An integrated analysis of microRNAs involved in fat deposition in different pig breeds

The domestic pig, an important species in the animal production industry, is also a model system for studying fat deposition and fat-associated diseases in humans. In order to investigate the genetic relationships of the miRNAs that may regulate fat deposition, we performed a genome-wide analysis of miRNAs derived from subcutaneous adipose tissue of Laiwu and Large White pig using RNA-seq. A total of 26,486,906 reads were obtained. Further analysis indicated that 39 known miRNAs and 56 novel miRNAs had significantly differential expression between the two breeds of pigs. Gene Ontology and KEGG pathway analysis revealed that the predicted targets of these differentially expressed miRNAs were involved in several fat-associated pathways, such as the PPAR, MAPK and Wnt signaling pathways. In addition, we found that three miRNAs, ssc-miR-133a-3p, ssc-miR-486 and ssc-miR-1, had an impact on the development of porcine subcutaneous fat through the PPAR signaling pathway. This study provides clues to understand the potential mechanisms of adipogenesis and fat deposition between two different pig breeds. Furthermore, these results may also contribute to the research involving human obesity.


INTRODUCTION
Obesity is a heterogeneous disease that is associated with comorbidities, such as type 2 diabetes mellitus (T2DM), cardiovascular disease and cancer [1].It has been a major risk for human health.Therefore, it is urgent to elucidate the mechanism of obesity and develop effective treatment programs.In general, adipose tissue (AT) is considered a passive reservoir for energy storage that also functions in mechanical and heat insulation or participates in the regulation of thermogenesis.However, recently, it is recognized that AT synthesizes and secretes a variety of bioactive peptides, known as "adipokines," which participate in the regulation of glucose and lipid metabolism, energy homeostasis, feeding behavior, insulin sensitivity, inflammation, immunity, adipogenesis and vascular function or coagulation [2].Several studies have examined the role of dysfunctional adipose tissue in the pathogenesis of obesity.Therefore, the study of adipose tissue function is of great significance for revealing the mechanism of obesity.
The realization of adipose tissue function is closely related to adipocyte differentiation and fat metabolism.Previous studies showed that adipose cell differentiation and fat metabolism were regulated strictly by several transcription factors (TFs), including sterol regulatory element binding proteins (SREBPS), CCAAT/enhancer binding proteins (C/EBPS) and peroxisome proliferatoractivated receptor gamma (PPARγ).Recent studies indicated that microRNAs (miRNAs) also play critical roles in the adipose cell differentiation and fat metabolism.
miRNAs are a set of small noncoding RNAs (18-22nt)that bind to partially complementary sequences on target mRNAs resulting in the posttranscriptional regulation of gene expression [3].They play vital roles in cell differentiation, apoptosis, metabolism and other important biological processes [4][5][6][7].And these include their effects on adipocyte differentiation and fat deposition.For example, previous observations [6] argue that miR-14 is a dose-dependent regulator of DAG and TAG metabolism in Drosophila.Esau et al. reported miR-143 is involved in adipocyte differentiation [8].In addition, bta-miR-378 promotes the differentiation of bovine preadipocytes [9], ssc-miR-7134-3p regulates fat accumulation in castrated male pigs [10], miR-429 inhibits differentiation in porcine preadipocyte and so on [11].In this study, we identified the miRNAs which affect the adipocyte differentiation and fat metabolism in pigs.
The pig is an excellent model for studying obesity and diseases associated with obesity.One of the reasons for this is that pigs have the ability to generate fatty deposits in a manner that is similar to humans.Moreover, the development, morphology and function of the normal cardiovascular system in swine closely resemble that of humans [12].Pigs are also, like humans, omnivores, and develop spontaneous atherosclerosis with increased age and have lipoprotein profiles and a metabolism that is similar to humans [13].The Laiwu pig, a typical obese pig and that easily deposits fat, is a unique breed in the Shandong Province of China with characteristics of high adaptability, meat quality, fecundity and a strong combining ability.Large White pigs, also called Yorkshire pigs, originate from the United Kingdom and have the characteristics of a large size, high fecundity and adaptability, and high feed conversion rate and dressing percentage.Large White pigs are one of the world-class lean meat-type pigs.These two breeds show significant differences in fat accumulation and serve as ideal animal models for studying fat deposition and obesity.
Recently, there has been an increase in the utilization of Next Generation Sequencing (NGS) to examine the transcriptome for the identification of differential expression as well as for the opportunity to discover novel transcripts, including new alternative isoforms and miRNAs [14][15][16][17][18][19][20][21][22][23][24][25][26].Therefore, in the present study, a comparison of the miRNA profiles of the adipose tissue of the Laiwu pig and Large White pig was performed using RNA-seq.The data were further analyzed to identify the miRNAs that were related to adipopexis.This study provides information on the miRNAs that are involved in the accumulation of fat in pigs, and the target genes and pathways they regulate.Swine producers use genetics to rear pigs that have rapid lean growth.However, a balance has to be maintained as breeding pigs with a high proportion of lean meat has led to a decrease in the intramuscular fat content and therefore decreased pork quality.This study sheds light on the players in the adipose pathway that could be manipulated to rear pigs which show rapid lean growth accompanied by optimum intramuscular fat.Moreover, as the genetics of fat deposition is similar between pigs and humans, the data obtained in this study would also provide clues for solving the problems related to human obesity.

