A Key Gene, PLIN1, Can Affect Porcine Intramuscular Fat Content Based on Transcriptome Analysis

Intramuscular fat (IMF) content is an important indicator for meat quality evaluation. However, the key genes and molecular regulatory mechanisms affecting IMF deposition remain unclear. In the present study, we identified 75 differentially expressed genes (DEGs) between the higher (H) and lower (L) IMF content of pigs using transcriptome analysis, of which 27 were upregulated and 48 were downregulated. Notably, Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis indicated that the DEG perilipin-1 (PLIN1) was significantly enriched in the fat metabolism-related peroxisome proliferator-activated receptor (PPAR) signaling pathway. Furthermore, we determined the expression patterns and functional role of porcine PLIN1. Our results indicate that PLIN1 was highly expressed in porcine adipose tissue, and its expression level was significantly higher in the H IMF content group when compared with the L IMF content group, and expression was increased during adipocyte differentiation. Additionally, our results confirm that PLIN1 knockdown decreases the triglyceride (TG) level and lipid droplet (LD) size in porcine adipocytes. Overall, our data identify novel candidate genes affecting IMF content and provide new insight into PLIN1 in porcine IMF deposition and adipocyte differentiation.


Introduction
Intramuscular fat (IMF), corresponding to pork marbling, refers to the fat located within muscles, which is associated with the number and size of intramuscular adipocytes [1]. Previous studies have indicated that IMF is closely related to flavor, juiciness, and tenderness [2,3]; lower IMF content results in dry and less-flavorful meat, and IMF content is positively associated with sensory quality traits of meat and consumer acceptability [4]. When IMF content increases from 1 to 3%, sensory quality is significantly improved [5]. In the past few decades, breeders have focused on lean meat yield and backfat thickness, which result in decreased IMF content [6,7]. At present, the average IMF content of longissimus dorsi muscle is about 2% in many lean pig breeds [8][9][10]. A previous study indicated that consumer preference for appearance in pork is strongly associated with IMF content [11], therefore improving IMF content has become an important objective in modern pig breeding programs [12]. Although the heritability of IMF content is relatively high, varying from 0.21 to 0.86, it is difficult to determine the IMF content in livestock, resulting in its rare use in pig breeding programs [1]. Therefore, developing of molecular breeding technologies could provide a valid way to improve the IMF content in pigs.
In recent years, many quantitative trait loci (QTL) and candidate genes affecting IMF content have been identified [3,[13][14][15][16]. To date, 25,610 QTLs representing 646 different traits have been curated in a pig QTL database (PigQTLdb). Among these QTLs, 105 are associated with intramuscular fat content (https://www.animalgenome.org/cgi-bin/QTLdb/SS/index, [17], release 34, 21 December 2017). Multiple differentially expressed genes (DEGs) related to IMF content were identified in the pig using microarray technology, including acetyl-coenzyme A acyltransferase 1 (ACAA1), lipoprotein lipase (LPL), hormone-sensitive lipase (HLS), fatty acid binding protein 3 (FABP3), and FLJ36031 (pFLJ) genes [18]. A genome-wide association study (GWAS) demonstrated that a mutation (C > A) located in intron 9 of phosphorylase kinase catalytic subunit gamma 1 (PHKG1) can affect its expression and is associated with meat quality traits, including IMF content [19]. Moreover, stearoyl CoA desaturase (SCD) and leptin (LEP) genes were upregulated in semimembranosus muscle with high IMF content using transcriptome analysis [20]. Melanocortin 4 receptor (MC4R), mitochondrial phosphoenolpyruvate carboxykinase (PEPCK), and SCD genes expression levels are significantly correlated with IMF content in Duroc × Shanzhu commercial crossbred pigs [16]. However, whether these are the key genes affecting IMF content remains ambiguous. Moreover, some results are not consistent, due to the differences in genetic background, technologies, and breed, etc. As biological technology has developed, high-throughput sequencing has become the efficient and convenient method to identify candidate genes influencing IMF.
Herein, we used RNA sequencing (RNA-Seq) technology to identify the novel candidate genes affecting pork intramuscular fat deposition. We further demonstrated that the candidate gene, PLIN1, is involved in the regulation of IMF content in pigs, and can regulate triglyceride (TG) level and lipid droplet (LD) size in porcine adipocytes. Overall, this study provides valuable information for identifying novel candidate genes affecting IMF content, and clarifies the molecular mechanism of candidate gene PLIN1 affecting IMF content.

