Homologous recombination (HR) is a conserved pathway for repair of DNA double strand breaks (DSBs) and stalled or collapsed replication forks, and depends upon recognition of a homologous template on a sister chromatid or alternate parental copy. Non-allelic homologous recombination (NAHR) results from erroneous recognition of a similar but non-homologous template and can lead to lethal chromosomal deletions or rearrangements. To avoid errors, NAHR is actively suppressed by cellular mechanisms that disrupt heteroduplex recombination intermediates. Using a specialized type of recombination pathway single-strand annealing (SSA) as a model in yeast, I found that rejection of heteroduplex HR intermediate induces a RAD9dependent cell cycle delay in the G2 stage of the cell cycle. Strains lacking the RAD9 gene, and consequently a damage-induced G2 delay, less frequently allowed SSA between divergent sequences than identical ones. However, non-allelic SSA could be restored to wild-type levels if a G2 delay was induced by nocodazole treatment. These results indicate that that cell cycle delay induced by the Rad9-dependent DNA damage response can passively promote recombination between non-allelic sequences despite the potential for creating deleterious genome rearrangements. Secondly, following identification of the Msh2-Msh6 heterodimer and Sgs1 helicase as essential factors for unwinding of a heteroduplex intermediate during SSA (Sugawara et al., 2004; Goldfarb and Alani, 2005), our lab determined that these proteins interact through a direct physical interaction, similarly to mammalian homologs (Pedrazzi et al., 2003; Yang et al., 2004; Saydam et al., 2007). Next I asked whether other proteins that interact with iii Msh6 and Sgs1 contribute to heteroduplex rejection, including the topoisomerase Top3-Rmi1 which is known to stimulate Sgs1 activity (Cejka et al., 2010; Niu et al., 2010) and the replication clamp PCNA which enhances the activity of Msh6 in the mismatch repair pathway. I found that Top3-Rmi1 contributes to heteroduplex rejection, but appears to do so mainly by stabilizing Sgs1. Additionally, I show that PCNA is dispensible for heteroduplex rejection; three mutants of the catalytic subunit pol30 and a msh6 mutant lacking the Pol30 interaction domain rejected SSA heteroduplexes to the same extent as wild-type. Finally, two of the pol30 mutants displayed a reduction in SSA efficiency, revealing an unexpected role for PCNA in SSA. iv