non‐allelic suppressor clones that stood from the others for the best suppression phenotype identified single‐base pair substitutions either in the TOP1 gene, encoding the CPT target topoisomerase I, or in the PDR3, PDR10 and SAP185 genes, which encode for proteins involved in multi‐drug resistance. The mutation responsible for the suppression in the fifth clone was a single‐base pair substitution in the SGS1 gene (SGS1‐ss), causing the amino acid change G1298R in the HRDC domain that is conserved in the RecQ helicase family. The identity of the genes that are mutated in the six remaining suppressor clones remained to be determined.
The SGS1‐ss allele suppressed the sensitivity of the sae2∆ mutant not only to CPT, but also to phleomycin (phleo) and MMS, resulting in almost wild‐type survival of sae2∆ SGS1‐ss cells treated with these drugs (Fig 1A). The ability of Sgs1‐ss to suppress the sensitivity of sae2∆ to genotoxic agents was dominant, as sae2∆/sae2∆ SGS1/SGS1‐ss diploid cells were less sensitive to CPT, phleomycin and MMS compared to sae2∆/sae2∆ SGS1/SGS1 diploid cells (Fig 1B).
Figure 1
The alternative text for this image may have been generated using AI.
Suppression of the sensitivity to genotoxic agents of sae2∆ and mre11 nuclease‐defective mutants by Sgs1‐ss
A, B Exponentially growing cells were serially diluted (1:10), and each dilution was spotted out onto YEPD plates with or without camptothecin (CPT), phleomycin or MMS.
C, D Meiotic tetrads were dissected on YEPD plates that were incubated at 25°C, followed by spore genotyping.
E Exponentially growing cells were serially diluted (1:10), and each dilution was spotted out onto YEPD plates with or without CPT, phleomycin or MMS.
F, G Meiotic tetrads were dissected on YEPD plates that were incubated at 25°C, followed by spore genotyping.
Besides providing the endonuclease activity to initiate DSB resection, MRX also promotes stable association of Exo1, Sgs1 and Dna2 at the DSB ends 20, thus explaining the severe resection defect of cells lacking the MRX complex compared to cells lacking either Sae2 or the Mre11 nuclease activity. Sgs1‐ss suppressed the hypersensitivity to genotoxic agents of mre11‐H125N cells, which were specifically defective in Mre11 nuclease activity (Fig 1A). By contrast, mre11∆ SGS1‐ss double‐mutant cells were as sensitive to genotoxic agents as the mre11∆ single mutant (Fig 1A). Altogether, these findings indicate that Sgs1‐ss can bypass the requirement of Sae2 or MRX nuclease activity for survival to genotoxic agents, but it still requires the physical integrity of the MRX complex to exert its function.
Sgs1 promotes DSB resection by acting as a helicase 4, 5, prompting us to investigate whether Sgs1‐ss requires its helicase activity to exert the suppression effect. Both the lack of Sgs1 and its helicase‐dead Sgs1‐hd variant, carrying the K706A amino acid substitution 21, impaired viability of sae2∆ cells 5 (Fig 1C). This synthetic sickness is likely due to poor DSB resection, as it is known to be alleviated by making DNA ends accessible to the Exo1 nuclease 18, 19. The K706A substitution was therefore introduced in Sgs1‐ss, thus generating the Sgs1‐hd‐ss variant, and meiotic tetrads from diploid strains double heterozygous for sae2∆ and sgs1‐hd‐ss were analysed for spore viability on YEPD plates. All sae2∆ sgs1‐hd‐ss double‐mutant spores formed much smaller colonies than each single‐mutant spore (Fig 1D), with a colony size similar to that obtained from sae2∆ sgs1‐hd double‐mutant spores (Fig 1C). Thus, Sgs1‐ss appears to require its helicase activity to suppress the lack of Sae2 function.
Suppression of sae2∆ by Sgs1‐ss requires Dna2, but not Exo1
The ssDNA formed by Sgs1 unwinding is degraded by the nuclease Dna2, which acts in DSB resection in a parallel pathway with respect to Exo1 5. Thus, we asked whether the suppression of sae2∆ hypersensitivity to DNA damaging agents by Sgs1‐ss requires Exo1 and/or Dna2. Although the lack of Exo1 exacerbated the sensitivity of sae2∆ cells to some DNA damaging agents (Fig 1E), the SGS1‐ss allele was still capable to suppress the sensitivity to CPT, phleomycin and MMS of sae2∆ exo1∆ double‐mutant cells (Fig 1E), indicating the suppression of sae2∆ by Sgs1‐ss is independent of Exo1.
