The yeast Xrs2 complex functions in S phase checkpoint regulation

Abstract

The Nbs1 complex is an evolutionarily conserved multisubunit nuclease composed of the Mre11, Rad50, and Nbs1 proteins. Hypomorphic mutations in the NBS1 or MRE11 genes in humans result in conditions characterized by DNA damage sensitivity, cell cycle checkpoint deficiency, and high cancer incidence. The equivalent complex in the yeast Saccharomyces cerevisiae (Xrs2p complex) has been implicated in DNA double-strand break repair and in telomere length regulation. Here, we find that xrs2Delta, mre11Delta, and rad50Delta mutants are markedly defective in the initiation of the intra-S phase checkpoint in response to DNA damage. Furthermore, the absence of a functional Xrs2p complex leads to sensitivity to deoxynucleotide depletion and to an inability to efficiently slow down cell cycle progression in response to hydroxyurea. The checkpoint appears to require the nuclease activity of Mre11p and its defect is associated with the abrogation of the Tel1p/Mec1p signaling pathway. Notably, DNA damage induces phosphorylation of both Xrs2p and Mre11p in a Tel1p-dependent manner. These results indicate that the Tel1p/ATM signaling pathway is conserved from yeast to humans and suggest that the Xrs2p/Nbs1 complexes act as signal modifiers.

Figure 1

Hypersensitivity of xrs2Δ, mre11Δ, and rad50Δ mutants to replicative stress. Fivefold serial dilution of yeast cells were plated on YPAD containing various concentrations of HU. For the HU acute hypersensitivity experiment, exponential cultures were grown for 9 h in liquid YPAD containing 200 mM HU before being washed and plated on YPAD plates. Relevant genotypes are shown on the left side of the panels.

Figure 2

Replicative stress causes premature initiation of DNA synthesis in the absence of a functional Xrs2p complex. (A) α-factor-synchronized cells were released from G₁ arrest in medium containing nocodazole and no drug (left panels), 40 mM HU (middle panels), or 35 units/mL bleomycin (right panels), as described in Materials and Methods. Samples were collected at 10-min intervals and analyzed for DNA content by FACS. Yeast strains used were W303-1A (wild-type), DDY004 (xrs2Δ), and DDY022 (xrs2Δ mre11Δ rad50Δ). (B) Quantification of post-G₁ cells. Fraction of cells having a DNA content superior to G₁ are plotted in function of time below the respective FACS profiles. Numbers are corrected for the presence of cells blocked in G₂ at time 0. Typical FACS profiles are shown.

Figure 3

The intra-S phase checkpoint is functional in cells deficient in DNA DSB repair. (A) α-factor-synchronized cells were released from G₁ arrest in medium containing nocodazole and either no drug (top panels) or 35 units/mL bleomycin (bottom panels). Bleomycin was added 10 min after the release from G₁ and samples were collected at 30-min intervals and analyzed for DNA content by FACS, as described in Materials and Methods. Yeast strains used were W303-1A (wild-type), DDY008 (rad50Δ), and DDY062 (rad52Δ yku70Δ). (B) Quantification of post-G₁ cells. Fraction of cells having a DNA content superior to G₁ are plotted in function of time. Numbers are corrected for the presence of cells blocked in G₂ at time 0. Typical FACS profiles are shown.

Figure 4

The Xrs2p complex is required for efficient checkpoint signaling. G₁-synchronized cells were released from α-factor arrest and entered S phase synchronously in the presence either 20 mM (left panels) or 200 mM HU (right panels). Samples were collected at 15-min intervals and protein extracts were prepared. (A) In situ analysis of Rad53p activity. The position of phosphorylated Rad53p is indicated by a star. Western blot analysis of 3-phosphoglycerate kinase (3-PGK) was performed in parallel (shown below Rad53p ISAs) to confirm equal loading in each lane. (B) Quantification of Rad53p autophosphorylation. Bars represent the relative fold-activation of wild-type (black), xrs2Δ (white), and xrs2Δ mre11Δ rad50Δ (grey) strains. The activity of Rad53p was quantified using Fujifilm BAS-2500.

