Identification of single nucleotide polymorphisms (SNPs) potentially associated with residual feed intake in Qinchuan beef cattle by hypothalamus and duodenum RNA-Seq data

Abstract

The regulation of residual feed intake (RFI) in beef cattle involves brain-gut mechanisms due to the interaction between neural signals in the brain and hunger or satiety in the gut. RNA-Seq data contain an extensive resource of untapped SNPs. Therefore, hypothalamic and duodenal tissues from ten extreme RFI individuals were collected, and transcriptome sequenced in this study. All the alignment data were combined according to RFI, and the SNPs in the same group were identified. A total of 270,410 SNPs were found in the high RFI group, and 255,120 SNPs were found in the low RFI group. Most SNPs were detected in the intronic region, followed by the intergenic region, and the exon region accounts for 1.11% and 1.38% in the high and low RFI groups, respectively. Prediction of high-impact SNPs and annotation of the genes in which they are located yielded 83 and 97 genes in the high-RFI and low-RFI groups, respectively. GO enrichment analysis of these genes revealed multiple NADH/NADPH-related pathways, with ND4, ND5, and ND6 significantly enriched as core subunits of NADH dehydrogenase (complex I), and is closely related to mitochondrial function. KEGG enrichment analysis of ND4, ND5, and ND6 genes was enriched in the thermogenic pathway. Multiple genes, such as ATP1A2, SLC9A4, and PLA2G5, were reported to be associated with RFI energy metabolism in the concurrent enrichment analysis. Protein-protein interaction analysis identified multiple potential candidate genes related to energy metabolism that were hypothesized to be potentially associated with the RFI phenotype. The results of this study will help to increase our understanding of identifying SNPs with significant genetic effects and their potential biological functions.

Keywords: Beef cattle; Duodenum; Hypothalamus; RNA-Seq; Single nucleotide polymorphisms.

Figure 1. Sample collection and bioinformatics analysis.

Hypothalamic and duodenal tissues from extreme RFI individuals were collected from 30 Qinchuan cows. After transcriptome sequencing, quality control, alignment, and deduplication, the alignment files were merged. The data from the two tissues were merged according to the RFI group, resulting in two alignment files. The merger greatly increased the depth (DP) of the reads, and the average reads DP of the two groups was basically consistent. Finally, the BCFtools software was used to identify SNPs in the merged data.

Figure 2. Transcript expression (TPM) analysis of each sample.

(A) Transcript expression density in each sample. (B) Transcript expression in each sample.

Figure 3. High RFI group and low RFI group combined data SNPs type statistics.

(A) Statistics on SNPs types in the high RFI group. (B) Statistics on SNPs types in the low RFI group.

Figure 4. SNPs distribution statistics on chromosomes.

(A) Statistics on the number of SNPs on different chromosomes. (B) SNPs number and length ratio statistics on different chromosomes. (C) Distribution of SNPs on chromosome locations in the high RFI group. (D) Distribution of SNPs on chromosome locations in the low RFI group. (E) Percentage of the genome comprising each type of feature (top) and the proportion of SNPs detected by HRFI group-specific SNPs (middle) and LRFI group-specific SNPs (bottom) across these genomic features.

Figure 5. Influence prediction and amino acid change.

(A) High RFI group specific SNPs influence prediction statistics. (B) Low RFI group specific SNPs influence prediction statistics. (C) The amino acid changes caused by SNPs in the high RFI group were reference amino acids horizontally and altered amino acids vertically. Red background colors indicate that more changes happened (heat-map). (D) The amino acid changes caused by SNPs in the low RFI group.

Figure 6. Gene function annotation of high-impact SNP loci.

(A) Gene GO enrichment analysis (p < 0.05) of HRFI group-specific high-impact SNP sites, and select the top 10 for each term type based on p-value. (C: cellular component; F: molecular function; P: biological process). (B) Gene GO enrichment analysis (p < 0.05) of LRFI group-specific high-impact SNP sites, and select the top 10 for each term type based on p value. (C) Gene KEGG enrichment analysis (p < 0.05) of HRFI group-specific high-impact SNP sites, and select the top 20 based on p value. (D) Gene KEGG enrichment analysis (p < 0.05) of LRFI group-specific high-impact SNP sites, and select the top 20 based on p value.

Figure 7. Protein-protein interaction analysis of high-impact SNP loci.

(A) Protein-protein interaction analysis of high-impact SNP sites in HRFI group, using string database, composed of 29 nodes and 23 edges. (B) Protein-protein interaction analysis of high-impact SNP sites in LRFI group, using string database, composed of 42 nodes and 41 edges.