A Novel lnc-RNA, Named lnc-ORA, Is Identified by RNA-Seq Analysis, and Its Knockdown Inhibits Adipogenesis by Regulating the PI3K/AKT/mTOR Signaling Pathway

Obesity is closely associated with numerous adipogenic regulatory factors, including coding and non-coding genes. Long noncoding RNAs (lncRNAs) play a major role in adipogenesis. However, differential expression profiles of lncRNAs in inguinal white adipose tissue (iWAT) between wild-type (WT) and ob/ob mice, as well as their roles in adipogenesis, are not well understood. Here, a total of 2809 lncRNAs were detected in the iWAT of WT and ob/ob mice by RNA-Sequencing (RNA-Seq), including 248 novel lncRNAs. Of them, 46 lncRNAs were expressed differentially in WT and ob/ob mice and were enriched in adipogenesis signaling pathways as determined by KEGG enrichment analysis, including the PI3K/AKT/mTOR and cytokine–cytokine receptor interaction signaling pathways. Furthermore, we focused on one novel lncRNA, which we named lnc-ORA (obesity-related lncRNA), which had a seven-fold higher expression in ob/ob mice than in WT mice. Knockdown of lnc-ORA inhibited preadipocyte proliferation by decreasing the mRNA and protein expression levels of cell cycle markers. Interestingly, lnc-ORA knockdown inhibited adipocyte differentiation by regulating the PI3K/AKT/mTOR signaling pathway. In summary, these findings contribute to a better understanding of adipogenesis in relation to lncRNAs and provide novel potential therapeutic targets for obesity-related metabolic diseases.


Introduction
Obesity has become a public health hazard worldwide and is the main cause of cardiovascular diseases, type 2 diabetes, and obesity-associated metabolic syndrome [1]. Adipogenesis is mediated by a series of complex processes, including commitment of mesenchymal stem cells into preadipocytes and the induction of preadipocytes to mature adipocytes [2,3]. The exploration of this process and its regulation mechanism is of great significance for the prevention and therapy of obesity-related diseases. The ob/ob mouse is a genetic obesity mouse with a deficiency of the leptin gene that constitutively develops obesity [4]. Therefore, it is a good model for investigating the gene regulatory network of obesity.
Leptin is a hormone that is primarily made and secreted by mature adipocytes and binds to its receptor in the hypothalamus, with positive effects on energy homeostasis and weight loss [5].

Quality Control
Raw reads of fastq format were firstly processed through in-house perl scripts. In this step, clean reads were obtained by removing reads containing adapter, reads containing ploy-N, and low-quality reads from raw data. At the same time, Q20, Q30, and GC content of the clean data were calculated. All downstream analyses were based on clean, high-quality data.

Mapping to the Reference Genome and Transcriptome Assembly
We used Bowtie v2.0.6 to build an index of the reference genome and TopHat v2.0.9 to align paired-end clean reads to the reference genome (Mus musculus.GRCm38). The mapped reads of each sample were assembled by both Scripture (beta2) and Cufflinks (v2.1.1) in a reference-based approach [26,27]. We then adopted five steps to identify novel lncRNAs from the assembled transcripts: (1) transcripts with a length of <200 bp were removed; (2) transcripts with exon number <2 were removed; (3) transcripts were compared with annotated transcripts using Cuffcompare v2.1.1; (4) transcripts with Fragments Per Kilobase of transcript per Million fragments mapped (FPKM) <0.5 were removed; (5) transcripts that did not pass the protein-coding score test were removed. We used CNCI (coding-non-coding index) (v2) profiles adjoining nucleotide triplets to effectively distinguish protein-coding and non-coding sequences independent of known annotations and the CPC (coding potential calculator) (0.9-r2), which mainly assesses the extent and quality of open reading frames (ORFs) in a transcript and searches the sequences in a known protein sequence database to clarify the coding and non-coding transcripts. We translated each transcript in all three possible frames and used Pfam Scan (v1.3) to identify the occurrence of any of the known protein family domains documented in the Pfam database (using both Pfam A and Pfam B). Any transcript with a Pfam hit was excluded in the following steps. PhyloCSF (phylogenetic codon substitution frequency) (v20121028) was used to examine the evolutionary signatures characteristic of alignments of conserved coding regions, such as the high frequencies of synonymous codon substitutions and of conservative amino acid substitutions as well as the low frequencies of other missense and non-sense substitutions to distinguish protein-coding and non-coding transcripts.

