A novel long noncoding RNA regulating cholesterol and bile acid homeostasis: A new kid on the block and a potential therapeutic target?

Potential conflict of interest: Nothing to report. See Article on Page 58 “There were long stretches of DNA in between genes that didn't seem to be doing very much; some even referred to these as ‘junk DNA,’ though a certain amount of hubris was required for anyone to call any part of the genome ‘junk,’ given our level of ignorance.” Collins1 The availability of complete genome sequences of humans, rodents, and multiple other organisms (including data from epigenome analysis and whole‐transcriptome RNA sequencing) have revealed that the coding sequences account for only 2% of the genome. With this fact came the realization that the so‐called junk DNA is not “junk” but consisted of microRNAs, long noncoding RNAs (lncRNAs), small interfering RNAs, and piwi‐interacting RNAs. The role of microRNAs in a variety of cellular pathways continues to be elucidated. lncRNAs are RNAs longer than 200 nucleotides lacking identifiable open reading frames (with rare exceptions). lncRNAs can be classified into sense, antisense, intronic, divergent, and intergenic depending on their location.2 Pervasive transcription from intergenic regions generates lncRNAs with a conservative annotated estimate of 9277 genes producing 14,880 transcripts (Gencode, version 7).3 A major difference between lncRNAs and protein‐coding genes is their poor conservation between humans and rodents, complicating interpretation of animal studies to human biology. However, lncRNAs are expressed in a tissue‐specific and developmentally regulated manner, transcribed by RNA polymerase II and localized to nucleus, cytosol, or both. Among the best‐studied and most understood lncRNAs are ribosomal RNAs, X‐inactive specific transcript, and H19 (serving as a sponge for microRNAs). In the context of liver carcinoma, lncRNA highly up‐regulated in hepatocellular carcinoma4 was shown to be elevated in hepatocellular carcinoma. lncRNA highly up‐regulated in hepatocellular carcinoma promoted tumor growth by up‐regulation of acyl coenzyme A‐synthetase long chain family member 1 and disruption of the circadian clock. Despite these advances, our current understanding of the biology and disease relevance of lncRNAs is still in its infancy. For an in‐depth understanding of the current status of lncRNAs in liver disease, the reader is referred to a recent review in hepatology.5 This issue of hepatology contains a report by Lan et al.6 on the identification of a novel lncRNA, named lnc‐HC (long noncoding RNA up‐regulated by high cholesterol), that is increased in response to high‐fat diet (HFD) feeding in rats. Prior to the publication of this report, there have been few studies on the participation of lncRNAs as regulators of metabolism and lipid homeostasis. Li et al.7 recently identified by microarray analysis an lncRNA which they named liver‐specific triglyceride (TG) regulator (lncLSTR), specifically expressed in liver. They showed that in vivo knockdown of lncLSTR lowered plasma TGs by inducing apoC2 (by farnesoid X receptor [FXR]), which promotes lipoprotein lipase‐mediated TG degradation. lncLSTR knockdown led to increased muricholic acid (MCA), which in turn led to an increase in the MCA/cholic acid ratio. This change in ratio was due to the absence of repression by lncLSTR bound to transactive response DNA‐binding protein 43, a transcriptional repressor. Transactive response DNA‐binding protein 43 (bound to lncLSTR) repressed Cyp8b1 transcription (by down‐regulating hepatocyte nuclear factor 4‐alpha activation of Cyp8b1), leading to reduced MCA synthesis. The potency of various bile acids to activate FXR indicated that MCA was more potent than cholic acid. Thus, a weaker FXR activation (low MCA levels) led to decreased ApoC2 expression, a target gene for FXR, explaining the raised serum TG levels. In a second study8 an antisense transcript of apolipoprotein A1 (ApoA1), another lncRNA that down‐regulated ApoA1 expression, was identified. Further analysis showed that ApoA1 antisense acted by recruiting histone‐modifying enzymes lysine‐specific demethylase1 and Suz12 (both epigenetic repressors) to the ApoA1 promoter, resulting in reduced ApoA1 transcription (Fig. 1).Figure 1: Mechanisms of action of lncRNAs regulating lipid, cholesterol, and bile acid homeostasis. (A) The pathways involved in regulation by lncLSTR and ApoA1 antisense7 as described. (B) The findings