Genome-wide association study reveals 14 new SNPs and confirms two structural variants highly associated with the horned/polled phenotype in goats

Abstract

Background: There is a long-term interest in investigating the genetic basis of the horned/polled phenotype in domestic goats. Here, we report a genome-wide association study (GWAS) to detect the genetic loci affecting the polled phenotype in goats.

Results: We obtained a total of 13,980,209 biallelic SNPs, using the genotyping-by-sequencing data from 45 Jintang Black (JT) goats, which included 32 female and nine male goats, and four individuals with the polled intersex syndrome (PIS). Using a mixed-model based GWAS, we identified two association signals, which were located at 150,334,857-150,817,260 bp (P = 5.15 × 10^{- 119}) and 128,286,704-131,306,537 bp (P = 2.74 × 10^{- 15}) on chromosome 1. The genotype distributions of the 14 most significantly associated SNPs were completely correlated with horn status in goats, based on the whole-genome sequencing (WGS) data from JT and two other Chinese horned breeds. However, variant annotation suggested that none of the detected SNPs within the associated regions were plausible causal mutations. Via additional read-depth analyses and visual inspections of WGS data, we found a 10.1-kb deletion (CHI1:g. 129424781_129434939del) and a 480-kb duplication (CHI1:150,334,286-150,818,098 bp) encompassing two genes KCNJ15 and ERG in the associated regions of polled and PIS-affected goats. Notably, the 10.1-kb deletion also served as the insertion site for the 480-kb duplication, as validated by PCR and Sanger sequencing. Our WGS genotyping showed that all horned goats were homozygous for the reference alleles without either the structural variants (SVs), whereas the PIS-affected goats were homozygous for both the SVs. We also demonstrated that horned, polled, and PIS-affected individuals among 333 goats from JT and three other Chinese horned breeds can be accurately classified via PCR amplification and agarose gel electrophoresis of two fragments in both SVs.

Conclusion: Our results revealed that two genomic regions on chromosome 1 are major loci affecting the polled phenotypes in goats. We provided a diagnostic PCR to accurately classify horned, polled, and PIS-affected goats, which will enable a reliable genetic test for the early-in-life prediction of horn status in goats.

Keywords: CNV; GWAS; Genetic testing; Goat; Horn; SNP; Whole-genome sequencing.

Fig. 1

Genome-wide association study for the polled phenotype in JT goats based on a linear mixed model. A mixed-model based GWAS for the polled phenotype in 45 JT goats that included 32 female, nine male, and four PIS-affected individuals. Manhattan plot of the association of 13,980,209 biallelic SNPs on autosomes 1–29 with the polled phenotype. The chromosomes are plotted separately by color. The horizontal dashed line indicates the genome-wide Bonferroni-corrected significance level

Fig. 2

Summary of the genotype distributions of the 14 genome-wide most significantly associated SNPs and the loci showing highest LD with them. a The genotype distributions of the 14 most significantly associated SNPs in three domestic goats breeds (i.e., JT, CB, and TC) and Bezoars identified by WGS. b The genotype distributions of one SNP showed the highest LD with the 14 most associated SNPs, and its adjacent two Indels in JT, CB, TC, and Bezoars identified by WGS

Fig. 3

Characterization of two structural variants in the two associated regions. a Genomic coverage of different phenotype groups at the 10.1-kb deletion site (adjusted for the chromosome-wide coverage and calculated in 1-kb sliding windows). b The screen capture of aligned short reads featuring a 10.1 kb deletion between 129,424,781 and 129,434,939 bp on chromosome 1 using IGV. c Genomic coverage of different phenotype groups at the 480-kb duplication site (adjusted for the chromosome-wide coverage and calculated in 1-kb sliding windows). d The screen capture of aligned short reads featuring a 10.1 kb deletion between 150,334,286 and 150,818,098 bp on chromosome 1 using IGV. e A diagram showed the inverse insertion of the 480-kb duplication at the PIS location in the genomes of polled and PIS goats. f Confirmation of the inverse insertion of the 480-kb duplication into the PIS deletion in PIS goats by PCR amplification and Sanger sequencing

Fig. 4

Validation of two structural variants in 333 animals from four Chinese goat breeds by PCR amplification and agarose gel electrophoresis. a Agarose gel picture of the two PCR products (i.e., the 369-bp and 1822-bp fragments) for four horned, four polled, and four PIS-affected goats. Pattern 1 represented that the presence of only the 369-bp fragment; Pattern 2 represented the presence of only the 1822-bp fragment; the pattern 3 represented the presence of both the 369-bp and 1822-bp fragments. b Summary of the validation of two SVs in 333 samples from four Chinese goat breeds (i.e., JT, CB, TC, and NJ). Note: the total 333 samples included the animals with the WGS data, except for Bezoars