Gestational choline supplementation improves cross-generational mood by epigenetic upregulation of Nr3c1 [RNA-Seq]. Mus musculus

Environmental cues during gestation could exert transgenerational effects on behavior mainly through changes in epigenetic marks such as DNA methylation. However, the role of histone modifications in the inheritance of acquired phenotypes remains largely unknown in mammals. Here, we report that gestational choline supplementation (GCS) in maternal mice exerts anxiolytic effects on male offspring across 2 generations. Correspondingly, GCS-induced H3K9 hyperacetylation in the Nr3c1 promoter in male hippocampus leads to upregulation of Nr3c1 and its encoded protein, glucocorticoid receptor (GR), across 2 generations through the male germline. Furthermore, inhibition of CREB-binding protein (CBP) histone acetyltransferase (HAT) function restored GCS-induced epigenetic and behavioral alterations, suggesting that CBP function as a HAT to increase Nr3c1 expression through H3K9 hyperacetylation. Thus, GCS-induced GR upregulation through CBP-mediated H3K9 hyperacetylation in the Nr3c1 promoter is associated with anxiolytic behavior in male offspring, highlighting the role of histone modification in acquired phenotypes across mammalian generations. Overall design: Total RNA was extracted from the hippocampi of F1m-GCS and F1m-CON mice using the RNeasy Mini Kit (QIAGEN, Germany). The RNA quality was assessed using the Bioanalyser 2100 RNA 6000 Nano Kit (Agilent Technologies, USA). mRNA-Seq libraries were generated from total RNA with polyA+ selection of mRNA using the TruSeq RNA Sample Prep Kit v2 (Illumina, USA), and transcriptomes were sequenced using the Hiseq 2000 Sequencing System (Illumina) in paired-end mode. Sequencing adapters and low quality sequencing reads were excluded using the Trim Galore program (http://www.bioinformatics.babraham.ac.uk/projects/trim_galore/). TopHat2(Kim et al., 2013) was used to align the reads to the University of California Santa Cruz (UCSC) mouse mm10 reference genome, allowing up to two mismatches and a Phred-scaled mapping quality ≥ 4. The bioconductor package edgeR(Robinson et al., 2010) was used to identify genes that were differentially expressed between CON and GCS groups based on the number of mapped paired-end reads for each gene counted by the Python package HTSeq (http://www.htseq.org/). The genes for which p values and false discovery rate (FDR) values were below 0.05 were considered differentially expressed between CON and GCS groups. The same RNA-seq and analysis protocols were performed on CON- and SUP-treated HT22 cells.