Long Non-coding RNA ADNCR Suppresses Adipogenic Differentiation By Competing For MiR-204

Fat and fatty acid composition, whether in adipose tissue or muscle, are positively contributed to the appearance, firmness, flavor, hardness and shelf-life of meat products, and are central of the nutritional value of meat. Therefore, unraveling the regulatory mechanisms of adipogenic differentiation is important for improving beef quality. Adipogenesis is a complex and precisely orchestrated process mediated by a network of adipogenic regulatory factors. At the center of this network, the peroxisome proliferator-activated receptor γ(PPARγ) and CCAAT/enhancer-binding protein α(C/EBPα) are considered the crucial determinants of adipocyte fate. While the major molecular pathways of adipogenesis are understood, this scenario is far from complete, as new players are gradually characterized as critical regulators of adipogenesis. In fact, the recent dramatic increase in knowledge has demonstrated that long noncoding RNA(lncRNAs) are involved in adipogenic gene regulatory network, but the precise molecular mechanism has largely remained elusive. In the present study, we performed Ribo-Zero RNA-Seq to investigate both the poly(A)+ and poly(A)- lncRNAs of in vitro cultured bovine preadipocytes and differentiated adipocytes. Inaddition, we explored the regulatory mechanism of lncRNA ADNCR on adipogenic differentiation by using RACE, RNA FISH, gain- and loss-of-function, dual luciferase reporter assay and site-directed mutagenesis. Our main results are as follows: 1. ADNCR is the most downregulated lncRNA in adipocyte differentiationTo comprehensively identify lncRNAs with potential functions during adipogenesis, we performed Ribo-Zero RNA-Seq to survey the transcriptome landscape of in vitro cultured bovine preadipocytes and mature adipocytes(two biological replicates for each condition). In total, ~393 million clean reads were obtained, of which 89.7% could be aligned to the reference bovine genome(bosTau7). A large fraction(52%) of all mapped sequence reads were located in intergenic or intronic regions, indicating an impressive number of unannotated transcripts. To identify novel reliable lncRNA models, we considered only multi-exonic transcripts and filtered them through the following highly stringent criterion:(1) Size selection: Only transcripts with ≥ 200 bp were kept.(2) Reads coverage threshold: Transcripts with ≤ 3 reads were removed from our data.(3) Open reading frame(ORF) filter: Transcripts with a predicted ORF longer than 100 aa were removed.(4) Known protein domain filter: Transcripts were aligned to the Pfam and Swiss-Protein database to eliminate transcripts with significant homology to known protein domains.(5) Protein-coding-score test: Both the Coding-Non-Coding Index(CNCI) and Coding Potential Calculator(CPC) were used to evaluate the coding potential of candidate lncRNAs. Through these multilayered strict analyses, we finally identified 2,882 lncRNAs, including 1,037 novel lncRNAs. When classified based on their location with respect to the location of protein-coding genes, 55 percent of them belonged to the intergenic group, whereas only 4 percent of them were antisense lncRNAs. The length of the chromosome normally reflects lncRNAs content: more lncRNA loci were found in larger chromosomes than in shorter chromosomes. Genomic features analyses indicated that these lncRNAs are shorter in length, and have significantly lower expression levels and fewer exon number than the RefSeq protein coding transcripts. The primary sequence of lncRNAs is less conserved than mRNA, but they are obviously more conserved than random sequence, implying their essential role in organismal evolution. Compared the lncRNAs expression levels of preadipocytes with those of differentiated adipocytes revealed a total of 16 differentially expressed lncRNAs, of which, ADNCR(adipocyte differentiation-associated long noncoding RNA) the most downregulated lncRNA. 2. ADNCR inhibits adipogenesis by sponging miR-204The 5′ RACE and 3′ RACE analyses demonstrated that ADNCR is a transcript of 1,730 nucleotides and is composed of two exons. Bioinformatics analyses and in vitro translation experiments demonstrated that ADNCR is a bona fide lncRNA. The semi-quantitative PCR of nuclear and cytoplasmic fractions indicated that ADNCR is mainly localized in the cytoplasm. Overexpression of ADNCR significantly hampered adipogenesis, as indicated by a dramatic decrement in the number of mature adipocytes and in the expression of the respective adipogenic markers PPARγ, C/EBPα, and FABP4. To explore the underlying mechanism of ADNCR action and identify its targets that are important for adipogenesis, we analyzed the potential miRNAs binding sites in ADNCR using RNAhybrid and identified two putative complementary sequences for miR-204 in ADNCR at positions 205–250 and 1630–1659. Mechanistically, ADNCR inhibited adipocyte differentiation by functioning as a competing endogenous RNA(ceRNA) for miR-204, thereby augmenting the expression of the miR-204 target gene, SIRT1, which is known to inhibit adipocyte differentiation and adipogenic gene expression by docking with NCoR and SMART to repress PPARγ activity. 3. Resveratrol and nicotinamide regulate bovine adipogenesis through a SIRT1-dependent mechanism Due to the important role of SIRT1 in adipocyte differentiation and muscle development, we next analyzed the effect of resveratrol and nicotinamide on adipogenesis,which activates and inhibits SIRT1, respectively. Resveratrol-induced SIRT1 activation significantly decreased adipocyte number, and lessened expression of adipogenic gene. Inhibition of SIRT1 by nicotinamide resulted in increased number of adipocytes and adipocyte markers. By using an shRNA approach, we further demonstrated that resveratrol and nicotinamide modulate bovine adipogenesis through a SIRT1-dependent mechanism, because there were no drug-mediated fat changes in SIRT1 konckdown cells. 4. SNPs in the promoter region of SIRT1 regulate its activityBy using DNA pool sequencing, five novel SNPs in bovine SIRT1 gene was identified in five main Chinese cattle breeds.These SNPs are significantly associated with some growth traits in NY cattle.SNPs g.-382G>A and g.-274C>G are located in the core promoter region of SIRT1, and both of them can decrease the SIRT1 promoter activity. By using matinspector release professional 8.0, we predicted that SNP g.-382G>A generates a MEF2 A binding sites. To confirm this, we co-transfected pcDNA-MEF2 A with pGL3-GC or pGL3-AC into C2C12 cells and observed that pcDNA-MEF2 A significantly reduced the luciferase activity of pGL3-AC. Furthermore, qRT-PCR revealed that the expression levels of SIRT1 with AA genotype is lower than that with GG genotype.Our findings not only complement the reference genome annotation of cattle and provide a valuable genomic resource for the identification of lncRNAs with functional roles in adipocyte differentiation, but also reveal new insights into understanding the mechanisms of adipogenic differentiation.