Identification of key LncRNAs and mRNAs in fat deposition of Saba pigs based on comparative analysis of transcriptomics

ABSTRACT

Long non-coding RNAs (LncRNAs) are gaining attention in the context of pork production due to their role in regulating fat deposition, a crucial economic trait. This study focused on profiling LncRNAs in back fat deposition in Saba and Yorkshire pigs, which exhibit significant differences in fat accumulation. Using transcriptome sequencing and bioinformatics, we identified 704 differentially expressed mRNAs and 50 LncRNAs. We employed cis-regulatory analysis to explore the function of LncRNAs. Gene Ontology (GO) enrichment revealed pathways and enzyme activity-related terms, indicating that LncRNAs may modulate gene expression to influence biological processes. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis showed enrichment in fat deposition pathways, underscoring the critical role of LncRNAs in this process. Further analysis and construction of LncRNA-mRNA interaction networks revealed 5 potential regulatory networks. Combining the data of differentially expressed LncRNAs and mRNAs, we identified candidate LncRNAs that regulate the expression of ACACA and FASN, which are functionally important genes affecting fat deposition. In summary, this study provides a theoretical foundation and valuable insights into the molecular mechanisms underlying LncRNA-mediated regulation of fat deposition in pigs. These findings are essential for advancing our understanding of porcine fat deposition and enhancing pork production.

KEYWORDS:

• Saba pigs
• transcriptomics
• fat deposition
• LncRNA
• mRNA

Introduction

Long non-coding RNAs (LncRNAs) are defined as non-coding RNAs of at least 200 nucleotides, which is mainly located in the nucleus and involved in the regulation of gene expression, including chromatin reorganization, RNA processing, cell differentiation and development, et al. (Fatica and Bozzoni 2014; Zhao et al. 2014; van Solingen et al. 2018). LncRNAs are playing an increasingly important role in the occurrence and treatment of human diseases, especially in cancer epigenetic and cardiovascular disease regulation and so on (Schonrock et al. 2012; Satpathy and Chang 2015; Zhang, Du, et al. 2019a). With the development of high-throughput sequencing technology, an increasing number of LncRNAs have been discovered and annotated by researchers. However, compared to the medical-related studies and reports on LncRNAs in human tumours and cancers, there are relatively few applications of LncRNAs in pig production. In recent years, the regulatory mechanism of LncRNA in animal adipose tissue has become a hotspot (Knoll et al. 2015; Huang et al. 2017).

Fat deposition is one of the important economic traits in pork production. Subcutaneous fat deposition, known as backfat thickness is an important indicator for evaluating carcass traits reference. Studies have shown that there was a significant negative correlation between backfat thickness and lean meat rate, which was significantly correlated with fat percentage (Yun et al. 2003; Jankowiak et al. 2019), and has an important impact on pig growth efficiency and meat quality (Zambonelli et al. 2016). In addition, the problem of human health caused by obesity has attracted more and more attention. Obesity is related to a variety of metabolic disorders such as insulin resistance (Hajer et al. 2008) and type II diabetes (Willer et al. 2009). Importantly, pigs are anatomically and physiologically similar to humans (Walters et al. 2012), so they can serve as a good medical model for studying diseases of fat metabolism. Studies on differentially expressed mRNAs have identified many important candidate genes related to influencing fat deposition (Lyu et al. 2022; Xu et al. 2022), but the regulatory mechanisms behind the differential expression of molecules are unknown, and this has to some extent limited the use of molecular breeding in the selection and breeding of high-fat pig breeds.

