X chromosome control of meiotic chromosome synapsis in mouse inter-subspecific hybrids

Abstract

Hybrid sterility (HS) belongs to reproductive isolation barriers that safeguard the integrity of species in statu nascendi. Although hybrid sterility occurs almost universally among animal and plant species, most of our current knowledge comes from the classical genetic studies on Drosophila interspecific crosses or introgressions. With the house mouse subspecies Mus m. musculus and Mus m. domesticus as a model, new research tools have become available for studies of the molecular mechanisms and genetic networks underlying HS. Here we used QTL analysis and intersubspecific chromosome substitution strains to identify a 4.7 Mb critical region on Chromosome X (Chr X) harboring the Hstx2 HS locus, which causes asymmetrical spermatogenic arrest in reciprocal intersubspecific F1 hybrids. Subsequently, we mapped autosomal loci on Chrs 3, 9 and 13 that can abolish this asymmetry. Combination of immunofluorescent visualization of the proteins of synaptonemal complexes with whole-chromosome DNA FISH on pachytene spreads revealed that heterosubspecific, unlike consubspecific, homologous chromosomes are predisposed to asynapsis in F1 hybrid male and female meiosis. The asynapsis is under the trans- control of Hstx2 and Hst1/Prdm9 hybrid sterility genes in pachynemas of male but not female hybrids. The finding concurred with the fertility of intersubpecific F1 hybrid females homozygous for the Hstx2(Mmm) allele and resolved the apparent conflict with the dominance theory of Haldane's rule. We propose that meiotic asynapsis in intersubspecific hybrids is a consequence of cis-acting mismatch between homologous chromosomes modulated by the trans-acting Hstx2 and Prdm9 hybrid male sterility genes.

Figure 1. Single QTL mapping of Hstx2 on Chr X.

(A) QTL analysis of testes weight in the (B6.PWD-Chr X×B6)F1×PWD cross showed a 1.5-LOD support interval between 34.6 cM to 35.3 cM on Chr X. (B) Distribution of testes weight of males carrying PWD or B6 allele of DXMit87 marker with LOD score 30. (C) QTL analysis of sperm count in ductus epididymis shows the same 1.5 LOD support interval as for testes weight. (D) Distribution of sperm count of males carrying PWD or B6 allele of DXMit87 marker (Chr X: 66.65 Mb, GRCm38) with LOD score above 20.

Figure 2. Fine mapping of Hstx1 and Hstx2 HS loci on Chr X using partial consomic strains.

The partial consomic B6.PWD-Chr X.# females were crossed with B6 or PWD males for mapping Hstx1 and Hstx2, respectively. Testes weight and sperm count were used as fertility phenotypes. The borders of introgressed PWD sequence (black) were determined by MegaMUGA genotyping for B6.PWD-Chr X.1 (abbreviated here X.1), B6.PWD-Chr X.1s (X.1s) and B6.PWD-Chr X.2 (X.2). For B6.PWD-Chr X.3 (X.3) mapping see . The map positions correspond to genome assembly GRCm38, megabase scale, for details see Table S1.

Figure 3. Expression profiling of a cluster of MiRNA genes within the Hstx1/Hstx2 critical region.

(A) Log fold-change of expression ratio of PWD versus B6 MiRNA genes in flow-sorted testicular cells. Significant overexpression of Mir465 cluster in PWD germ cells is shown in red. (B) Validation of Mir465 overexpression in PWD primary spermatocytes by qRT PCR. Data normalized to U6 non-coding RNA (C) Log fold-change of expression ratio of MiRNA genes in (PWD×B6)F1 versus (B6×PWD)F1 (abbreviated PB6F1 and B6PF1) 14.5 d old testes. Significant upregulation in red (P<0.05). (D) qRT PCR validation of differences in Mir465 expression. Data normalized to Mir152.

Figure 4. Single QTL scan for autosomal loci supporting Hstx2 independent intrameiotic arrest of (Mmd×Mmm)F1 intersubspecific hybrids.

(A) Testes weight QTLs (red) reached significance on Chrs 3 (marker UNC030163295, Chr3: 104,320,699) and 13 (JAX00357337 Chr 13:47,975,634) and QTLs for sperm count on Chrs 3, 9 (JAX00171568, Chr 9:56,491,601) and 13. Sperm count was evaluated as a binary trait (SC = 0, SC>0). (B) and (C) Additive effect of QTLs on testes weight and sperm count. For map positions and possible candidate genes see Table S4.

Figure 5. Super-resolution microscopy of synaptonemal complexes on spreads of (B6.PWD-Chr X.1s×PWD) pachytene spermatocytes.

(A) Detail of a pachytene spermatocyte of a sterile male immunostained by SYCP3 (red) and SYCP1 (green) antibodies. Properly synapsed bivalents (arrows) show two parallel threads of transverse filaments decorated by SYCP1 antibody which form the central region embedded in SYCP3 lateral elements. Unsynapsed chromosomes lack transverse filaments but display some irregular SYCP1 spots (arrowheads). (B) Example of a nonhomologous pairing and/or translocations, and asynapsis in pachynema of (B6.Chr X.1s×PWD)F1 sterile male. Bar 2000 nM.

Figure 6. Meiotic asynapsis in female hybrids.

Abbreviations of consomic strains and their hybrids: DX.1 – B6.PWD-Chr X.1; DX.1PF1 – (B6.PWD-Chr X.1×PWD)F1; DX.1sPF1 – (B6.PWD-Chr X.1s×PWD)F1; DX.1B6F1 – (B6.PWD-Chr X.1×B6)F1; DX.1sB6F1 – (B6.PWD-Chr X.1s×B6)F1; DX.1sD17F1 – (B6.PWD-Chr X.1s×B6.PWD-Chr 17)F1; D17B6F1 – (B6.PWD-Chr 17×B6)F1; B6PF1 – (B6×PWD)F1, PB6F1 – (PWD×B6)F1; PD17F1 – (PWD×B6.PWD-Chr 17)F1. (A) Chromosome synapsis in pachytene oocytes of B6 and (B6.PWD-Chr X.1s×PWD)F1 18.5–19.5 dpc female fetuses was analyzed by combination of SYCP1, SYCP3 and CREST (centromeric heterochromatin) immunostaining or by HORMAD2 and SYCP3 to detect unsynapsed chromosomes. Bar, 10 µm. (B) The frequency of oocytes showing one or more asynaptic chromosomes is similar (>40%) irrespective of Hstx2 and Prdm9/Hst1 genotype. (C) Although the (PWD×B6), (B6.PWD-Chr X.1s×PWD), (B6.PWD-Chr X.1×PWD) and (B6.PWD-Chr 17×PWD)F1 hybrid females do not differ in percentage of pachytene oocytes with asynapsis, the (B6.PWD-Chr 17×PWD)F1 females, conspecific for Chr 17^PWD, carry significantly less asynapsed chromosomes per cell. (D) The frequency of diplonemas in spread oocyte preparations was significantly lower (p<0.01, χ² test) in intersubspecific hybrids than in parental inbred strains..

Figure 7. The proposed sequence of events leading to male limited sterility of intersubspecific hybrids of house mouse.

Susceptibility of heterosubspecific homologs to asynapsis is common to both sexes. Prdm9/Hst1 and Hstx2 hybrid sterility genes can modulate this sensitivity from 0% to >95% in spermatogenesis but not in oogenesis, depending on allelic combinations of epistatic DMIs. Multiple asynaptic autosomes provoke MSCI, contributing to hybrid male sterility. Approximately one half of unaffected oocytes ensure fertility of hybrid females. It remains to be established what is the cause of asynapsis of heterosubspecific homologs.