Identification of differentially expressed lncRNAs in different daily weight gains of Jiangquan black pigs and functional analysis of LOC100518120

Abstract

Background

Jiangquan black pig is a novel breed under development, characterized by rapid growth, high-quality meat, strong stress resistance, and high reproductive performance. However, the molecular mechanisms underpinning their muscle development remain poorly characterized. To address this gap, we conducted whole-transcriptome RNA-seq on longissimus dorsi muscle samples from Jiangquan black pigs with divergent growth rates (fast-growing vs. slow-growing) to identify key regulatory factors of muscle development.

Results

We identified a total of 5,329 known and 331 novel lncRNAs from the transcriptomic data, among which 70 lncRNAs showed differential expression between the fast-growing and slow-growing groups. Functional enrichment analysis of these differentially expressed (DE) lncRNAs revealed that multiple DE-lncRNAs were associated with PPP3CB; three candidate lncRNAs (ENSSSCG00000036609, ENSSSCG00000047932, and the novel lncRNA LOC100518120) were further predicted as potential regulators of muscle growth. Subsequent functional validation in porcine skeletal muscle satellite cells confirmed that LOC100518120 significantly upregulates the expression of its downstream target gene G0S2, promotes cell proliferation and differentiation, and suppresses apoptosis.

Conclusions

Our study identifies key lncRNA candidates associated with muscle growth in Jiangquan black pig, and elucidates the pro-myogenic functional role of the novel lncRNA LOC100518120. These findings provide a theoretical foundation for optimizing molecular breeding strategies and further improving meat quality in this breed.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12864-026-13151-6.

Introduction

The Jiangquan black pig is a high-quality meat-type pig which was bred based on the original Yimeng black pig, importing the genes of Duroc pig for meat traits, and cultivated through years of scientific selection and breeding [1]. Approximately 35% of the genetic background of the Jiangquan black pig derives from the Duroc pig. The Yimeng black pig is highly regarded by consumers for its strong disease resistance, tolerance to coarse feed, and excellent meat quality [2]. The Jiangquan black pig retains these advantages of the Yimeng black pig while also inheriting fast weight gain and high lean meat rate content traits of the Duroc pig. Therefore, Jiangquan black pig have great economic potential and high comprehensive utilization value.

Skeletal muscle, the largest organ in mammals, undergoes growth and development influenced by a combination of genetic and environmental factors, significantly impacting pork quality [3]. Skeletal muscle is a highly complex and heterogeneous tissue formed by myogenic progenitor cells [4]. Myogenic progenitor cells proliferate during embryonic development, but their numbers decrease while the myonuclei reach a steady state and myofibrillar protein synthesis peaks [5]. Once the muscle matures, these myogenic progenitor cells enter a quiescent state and subsequently reside in the muscle as satellite cells [6]. Satellite cells will be activated in response to muscle damage, proliferate by mitosis, and differentiate into produce myoblasts. These myoblasts then fuse to form myotubes and ultimately mature myofibers, facilitating the repair of damaged muscle fibers [7]. Upon completion of the repair process, these satellite cells revert to a quiescent state [8]. Unraveling the molecular mechanisms governing skeletal muscle growth and development holds significant importance for both animal husbandry and biomedicine.

As a non-coding RNA produced by RNA polymerase II, lncRNAs are transcripts that are more than 200 nucleotides (nt) in length and lack an efficient open reading frame for translation [9]. They exhibit distinct features compared to mRNAs, such as lncRNAs act as scaffolds, decoys or signals and can act through genomic targeting, regulation in cis or trans, and antisense interference [10]. These non-coding transcripts typically modulate mRNA expression levels and play a crucial role in gene regulation during skeletal myogenesis [11, 12]. 29 lncRNAs were found to be associated with muscle development, metabolism, cell proliferation and apoptosis in sheep skeletal muscle [13]. The lncRNA has2os exhibits high expression in skeletal muscle and undergoes significant upregulation during skeletal cell differentiation. Knockdown of has2os has been shown to inhibit myoblast fusion and hinder the expression of key myogenic factors such as MyHC and Mef2C [14]. Another notable lncRNA, H19, in mice encodes two conserved miRNAs: miR-675-3p and miR-675-5p that facilitate muscle differentiation [15]. In summary, it is clear that lncRNAs play an indispensable role in muscle growth and development.

G0/G1 switch gene 2 (G0S2) was identified during the induction of cell-cycle arrest in blood monocytes [16]. Subsequent work has shown that G0S2 participates in mammalian cell proliferation, apoptosis, and differentiation. CRISPR/Cas9-mediated knockout of G0S2 in chicken pre-adipocytes markedly inhibits adipogenic differentiation while promoting proliferation and exerting a modest pro-apoptotic effect [17]. Most studies to date have focused on the role of G0S2 in adipose tissue and adipocytes, as well as on certain disease-related contexts [18]. The prevailing view is that G0S2 functions as a lipid-droplet-associated protein that potently suppresses lipolysis by directly inhibiting the rate-limiting lipase adipose triglyceride lipase (ATGL) [19]. Skeletal muscle is an organ that enables movement, drives energy metabolism, and stores intramuscular fat (IMF). In this tissue, G0S2 controls the rate of intracellular lipolysis, which in turn affects the release of free fatty acids and helps keep muscle energy homeostasis stable [20]. G0S2 expression correlates with IMF content and may affect meat juiciness and flavor [21]. Moreover, G0S2 is implicated in the regulation of apoptosis, autophagy, and mitochondrial function, processes that are closely linked to muscle homeostasis, atrophy, and regeneration. Integrated ATAC-seq and RNA-seq analyses of the longissimus dorsi muscle from Landrace and Duroc pigs have suggested a tight association between G0S2 and porcine skeletal muscle growth and development [22]. In models of exercise physiology and metabolic disease (e.g., insulin resistance), altered G0S2 expression in muscle has been observed [23], indicating that G0S2 may serve as a pivotal node linking lipid metabolism, energy balance, and muscle function.