Saba pig (SS) is a native breed prevalent in Yunnan, China, like the species of pigs in other parts of China, it has great pork quality and fat deposition capacity. The average backfat thickness of Saba pig was 3.82 cm, lean meat rate was 45.68% and the backfat thickness of Yorkshire (YS) was 1.91 cm, lean meat rate was 73.80% (Zhang et al. 2016), average backfat thickness is determined by accurately measuring the backfat thickness at three key areas of the pig – the shoulder, the chest loin joint, and the loin sponsor joint – using vernier calipers, and calculating the average of these measurements. The fat deposition capacity between the two pig breeds is quite different, therefore, Saba and Yorkshire pigs provide good resources for studying the molecular mechanism of fat deposition. In this study, RNA-seq were adopted to comparatively analyse the back fat in Saba and Yorkshire pigs to identify key LncRNAs and genes associated with fat deposition. Furthermore, by GO (Gene Ontology), KEGG (Kyoto Encyclopedia of Genes and Genomes pathway), and analysis of the cis-regulatory of LncRNA, we investigated the molecular mechanism of differentially expressed LncRNAs and genes regulating fat deposition. The results might provide useful information for understanding the mechanism of LncRNA regulating pig fat deposition.

Materials and methods

Samples preparation

We selected Saba and Yorkshire pigs with the same management and months of age (3 months), the picture of the Saba pig is shown in Figure 1. After slaughter, we collected the back fat tissue from all the 6 pigs (3 Saba pigs and 3 Yorkshire pigs), Back fat tissue is collected from the back of the pig, specifically the portion located on either side of the spine, and stored at −80°C. The pig population was constructed in National Saba pig breeding farm, Yunnan, China. All the experiment procedures were approved by the Administration of Affairs Concerning Experimental Animals and supported by the ethics committee of Yunnan Agricultural University, Kunming, Yunnan, China.

Figure 1. Saba pig A. Saba pig, boar. B. Saba pig, sow.

Total RNA extraction and sequencing

Total RNA was extracted from each individual by TRIzol (Invitrogen, USA). RNA degradation and contamination was monitored on 1% agarose gels. RNA purity was checked by using the Nano Photometer spectrophotometer (IMPLEN, CA, USA). RNA concentration was measured with using Qubit RNA Assay Kit in Qubit 2.0 Flurometer (Life Technologies, CA, USA). RNA integrity was assessed by using the RNA Nano 6000 Assay Kit of the Agilent Bioanalyzer 2100 system (Agilent Technologies, CA, USA). After the library was constructed, a double-terminal sequencing programme was run on Illumina Hiseq 4000 (PE150) for high-throughput sequencing. The original sequence was obtained and high-quality clean reads were obtained after filtering.

Genome mapping and LncRNA screening

The raw data from the machine was quality controlled and filtered using FASTP 0.23.4 (Chen et al. 2018) software to obtain clean data. Then, the Q20, Q30 and CG content information of the samples were counted to perform preliminary tests on the high-throughput sequencing data. Next, HISAT2 2.2.1 (Kim et al. 2015) software was utilized to align the clean reads with the pig reference genes (Sus scrofa 11.1: https://ftp.ensembl.org/pub/release-110/fasta/sus_scrofa/dna/Sus_scrofa.Sscrofa11.1.dna.toplevel.fa.gz) for alignment. Based on the alignment results, transcripts were assembled and reconstructed for each sample using STRINGTIE 2.2.1 (Pertea et al. 2015) software, and then merged using the Cuffmerge command. Based on the transcriptome assembly results, we took the following steps to screen LncRNAs according to their characteristics: Step 1: Filter out single-exon transcripts with low expression and low confidence in the transcriptome splice results, and select only transcripts with exon number ≥ 2. Step 2: Select transcripts with transcript length > 200 bp. Step 3: Exclude transcripts that overlap with exons with coding capacity in the pig genome annotation file (https://ftp.ensembl.org/pub/release-110/gtf/sus_scrofa/Sus_scrofa.Sscrofa11.1.110.gtf.gz). Step 4: Calculate the expression of each transcript and select transcripts with FPKM ≥ 0.5, the filtration conditions were referred to Kumar et al. (2019). Additionally, we utilized the non-protein coding nature of LncRNAs to process transcript sequences from the initial four screening steps. These were inputted into coding potential prediction software (CPC (Kong et al. 2007), PFAM (Finn et al. 2007), phyloCSF (Lin et al. 2011), CNCI (Sun et al. 2013)), isolating sequences lacking coding potential. The intersecting results from these four software tools determined our final selection. Through these 5 steps of screening, we obtained transcripts that will be used as candidate LncRNAs in this study for subsequent analysis. We also analysed the structure as well as the conserved sequences of the screened LncRNAs and mRNAs, and the comparison of the length of transcripts, the number of exons, the length of ORFs, and the sequence conservatism can be used to obtain the difference between them, and to detect whether the candidate LncRNAs conform to the general characteristics.

