Fine-scale variation in meiotic recombination in Mimulus inferred from population shotgun sequencing

Meiotic recombination rates can vary widely across genomes, with hotspots of intense activity interspersed among cold regions. In yeast, hotspots tend to occur in promoter regions of genes, whereas in humans and mice, hotspots are largely defined by binding sites of the positive-regulatory domain zinc finger protein 9. To investigate the detailed recombination pattern in a flowering plant, we use shotgun resequencing of a wild population of the monkeyflower Mimulus guttatus to precisely locate over 400,000 boundaries of historic crossovers or gene conversion tracts. Their distribution defines some 13,000 hotspots of varying strengths, interspersed with cold regions of undetectably low recombination. Average recombination rates peak near starts of genes and fall off sharply, exhibiting polarity. Within genes, recombination tracts are more likely to terminate in exons than in introns. The general pattern is similar to that observed in yeast, as well as in positive-regulatory domain zinc finger protein 9–knockout mice, suggesting that recombination initiation described here in Mimulus may reflect ancient and conserved eukaryotic mechanisms.Meiotic recombination is a highly regulated process that enables pairing of homologous chromosomes and, by the formation of crossovers, ensures proper segregation (1). Along with mutation, drift, and selection, recombination is a critical factor in shaping genome-wide sequence variation. Recombination rates vary substantially across eukaryote genomes (2) in a manner that we are only beginning to understand. In humans and mice, the location of regions of strong recombination (“hotspots”) is largely determined by positive-regulatory domain zinc finger protein 9 (PRDM9) binding sites (3), whereas in yeast such regions are associated with nucleosome-depleted open chromatin often associated with gene promoters (4). When PRDM9 is disabled in mice, hotspots tend to relocalize to promoter regions (5). In flowering plants, at least one example of a promoter-associated hotspot has been reported (6), but it remains an open question whether this is a general tendency in plants.The positions of crossovers and the boundaries of gene conversion tracts resulting from meiotic recombination are often imprecisely known, because they can only be identified based on the location of nearby segregating markers. Within a species, genome-wide variation in recombination rates can be determined by following the inheritance of such genetic markers in crosses or pedigrees (7–10) or by examining patterns of linkage disequilibrium within a population (11–15). Population-based approaches have the advantage that in diverse populations, hundreds of thousands of historical recombination events can be sampled, compared with only hundreds in the largest pedigrees.The monkeyflower Mimulus guttatus has an exceptionally high nucleotide diversity, which makes it a particularly appealing system for characterizing the boundaries of recombination events. We observed an average pairwise nucleotide difference of π = 2.9% in a sample of 98 wild plants (196 haploid genomes) from four locations within a 16-km radius in the Sierra Nevada foothills in Northern California. At such high diversity, pairs of adjacent SNPs defining local haplotypes are often found on the same Illumina sequencing read (e.g., within 50 bases). Thus, short-range haplotypes can be determined cost-effectively by shotgun sequencing pooled samples rather than by sequencing each plant individually.For a pair of nearby segregating biallelic SNPs we expect to observe only three of the four possible haplotypes unless recombination and/or parallel mutation has occurred since the originating mutations. This is the essence of Hudson’s four-gamete test (16) and allows us to identify putative boundaries of historical crossovers and gene conversion tracts (Fig. 1) to within a fraction of a read length. If this information is combined with population genetic models, local recombination rates can then be inferred (17).Open in a separate windowFig. 1.Appearance of four haplotypes at a pair of SNP loci by recombination. From a single ancestral sequence (Top) a single mutation produces a second haplotype. A second mutation at a nearby site (Middle) generates a third haplotype In the population. Finally, a recombination boundary between the two SNP loci (Bottom) generates a fourth haplotype. Note that the recombination boundary can be due to a crossover event or a gene conversion tract. A fourth haplotype can also appear due to a parallel mutation (not shown), but this scenario can be distinguished from recombination because parallel mutation at a site should not depend on the distance to the nearest SNP.