Quantification of Gene Expression Level
FPKM reads of both lncRNAs and coding genes were calculated by Cuffdiff (v2.1.1) in each sample. Gene FPKMs were counted by summing the FPKMs of transcripts in each gene group.

Differential Expression Analysis and KEGG Enrichment Analysis
Cuffdiff was used to determine differential expression in digital transcript or gene expression data by a model based on the negative binomial distribution. The p-values were adjusted using the Benjamini and Hochberg method. A corrected p-value of 0.05 was set as the threshold for significantly differential expression. We used the KOBAS software to detect the statistical enrichment of differentially expressed lncRNAs and differentially expressed genes (DEGs) in KEGG pathways.

Cell Culture and Transfection
The 3T3-L1 cell line, obtained from American Type Culture Collection (ATCC, Manassas, VA, USA), was cultured in high-glucose Dulbecco's Modified Eagle Medium (DMEM) with 10% fetal bovine serum (FBS), 100 µg/mL streptomycin, and 100 U/mL penicillin under an atmosphere of 5% CO2 at 37 • C. After two days of culture, when the cells reached confluence, the cells were induced by substituting growth medium (GM) with the differentiation medium (DM), composed of DMEM, 1 µM dexamethasone (Sigma, St. Louis, MO, USA), 10 µg/mL insulin (Sigma), and 0.5 mM 3-isobuty-1-methxlzanthine (IBMX). After two days, the medium was changed with the induction medium containing 10% FBS and 10 µg/mL insulin. For proliferating preadipocytes transfection, preadipocytes were seeded in 6-well or 12-well plates, and lncRNA siRNA or the negative control (NC) (Ribobio, Guangzhou, China) at a concentration of 50 nM was transfected into cells at 40% density, using X-tremeGENE siRNA Transfection Reagent (Roche, Basel, Switzerland) and Opti-MEM (Gibco, Grand Island, NY, USA) culture medium, according to the manufacturer's protocol. The sequence of NC was 5 -UUCUCCGAACGUGUCACGUTT-3 , 5 -ACGUGACACGUUCGGAGAATT-3 ; the sequence used for siRNA was 5 -GGAGAUCUAACUAGUUACATT-3 , 5 -UGUAACUAGUUAGAUCUCCTT-3 . The cells were harvested 24 h or 48 h after transfection. For adipocytes differentiation studies, the cells were transfected when their density reached 70-80%; the transfection protocol was the same as for proliferating preadipocytes transfection.

RNA-Fluorescent In Situ Hybridization
The subcellular localization of lnc-ORA was detected by a fluorescent in situ hybridization (FISH) kit (Ribobio, China) according to the manufacturer's instruction. Briefly, the preadipocytes were washed with PBS and fixed in 4% paraformaldehyde for 10 min. Then, the cells were permeabilized with a 0.5% Triton X-100 solution at 4 • C for 5 min, washed with PBS three times for 5 min, and pre-hybridized at 37 • C for 30 min. Then, an oligodeoxynucleotide probe of anti-lnc-ORA, anti-18s, or anti-U6 was used in the hybridization solution at 37 • C overnight in the dark. The next day, the cells were stained with DAPI and imaged using a confocal laser-scanning microscope.

Cytoplasmic and Nuclear RNA Extraction
For the detection of the lnc-ORA location in cytoplasmic and nuclear RNA, cultured 3T3-L1 preadipocytes were washed with PBS, suspended in lysis buffer, and then incubated on ice for 5 min. After centrifuging at 8000 rpm for 5 min, the supernatant was transferred to a new microcentrifuge tube and subjected to cytoplasmic RNA extraction, while the pellet was resuspended in lysis buffer and subjected to nuclear RNA extraction. For RNA extraction, the fractions were first incubated with Proteinase K (10 mg/mL) at 37 • C for 20 min and then mixed with Trizol (Carsbad, CA, USA). RNA was separated by chloroform and precipitated by ethanol with 3 M sodium acetate [28].