Quantification and differential expression analysis of LnRNA and mRNA expression

The screened LncRNAs and mRNAs were quantified using STRINGTIE 2.2.1 software, and the correlation analysis between the samples was performed after normalizing the gene expression using the FPKM method. BALLGOWN 2.32.0 (Pertea et al. 2015) software was used to perform differential expression analysis for LncRNAs and mRNAs separately, with log₂|FoldChange| ≥ 1 and p value < 0.05 were used as thresholds to screen differentially expressed LncRNAs and mRNAs, respectively.

LncRNA target gene prediction and GO and KEGG enrichment analysis of target genes

Cis-regulation of LncRNAs refers to the possibility that LncRNAs may have an effect on the expression of genes coding for them in their neighbourhood as a way to regulate the expression of these genes. We chose to look for potentially regulated target genes centred on LncRNAs in a 100 kb chromosomal region upstream and downstream of them, the selection of genes within the 100 kb region upstream and downstream of LncRNAs as target genes for LncRNA cis-regulation was informed by Kumar et al. (2019). The target genes were analysed by GO (Gene Ontology) and KEGG (Kyoto Encyclopedia of Genes and Genomes) enrichment to gain insights into the possible functions of the LncRNA.

Construction of the LnRNA-mRNA regulatory network

Based on the obtained LncRNA-mRNA-targeted regulatory relationships, network maps were drawn using Gephi 0.10.1 software to find the key regulatory relationships associated with the target traits.

qRT-PCR verification

In order to verify the accuracy of high-throughput sequencing results, real-time quantitative PCR was used to verify the expression. We used the Tengen SuperReal PreMix Plus kit for qRT-pcr. 2 mRNAs and 2 LncRNAs were randomly selected. The real-time PCR primer information is shown in Table 1. The reaction system consisted of 2.5 μL of cDNA, 12.5 μL 2 qPCR mix, 1 μL of forward primer, 1 μL of reverse primer and 8.0 μL of nuclease free water. The PCR reaction conditions were as follow: pre-denaturation at 95°C for 10 min, followed by 40 cycles, denaturation at 95°C for 15s; renaturation at 60°C and extension for 60s. The expression of LncRNAs and mRNAs between samples was calculated by 2-△△Ct method.

Table 1. Real-time quantitative PCR primer information.

Genes	Primer Sequences (5′∼3′)	Product size (bp)	Annealing temperature (°C)
ACTIN	F:TGGTTCTGGGCTCTGTAAGGC R:TGATGCCGTGTTCTATTGGGTA	191	60
ACACA	F:AAGCAAGGTCGAGACCGAAAG R:GTATTCAGCGTTGGCTTTCAGG	249	60
ACAT1	F:CTATTGGAGGAGGCAGCCTGTAC R:TTCGCTCCACATATCTTATTTCCTG	192	60
LNC_002930	F:ATAAGGCTTCAACATTCG R:CACCCACTTCTTCTCACC	175	50
LNC_000258	F:ATACCGATTGAAGGCATTGGAG R:CAAACCACCGCATTAACTTGGA	96	60

Results

Quality control and genome mapping

In the study, 6 LncRNA libraries (SS1, SS2 and SS3; YS1, YS2 and YS3) using backfat with SS and YS pigs were constructed. Through LncRNA sequencing analysis, read trimming and subsequent filtering steps, about 89480450∼131401398 clean reads were obtained for each sample. Approximately 92.52–93.68% of the reads were successfully aligned to the Sus scrofa reference genome (Table 2).