This study aims to refine the molecular breeding strategy for Jiangquan black pig. To achieve this goal, it is essential to clarify the impact of lncRNAs on muscle growth and development in this breed. lncRNAs play a crucial role in the myogenic regulatory network [24]; however, their specific functions in Jiangquan black pig remain largely unexplored. For the first time, this paper systematically analyzes the basic characteristics of lncRNAs in the skeletal muscles of Jiangquan black pig, identifies differential lncRNAs associated with varying daily weight gains, and validates the function of the novel lncRNA LOC100518120. The functional validation of LOC100518120 fills a research gap regarding its role in porcine skeletal muscle development, thereby providing a molecular basis for improving breeding efficiency.

Materials and methods

Experimental animals

In this study, Jiangquan black pigs were obtained from Shandong Linyi Jiangquan Agricultural and Animal Husbandry Co., Ltd. We selected 50 healthy sows with clear pedigrees. and recorded their age, weight, and live backfat thickness until they reached around 100 kg in weight. All sows were raised under uniform growth conditions and feeding environments. After adjusting for age at the final measurement, 8 sows with half-sibling relationships were selected using average daily weight gain as a benchmark. 4 sows were categorized into a high average daily gain group (Group F, Avg. = 597.18 g), and 4 sows were categorized into a low average daily gain group (Group S, Avg. = 525.47 g), the specific sample details are shown in Table 1. Subsequently, the longissimus dorsi muscle from each of these eight sows was selected for sequencing analysis.

Table 1.

Sample details

Sports event	100 kg body weight at day of age(d)	Average daily weight gain(g)
F-1	225.2372422	592.9281218
F-2	204.1514714	606.2628437
F-3	221.2398973	597.0663707
F-4	204.2017631	592.4834349
S-1	237.368804	536.6089827
S-2	228.9644884	541.933069
S-3	231.2017631	523.1620152
S-4	228.3505238	500.2125052

RNA extraction, strand-specific library construction and sequencing

Total RNA was extracted using Trizol reagent kit (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. RNA quality was assessed on an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA) and checked using RNase free agarose gel electrophoresis. After total RNA was extracted, rRNAs were removed to retain mRNAs and ncRNAs. The enriched mRNAs and ncRNAs were fragmented into short fragments by using fragmentation buffer and reverse transcribed into cDNA with random primers. The second- strand cDNA were synthesized by DNA polymerase I, RNase H, dNTP (dUTP instead of dTTP) and buffer. Next, the cDNA fragments were purified with QiaQuick PCR extraction kit (Qiagen, Venlo, The Netherlands), end repaired, poly(A) added, and ligated to Illumina sequencing adapters. Then UNG (Uracil-N-Glycosylase) was used to digest the second- strand cDNA. The digested products were size selected by agarose gel electrophoresis, PCR amplified, and sequenced using Illumina HiSeqTM 4000 (or other platforms) by Gene Denovo Biotechnology Co. (Guangzhou, China).

Quality control for raw data

To ensure data quality, it is essential to filter the raw data before conducting data analysis, thereby mitigating potential analysis interference caused by invalid data. For this purpose, we use fastp [25] to remove reads with adapters, reads with N ratio more than 10%, reads composed entirely of A bases, and low-quality reads (defined as those where bases with a quality value Q ≤ 20 constitute over 50% of the total reads). Subsequently, this process yielded clean reads ready for further analysis.

Read alignment analysis

In our study, we used the Sus scrofa (Sus_scrofa.Sscrofa11.1.108.chr.gtf.gz) genome as the reference for our analysis.

In typical organisms, mRNA makes up only a small fraction of total RNA, with the majority being ribosomal RNA (rRNA). Despite our efforts to remove rRNA using the de-rRNA kit, residual rRNA may still be present due to sample quality and species variations. To safeguard subsequent experiments from rRNA interference, we employed the short reads comparison tool bowtie2 [26]. This tool compares clean reads to the species’ ribosomal database, eliminating reads that match ribosomes without allowing any mismatches. The remaining unmapped reads were then used for subsequent transcriptome analysis.

We conducted a reference genome-based comparison analysis using the HISAT2 software [27]. This software offers the, efficiently comparing spliced reads in RNA Seq data through both global and local search methods.

Identification of lncRNA

Because lncRNA originates from a complex source and different gene transcripts yield lncRNA with significant variations, we analyze lncRNA based on transcripts. Given the limited knowledge of lncRNA, new predictions require CNCI [28]、CPC2 [29]、feelnc [30] software. Initially, we assembled and merged reads using stringtie (v2.2.1) [31], reconstructed transcripts, and retained those with lengths ≥ 200 bp and at least 2 exons. Subsequently, using the transcripts reconstructed by stringtie, we predicted the coding potential of new transcripts using CPC2, CNCI, and feelnc software. The reliable prediction results were obtained by considering the intersection of transcripts lacking coding potential.

Differential expression analysis

The reads count data from the lncRNA expression level analysis underwent gene differential expression analysis using DESeq2 software [32]. Following read count normalization, the software calculated the hypothesis testing probability (pvalue < 0.05, |log2FC|> log2(1.5)) based on the model. Visualization of the differential expression was done through volcano plots, heat maps, and other methods based on the analysis results.

lncRNA-mRNA association analysis

In our study, we examined how lncRNAs are linked to mRNAs in three ways: through base complementary pairing, regulation of neighboring protein-coding genes’ transcription by lncRNAs, or correlation analysis of lncRNAs with co-expressed protein-coding genes [33].

