Regulation of gene expression in mammalian cells is

Regulation of gene expression in mammalian cells is governed by the epigenetic machinery, which includes several distinctive yet entangled branches. DNA and histone modifying enzymes, non-coding regulatory RNAs, and ATPase-dependent chromatin remodeling proteins all contribute to the dynamic alteration of transcriptome in HSCs undergoing phenotypic transition [6,7]. Transcriptional regulation by differential histone modifications represents a most well-characterized example in this regard. Gene expression patterns based on both singular locus and genomewide analysis reveal that histone lysine acetylation and H3K4 methylation correlate with transcriptional activation while H3K9/H3K27 methylation predicts transcriptional repression [8]. The methylation status of any specific lysine residue is determined by the delicate balance between a methyltransferase, the “reader”, and a demethylase, the “eraser” [9]. In mammals, H3K9 trimethylation is catalyzed by KMT1A/1B and removed by KDM4, whose role in HSC activation is yet to be investigated. Here, we present evidence to show that KDM4 regulates (suppresses) HSC activation by contributing to SREBP2-mediated activation of miR-29 transcription.
Methods
Results
Discussion The epigenetic machinery plays key roles in the phenotypic modulation of hepatic stellate cells [26]. Mechanistic investigation of HSC activation by the epigenetic machinery not only boosts our understanding of the pathogenesis of liver fibrosis, but sheds lights on the development of novel therapeutic strategies to treat end-stage liver diseases. Here we delineate a novel SREBP2-KDM4 pathway that contributes to the maintenance of HSC quiescence by activating miR-29 transcription. We show here that expression levels of SREBP2 and KDM4 are down-regulated in activated HSCs compared to quiescent HSCs both in vitro and in vivo. It is unclear at this point what mechanism contributes to the down-regulation of SREBP2/KDM4 during HSC activation. It has been previously shown that the nuclear receptors retionic Mirtazapine mg X receptor (RXR) and thyroid hormone receptor (TR) activate SREBP2 transcription by forming a heterodimer on the SREBP2 promoter [27]. RXR expression is down-regulated in activated HSCs compared to quiescent HSCs [28], which may potentially result in the decrease of SREBP2 expression. Alternatively, SREBP2 transcription is repressed by the forkhead family of transcription factors (FOXO1/FOXO3a) via a conserved insulin response element (IRE) mapped to the proximal SREBP2 promoter [29]. Since FOXO1 and FOXO3a are post-translationally activated in mature HSCs [30,31], we speculate that FOXO1 and/or FOXO3a may be responsible for SREBP2 down-regulation during HSC activation. On the other hand, little is known regarding the transcription regulation of the KDM4 genes. Detailed analysis of SREBP2/KDM4 transcription during HSC activation would further our understanding of this subject matter. The methylation status of any given lysine residue is determined by the dynamic interplay between the methyltransferase and the demethylase. While we focused on KDM4 in the regulation of miR-29 trans-repression, the alternative scenario wherein the H3K9 dimethyltransferase G9a might play a role in miR-29 repression remains unexplored. Indeed, pharmaceutical inhibition of G9a in LX-2 cells attenuates TGF-β induced Acta2 expression [32]. In addition, mounting evidence demonstrates that G9a programs pro-fibrogenic response in multiple organs [[33], [34], [35]]. It is not known whether the pro-fibrogenic effect of G9a is mediated through miR-29 repression in HSCs. We propose that loss of SREBP2 and KDM4 and consequently miR-29 during HSC activation is responsible for liver fibrosis. However, SREBP2 and/or KDM4 might control the expression of target genes other than miR-29 to modulate HSC phenotype. For instance, SREBP2 can stimulate the transcription of PPARγ [36], a well-known anti-fibrogenic nuclear receptor. Coincidently, lysine methylation levels surrounding the PPARγ promoter region are augmented during HSC activation [17,37], suggesting of potential involvement of KDM4. Further, SREBP2 has been identified as a binding partner for SMAD3, the master regulator of cellular pro-fibrogenic response; the interaction of SREBP2 with SMAD3 inhibits SMAD3 activity [38]. Therefore, it is possible that decrease of SREBP2 expression liberates SMAD3 from the inhibition allowing it to promote HSC activation.