Table 2. Summary of raw reads after quality control and mapping to the reference genome.

Sample name	Raw reads	Clean reads	Q20 (%)	Q30 (%)	GC content (%)	Total mapped
SS1	91878298	89480450	97.05	92.63	53.81	82930871 (92.68%)
SS2	95189340	92553500	97.16	92.84	55.19	86229582 (93.17%)
SS3	135254658	131401398	96.55	91.50	56.51	121576377 (92.52%)
YS1	107851692	104921806	97.05	92.65	56.61	98292410 (93.68%)
YS2	102865486	100116454	97.13	92.80	54.41	93429440 (93.32%)
YS3	105123322	101968640	96.65	91.82	53.45	95449636 (93.61%)

Identification and characterization of LncRNAs

Transcripts were spliced and assembled from the sequencing data of six samples based on STRINGTIE software. Firstly, based on the characteristics of LncRNAs, we designed four screening steps. Our analysis successfully identified 15,184 LncRNAs, as illustrated in Figure 2(A).

Figure 2. Overview of LncRNA Screening Process (A) Initial screening of 4,424,441 transcripts based on LncRNA structural characteristics, followed by further refinement focusing on non-protein coding functions of LncRNAs. (B) Out of 15,184 candidate LncRNAs identified via structural analysis, their coding potential was assessed using four prediction tools. The final candidate LncRNAs were selected based on their non-coding predictions across all tools.

Based on the final identified LncRNAs and mRNAs obtained by sequencing, we performed the comparison of transcript length, exon number, and open reading frames in order to verify whether the predicted LncRNAs possessed general characteristics by comparing the differences between them. In this study, 12,290 Novel LncRNAs, 591 known LncRNAs, and 45,788 mRNAs were compared in terms of exon, transcript length, and ORF length. In terms of exon count, novel LncRNAs showed an average of 4.2 exons, known LncRNAs had 2.8 exons, and mRNAs possessed 11.6 exons (Figure 3(A)). Regarding open reading frame (ORF) length, novel LncRNAs, known LncRNAs, and mRNAs exhibited lengths of 186, 118, and 594 nt, respectively (Figure 3(B)). When considering transcript length, novel LncRNAs had a mean length of 1,679 nt, known LncRNAs were on average 1,551 nt long, and mRNAs had an average length of 3,296 nt (Figure 3(C)). We conducted a conservative scoring analysis of mRNAs and LncRNAs using the phastCons software, resulting in a cumulative distribution map for both LncRNAs and mRNAs. Remarkably, the sequence conservation of LncRNAs was observed to be lower in comparison to that of mRNAs, as depicted in Figure 3(D).

Figure 3. Comparative characterization of LncRNAs and protein-coding genes (A) LncRNA and mRNA exon number density plots (B) LncRNA and mRNA ORF length density plots (C) Comparison of LncRNA and mRNA lengths (D) Conservative analysis of LncRNA and mRNA.

Sample correlation analysis

Before performing differential expression analysis, we first performed sample correlation analysis on the FPKM-corrected gene expression matrix. The results of the analysis showed (Figure 4) that the intra-group correlation coefficients were higher than the inter-group correlation coefficients in the comparisons between samples. This indicates better reproducibility between samples and meets the requirements for subsequent differential expression analysis.

Figure 4. Expression correlation plot between samples.

