Y2H screens suggest interaction between ScRev7 and the MRX subunits. The Y2H assay was performed in: (A) WT, (B) rev3Δ, and (C) mre11Δ rad50Δ xrs2Δ mutant strains. These strains were cotransformed with pairwise combinations of empty vector, bait (pGBKT7-REV7) and prey (pGADT7/MRE11, RAD50, XRS2 or REV7) plasmids. Equal number of mid-log phase cells were spotted onto the SC/-Trp -Leu agar plates (upper panels) or SC/-Trp -Leu -His agar plates containing 3-AT (bottom panels). Cells were imaged after 48 h of growth at 30 °C. The images shown in panels (A), (B) and (C) are representative of three independent experiments. (D) and (E), normalized MST binding curves obtained for the titration of Mre11, Rad50, Xrs2 and Rev1 proteins against Rev7-eGFP or eGFP, respectively. Error bars indicate SEM, and data are representative of three independent experiments. (F) Quantitative parameters for interaction between ScRev7and Mre11, Rad50, Xrs2 or Rev1 proteins.

Deletion analysis revealed that the C-terminal 42 amino acids of ScRev7 are critical for its binding to the MRX subunits. (A) Schematic representation of the full-length and truncated ScRev7 variants. The truncated species lacking the indicated number of amino acids (aa) in the N-terminal domain (NTDΔ) or C-terminal domain (CTDΔ) is indicated on the right-hand side of the figure. (B) Representative images of spot assays of cells carrying pairwise combination of empty vector, bait and prey plasmids expressing N-terminally truncated species of Rev7 and full-length Rev7, Mre11, Rad50, Xrs2, respectively. (C) Same as panel (B), but with the bait plasmids encoding C-terminally truncated Rev7 variants. Cells were imaged after 48h of growth at 30 °C. Y2H assay was performed as described in the legend to Figure 1. Data are representative of three independent experiments.

A 42-amino acid segment at the extreme C-terminus of ScRev7 renders cells resistant to the synergistic adverse effect of G-quadruplex DNA and HU. (A) REV7 deletion does not affect HU sensitivity of rev7Δ cells. Representative images of YPD pates showing spot assay of WT and rev7Δ cells in the absence or presence of 50 or 100 mM HU. (B) Representative images of SC/-Leu agar plates showing spot assay of WT, rev7Δ, rev7-C1 or rev7-42 aa cells harboring the indicated plasmids carrying G-quadruplex capable sequences derived from chromosome IV (Chr.IVG4). The cells were grown on SC/-Leu agar plates in the absence of HU. (C) Same as panel (B), but the growth medium contained 100 mM HU. For serial dilutions, each strain was grown in SC/-Leu medium, normalized to OD600 = 1.0, and serially diluted using yeast nitrogen base medium. Five μl aliquots from the serial dilutions were spotted onto the SC/-Leu agar plates with or without HU. FT10/Control plasmid lacks the G-quadruplex DNA insert. The abbreviations “le” and “lg” stand for leading and lagging strands, respectively. Cells were imaged after 4 days of growth at 30 °C. Data are representative of three independent experiments.

ScRev7, but not the ScRev7-C1 variant, binds preferentially to G-quadruplex structures. (A) SDS-PAGE analysis of protein samples from different stages of ScRev7 purification. Lane 1: standard protein markers; 2, uninduced cell lysate (10 μg); 3, induced cell lysate (10 μg); 4, Ni2+-NTA column eluate (3 μg); 5, Superdex S75 column eluate (1 μg); 6, eluate from heparin column (0.8 μg). (B) SDS-PAGE analysis of purified truncated ScRev7-C1 variant. Lane 1: standard protein markers. Lanes 2 and 3, purified full length ScRev7 and ScRev7-C1 variant. (C) Quantification of DNA binding specificities of ScRev7 or the variant ScRev7-C1. Each point in the graph represents mean values of three independent experiments. Four symbols [square (□), cross (x), black circle (•) and red circle (•)] overlap with each other, and are connected by a horizontal line along the x-axis. The best-fit curve was obtained by non-linear regression analysis by applying one-site binding model using Graph pad Prism (v 5.0). Error bars represent SEM of three independent experiments. (D) Apparent equilibrium dissociation constants (Kd values) for the binding of Rev7 or Rev7-C1 variant to the indicated DNA substrates. While G-quadruplex structures 4G3-G4 and 6G3-G4 contain four and six arrays of three guanine residues, respectively, TP-G4, contains one each of five, four and three arrays of guanine residues.

