Inheritance of social dominance is associated with global sperm DNA methylation in inbred male mice

Abstract

Dominance relationships between males and their associated traits are usually heritable and have implications for sexual selection in animals. In particular, social dominance and its related male pheromones are heritable in inbred mice; thus, we wondered whether epigenetic changes due to altered levels of DNA methylation determine inheritance. Here, we used C57BL/6 male mice to establish a social dominance-subordination relationship through chronic dyadic encounters, and this relationship and pheromone covariation occurred in their offspring, indicative of heritability. Through transcriptome sequencing and whole-genome DNA methylation profiling of the sperm of both generations, we found that differential methylation of many genes was induced by social dominance-subordination in sires and could be passed on to the offspring. These methylated genes were mainly related to growth and development processes, neurodevelopment, and cellular transportation. The expression of the genes with similar functions in whole-genome methylation/bisulfite sequencing was also differentiated by social dominance-subordination, as revealed by RNA-seq. In particular, the gene Dennd1a, which regulates neural signaling, was differentially methylated and expressed in the sperm and medial prefrontal cortex in paired males before and after dominance-subordination establishment, suggesting the potential epigenetic control and inheritance of social dominance-related aggression. We suggest that social dominance might be passed on to male offspring through sperm DNA methylation and that the differences could potentially affect male competition in offspring by affecting the development of the nervous system.

Keywords: WGBS; aggression; epigenetic inheritance; pheromone; social hierarchy.

Figure 1.

Body weights of paired male mice and their behavior during daily dyadic encounters for 21 consecutive days. (A) Body weights of F0 males before and after 21 days of dyadic encounters. (B and D) Frequency of attacks of respective F0 and F1 males. (C and E) Frequency of defences of respective F0 and F1 males. The individual that attacks more frequently in 21 days is defined as the dominant, and its opponent is defined as the subordinate (mean ± SE, n = 15, *P < 0.05, using paired t-test).

Figure 2.

Locomotion of F0 experimental group (dominant-subordinate pairs which have been subjected to 21 days of daily dyadic encounters) in open field. (A) Time spent in the centre zone. (B) Distance travelled in the open field (mean ± SE, n = 15, *P < 0.05, using paired t-test).

Figure 3.

Comparisons of hepatic Darcin expression levels (A) and urinary Darcin protein levels (B) between dominant and subordinate males of F0 experimental group (repeated dyadic encounters for 21 consecutive days), and between winners and losers of F0 control group (a single dyadic encounter on the 21st day). Comparisons of hepatic Darcin expression levels (C) and urinary Darcin protein levels (D) between dominant and subordinate males of F1 experimental group, and between winners and losers of F1 control group (mean ± SE, n = 3 for hepatic mRNA, n = 4 for urinary protein, *P ≤ 0.05, **P < 0.01, using paired t-test).

Figure 4.

Distribution of annotated DMRs on genomic regions for F0 experimental group (dominant-subordinate pairs which have been subjected to 21 days of daily dyadic encounters) (A) and control group (winner-loser pairs which have been subjected to a single dyadic encounter on the 21st day) (B). Distribution of annotated differentially methylated CpG sites for F0 experimental group (C) and F0 control group (D) (n = 2).

Figure 5.

GO (A and B) and KEGG (C) enrichment analyses of the DMGs between dominant and subordinate F0 males. Enrichment analyses of the hyper-methylated genes in dominant (D) and subordinate (E) F0 males. Complete list is shown in Supplementary Table S1.

Figure 6.

Enrichment analysis of the DMGs of F1 control group (winner-loser pairs which have been subjected to a single dyadic encounter). Complete list is shown in Supplementary Table S2.

Figure 7.

Enrichment analyses of the common DMGs between F0 experimental group (dominant-subordinate pairs which have been subjected to 21 days of daily dyadic encounters) and F1 control group (winner-loser pairs which have been subjected to a single dyadic encounter). Enrichment analyses are based on GO, KEGG (A and B) and protein-protein interaction (C).

Figure 8.

DNA methylation levels (A) and sperm mRNA expression levels (B) of Dennd1a in F0 experimental group (dominant-subordinate pairs which have been subjected to 21 days of daily dyadic encounters) and F0 control group (winner-loser pairs which have been subjected to a single dyadic encounter on the 21st day). The expression levels of Rab35 in hippocampus (C) and mPFC (D) of F0 experimental group and F0 control group (DNA methylation, n = 2, *differences > 25%, **differences > 50%; mRNA expression, n = 3, *P < 0.05, using paired t-test).

Figure 9.

Enrichment analyses of sperm DEGs of F0 experimental group (dominant-subordinate pairs which have been subjected to 21 days of daily dyadic encounters). Enrichment analyses for the genes which had higher (A) or lower (B) expression levels in dominant males than in subordinate males based on GO; and enrichment analysis for the DEGs based on protein-protein interaction (C). Complete list is shown in Supplementary Table S3.

Figure 10.

Enrichment analyses of sperm DEGs of F1 experimental group (dominant-subordinate pairs which have been subjected to 21 days of daily dyadic encounters) based on GO and KEGG (A), and protein-protein interactions (B). Complete list is shown in Supplementary Table S4.

Figure 11.

Methylation signal distribution of Dennd1a gene in F0 experimental group. D1 and D2 represent the dominant group, S1 and S2 represent the subordinate group.