Resection activity of the Sgs1 helicase alters the affinity of DNA ends for homologous recombination proteins in Saccharomyces cerevisiae

Abstract

The RecQ helicase family is critical during DNA damage repair, and mutations in these proteins are associated with Bloom, Werner, or Rothmund-Thompson syndromes in humans, leading to cancer predisposition and/or premature aging. In the budding yeast Saccharomyces cerevisiae, mutations in the RecQ homolog, SGS1, phenocopy many of the defects observed in the human syndromes. One challenge to studying RecQ helicases is that their disruption leads to a pleiotropic phenotype. Using yeast, we show that the separation-of-function allele of SGS1, sgs1-D664Δ, has impaired activity at DNA ends, resulting in a resection processivity defect. Compromising Sgs1 resection function in the absence of the Sae2 nuclease causes slow growth, which is alleviated by making the DNA ends accessible to Exo1 nuclease. Furthermore, fluorescent microscopy studies reveal that, when Sgs1 resection activity is compromised in sae2Δ cells, Mre11 repair foci persist. We suggest a model where the role of Sgs1 in end resection along with Sae2 is important for removing Mre11 from DNA ends during repair.

Keywords: RecQ helicases; Sae2; Sgs1; homologous recombination; resection.

Figure 1

Cells with sgs1-D664Δ have a synthetic sick interaction when combined with sae2Δ. (A) Yeast with sgs1-D664Δ or sgs1Δ were mated to sae2Δ, mre11Δ, rad50Δ, or xrs2Δ haploid cells. The diploids were sporulated and the meiotic products analyzed by tetrad analysis. Pictures of representative spores from a single tetrad are shown. The size of the spore colonies were quantitated relative to WT (set at one), and their sizes are plotted on the graph with standard deviations shown. (B) Summary of synthetic interactions for sgs1-D664Δ with HR mutant genes. The synthetic interactions marked with an asterisk (*) were previously reported (Bernstein et al. 2009).

Figure 2

In the absence of Rad51, sgs1-D664Δ exo1Δ cells display a resection defect. (A) Schematic of the SSA assay used with two ade2 heteroalleles with an intervening TRP1 gene. A unique I-SceI cut site is inserted in one of the ade2 alleles. When I-SceI is conditionally expressed, SSA occurs and the repair products lead to ADE2+ trp1- recombinants. The sizes of the bands generated by restriction digestion of genomic DNA are indicated below each schematic. (B) Yeast strains containing the SSA assay diagrammed with the indicated genotypes were grown in rich medium (YPD) to early log phase (0.5 OD₆₀₀) and then fivefold serially diluted onto rich medium (YPD) or medium containing galactose and raffinose (YPGal/Raf), which induces expression of the I-SceI enzyme from a galactose-inducible promoter. Plates were photographed after 3 days of growth at 30°. The same cells were analyzed by DNA blotting to detect the cut fragments and the ADE2+ repair product from 0 to 6 hr after I-SceI cutting.

Figure 3

When RAD51 is disrupted, sgs1-D664Δ exo1Δ cells exhibit a resection processivity defect. (A) Schematic of the mating-type locus on chromosome III before and after an HO-induced DSB. When RAD51 is disrupted, the formation of ssDNA is favored and can be monitored by alkaline electrophoresis of StyI (S)/BstXI (B)-digested genomic DNA (see text for discussion). The ssDNA products, designated, r1–r7, can be observed by DNA blot hybridization with a 3′ strand-specific probe. (B) Cells with the indicated genotypes were analyzed by DNA blot for the r1–r7 ssDNA fragments 0, 0.5, 1, 1.5, 2, 3, and 4 hr after induction of HO endonuclease.

Figure 4

The slow growth of sae2Δ sgs1-D664Δ cells is suppressed by yku70Δ or by Exo1 overexpression. (A and B) Yeast strains with the indicated genotypes were fivefold serially diluted onto rich medium, grown at 30° for 1 day, and photographed. (C) WT or sgs1-D664Δ sae2Δ cells were transformed with an empty plasmid or one that expressed Exo1. The respective strains were fivefold serially diluted onto minimal medium to maintain a plasmid (SC-TRP), grown at 30° for 2 days, and photographed. (D) Spore colony size of the strains of the indicated genotypes. The spore colony size relative to WT was calculated and plotted with standard deviations. The exo1Δ sae2Δ sgs1-D664Δ spores did not generate viable colonies. (E) Yeast strains containing the SSA assay diagrammed in Figure 2A with the indicated genotypes were grown in rich medium (YPD) to early log phase (0.5 OD₆₀₀) and then fivefold serially diluted onto rich medium (YPD) or medium containing galactose and raffinose (YPGal/Raf), which induces expression of the I-SceI enzyme from a galactose-inducible promoter. Plates were photographed after 3 days of growth at 30°. (F) The sae2Δ and sae2Δ sgs1-D664Δ cells were analyzed by DNA blot for resection products and cut fragments were diagrammed as in Figure 2A.

Figure 5

Persistent Mre11 foci form in response to IR in sae2Δ sgs1-D664Δ double mutants and is alleviated by disrupting YKU70. Cells expressing Mre11-YFP were analyzed in WT, sgs1-D664Δ, sae2Δ, sae2Δ sgs1-D664Δ, yku70Δ, or sae2Δ sgs1-D664Δ yku70Δ after exposure to 40 Gy of IR for 0.2–4 hr. For each time point, a single Z-stack is shown; white arrowheads indicate a focus. Each experiment was done in triplicate with a total of 200–650 cells analyzed with standard errors plotted.

Figure 6

Mre11 foci persist in sae2Δ sgs1-D664Δ double mutants at an inducible DSB site. An I-SceI cut site integrated at the URA3 locus on chromosome V adjacent to 112 copies of the Tet repressor-binding site (112xtetO) was analyzed (Lisby et al. 2004). Recruitment of Mre11-YFP to the cut site was monitored in strains expressing a GAL-I-SceI plasmid either immediately (0 hr) or 4 hr after I-SceI expression was inhibited by the addition of glucose. The experiment was done in duplicate, and the results are quantitated in the graph with standard error plotted.