• I.Antisense lncRNA Analysis: lncRNAs are involved in many post-transcriptional regulation processes, same as some other small RNAs like miRNA and snoRNA, those regulations are always related with complementary base pairing. Some antisense lncRNAs may regulate gene silencing, transcription and mRNA stability. In order to reveal the interaction between antisense lncRNA and mRNA, the software RNAplex [34] was used to predict the complement arycorrelation of antisense lncRNA and mRNA. The program contains Vienna RNA package [35], and the prediction of best base pairing was based on the alculation of minimum free energy through thermodynamics structure.
• II.lncRNA cis-regulation Analysis: One of the functions of lncRNAs is cis-regulation of their neighboring genes on the same allele. The up-stream lncRNAs which have intersection of promoter or other cis-elements may regulate gene expression in transcriptional or post-transcriptional level. The down-stream or 3’UTR region lncRNAs may have other regulatory functions. Thus, lncRNAs which had been previously annotated as “unknown region” were annotated again. lncRNAs in less than 10 kb up/down stream of a gene were likely to be cis-regulators. The cis target genes were then subjected to enrichment analysis of GO functions and KEGG pathways.
• III.lncRNA trans-regulation Analysis: Another function of lncRNAs is trans-regulation of co-expressed genes not adjacent to lncRNAs. We analysed the correlation of expression between lncRNAs and protein-coding genes to identify target genes of lncRNAs. Pearson correlation coefficient was used for samples ≥ 6, and protein-coding genes with absolute correlation more than 0.95 were then subjected to enrichment analysis of GO functions and KEGG pathways. Then enrichment analysis of GO functions and KEGG pathways were conducted in protein-coding genes in the network. Trans-regulation analysis would not be recommended if samples < 6.

Alternative splicing analysis

Alternative splicing (AS) is an important gene regulatory mechanism in eukaryotes. rMATS (version 4.0.1) [36] was used to classify 5 types of alternative splicing events (SE-Skipped exon; RI-Retained intron; MXE- Mutually exclusive exon; A5SS- Alternative 5’ splice site; A3SS- Alternative 3’ splice site) for each sample, and analyze differential alternative splicing events between samples. We identified AS events with a false discovery rate (FDR) < 0.05 in a comparison as significant AS events.

Cell culture and differentiation

The cells used in this study were all immortalized porcine skeletal muscle satellite cells (iCell Bioscience, Shanghai, China). Porcine skeletal muscle satellite cells were cultured in growth medium, at 37℃ in humid air containing 5% CO₂. When the cell density reaches 70%, the cells were induced to differentiate by using differentiation medium. Growth medium: high glucose Dulbecco’s modified Eagle medium (DMEM, Gibco, New York, NY, USA) add 10% fetal bovine serum (FBS, Gibco, New York, NY, USA) and 0.5% penicillin/streptomycin (Solarbio, Beijing, China). Differentiation medium: DMEM add 2% horse serum (Gibco, New York, NY, USA).

Cell transfection

Cell transfection was performed using Hieff Trans^® Liposomal 2000 Transfection Reagent (Yeasen, Shanghai, China). The overexpression-vector and siRNA were provided by genepharma (Shanghai, China). Cells were seeded in six-well plate and transfected when the cell density reached 70%-90%. siRNA or the overexpression vector was diluted with 250 µL of Opti-MEM, and another 250 µL of Opti-MEM was used to dilute the transfection reagent. After a 5-minute incubation, the diluted transfection reagent was slowly added dropwise to the siRNA or overexpression-vector solution. The mixture was gently mixed and incubated for 20 min to form the transfection complex. The growth medium in the six-well plate was discarded and replaced with 2 mL of DMEM containing 10% fetal bovine serum. 500 µL transfection complex was added to each well and mixed gently. After 4–6 h, the medium was replaced with growth medium.

CCK-8 cell proliferation assay

The CCK-8 cell proliferation assay kit was provided by beyotime (Shanghai, China). Cells were seeded in a 96-well plate and cultured until the cell density reached 80% before transfection. After transfection, 10 µL of CCK-8 reagent was added to each well at 12 h, 24 h, 36 h, and 48 h post-transfection. The cells were incubated in the dark for 2 h in a cell culture incubator [37]. The absorbance at 450 nm was measured, and cell viability was calculated using the following formula: [Experimental group - Blank group] / [Control group - Blank group].

Laser confocal microscopy

EdU-488 Cell Proliferation Assay Kit was provided by beyotime (Shanghai, China). Cells were seeded on 6-well plate cell crawls, add the EdU working solution to the 6-well plate and incubate for 2 h. Remove the culture medium, and add 4% paraformaldehyde to each well for fixation at room temperature for 15 min. Remove the fixation solution and rinsed three times by pre-cooled phosphate buffered saline (PBS). Add 0.3% Triton X-100 permeabilization solution and incubate at room temperature for 10 min. Remove the permeabilization solution and wash three times with PBS. Add 500 µL of Click reaction solution to each well and incubate at room temperature in the dark for 30 min. Remove the Click reaction solution and wash three times with PBS. Add an anti-fade mounting medium containing DAPI to the slides, use tweezers to place the coverslip, and quickly invert it onto the slide with the mounting medium. Observe the slide under Laser Confocal Microscopy soon as possible.

Flow cytometry analysis

The cell cycle analysis was performed using the Cell Cycle and Apoptosis Analysis Kit (Beyotime, Shanghai, China). The apoptosis analysis was performed using the Annexin V-FITC Apoptosis Detection Kit (Beyotime, Shanghai, China) [38].

