Recombination of yeast mitochondrial DNA does not require mitochondrial protein synthesis

Mitochondrial genes recombine extensively in yeast zygotes. In heteropolar crosses (ω(+) × ω(-)) in which the ω(-) "allele" consists of an insertion, there is preferential recovery of ω(+) and markers closely linked to it. This polarity has been postulated to be a consequence of one-way gene conversion beginning at the ω locus (ω- to ω(+)). We have shown that most or all mitochondrial recombination in homopolar and heteropolar crosses, and the phenomenon of polarity itself, does not require products of protein synthesis on mitochondrial ribosomes. (i) Yeast strains were grown and mated, and the zygotes plated and grown, on glucose medium with erythromycin to inhibit and dilute out the products of mitochondrial protein synthesis. Recombination frequencies and polarity at the cap1 and oli1 loci were normal compared to controls in some homopolar (ω(+) × ω(-)) and heteropolar crosses. Apparent changes in recombination frequencies and polarity were seen in other crosses but are attributable to locus-specific petite induction by erythromycin. (ii) Homopolar (ω(+) × ω(+)) and heteropolar crosses between pairs of petite mutants retaining the cap1, ery1, and oli1 loci also showed nearly normal recombination at the cap1 and oli1 loci, as determined by test-crossing the petite progeny. The petite mutants and zygotes cannot do mitochondria) protein synthesis. These results support the recombinational model of polarity.