Maternal diets trigger sex-specific divergent trajectories of gene expression and epigenetic systems in mouse placenta

Abstract

Males and females responses to gestational overnutrition set the stage for subsequent sex-specific differences in adult onset non communicable diseases. Placenta, as a widely recognized programming agent, contibutes to the underlying processes. According to our previous findings, a high-fat diet during gestation triggers sex-specific epigenetic alterations within CpG and throughout the genome, together with the deregulation of clusters of imprinted genes. We further investigated the impact of diet and sex on placental histology, transcriptomic and epigenetic signatures in mice. Both basal gene expression and response to maternal high-fat diet were sexually dimorphic in whole placentas. Numerous genes showed sexually dimorphic expression, but only 11 genes regardless of the diet. In line with the key role of genes belonging to the sex chromosomes, 3 of these genes were Y-specific and 3 were X-specific. Amongst all the genes that were differentially expressed under a high-fat diet, only 16 genes were consistently affected in both males and females. The differences were not only quantitative but remarkably qualitative. The biological functions and networks of genes dysregulated differed markedly between the sexes. Seven genes of the epigenetic machinery were dysregulated, due to effects of diet, sex or both, including the Y- and X-linked histone demethylase paralogues Kdm5c and Kdm5d, which could mark differently male and female epigenomes. The DNA methyltransferase cofactor Dnmt3l gene expression was affected, reminiscent of our previous observation of changes in global DNA methylation. Overall, this striking sexual dimorphism of programming trajectories impose a considerable revision of the current dietary interventions protocols.

Figure 1. A plot of the RT-qPCR and microarray values.

for the gene expression coefficient (log2 of ratios) for the 4 different comparisons (F HFD vs F CD, M HFD vs M CD, M CD vs F CD and M HFD vs F HFD) for 21 genes (Hsd11b1, Hsd11b2, Cxcl1, Prap1, Slc13a4, Slc22a3, Sult1e1, Tph1, Gcm1, Jarid1c, Suv39h1, Suv39h2, Dnmt3l, Maoa, C3, Pdgfß, Drd4, Vdr, Prl7a1, Gzmb, CD81). The RT-qPCR data correlated with the microarray data for most of the 84 coefficient, with a linear equation: y = 1.07x and a high correlation coefficient (r = 0.775). Spearman’s test indicated that this correlation was highly significant (p<10-5).

Figure 2. Venn diagram representing the number of genes in LIMMA statistical analysis.

(A) displaying significant sexual dimorphism or (B) significantly dysregulated under the influence of maternal high-fat diet.

Figure 3. Western blotting with antibodies specific for H3K4me3 and H3K9me3.

was used to determine the degree of lysine-specific methylation, in relationship with Kdm5c/5d (H3K4me3 demethylase) and Kmt1a/1b (H3K9 trimethylases). No sex- or diet-specific signal was observed.