Rapid Amplification of 5 and 3 cDNA Ends (RACE)
RACE was performed using the SMARTer RACE Kit with SMARTer II A Oligonucleotide and SMARTScribe™ Reverse Transcriptase (Takara) according to the instruction of the manufacturer. Ex Taq and LA Taq with GC Buffer (Takara) were used for PCR amplification following the manufacturer's protocol. PCR products were gel-purified with QIAquick (QIAGEN, Duesseldorf, Germany), cloned into the pGEM-T easy vector, and sequenced. The primer used for 5 RACE was 5 -GATTACGCCAAGCTTCCTTCAGCCCATCTTGTGTCCCGT-3 . The primer for 3 RACE was 5 -GATTACGCCAAGCTTGGGACACAAGATGGGCTGAAGGGG-3 .

Flow Cytometry, Edu, and CCK-8 Assays
Flow cytometry, Edu staining, and cell count kit 8 (CCK-8) assays were performed according to our previously described method [29]. Preadipocytes were transfected with NC and siRNA when the cell densities were 40%. Then, 24 h or 48 h after transfection, the preadipocyte proliferation was assessed.

Statistical Analysis
Statistical analysis was performed using Graph Pad Prism 6 (Graph pad software, La Jolla, CA, USA). Comparisons among the individual means were made by a Student's t test. Data are presented as the mean ± SEM. A p value <0.05 was considered significant.

Identification and Characterization of lncRNAs in WT and ob/ob Mice
To illustrate the differential lncRNA profiles in iWAT between WT and ob/ob mice and to identify obesity-related lncRNAs, we performed RNA-seq of iWAT samples, which were isolated from eight-week-old WT and ob/ob mice kept with identical feeding conditions. The identification flow chart of novel lncRNAs is presented in Figure 1A. In total, 248 novel lncRNAs were identified in WT and ob/ob mice after transcript-coding potential analysis of the four prediction methods: the coding potential calculator (CPC), Pfam Scan database, phylogenetic codon substitution frequency (phyloCSF), and the coding-non-coding index (CNCI) ( Figure 1B). The lncRNA classification showed that 89.5% of novel lncRNAs transcripts were long intergenic noncoding RNAs (lincRNA), and 10.5% of novel lncRNAs transcripts were antisense lncRNA (AS lncRNAs); there were no intronic lncRNAs ( Figure 1C). Moreover, the number of exons of annotated and novel lncRNAs was less than that of mRNAs ( Figure 1D). The expression levels of lncRNAs were slightly lower than those of mRNAs ( Figure 1E). In addition, the ORF lengths of lncRNAs were shorter than those of mRNAs ( Figure 1F).

Differentially Expressed lncRNAs and Genes in WT and ob/ob Mice
To screen the DEGs and differentially expressed lncRNAs in WT and ob/ob mice, Cuffdiff v2.1.1 was used to analyze differential expression levels of lncRNAs and mRNAs. The results showed that there were 46 differentially expressed lncRNAs and 971 DEGs in WT and ob/ob mice, respectively (Figure 2A

Differentially Expressed lncRNAs and Genes in WT and ob/ob Mice
To screen the DEGs and differentially expressed lncRNAs in WT and ob/ob mice, Cuffdiff v2.1.1 was used to analyze differential expression levels of lncRNAs and mRNAs. The results showed that there were 46 differentially expressed lncRNAs and 971 DEGs in WT and ob/ob mice, respectively (Figure 2A

Pathway Analysis of DEGs
In order to further confirm the potential function of obesity-related lncRNAs in adipogenesis, KEGG analyses of the DEGs were performed. The results showed that the DEGs were enriched in 122 signaling pathways. The top 20 upregulated and top 20 downregulated signaling pathways are shown, respectively, in Figure 3A,B. The results showed that many pathways related to fat metabolism and adipogenesis were significantly enriched. Of them, the PI3K/AKT signaling pathway had the highest enrichment, and the cytokine-cytokine receptor interaction showed the greatest downregulation, suggesting that the two signaling pathways may be associated with obesity-induced lncRNAs in adipogenesis.