RESULTS miRNAs sequencing and mapping
We performed the miRNA sequencing on an Illumina Hiseq 2500 platform, and the reads were generated from the L library and D library.A total of approximately 26,486,906 raw reads were obtained.After removing the low-quality reads, adaptors and all possible contaminations, we obtained a total of 16,813,692 uniquely mapped reads, among which 8,259,953(64.09%) were from the L library and 8,553,739 (62.89%) were from the D library.The clean reads were mapped to the Rfam database and were distributed among several main categories, including miRNAs, rRNAs, snRNAs, tRNAs, other small RNAs and unannotated small RNAs (Table 1).The reads mapped to the unannotated small RNAs made up the highest proportion.The reads mapped to the miRNAs were the second most common (12.43% in L library and 14.28% in D library).The reason for the majority of the reads mapped to unannotated small RNAs may be due to the fact that these are not well studied and annotated.

Identification of known miRNAs and novel miRNAs
After mapping to the Rfam database, the remaining sequences were used for annotation of the conserved and novel miRNAs.In the Laiwu pig, 2660 known miRNAs and 584 novel miRNAs were identified.In the Large White pig, 2634 known miRNAs and 561 novel miRNAs were identified.Among them, 1851 miRNAs were common in both breeds.The novel miRNAs in this study were named by the following rules: species name-m (four-digit)-5p/3p, for example, ssc-m0644-3p.A statistical analysis of the sequence length of the mature miRNAs indicated that the lengths were mainly distributed at about 22nt, which is a typical size range for Dicer-derived products and is in line with the principle of miRNA processing.

Identification of differentially expressed miRNAs
The Laiwu pig and the Large White pig showed distinct fat depositions.To investigate the miRNAs that may be responsible for the significant differences in the fat characteristics, we identified the differentially expressed miRNAs between the Laiwu pig and Large White pig.The strict screening criteria are mentioned in the method section.The results indicated that 17 known miRNAs, including ssc-miR-133a-3p, ssc-miR-486, ssc-miR-1 and ssc-miR-204, were up-regulated in the Laiwu pig compared to the Large White pig (Supplementary Table 1).Meanwhile, 22 known miRNAs, including ssc-let-7i and ssc-miR-27b-3p, were down-regulated in the Laiwu pig compared to the Large White pig (Supplementary Table 2).In addition, ssc-miR-133a-3p, ssc-miR-486, ssc-miR-1, ssc-let-7i and ssc-miR-27b-3 were abundantly expressed in the Laiwu pig compared to the Large White pig.In addition, we identified 55 differentially expressed novel miRNAs between the Laiwu pig and the Large White pig.These differentially expressed miRNAs were used for further prediction of the targets.

Target gene prediction of the differentially expressed miRNAs
The identification of miRNA targets helps to illustrate the functions of the differentially expressed miRNAs identified in the present study.A total of 36,259 genes were predicted to be the potential targets of the 94 differentially expressed miRNAs in the Laiwu and Large White pig (Supplementary Table 3 and Supplementary Table 4).miRNAs can bind to partially complementary sequences on target mRNAs and result in the posttranscriptional regulation of gene expression, thus changing the biological processes in which their target genes are involved.Therefore, in order to further explore the function of the differentially expressed miRNAs, we mainly utilised the targeted genes for the subsequent functional analysis.

