Genome-Wide Detection of Copy Number Variations and Their Association With Distinct Phenotypes in the World's Sheep

Abstract

Copy number variations (CNVs) are a major source of structural variation in mammalian genomes. Here, we characterized the genome-wide CNV in 2059 sheep from 67 populations all over the world using the Ovine Infinium HD (600K) SNP BeadChip. We tested their associations with distinct phenotypic traits by conducting multiple independent genome-wide tests. In total, we detected 7547 unique CNVs and 18,152 CNV events in 1217 non-redundant CNV regions (CNVRs), covering 245 Mb (∼10%) of the whole sheep genome. We identified seven CNVRs with frequencies correlating to geographical origins and 107 CNVRs overlapping 53 known quantitative trait loci (QTLs). Gene ontology and pathway enrichment analyses of CNV-overlapping genes revealed their common involvement in energy metabolism, endocrine regulation, nervous system development, cell proliferation, immune, and reproduction. For the phenotypic traits, we detected significantly associated (adjusted P < 0.05) CNVRs harboring functional candidate genes, such as SBNO2 for polycerate; PPP1R11 and GABBR1 for tail weight; AKT1 for supernumerary nipple; CSRP1, WNT7B, HMX1, and FGFR3 for ear size; and NOS3 and FILIP1 in Wadi sheep; SNRPD3, KHDRBS2, and SDCCAG3 in Hu sheep; NOS3, BMP1, and SLC19A1 in Icelandic; CDK2 in Finnsheep; MICA in Romanov; and REEP4 in Texel sheep for litter size. These CNVs and associated genes are important markers for molecular breeding of sheep and other livestock species.

Keywords: CNVs; GWAS; genetic adaptation; selection; sheep.

FIGURE 1

(A) Geographic origin of the domestic sheep populations studied. The domestic sheep populations are color coded according to geographical origin. (B) Neighbor-joining clustering of the sheep populations based on the Reynolds’ distances. (C) Principal component analysis (PCA) of the 67 domestic sheep populations. The circle of yellow, blue, green, and red represent Eastern-Central Asian, Western Asian, African, and European populations, respectively.

FIGURE 2

(A) Genomic distribution and frequencies of copy number variation regions (CNVRs) on autosomes in the four genetic groups of domestic sheep populations. The tracks from outside to inside (A–E) are CNVRs shared across all the five groups: African, Eastern-Central Asian, European, and Western Asian populations. (B) CNVR frequency heatmap based on hierarchical clustering analysis for the top 5% of CNVRs (CNV events ≥ 2 CNV) in the four genetic groups of domestic sheep populations: African, Eastern-Central Asian, European, and Western Asian populations. The value is the number of CNVRs per individual (total of CNVRs/sample size based on population size in each CNVR).

FIGURE 3

Validation of copy number variations (CNVs) (i.e., five duplications and eight deletions) by quantitative PCR (qPCR). The x- and y-axes represent sample ID and relative CNVs expression level [2 × 2^{–(ΔΔCT±SD)}, n = 3].

FIGURE 4

(A) Venn diagram of the number of copy number variations (CNVs) shared by geographic groups of domestic sheep, including African, Eastern-Central Asian, European, and Western Asian populations. (B) Histogram of genomic coverage by CNV regions (CNVRs) shared by two geographic groups of domestic sheep populations.

FIGURE 5

Manhattan plots of genome-wide association study (GWAS) of large litter size for Wadi, Hu, Finnsheep, Icelandic, and Romanov and smaller litter size for Texel. The threshold of adjusted P < 0.05 is indicated by a dotted line.