As DNA2 is essential for cell viability, dna2∆ cells were kept viable by the pif1‐M2 mutation, which impairs the ability of Pif1 to promote formation of long flaps that are substrates for Dna2 22. Diploids homozygous for the pif1‐M2 mutation and heterozygous for sae2∆, dna2∆ and SGS1‐ss were generated, followed by sporulation and tetrads dissection. No viable sae2∆ dna2∆ pif1‐M2 cells could be recovered, and the presence of the SGS1‐ss allele did not restore viability of sae2∆ dna2∆ pif1‐M2 triple‐mutant spores (Fig 1F). By contrast, tetrads from a diploid homozygous for the pif1‐M2 mutation and heterozygous for sae2∆, dna2∆ and ku70∆ showed that the lack of Ku70, which relieved Exo1 inhibition 18, 19, restored viability of sae2∆ dna2∆ pif1‐M2 spores (Fig 1G). These findings indicate that Sgs1‐ss requires Dna2 to bypass Sae2 requirement.
Sgs1‐ss suppresses the adaptation defect of sae2∆ cells
A single irreparable DSB triggers a checkpoint‐mediated cell cycle arrest. Yeast cells can escape an extended checkpoint arrest and resume cell cycle progression even with an unrepaired DSB (adaptation) 23, 24. Sae2 lacking cells, like other resection deficient mutants, fail to turn off the checkpoint triggered by an unrepaired DSB and remain arrested at G2/M as large budded cells 12, 25–27. To investigate whether Sgs1‐ss suppresses the adaptation defect of sae2∆ cells, we used JKM139 derivative strains carrying the HO endonuclease gene under the control of a galactose‐inducible promoter. Galactose addition leads to generation at the MAT locus of a single DSB that cannot be repaired by HR, because the homologous donor loci HML or HMR are deleted 23. When G1‐arrested cell cultures were spotted on galactose‐containing plates, sae2∆ SGS1‐ss cells formed microcolonies with more than two cells more efficiently than sae2∆ cells, which were still arrested at the two‐cell dumbbell stage after 24 h (Fig 2A). Checkpoint activation was monitored also by following Rad53 phosphorylation, which is required for Rad53 activation and is detectable as a decrease of its electrophoretic mobility. When galactose was added to exponentially growing cell cultures of the same strains, sae2∆ and sae2∆ SGS1‐ss mutant cells showed similar amounts of phosphorylated Rad53 after HO induction (Fig 2B), indicating that Sgs1‐ss did not affect checkpoint activation. However, Rad53 phosphorylation decreased in sae2∆ SGS1‐ss double‐mutant cells within 12–14 h after galactose addition, whereas it persisted longer in sae2∆ cells that were defective in re‐entering the cell cycle (Fig 2B). Thus, Sgs1‐ss suppresses the inability of sae2∆ cells to turn off the checkpoint in the presence of an unrepaired DSB.
Figure 2
The alternative text for this image may have been generated using AI.
Suppression of the adaptation defect of sae2∆ cells by Sgs1‐ss
A. YEPR G1‐arrested cell cultures of wild‐type JKM139 and otherwise isogenic derivative strains were plated on galactose‐containing plates (time zero). At the indicated time points, 200 cells for each strain were analysed to determine the frequency of large budded cells and of cells forming microcolonies of more than two cells. The mean values from three independent experiments are represented (n = 3).
B. Exponentially growing YEPR cultures of the strains in (A) were transferred to YEPRG (time zero), followed by Western blot analysis with anti‐Rad53 antibodies.
C. ChIP analysis. Exponentially growing YEPR cell cultures of JKM139 derivative strains were transferred to YEPRG, followed by ChIP analysis of the recruitment of Mre11–Myc at the indicated distance from the HO‐cut compared to untagged Mre11 (no tag). In all diagrams, the ChIP signals were normalized for each time point to the corresponding input signal. The mean values are represented with error bars denoting s.d. (n = 3). *P < 0.01, _t_‐test.