Gene ontology (GO) and KEGG pathway analysis of the targeted genes of differentially expressed miRNAs
GO terms were assigned to the differentially expressed miRNAs targeted genes based on their sequence similarities to known proteins in the UniProt database annotated with GO terms as well as the InterPro and Pfam domains they contain.The GO annotation and enrichment analysis of these targeted genes were done using the DAVID software (Supplementary Table 5 and Supplementary Table 6), in which the gene length bias was corrected.The GO terms with corrected P-value less than 0.05 were considered significantly enriched by the targeted genes.The distribution bar charts of the biological processes, cellular components and molecular functions are shown in Figure 1.From the perspective of the biological processes, the targeted genes of the up-regulated differentially expressed miRNAs were significantly enriched in cellular protein complex assembly (GO: 0043623) and protein complex biogenesis (GO: 007027).The targeted genes of the down-regulated differentially expressed miRNAs were significantly enriched in gluconeogenesis (GO: 000609), hexose biosynthetic processes (GO: 0019319) and carbohydrate biosynthetic process (GO: 0034637) and other GO terms related with material synthesis and energy metabolism.From the molecular function perspective, the targeted genes of the up-regulated and down-regulated differentially expressed miRNAs were mainly enriched in steroid binding (GO: 0005496), lipid binding (GO: 0008289) and protein phosphatase activity, which are related to the functions of enzyme activity and lipid binding.From the cellular component perspective, cytoskeleton (GO: 0005856) and organelle (GO: 0043228) were significantly enriched by the targeted genes of both the up-regulated and downregulated differentially expressed miRNAs.
To further probe the potential function of the miRNAs with differential expression, we used the DAVID software to test the statistical enrichment of the targeted genes of the differentially expressed miRNAs in the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway (Figure 2 and Supplementary Table 7).Figure 2 shows the results of the pathway enrichment, which clearly displays that the MAPK, PPAR and Wnt signaling pathways are the significantly enriched terms.This suggests that by targeting the genes in these pathways known to be involved in fatty acid metabolism, the differentially expressed miRNAs might regulate adipogenic differentiation and lipid metabolism.

Construction of the miRNA target network
To better display the interaction between the miRNAs and their target genes, we selected the target genes enriched in both significant GO BP terms and KEGG pathways and the genes reported to be associated with fat deposition to construct the miRNA target networks [27][28][29].As is shown in Figure 3 and Supplementary Table 8, one gene can be targeted by multiple miRNAs, and most miRNAs target more than one target gene.For example, ssc-miR-455-3p was directed to the targeted genes CPT1A and DKK3 and ssc-miR-874 was directed to STARD3 and MGLL.Importantly, among these targeted genes, SCD, SCD5, STARD3 and CPT1A are reported to be involved in adipocyte differentiation and fat deposition [27][28][29].

The relationship between differentially expressed miRNAs and the targeted genes in the PPAR pathway
As mentioned above，three important factors in the PPAR signaling pathway, SCD, SCD5 and CPT1A, participate in adipocyte differentiation and fat deposition and also are predicted to be the targets of ssc-miR-1, ssc-miR-486 and ssc-miR-133a-3p, respectively (Figure 4A).In addition, the expressions of these three miRNAs were highly abundant.Therefore, we described the relationships between the three differentially expressed miRNAs and the targeted genes in the PPAR pathway based on the STRING database.Figure 4B shows the complicated relationships between glycerol kinase (GK), peroxisome proliferator-activated receptor delta (PPARD), diazepam binding inhibitor(DBI), acyl-CoA binding protein, stearoyl-CoA desaturase (SCD), SCD5, cytochrome P450 family 7 subfamily A, polypeptide 1(CYP7A1), solute carrier family 27 member 2 (SLC27A2) and Carnitine palmitoyltransferase 1A (CPT1A) in the PPAR pathway.The miRNAs were also included in the analysis in order to illustrate the potential regulators.Therefore, as Figure 4B displays, ssc-miR-133a-3p, ssc-miR-1 and ssc-miR-486 might change the signal transduction in the PPAR signaling pathway by targeting the important genes CPT1A, SCD and SCD5, respectively, and thus, take part in the regulation of biological processes related to adipogenesis.

