Highlights d Wild-type spatial positioning and timely initiation of recombination require ANKRD31 d Selective use of PRDM9 binding sites as DSB sites requires ANKRD31 d Enrichment of pro-DSB factors in the PAR requires ANKRD31 but not IHO1 d Recombination in the PAR depends on ANKRD31

F 1 hybrids between mouse inbred strains PWD and C57BL/6 represent the most thoroughly genetically defined model of hybrid sterility in vertebrates. Hybrid male sterility can be fully reconstituted from three components of this model, the Prdm9 gene, intersubspecific homeology of Mus musculus musculus and Mus musculus domesticus autosomes, and the X-linked Hstx2 locus. Hstx2 modulates the extent of Prdm9-dependent meiotic arrest and harbors two additional factors responsible for intersubspecific introgression-induced oligospermia (Hstx1) and meiotic recombination rate (Meir1). To facilitate positional cloning and to overcome the recombination suppression within the 4.3 Mb encompassing the Hstx2 locus, we designed Hstx2-CRISPR and SPO11/Cas9 transgenes aimed to induce DNA double-strand breaks specifically within the Hstx2 locus. The resulting recombinant reduced the Hstx2 locus to 2.70 Mb (chromosome X: 66.51-69.21 Mb). The newly defined Hstx2 locus still operates as the major X-linked factor of the F 1 hybrid sterility, and controls meiotic chromosome synapsis and meiotic recombination rate. Despite extensive further crosses, the 2.70 Mb Hstx2 interval behaved as a recombination cold spot with reduced PRDM9-mediated H3K4me3 hotspots and absence of DMC1defined DNA double-strand-break hotspots. To search for structural anomalies as a possible cause of recombination suppression, we used optical mapping and observed high incidence of subspecies-specific structural variants along the X chromosome, with a striking copy number polymorphism of the microRNA Mir465 cluster. This observation together with the absence of a strong sterility phenotype in Fmr1 neighbor (Fmr1nb) null mutants support the role of microRNA as a likely candidate for Hstx2.

Orderly segregation of chromosomes during meiosis requires that crossovers form between homologous chromosomes by recombination. Programmed DNA double-strand breaks (DSBs) initiate meiotic recombination. We identify ANKRD31 as a critical component of complexes of DSB-promoting proteins which assemble on meiotic chromosome axes. Genome-wide, ANKRD31 deficiency causes delayed recombination initiation. In addition, loss of ANKRD31 alters DSB distribution owing to reduced selectivity for sites that normally attract DSBs.Strikingly, ANKRD31 deficiency also abolishes uniquely high rates of recombination that normally characterize pseudoautosomal regions (PARs) of X and Y chromosomes.Consequently, sex chromosomes do not form crossovers leading to chromosome segregation failure in ANKRD31-deficient spermatocytes. These defects are accompanied by a genome-wide delay in assembling DSB-promoting proteins on axes and a loss of a specialized PAR-axis domain that is highly enriched for DSB-promoting proteins. Thus, we propose a model for spatiotemporal patterning of recombination by ANKRD31-dependent control of axis-associated complexes of DSB-promoting proteins. Highlights (85 characters max)Temporal and spatial patterning of recombination are regulated by ANKRD31 Selective use of PRDM9 binding sites as DSB hotspots requires ANKRD31 Enrichment of pro-DSB factors in the PAR requires ANKRD31 but not IHO1 Recombination in the PAR critically depends on ANKRD31 are consistent with unpublished data of Acquaviva, Jasin & Keeney (personal communication).Acquaviva et al. observed ANKRD31 aggregates on PARs, and identified chromosome 4, 9 and 13 as autosomes whose non-centromeric ends carry arrays of PAR-like sequences that associate with ANKRD31 aggregates. Distinct molecular requirements for ANKRD31 aggregates and ANKRD31 fociGiven their distinct behaviour in synapsed regions, foci and aggregates of ANKRD31/MEI4/REC114 might represent qualitatively different protein complexes with distinct underlying molecular requirements. To test this possibility we compared localization of ANKRD31, REC114 and MEI4 in Mei4 -/-, Rec114 -/and Iho1 -/spermatocytes ( Figure 2J and S3E-I). The numbers of REC114 and MEI4 foci are strongly reduced in Mei4 -/and Rec114 -/mice, respectively (Kumar et al., 2018). In addition, we found that both focus and aggregate formation of ANKRD31, REC114 and MEI4 were each disrupted in Mei4 -/and Rec114 -/spermatocytes ( Figure S3E-G and Table S3). Remarkably, only the formation of ANKRD31 ( Figure 2I), MEI4 and REC114 foci (Stanzione et al., 2016), but not aggregates, were disrupted in Iho1 -/spermatocytes ( Figure 2J). ANKRD31 aggregates formed efficiently in Iho1 -/spermatocytes; median numbers of ANKRD31 aggregates were four in both wild type (n=62) and Iho1 -/-(n=57) spermatocytes in zygotene. These aggregates always colocalized with aggregates of MEI4 (n=100) and REC114 (n=100) in Iho1 -/spermatocytes ( Figure 2J and S3H).ANKRD31 aggregates also colocalized with PAR FISH signals in late zygotene-like Iho1 -/spe...

Hybrid sterility contributes to speciation by preventing gene flow between related taxa. Prdm9, the first and only hybrid male sterility (HMS) gene known in vertebrates, predetermines the sites of recombination between homologous chromosomes and their synapsis in early meiotic prophase. The asymmetric binding of PRDM9 to heterosubspecific homologs of Mus m. musculus x Mus m. domesticus F1 hybrids and increase of PRDM9-independent DNA double-strand break (DSB) hotspots results in difficult to repair DSBs, incomplete synapsis of homologous chromosomes and meiotic arrest at the first meiotic prophase. Here we show that Prdm9 behaves as a major HMS gene in mice outside the Mus m. musculus x Mus m. domesticus F1 hybrids, in the genomes composed of Mus m. castaneus and Mus m. musculus chromosomes segregating on the Mus m. domesticus background. The Prdm9cst/dom2 (castaneus/domesticus) allelic combination secures meiotic synapsis, testes weight and sperm count within physiological limits, while the Prdm9msc1/dom2 (musculus/domesticus) males show a range of fertility impairment. Out of five quantitative trait loci contributing to the Prdm9msc1/dom2-related infertility, four control either meiotic synapsis or fertility phenotypes and one controls both, synapsis and fertility. Whole-genome genotyping of individual chromosomes showed preferential involvement of nonrecombinant musculus chromosomes in asynapsis in accordance with the chromosomal character of HMS. Moreover, we show that the overall asynapsis rate can be estimated solely from the genotype of individual males by scoring the effect of nonrecombinant musculus chromosomes. Prdm9-controlled HMS represents an example of genetic architecture of HMS consisting of genic and chromosomal components.