The adaptation defect of _sae2_Δ cells has been proposed to be due to an increased persistence at DSBs of the MRX complex, which in turn causes unscheduled Tel1 activation 26, 27. We then asked by chromatin immunoprecipitation (ChIP) and quantitative real‐time PCR (qPCR) analysis whether Sgs1‐ss can reduce the binding of MRX to the DSB ends in _sae2_Δ cells. When HO was induced in exponentially growing cells, the amount of Mre11 bound at the HO‐induced DSB end was lower in sae2∆ SGS1‐ss than in sae2∆ cells (Fig 2C). As MRX persistence at the DSB in sae2∆ cells has been proposed to be due to defective DSB resection, this finding suggests that Sgs1‐ss suppresses the resection defect of _sae2_Δ cells.
Sgs1‐ss suppresses the resection defect of sae2∆ cells
To investigate whether Sgs1‐ss suppresses the sensitivity to genotoxic agents and the adaptation defect of sae2∆ cells by restoring DSB resection, we used JKM139 derivative strains to monitor directly generation of ssDNA at the DSB ends 23. Because ssDNA is resistant to cleavage by restriction enzymes, we directly monitored ssDNA formation at the irreparable HO‐cut by following the loss of SspI restriction fragments after galactose addition by Southern blot analysis under alkaline conditions, using a single‐stranded probe that anneals to the 3′ end at one side of the break (Fig 3A). Resection in sae2∆ SGS1‐ss cells was markedly increased compared to sae2∆ cells, indicating that Sgs1‐ss suppresses the resection defect caused by the lack of Sae2 (Fig 3B and C).
Figure 3
The alternative text for this image may have been generated using AI.
Sgs1‐ss suppresses the resection defect of sae2∆ cells
A. Method to measure double‐strand break (DSB) resection. Gel blots of SspI‐digested genomic DNA separated on alkaline agarose gel were hybridized with a single‐stranded MAT probe (ss probe) that anneals to the unresected strand. 5′–3′ resection progressively eliminates SspI sites (S), producing larger SspI fragments (r1 through r7) detected by the probe.
B. DSB resection. YEPR exponentially growing cell cultures of JKM139 derivative strains were transferred to YEPRG at time zero. Genomic DNA was analysed for ssDNA formation at the indicated times after HO induction as described in (A).
C. Densitometric analyses. The experiment as in (B) has been independently repeated three times, and the mean values are represented with error bars denoting s.d. (n = 3).
D. DSB repair by single‐strand annealing (SSA). In YMV45 strain, the HO‐cut site is flanked by homologous leu2 sequences that are 4.6 kb apart. HO‐induced DSB formation results in generation of 12‐ and 2.5‐kb DNA fragments (HO‐cut) that can be detected by Southern blot analysis with a LEU2 probe of KpnI‐digested genomic DNA. DSB repair by SSA generates an 8‐kb fragment (product).
E. Densitometric analysis of the product band signals. The intensity of each band was normalized with respect to a loading control (not shown). The mean values are represented with error bars denoting s.d. (n = 3).
Repair of a DSB flanked by direct repeats occurs primarily by single‐strand annealing (SSA), which requires nucleolytic degradation of the 5′ DSB ends to reach the complementary DNA sequences that can then anneal 28. To assess whether the Sgs1‐ss‐mediated suppression of the resection defect caused by the lack of Sae2 was physiologically relevant, we asked whether Sgs1‐ss suppresses the SSA defect of sae2∆ cells. To this end, we introduced the SGS1‐ss allele in YMV45 strain, which carries two tandem leu2 repeats located 4.6 kb apart, with a HO recognition site adjacent to one of the repeats 28. This strain also harbours a GAL‐HO construct for galactose‐inducible HO expression. As expected, accumulation of the repair product was reduced in sae2∆ compared to wild‐type cells, whereas it occurred with almost wild‐type kinetics in sae2∆ SGS1‐ss double‐mutant cells (Fig 3D and E), indicating that Sgs1‐ss improves SSA‐mediated DSB repair in the absence of Sae2.
Altogether, these findings indicate that Sgs1‐ss suppresses both the sensitivity to genotoxic agents of sae2∆ cells and the MRX persistence at DSBs by restoring DSB resection. Interestingly, the effects of the SGS1‐ss mutation are opposite to those of the separation‐of‐function sgs1‐D664∆ allele, which specifically impairs viability of sae2∆ cells and DSB resection without affecting other Sgs1 functions 29.