Validation of the sequencing data by qRT-PCR
To validate the sequencing data, six differentially expressed miRNAs were randomly selected and were further examined by qRT-PCR (Supplementary Table 9).The result showed that ssc-miR-486, ssc-miR-133a-3p, ssc-miR-122, ssc-miR-1, ssc-miR-204 and ssc-m0092-3p were significantly higher in the Laiwu pig.These data were consistent with the sequencing results.

DISCUSSION
Over one-third of the world population is now struggling with overweight or obesity issues, and a disease burden frequently associated with these conditions [30].Obesity is a well-known risk factor for metabolic diseases, such as T2DM, dyslipidemia, coronary heart disease, hypertension, non-alcoholic fatty liver disease (NAFLD) and stroke.It is also linked to dementia, obstructive sleep apnea and numerous types of cancers [31].Therefore, revealing the mechanism of fat deposition and adipocyte differentiation is of great significance for the treatment of obesity.Previous studies have focused on the mechanism of fatty deposits using rodents.For example, the rat 3T3-L1 cell line is usually used to study the differentiation of fat cells [32,33].Moreover, fat mice are also considered as animal models to study fat metabolism diseases [34].However, there is a gap in the fatty deposit locations and fatty deposits formation between humans and rodents.Pigs, with a maximum ability for fat deposition, are the most obese animals and are similar to humans in their manner of fat deposition.Therefore, pigs are ideal animal models to study human diseases related to adipose [13].Recently, researchers have studied the mechanism of obesity using pigs and made some progress [35].In this study, the Laiwu pig and Large White pig, with significant differences in adipopexis, were chosen as the animal models.The profiles of the miRNAs in the adipose tissue of these two different breeds were generated using RNAseq to further study the potential molecular mechanism of adipopexis.The results indicated that 40 miRNAs were up-regulated in the Laiwu pig, and meanwhile, 54 miRNAs were down-regulated in the Laiwu pig.The bioinformatics analysis of these differentially expressed miRNAs provided more information about the identification, classification and the possible targeted genes that might be related to adipopexis.
The GO enrichment analysis suggested that the targeted genes of the differentially expressed miRNAs were enriched in carbohydrate biosynthesis, lipid biosynthesis, steroid binding, enzyme activity and other functions involved in lipometabolism and energy metabolism.The KEGG enrichment analysis indicated that the PPAR, MAPK, Wnt and TGF-beta signaling pathways were enriched.Previous studies showed that these signaling pathways are related to adipocyte differentiation and fat metabolism.MAPK signaling pathway involves four pathways, including ERK1/2, JNK, p38 MAPK and ERK5/BMK1, which have played an important role in adipopexis [52][53][54][55].The Wnt signaling pathway is involved in the fate determination of the fat cell, adipogenic differentiation and lipometabolism [56,57].Kennsell et al. argued that the Wnt signaling pathway inhibited adipogenic differentiation either by depending on beta-catenin or not [58].The TGF-beta signaling pathway is also closely related with lipometabolism.Kim et al. suggested that miR-21 targeted the gene transforming growth factor beta receptor 2 (TGFBR2) to regulate adipogenic differentiation in human stem cells [36].Moreover, we noticed that PPAR signaling pathway is important in fatty acid metabolism, sterol metabolism and adipogenic differentiation.Figure 4 shows that CPT1A, SCD5 and SCD (SCD1) are enriched in this pathway, and they are targeted by ssc-miR-133a-3p, ssc-miR-486 and ssc-miR-1, respectively.SCD, with highly conserved sequences, exists in two forms of isomers, SCD1 and SCD5, in most vertebrates [59].SCD1 is the limiting enzyme in the transformation of saturated fatty acids to monounsaturated fatty acids, which plays an important role in fatty acid biosynthesis [14,60].SCD5 affects the components of aliphatic acid in the milk of Holstein cows, which is also an important regulatory gene in lipometabolism [61].CPT1 has three different  hypotypes, including CPT-1A, CPT-1B and CPT-1C.CPT1 is the limiting enzyme in the beta-oxidation of free fatty acids.CPT1 catalyzes long-chain fatty acyl coenzyme A and carnitine into acyl carnitine, which facilitates in the oxidation of fatty acids in mitochondria matrix [29].Malandrino et al. found that CPT1A promotes the oxidation of fatty acids in fat cells and macrophages, thus, reducing lipid accumulation and inflammation [62].Therefore, CPT1A plays a role in the inhibition of fat deposition.Based on the evidence mentioned above, ssc-miR-133a-3p, ssc-miR-486 and ssc-miR-1 may regulate lipid metabolism by targeting the genes CPT1A, SCD5 and SCD, respectively.Among these, ssc-miR-133a-3p may promote the fat deposition in pigs by targeting the gene CPT1A, which is in line with the high expression and up-regulation of ssc-miR-133a-3p in Laiwu pigs.However, how ssc-miR-486 and ssc-miR-1 effect lipid metabolism and deposition in the pig by targeting SCD5 and SCD requires further investigation.To validate the RNA-seq data, we selected six differentially expressed miRNAs and determined the expression of these RNAs by RT-PCR (Supplementary Table 9).We used three animals in each of the two breeds for the miRNA validation.Each miRNA's expression/fold changes for the Large White and Laiwu pig groups were compared.The expression levels obtained by RT-PCR were consistent with the RNA-seq results described above.Taken together, these findings suggest that the identified miRNA-targeted genes contribute to regulating genes and pathways that are involved in adipogenesis, which may help to define the fat deposition differences between two groups of pigs.
Recently, it has been shown that targeting PPARs is a potential way to treat NAFLDs [63,64].Since the miRNAs that we uncovered in our study regulate the PPAR pathway, this study helps us understand the players involved in the regulation of this pathway and could provide insights into better treatment for NAFLDs.