Figure 5

Hypersensitivity of mre11 nuclease mutants to replicative stress. (A) Fivefold serial dilution of mre11Δ yeast containing single-copy plasmids expressing wild-type Mre11–ProA or various nuclease mutants were plated on SC-uracil containing no drug, 20 mM HU, or 200 mM HU. Exponential cultures were also grown for a further 9 h in liquid SC-uracil containing 200 mM HU before being washed and plated on SC-uracil plates, as described in Materials and Methods. Wild-type (WT) cells carrying an empty plasmid were included for comparison. Δ indicates mre11Δ cells carrying various MRE11 alleles or an empty plasmid. (B) The nuclease activity of Mre11p is required for efficient checkpoint signaling. G₁-synchronized mre11Δ cells expressing wild-type Mre11p and nuclease mutants (H213Y and P162S) were released into a synchronous S phase in the presence of 20 mM HU. Rad53p activity was evaluated in situ, as described in Figure 4. The lower right panel is a quantification of Rad53p autophosphorylation; bars represent the relative fold activation of wild-type (black), H213Y mutant (white), and P162S mutant (grey) strains. (C) Mutations in the nuclease domain of Mre11p result in premature initiation of DNA synthesis in the presence of bleomycin. Cells expressing wild-type Mre11p and nuclease mutants (H213Y and P162S) were synchronously released into S phase in the presence of nocodazole and 25 units/mL bleomycin (added 5 min following release), and processed for FACS analysis.

Figure 6

Xrs2p and Mre11p are phosphorylated in response to DNA damage. (A) Exponentially growing cultures of yeast expressing Mre11–13Myc, Xrs2–13Myc and Rad50–3FLAG were analyzed for protein phosphorylation in response to DNA damage (left panels). Cells were untreated (−) or treated with either 4-NQO (25 μM), bleomycin (350 units/mL), MMS (0.02%), or UV (60 J/m²). Arrows indicate the position of the phosphorylated and unphosphorylated bands (left). For dephosphorylation experiments (right panels), Mre11p, Xrs2p, and Rad50p were immunoprecipitated from soluble extracts of yeast treated with 4-NQO. Half of each immunoprecipitate was dephosphorylated with λ phosphatase; the remaining half was mock treated. (B) Genetic requirements for phosphorylation of Mre11p and Xrs2p. Exponential cultures of various checkpoint mutants expressing Myc-tagged Xrs2p or Mre11p were treated with 4-NQO (25 μM). Untreated (−) and treated (+) wild-type cultures were also included as controls. Protein extracts were prepared 1 h after the addition of 4-NQO and analyzed by Western blotting. Arrows indicate the position of the basal and phosphorylated bands (left). Mutant strains are as follows: tel1Δ (Y662), mec1–21 (Y663), tel1Δ mec1–21 (Y664), rad53Δ sml1–1 (U960–5C), rad9Δ rad24Δ and lcd1Δ. The mec1–21 allele used here is a hypomorphic allele that does not require a second site mutation for the suppression of mec1Δ-associated lethality (Desany et al. 1998). Although it is not known whether this allele is completely defective for checkpoint functions, the similar results obtained with the lcd1Δ/ddc2Δ mutant (which is fully epistatic with MEC1; Rouse and Jackson 2000) and the mec1-21 mutant supports the view that we observe the full effect of the loss of Mec1p in our phosphorylation assay.

Figure 7

Functional and epistatic interactions between the Xrs2p complex and Tel1p. (A) HU sensitivity of various mutants affected in the XRS2, RAD53, TEL1, and MEC1 genes. Fivefold serial dilution of yeast cultures were plated on YPAD medium containing low concentrations of HU. Survival was scored after 2–3 d of growth at 30°C. (We consistently observe that a tel1Δ mutation partially suppresses the slow growth phenotype associated with a xrs2Δ mutation, possibly because Tel1p is inappropriately activated in the absence of Xrs2p.) Yeast genotypes are on the left side of the panels. All the strains tested have a sml1-1 mutation. (B) Nuclease mutants of Mre11p are not phosphorylated in response to DNA damage. Exponential cultures of mre11Δ mutants expressing wild-type or nuclease-defective (H213Y and P162S) Mre11–ProA were damaged with 4-NQO (25 μM) and samples were taken at timed intervals. Protein extracts and Western blot analysis were performed as described in Figure 6.