The role of heteroduplex correction in gene conversion in Saccharomyces cerevisiae

Abstract

Two different models have been proposed to explain the relative frequencies of the non-mendelian allelic segregations which are detected by tetrad analysis after meiosis in fungi^1–7. The first model maintains that 6:2 type tetrads result from correction of heteroduplexes containing mismatched sites and 5:3 type tetrads result from failure to correct mismatched sites. The second model suggests that 6:2 segregations result from the filling-in of double-strand gaps using information obtained from both strands of a homologous duplex. In this model 5:3 type tetrads result if the allele is included in the heteroduplex regions flanking the gap and the resulting mismatched nucleotides are not corrected⁷. We have studied the correction of heteroduplex plasmid DNA in pms1 mutant strains of Saccharomyces cerevisiae, which are known to exhibit higher frequencies of 5:3 type tetrads and lower frequencies of 6:2 tetrads than wild-type strains. Our results suggest that the pms1 mutation causes a defect in mismatch correction, supporting the hypothesis that meiotic gene conversion in wild-type yeast cells often results from the correction of heteroduplex DNA.