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

Abstract

Sex differences in gene expression are widespread in the liver, where many autosomal factors act in tandem with growth hormone signaling to regulate individual variability of sex differences in liver metabolism and disease. Here, we compare hepatic transcriptomic and epigenetic profiles of mouse strains C57BL/6J and CAST/EiJ, representing two subspecies separated by 0.5–1 million years of evolution, to elucidate the actions of genetic factors regulating liver sex differences. We identify 144 protein coding genes and 78 lncRNAs showing strain-conserved sex bias; many have gene ontologies relevant to liver function, are more highly liver-specific and show greater sex bias, and are more proximally regulated than genes whose sex bias is strain-dependent. The strain-conserved genes include key growth hormone-dependent transcriptional regulators of liver sex bias; however, three other transcription factors, Trim24, Tox, and Zfp809, lose their sex-biased expression in CAST/EiJ mouse liver. To elucidate the observed strain specificities in expression, we characterized the strain-dependence of sex-biased chromatin opening and enhancer marks at cis regulatory elements (CREs) within expression quantitative trait loci (eQTL) regulating liver sex-biased genes. Strikingly, 208 of 286 eQTLs with strain-specific, sex-differential effects on expression were associated with a complete gain, loss, or reversal of the sex differences in expression between strains. Moreover, 166 of the 286 eQTLs were linked to the strain-dependent gain or loss of localized sex-biased CREs. Remarkably, a subset of these CREs apparently lacked strain-specific genetic variants yet showed coordinated, strain-dependent sex-biased epigenetic regulation. Thus, we directly link hundreds of strain-specific genetic variants to the high variability in CRE activity and expression of sex-biased genes and uncover underlying genetically-determined epigenetic states controlling liver sex bias in genetically diverse mouse populations.

Chromosomal and gonadal sex drive sex differences in lipids and hepatic gene expression in response to hypercholesterolemia and statin treatment

“…We focused on liver because of its central role in regulating homeostatic lipid levels and as the primary target for statin inhibition of cholesterol synthesis. Our study extends and augments previous work identifying sex differences in the hepatic transcriptome by assessing the role of sex chromosomes in hepatic gene regulation in physiological states that are relevant to human disease—hypercholesterolemia and statin treatment [ 29 – 32 ].…”

Regulation of Sexually Dimorphic Expression of Major Urinary Proteins

“…Determining the origins of sexually dimorphic MUP expression requires comparing MUP expression in both sexes among rodent species. The phylogeny MUP genes have been described ( Stopka et al, 2012 ); however, few studies have compared urinary protein output ( Nazarova et al, 2018 ) or MUP gene expression between the sexes in different Mus species ( Sheehan et al, 2019 ; Matthews et al, 2021 ). Several studies have compared the MUPs of two European Mus subspecies and examined the hypothesis that divergence of these genes among populations promotes speciation.…”

The relationship between sexual dimorphism and androgen response element proliferation in primate genomes

“…In some cases, data were available from a congener or were gathered prior to a taxonomic revision, splitting a formerly single species into multiple species. For Nomascus leucogenys , testis dimensions were taken (Hill and Kanagasuntheram, 1959) and used to estimate testis mass (Dahl et al., 1993). For three species ( Rhinopithecus roxellana , Cercopithecus mona , and Piliocolobus tephrosceles ), no comparable data were available for testis weight.…”

“…Furthermore, duplications of cis-regulatory elements may be important in creating new opportunities for receptor regulation (Jimenez-Delgado et al, 2009) or in altering the transcriptomic response in an additive fashion (Geserick et al, 2005, Cleutjens et al, 1996, Wilson et al, 2016. There is evidence that sex-biased gene expression is altered as physically proximal cis-regulatory elements are gained or lost (Matthews et al, 2021). In the human genome, there is a correlation between the number of estrogen-mediated genes and the number of putative estrogen binding sites across chromosomes (Anderson and Jones, 2019).…”