Double-strand breaks (DSBs) represent the most severe DNA lesion a cell can suffer, as they pose the risk of inducing loss of genomic integrity and promote oncogenesis in mammals. Two pathways repair DSBs, nonhomologous end joining (NHEJ) and homologous recombination (HR). With respect to mechanism and genetic requirements, characterization of these pathways has revealed a large degree of functional separation between the two. Nej1 is a cell-type specific regulator essential to NHEJ in Saccharomyces cerevisiae. Srs2 is a DNA helicase with multiple roles in HR. In this study, we show that Nej1 physically interacts with Srs2. Furthermore, mutational analysis of Nej1 suggests that the interaction was strengthened by Dun1-dependent phosphorylation of Nej1 serines 297/298. Srs2 was previously shown to be recruited to replication forks, where it promotes translesion DNA synthesis. We demonstrate that Srs2 was also efficiently recruited to DSBs generated by the HO endonuclease. Additionally, efficient Srs2 recruitment to this DSB was dependent on Nej1, but independent of mechanisms facilitating Srs2 recruitment to replication forks. Functionally, both Nej1 and Srs2 were required for efficient repair of DSBs with 15-bp overhangs, a repair event reminiscent of a specific type of HR called single-strand annealing (SSA). Moreover, absence of Rad51 suppressed the SSA-defect in srs2 and nej1 strains. We suggest a model in which Nej1 recruits Srs2 to DSBs to promote NHEJ/SSA-like repair by dismantling inappropriately formed Rad51 nucleoprotein filaments. This unexpected link between NHEJ and HR components may represent cross-talk between DSB repair pathways to ensure efficient repair.

Yen1 is a nuclease identified in Saccharomyces cerevisiae that cleaves the Holliday junction (HJ) intermediate formed during homologous recombination. Alternative routes to disjoin HJs are performed by the Mus81/Mms4- and Sgs1/Top3/Rmi1-complexes. Here, we investigate the role of the Yen1 protein in the yeast Kluyveromyces lactis. We demonstrate that both yen1 mus81 and yen1 sgs1 double mutants displayed negative genetic interactions in the presence of DNA-damaging chemicals. To test if these phenotypes required the catalytic activity of Yen1, we introduced point mutations targeting the catalytic site of Yen1, which abolished the nuclease activity in vitro. Remarkably, catalytically inactive Yen1 did not exacerbate the hydroxyurea sensitivity of the sgs1Δ strain, which the yen1Δ allele did. In addition, overexpression of catalytically inactive Yen1 partially rescued the DNA damage sensitivity of both mus81 and sgs1 mutant strains albeit less efficiently than WT Yen1. These results suggest that Yen1 serves both a catalytic and non-catalytic role in its redundant function with Mus81 and Sgs1. Diploids lacking Mus81 had a severe defect in sporulation efficiency and crossover frequency, but diploids lacking both Mus81 and Yen1 showed no further reduction in spore formation. Hence, Yen1 had no evident role in meiosis. However, overexpression of WT Yen1, but not catalytically inactive Yen1 partially rescued the crossover defect in mus81/mus81 mutant diploids. Yen1 is a member of the RAD2/XPG-family of nucleases, but genetic analyses revealed no genetic interaction between yen1 and other family members (rad2, exo1 and rad27). In addition, yen1 mutants had normal nonhomologous end-joining efficiency. We discuss the similarities and differences between K. lactis Yen1 and Yen1/GEN1 from other organisms.

Yen1 is nuclease that can cleave the Holliday junction (HJ), an important DNA intermediate formed during homologous recombination. Here, we show that Yen1 interacts molecularly with Uls1, a SUMO targeted ubiquitin ligase that also belongs to the SWI/SNF-family of DNA-dependent ATPases. We demonstrate that Yen1 is SUMO modified in its carboxyl terminus and that this modification strengthens the interaction between Yen1 and Uls1. Absence of Uls1 increased the steady-state levels of Yen1, but only after extensive DNA damage, suggesting that Uls1 has a role in damage-induced degradation of Yen1. Consistent with a shared role for Uls1 and Yen1, mutations in the two enzymes display similar phenotypes. Both uls1 and yen1 have a negative genetic interaction with the alternative HJ-cleaving nuclease Mus81. This negative genetic interaction is manifested in supersensitivity to DNA damaging agents, but also in a meiotic defect. Neither mus81 uls1 nor mus81 yen1 double mutant diploids can complete meiosis. Moreover, both uls1 and yen1 exacerbates the chromosome mis-segregation phenotype of mus81. However, the mus81 uls1 yen1 triple mutant strain was slightly more sensitive to DNA damage compared to any double mutant combination, indicating that Uls1 and Yen1 also have independent roles in DNA repair. Point mutant alleles of Uls1 (uls1K975R and uls1C1330S/C1333S) that inactivates the ATPase and potential ubiquitin ligase activities are also supersensitive to DNA damage when combined with mus81, indicating that both activities of Uls1 are essential for function. We suggest that Yen1 and Uls1 are involved in an alternative pathway that is responsible for resolving complex DNA repair intermediates in the absence of Mus81.

Condensed chromatin hinders proteins from accessing the DNA, hence posing a block to processes like DNA repair. In this study, we investigate how histone modifications influence DNA double-strand break (DSB) repair. We show that blocking phosphorylation of serine 129 of histone H2A impairs DSB-repair, probably by reducing the efficiency of homologous recombination (HR). The lysine residues of histone H3 and H4 are subjected to reversible acetylation and methylation and we exchanged the lysines for either arginine (mimicking non-acetylated lysine) or glutamine (mimicking acetylated lysine). A histone H3 mutant with five N-terminal lysines exchanged for arginine showed reduced gene conversion and perturbed cell cycle progression. Leaving a single lysine residue intact was sufficient for protecting cells from DNA damage. In addition, exchanging the five lysines for glutamine did not result in these defects, indicating that one lysine residue in the histone H3 N-terminus must be acetylated for efficient DSB-repair. We find no evidence for that histone modification reduces the efficiency of nonhomologous end joining. Furthermore, the histone H3 K9, 14, 18, 23, 27R mutation is not defective in transcription of DSB repair genes indicating that the defects we observe in DSB-repair is unlikely to be due to indirect regulatory effects. These findings indicate that both histone H2A phosphorylation and histone H3 acetylation is important for the efficiency of the HR-pathway.