Harnessing natural variation to identify cis regulators of sex-biased gene expression in a multi-strain mouse liver model

摘要: Sex differences in gene expression are widespread the liver, where a large number of autosomal factors act tandem with growth hormone signaling to regulate individual variability sex liver metabolism and disease. Here, we compare hepatic transcriptomic epigenetic profiles mouse strains C57Bl/6J CAST/EiJ, representing two subspecies separated by 0.5-1 million years evolution, elucidate actions genetic regulating differences. We identify 144 protein coding genes 78 lncRNAs showing strain-conserved bias; many have ontologies relevant function, more highly liver-specific show greater bias, proximally regulated than whose bias is strain-dependent. The include key hormone-dependent transcriptional regulators however, three other transcription factors, Trim24, Tox, Zfp809, lose sex-biased CAST/EiJ liver. To these strain specificities expression, characterized strain-dependence chromatin opening enhancer marks at cis regulatory elements (CREs) within quantitative trait loci (eQTL) genes. Strikingly, 208 286 eQTLs strain-specific, sex-differential effects on were associated complete gain, loss, or reversal between strains. Moreover, 166 linked strain-specific gain loss localized CREs. Remarkably, subset CREs lacked variants yet showed coordinated, strain-dependent regulation. Thus, directly link hundreds high CRE activity genes, uncover underlying genetically-determined states controlling genetically diverse populations. Author summaryMale-female confer biological processes human health disease, but difficult model inbred mice given their identical backgrounds. Outbred provide some variability, cross-strain studies rodents not been well studied. Here Diversity founder strains, C57Bl/6 CAST/EiJ. find that can be element one sequence variants. cases, variants, was subject harnessing power naturally occurring diversity mice, integrated data genetic, epigenetic, levels across evolutionary divergent discover genomic regions control phenotypic These findings may serve as for variation effect population-wide