Ethics Statement
All experiments were undertaken in accordance with the guidelines of the regional Animal Ethics Committee and were approved by the Institutional Animal Care and Use Committee of Nanjing Agricultural University (SYXK2011-0036, 6 December 2011).

RNA Extraction, Library Preparation and Sequencing
A total of 18 longissimus dorsi samples were used to perform the RNA-Seq, which were derived from 3 batches (Nos. 2, 3, and 5) randomly selected from 11 batches; three individuals with higher (H) (n = 3) and lower (L) (n = 3) IMF content were picked out from each batch. Total RNA was extracted using TRIzol (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's protocol, and RNA concentration and integrity were evaluated by an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Subsequently, an equal amount of RNA from 3 samples of each batch (3 with H IMF content and 3 with L IMF content) were pooled to generate one sequencing library. A total of 6 libraries were constructed using the NEBNext ® Ultra™ RNA Library Prep Kit for Illumina (NEB, Ipswich, MA, USA), following the manufacturer's instructions. Size, purity, and concentration of the sample library were evaluated by Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA), and sequencing was performed on an Illumina HiSeq 4000 (Illumina, San Diego, CA, USA) to produce 150 bp paired-end reads. All the RNA-Seq data related to this study have been deposited in Sequence Read Archive (SRA, National Institute of Health, Bethesda, MD, USA) with the accession code SRP125462.

Raw Data Filtering and Transcript Analysis
High-quality clean reads were obtained by removing low-quality (the percentage of low-quality bases of quality value ≤ 15 is more than 50% in a read), adapter-polluted, and high content of unknown base (N) reads (more than 5%) from raw reads. Then, clean reads were aligned to the pig reference genome (Sscrofa10.2) using HISAT [22]. Sample transcripts were reconstructed using StringTie [23], then reconstructed transcripts were compared to the reference annotation using Cuffcompare [24]. Coding potential of novel transcripts was predicted using the Coding Potential Calculator (CPC) [25]. Then, coded novel transcripts were merged with reference transcripts to obtain a complete reference, and downstream analysis was based on this reference.

Screening of Differentially Expressed Genes, and Gene Ontology and Kyoto Encyclopedia of Genes and Genomes Analysis
Clean reads were mapped to reference using Bowtie2 [26], and then gene expression levels were calculated with RSEM [27] and described by fragments per kilobase of exon per million fragments mapped values. Differentially expressed genes between two groups were determined with NOIseq [28], and the parameters were set to log2 (fold change) ≥ 1.00 and probability ≥ 0.7. To further understand the function of DEGs, Gene Ontology (GO) analysis was performed using the Blast2GO program [29]. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotation was conducted using the BLASTX algorithm against the KEGG database (http://www.genome.jp/kegg/, Kanehisa Laboratories, Kyoto, Japan). Functional enrichment of GO and KEGG was performed using the OmicShare tools, a free online platform for data analysis (http://www.omicshare.com/tools). Corrected p-value ≤ 0.05 was defined as significantly enriched.

Quantitative Reverse-Transcription PCR Analysis
Total RNA from longissimus dorsi muscle, other tissues (liver, spleen, kidney, heart, and adipose), and cells were extracted using TRIzol (Invitrogen, Carlsbad, CA, USA), and then reverse transcribed into complementary DNA (cDNA) using Primescript RT Master Kit (Takara, Dalian, China) according to the manufacturer's instructions. Primers of DEGs were designed by Primer 3 (http://bioinfo.ut.ee/ primer3/) [31,32] and are listed in Table S1. Real-time quantitative reverse transcription polymerase chain reaction (qRT-PCR) was performed with AceQ qPCR SYBR Green Master Mix (Vazyme, Nanjing, China) in a reaction volume of 20 µL. The cycling parameters were as follows: 95 • C for 5 min, followed by 40 amplification cycles, each at 95 • C for 10 s, then 60 • C for 30 s. All reactions were performed in triplicate for each sample. The glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene was used as internal control to normalize the relative expression of genes. The gene expression levels were calculated by the ∆∆CT value method [33].

