Discovery of candidate genes for muscle traits based on GWAS supported by eQTL-analysis

Abstract

Biochemical and biophysical processes that take place in muscle under relaxed and stressed conditions depend on the abundance and activity of gene products of metabolic and structural pathways. In livestock at post-mortem, these muscle properties determine aspects of meat quality and are measurable. The conversion of muscle to meat mimics pathological processes associated with muscle ischemia, injury or damage in humans and it is an economic factor in pork production. Linkage, association, and expression analyses independently contributed to the identification of trait-associated molecular pathways and genes. We aim at providing multiple evidences for the role of specific genes in meat quality by integrating a genome-wide association study (GWAS) for meat quality traits and the detection of eQTL based on trait-correlated expressed genes and trait-associated markers. The GWAS revealed 51 and 200 SNPs significantly associated with meat quality in a crossbred Pietrain×(German Landrace×Large White) (Pi×(GL×LW)) and a purebred German Landrace (GL) population, respectively. Most significant SNPs in Pi×(GL×LW) were located on chromosomes (SSC) 4 and 6. The data of 47,836 eQTLs at a significance level of p<10(-5) were used to scale down the number candidate genes located in these regions. These SNPs on SSC4 showed association with expression levels of ZNF704, IMPA1, and OXSR1; SSC6 SNPs were associated with expression of SIGLEC10 and PIH1D1. Most significant SNPs in GL were located on SSC6 and associated with expression levels of PIH1D1, SIGLEC10, TBCB, LOC100518735, KIF1B, LOC100514845, and two unknown genes. The abundance of transcripts of these genes in muscle, in turn, is significantly correlated with meat quality traits. We identified several genes with evidence for their candidacy for meat quality arising from the integrative approach of a genome-wide association study and eQTL analysis.

Keywords: GWAS; SNP chip; eQTL; microarray; pork quality..

Figure 1

Genome-wide association with meat quality traits in Pi×(GL×LW). Manhattan plots of genome-wide association analyses of six meat quality traits (PH24MLD, PH24MSM, PH45MLD, LF24MLD, IMP24MLD, and meat colour a*) in Pi×(GL×LW). Black lines indicate the significance threshold corresponding to negative log₁₀ (NegLog10)>5.

Figure 2

Genome-wide association with the abundance of selected transcripts in Pi×(GL×LW). Manhattan plots of genome-wide association analyses of transcript abundance of ZNF704, IMPA1, and OXSR1 in Pi×(GL×LW). Black lines indicate the significance threshold corresponding to negative log₁₀ (NegLog10)>5.

Figure 3

Genome-wide association with meat quality traits in GL. Manhattan plots of genome-wide association analyses of eight meat quality traits (meat colour a*, meat colour b*, PH24MLD, PH24MSM, PH45MLD, IMP24MLD, DVMLD, and LF24MLD) in GL. Black lines indicate the significance threshold corresponding to negative log₁₀ (NegLog10)>5.

Figure 4

Linkage disequilibrium of single nucleotide polymorphisms (SNPs) associated with meat quality. Linkage disequilibrium structures of SNPs located on SSC6 between 38-75 Mb that significantly associated with meat quality. Colour indicates the level of linkage disequilibrium between loci (correlation coefficient) as depicted in the key.

Figure 5

Genome-wide association with the abundance of selected transcripts in GL. Manhattan plots of genome-wide association analyses of transcript abundance of SIGLEC10, LOC100514845, LOC100518735, PIH1D1, unknown, KIF1B, TBCB and two non-annotated transcripts (unknown) in GL. Black lines indicate the significance threshold corresponding to negative log₁₀ (NegLog10)>5.