The functions of two interacting proteins, MEI-9 and MUS312, in DNA repair and meiotic recombination pathways

Meiotic recombination and DNA repair pathways are important in maintaining genome stability. MEI-9 is the Drosophila homolog of the human and yeast structure-specific DNA endonucleases XPF and Rad1p. Like XPF and Rad1p, MEI-9 functions in nucleotide excision repair (NER). MEI-9 is also required to repair interstrand crosslinks (ICLs) and to generate meiotic crossovers. To understand how MEI-9 carries out its different functions, I took two approaches. First I carried out a structure-function analysis of the MEI-9 protein. We collected 10 different existing mutant alleles of mei-9. By sequence analysis I found the molecular lesions responsible for each. I also characterized each allele genetically assessing their defects in DNA repair and in meiotic recombination. Genetic characterization of mei-9 mutants identified two separation-of-function alleles: a repair-specific separation-of-function allele, mei-9RT1, and a meiosis-specific separation-of-function allele, mei-9 12. Stucture-function studies also enabled me to elucidate how different motifs in the MEI-9 protein contribute to different functions of MEI-9.; The identification of additional components of mei-9 dependent pathways is another approach that I took to understand the roles of mei-9. A meiotic phenotype similar to mei-9 and a similar sensitivity to NH2 made mus312 a good candidate for being a member of at least two mei-9 dependent pathways: meiotic recombination and ICL repair. I measured the recombination frequency in mus312 mutant flies and found that it is identical to mei-9 mutants. This suggests that both gene products act at the same step in the recombination pathway. We isolated two additional mus312 alleles. They both failed to complement the original mus312 allele in the nondisjunction assay and showed hypersensitivity to ICL agent nitrogen mustard. To clone mus312, we used deficiencies to map it to a small region on chromosome 3 containing 50 predicted genes. At that point an ongoing yeast two hybrid screen in our lab, which we used MEI-9 as a bait, yielded one interacting clone within this region: CG8601. I sequenced this gene from our three mus312 mutant stocks and found that all three alleles have nonsense mutations. Thus I concluded that this mei-9 interacting clone, CG8601, is indeed mus312. The separation-of-function allele mei-912 has high levels of chromosome nondisjunction but is not defective in NER, and thus has a similar phenotype to mus312 mutants. This missense mutation might prevent MEI-912 from interacting with a protein partner required in recombination, but not in NER. MUS312 is a good candidate for such a protein. I tested this hypothesis in a yeast two hybrid assay. There is no detectable interaction between MEI-9 12 and MUS312. Thus, it is likely that mei-912 imparts a recombination defect because the mutant protein cannot interact with MUS312, and therefore, the physical interaction is important in vivo. This interaction suggests that MUS312 might augment the function of MEI-9 in meiotic recombination.