ScRev7 impedes both exo- and endonucleolytic activities of the MRX subunits. (A) A representative image showing the effect of ScRev7 on the exonuclease activity of MRX complex. Reaction mixtures containing 5 nM 32P-labelled 60 bp dsDNA were incubated in the absence or presence of M/R/X subunits and various amounts of ScRev7. Lane 1, 32P-labelled dsDNA. Lane 2, same as in lane 1, but with 2.5 μM ScRev7. Lane 3, same as in lane 1, but with M/R/X subunits (0.1 μM each). Lanes 4-12, same as in lane 3, but with 0.05, 0.1, 0.3, 0.5, 0.7, 1, 1.5, 2 and 2.5 μM of ScRev7, respectively. (B) A representative image showing the effect of Rev7-C1 variant on the exonuclease activity of the MRX complex. Assay was performed as in panel (A), but with ScRev7-C1 variant. Lane 1, 32P-labelled dsDNA. Lane 2, same as in lane 1, but with 2.5 μM ScRev7-C1 variant. Lane 3, same as in lane 1, but with M/R/X subunits (0.1 μM each). Lanes 4-5, same as in lane 3, but with 1.5 and 3.0 μM of ScRev7-C1 variant, respectively. (C) A representative image showing Mre11 endonuclease activity on circular ssDNA. Reaction mixtures containing 100 ng M13 circular ssDNA were incubated without (lane 1) or with 0.1, 0.5, 1, 1.5 and 2 μM Mre11, respectively (lanes 2-6). (D) A representative image showing the effect of ScRev7 on Mre11 endonuclease activity by ScRev7. Reaction mixtures containing 1 μM Mre11 were pre-incubated with 0.5, 1.0 1.5 and 2 μM of ScRev7 (lanes 3-6 respectively), prior to the addition of 100 ng of M13 circular ssDNA. Lane 1, circular ssDNA alone. Lane 2, same as in lane 1, but with ScRev7. Increasing concentrations of the indicated protein is represented by open triangles at the top of the gel.

Rev7 inhibits the ATPase activity of Rad50 without impacting its ability to bind ATP. (A), Rad50 binds [γ-32P]ATP in a dose-dependent manner. Lane 1, no protein. Lanes 2-7, reactions were performed with 0.1, 0.2, 0.4, 0.6, 0.8 and 1 µM of Rad50 and 400 pmol γ-32P]ATP. (B) Quantification of ATP binding by Rad50 as a function of its concentration. (C) ScRev7 does not affect the ability of Rad50 to bind ATP. Lane 1, Rad50 and 400 pmol γ-32P]ATP. Lanes 2-7, same as in lane 1, but with 0.5, 1, 2, 4, 5, and 6 µM of ScRev7, respectively. (D) Quantification of the effect of ScRev7 on ATP binding by Rad50. (E) ATPase activity of Rad50 as a function of its concentration. Reactions were performed in the absence (lane 1) or presence (lanes 2-9) of 0.05, 0.1, 0.2, 0.3, 0.5, 0.6, 0.8 and 1 µM of Rad50, respectively. (F) Quantification of Rad50 ATPase activity as a function of its concentration. (G) ScRev7 abrogates ATP hydrolysis catalyzed by Rad50. (H) ScRev7-C1 variant impedes ATP hydrolysis catalyzed by Rad50. In panels (G) and (H) lane 1 contained 400 pmol γ-32P]ATP; lane 2, same as in lane, but 2 µM ScRev7/ScRev7-C1 variant; lane 3, as in lane 1, but with 0.5 µM Rad50; lanes 4-9, as in lane 3, but with 0.5 µM Rad50 and 0.1, 0.3, 0.5, 1, 1.5 and 2 µM ScRev7/ScRev7-C1 variant, respectively. (I) Quantification of the inhibitory effect of ScRev7/ScRev7-C1 on ATP hydrolysis catalyzed by Rad50. The closed triangles on the top of gel images in panels (A), (C), (E), (G) and (H) represent increasing concentrations of Rad50, ScRev7 or ScRev7-C1. Error bars indicate SEM, and data are representative of three independent experiments.