Pathway Analysis of DEGs
In order to further confirm the potential function of obesity-related lncRNAs in adipogenesis, KEGG analyses of the DEGs were performed. The results showed that the DEGs were enriched in 122 signaling pathways. The top 20 upregulated and top 20 downregulated signaling pathways are shown, respectively, in Figure 3A,B. The results showed that many pathways related to fat metabolism and adipogenesis were significantly enriched. Of them, the PI3K/AKT signaling pathway had the highest enrichment, and the cytokine-cytokine receptor interaction showed the greatest downregulation, suggesting that the two signaling pathways may be associated with obesity-induced lncRNAs in adipogenesis.

Validation of the Differentially Expressed lncRNAs by qPCR
The adipogenic ability of obese mice was obviously higher than that of WT mice. First, we detected the expression levels of several adipogenic marker genes in the iWAT of WT and ob/ob mice, such as PPARγ, fatty acid binding protein 4 (FABP4), fatty acid synthase (FASN), leptin, and peroxisome proliferator-activated receptor gamma coactivator 1-α (PGC-1α) ( Figure 4A). The results showed that the expression levels of PPARγ, FABP4, and FASN in ob/ob mice were higher than in WT mice. However, the expression levels of the thermogenesis gene PGC1-α was lower than in WT mice ( Figure 4B), indicating that our sequencing samples were suitable for an obesity model. To confirm the accuracy of the sequencing data, nine randomly selected lncRNAs were validated by qPCR. The results showed that the expression patterns of lncRNAs were similar to those of the sequencing data ( Figure 4C,D). Specifically, we found that LNC_0000092, the newly named lnc-ORA (obesity-related lncRNA), was abundant in the iWAT of ob/ob mice, with a seven-fold expression change compared to its expression in WT mice, suggesting it may be an important lncRNA during adipogenesis.

Validation of the Differentially Expressed lncRNAs by qPCR
The adipogenic ability of obese mice was obviously higher than that of WT mice. First, we detected the expression levels of several adipogenic marker genes in the iWAT of WT and ob/ob mice, such as PPARγ, fatty acid binding protein 4 (FABP4), fatty acid synthase (FASN), leptin, and peroxisome proliferator-activated receptor gamma coactivator 1-α (PGC-1α) ( Figure 4A). The results showed that the expression levels of PPARγ, FABP4, and FASN in ob/ob mice were higher than in WT mice. However, the expression levels of the thermogenesis gene PGC1-α was lower than in WT mice ( Figure 4B), indicating that our sequencing samples were suitable for an obesity model. To confirm the accuracy of the sequencing data, nine randomly selected lncRNAs were validated by qPCR. The results showed that the expression patterns of lncRNAs were similar to those of the sequencing data ( Figure 4C,D). Specifically, we found that LNC_0000092, the newly named lnc-ORA (obesity-related lncRNA), was abundant in the iWAT of ob/ob mice, with a seven-fold expression change compared to its expression in WT mice, suggesting it may be an important lncRNA during adipogenesis.