Differential expression analysis

We performed differential analysis of different types of transcripts (LncRNAs and mRNAs) using Ballgown software and used the following thresholds to screen for differentially expressed LncRNAs and mRNAs: log₂|FoldChange| > 1 and p value < 0.05. Among LncRNAs, we screened 50 differentially expressed LncRNAs, of which 22 were up-regulated and 27 were down-regulated. And in mRNA, we identified 704 differentially expressed genes, of which 350 were up-regulated and 354 down-regulated (Figure 5).

Figure 5. Volcano plot of differentially expressed LncRNAs and mRNAs (A) Volcano plot of differentially expressed LncRNAs (B) Volcano plot of differentially expressed mRNAs.

Functional enrichment analysis of LncRNA adjacent genes

The expression of LncRNAs may have an effect on the transcription of genes in their nearby areas. Genes located within a 100 kb radius upstream and downstream of differentially expressed LncRNAs were considered potential targets regulated by LncRNAs, and the biological processes and functional pathways involved in the cis-targets of LncRNAs were explored by functional enrichment analysis of these target genes. The results showed that there were a total of 229 functional genes encoding proteins near the 50 differentially expressed LncRNAs. We performed GO and KEGG enrichment analyses on these genes, respectively. GO enrichment analysis showed (Figure 6) that the target genes could be enriched in GO entries that regulate and affect the biological functions, such as protein kinase activity, chitinase activity, potassium ion leakage channel activity, and lipid phosphatase activity, which suggests that LncRNAs play an important role in regulating the process of gene expression. And the KEGG enrichment results (Figure 7) showed that this part of functional genes mainly involved in pathways related to the production and deposition of fat, such as fatty acid biosynthesis, fat digestion and absorption, AMPK signalling pathway and insulin signalling pathway. This result also reveals the differences in fat deposition traits between the two species of pigs, and also suggests that post-transcriptional regulation is an important process affecting fat deposition in pigs.

Figure 6. GO enrichment analysis of cis-regulated target genes of differentially expressed LncRNAs.

Figure 7. KEGG enrichment analysis of cis-regulated target genes of differentially expressed LncRNAs.

LncRNA-mRNA interaction network construction

Based on the KEGG enrichment analysis results of the pathways associated with fat deposition traits, the six functional genes that may have an impact on fat deposition were identified, which were ACACA, FASN, CAB39L, RPS6KB2, PPP1CA, and PLPP2. In order to deeply investigate the potential regulatory mechanisms between the molecules, according to the cis-regulatory mechanism of LncRNAs, i.e. LncRNAs may regulate the expression of their neighbouring genes, we constructed a LncRNA-mRNA interaction network (Figure 8). In this network, the FASN is regulated by 6 LncRNAs, and the CAB39L interacts with 7 LncRNAs. the ACACA may be regulated by 2 LncRNAs, and the PLPP2 is regulated by 1 LncRNA. Notably, 5 LncRNAs may be involved in regulating the expression of both PPP1CA and RPS6KB2. On the basis of the differentially expressed LncRNAs and mRNAs identified in previous analyses, we not only established a LncRNA-mRNA cis-regulatory network map (Figure 8), but also successfully constructed a cis-regulatory network map involving differentially expressed LncRNAs with differentially expressed mRNAs (Figure 9). We found that the FASN and the ACACA were differentially expressed in the subdorsal fat of two different pig breeds, suggesting that the LncRNA-mRNA interaction network regulating the ACACA and FASN deserves further investigation, the sequence information for LNC_002572 and LNC_002390 is shown in Supplemental Table 1.

Figure 8. Network of regulatory relationships between LncRNAs and their cis-regulated target genes.

Figure 9. Cis-regulatory network map of differentially expressed LncRNAs and mRNAs.

qPCR verification

First, we determined the relative expressions of 2 mRNASs and 2 LncRNAs in the back fat tissues of Saba and Yorkshire pigs by Real Time Quantitative PCR (RT-qpcr) technology, and the specific results are shown in the Figure 10. In the back fat tissue, the expression levels of all 4 genes, ACAT1, ACACA, LNC_002930 and LNC_000258, showed significant differences. In addition, we also calculated the multiplicity of differences (expressed as log₂(FoldChange)) of these 4 genes in the RT-qpcr assay based on the relative expression results, which showed that the gene expression trends in the RT-qpcr results were consistent with those in the transcriptome sequencing results, which further verified the reliability of the transcriptome sequencing results.

