Whole-genome sequence analysis reveals selection signatures for important economic traits in Xiang pigs

Abstract

Xiang pig (XP) is one of the best-known indigenous pig breeds in China, which is characterized by its small body size, strong disease resistance, high adaptability, favorite meat quality, small litter sizes, and early sexual maturity. However, the genomic evidence that links these unique traits of XP is still poorly understood. To identify the genomic signatures of selection in XP, we performed whole-genome resequencing on 25 unrelated individual XPs. We obtained 876.70 Gb of raw data from the genomic libraries. The LD analysis showed that the lowest level of linkage disequilibrium was observed in Xiang pig. Comparative genomic analysis between XPs and other breeds including Tibetan, Meishan, Duroc and Landrace revealed 3062, 1228, 907 and 1519 selected regions, respectively. The genes identified in selected regions of XPs were associated with growth and development processes (IGF1R, PROP1, TBX19, STAC3, RLF, SELENOM, MSTN), immunity and disease resistance (ZCCHC2, SERPINB2, ADGRE5, CYP7B1, STAT6, IL2, CD80, RHBDD3, PIK3IP1), environmental adaptation (NR2E1, SERPINB8, SERPINB10, SLC26A7, MYO1A, SDR9C7, UVSSA, EXPH5, VEGFC, PDE1A), reproduction (CCNB2, TRPM6, EYA3, CYP7B1, LIMK2, RSPO1, ADAM32, SPAG16), meat quality traits (DECR1, EWSR1), and early sexual maturity (TAC3). Through the absolute allele frequency difference (ΔAF) analysis, we explored two population-specific missense mutations occurred in NR6A1 and LTBP2 genes, which well explained that the vertebrae numbers of Xiang pigs were less than that of the European pig breeds. Our results indicated that Xiang pigs were less affected by artificial selection than the European and Meishan pig breeds. The selected candidate genes were mainly involved in growth and development, disease resistance, reproduction, meat quality, and early sexual maturity. This study provided a list of functional candidate genes, as well as a number of genetic variants, which would provide insight into the molecular basis for the unique traits of Xiang pig.

Figure 1

The distribution and annotation of genomic variants in XP. (a) Circos plot of distribution of variants density after filtration. The density of variants was calculated in each 100 kb step size. The circles from inside to outside display INDEL and SNP density per window, respectively. (b) SNP numbers in genomic annotation according to ANNOVAR. (c) The pie plot shows annotated SNPs at exonic regions.

Figure 2

Population genetic analysis. (a) Plots of principal components 1 and 2 for the 125 individuals. (b) Neighbor-joining tree constructed from SNPs data among five pig populations. (c) Genetic structure analysis of samples using Admixture, with changing ancestral populations from K = 2 to K = 5.

Figure 3

Decay of linkage disequilibrium in XP, TT, MS, DU and LW breeds.

Figure 4

Manhattan plots of genome-wide of Fst and θπ values across all 18 autosomes identified in XP population from four different comparisons. (a) TT vs XP. (b) MS vs XP. (c) DU vs XP. (d) LW vs XP. Dashed lines represented the threshod of top 5% Fst and θπ values.

Figure 5

Venn diagrams represented the numbers of unique and shared (a) positively selected regions and (b) the candidate genes between four comparisons.

Figure 6

Characterization of selection signals around ZCCHC2 gene locus in XPs. (a) The Fst and θπ values around the ZCCHC2 locus. (b) Haplotype plots spanning ZCCHC2 gene among the five pig populations. The allele consistent with reference genome was indicated in lightyellow and derived allele in red color.

Figure 7

Haplotype plots spanning SDR9C7 and RDH16 genes among the five pig populations. The allele consistent with reference genome was indicated in lightyellow and derived allele in red.

Figure 8

Putative functional variant (p.R84C) in TAC3 gene. (a) Multispecies alignment of the protein sequences around the variant. The dots in the alignment indicated the amino acids which were identical with those in XP. (b) Distribution of the allele frequency of the variant in five pig populations.

Figure 9

Characterization of selection signals around TBX19 gene locus in XPs, and three putative functional variants (p.N105T, p.M158V, p.T189A). (a) The Fst and θπ values around the TBX19 locus. (b) Cross-species alignment of the protein sequences around three functional variants (p.N105T, p.M158V, p.T189A) in TBX19 gene. The dots in the alignment indicated the amino acids which were identical with those in XPs, and the dashes indicated the missing data. (c) Distribution of the allele frequency of the three variants in three populations (XP, DU, and LW).

Figure 10

The significant KEGG pathway enrichment of the candidate genes under selection in XPs. (a) TT vs XP, (b) MS vs XP (c) DU vs XP, and (d) LW vs XP. Note: KEGG pathways were analyzed via KEGG database (https://www.kegg.jp/kegg/kegg1.html).