Identification, Subcellular Locations, and Expression Pattern Analysis of lnc-ORA
lnc-ORA was transcribed from locus 89,767,653 to locus 89,768,886 on chromosome 15 and consisted of two exons and one intron. The alignment track showed that lnc-ORA is an unannotated lncRNA in the mouse genome ( Figure S1A). The rapid amplification of cDNA ends (RACE) and PCR experiments indicated that the full length of lnc-ORA was 836 bp (Figure S1B-D). To confirm the subcellular locations of lnc-ORA in preadipocyte, FISH, and cytoplasmic and nuclear RNA separated extraction experiments were performed. The results showed that lnc-ORA was expressed in both cytoplasmic and nuclear locations (Figures 5A and S2B). Then, we used the Coding Potential Assessment Tool (CPAT) to evaluate the protein-coding potential of lnc-ORA. The results showed that lnc-ORA had no protein-coding potential ( Figure S2A). The cell-specific localization of lnc-ORA in adipose tissue (AT) was determined by the isolation of preadipocytes, macrophages, and adipocytes from the AT of obese mice. The results showed that lnc-ORA was mainly expressed in adipocytes and preadipocytes and almost not expressed in macrophages ( Figure S2C). lnc-ORA was expressed in various tissues, including heart, liver, spleen, kidney, BAT, and muscle and was expressed at the highest levels in WAT compared to other tissues ( Figure S2D). 3T3-L1 cells were differentiated and generated lipid droplets after 8 d induction (Figure S3A), which is consistent with the expression of adipogenic marker genes ( Figure S3B-D). Moreover, the expression levels of lnc-ORA gradually increased during proliferation and differentiation of 3T3-L1 cells ( Figure 5B,C),  Figure S1A). The rapid amplification of cDNA ends (RACE) and PCR experiments indicated that the full length of lnc-ORA was 836 bp (Figure S1B-D). To confirm the subcellular locations of lnc-ORA in preadipocyte, FISH, and cytoplasmic and nuclear RNA separated extraction experiments were performed. The results showed that lnc-ORA was expressed in both cytoplasmic and nuclear locations ( Figure 5A and Figure S2B). Then, we used the Coding Potential Assessment Tool (CPAT) to evaluate the protein-coding potential of lnc-ORA. The results showed that lnc-ORA had no protein-coding potential ( Figure S2A). The cell-specific localization of lnc-ORA in adipose tissue (AT) was determined by the isolation of preadipocytes, macrophages, and adipocytes from the AT of obese mice. The results showed that lnc-ORA was mainly expressed in adipocytes and preadipocytes and almost not expressed in macrophages ( Figure S2C). lnc-ORA was expressed in various tissues, including heart, liver, spleen, kidney, BAT, and muscle and was expressed at the highest levels in WAT compared to other tissues ( Figure S2D). 3T3-L1 cells were differentiated and generated lipid droplets after 8 d induction (Figure S3A), which is consistent with the expression of adipogenic marker genes ( Figure S3B-D). Moreover, the expression levels of lnc-ORA gradually increased during proliferation and differentiation of 3T3-L1 cells ( Figure 5B,C), indicating that this lncRNA could regulate preadipocytes' proliferation and differentiation during lipid formation.
indicating that this lncRNA could regulate preadipocytes' proliferation and differentiation during lipid formation.

Knockdown of lnc-ORA Inhibits the Proliferation of Preadipocytes
To confirm the role of lnc-ORA in preadipocyte proliferation, 3T3-L1 cells were transfected with siRNA or an NC. The results showed that the expression levels of lnc-ORA obviously decreased after transfection for 24 h and 48 h ( Figure 6A). Cell viability was not significantly affected by siRNA transfection ( Figure 6I). The mRNA levels of cell cycle marker genes, including proliferating cell nuclear antigen (PCNA) (Figure 6B), cyclin-dependent kinases B (cyclin B) ( Figure 6C), cyclin D1 ( Figure 6D), and cyclin E ( Figure 6E), were significantly decreased after transfection for 24 h but significantly increased the mRNA levels of cyclin-dependent kinase inhibitor 1B (p27) ( Figure 6F) and were consistent with the protein levels ( Figure 6G,H). Furthermore, the Edu staining assay indicated that knockdown of lnc-ORA decreased the number of Edu-labeled cells ( Figure 7A,B). CCK-8 showed that the total cell number after lnc-ORA knockdown was less than the number of NC-treated cells ( Figure 7C). In addition, the knockdown of lnc-ORA decreased the number of cells in S-phase and increased the number of cells in G1-phase, as determined by flow cytometer analysis ( Figure 7D,E),

