Interacting impact of maternal inflammatory response and stress on the amygdala transcriptome of pigs

Abstract

Changes at the molecular level capacitate the plasticity displayed by the brain in response to stress stimuli. Weaning stress can trigger molecular changes that influence the physiology of the offspring. Likewise, maternal immune activation (MIA) during gestation has been associated with behavior disorders and molecular changes in the amygdala of the offspring. This study advances the understanding of the effects of pre- and postnatal stressors in amygdala gene networks. The amygdala transcriptome was profiled on female and male pigs that were either exposed to viral-elicited MIA or not and were weaned or nursed. Overall, 111 genes presented interacting or independent effects of weaning, MIA, or sex (FDR-adjusted P-value <0.05). PIGY upstream reading frame and orthodenticle homeobox 2 are genes associated with MIA-related neurological disorders, and presented significant under-expression in weaned relative to nursed pigs exposed to MIA, with a moderate pattern observed in non-MIA pigs. Enriched among the genes presenting highly over- or under-expression profiles were 24 Kyoto Encyclopedia of Genes and Genomes pathways including inflammation, and neurological disorders. Our results indicate that MIA and sex can modulate the effect of weaning stress on the molecular mechanisms in the developing brain. Our findings can help identify molecular targets to ameliorate the effects of pre- and postnatal stressors on behaviors regulated by the amygdala such as aggression and feeding.

Keywords: RNA-seq; double-hit hypothesis; gestational stress; weaning stress.

Figure 1

Heatmap listing the differential expression [log₂(fold change between pig groups)] of genes that presented at least one significant interaction or main effect of weaning, maternal immune activation at P-value <5.0E-05 and |log₂(fold change between pig groups)|>2.0. ^aFunctional annotation of the gene; N, neuro-associated or I, immune-associated function. ^bWeaning×MIA: log₂(fold change between pig groups) characterizing the weaning-by-maternal immune activation effect; Weaning×Sex: log₂(fold change between pig groups) characterizing the weaning-by-sex effect; MIA×Sex: log₂(fold change between pig groups) characterizing the maternal immune activation-by-sex effect. ^cLog₂(fold change of pig group X relative to group Y)=log₂(X/Y)=log₂(X) – log₂(Y). The pig groups compared included: non-MIA or control nursed pigs (CN), non-MIA weaned pigs (CW), PRRSV-elicited MIA nursed pigs (PN), MIA weaned pigs (PW), nursed females (NF), nursed males (NM), weaned females (WF), weaned males (PM), non-MIA females (CF), non-MIA males (CM), PRRSV-elicited MIA females (PF), MIA males (PM). ^dRed (green) denotes under(over)-expression in the X or first pig group relative to the Y or second pig group.

Figure 2

Gene network submodules of the gene orthodenticle homeobox 2 (A, C) and the gene vasoactive intestinal peptide receptor 2 (B, D). Modules (A) and (B) depict under- (blue) and over- (red) expression between MIA weaned and nursed pigs while modules (C) and (D) depict under- (blue) and over- (red) expression between non-MIA weaned and nursed pigs.