Western Blotting
Longissimus dorsi muscle from pig cells were lysed with RIPA lysis buffer (Beyotime, Nanjing, China) supplemented with 1% protease inhibitor cocktail (Selleck, Houston, TX, USA). Protein concentration was measured using bicinchoninic acid (BCA) Protein Assay Kit (Beyotime, Nanjing, China). An equal amount of protein was run on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), then transferred to polyvinylidene difluoride (PVDF) membranes (Millipore, Billerica, MA, USA). The membranes were blocked with 5% bovine serum albumin (BSA) (Sigma, St Louis, MO, USA) for 1 h at room temperature and incubated with rabbit polyclonal to anti-pig PLIN1 antibody (1:1000) (Abcam, Cambridge, MA, USA) or GAPDH antibody (1:1000) (Abcam, Cambridge, MA, USA) at 4 • C overnight. Next, the membranes were incubated with HRP-conjugated goat anti-rabbit secondary antibody (1:2000) (Abcam, Cambridge, MA, USA) for 1 h at room temperature. The signals were detected with Pierce Fast Western Blot Kit (ThermoFisher Scientific, Waltham, MA, USA) on a chemiluminescence detection system (LAS-4000 Imager, Tokyo, Japan). The band intensities were quantified using ImageJ 1.42q software (National Institutes of Health, Bethesda, MD, USA), and the level of GAPDH was used to normalize the expression of the target protein.

Bodipy Staining
To stain for neutral lipids of porcine adipocytes, cells were washed twice with PBS, fixed in 4% PFA for 30 min, incubated with Bodipy (1:1000) for 10 min, and imaged by confocal microscopy.

Cellular Triglyceride Determination
For cellular TG level measurement, adipocyte cells were lysed with cell lysis buffer (Nanjing Jiancheng, Nanjing, China). Protein concentrations in extracts were measured by BCA Protein Assay Kit. Then, TG levels were determined by TG determination kit (Nanjing Jiancheng, Nanjing, China) according to the manufacturer's protocols. TG levels were normalized to the protein concentration of each sample.

Statistical Analyses
Statistical analyses were conducted using SPSS 20.0 (SPSS Inc., Chicago, IL, USA). Statistical significance of the difference between two means was calculated using Student's t test or ANOVA test. A value of p < 0.05 was considered to represent a statistically significant difference.

Phenotypic Variations between Extreme Groups
In this study, we measured the IMF content of longissimus dorsi muscle of 279 pigs in 11 batches. The values of mean, standard deviation (SD), minimum, maximum, and coefficient of variation (CV) of IMF content for each group and all groups are shown in Table S2. The results show that the mean and standard deviation of IMF content for the entire pig population were 2.64% and 0.91%, respectively. Moreover, IMF content maximum and minimum were 6.56% and 1.09%, respectively. To reduce the effect of intergroup differences in the RNA-Seq, we randomly chose three groups from 11 batches, and picked out three extremely high and three extremely low IMF content individuals from each group, and mixed RNA from three individuals equally to generate a pool. A total of six libraries were constructed ( Figure S1), which were named H2, H3, H5, L2, L3, and L5. Statistical analysis showed that the IMF content between higher (H) and lower (L) in each group was significantly different (p < 0.01; Table 1), while carcass weight was not different between the H and L IMF groups.

Summary of RNA Sequencing Data
Output of RNA-Seq data is shown in Table 2. After filtering low-quality, adapter-polluted, and high content of unknown base (N) reads, we acquired 268,826,758 clean reads (from 44,222,498 to 45,247,560 for each sample) and 40.33 giga (G) clean bases (from 6.63 G to 6.79 G for each sample) in six libraries. The average quality value of Q20 (Phred quality score > 20 and an error rate < 0.01) was 94.08% for each library. An average of approximately 65% of clean reads were mapped to the pig reference genome, in which 58.80% were uniquely mapped to the reference genome (Table 2). Moreover, we identified 13,568 novel transcripts, which contained 9628 coding transcripts and 3940 noncoding transcripts.