Quantitative real-time polymerase chain reaction (RT-qPCR) analysis

Total RNA was extracted from samples by using RNAsimple Total RNA Kit (Tiangen Biotech Co., Ltd, Beijing, China). The cDNA was synthesized by using Evo M-MLV RT Mix Kit with gDNA Clean for qPCR Ver.2 (Accurate Biotechnology Co., Ltd, Hunan, China). We used the SYBR^® Green Premix Pro Taq HS qPCR kit (Accurate Biotechnology Co., Ltd, Hunan, China) as the reaction reagent for RT-qPCR, the specific reaction system is shown in Table 2. Run the formulated 20µL system on a Light Cycler 96 real-time PCR system (Roche, Basel, Switzerland), the specific reaction program is shown in Table 3. The primer sequence information is shown in Table 4. The relative gene expression levels were calculated by using the 2^−ΔΔCt method. Graphs were created by using GraphPad Prism 10.1.2.

Table 2.

Reaction system for real-time quantitative PCR

Component	Addition
2× SYBR Green Pro Taq HS Premix	10µL
Template	2µL
Primer F (10µM)	0.4µL
Primer R (10µM)	0.4µL
RNase free water	7.2µL

Table 3.

Reaction program for real-time quantitative PCR

Steps	Reaction Temperature	Time	Cycles
Step 1	95℃	30 s	1
Step 2	95℃	5 s	40
	60℃	30 s
Step 3	Dissociation Stage

Table 4.

Primers for real-time quantitative PCR and RACE

Genes	Sequence (5’-3’)
MSTRG.10850.1(LOC100518120)	F: AACTTTTGAAAACTGTCGGGGTTCC R: AGAGATTCTGCTTCCTGCTCCATTG
MSTRG.7493.1	F: TTTACTATTCCAGCAGGACACC R: GATTACAGACCAGTTCGCCAC
ENSSSCT00000102851	F: TCAGTGCTGCGGTCATTGCTT R: TCGCCGATTCCCATCTGCTTT
MSTRG.16372.2	F: GTGAGACAAGTCGTCCCATAGTTC R: CCCGCTGCTCCTACGAAAGTC
MSTRG.3088.2	F: CAACTCATTTGGCCTGTCATCC R: CCATCAACCCTTTGGCACTCT
MSTRG.8291.2	F: TTACCAAGCCACAATCACCAA R: ATCCACTCATTTCCTCCACCC
GADPH	F: AAGTTCCACGGCACAGTCAAG R: ACCAGCATCACCCCATTT
G0S2	F: CTCTGGCCAAGGAGCTGATG R: CGCCGCTTGTCCTTCTCCA
MYOD	F: CCGCTCCGCGACGTAGATT R: GGCAGGGAAGTGCGAGTGTT
MYOG	F: GAGACATCCCCCTATTTCTACCA R: GCTCAGTCCGCTCATAGCC
MEF2C	F: TTGGGTTGATGAAGAAGGCTTATGAG R: GCTTGTTGGTGCTGTTGAAGATG
PCNA	F: GGTCCAGGGCTCTATC R: TTCATTGCCAGCACAT
CCND	F: GGCCTCGAAGATGAAGGAG R: GGCGGGTTGGAAATGAA
CDK4	F: CATCCCAATGTTGTCCG R: TCTGGCTTCAGGTCTCG
RACE-5’	R1: AGTTGGTTCTGGCTGATGGTT R2: GGATCTGGAGAAGAGATTCTGCTT
RACE-3’	F1: TTCAATGGAGCAGGAAGCAGAA F2: GAGGTCCAGGAGAATACCAGAGA

Statistical analysis

All cellular experiments were conducted with 3 independent biological replicates, and RT‑qPCR was performed with 3 technical replicates per biological sample. All quantitative data were analyzed using SPSS Statistics 22.0 software (IBM Corp., Armonk, NY, USA). Differences between two independent groups were compared using the unpaired Student’s t‑test, while comparisons among three or more independent groups were assessed by one‑way analysis of variance (ANOVA). Results are presented as mean ± standard error of the mean (SEM). Statistical significance was predefined as *P < 0.05, **P < 0.01, and ***P < 0.001.

Results

Quality analysis of sequencing data

We conducted quality control analysis on the results of RNA-seq (CRA024215). Following the results of quality control, the sequencing results revealed that Q20% was exceeded 97.47%, Q30% was surpassed 93.46%, N% was below 0.01%, and the GC content was above 55.90% (Table 5). These findings suggest that the obtained CleanData is of high quality and suitable for subsequent transcriptome analysis. In each individual samples, 61.73% to 78.56% of the clean reads were mapped to the Sus scrofa reference genome (Table 6). The mapping results indicated that the quality and reliability of the sequencing data were relatively high.

Table 5.

Statistics of sequencing read quality control

Sample	F-1	F-2	F-3	F-4	S-1	S-2	S-3	S-4
CleanData (bp)	14,169,221,396	15,455,529,768	14,920,893,046	15,350,988,150	15,442,610,918	14,437,471,550	14,095,330,254	14,017,378,016
Q20 (%)	13,820,127,751 (97.54%)	15,078,387,256 (97.56%)	14,571,191,202 (97.66%)	15,024,061,510 (97.87%)	15,051,361,427 (97.47%)	14,082,357,667 (97.54%)	13,752,865,783 (97.57%)	13,671,090,868 (97.53%)
Q30 (%)	13,247,777,202 (93.50%)	14,478,846,918 (93.68%)	14,003,985,159 (93.85%)	14,489,584,711 (94.39%)	14,432,699,483 (93.46%)	13,524,312,905 (93.68%)	13,201,990,565 (93.66%)	13,125,187,571 (93.64%)
N (%)	19,211 (0.00%)	194,491 (0.00%)	18,745 (0.00%)	193,383 (0.00%)	195,029 (0.00%)	181,158 (0.00%)	175,519 (0.00%)	176,379 (0.00%)
GC (%)	7,920,814,869 (55.90%)	8,879,556,177 (57.45%)	8,647,668,938 (57.96%)	8,869,412,762 (57.78%)	8,692,984,968 (56.29%)	8,391,017,741 (58.12%)	7,982,128,285 (56.63%)	8,068,776,013 (57.56%)

CleanData (bp) Total number of high-quality data bases after filtering (bp)

Q20 (%) The number of bases in the CleanData with quality values at Q20 level or higher, and its percentage in CleanData

Q30 (%) The number of bases in the CleanData with quality values at Q30 level or higher, and its percentage in CleanData

N (%) Number of single-ended reads containing N bases, and its percentage in CleanData

GC (%) The number of G and C bases out of the total bases, and its percentage in the total bases

F Fast daily gain group

S Slow daily gain group

Table 6.

