Long Noncoding RNA–MicroRNA Pathway Controlling Nuclear Factor IA, A Novel Atherosclerosis Modifier Gene
In the current issue, Hu et al1 provide evidence for the transcription factor nuclear factor 1A (NFIA) as a novel atherosclerosis candidate gene, which is regulated through expression of the long noncoding RNA (lncRNA) RP5-833A20.1, embedded in intron 2 of NFIA, and microRNA (miRNA) hsa-miR-382-5p. In addition to providing compelling evidence for a role of NFIA in atherogenesis, the article is important for strengthening the concept of lncRNAs as master-regulators of this frequent disease.
See accompanying article on page 87
lncRNAs are a class of RNAs gaining increasing attention in biomedical research. They lack an open reading frame and are distinguished from other classes of noncoding RNAs, such as miRNAs, by an arbitrary size limit of >200 bp.2 Until now, >10 000 mammalian lncRNAs have been discovered across the genome and it has only been recognized recently that lncRNAs constitute an important layer of transcriptional regulation.3
In a hypothesis-free approach, the authors used expression arrays in a THP-1 foam cell model and identified several differentially expressed mRNAs and lncRNAs. They subsequently focused on NFIA, which overlapped with the genomic position of a yet uncharacterized lncRNA, designated RP5-833A20.1 (synonym NFIA-A1), on chromosome 1p31. THP-1–derived foam cells showed increased RP5-833A20.1 and decreased NFIA expression (Figure). Because lncRNAs have been shown to regulate target genes by interacting with miRNAs,4 the authors followed this hypothesis by searching for miRNAs targeting NFIA. They identified hsa-miR-382-5p as potential effector miRNA, which was induced by RP5-833A20.1 lncRNA and led to reduced NFIA mRNA and protein expression. The identified RP5-833A20.1/hsa-miR-382-5p/NFIA pathway was elegantly confirmed in a series of in vitro experiments using overexpression and knockdown technologies. In addition, Hu et al provide evidence for a feedback regulation of NFIA through induction of RP5-833A20.1 expression. On the functional level, the authors showed that the pathway affected central mechanisms of atherogenesis such as cholesterol homeostasis and inflammation in vitro.
The potential in vivo role of NFIA in atherosclerosis was further elucidated by lentiviral overexpression in apolipoprotein E–deficient (Apoe−/−) mice. NFIA overexpression resulted in increased retrograde cholesterol transport and excretion as well as improved plasma lipid profiles (increased high-density lipoprotein cholesterol, decreased low-density lipoprotein cholesterol, and very-low-density lipoprotein cholesterol). Furthermore, the circulating levels of proinflammatory cytokines, such as tumor necrosis factor-α and interleukin-6, were significantly reduced in NFIA-overexpressing Apoe−/− mice. Importantly, the authors noted a significant reduction of atherosclerotic plaque size in NFIA-treated Apoe−/−animals. A potentially confounding aspect of the interpretation of these data is the serum high-density lipoprotein cholesterol concentrations that are much higher than commonly reported in B6.Apoe−/− mice.5
In humans, the chromosome 1p31 gene locus containing NFIA has not been identified in genome-wide association studies of atherosclerosis to date.6 This might be because of a rather subtle effect of NFIA on atherosclerosis in humans or because the locus does not contain allelic variation, leading to functionally relevant differences of human NFIA expression. However, it should be noted that NFIA has indeed been identified as a disease locus in several genome-wide association studies for other phenotypes (genome-wide association studies catalogue at http://www.genome.gov/gwastudies/, accessed August 27, 2014) such as electrocardiographic conduction measures, thyroid hormone levels, plasma fatty acid concentrations, celiac disease, and bipolar disorder. The latter might be related to the more severe phenotype seen in patients with NFIA deletions and in knockout mice, both showing malformations of the central nervous system.7,8 It therefore seems that NFIA has a large number of functions, which is plausible for its role as a transcription factor.
In addition to providing evidence for a role of NFIA as a novel atherosclerosis modifier gene, the article by Hu et al is important for supporting the concept of lncRNAs as master-regulators of this disease. Until now, the probably best characterized lncRNA in atherogenesis is the antisense noncoding RNA at the INK4 locus (ANRIL), located at the most robust yet identified genetic locus of atherosclerosis on chromosome 9p21. Whereas ANRIL affected atherosclerosis susceptibility through epigenetic transregulation of target genes,9 Hu et al proposed induction of hsa-miR-382-5p expression by RP5-833A20.1 as effector mechanism of this lncRNA. To date, known mechanisms of lncRNA–miRNA interaction involve lncRNAs as miRNA sponges/decoys, lncRNA–miRNA competition for target mRNAs, and lncRNAs generating miRNAs as part of their sequence.4 To our knowledge, lncRNA-mediated induction of miRNAs, which are encoded by a different genomic location (hsa-miR-382-5p on chromosome 14q32 versus RP5-833A20.1 on chromosome 1p31) and which do not share sequence homology, has not been described, yet. Unfortunately, the precise molecular mechanism of hsa-miR-382-5p induction on RP5-833A20.1 lncRNA overexpression remains obscure and future work is warranted to unravel the molecular mechanisms of the appealing hypothesis of miRNA transinduction through lncRNAs.
Taken together, Hu et al identified NFIA as a novel target in atherosclerosis. Moreover, they provide further evidence for a role of lncRNAs in atherogenesis and propose a novel mechanism for lncRNA-mediated regulation of miRNA expression, providing an exciting topic for further research.
Sources of Funding
Dr L.M. Holdt and Dr Teupser are supported by the Collaborative Research Centre SFB 1123, project B1 funded by the German Research Foundation (DFG).
- © 2014 American Heart Association, Inc.
- Hu YW,
- Zhao J-Y,
- Li S-F,
- et al
- Lu W,
- Quintero-Rivera F,
- Fan Y,
- et al