S. cerevisiae REV7 promotes NHEJ-mediated DSB repair.

(A) Map of the NHEJ reporter plasmid and the experimental workflow. Sixty ng of uncut or linearized pRS416 plasmid DNA was transformed into the WT and indicated isogenic mutants carrying single or double deletions. Transformants were selected on SC medium lacking uracil. (B) Quantification of NHEJ efficiency relative to WT cells. (C) Quantification of fold decrease in the efficiency of NHEJ relative to the WT. The boxes represent mean; whiskers, minimum and maximum values. “LOD” denotes below the level of detection compared with the WT. The data are presented as the mean ± SEM of three independent experiments. n.s.: not significant, *p < 0.05, **p < 0.005, ****p < 0.0001 vs. control. The exact p-values are presented in Supplementary Table S6.

S. cerevisiae REV7 facilitates chromosomal DSB repair via NHEJ pathway. (A) A schematic diagram showing the “suicide deletion” cassette at the ADE2 locus of chromosome XV [redrawn from ref. Karathanasis and Wilson, 2002]. The arrows indicate the locations of PCR primers. Brown and staggered red boxes correspond to direct repeats and I-SceI cleavage sites, respectively. (B) Quantification of NHEJ efficiency in the WT and mutant strains. (C) Representative gel images of PCR-amplified DNA products from Ade2+ transformant cells. Data are means ± SEM from three independent experiments. n.s.: not significant, **p < 0.005, ***p < 0.001, ****p < 0.0001 vs. control, as assessed by one-way ANOVA Dunnett’s multiple comparison test. The exact p-values are presented in Supplementary Table S7.

Deletion of REV7 increases the frequency of mitotic HR. (A) Schematic representation of plasmid-chromosome recombination assay. The ura3-1 and ura3-G4 alleles are located on chromosome V and plasmid pFAT10-G4, respectively. Recombination between a plasmid borne ura3-G4 allele and the chromosomal borne ura3-1 allele would result in Ura+ prototrophs. (B) Representative images of Ura3+ papillae on SC/-Ura agar plates in the absence or presence of 100 mM HU. (C) Quantification of the rate of HR frequency in different strains. Data are presented as mean ± SD from three different experiments. ns, not significant, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. control, as assessed using one-way ANOVA and Tukey’s post hoc test.

SDS-PAGE analysis of purified ScRev7-GFP.

SDS-PAGE and western blot analysis of purified ScRev7-GFP. Lane 1, standard molecular weight markers; 2, purified GFP; 3, purified Rev7-GFP; Western blot analysis of (4) purified GFP and (5) Rev7-GFP, using mouse anti-GFP antibodies.

Representative spectra for unique peptides obtained after trypsin digestion of Rev7-GFP protein that match with (A) GFP sequence, (B) ScRev7 sequence. (C) Snapshot of sequence coverage (shown in green) obtained for Rev7 and (D) GFP. (E) Table summarizes data obtained for Orbitrap LC-MS performed using purified Rev7-GFP protein.

SDS-PAGE analysis of purified GST-tagged ScRev1.

Lane 1, standard molecular weight markers; 2, purified GST tagged Rev1. The faster migrating band corresponds to ScRev7, which co-purifies with Rev1-GST (Acharya et al., 2005).

AlphaFold-Multimer generated models of Rev7-Mre11 and Rev7-Rad50 protein complexes.

(A) Model of the Rev7-Mre11 heterodimer, Rev7, light blue, Mre11, deep blue. (B) Model of Rev7-Rad50 heterodimer, Rev7, light blue, Rad50, cyan. Alongside is the two-dimensional view of the corresponding protein-protein binding interface as generated by the LigPlot software. Numbers indicated on the dotted lines represent bond-lengths in Angstroms. The C-terminal safety belt region of Rev7 is coloured in green in both complexes. By convention, red, black and blue circles represent oxygen, carbon and nitrogen atoms respectively. The dashed line represents the interaction interface.

Western blot showing the abundance of N- and C-terminally truncated variants of Rev7.

S. cerevisiae PJ69-4A cells were co-transformed with the indicated bait plasmid expressing c-Myc epitope tagged N- or C-terminally truncated variant of Rev7 (REV7-C1, REV7-C2, REV7-C3, REV7-N1, REV7-N2 or REV7-N3) with and a prey plasmid expressing REV7, MRE11, RAD50 or XRS2. PGK-1 was used as the loading control and was probed with anti-PGK-1 antibodies. Sample loading was normalized to similar amounts by determining the protein concentration utilizing the dye-binding assay. Lane ‘C’ represents lysates from cells harboring empty bait and prey vectors.

