Phenotypes of mutations in the three groups of genes affecting meiotic recombination in

In the yeast, Saccharomyces cerevisiae, several genes appear to act early in meiotic recombination. HOPl and REDl have been classified as such early genes. The data in this paper demonstrate that neither a redl nor a hopl mutation can rescue the inviable spores produced by a rad52 spol3 strain; this phenotype helps to distinguish these two genes from other early meiotic recombination genes such as SPOll, REC104, or A4EI4. In contrast, either a redl or a hopl mutation can rescue a rad50S spol3 strain; this phenotype is similar to that conferred by mutations in the other early recombination genes (e.g., REC104). These two different results can be explained because the data presented here indicate that a rad50S mutation does not diminish meiotic intrachromosomal recombination, similar to the mutant phenotypes conferred by redl or hopl. Of course, REDl and HOPl do act in the normal meiotic interchromosomal recombination pathway; they reduce interchromosomal recombination to -10% of normal levels. We demonstrate that a mutation in a gene (REC104) required for initiation of exchange is completely epistatic to a mutation in REDl. Finally, mutations in either HOPl or RED1 reduce the number of doublestrand breaks observed at the HIS2 meiotic recombination hotspot. A unique event that affects chromosomes during meiosis is the high frequency recombination and synapsis between homologs that occurs in prophase I. Recombination not only generates variation among genes that reside on the same chromosome, but is, in almost all eucaryotes, necessary for the proper execution of the second unique meiotic event, the reductional division. Much progress has been made in defining the genes required for meiotic recombination and synapsis in the yeast Saccharomyces cermisiae ( e.g., PETES et al. 1991). Many of the advances in understanding meiotic recombination are due to the existence of a powerful tool in yeast that allows (at least some types of) Reccells to produce viable meiotic products (spores). This tool is a mutation in the SPOI3 gene; a spol3mutation can essentially eliminate the reductional division so that there is no longer an absolute demand for recombination for proper chromosome s gregation ( KLMHOLZ and ESPOSITO 1980; MALONE and ESPOSITO 1981; KLAPHOLZ et al. 1985). A $013 mutant diploid produces two diploid spores ( a “dyad” ) . More recently, HUGERAT and SIMCHEN ( 1993) demonstrated that, in the presence of ~$013, some chromosomes can undergo a reductional division while others divide equationally. However, in the presence of a Rec mutation that eliminates meiotic recombination (such as recl04), essentially all chromosomes undergo an equational division in a spol3 strain and produce almost no ( <0.5% ) recombinant (or reductional) type dyads for all chromosomes and intervals examined ( e . 6 , GAL,. Corresponding author: Robert E. Malone, Department of Biological Sciences, University of Iowa, Iowa City LA 52242. E-mail: [email protected] Genetics 144: 71-86 (September, 1996) BRAITH and MALONE 1992). One can conclude from this that any reductional divisions that may occur in a spol3 meiosis are dependent upon recombination. It is possible to place genes affecting meiotic recombination in S. cerevisiae into two basic classes, which have been called “Early” and “Late” ( MALONE 1983; see also PETES et al. 1991 ) . In combination with spol3, mutations in early genes (e.g., spoll) produce viable, recombinationless spores after meiosis. Mutations in late recombination genes (e.g., rad52) produce dead spores even if spol3 is present. To demonstrate that the late mutations are indeed in the recombination pathway, it is possible to “rescue” them by mutating an early gene like RAD50 or SPOll ( e.g., MALONE 1983; ENGEBRECHT and ROEDER 1989). These observations are consistent with the idea that recombination occurs in a linear dependent pathway; blocks in the early genes prevent exchange from starting, whereas blocks in the late recombination genes stop exchange in the middle of the process ( MALONE 1983). Bivalents caught in the middle of exchange would presumably have difficulty with either an equational or a reductional division and hence would produce dead spores even in a spol3 meiosis. Consistent with this view are the observations that cells with mutations in late exchange genes generate some recombinational intermediates ( e.g., heteroduplex D N A ) , while mutations in early exchange genes do not generate any intermediates at all (e.g., BORTS et al. 1986; NAG and PETES 1993; NAG et aZ. 1995). The picture became more complex with the discovery of a second type of early meiotic recombination gene. In a clever screen for mutations that affected interchromosomal, but not intrachromosomal, meiotic ex72 Y . Mao-Draayer et al.