Figure 10. RT-qpcr validation of transcriptome sequencing results. A. Relative expression of ACACA, ACAT1, LNC_002930 and LNC_000258 in back fat of Saba(SS) pigs vs. Yorkshire pigs(YS). B. Validation of expression trends of ACACA, ACAT1, LNC_002390 and LNC_000258 in back fat of Saba and Yorkshire pigs.

Discussion

Fat deposition can have an effect on both meat quality and appearance in livestock. For example, increased intramuscular fat content can improve meat texture and flavour (Miller 2020). However, excessive fat deposition not only adversely affects meat quality, but may also negatively impact the health of the individual. Therefore, exploring the regulatory mechanisms of fat deposition in livestock is of great significance in guiding the improvement of meat quality and the selection of breeds. Researchers have initiated numerous studies to investigate the molecular mechanisms of fat deposition in pigs. Comparative transcriptome analysis of liver, longest back muscle and back subcutaneous adipose tissue from pig breeds with different fat deposition ability or from individuals with significant differences in fat deposition ability within the same breed has identified a number of key candidate genes that affect fat deposition in pigs (Wang et al. 2015; Xu et al. 2018; Zhang et al. 2021; Zhang et al. 2022). The expression of coding genes plays a crucial role in life activities, and at the same time, the molecular mechanisms that regulate the differential expression of mRNAs have attracted much attention. Analysing the regulatory networks behind these differential expressions is of great significance for the in-depth study of life activities.

Long non-coding RNAs (LncRNAs) have been shown to play an important role in regulating the expression of mRNAs. LncRNAs play key roles in several areas of life activities, such as in fat deposition (Gong et al. 2021; Feng et al. 2023; Yue et al. 2023). Therefore, in this study, not only mRNAs but also LncRNAs were sequenced. We screened for differentially expressed LncRNAs and performed GO and KEGG enrichment analysis on the target genes of significantly differentially expressed LncRNAs. In the enrichment analysis results, we found that the GO enrichment results indicated that LncRNAs could be involved in the regulatory processes of multiple life activities. And the KEGG enrichment results showed significant enrichment of pathways related to fat deposition and metabolism, which indicated that the two pig breeds selected for this study differed significantly in fat deposition, and thus could be used as an ideal model for the study of fat deposition in pigs. Also, these results suggest that LncRNA has an important effect on fat deposition.