Ethics statement
All of the procedures involving animals were approved by the animal care and use committee at Institute of Animal Sciences, Chinese Academy of Agricultural Sciences where the experiment was conducted.All of the experiments were performed in accordance with the relevant guidelines and regulations set by the Ministry of Agriculture of the People's Republic of China.

Sample preparation
All the animals in this study were maintained in Laiwu Pig Farm (Shandong, China), and two different breeds were analyzed: Laiwu pig and Large White pig.They were housed under the same conditions, including free access to water and food in natural lighting.Three healthy pigs from each breed (150-day-old) were selected for the study.At slaughter age, the mean body weights of Laiwu pigs and Large White pigs were 35 kg and 90 kg respectively.The subcutaneous adipose tissues were collected after the animals were butchered, and these tissues were frozen in liquid nitrogen and stored immediately at -80°C for further RNA extraction.All the experimental procedures were carried out according to authorization granted by the Chinese Ministry of Agriculture.

Library preparation and miRNA sequencing
Total RNA was extracted from the subcutaneous adipose tissues of two different breeds using TRIzol (Invitrogen) as described by the manufacturer.The RIN or RNA integrity number was used to determine the RNA integrity quality of the extraction.Qualified total RNA was stored at -80°C for later miRNA separation.The deep-sequencing technology was used for the RNA-Seq [65].Two small RNA libraries representing the two species (coming from a pool of 3 Laiwu pigs and a pool of 3 Large White pigs) were constructed according to the Illumina ® TruSeq™ Small RNA Sample Preparation protocol.Briefly, small RNAs were isolated from the total RNA by a denaturing polyacrylamide gel electrophoresis (PAGE).The 5′-adaptor and 3′-adaptor were ligated to these small RNAs, followed by reverse transcription with SuperScript II Reverse Transcriptase (Invitrogen), and then, the cDNA was amplified by PCR and was purified to generate two small RNA libraries.Quality control analysis on the library was performed with the Qubit™ dsDNA HS kit and a Qubit ® 2.0 Fluorometer.High Sensitivity DNA Chip and Agilent 2100 system were used to determine the library size and purity.Finally, the qualified small RNA libraries were sequenced using the single-read multiplex program at the Beijing Biotechnology Corporation (Beijing, China).The two cDNA libraries from the two species (coming from a pool of 3 Laiwu pigs and a pool of 3 Large White pigs) were named as L and D, respectively.