described by Lan et al.,6 in which lnc‐HC down‐regulates cyp7a1 and abca1 mRNA levels through association with hnRNPA2B1. Abbreviations: AS; antisense transcript; BA, bile acid; Chol, cholesterol; CEBPß, CCAAT/enhancer binding protein ß; LPL, lipoprotein lipase; LSD1, lysine‐specific demethylase 1; LXR, liver X receptor; TDP‐43, transactive response DNA‐binding protein 43.Although the above two studies did report effects of lncRNAs on lipid metabolism including demonstrating effects on a bile acid biosynthetic enzyme (Cyp8b1), no clinical/translational relevance to obesity and/or metabolic syndrome in an animal model was examined. This is exactly the area that is addressed by the Lan et al. study, thus highlighting its significance to liver disease/metabolic syndrome. In an earlier study,9 they identified a 1410‐nucleotide‐long EST(EH05) by subtraction library employing HFD‐fed mice. To verify that this expressed sequence tag is indeed lncRNA not coding for a protein, the authors carried out open reading frame prediction, conservation between species (only 31.3%, rodents to humans), location on orthologous chromosomes in rodents and humans, absence of a Kozak sequence, and in vitro translation with a green fluorescent protein tag with positive controls. Together these studies suggested that this DNA is indeed lncRNA. Precise chromosomal localization studies pointed to lnc‐HC at the 3′ end of the Slc25a15 open reading frame (mitochondrial carnitine transporter), indicating it as an overlapping transcript. lnc‐HC was expressed highly in the hepatocyte nucleus (up‐regulated by HFD with no increase in Slc25a15 messenger RNA [mRNA]), with weaker levels in muscle and pancreas. Using CBRH‐7919 cells, a rat hepatoma cell line, knockdown and overexpression of lnc‐HC led to an increase and a decrease of cyp7a1 and abca1, respectively, at mRNA and protein levels. These data gave the first clue that lnc‐HC acts by down‐regulating bile acid biosynthesis and cholesterol export. Interestingly, lnc‐HC silencing of cholesterol‐fed cells led to further elevation in the target gene expression. In vivo silencing of lnc‐HC in HFD‐fed rats led to amelioration of insulin resistance and improvement of serum TG, cholesterol, and high‐density lipoprotein cholesterol, implicating a potential translational applicability of anti‐lnc‐HC. If these data can be replicated in other models including human cells, using anti‐lnc‐HC for lowering plasma cholesterol as a therapeutic modality is feasible, assuming safety and efficacy of such therapies. In an effort to understand further how lnc‐HC mediated its effect, RNA pulldown followed by mass spectrometry revealed its interaction with hnRNAPA2B1. Association between hnRNAPs and lncRNAs has been shown,10 in adipose tissue, between lnc‐BATE1 and hnRNPU. lnc‐HC bound to hnRNAPA2B1 was capable of binding to cyp7a1 and abca1. Consistent with this association, knockdown of hnRNAPA2B1 up‐regulated while overexpression moderately decreased cyp7a1 and abca1 expression. mRNA turnover studies revealed that hnRNAP2AB1 decreased the half‐lives of cyp7a1 and abca1, suggesting this as the mechanism by which lnc‐HC exerted its effect. Despite these findings, it remained uncertain how lnc‐HC is regulated. Given the fact that lncRNAs are transcribed similar to protein‐coding genes and regulated by upstream promoters, the authors cloned the proximal and distal (Slc25a15) promoters and analyzed their regulation. The results revealed that basal promoter activity was controlled by the distal promoter. However, the proximal promoter contained two CCAAT/enhancer binding protein ß‐binding sites, of which the first (close to the transcription start site) was functional. Upstream of CCAAT/enhancer binding protein ß was liver X receptor, which activated CCAAT/enhancer binding protein ß transcription through cholesterol metabolites. Taken together with the two earlier studies indicated above, homeostatic mechanisms for regulating cholesterol and bile acid levels are mediated by multiple lncRNAs. Future studies using lnc‐HC knockout mice are necessary to confirm and extend these data. Additional lncRNAs might be uncovered that act on the same or other targets in the cholesterol and bile acid biosynthetic pathways. It is not inconceivable that in the not too distant future therapies using lncRNAs or their antisense nucleotides will be added to the currently existing armamentarium of cholesterol‐lowering drugs. Author names in bold denote shared co‐first authorship.