Identification of General Patterns of Sex-Biased Expression in Daphnia, a Genus with Environmental Sex Determination

Abstract

Daphnia reproduce by cyclic-parthenogenesis, where phases of asexual reproduction are intermitted by sexual production of diapause stages. This life cycle, together with environmental sex determination, allow the comparison of gene expression between genetically identical males and females. We investigated gene expression differences between males and females in four genotypes of Daphnia magna and compared the results with published data on sex-biased gene expression in two other Daphnia species, each representing one of the major phylogenetic clades within the genus. We found that 42% of all annotated genes showed sex-biased expression in D. magna This proportion is similar both to estimates from other Daphnia species as well as from species with genetic sex determination, suggesting that sex-biased expression is not reduced under environmental sex determination. Among 7453 single copy, one-to-one orthologs in the three Daphnia species, 707 consistently showed sex-biased expression and 675 were biased in the same direction in all three species. Hence these genes represent a core-set of genes with consistent sex-differential expression in the genus. A functional analysis identified that several of them are involved in known sex determination pathways. Moreover, 75% were overexpressed in females rather than males, a pattern that appears to be a general feature of sex-biased gene expression in Daphnia.

Keywords: Daphnia; Differential gene expression; ESD; Genetics of Sex; females; males.

Figure 1

Heatmap showing the normalized expression levels of the sex-DE genes (p < 0.05) with at least a two- fold expression difference between males and females. Each line represents a gene and each column specific sample (genotype and sex), with relative expression levels indicated by color (from highly overexpressed, red to highly underexpressed, blue, as indicated by the scale to the right). The dendrogram above the sample columns indicates clustering according to the Euclidean distance matrix implemented in the pheatmap R.

Figure 2

A. Venn diagram showing the number of sex-DE genes among the 7453 one-to one orthologs in each of the three species of Daphnia. B. Number of sex-DE genes being biased in the same vs. opposite directions. Panels from left to right: genes being sex-DE all three species (707 genes), genes being sex-DE only in D. magna and D. pulex (1402 genes), D. magna and D. galeata, (349 genes), and D. pulex and D. galeata (142 genes). C. Number of female-biased and male-biased genes in each of the four categories depicted in Figure 2 B. (only genes being biased in the same direction).

Figure 3

Proportion of genes with different degrees of sex-bias (the degree of sex-bias is summarized in four categories of fold change). Panels from left to right: genes being sex biased in all three species (707 genes), in two species (1751 genes), and only in D. magna (2360 genes). Only the 7453 genes, for which single-copy, one-to-one orthologs could be identified in all three species were considered for this analysis.

Figure 4

Composition and hierarchical organization of GO terms associated to the 675 genes with consistent sex-biased gene expression in all three species. n_obs: number of genes in the given GO category.

Figure 5

Enrichment analysis of GO terms among the core-set of genes compared to the entire list of D. magna genes. Shown are over-represented (Bonferroni-corrected p < 0.05) terms and functional categories (the latter distinguished by color) for molecular functions (A) and biological processes (B). The size of each rectangle is proportional to the –log(p-value) for its category.