Why budding yeast overrides the DNA damage checkpoint

Checkpoints arrest biological processes and enhance chances for error correction. In many species ranging from budding yeast to human, checkpoints are eventually overridden despite persistent dam-age. Whether checkpoint override serves a biological function remains unclear. Here, we investigate this question in the context of the DNA damage checkpoint (DDC) in budding yeast. To demonstrate that DDC override increases fitness, we pursued a novel approach: To avoid inherent ambiguities when comparing genetic mutants, we instead employed a light-controlled trigger to finely tune the timing of checkpoint override in a consistent wild-type DDC and DNA repair background. We show that override is beneficial and wild-type override timing maximizes fitness. We formulate two specific hypotheses to explain the fitness benefit: i) override enables multiple rounds of replication, including of broken chromosomal fragments, statistically increasing the chance of at least one successful repair; or ii) override may enhance specific DNA repair pathways. Testing the first hypothesis, we tracked broken chromosome fragments using an optogenetic reporter and found their segregation pattern to be inconsistent with a probabilistic increase in post-override repair opportunities. To test the second hypothesis, we dynamically depleted key repair pathway proteins - individually and combinatorially - without interfering with the establishment of checkpoint arrest. Strikingly, we found that proteins in-volved in microhomology-mediated end joining (MMEJ) substantially enhanced override-associated break repair. Together, these results provide direct evidence of a fitness advantage conferred by checkpoint override and uncover MMEJ-associated repair proteins as the mechanistic basis.