Sgs1‐ss accelerates DSB resection by escaping Rad9 inhibition
The Sgs1‐ss mutant variant can bypass Sae2 requirement in initiation of DSB resection either because it allows Dna2 to substitute for Sae2/MRX endonuclease activity or because it increases the resection efficiency. To distinguish between these two possibilities, we asked whether Sgs1‐ss could bypass Sae2 requirement in resecting meiotic DSBs, where the Sae2/MRX‐mediated endonucleolytic cleavage is absolutely required to initiate DSB resection by allowing the removal of Spo11 from the DSB ends 15, 16. A _sae2_Δ/_sae2_Δ SGS1‐ss/SGS1‐ss diploid strain was constructed and its kinetics of processing/repair of meiotic DSBs generated at the THR4 hotspot was compared to those of a _sae2_Δ/_sae2_Δ diploid. DSBs disappeared in both wild‐type and SGS1‐ss/SGS1‐ss cells about 4 h after transfer to sporulation medium, while they persisted until the end of the experiment in both _sae2_Δ/_sae2_Δ and _sae2_Δ/_sae2_Δ SGS1‐ss/SGS1‐ss diploid cells (Supplementary Fig S1). Thus, Sgs1‐ss cannot substitute the endonucleolytic clipping by Sae2/MRX when this is absolutely required to initiate DSB resection.
Interestingly, the Sgs1‐ss mutant variant accelerates both DSB resection and SSA compared to wild‐type Sgs1 (Fig 3B–E), suggesting that Sgs1‐ss might increase the resection efficiency by escaping the effect of negative regulators of this process. In particular, Rad9 provides a barrier to resection through an unknown mechanism 13, 14. As shown in Fig 4A and B, both SGS1‐ss and rad9∆ mutant cells accumulated the resection products more efficiently than wild‐type cells, and the presence of Sgs1‐ss did not accelerate further the generation of ssDNA in rad9∆ cells. Thus, the lack of Rad9 and the presence of Sgs1‐ss appear to increase the efficiency of DSB resection through the same mechanism. Furthermore, cells lacking Rad9 displayed sensitivity to CPT and phleomycin (Fig 4C). Consistent with the finding that the SGS1‐ss and rad9∆ alleles affect the same process, rad9∆ was epistatic to SGS1‐ss with respect to the survival to genotoxic agents, as sae2∆ rad9∆ SGS1‐ss cells were as sensitive to CPT and phleomycin as sae2∆ rad9∆ and rad9∆ cells (Fig 4C).
Figure 4
The alternative text for this image may have been generated using AI.
Double‐strand break (DSB) resection is accelerated by the same mechanism in SGS1‐ss and rad9∆ cells
A. DSB resection. YEPR exponentially growing cell cultures of JKM139 derivative strains were transferred to YEPRG at time zero. Genomic DNA was analysed for ssDNA formation as described in Fig 3A.
B. Densitometric analyses. The experiment as in (A) has been independently repeated three times, and the mean values are represented with error bars denoting s.d. (n = 3).
C. Exponentially growing cells were serially diluted (1:10), and each dilution was spotted out onto YEPD plates with or without camptothecin (CPT) or phleomycin.
D. DSB resection. HO was induced at time zero in α‐factor‐arrested JKM139 derivative cells that were kept arrested in G1 with α‐factor throughout the experiment. Genomic DNA was analysed for ssDNA formation as described in Fig 3A.
E. Densitometric analyses. The experiment as in (D) has been independently repeated three times, and the mean values are represented with error bars denoting s.d. (n = 3).