Gene Expression Analysis, and Differentially Expressed Genes Identification
Expression levels of all genes in the six sequencing libraries are listed in Table S3. A total of 16,798 genes were identified in all samples, and number of genes of each sample ranged from 13,851 to 14,686. To identify candidate genes affecting IMF content, we used NOIseq software to detect DEGs between the H and L IMF content groups. A total of 75 DEGs were identified, 28 upregulated and 47 downregulated genes (Figure 1). Detailed information for all DEGs is shown in Table S4. To validate the RNA-Seq results, 10 DEGs (coenzyme Q10A (COQ10A), dynactin subunit 2 (DCTN2), cell death inducing DFFA like effector c (CIDEC), fatty acid synthase (FASN), SCD, PLIN1, aldolase C fructose-bisphosphate (ALDOC), pentraxin 3 (PTX3), NADH ubiquinone oxidoreductase subunit A6 (NDUFA6), and COMM domain-containing protein 5 (COMMD5)) were randomly selected to examine the gene expression level using qRT-PCR. The results showed that expression patterns of these genes from qRT-PCR were consistent with RNA-Seq (Figure 2A), and the correlation of the two methods was relatively high, with a correlation coefficient of 0.91 ( Figure 2B), suggesting that DEGs identified from RNA-Seq were reliable in this study.

Gene Ontology and Kyoto Encyclopedia of Genes and Genomes Enrichment Analysis of Differentially Expressed Genes
To clarify the functional classification of DEGs in the regulation process of IMF, we performed GO analysis for DEGs. All GO terms of DEGs were summarized into three main GO categories, biological process, cellular component and molecular function, and 34 subcategories ( Figure 3A). Among these subcategories, the "cell" and "cell part" belonging to the cellular component contained the most DEGs. The enriched analysis results showed that 27, 27, and 39 GO terms were significantly enriched in biological process, cellular component, and molecular function, respectively (Table S5 and Figure 3B). In order to elucidate the biological pathways of DEGs, we further performed pathway enrichment analysis, and detailed information on pathway enrichment is listed in Table S6. In total, 13 pathways were significantly enriched ( Figure 4).

Association of PLIN1 Expression Level with Porcine Intramuscular Fat Content
Previous research indicated that the peroxisome proliferator-activated receptor (PPAR) signaling pathway has an important role in adipogenesis [34]. Intriguingly, the DEG PLIN1 was classified in the PPAR signaling pathway, which attracted our attention in the current study. Moreover, previous study has shown that PLIN1 is a lipid droplet protein and plays an important role in the lipolysis [35]. In addition, PLIN1 was chosen as an important candidate gene affecting IMF according to its expression abundance (medium), fold change (>2.0), and probability (>0.70). To demonstrate the association of PLIN1 expression level with IMF content in pigs, we detected PLIN1 messenger RNA (mRNA) and protein levels between the H and L IMF content groups. First, we examined the PLIN1 mRNA levels in 60 samples, 30 higher IMF individuals and 30 lower IMF individuals, and IMF content was significantly different between the two groups (p < 0.01) ( Figure 5A). Correspondingly, the mRNA level of PLIN1 was significantly higher in the H IMF content group than in the L IMF content group (p < 0.01) ( Figure 5B). We further confirmed that the PLIN1 protein level was significantly increased (p < 0.01) in the H IMF content group when compared to the L IMF content group ( Figure 5C-E).

PLIN1 Is Induced in the Process of Adipocyte Differentiation
To verify the relationship of PLIN1 with adipocyte differentiation, we detected the PLIN1 expression levels in different tissues and stages of adipocyte differentiation. Our results indicate that the PLIN1 gene was highly expressed in adipose tissue compared with other tissues ( Figure 6A). The expression of PLIN1 at mRNA and protein levels was significantly upregulated at 3, 7, and 11 days after induction of differentiation, while PLIN1 was not expressed in the preadipocytes ( Figure 6B,C).

Location of PLIN1 in the Longissimus Dorsi Muscle and Adipocytes
To confirm the location of PLIN1 in longissimus dorsi muscle of pigs, we conducted immunohistochemistry. The results show that porcine PLIN1 was localized in adipocytes of porcine longissimus muscle ( Figure 7A). Furthermore, the location of PLIN1 was examined in adipocytes of pigs using immunofluorescence staining, and it was localized around lipid droplets stained with Bodipy ( Figure 7B,C). immunofluorescence at the surface of lipid droplets (LDs) in porcine adipocytes. PLIN1 was stained with antibody PLIN1 (red). Lipid droplets were stained with Bodipy (green). Nuclei were stained with DAPI (blue). Scale bars: 10 µm.