Mapping and annotation
Reference genome and gene model annotation files were downloaded from the genome website directly.To analyze the miRNA sequencing results, the adaptor sequences and low quality sequences were initially filtered out from the raw reads by the fastx_toolkit (http:// hannonlab.cshl.edu/fastx_toolkit/).The clean reads were first aligned to the pig genome (NCBI, Sscrofa10.2) using the bowtie program (version 1.0.0) and were then searched against the Rfam database (http://rfam.xfam.org/) to remove rRNA, tRNA and snRNA.Moreover, the remaining sequences were compared with the mature miRNAs of human and animals, including cow, sheep, www.impactjournals.com/oncotargetand pig in miRBase (Release 21; http://www.mirbase.org/) to identify the conserved miRNAs.The hits were considered a real match if there was a minimum 16-nucleotide match between the sequence read and the miRNA from the database.The read sequences that matched with the reference mature miRNAs (allowing for one or two mismatching end-nucleotides) were annotated as miRNA candidates.The miRNA candidates were then clustered into categories according to sequence similarity, and the sequences varying only in length and/or a few end nucleotides were grouped under the same miRNA identifier or related keywords, which is typically used to identify miRNAs (https://www.mdc-berlin.de/36105849/en/research/research_teams/systems_biology_of_gene_ regulatory_elements/projects/miRDeep/documentation).

Identification of novel miRNAs
To identify potential novel miRNAs in pigs, 150 nucleotides of a sequence flanking each side of these miRNA candidate sequences were extracted.The secondary structures were then predicted using the program MIREAP (http://sourceforge.net/projects/ mireap/).If a hairpin structure with a free energy of hybridization lower than -20 kcal/mol was predicted, the RNA sequence was subjected to the miReap analysis, which predicts whether the input RNA sequence is a genuine pre-miRNA-like hairpin sequence.Any sequence that met the following three criteria was harvested: (1) mature miRNAs reside in one arm of the hairpin precursor, which are short of large internal loops or bulges; (2) the stem-loop structure is steady, with the free energy hybridization lower than -20 kcal/mol; and (3) hairpins located in intergenic regions or introns were considered as potential new miRNAs [66].

Differential expression analysis
In order to obtain the differentially expressed miRNAs, the RPM method was used to judge the expression level of the two samples.The differential expression analysis of miRNAs was performed using the DEGSeq R package (1.10.1).DEGSeq provided statistical routines for determining the differential expression in the digital gene expression data using a model based on the binomial distribution.The resulting P-values were corrected using the Benjamini and Hochberg's approach for controlling the false discovery rate.Differentially expressed miRNAs with a corrected P-value < 0.05 and a more than 2-fold difference were selected for further analysis.

Target prediction of the differentially expressed miRNAs
In mammals, miRNAs incompletely complementarily pair with the 3′-UTR region of the target genes to repress the transcription of key regulators related to growth, differentiation and other important pathways.The second base to the eighth base of the miRNAs are generally considered the 'seed' sequence, which is important to recognize and bind the 3′-UTR region of their targets.Therefore, to explore the potential function of the miRNAs with the significantly differential expression in porcine subcutaneous adipose tissue, the miRanda program (http://www.microrna.org/microrna/getDownloads.do)was applied to predict their target mRNAs in the linux system based on the complementary pairing of bases.The 3′-UTR sequences were extracted according to gene annotation information.The parameters were set as follows: score threshold at 140; energy threshold at-19 kcal/mol; scaling parameter at 4 and gapopen penalty at-9.A set of predicted target genes was designed for further analysis.

GO and KEGG pathway analyses of the miRNA target genes
The GO enrichment analysis of the target genes of the differentially expressed miRNAs was implemented by the DAVID software (https://david.ncifcrf.gov/).With all the porcine genes as the list of the background genes, the GO terms with corrected P-value less than 0.05 were considered significantly enriched terms.KEGG is a database resource for understanding highlevel functions and utilities of the biological system, such as the cell, the organism and the ecosystem, from molecular-level information, especially for large-scale molecular datasets generated by genome sequencing and other high-throughput experimental technologies (http:// www.genome.jp/kegg/).We used the DAVID software to test the statistical enrichment of the target genes of the differentially expressed miRNAs in the KEGG pathways.The KEGG terms with corrected P-values less than 0.05 were considered significantly enriched terms with the list of background genes of all porcine genes.