Double‐strand break resection in the G1 phase of the cell cycle is specifically inhibited by the Ku complex, whose lack allows nucleolytic processing in G1 cells independently of Cdk1 activity 30. RAD9 deletion does not allow DSB resection in G1, but it enhances resection in G1‐arrested ku∆ cells 31, indicating that Rad9 inhibits DSB resection in G1, but this function becomes apparent only when Ku is absent. To investigate whether Sgs1‐ss was capable to counteract the inhibitory function of Rad9 in G1, we monitored DSB resection in SGS1‐ss and ku70∆ SGS1‐ss cells that were kept arrested in G1 by α‐factor during HO induction. Consistent with the requirement of Cdk1 activity for efficient DSB resection, the 3′‐ended resection products were barely detectable in wild‐type G1 cells, whereas their amount increased in ku70∆ G1 cells that, as previously reported 30, accumulated mostly 1.7‐, 3.5‐ and 4.7‐kb ssDNA products (r1, r2, r3) (Supplementary Fig S2). By contrast, DSB resection in SGS1‐ss cells was undistinguishable from that observed in wild‐type cells (Supplementary Fig S2), indicating that Sgs1‐ss does not allow DSB resection in G1. Furthermore, while RAD9 deletion enhanced the resection efficiency of ku70∆ G1 cells, G1‐arrested ku70∆ and ku70∆ SGS1‐ss cells accumulated resection products with similar kinetics (Fig 4D and E). Altogether, these findings indicate that Sgs1‐ss is not capable to allow DSB resection in G1 either in the presence or in the absence of Ku. As Sgs1‐ss function in DSB resection depends on Dna2, whose activity requires Cdk1‐mediated phosphorylation 8, the inability of Sgs1‐ss to overcome both Ku‐ and Rad9‐mediated inhibition in G1 may be due to the requirement of Cdk1 activity to support Dna2 and therefore Sgs1‐ss function in DSB resection.
Rapid DSB resection in rad9∆ cells depends mainly on Sgs1
Generation of ssDNA at uncapped telomeres in rad9∆ cells has been shown to be more dependent on Dna2/Sgs1 than on Exo1 32. This observation, together with the finding that SGS1‐ss does not accelerate further the generation of ssDNA in rad9∆ cells (Fig 4A and B), raises the possibility that Rad9 inhibits DSB resection by limiting Sgs1 activity and that the Sgs1‐ss variant can escape this inhibition. We tested this hypothesis by investigating the contribution of Sgs1 and Exo1 to the accelerated DSB resection displayed by rad9∆ cells. As shown in Fig 5A and B, sgs1∆ was epistatic to rad9∆ with respect to DSB resection, as sgs1∆ rad9∆ double‐mutant and sgs1∆ single‐mutant cells resected the HO‐induced DSB with similar kinetics. By contrast, DSB resection in exo1∆ rad9∆ cells was more efficient than in exo1∆ cells, although it was delayed compared to rad9∆ cells (Fig 5C and D). Thus, the rapid resection in the absence of Rad9 depends mainly on Sgs1, although also Exo1 contributes to resect the DSB in the absence of Rad9. Consistent with the finding that Sgs1‐ss overrides Rad9 inhibition, SGS1‐ss exo1∆ cells resected the DSB with kinetics similar to that of rad9∆ exo1∆ cells (Supplementary Fig S3).
Figure 5
The alternative text for this image may have been generated using AI.
Rapid resection in rad9∆ cells depends mainly on Sgs1
A. Double‐strand break (DSB) resection. YEPR exponentially growing cell cultures of JKM139 derivative strains were transferred to YEPRG at time zero. Genomic DNA was analysed for ssDNA formation as described in Fig 3A.
B. Densitometric analyses. The experiment as in (A) has been independently repeated three times, and the mean values are represented with error bars denoting s.d. (n = 3).
C. DSB resection. The experiment was performed as in (A).
D. Densitometric analyses. The experiment as in (C) has been independently repeated three times, and the mean values are represented with error bars denoting s.d. (n = 3).
Rad9 inhibits DSB resection by limiting Sgs1 association at DNA breaks
If loss of end protection by Rad9 allowed Sgs1 to initiate DSB resection, which normally requires Sae2, then RAD9 deletion, like Sgs1‐ss, should suppress the resection defect of sae2∆ cells. Indeed, DSB resection in sae2∆ rad9∆ cells was as fast as in rad9∆ cells, which resected the DSB more efficiently than wild‐type and sae2∆ cells (Fig 6A and B), indicating that the lack of Rad9 bypasses Sae2 function in DSB resection.
Figure 6
The alternative text for this image may have been generated using AI.
Rad9 inhibits Sgs1 association at the double‐strand breaks (DSBs)
A. DSB resection. YEPR exponentially growing cell cultures of JKM139 derivative strains were transferred to YEPRG at time zero. Genomic DNA was analysed for ssDNA formation as described in Fig 3A.