PLIN1 Knockdown Decreases Triglyceride Level and Lipid Droplet Sizes in Pig Adipocytes
To confirm the molecular mechanisms of PLIN1 in lipid droplets formation and TG level, PLIN1 was knocked down in adipocytes using siRNA. We first detected the efficiency of PLIN1 depletion in pig adipocytes, and data indicated that mRNA and protein levels of PLIN1 were markedly decreased in the PLIN1 knockdown group compared to control ( Figure 8A-C). Triglyceride level was significantly reduced in PLIN1 knockdown adipocytes ( Figure 8D). Moreover, knockdown of PLIN1 resulted in a significant decrease of lipid droplet size in pig adipocytes ( Figure 8E,F).

Discussion
Intramuscular fat content is significantly related to the sensory quality of cooked meat, such as juiciness and tenderness [36]. The range of measured IMF content of longissimus dorsi muscle derived from 279 pigs was large. Therefore, extremely higher and lower IMF content groups are suitable for identifying the novel candidate genes affecting IMF content by RNA-Seq analysis.
A total of 13,568 novel transcripts were obtained in the study, which was consistent with the number of novel transcripts obtained from porcine digitorum longus and soleus muscles (10,962 and 9686, respectively) [37]. Furthermore, a previous study showed that 18,959 genes were detected in biceps femoris and soleus muscles [38]. Our results also indicated that the number of genes from longissimus dorsi muscle was similar to that from other muscles.
Moreover, GO enrichment analysis results indicated that most significant GO terms, including "acyl-(acyl-carrier-protein) hydrolase activity", "fatty acid synthase activity", and "myosin filament", may play an important role in IMF deposition. Interestingly, acyl-(acyl-carrier-protein) hydrolase activity and fatty acid synthase activity contained the differentially expressed gene FASN. The FASN gene plays a crucial role in the process of lipogenesis, and it is a large multienzyme involved in all steps of fatty acid synthesis in mammals [39]. FASN synthesizes long-chain fatty acids by using acetyl-CoA as a primer, malonyl-CoA as a two-carbon donor, and nicotinamide adenine dinucleotide phosphate (NADPH) as a reducing equivalent [40]. The mRNA abundance of FASN has a significant positive correlation with IMF content in longissimus dorsi muscle of Korean cattle [41]. Zhu et al. reported that FASN located at 51.3 Mb on Bos taurus autosome 19 (BTA19) is related to fatty acid composition using high-density single nucleotide polymorphism (SNP) array in Chinese Simmental beef cattle [42]. In the present study, FASN expression level was associated with IMF content, which was consistent with previous results [43].
Furthermore, we observed several significantly enriched pathways, including insulin signaling pathway, PPAR signaling pathway, fatty acid biosynthesis, fatty acid metabolism, and terpenoid backbone biosynthesis, suggesting that DEGs in these pathways may be related to IMF biosynthesis and metabolism. DEGs in these pathways, such as FASN, SCD, PLIN1, 3-phosphoinositide dependent protein kinase 1 (PDK1), and protein kinase cAMP-dependent type II regulatory subunit alpha (PRKAR2A) genes may be involved in IMF deposition. Notably, porcine SCD was reported to take part in the regulation and biosynthesis process of unsaturated fatty acids [43,44], and its polymorphisms were associated with fatty acid composition in pigs using genome-wide association studies [45]. Moreover, previous studies reported that the expression level of SCD is positively correlated with monounsaturated fatty acid in Duroc × Shanzhu commercial crossbreed [16], and mRNA and protein expression levels of SCD are significantly higher in fatty Wujin pigs with high IMF content than that in lean Landrace pigs with low IMF content [46]. In the present study, SCD expression level also showed a significant difference between the H and L IMF content groups, suggesting that it is an important candidate gene affecting IMF deposition.