Statistics of sequencing read alignment

Sample	F-1	F-2	F-3	F-4	S-1	S-2	S-3	S-4
Total Reads	93,764,112	102,461,148	98,837,426	101,381,450	102,496,124	95,468,576	93,291,772	92,882,370
Unmapped (%)	20,100,518 (21.44%)	33,641,011 (32.83%)	35,378,899 (35.80%)	31,696,984 (31.27%)	39,220,883 (38.27%)	34,854,881 (36.51%)	28,027,027 (30.04%)	29,182,413 (31.42%)
Unique Mapped (%)	59,456,443 (63.41%)	50,766,830 (49.55%)	49,136,059 (49.71%)	49,469,362 (48.80%)	48,499,989 (47.32%)	45,041,864 (47.18%)	52,510,876 (56.29%)	50,245,292 (54.10%)
Multiple Mapped (%)	14,207,151 (15.15%)	18,053,307 (17.62%)	14,322,468 (14.49%)	20,215,104 (19.94%)	14,775,252 (14.42%)	15,571,831 (16.31%)	12,753,869 (13.67%)	13,454,665 (14.49%)
Total Mapped (%)	73,663,594 (78.56%)	68,820,137 (67.17%)	63,458,527 (64.20%)	69,684,466 (68.73%)	63,275,241 (61.73%)	60,613,695 (63.49%)	65,264,745 (69.96%)	63,699,957 (68.58%)

Prediction of novel lncRNAs

The transcripts were aligned with the Sus scrofa (Sus_scrofa.Sscrofa11.1.108.chr.gtf.gz) reference genome, a total of 5,329 known lncRNA genes were identified. After transcripts reconstructed by using stringtie [31], 7453 novel transcripts were obtained. A total of 1,192 transcripts were retained through CNCI filtering [28], 1,053 transcripts were retained through CPC2 filtering [29], and 443 transcripts were retained through Feelnc filtering [30]. The intersection of these three software programs preserved 331 transcripts (Fig. 1A). These 331 transcripts were considered as novel lncRNAs. Together with the known lncRNAs, a total of 12,051 lncRNAs were identified.

Fig. 1.

The Venn diagram and genomic characterization of lncRNAs. A The Venn diagram of CPC2, CNCI and Feelnc software prediction. Pink represents CNCI, blue represents CPC2, yellow represents Feelnc. B Number of lncRNAs exons. C The chromosome distribution of lncRNAs. D Transcript length of lncRNAs. E Different types and quantities of lncRNAs

Genomic characterization of lncRNAs

The statistics on the distribution of exon numbers in lncRNAs revealed that the majority of exons consisted of 1 or 2 exons (Fig. 1B). In muscle tissue, the predominant chromosomal localization for most lncRNAs is on chromosome 1, with subsequent notable occurrences on chromosomes 6 and 13 (Fig. 1C).

The statistical analysis of lncRNA lengths revealed that the prevalent length interval among the samples was above 5000 nt, closely followed by the range of 3000 nt to 3500 nt (Fig. 1D).

Considering the genomic positioning of lncRNAs in relation to protein-coding genes, the Intergenic LncRNA category was the most prevalent, constituting 89% of the total (Figure 1E). Among the novel genes, Intergenic lncRNAs, Sense lncRNAs, and Antisense lncRNAs were the most abundant, collectively comprising 76% of the total.

Identification of differentially expressed lncRNAs

Following the completion of a comparative analysis of expression levels between groups F and S, 70 differentially expressed lncRNAs were identified, with 40 lncRNAs upregulated and 30 lncRNAs downregulated (Fig. 2A). The heatmap illustrates the significantly different in expression levels of these differentially expressed lncRNAs between Group F and Group S (Fig. 2B).

Fig. 2.

Statistics of differentially expressed lncRNAs, and interaction network of lncRNA-mRNA gene pairs. A Volcano plot analysis of 70 DE lncRNAs. The red dot represents the up-regulated gene, the blue dot represents the down-regulated gene. B Heatmap plots of DE lncRNAs. Rows represent lncRNAs, columns represent samples. The red color represents the higher expression of the gene in the sample and the green color represents the lower expression of the gene in the sample. C Interaction network for top 50 of lncRNA-mRNA gene pairs. The green nodes represent lncRNA. The yellow nodes represent trans-target gene

Target gene prediction of differentially expressed lncRNA

We conducted association analysis between the differentially expressed lncRNAs and mRNAs in RAN-seq. In the association analysis between lncRNAs and mRNAs, we identified 3,610 lncRNAs had cis-regulation target genes within 10,000 bp upstream and downstream. These 3,610 lncRNAs formed 4,603 lncRNA-mRNA gene pairs with 3,463 mRNAs. But only 1 differentially expressed lncRNAs formed 1 lncRNA-mRNA gene pairs with 1 mRNA. Using the Pearson correlation coefficient method, we identified 52 differentially expressed lncRNAs formed 119 lncRNA-mRNA gene pairs with 83 trans-acting target mRNAs. The top 50 lncRNA-mRNA gene pairs with the most significant p-values were shown in Figure 2C.

