Molecular evolution of mammalian genes with epistatic interactions in fertilization

Abstract

Background: Genes that encode proteins associated with sperm competition, fertilization, and sexual conflicts of interest are often among the most rapidly evolving parts of animal genomes. One family of sperm-expressed genes (Zp3r, C4bpa) in the mammalian gene cluster called the regulator of complement activation (RCA) encodes proteins that bind eggs and mediate reproductive success, and are therefore expected to show high relative rates of nonsynonymous nucleotide substitution in response to sexual selection in comparison to other genes not involved in gamete binding at fertilization. We tested that working hypothesis by using phylogenetic models of codon evolution to identify episodes of diversifying positive selection. We used a comparative approach to quantify the evidence for episodic diversifying selection acting on RCA genes with known functions in fertilization (and sensitivity to sexual selection), and contrast them with other RCA genes in the same gene family that function in innate immunity (and are not sensitive to sexual selection).

Results: We expected but did not find evidence for more episodes of positive selection on Zp3r in Glires (the rodents and lagomorphs) or on C4BPA in Primates, in comparison to other paralogous RCA genes in the same taxon, or in comparison to the same orthologous RCA gene in the other taxon. That result was not unique to RCA genes: we also found little evidence for more episodes of diversifying selection on genes that encode selective sperm-binding molecules in the egg coat or zona pellucida (Zp2, Zp3) in comparison to members of the same gene family that encode structural elements of the egg coat (Zp1, Zp4). Similarly, we found little evidence for episodic diversifying selection acting on two other recently discovered genes (Juno, Izumo1) that encode essential molecules for sperm-egg fusion.

Conclusions: These negative results help to illustrate the importance of a comparative context for this type of codon model analysis. The results may also point to other phylogenetic contexts in which the effects of selection acting on these fertilization proteins might be more readily discovered and documented in mammals and other taxa.

Keywords: Coevolution; Gamete recognition; Innate immunity; Positive selection; Sexual selection; Zona pellucida.

Fig. 1

Species trees used in the codon model analyses, including common names for species and higher taxon names for clades, from Springer et al. [65] and Fabre et al. [16]. Organism icons from phylopic.org

Fig. 2

Episodes of diversifying positive selection on three genes from the RCA cluster, including Zp3r and C4bpa (in Glires), and C4BPA (in Primates). Zp3r and C4BPA are known or expected to be expressed in gametes and sensitive to sexual selection, but C4bpa is not. Red branches in each phylogeny show lineages under positive selection in aBSREL analyses. Numbered red vertical bars in each gene cartoon show the locations of codons under positive selection in MEME analyses relative to the total length of the alignment (the grey bar), and relative to several different protein coding domain types. No episodes of diversifying selection were found in C4bpb (in Glires) or C4BPB (in Primates; not shown)

Fig. 3

Episodes of diversifying positive selection on two genes that encode egg coat proteins that bind sperm, including Zp3 and Zp2 (in Glires), and ZP2 (in Primates). Both genes in both taxa are known to be expressed in the egg coat and sensitive to sexual selection. Note that no positively selected sites were identified in ZP2 in Primates. Branches and sites under positive selection, and protein coding domain types, are shown as in Fig. 2. No episodes of diversifying selection were found in ZP3 in Primates (not shown)

Fig. 4

Episodes of diversifying positive selection on an egg coat structural gene (Zp1) that does not bind sperm and is not expected to be sensitive to sexual selection. Branches and sites under positive selection, and protein coding domain types, are shown as in Fig. 2

Fig. 5

Episodes of diversifying positive selection on a second egg coat structural gene (Zp4) that does not bind sperm and is not expected to be sensitive to sexual selection. Branches and sites under positive selection, and protein coding domain types, are shown as in Fig. 2

Fig. 6

Episodes of diversifying positive selection on three genes that encode gamete-fusion proteins, including Juno and Izumo1 (in Glires), and IZUMO1 (in Primates). Both genes are known or expected to be expressed in gametes and sensitive to sexual selection. Note that no positively selected lineages could be identified in Juno (in Glires), and no positively selected sites could be identified in IZUMO1 (in Primates). Branches and sites under positive selection, and protein coding domain types, are shown as in Fig. 2. No episodes of diversifying selection were found in JUNO in Primates (not shown)