B. Densitometric analyses. The experiment as in (A) has been independently repeated three times, and the mean values are represented with error bars denoting s.d. (n = 3).
C. ChIP analysis. Exponentially growing YEPR cell cultures of JKM139 derivative strains were transferred to YEPRG, followed by ChIP analysis of the recruitment of Sgs1‐HA and Sgs1‐ss‐HA at the indicated distance from the HO‐cut compared to untagged Sgs1 (no tag). In all diagrams, the ChIP signals were normalized for each time point to the corresponding input signal. The mean values are represented with error bars denoting s.d. (n = 3). *P < 0.01, _t_‐test.
D. ChIP analysis in G1‐arrested cells. As in (C), but showing ChIP analysis of the recruitment of Sgs1‐HA and Sgs1‐ss‐HA in cells that were kept arrested in G1 by α‐factor. The mean values are represented with error bars denoting s.d. (n = 3). *P < 0.01, _t_‐test.
E. ChIP analysis in G1‐arrested cells. As in (C), but showing ChIP analysis of the recruitment of Exo1–Myc in cells that were kept arrested in G1 by α‐factor. The mean values are represented with error bars denoting s.d. (n = 3). *P < 0.01, _t_‐test.
We then asked by ChIP and qPCR analysis whether Rad9 limits Sgs1 activity by regulating Sgs1 binding/persistence to the DSB ends. When HO was induced in exponentially growing cells, the amount of Sgs1 bound at the HO‐induced DSB was higher in rad9∆ than in wild‐type cells (Fig 6C), indicating that Rad9 counteracts Sgs1 recruitment to the DSB. Interestingly, the Sgs1‐ss variant was recruited at the DSB with equivalent efficiencies in both exponentially growing wild‐type and rad9∆ cells (Fig 6C). These differences were not due to different resection kinetics, as we obtained similar results also when the HO‐induced DSB was generated in G1‐arrested cells (Fig 6D), which resected the DSB very poorly due to the low Cdk1 activity 6. Interestingly, the amount of Sgs1‐ss bound to the DSB was higher than the amount of wild‐type Sgs1 in rad9∆ cells (Fig 6C and D), suggesting that Sgs1‐ss has a higher intrinsic ability to bind/persist at the DSB. Altogether, these results indicate that Rad9 limits the association of Sgs1 to the DSB ends and that the Sgs1‐ss variant escapes this inhibition possibly because it binds more tightly the DSB. Interestingly, the robust association of Sgs1‐ss to the DSB in G1‐arrested cells (low Cdk1 activity) did not result in DSB resection (Supplementary Fig S2) possibly because Sgs1 acts in DSB resection together with Dna2, whose activity requires Cdk1‐mediated phosphorylation 8. Consistent with a contribution of Exo1 in promoting DSB resection in the absence of Rad9, rad9∆ cells showed an increased Exo1 recruitment to the DSB compared to wild‐type cells (Fig 6E).
In summary, we show that Rad9 increases the requirement for the MRX/Sae2 activities in DSB resection by inhibiting the action of the Sgs1/Dna2 long‐range resection machinery. Extensive resection in Rad9‐deficient cells is mainly dependent on Sgs1, whose recruitment at DSBs is inhibited by Rad9. By contrast, Sgs1‐ss, which suppresses the resection defect of sae2∆ cells, is robustly associated with the DSB ends both in the presence and in the absence of Rad9 and resects the DSB more efficiently than wild‐type Sgs1. These findings indicate that Rad9 inhibits the activity of Sgs1/Dna2 by limiting Sgs1 binding/persistence at DSB ends and that the Sgs1‐ss mutant variant escapes this inhibition possibly because it is more tightly bound to DNA. Thus, while Ku increases the requirement for the MRX/Sae2 activities in DSB resection by inhibiting preferentially Exo1 20, Rad9 mainly restricts the action of Sgs1/Dna2. As MRX and Sae2 are especially important for initial processing of DNA ends that contain adducts, the Rad9‐ and Ku‐mediated inhibitions of Sgs1/Dna2 and Exo1 activities in initiating DSB resection ensure that all DSBs are processed in a similar manner independently of their nature.