GO enrichment and KEGG pathway analyses

In the GO enrichment analyses, the top Biological Progress terms mainly included cellular process, biological regulation, regulation of biological process, metabolic process, response to stimulus, and other terms. The top Cellular Component terms mainly included cellular anatomical entity, protein-containing complex. The top Molecular Function terms mainly included binding, catalytic activity, transcription regulator activity, and other terms (Fig. 3A-Antisense, 3A- Cis, 3A- Trans).

Fig. 3.

GO enrichment analysis and KEGG enrichment analysis. A GO enrichment top 20. B KEGG enrichment top 20. Outer Circle: The top 20 enriched GO terms/pathway ID. Different colors represent different Ontologies/A class. Second Circle: The number of genes in this GO/pathway term (length) and the Q value (colour). Third Circle: The number of differentially expressed genes in this GO term/pathway. Inner Circle: The RichFactor value of each GO term/pathway, each one grid represents 0.1

The top 20 antisense-regulation KEGG pathways mainly included CGMP-PKG signaling pathway, Dopaminergic synapse, VEGF signaling pathway, Proteoglycans in cancer, and other pathways (Fig. 3B-Antisense).

The top 20 cis-regulation KEGG pathways mainly included Herpes simplex virus 1 infection, Metabolic pathways, Cysteine and methionine metabolism, Toll and Imd signaling pathway, and other pathways (Fig. 3B-Cis).

The top 20 trans-regulation KEGG pathways mainly included RIG-I-like receptor signaling pathway, hepatitis C, Alcoholism, influenza A, and other pathways (Fig. 3B-Trans).

In the above analysis, we identified several potential key regulators of muscle development, including ENSSSCG00000036609, ENSSSCG00000047932, MSTRG.5248.1, and the novel lncRNA LOC100518120. Notably, the target gene of MSTRG.5248.1 was PPP3CB, which was significantly enriched in the Gene Ontology terms muscle cell development (GO:0055001), muscle cell differentiation (GO:0042692), and striated muscle cell differentiation (GO:0051146).

Alternative splicing analysis

In our investigation, a total of 5 alternative splicing events were detected (Fig. 4A, B), with skipped exon (SE) being the most prevalent, constituting 78% of all alternative splicing (AS) events observed. And SE accounted for 77% of the differentially expressed AS events. So, SE were the most prevalent AS events observed in this research.

Fig. 4.

Statistics of AS events and sequencing data validation. A The proportion of novel lncRNAs to known lncRNAs in AS events (B) The number of differentially expressed AS events. C Comparison of sequencing data and RT-qPCR results. D Correlation of RT-qPCR and sequencing data

RT-qPCR validation of the gene expression data from RNA-seq

To validate the precision of RNA-seq results, six differentially expressed lncRNAs were randomly chosen for RT-qPCR analysis. The correlation and p-value were obtained after linear fitting of the log2 (fold change). The experimental results of these genes were consistent with the sequencing results, indicating that the sequencing results in this study were reliable (Figure 4C, D).

The gene characteristics of LOC100518120

Building on the aforementioned studies, in GO enrichment and KEGG pathway analyses, we identified the most significantly differentially expressed gene G0S2, which is a key regulator of the G0/G1 cell cycle switch. We predict that G0S2 functions as the target gene of the novel lncRNA MSTRG.10850.1. After a BLAST on NCBI, we found that MSTRG.10850.1 shares a 100% similarity with LOC100518120. However, LOC100518120 is a predicted lncRNA from genomic analysis, and there are no existing studies on this gene.

Therefore, we did further research on LOC100518120. Using known core sequences, primers were designed for rapid-amplification of cDNA ends (RACE) and sequencing, which yielded the full-length sequence of this gene. In the 5’ and 3’ amplifications, gene fragments of 206 bp and 321 bp were obtained (Fig. 5A). These fragments were then spliced together, resulting in a full-length sequence of 557 bp. Through algorithmic prediction (Fig. 5B) and nuclear-cytoplasm separation experiments (Fig. 5C), it was found that the gene is primarily localized in the cytoplasm.

Fig. 5.

The gene characteristics of LOC100518120. A Gel electrophoresis of RACE. B Algorithmic prediction of LOC100518120. (C) RT-qPCR results of nuclear-cytoplasm separation

Functional validation of LOC100518120

LOC100518120 promotes proliferation of porcine skeletal muscle satellite cells

We used RT-qPCR to examine the targeting of LOC100518120 to G0S2, as well as its effect on the expression of proliferation marker genes in porcine skeletal muscle satellite cells. The results revealed that overexpression of LOC100518120 significantly increased the expression of its target gene G0S2, which in turn significantly upregulated the expression of proliferation marker genes PCNA and CCND (Fig. 6A). Conversely, knockdown of LOC100518120 significantly reduced the expression of its target gene G0S2 and significantly decreased the expression levels of marker genes PCNA, CCND, and CDK4 (Fig. 6B).

Fig. 6.

The effect of LOC100518120 on cell proliferation. A (B) The effect of LOC100518120 overexpression/knockdown on G0S2 and proliferation marker genes. C (D) The effect of LOC100518120 overexpression/knockdown on the cell proliferation cycle. E (F) CCK-8 assess the effect of LOC100518120 overexpression/knockdown on cell activity. G (H) (I) (J) Observation of EdU staining by the InCell‑2000 confocal microscope (Scale bar: 100 μm). (*P < 0.05, **P < 0.01, ***P < 0.001, n = 3)

After treating the cells with a cell cycle detection kit, flow cytometry was used to analyze the results. We found that, compared to the control group, overexpression of LOC100518120 reduced the number of cells arrested in the G0/G1 phase and significantly increased the proportion of cells in the S phase, thereby promoting cell proliferation (Fig. 6C). In contrast, interference with LOC100518120 led to an increase in the number of cells arrested in the G0/G1 phase and a significant reduction in the proportion of cells in the S phase, thereby inhibiting cell proliferation (Fig. 6D).