Interaction networks of miRNA and the target genes
Based on the String database (http://string-db.org/),we displayed the interaction networks of the miRNAs and their target genes in the PPAR signaling pathway and illustrated the relationships of the miRNAs with their target genes in this pathway.Moreover, in order to better demonstrate the relationships between differentially expressed miRNAs with their target genes, Cytoscape software was used to assess the relationships between miRNAs and their target genes that were significantly enriched in the GO and KEGG analyses and the target genes that were reported to be related to lipid metabolism in previous studies.

qRT-PCR identification
The qRT-PCR method was utilized to validate the expression of the miRNAs.For each miRNA, three biological replications were performed.Briefly, 2 µg of total RNA was reverse transcribed to cDNA using SG One-Step miRNA RT Kit (#Q1014, SinoGene, China), and then, SYBR Green real-time PCR was carried out using the synthesized cDNA as the template.The qRT-PCR reaction system comprised of 7.5 µl of 2 × SG Green qRT-PCR Mix, 0.25 µl Forward Primer, 0.25 µl of Reverse Primer, 1 µl cDNA and 6 µl nuclease-free Water, and the qRT-PCR reactions were performed by Step One PLUS (Applied Biosystems, Foster city, CA), with a condition of 95°C for 10 min, followed by 45 cycles of 95°C for 15s and 60°C for 15s, and a final stage of dissociation analysis.The porcine U6 gene was used as an internal control, and the ΔΔCt method was used to calculate the relative expression level of the miRNAs between the samples.The qRT-PCR results produced the amplification plot of an S type and the dissolution curve of a single peak, indicating that specific amplifications were attained with the primers designed by Sinogene Scientific (Beijing, China).

Statistical analyses
All of the data are presented as the means ± SD.For the comparisons, a Student's t-test was performed, and a corrected p < 0.05 was considered statistically significant.

CONCLUSIONS
Taken together, our study explored the potential role of differentially expressed miRNAs in fat deposition in pigs.Thirty-nine known miRNAs and 55 novel miRNAs were identified.The Gene Ontology and KEGG analyses indicated that the target genes of the differentially expressed miRNAs were involved in the PPAR, MAPK and Wnt signaling pathways and were related with lipometabolism.In addition, we first reported that ssc-miR-133a-3p, ssc-miR-486 and ssc-miR-1 regulate the target genes CPT1A, SCD5 and SCD, respectively, in the PPAR signaling pathway so that they can have an impact on lipidosis.These three miRNAs were upregulated in the Laiwu pig compared to the Large White pig.Regulation of the target genes CPT1A, SCD5 and SCD by ssc-miR-133a-3p, ssc-miR-486 and ssc-miR-1, respectively, may contribute to the molecular mechanism by which Laiwu pigs accumulate more fat when compared to the Large White pigs.Of course, these three targets and the mechanism by which these miRNAs regulate fat deposition still need to be verified in future experiments.Our findings, both computational and experimental, provide valuable information regarding the miRNAs involved in fat deposition and contribute to revealing the molecular mechanism of the gene regulation governing the development and physiology of adipose tissue in pigs.Moreover, our study provides data that might reveal insight into obesity-related diseases.

Figure 1 :
Figure 1: GO analysis of the targeted genes of the differentially expressed miRNAs.(A) represents the GO analysis of the targeted genes of the up-regulated differentially expressed miRNAs.(B) represents the GO analysis of the targeted genes of the downregulated differentially expressed miRNAs.The figure is composed of three parts: biological processes (BP), molecular functions (MF), and cellular components (CC).The significance level of enrichment was set at a corrected p < 0.05.

Figure 2 :
Figure2: KEGG enrichment analysis of the targeted genes of the differentially expressed miRNAs.The horizontal axis refers to the -log2P, while the vertical axis refers to the significantly enriched pathway with a corrected p < 0.05.A higher number for -log2P value indicates the promoted performance of the pathway enrichment.

Figure 3 :
Figure 3: Relationships between the differentially expressed miRNAs and their targeted genes.(A) represents the relationships between the up-regulated miRNAs with their targeted genes.(B) represents the relationships between the down-regulated miRNAs with their targeted genes.The circular nodes represent the targeted genes, while the square nodes represent the differentially expressed miRNAs.

Figure 4 :
Figure 4: Relationships between ssc-miR-133a-3p, ssc-miR-1 and ssc-miR-486 and the targeted genes CPT1A, SCD and SCD5 and the transcription factors binding to these miRNAs.(A) represents the result of the target gene prediction of ssc-miR-133a-3p, ssc-miR-1 and ssc-miR-486 using miRanda software.(B) represents the protein-protein interactions and the interactions of the miRNAs with the CPT1A, SCD and SCD5 genes in the PPAR pathway.(C) represents the interactions of ssc-miR-133a-3p, ssc-miR-1 and ssc-miR-486 with the transcription factors binding to them.The yellow nodes indicate the transcription factors, and the pink nodes indicate the miRNAs.