Convergent recombination suppression suggests role of sexual selection in guppy sex chromosome formation

Abstract

Sex chromosomes evolve once recombination is halted between a homologous pair of chromosomes. The dominant model of sex chromosome evolution posits that recombination is suppressed between emerging X and Y chromosomes in order to resolve sexual conflict. Here we test this model using whole genome and transcriptome resequencing data in the guppy, a model for sexual selection with many Y-linked colour traits. We show that although the nascent Y chromosome encompasses nearly half of the linkage group, there has been no perceptible degradation of Y chromosome gene content or activity. Using replicate wild populations with differing levels of sexually antagonistic selection for colour, we also show that sexual selection leads to greater expansion of the non-recombining region and increased Y chromosome divergence. These results provide empirical support for longstanding models of sex chromosome catalysis, and suggest an important role for sexual selection and sexual conflict in genome evolution.

Figure 1. Distribution of sex differences in coverage and SNP density for all chromosomes.

The X chromosome is in purple. Horizontal and vertical lines denote interquartile ranges.

Figure 2. Male and female coverage characteristics of guppy sex chromosome.

(a) Moving average of coverage differences between male and female reads based on sliding window analysis (window size of 40 scaffolds). Ninety-five per cent confidence intervals based on bootstrapping autosomal estimates are in grey. (b) Male (blue) and female (red) coverage for the X chromosome. For both panels, dark purple indicates the region of the sex chromosomes with the greatest X-Y sequence divergence, where coverage is significantly less in males (Stratum I, 22–25 Mb).

Figure 3. Male and female SNP density and expression differences on guppy sex chromosome.

(a) Moving average of male:female SNP density based on sliding window analysis (window size of 40 scaffolds). Ninety-five per cent confidence intervals based on bootstrapping autosomal estimates are in grey. (b) Male (blue) and female (red) expression of genes along the X chromosome (window size of 40 genes). Dark purple indicates the region of the sex chromosomes with the greatest X-Y sequence divergence, where coverage is significantly less in males (Stratum I, 22–25 Mb) (see Fig. 2), light purple indicates the region with less X-Y differentiation, where there is a significant excess of male SNPs (Stratum II, 15–22 Mb).

Figure 4. Male:female SNP density for the X chromosome across upstream (orange) and downstream (black) guppy populations.

(a,c and e) Moving averages of normalized SNP density across the X chromosome based on sliding window analysis (window size of 40 genes) for Yarra (a), Quare (c) and Aripo (e) watersheds. Ninety-five per cent confidence intervals based on bootstrapping autosomal estimates are in grey. Dark purple indicates the region of the sex chromosomes with the greatest X-Y sequence divergence, where coverage is significantly less in laboratory population males (Stratum I, 22–25 Mb) (see Fig. 2), light purple indicates the region with less X-Y differentiation, where there is a significant excess of male SNPs in laboratory populations (Stratum II, 15–22 Mb) (see Fig. 3). (b,d and f) Distribution of sex differences in normalized SNP density for the X-Y diverged region (Strata I and II, 15–25 Mb) for Yarra (b), Quare (d) and Aripo (f) watersheds. **P-value<0.020, *P-value<0.050 based on permutation tests.