Single nucleotide polymorphisms in the other DEGs associated with IMF content trait, including myosin, heavy chain 4 (MYH4), and cardiomyopathy associated 5 (CMYA5), have been reported in previous studies. Luo et al. performed GWAS of meat quality traits in pigs, and reported that the SNP (ALGA0067099) located at the MYH4 intron was significantly associated with IMF content [47]. CMYA5, named myospryn, belongs to the tripartite motif superfamily of proteins, and is highly expressed in the skeletal muscle and heart [48,49]. The SNP (A383C) of the CMYA5 gene is significantly correlated with meat quality traits, including drip loss and IMF in pigs [50]. Therefore, these DEGs act as candidates, and their SNPs are useful as genetic markers for IMF content. The DEG ALDOC, mapped to the chromosome 12 in pigs [51], activates Wnt pathway signaling by binding with GSK-3β [52]. Recent research found that ALDOC is differently expressed protein between Chinese native breeds and introduced Western breeds, identified using isobaric tag for relative and absolute quantitation (iTRAQ) in longissimus dorsi muscle [53].
The PPAR signaling pathway was significantly enriched in this study, and previous studies indicated that it is a key regulator of lipid metabolism [54]. Therefore, the PPAR signaling pathway is involved in IMF metabolism. Interestingly, the differentially expressed gene PLIN1, a lipid droplet coat protein, was included in this pathway. PLIN1 belongs to the perilipin family proteins, which share highly conserved N-terminal sequences (PAT domain) and a common affinity for intracellular neutral lipid storage droplets [55]. The PLIN1 gene is located on chromosome 15 in humans [56], while it is localized to chromosome 7 in pigs. In our study, the PLIN1 gene was also highly expressed in the adipose tissue, and was much lower in other tissues. Previous studies also demonstrated that PLIN1 is highly expressed in adipocytes and adipose tissues in many species, including mice and humans [57]. PLIN1-mediated LD formation can positively regulate sterol regulatory element-binding protein-1 (SREBP-1) activation during adipogenesis, resulting in TG accumulation [58]. Nuclear receptor peroxisome proliferator-activated receptor-γ (PPARγ) enhances PLIN1 expression through DNA demethylation on PPAR-response elements of PLIN1 gene promoter upon differentiation [59]. Our results show that the mRNA and protein levels of PLIN1 were significantly increased in adipocytes after induction differentiation, suggesting that PLIN1 is involved in the process of porcine adipogenesis. PLIN1-null mice had higher intake and normal body weight, while they had less body fat compared with wild-type mice [60]. In our study, the PLIN1 expression level in the H IMF group was significantly higher than in the L IMF group in pigs. The results indicate that PLIN1 may promote IMF content and deposition.
We confirmed that PLIN1 localizes to lipid droplets in porcine adipocytes, which is consistent with previous studies in human [61] and mouse [35] adipocytes. The results indicate that PLIN1 is an adipocyte-specific lipid droplet-association protein in porcine adipocytes. Furthermore, PLIN1 also localizes to adipocytes of porcine longissimus muscle in the study and a similar result is found in semimembranosus muscle [62]. These results suggest that PLIN1 functions as a lipid droplet-associated protein to regulate IMF deposition. Previous studies have shown that comparative gene identification-58 (CGI-58) binds to PLIN1 on lipid droplets under basal conditions in adipocytes, preventing interaction with adipose triglyceride lipase (ATGL) [63][64][65][66]. However, in times of energy deficit, CGI-58 can bind and activate ATGL, which catalyzes TG lipolysis [67]. Our data suggest that TG level was significantly decreased in the PLIN1 knockdown group compared to the control. Therefore, PLIN1 may increase the TG content of porcine adipocytes by inhibiting TG lipolysis. PLIN1 interacts with CIDEC to promote lipid droplet formation in 3T3-L1 pre-adipocytes [35] and human adipocytes [61]. Intriguingly, we found that the CIDEC mRNA expression level was significantly higher in the H IMF group than the L IMF group. These results suggest that PLIN1 may promote larger lipid droplets formation in the H IMF group by interacting with CIDEC, which increased the TG content in the H IMF group.

Conclusions
In summary, we used RNA-Seq to perform the comparative transcriptome analysis of porcine longissimus dorsi muscle with extreme phenotypes for IMF content in the study. The DEGs identified in our research provide a foundation for elucidating the molecular mechanisms of IMF deposition. Furthermore, our results provide new insights into PLIN1 influencing the porcine IMF trait.