Knockdown of lnc-ORA Inhibits the Proliferation of Preadipocytes
To confirm the role of lnc-ORA in preadipocyte proliferation, 3T3-L1 cells were transfected with siRNA or an NC. The results showed that the expression levels of lnc-ORA obviously decreased after transfection for 24 h and 48 h ( Figure 6A). Cell viability was not significantly affected by siRNA transfection ( Figure 6I). The mRNA levels of cell cycle marker genes, including proliferating cell nuclear antigen (PCNA) (Figure 6B), cyclin-dependent kinases B (cyclin B) ( Figure 6C), cyclin D1 ( Figure 6D), and cyclin E ( Figure 6E), were significantly decreased after transfection for 24 h but significantly increased the mRNA levels of cyclin-dependent kinase inhibitor 1B (p27) ( Figure 6F) and were consistent with the protein levels ( Figure 6G,H). Furthermore, the Edu staining assay indicated that knockdown of lnc-ORA decreased the number of Edu-labeled cells ( Figure 7A,B). CCK-8 showed that the total cell number after lnc-ORA knockdown was less than the number of NC-treated cells ( Figure 7C). In addition, the knockdown of lnc-ORA decreased the number of cells in S-phase and increased the number of cells in G1-phase, as determined by flow cytometer analysis ( Figure 7D,E), indicating that the knockdown of lnc-ORA inhibited preadipocytes proliferation, possibly through the suppression of the DNA replication process.
indicating that the knockdown of lnc-ORA inhibited preadipocytes proliferation, possibly through the suppression of the DNA replication process.

Knockdown of lnc-ORA Inhibits Adipocytes Differentiation through the PI3K/AKT/mTOR Signaling Pathway
To explore the function of lnc-ORA regulation on adipogenic differentiation, 3T3-L1 cells were transfected with siRNA or NC at a confluence of 70-80%. The expression of lnc-ORA was decreased to 30-40% by siRNA-mediated lnc-ORA knockdown ( Figure 8B). BODIPY staining indicated that the knockdown of lnc-ORA inhibited lipid accumulation in 3T3-L1 cells ( Figure 8A). The knockdown of lnc-ORA significantly reduced the mRNA expression levels of the adipogenesis markers PPARγ, FASN, and FABP4 at Days 2 and 4 ( Figure 8C-E), which was consistent with the protein levels ( Figure  8F,G). Collectively, these data demonstrated that the knockdown of lnc-ORA markedly inhibited adipocytes differentiation. In addition, we further examined the PI3K/AKT/mTOR, ERK, and p38 signaling pathways during adipogenesis. The results showed that the knockdown of lnc-ORA reduced the ratios of p-PI3K/PI3K, p-AKT/AKT, and p-mTOR/mTOR during adipocyte differentiation, whereas the ERK and p38 signaling pathways were not affected ( Figures 9A,B and

Knockdown of lnc-ORA Inhibits Adipocytes Differentiation through the PI3K/AKT/mTOR Signaling Pathway
To explore the function of lnc-ORA regulation on adipogenic differentiation, 3T3-L1 cells were transfected with siRNA or NC at a confluence of 70-80%. The expression of lnc-ORA was decreased to 30-40% by siRNA-mediated lnc-ORA knockdown ( Figure 8B). BODIPY staining indicated that the knockdown of lnc-ORA inhibited lipid accumulation in 3T3-L1 cells ( Figure 8A). The knockdown of lnc-ORA significantly reduced the mRNA expression levels of the adipogenesis markers PPARγ, FASN, and FABP4 at Days 2 and 4 ( Figure 8C-E), which was consistent with the protein levels ( Figure 8F,G). Collectively, these data demonstrated that the knockdown of lnc-ORA markedly inhibited adipocytes differentiation. In addition, we further examined the PI3K/AKT/mTOR, ERK, and p38 signaling pathways during adipogenesis. The results showed that the knockdown of lnc-ORA reduced the ratios of p-PI3K/PI3K, p-AKT/AKT, and p-mTOR/mTOR during adipocyte differentiation, whereas the ERK and p38 signaling pathways were not affected ( Figure 9A,B and Figure S4), indicating that lnc-ORA regulates adipocytes differentiation through the PI3K/AKT/mTOR signaling pathway ( Figure 9C). S4), indicating that lnc-ORA regulates adipocytes differentiation through the PI3K/AKT/mTOR signaling pathway ( Figure 9C).

