Changes of gene expression but not cytosine methylation are associated with male parental care reflecting behavioural state, social context and individual flexibility

Abstract

Behaviour is often a front line response to changing environments. Recent studies show behavioural changes are associated with changes of gene expression; however, these studies have primarily focused on discrete behavioural states. We build on these studies by addressing additional contexts that produce qualitatively similar behavioural changes. We measured levels of gene expression and cytosine methylation, which is hypothesized to regulate the transcriptional architecture of behavioural transitions, within the brain during male parental care of the burying beetle Nicrophorus vespilloides in a factorial design. Male parenting is a suitably plastic behaviour because although male N. vespilloides typically do not provide direct care (i.e. feed offspring) when females are present, levels of feeding by a male equivalent to the female can be induced by removing the female. We examined three different factors: behavioural state (caring versus non-caring), social context (with or without a female mate) and individual flexibility (if a male switched to direct care after his mate was removed). The greatest number of differentially expressed genes were associated with behavioural state, followed by social context and individual flexibility. Cytosine methylation was not associated with changes of gene expression in any of the factors. Our results suggest a hierarchical association between gene expression and the different factors, but that this process is not controlled by cytosine methylation. Our results further suggest that the extent a behaviour is transient plays an underappreciated role in determining its underpinning molecular mechanisms.

Keywords: DNA methylation; Epigenetics; Nicrophorus vespilloides; Social behaviour; Social neuroscience.

Fig. 1.

Principal component analysis of gene expression of male Nicrophorus vespilloides with individual samples coloured by behavioural state; caring (black, n=10) versus non-caring (red, n=12) normalized with DESeq2. The graph clearly shows principal component 1 as an axis of separation for this contrast.

Fig. 2.

Venn diagram showing the overlap of significantly differentially expressed genes in male N. vespilloides between the three contrasts analyzed: behavioural state, social context and individual flexibility.

Fig. 3.

Venn diagram showing the large overlap between the methylated genes of adult and larval N. vespilloides, using only genes that had high sequencing coverage amongst all samples to adjust for differences of sequencing depth between adult and larval samples.

Fig. 4.

Venn diagram showing the overlap between methylated adult genes in male N. vespilloides and the differentially expressed genes (DEGs) between the caring versus non-caring contrast.