Hybridization and introgression between toads with different sex chromosome systems

Abstract

The growing interest in the lability of sex determination in non-model vertebrates such as amphibians and fishes has revealed high rates of sex chromosome turnovers among closely related species of the same clade. Can such lineages hybridize and admix with different sex-determining systems, or could the changes have precipitated their speciation? We addressed these questions in incipient species of toads (Bufonidae), where we identified a heterogametic transition and characterized their hybrid zone with genome-wide markers (RADseq). Adult and sibship data confirmed that the common toad B. bufo is female heterogametic (ZW), while its sister species the spined toad B. spinosus is male heterogametic (XY). Analysis of a fine scale transect across their parapatric ranges in southeastern France unveiled a narrow tension zone (∼10 km), with asymmetric mitochondrial and nuclear admixture over hundreds of kilometers southward and northward, respectively. The geographic extent of introgression is consistent with an expansion of B. spinosus across B. bufo's former ranges in Mediterranean France, as also suggested by species distribution models. However, widespread cyto-nuclear discordances (B. spinosus backrosses carrying B. bufo mtDNA) run against predictions from the dominance effects of Haldane's rule, perhaps because Y and W heterogametologs are not degenerated. Common and spined toads can thus successfully cross-breed despite fundamental differences in their sex determination mechanisms, but remain partially separated by reproductive barriers. Whether and how the interactions of their XY and ZW genes contribute to these barriers shall provide novel insights on the debated role of labile sex chromosomes in speciation.

Keywords: Bufo bufo; Bufo spinosus; RADseq; hybrid zone; reproductive isolation; sex chromosome turnover; speciation.

Figure 1

Sex‐specific presence of RAD variants in B. bufo (from the adult and sib datasets) and B. spinosus (adults only) as inferred by the RADsex approach. Differences reaching statistical significance are squared in red. In B. bufo, the analysis counted more tags specific to females (vertical axis) than males (horizontal axis), but a single one in significant proportions. In B. spinosus, tens of tags are significantly specific to males.

Figure 2

The hybrid zone between B. bufo (blue) and B. spinosus (red) in southern France. The left panel presents our nuclear clustering analyses based on RADseq data (950 species‐diagnostic loci, i. e. fixed between edge populations); the left inset map zooms in the area of contact. The middle panel reports population ancestry obtained from the analysis of intron markers BDNF, POMC, RAG1 and RPL3 (Arntzen et al. 2017); many individuals with B. spinosus or intermediate ancestries at these loci were sampled in ranges that clearly belong to B. bufo (according to the RADseq data). The right panel combines mitochondrial barcoding from ours and Arntzen et al. (2017). The top right inset map shows the distribution of the two species in Western Europe (adapted from Dufresnes 2019).

Figure 3

Geographic clines fitted to the proportion of mitochondrial DNA, nuclear genome average (STRUCTURE Q), and allele frequency at 950 species‐diagnostic SNPs (i.e., fixed between edge populations) along our B. bufo/spinosus transect running from Switzerland (NE) to French Catalonia (SW). Distances are given as the deviation from the median center.

Figure 4

Projected distributions of B. bufo and B. spinosus according to their bioclimatic models. Thick lines show range limits. The area of contact studied here (frame) is suitable for both species, particularly along the Rhône valley for B. spinosus.