Discussion
In this study, a total of 2809 lncRNA transcripts were identified in the iWAT from WT and ob/ob mice, of which 248 were novel lncRNAs. Further, 46 differentially expressed lncRNAs in WT and ob/ob mice were identified. Specifically, a novel lncRNA, named lnc-ORA, enhanced preadipocytes proliferation and adipogenesis through the PI3K/AKT/mTOR signaling pathway. Overall, our findings provide direct evidence that lnc-ORA is a novel positive regulator of adipogenesis.
Many different processes contribute to the commitment of a mesenchymal stem cell to the adipocyte lineage, including the regulation of a complex network of transcription factors, cofactors, and numerous signaling pathways [30][31][32][33][34][35]. Alvarez-Dominguez et al. used RNA-seq to reconstruct de novo transcriptomes of mouse BAT, iWAT, and eWAT and identified lnc-BATE1 as a regulator of BAT development [36]. Lo et al. performed an analysis of lncRNA transcriptome in adipocytes isolated from BAT and WAT in high-fat-diet-induced obese mice. The results identified a series of obesity-related lncRNAs, among which the most crucial was lnc-leptin, transcribed from the promoter of leptin. lnc-leptin promoted adipogenesis through direct interaction between its DNA loci and leptin [8]. Our previous study investigated the difference of lncRNAs in fat-type bamei pigs and lean-type large white pigs [25]. Although many studies have investigated lncRNAs related to obesity [37][38][39], the different expression of lncRNAs in WAT between ob/ob and WT mice is unclear.
Here, we identified 248 novel lncRNAs in WT and ob/ob mice. The characterization of the novel lncRNAs was consistent with basic features of lncRNA, as determined by analyzing exon number, FPKM distribution, and ORF length. Specifically, we focused on an obesity-related lncRNA, lnc-ORA, which showed higher expression in the adipose tissue and more than seven-fold upregulation in ob/ob mice compared with WT mice. Given the higher expression levels in the adipose tissue, we used the knockdown method to eliminate its function in 3T3-L1 cells. Interestingly, the results showed that

