Construction of a co-expression network affecting intramuscular fat content and meat color redness based on transcriptome analysis

Abstract

Introduction: Intramuscular fat content (IFC) and meat color are vital indicators of pork quality. Methods: A significant positive correlation between IFC and redness of meat color (CIE a* value) indicates that these two traits are likely to be regulated by shared molecular pathways.To identify candidate genes, hub genes, and signaling pathways that regulate these two traits, we measured the IFC and CIE a* value in 147 hybrid pigs, and selected individuls with extreme phenotypes for transcriptome analysis. Results: The results revealed 485 and 394 overlapping differentially expressed genes (DEGs), using the DESeq2, limma, and edgeR packages, affecting the IFC and CIE a* value, respectively. Weighted gene co-expression network analysis (WGCNA) identified four modules significantly correlated with the IFC and CIE a* value. Moreover, we integrated functional enrichment analysis results based on DEGs, GSEA, and WGCNA conditions to identify candidate genes, and identified 47 and 53 candidate genes affecting the IFC and CIE a* value, respectively. The protein protein interaction (PPI) network analysis of candidate genes showed that 5 and 13 hub genes affect the IFC and CIE a* value, respectively. These genes mainly participate in various pathways related to lipid metabolism and redox reactions. Notably, four crucial hub genes (MYC, SOX9, CEBPB, and PPAGRC1A) were shared for these two traits. Discussion and conclusion: After functional annotation of these four hub genes, we hypothesized that the SOX9/CEBPB/PPARGC1A axis could co-regulate lipid metabolism and the myoglobin redox response. Further research on these hub genes, especially the SOX9/CEBPB/PPARGC1A axis, will help to understand the molecular mechanism of the co-regulation of the IFC and CIE a* value, which will provide a theoretical basis for improving pork quality.

Keywords: RNA-seq; functional enrichment analysis; hub gene; intramuscular fat content; meat color redness.

FIGURE 1

Statistical analysis of intramuscular fat content (IFC) and meat color redness (CIE a*) values. (A) Phenotypic values of the IFC and CIE a* value. (B) Correlation analysis of the IFC and CIE a* value.

FIGURE 2

IFC and CIE CIE a* value comparison between the high and low groups. (A) Representative plots of latissimus dorsi (LD) tissue and H&E staining of shared samples from the low IFC group and low CIE a* group, scale bar = 100 μm. (B) Representative plots of LD tissue and H&E staining of shared samples from the high IFC group and high CIE a* group, scale bar = 100 μm. (C–E) Comparasion of the IFC and CIE a* value in different groups. The H_group represents the sample combination of H_IFC and H_a*, n = 10; The L_group represents the sample combination of the L_IFC and L_a* value group, n = 9. Error bars represent the standard deviation (SD), where yellow bars represent the IFC and blue bars represent the CIE a* value. *p < 0.05, **p < 0.01, ***p < 0.001, two-tailed Student’s t-test.

FIGURE 3

Identification of differentially expressed genes (DEGs). (A) Venn diagram of DEGs identified using the DESeq2, limma, and edgeR packages for the IFC. (B) Heatmap of overlapping DEGs between the H_IFC and L_IFC groups. (C) Venn diagram of DEGs identified using the DESeq2, limma, and edgeR packages for the CIE a* value. (D) Heatmap of overlapping DEGs between the H_a* and L_a* groups.

FIGURE 4

GO and KEGG enrichment analysis of overlapping DEGs. (A) Top five GO terms of overlapping DEGs for the IFC in the BP, CC, and MF categories. (B) Significantly enriched KEGG pathways of overlapping DEGs for IFC. (C) Top five GO terms of overlapping DEGs for CIE a* value in the BP, CC, and MF categories. (D) Significantly enriched KEGG pathways of overlapping DEGs for the CIE a* value. The size of the dot represents the number of overlapping DEGs enriched to this GO term or pathway. The colour of the dot represents the significance of the enrichment, where a redder dot indicates greater significance.

FIGURE 5

Weighted gene co-expression network analysis (WGCNA). (A) The gene dendrogram was obtained by clustering the dissimilarity based on consensus Topological Overlap with the corresponding module colors indicated by the color row. (B) Matrix with module trait relationships (MTRs) and corresponding p values between the detected modules on the y-axis and traits (IFC and CIE a* value) on the x-axis, where blue represents a negative correlation, red represents a positive correlation, and white represents no correlation. (C) The number of genes contained in each module.

FIGURE 6

Venn network diagrams of enrichment analysis. (A) Venn network diagram of significantly enriched GO terms under three conditions for IFC. (B) Venn network diagram of significantly enriched KEGG pathways under three conditions for IFC. (C) Venn network diagram of significantly enriched GO terms under three conditions for the CIE a* value. (D) Venn network diagram of significantly enriched KEGG pathways under three conditions for CIE a* value.

FIGURE 7

Protein protein interaction (PPI) network for the candidate genes affecting the IFC and CIE a* value. Edges (gray lines) between nodes indicate the interaction of genes in the network. Green circles represent candidate DEGs for IFC, and blue circles represent candidate DEGs for CIE a*. Brown hexagos represent overlapping candidate DEGs for IFC and CIE a*. Red circles and V-shapes represent DEGs with connectivity greater than 10 and are considered hub genes. Hub genes shared by the IFC and CIE a* value are represented by V-shapes.