Patterns of molecular evolution in a parthenogenic terrestrial isopod (Trichoniscus pusillus)

Abstract

The "paradox of sex" refers to the question of why sexual reproduction is maintained in the wild, despite how costly it is compared to asexual reproduction. Because of these costs, one might expect nature to select for asexual reproduction, yet sex seems to be continually selected for. Multiple hypotheses have been proposed to explain this incongruence, including the niche differentiation hypothesis, the Red Queen hypothesis, and accumulation of harmful mutations in asexual species due to inefficient purifying selection. This study focuses on the accumulation of mutations in two terrestrial isopods, Trichoniscus pusillus, which has sexual diploid and parthenogenic triploid forms, and Hyloniscus riparius, an obligately sexual relative. We surveyed sex ratios of both species in an upstate New York population and obtained RNA-seq data from wild-caught individuals of both species to examine within- and between-species patterns of molecular evolution in protein-coding genes. The sex ratio and RNA-seq data together provide strong evidence that this T. pusillus population is entirely asexual and triploid, while the H. riparius population is sexual and diploid. Although all the wild T. pusillus individuals used for sequencing shared identical genotypes at nearly all SNPs, supporting a clonal origin, heterozygosity and SNP density were much higher in T. pusillus than in the sexually reproducing H. riparius. This observation suggests this parthenogenic lineage may have arisen via mating between two divergent diploid lineages. Between-species sequence comparisons showed no evidence of ineffective purifying selection in the asexual T. pusillus lineage, as measured by the ratio of nonsynonymous to synonymous substitutions (dN/dS ratios). Likewise, there was no difference between T. pusillus and H. riparius in the ratios of nonsynonymous to synonymous SNPs overall (pN/pS). However, pN/pS ratios in T. pusillus were significantly higher when considering only SNPs that may have arisen via recent mutation after the transition to parthenogenesis. Thus, these recent SNPs are consistent with the hypothesis that purifying selection is less effective against new mutations in asexual lineages, but only over long time scales. This system provides a useful model for future studies on the evolutionary tradeoffs between sexual and asexual reproduction in nature.

Keywords: Oniscidea; Paradox of sex; Parthenogenesis; Trichoniscidae; dN/dS; pN/pS.

Figure 1. Allele read count ratio distribution.

Histogram illustrating the distribution of the proportion of reads containing the reference allele (as opposed to the alternate allele) at each SNP in heterozygous individuals within each species. In T. pusillus, there are two peaks at approximately 0.33 and 0.66 (vertical dashed lines), consistent with triploid genotypes. In H. riparius and T. rathkei, there is one peak at approximately 0.5 (vertical dashed lines), consistent with diploid genotypes.

Figure 2. dN, dS, and dN/dS.

Violin plots of dN, dS, and dN/dS ratios, resulting from pairwise comparisons between T. pusillus vs. T. rathkei, and H. riparius vs. T. rathkei. For dN, the T. pusillus - T. rathkei median was 0.161; H. riparius - T. rathkei median was 0.152; Wilcoxon signed rank test p = 1.4 × 10⁻⁶⁹. For dS, the T. pusillus - T. rathkei median was 7.22; H. riparius - T. rathkei median was 6.60; Wilcoxon signed rank test p = 4.0 × 10⁻¹¹. For dN/dS, the T. pusillus - T. rathkei median was 0.019; H. riparius - T. rathkei median was 0.021; Wilcoxon signed rank test p = 1.2 × 10⁻⁶.