Discussion
In this study, a total of 2809 lncRNA transcripts were identified in the iWAT from WT and ob/ob mice, of which 248 were novel lncRNAs. Further, 46 differentially expressed lncRNAs in WT and ob/ob mice were identified. Specifically, a novel lncRNA, named lnc-ORA, enhanced preadipocytes proliferation and adipogenesis through the PI3K/AKT/mTOR signaling pathway. Overall, our findings provide direct evidence that lnc-ORA is a novel positive regulator of adipogenesis.
Many different processes contribute to the commitment of a mesenchymal stem cell to the adipocyte lineage, including the regulation of a complex network of transcription factors, cofactors, and numerous signaling pathways [30][31][32][33][34][35]. Alvarez-Dominguez et al. used RNA-seq to reconstruct de novo transcriptomes of mouse BAT, iWAT, and eWAT and identified lnc-BATE1 as a regulator of BAT development [36]. Lo et al. performed an analysis of lncRNA transcriptome in adipocytes isolated from BAT and WAT in high-fat-diet-induced obese mice. The results identified a series of obesity-related lncRNAs, among which the most crucial was lnc-leptin, transcribed from the promoter of leptin. lnc-leptin promoted adipogenesis through direct interaction between its DNA loci and leptin [8]. Our previous study investigated the difference of lncRNAs in fat-type bamei pigs and lean-type large white pigs [25]. Although many studies have investigated lncRNAs related to obesity [37][38][39], the different expression of lncRNAs in WAT between ob/ob and WT mice is unclear. Here, we identified 248 novel lncRNAs in WT and ob/ob mice. The characterization of the novel lncRNAs was consistent with basic features of lncRNA, as determined by analyzing exon number, FPKM distribution, and ORF length. Specifically, we focused on an obesity-related lncRNA, lnc-ORA, which showed higher expression in the adipose tissue and more than seven-fold upregulation in ob/ob mice compared with WT mice. Given the higher expression levels in the adipose tissue, we used the knockdown method to eliminate its function in 3T3-L1 cells. Interestingly, the results showed that the knockdown of lnc-ORA inhibited both preadipocyte proliferation and processes to alleviate adipogenesis, as indicated by a decrease in the number of cells in S-phase and the expression of cell cycle marker genes PCNA, cyclin B, cyclin D1, and cyclin E, as well as by a decrease in the expression of adipogenesis markers PPARγ, FASN, and FABP4. The mitotic clonal expansion (MCE) is a prerequisite for adipocyte differentiation that occurs within 48 h of adipogenic stimulation [40]. After induction of differentiation, postconfluent and growth-arrested 3T3-L1 preadipocytes synchronously reenter the cell cycle, undergo several rounds of mitotic clonal expansion, and then express genes that produce the adipocyte phenotype [41]. Here, we found that knockdown of lnc-ORA inhibited preadipocyte proliferation by analysis of flow cytometry, Edu, CCK-8, and the levels of proliferating key genes, implying lnc-ORA may be involved in the mitotic clonal expansion phase and hence impair terminal adipocyte differentiation. These data indicate that lnc-ORA is a positive regulator of adipogenesis and provide an explanation for its higher levels in ob/ob mice than in WT mice.
lncRNAs not only regulate the transcription of target genes in the nucleus but also participate in the post-transcriptional regulation of genes in the cytoplasm, so the subcellular localization of lncRNAs is important for elucidating their function [42,43]. We found that lnc-ORA was located in both the nucleus and the cytoplasm of preadipocytes, implying that lnc-ORA could participate in target gene regulation at both the transcription and the post-transcription levels.
The leptin-deficient ob/ob mouse is a key animal model for the investigation of obesity and type 2 diabetes. Leptin receptors are members of the cytokine family of receptors. The long-form receptor has been shown to activate the STAT, MAPK, and PI3K signaling pathways [44][45][46]. In addition, it has been demonstrated that PI3K participates in obesity-related biological processes [47]. Leptin-stimulated secretion of insulin requires calcium and the activation of the PI3K/AKT signaling pathway [48][49][50]. Previous studies have indicated that leptin acts on POMC and NPY/AgRP neurons to suppress food intake and promote energy expenditure [51]. One mechanism is leptin-induced PI3K signaling regulation of BAT thermogenesis and WAT browning [52]. PI3Ks are members of a unique and conserved family of enzymes responsible for the phosphorylation of proteins. The insulin-mediated PI3K/AKT signaling pathway plays an important role within adipocytes of obese patients and leads to an excess of lipids that has to be properly stored in the fat tissue [53]. In this study, KEGG analyses of DEGs were performed. We predicted that the DEGs were enriched in 122 signaling pathways. The top 20 upregulated and downregulated KEGG pathways were involved in fat metabolism and adipogenesis processes. Among them, the PI3K/AKT signaling pathway had the highest enrichment, and the cytokine-cytokine receptor interaction showed the greatest downregulation, suggesting that the two signaling pathways may play important roles during adipogenesis. Therefore, we further examined the PI3K/AKT/mTOR, ERK, and p38 signaling pathways after knockdown of lnc-ORA in 3T3-L1 cells. We found that knockdown of lnc-ORA only decreased the ratios of p-PI3K/PI3K, p-AKT/AKT, and p-mTOR/mTOR during adipocytes differentiation, indicating that lnc-ORA is required for adipocytes differentiation through the PI3K/AKT/mTOR signaling pathway. The findings further confirmed our prediction on the basement of KEGG analysis of the DEGs. However, we here must point out that other undetected adipogenic-related signaling pathways may be involved in adipogenic regulation of lnc-ORA.

Conclusions
In conclusion, we found 46 differentially expressed lncRNAs in the iWAT of ob/ob and WT mice and identified a novel lnc-ORA that regulated adipogenesis through the PI3K/AKT/mTOR signaling pathway. These findings contribute to a better understanding of adipogenesis by lncRNAs and provide novel potential therapeutic targets for obesity-related metabolic diseases. Author Contributions: W.P. and R.C. conceived and designed the experiments; R.C. and G.T. performed the experiments; Q.Z., W.Y., W.Z., J.X., C.W., and C.H. analyzed the data; G.Y. and W.P. resources, contributed reagents, materials, analysis tools; R.C. wrote the paper.