ScRev7 preferentially binds to G4 DNA, as opposed to ssDNA, dsDNA or mutated G4 DNA.

Panels (A-C), Reaction mixtures containing 20 mM Tris-HCl (pH 6.8), 0.1 mM DTT, 0.2 μg/ml BSA, 5% glycerol, 0.5 nM of 32P-labeled G-quadruplex structures with varying G4 motifs (indicated at the top of the gel) with varying concentrations of ScRev7. Lane 1 in all panels, the indicated DNA substrate without protein. Lanes 2-11, reactions performed with ScRev7 at 5, 20, 40, 50, 70, 90, 100, 120, 140 and 160 nM, respectively. Panels (D-F), similar to panels (A-C), but the reaction mixtures contained 32P-labeled ssDNA, dsDNA or mutated G4 DNA (G), similar to panels (D-F), but reaction mixtures contained 25, 100, 200, 250, 350, 450, 500, 600, 650, 800 nM ScRev-C1 variant.

ScRev7 inhibits the ATPase activity of ScRad50.

ATPase activity was measured using malachite green phosphate detection assay that quantifies the amount of Pi released upon ATP hydrolysis. (A) Increasing concentrations of ScRad50 (0, 0.05, 0.1, 0.2, 0.3, 0.5, 0.6, 0.8 and 1 µM) were incubated with reaction mixtures containing 20 mM Tris-HCl, pH 7.5, 50 mM KCl, 5% glycerol, 0.1 mM DTT, 0.2 mg/ml BSA, 150 µM ATP and 1 mM MgCl2. Absorbance values were then monitored spectrophotometrically at 620 nm. (B) Same as (A), but fixed concentrations of ScRad50 (0.25 µM) were pre-incubated with increasing concentrations of ScRev7 or ScRev7-C1 (0, 0.1, 0.3, 0.5, 1, 1.5, 2 µM). The graphs were generated by non-linear regression analysis using GraphPad Prism (V. 5.0). Error bars represent standard deviation of data from two independent experiments.

Mutation of G-quadruplex-forming sequences markedly attenuate the rate of HR frequency in rev7Δ cells. (A) Representative images showing Ura3+ papillae on SC/-Ura agar plates in the absence or presence of 100 mM HU. Cells were imaged after six days of growth at 30 °C. (B) Quantification of the rate of HR frequency in ura3-1 and ura3-1 rev7Δ strains carrying plasmids with unmutated and mutated ura3-G4 inserts. Data are presented as mean ± SD from three different experiments. n.s., not significant, *p<0.05, and ***p < 0.001 vs. control, as assessed using two-way ANOVA and Sidaks multiple comparison test.

G-quadruplex capable sequences are stable in rev7Δ cells during the HR assay. Total plasmid pFAT10-G4 DNA was isolated from from cultures of WT and rev7Δ cells. Samples of undigested and BamHI and Sph1 digested plasmid pFAT10-G4 DNA were analyzed by electrophoresis on a 0.6 % agarose gel and visualized after staining with ethidium bromide. Lane1: DNA marker; 2 and 5, represent undigested plasmid DNA from the WT and rev7Δ cells respectively. Lanes 3 and 4, plasmid pFAT10-G4 DNA from the WT cells digested by BamHI and Sph1; 6 and 7, same as lane 3 and 5, but plasmid isolated from rev7Δ cells. + RE and –RE indicate DNA digested by BamHI and Sph1 and undigested plasmid DNA respectively.

Inter-atomic distances and confidence parameters of amino acid residues mediating Rev7-Mre11 interactions.

Inter-atomic distances and confidence parameters of amino acid residues mediating Rev7-Rad50 interactions

The exact p-values for Figure 7B.

The p-values were obtained by comparing the percentage of non-homologous end-joining observed for the indicated single- or double-gene deletions versus either the wild-type (WT) or rev7Δ strain, using non-parametric one-way ANOVA Dunnett test.

The exact p-values for Figure 8B.

The p-values were obtained by comparing the percentage of non-homologous end-joining observed for the indicated single- or double-gene deletions versus either the wild-type (WT) or rev7Δ strain, using non-parametric one-way ANOVA Dunnett test.

S. cerevisiae strains used in this study.

Sequences of primers

Sequences of oligonucleotides used in the preparation of DNA substrates.