Male caregiving experience alters hippocampal neuroplasticity and transcription independent of reproduction in a biparental species

Abstract

In mammals, measurable changes in brain and behavior accompany the transition to parenthood. In the biparental California mouse, Peromyscus californicus, fathers experience enhanced neuroplasticity, including increased hippocampal dendritic spine density, reduced anxiety, and improved memory. Here, we first investigate whether siring offspring or pup interaction drives structural neuroplasticity in fathers and find that hippocampal spine density is positively associated with caregiving experience, even in the absence of reproduction. Next, we evaluate the transcriptional response to caregiving in the hippocampus of these males and identify 158 differentially expressed genes between fathers and non-fathers, many of which are associated with neurogenesis, neuronal signaling, and dendritic plasticity. Importantly, 11 of the differentially expressed genes in fathers are similarly regulated in non-fathers that interacted with unrelated pups, suggesting that even limited interaction with pups, in the absence of siring a litter, can induce transcriptional changes in the male brain. Moreover, of the differentially expressed genes that encode for transcription factors and cofactors, we repeatedly find that their expression correlates with spine density in males with varied caregiving experience, thus linking these transcriptional changes to neuroplasticity. Together, this work demonstrates that caregiving-regardless of whether the pups are the male's own biological offspring-drives hippocampal neuroplasticity and transcriptional changes in males of a biparental species.

Fig. 1. Association between caregiving behavior and hippocampal neuroplaticity and changes in transcription.

Dendritic spine density and differential gene expression across (A) experimental groups. Virgin males (shown in green) were socially housed with another virgin male for the duration of the study; thus, virgins did not mate or reproduce. Non-fathers (shown in blue) were housed with a sterile, tubally-ligated female, which allows for estrous cycling and mating behavior, thus non-fathers had the opportunity to mate but not reproduce and had no exposure to pups. Pup-sensitized non-fathers (shown in purple) were housed with sterile females; however, these males interacted with unrelated pups to evaluate the effects of caregiving in the absence of reproduction. First-time fathers (shown in red) were housed with a reproductive female and offspring from their first litter, whereas experienced fathers (shown in orange) were housed with a reproductive female and offspring from their second litter, thus fathers were able to mate, reproduce, and provide parental care. B Left and center images are a representation of granule cell neurons in the dentate gyrus of the hippocampus (outlined in red) and arrows in the right image identify individual dendritic spines. C Dendritic spine density of granule cell neurons in the dentate gyrus of the hippocampus among virgin (N = 13), non-father (N = 13), pup-sensitized non-father (N = 9), first-time father (N = 14), and experienced father (N = 12) California mouse males with variable mating, reproductive, and caregiving experience. Boxplots display the median (center line), interquartile range (IQR; 25th to 75th percentile, shown as the box), and whiskers extending to the smallest and largest data points within 1.5×IQR. Different letters above data points indicate statistically significant differences between experience groups (P < 0.05) determined using a one-way ANOVA followed by Tukey’s HSD. Groups that share the same letter are not significantly different from each other. D Number of differentially expressed genes (DEGs) between first-time fathers (N = 12) and non-fathers (N = 6), first-time fathers and pup-sensitized non-fathers (N = 6), and pup-sensitized non-fathers and non-fathers. The gene names in the box are DEGs that are shared between first-time fathers versus non-fathers and pup-sensitized non-fathers versus non-fathers. E DESeq2 DEGs between non-fathers and first-time fathers (top) and non-fathers and pup-sensitized non-fathers (bottom). In these volcano plots, statistically significant genes (P < 0.05) are represented above the horizontal dashed line and the log₂ fold change cutoff at −1 and 1 is indicated by the vertical dashed lines. The identified genes are transcription factors and cofactors that are correlated with spine density. F Scatterplots depicting correlations between dendritic spine density and the expression of DEGs (non-fathers versus first-time fathers DEGs) that encode transcription factors and co-factors, in non-fathers, pup-sensitized non-fathers, and first-time fathers. Non-fathers are shown in blue, pup-sensitized non-fathers are shown in purple, and first-time fathers are shown in red. P value is adjusted for multiple comparisons testing and Spearman rho correlation coefficient is provided.

Fig. 2. Gene ontology and functional annotation clustering of genes differentially expressed in first-time fathers and non-fathers.

A GO terms associated with differentially expressed genes between non-fathers and first-time fathers identified in the GO enrichment analysis. List of cellular components includes parent terms and subterms, whereas biological processes only include parent terms. B GO terms, Swiss-Prot keywords, and protein domains enriched in first-time fathers compared to non-fathers identified in the DAVID Functional Annotation Clustering analysis. The P value cutoff of 0.05 is indicated by the hashed line and the fold enrichment is indicated by the shade of red with larger fold enrichments depicted in dark red and smaller fold enrichments depicted in light red.

Fig. 3. Functional pathways enriched in first-time fathers and pup-sensitized non-fathers.

KEGG pathways and Reactome terms enriched in first-time fathers (red arrows) compared to non-fathers (blue arrows) and pup-sensitized fathers (purple arrows) compared to non-fathers, as predicted by KOBAS analysis. Up arrows indicate upregulation, and down arrows indicate downregulation of gene expression in the respective group. Pathways enriched in C57BL/6 J mothers in Ray et al. and BXD recombinant inbred mothers in Ashbrook et al. are highlighted in yellow.

Fig. 4. Modules of co-expressed genes associated with caregiving as well as behavioral and neurophysiological traits.

A WGCNA correlations between novel modules of co-expressed genes and experimental groups. B Duration of pup grooming and (C) nest building in pup-sensitized non-fathers and first-time fathers during caregiving assay from Colt et al.. D WGCNA correlations between the same modules of co-expressed genes and the behavioral traits and dendritic spine density. The heatmap color scale represents the correlation coefficient of each module’s eigengene expression and the experimental group (blue represents a negative correlation, white represents no correlation, and red represents a positive correlation).