Cell proliferative activity was assessed using the CCK-8 assay, and the results showed that, compared to the control group, overexpression of LOC100518120 significantly enhanced cell proliferative activity (Fig. 6E), whereas interference with this gene suppressed cell proliferative activity (Fig. 6F).

EdU staining was used to observe the effect of LOC100518120 on cell proliferation. After overexpression of LOC100518120, the EdU-positive cell (green fluorescence) rate was significantly increased (Fig. 6G, H). In contrast, interference with LOC100518120 led to a significant decrease in the EdU-positive cell rate (Fig. 6I, J).

LOC100518120 promotes differentiation of porcine skeletal muscle satellite cells

LOC100518120 overexpression vector and siRNA was transfected into porcine skeletal muscle satellite cells and induced differentiation. the cells were collected at D0 and D5. The results showed that on day 0 of differentiation, overexpression of LOC100518120 significantly increased the expression of differentiation marker genes MYOD, MYOG, and MEF2C (Fig. 7A), while interference with LOC100518120 significantly decreased the expression of these genes (Fig. 7B). On day 5 of differentiation, compared to the control group, the expression of differentiation marker genes MYOD, MYOG, and MEF2C was significantly higher in the cells treated with LOC100518120 overexpression (Fig. 7C). In contrast, cells treated with LOC100518120 interference showed a significant reduction in the expression of these differentiation marker genes (Fig. 7D).

Fig. 7.

The effect of LOC100518120 on cell differentiation. A (B) The effect of LOC100518120 overexpression/knockdown on differentiation marker genes on day 0 of differentiation. C (D) The effect of LOC100518120 overexpression/knockdown on differentiation marker genes on day 5 of differentiation. (*P < 0.05, **P < 0.01, ***P < 0.001, n = 3)

LOC100518120 inhibits apoptosis in porcine skeletal muscle satellite cells

Cells were treated with the apoptosis detection kit, and cell apoptosis was analyzed using flow cytometer. After overexpressing the LOC100518120 gene (Figure 8A), the number of apoptotic cells significantly decreased (P < 0.001). In contrast, after silencing the LOC100518120 gene (Figure 8B), apoptotic cells significantly increased (P < 0.001). These results suggest that the LOC100518120 gene plays a role in inhibiting apoptosis.

Fig. 8.

The effect of LOC100518120 on cell apoptosis. A The effect of LOC100518120 overexpression on the cell apoptosis. B The effect of LOC100518120 knockdown on the cell apoptosis. (*P < 0.05, **P < 0.01, ***P < 0.001, n = 3)

Discussion

The growth rate of skeletal muscle plays a pivotal role in pork production efficiency. Muscle quality significantly influences the taste and texture of pork. Previous research has predominantly examined the impact of mRNA on muscle growth. To delve deeper into the regulatory effect of lncRNAs on mRNA during muscle growth, we employed RNA-seq to effectively identify differentially expressed genes. Significant differences in gene expression between groups F and S were observed in this study. A total of 70 differentially expressed lncRNAs were identified.

During sequencing data alignment, the mapping rate was found to range from 61.73% to 78.56%. This may be attributable to Jiangquan Black Pig is a novel pig breed currently under development, for which no dedicated reference genome has been established to date. However, the core of this study is the analysis of differentially expressed lncRNAs between group F and group S; therefore, the relatively low mapping rate exerts only minimal influence on the main findings and conclusions of this paper.

In this study, many growth-related terms were enriched in the GO enrichment analysis. The KEGG enrichment analysis revealed a significant enrichment of the glucagon signaling pathway, carbohydrate digestion and absorption, protein export, MAPK signaling pathway, and other pathways. These pathways are all related to muscle growth and development. Our research sets the stage for future validation of the influence of lncRNAs in muscle growth.

PPP3CB functions by dephosphorylating its substrate proteins to regulate their physiological activity [39]. ATOH8 and bHLH transcription factor, plays a pivotal role in muscle development [40]. PPP3CB exhibits specificity in binding to and modulating the physiological activity of ATOH8, thereby influencing ATOH8’s role in muscle development [41]. FGF23, a fibroblast growth factor, is subject to gene expression inhibition by PPP3CB [42]. In the GO enrichment analysis of antisense-regulation genes in our experiment, the antisense-regulation target gene of lncRNA MSTRG.5248.1, PPP3CB (ENSSSCG00000010301), was significantly enriched in muscle cell development (GO:0055001), muscle cell differentiation (GO:0042692), and striated muscle cell differentiation (GO:0051146), indicating its significant involvement in muscle growth and development. Notably, MSTRG.5248.1 exhibited higher expression levels in group F compared to group S, suggesting a potential positive regulatory effect on muscle growth and development.

It has been shown that differential expression of lncRNAs affects tumor progression and that genes associated with tumor progression are also highly correlated with cell proliferation [43]. Among the trans-regulation target genes, ENSSSCG00000036609 and ENSSSCG00000047932 correspond to TBXT and HSPA6, respectively, both of which are influenced by multiple lncRNAs identified in this study, as depicted in the interaction network diagram. TBXT and HSPA6 are implicated in cancer, with TBXT suppression hindering epithelial-mesenchymal transition (EMT) and cell migration in breast cancer, while induced HSPA6 expression promotes proliferation and migration of osteosarcoma cells [44]. This predicts that multiple lncRNAs associated with both target genes may affect tumor cell development, and implies that these lncRNAs may affect muscle cells proliferation.