According to the results of KEGG enrichment analysis of genes enriched in the pathway related to fat deposition (fatty acid biosynthesis, fat digestion and absorption, AMPK signalling pathway and insulin signalling pathway), we screened six candidate genes, ACACA, FASN, CAB39L, PLPP2, RPS6KB2 and PPP1CA, which affect fat deposition in Saba pigs. With the help of molecular biological experiments and histological techniques researchers found and concluded that two genes, ACACA and FASN, play important roles in the process of fat deposition in pigs (Eusebi et al. 2017; Zappaterra et al. 2019; Zhang, Zhang, et al. 2019b; Wang et al. 2021). The ACACA is responsible for encoding the acetyl coenzyme A carboxylase (ACC) alpha-type enzyme, which plays a central regulatory role in the ab initio synthesis of fatty acids. By regulating its expression level, ACC alpha-type enzyme is able to influence the rate of fatty acid synthesis, which in turn affects the deposition of fat in the organism (Harris et al. 2011). Fatty acid synthases (FASN) encoded by the FASN are multi-functional proteins whose main role is to convert acetyl coenzyme A and malonyl coenzyme A to palmitic acid, an important class of long-chain saturated fatty acids, with the help of NADPH (Jensen-Urstad and Semenkovich 2012). This process is not only an essential fatty acid synthesis pathway for fat deposition in animals, but it has also been suggested that FASN is one of the key enzymes regulating the development of adipose tissue (Desert et al. 2018). CAB39L is a gene enriched in the AMPK pathway, which is an important signalling pathway in organisms that can sense the energy state of cells and then regulate a variety of physiological processes, and fat metabolism is also regulated by the AMPK pathway (Ke et al. 2018; Wang et al. 2018). Zhang et al. identified CAB39L as an important candidate gene affecting growth and carcass traits in chickens, and we hypothesized that the CAB39L gene affects carcass traits in pigs by influencing fat synthesis and metabolism in Saba pigs. The PLPP2 gene is a member of the phospholipase family, phosphatidic acid is an important precursor substance for fat synthesis, and there is a close relationship between phosphatidic acid metabolism and fat deposition (Zhou et al. 2021). The two genes, PPP1CA and RPS6KB2, are genes enriched in the insulin signalling pathway, which plays an important role in the regulation of fat metabolism and energy homeostasis, and abnormalities in the signalling pathway can lead to health problems such as obesity (Samuel and Shulman 2012; Boucher et al. 2014).

We constructed an LncRNA-mRNA interaction network according to the 6 functional genes screened to affect fat deposition in order to investigate their interrelationships under the cis-regulatory effect of LncRNAs. Among these six genes, we found that five different regulatory networks may exist. Among them, FASN, CAB39L, PLPP2 and ACACA genes were located in different networks, which implied that their expression was regulated by independent LncRNAs. In contrast, the two genes, RPS6KB2 and PPP1CA, are located in the same network, suggesting that they may be affected by the same LncRNA expression to co-regulate the expression of these two genes. Notably, both genes were enriched in the KEGG enrichment results for the insulin signalling pathway and belonged to the same pathway. Through the differential mRNA expression profile in this study, we also observed that ACACA and FASN were significantly differentially expressed between the two pig breeds, and thus we believe that the LncRNAs regulating these two genes are worthy of further attention and study.

Conclusion

In this study, the back fat tissues of Yorkshire and Saba pigs with significantly different fat deposition capacities were selected for transcriptome sequencing. By identifying differentially expressed LncRNAs and analysing their cis-regulated target genes by GO and KEGG enrichment, in GO enrichment we found that target genes cis-regulated by LncRNAs were enriched for certain pathways and enzyme activities, which suggests that LncRNAs may influence life activities in organisms through regulatory actions. In addition, KEGG enrichment results showed significant enrichment of pathways associated with fat deposition, emphasizing the important role of LncRNAs in the fat deposition process. By constructing the LncRNA-mRNA interaction network, we identified 5 potential regulatory networks, and combined with the results of differentially expressed mRNAs in back fat, we concluded that the LncRNAs regulating the differential expression of ACACA and FASN genes are worthy of further in-depth study.

Authors’ contributions

SL conceived and designed the study. XD and XL analysed the data and wrote the first draft. XW, QC and ML collected the samples and acquired the data. DY and GL were responsible for visualizing the data. All authors reviewed and approved the final manuscript.

Ethics approval and consent to participate

The research protocol was in accordance with the Declaration of Helsinki. Approved by the Ethical Review Committee of Yunnan Agricultural University. All participants provided written informed consent.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Yunnan major scientific and technological project (202202AE090005), Yunnan major scientific and technological project (202302AE090015), Leading Talents of Industrial Technology of Yunnan Province (YNWR-CYJS-2018-056), Yunnan Provincial University Science and Technology Innovation Team (Yunnan Provincial University Science and Technology Innovation Team for the Discovery and Utilization of Resistance Genes in Livestock and Poultry in the Highland Mountainous Areas (2020TD05)).