Paternal Obesity and SGLT2 Inhibition Alter Expression of Placental Regulatory Genes

Abstract

We previously demonstrated that paternal obesity is associated with offspring metabolic risk during later life, and that paternal SGLT2i treatment improves offspring metabolic phenotypes. Since the placenta is a key determinant of prenatal growth and development, we hypothesized the placenta could mediate the impact of paternal obesity and paternal SGLT2i treatment. Male C57BL/6J mice were fed standard chow (Purina 9F) or 60% high-fat diet (HFD, D12492, Research Diet), or 60% HFD plus the SGLT2 inhibitor canagliflozin (CANA, 25 mg/kg/d) for 4 weeks before mating with chow-fed females. Placenta were collected on E16.5, and RNA-seq was performed on placenta from male offspring (paternal chow, pChow, n=4, pHFD, n=5, and pHFD+CANA, n=4), and differentially expressed genes were identified using Limma. Placenta weight was significantly lower in pHFD (0.089±0.004 g, 7 litters from 6 fathers) vs. both pChow (0.108±0.011 g, 4 litters, 4 fathers) and pHFD+CANA (0.107±0.013 g, 5 litters, 5 fathers)(p<0.05). Litter size, fetal or liver weight, or fetal/placental weight ratio did not differ between groups. No genes were differentially expressed in pHFD vs. pChow (FDR<0.1). Gene set enrichment analysis (GSEA) identified significance (FDR<0.05, NES>1.8) for gene sets in steroid metabolic, drug catabolic, and protein-containing complex remodeling processes. Genes responsible for enrichment included cholesterol biosynthesis (Hmgcs1), transport (Apob, Apoa1/2/4, Apom, Apoc1, Vldlr, Pcsk9) and steroid hormone biosynthesis genes (Hsd3b1, Cyp11b1), all upregulated in pHFD by 1.5-3-fold. These results suggest pHFD could potentially affect maternal and fetal cholesterol homeostasis. pHFD+CANA altered expression of 154 genes vs. pHFD (7 up-, 147 down, FDR <0.1, FC >|1.5|); 18 gene sets were downregulated by pHFD+CANA (GSEA NES<-1.8 and FDR<0.05), including the 3 sets upregulated by pHFD. ChEA3 enrichment analysis (ENCODE library) predicted regulation by transcription factors important for cholesterol and sterol biosynthesis (Srebf1/2), embryonic development (Foxa2), & glucose homeostasis (Hnf4g), suggesting these pathways could mediate the “rescue” effect of pHFD+CANA (FDR<0.05). Expression of Foxa2 was significantly downregulated (4-fold) in pHFD+CANA vs. pHFD. We independently analyzed expression of the 78 detected imprinted genes. None were significantly different in pHFD, but both paternally expressed (Nnat) and maternally expressed genes (H19, Phlda2, Meg3, Meg8) were downregulated in pHFD+CANA vs. pHFD by 1.4 to 3.8 fold in pHFD+CANA (p<0.001,FDR<0.1). In summary, paternal SGLT2i reversed the impact of pHFD on placental weight. Robust impact of both pHFD and pSGLT2i on the transcriptome suggests that the placenta is a key mediator of paternal metabolic effects on offspring development and metabolic disease risk, potentially via modification of lipid transport.