Furthermore, reduced DNA binding of FXR/RXRα caused by FXR hyperacetylation may contribute to the decreased FXR-binding sites observed in obesity (5,272, compared to 15,263 sites in healthy mice). MK-1775 An important, unexpected finding in these studies is that binding of agonist-activated FXR was often associated with repression of gene expression. In a large fraction (8 of 16) of genes

examined, binding of ligand-activated FXR was associated with decreased mRNA levels, which was confirmed by decreased RNAPII occupancy and reduced acetylated histone H3K9/K14 levels. More important, levels of known histone gene-repression marks as well as H3K9 and H3K27 methylation, were markedly increased at those genes that were repressed in healthy mice after exposure to the FXR agonist, GW4064, for a short 1- or 3-hour treatment. Because mRNA levels were measured after 1-hour treatment, in addition to overnight treatment with GW4064, direct effects of FXR on gene transcription were likely detected. Although our follow-up epigenetic and gene-expression studies have suggested that gene repression by FXR is common, direct comparison of FXR binding with a comprehensive global transcriptome analysis using RNA sequencing or microarray will be necessary to definitively Ridaforolimus determine the extent of gene repression relative to gene activation by agonist-activated FXR. FXR is well known to repress its target genes

indirectly through the induction of SHP.11-14 These present studies suggest that FXR may also directly repress its target genes by unknown mechanisms. FXR could directly repress by binding to the

DNA as a FXR/RXRα heterodimer or as a monomer or homodimer, as previously shown in the regulation of apolipoprotein A1,29 which results in the inhibition of DNA binding of key transcription factors. In addition, FXR could directly inhibit genes by tethering to DNA-binding transcription factors and masking their interaction with coactivators and/or facilitating the interaction with corepressors. Sumoylation of peroxisome proliferator-activated receptor gamma and liver X receptor has been shown to be directly involved in the repression Tobramycin of inflammatory genes by the tethering of these nuclear receptors to DNA-binding activators, such as, nuclear factor kappa light-chain enhancer of activated B cells or activator protein 1.30 We have evidence that FXR is sumoylated in mouse liver extracts (D.H.K. and J.K.K., unpublished data), and FXR was shown to inhibit inflammatory responses,9, 10 so that this is a possible mechanism for FXR gene repression. Whether FXR directly suppresses its target genes by binding to DNA or tethering to other transcription factors is an important area of future investigation. In conclusion, these studies analyze, for the first time, a genome-wide comparison of FXR-binding sites in the livers of healthy and dietary obese mice.