The fat content and fat oxidation in skeletal muscle profoundly influence its quality and texture [45]. G0S2, a G0/G1 switch gene, acts to inhibit triglyceride lipase (ATGL) activity and reducing lipolysis in adipocytes [46]. Deletion of the G0S2 gene in mice leads to a significant reduction in relative body weight gain, accompanied by notably lower gonadal fat pad weight and elevated serum glycerol levels compared to controls [47]. In our GO enrichment analysis, ENSSSCG00000015617 (G0S2), the target gene of lncRNA LOC100518120, emerged as significantly associated with the positive regulation of cold-induced thermogenesis (GO:0120162), extrinsic apoptotic signaling pathway (GO:0097191), and positive regulation of extrinsic apoptotic signaling pathway (GO:2001238). lncRNA LOC100518120 ranked among the top 18 differentially expressed genes between groups, displaying significant highly expressed in group F. Consequently, we hypothesized that the heightened expression of lncRNA LOC100518120 could positively modulate G0S2, promoting faster weight gain in Jiangquan black pigs of group F compared to group S.

In the target‑prediction assay, we found that LOC100518120 regulates the expression of G0S2 in trans—i.e., after its synthesis the lncRNA diffuses from its site of transcription to act on a distant genomic locus. Trans‑acting lncRNAs can influence target genes through four principal modes: acting as a molecular sponge (competing endogenous RNA), a decoy, a guide, or a scaffold [48]. These four modes can be distinguished from one another based on differences in subcellular localization, interacting partner types, and effects on target gene transcription/translation, allowing us to narrow down the possible mechanism of LOC100518120 step by step. Bio‑informatic analysis failed to predict any miRNA‑binding sites for LOC100518120, allowing us to exclude the sponge and decoy mechanisms, which depend on sequestering miRNAs to relieve repression of their targets [49]. Subcellular fractionation showed that LOC100518120 is predominantly cytoplasmic, while qRT‑PCR demonstrated a significant increase in both G0S2 mRNA and protein levels. Because a scaffold function typically requires the simultaneous binding of multiple protein partners to nucleate a functional complex [50], and because LOC100518120 affects both the mRNA abundance of G0S2, we consider a scaffold role unlikely. Taken together, the data suggest that LOC100518120 may act as a guide, recruiting chromatin‑modifying complexes (e.g., PRC2, LSD1) or transcription factors to the promoter or enhancer of G0S2 to modulate its transcription [51, 52]. The precise mechanism remains to be elucidated by RNA‑immunoprecipitation (RIP) and pull‑down assays.

In the functional validation of the LOC100518120, we treated porcine skeletal muscle satellite cells through overexpression and knockdown. RT-qPCR was used to measure the expression levels of LOC100518120’s target genes and marker genes. The results validated that LOC100518120 can significantly upregulate the expression of the target gene G0S2. It was also found that this gene could promote the proliferation and differentiation of porcine skeletal muscle satellite cells. Flow cytometry analysis showed that LOC100518120 significantly increased the percentage of cells in the S phase and significantly reduced early apoptotic cells. The CCK-8 assay revealed that LOC100518120 could significantly enhance cell viability. After EdU staining, an increase in the number of proliferating cells was observed. In conclusion, LOC100518120 promotes the expression of its target gene G0S2, promotes the proliferation and differentiation of porcine skeletal muscle satellite cells, and inhibits cell apoptosis. This is consistent with the predicted function of LOC100518120 mentioned earlier. It also validates the hypothesis described in other studies regarding the association of G0S2 with porcine skeletal muscle growth and development [53].

Conclusions

In this study, we identified 70 differentially expressed lncRNAs by comparing Jiangquan black pigs with fast (group F) and slow (group S) weight gain rates. GO enrichment analysis revealed that the target genes of these lncRNAs are closely associated with muscle growth, while KEGG pathway analysis highlighted the involvement of G0S2 in multiple muscle development-related pathways. Functional validation demonstrated that the lncRNA LOC100518120 promotes proliferation and differentiation of porcine skeletal muscle satellite cells while suppressing apoptosis, suggesting its critical role in muscle regulation. These findings advance our understanding of the lncRNA-mediated molecular mechanisms underlying muscle growth in Jiangquan black pigs and provide a valuable theoretical basis for future breeding strategies and meat quality enhancement in this breed.

Authors’ contributions

Conceptualization, J.H. and Y.Z.; methodology, H.C., Y.W. and Y.Z.; software, J.H.; investigation, H.C., X.J., Q.S., W.C. and H.T.; resources, W.C. and H.T. data curation, J.H.; writing - original draft, J.H.; writing - reviewing & editing, J.H., H.C. and Y.Z. visualization, H.C., X.J., Q.S., W.C. and H.T.; project administration, Y.W.; funding acquisition, Y.W. and Y.Z.

Funding

This study was supported financially by the National Key R&D Program of China (No. 2021YFD1301203), the Agricultural Animal Breeding Project of Shandong Province (No. 2020LZGC012), Shandong Province pig Industry Technology System Project (No. SDAIT-08-02).

Data availability

All the original data involved in this experiment can be obtained from the author. Sequence data supporting the results of this study have been deposited in the Genome Sequence Archive (GSA) under the primary access code CRA024215.

Declarations

Ethics approval and consent to participate

The animal study was reviewed and approved by the Animal Ethics Committee of Shandong Agricultural University and performed in accordance with the Committee’s guidelines and regulations (No. 2004006).

All pig samples were obtained from a licensed slaughterhouse following the Chinese Regulations on the Administration of Pig Slaughtering (State Council Decree No. 742, 2021) and the national standard GB/T 17236 2019.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.