Saccharomyces cerevisiae DNA Ligase IV Supports Imprecise End Joining Independently of Its Catalytic Activity

摘要

DNA ligase IV (Dnl4 in budding yeast) is a specialized ligase used in non-homologous end joining (NHEJ) of DNA double-strand breaks (DSBs). Although point and truncation mutations arise in the human ligase IV syndrome, the roles of Dnl4 in DSB repair have mainly been examined using gene deletions. Here, Dnl4 catalytic point mutants were generated that were severely defective in auto-adenylation in vitro and NHEJ activity in vivo , despite being hyper-recruited to DSBs and supporting wild-type levels of Lif1 interaction and assembly of a Ku- and Lif1-containing complex at DSBs. Interestingly, residual levels of especially imprecise NHEJ were markedly higher in a deletion-based assay with Dnl4 catalytic mutants than with a gene deletion strain, suggesting a role of DSB-bound Dnl4 in supporting a mode of NHEJ catalyzed by a different ligase. Similarly, next generation sequencing of repair joints in a distinct single-DSB assay showed that dnl4 -K466A mutation conferred a significantly different imprecise joining profile than wild-type Dnl4 and that such repair was rarely observed in the absence of Dnl4. Enrichment of DNA ligase I (Cdc9 in yeast) at DSBs was observed in wild-type as well as dnl4 point mutant strains, with both Dnl4 and Cdc9 disappearing from DSBs upon 5′ resection that was unimpeded by the presence of catalytically inactive Dnl4. These findings indicate that Dnl4 can promote mutagenic end joining independently of its catalytic activity, likely by a mechanism that involves Cdc9. Author Summary Chromosomal rearrangements are common driver mutations in human genetic disease and cancer. The junctions observed at rearrangements typically show only a few base pairs in common between the partners, suggesting that they were formed by the end-to-end joining process, nonhomologous end joining (NHEJ). However, there is uncertainty about the mechanisms that actually create mutated junctions. DNA ligase IV catalyzes restorative double-strand break (DSB) joining in the canonical NHEJ pathway, but increasing evidence suggests that distinct NHEJ pathways that use DNA ligases I and/or III might be more important for mutations. We used yeast to study the in vivo consequence of having DNA ligase IV that was catalytically inactive but that nonetheless accumulated at DSBs normally. We detected mutated junctions in some assays that required DNA ligase IV protein but not its catalytic activity. This pattern suggests that DNA ligase I creates many mutated junctions when DNA ligase IV is present and that this can become a predominant mode of repair when DNA ligase IV activity is inefficient. Our yeast ligase IV mutations have properties similar to those observed in the human ligase IV syndrome, underscoring the relevance of these observations.