A total of 585 individuals of KNC (282 males and 303 females) comprising 68 F₁ nuclear families ranging in size from 3 to 20 chickens were used in this study. They were divided into five lines on the basis of plumage colors (Black, 88; Gray-Brown, 110; Red-Brown, 135; White, 122; and Yellow-Brown, 130) [25]. They were reared under controlled feeding and environmental conditions in the National Institute of Animal Sciences (NIAS), Republic of Korea. This study was also performed to meet recommendations described in “The Guide for the Care and Use of Laboratory Animals” published by the Institutional Animal Care and Use Committee of the National Institute of Animal Science (2012-C-037) in Korea. Body weights were measured at every two weeks from hatch to 20 weeks of age. Descriptive statistics of body weight and growth-related traits in the KNC population can be found elsewhere [26].

For the molecular genetic analysis, blood samples were taken from wing veins using 3 mL tubes containing ethylenediaminetetraacetic acid. Moreover, these blood samples were processed to isolate genomic DNA according to the method described by Miller et al [27]. The DNA concentration was measured using NanoDrop 2000c UV-Vis Spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA, USA). Isolated DNA genome was stored in the refrigerator at −20°C to maintain DNA quality until used.

Three pairs of primer sets and restriction enzyme information used in this study are shown in Supplementary Table S1. Two primers were designed using primer3 version 0.4.0 (http://frodo.wi.mit.edu/primer3/) based on GenBank reference NC_006088.3, and a primer pair for 9-bp indel detection that was previously reported by Wang et al [15]. The PCR was carried out in 20 μL volume containing 25 ng per μL DNA genome, 0.8 μL each primer, 1.6 μL 10 mM dNTP, 2.0 μL 10× reaction buffer, 0.2 μL Hot Start Taq Polymerase (GenetBio, Daejeon, Korea), and 12.6 μL double distilled water. The PCR reactions were carried out in the following steps: pre-denaturation at 94°C for 10 minutes, 35 cycles of 94°C for 30 seconds, annealing temperature (°C) for each primer for 30 seconds (Supplementary Table S1), and 72°C for 30 seconds, and followed by final extension at 72°C for 10 minutes. Reaction was performed using either GeneAmp PCR system 2700 (Applied Biosystems, Foster City, CA, USA) or C1000TM Thermal Cycler (BioRad, Hercules, CA, USA). The PCR products were visualized by 2% standard agarose gels stained with ethidium bromide (GenetBio, Korea). Initially, DNA pools containing 3 individual samples were used for PCR and sequencing to screen polymorphisms of THRSP. The PCR-RFLP was applied to confirm 3 single nucleotide polymorphism (SNPs) and a 3-bp indel. Approximately, 15 μL of PCR product was digested with 2 units of each restriction enzyme (Supplementary Table S1) based on the protocol provided by company (New England Biolabs Inc., Ipswich, MA, USA). Furthermore, the digested PCR product was separated on 3% agarose gels to identify genotype variations (Supplementary Figure S1). The 9-bp indel was confirmed using DNA Chip for The Agilent 2100 Bioanalyzer (Agilent Technologies Inc., Santa Clara, CA, USA) by following company protocol (Supplementary Figure S2). In addition, eleven of SNPs were genotyped using standard protocol of Fluidigm 192.24 SNPtype Genotyping Technology (Fluidigm, South San Francisco, CA, USA).

The phenotypic data of growth-related traits from the 585 F₁ progeny were in normal distribution (Data not shown). Single marker association analyses between the selected DNA markers and the traits using the F₁ progeny was conducted using the following linear mixed-model:

where, Y_ijklmn is the phenotype of the n^th animal, μ is overall mean, G_i is the fixed effect of THRSP genotype i, S_j is the fixed effect of sex j, B_k is the fixed effect of batch k, L_l is the fixed effect of line l, and P_m is random additive polygenic effect of animal m, ɛ_ijklmn is the random residual associated with the n^th animal. The mean and variance for random additive polygenic effects can be defined as: P~N(0, Aσ_a²), where A is based on relationship matrix computed from the nuclear pedigree in this study and σ_a² is the additive polygenic variance. The mean and variance for the residual random effect of individuals can be defined as: ɛ~N(0, Iσ_e²), where I is the identity matrix and σ_e² is the residual variance.

The significance level of fixed effects in the mixed-effects model above was computed by the Wald test in ASReml program (VSN International, Hemel Hempstead, UK). Nominal p<0.05 was regarded as significant for all tests.

The F₁ progeny were used to establish haplotypes at THRSP locus using the SNP markers showing nominal significance (p<0.05) for growth-related traits (Figure 1). The FImpute program based on pedigree information was applied to establish the haplotypes (REF. BMC Genomics 15, 478, 2014). Subsequently, the haplotype data of the F₁ chickens were used to conduct haplotype-based association study. The effect of haplotypes on the traits of interests were evaluated by the following general linear model using the MINITAB program (MINITAB, State College, PA, USA):

Y_ijklmn is the phenotype of the n^th animal, μ is overall mean, D_i is the fixed effect of THRSP diplotype i (Table 1), S_j is the fixed effect of sex j, B_k is the fixed effect of batch k, L_l is the fixed effect of line l, and Dam_(l,m) is fixed effect of m^th dam within l^th line, ɛ is the random residual term of the model.

Five sequences of THRSP were deposited in the National Center for Biotechnology Information (NCBI) website (GenBank accession numbers: KF574271-KF574275). A total of 30 polymorphisms were identified by aligning the sequences using ClustalW2 online software (http://www.ebi.ac.uk/Tools/msa/clustalw2/). These variations were spread from upstream to downstream of the gene. Eleven SNPs and an 3-bp indel at 1,355 bp from ATG were found in the flanking regions and untranslated regions of the THRSP gene, thirteen SNPs and an 3-bp indel at 965 bp from ATG were identified in intron, and three SNPs were also observed in the exon 1 of the THRSP gene. In addition, an 9-bp indel at 236 bp from start codon in the exon region of the THRSP gene which is identified in previous studies to be associated with body weight and abdominal fat traits in chickens was detected in present study (Table 1).

A total of 30 polymorphisms in the THRSP gene have been identified in this study. Of these, nine polymorphisms were investigated to be associated with the body weight traits using a mixed-effects model in 585 F₁ KNC progeny. Among the SNP markers listed in the Table 1, a genotype call rate less than 90% was not used for further association studies. Three SNPs showed significant association with growth-related traits (Figure 1). The g.−49G>T SNP of the THRSP gene was significantly associated with body weight at 20 weeks of age (BW20). Homozygote GG population had a higher BW20 than other genotypes. A SNP, namely g.451T>C, was also significantly associated with growth at 10 to 12 weeks (GR10–12). Moreover, g.1432A>C was statistically associated with growth 14 to 16 weeks and body weight at 18 weeks. The CC genotype of the g.1432A>C SNP was favorable for those traits. Additionally, four SNPs, namely g.−49G>T, g.696G>C, g.1419A>C, and g.1432A>C, were suggestively associated with some growth-related traits in F₁ KNC population (Figure 2).

We also constructed haplotypes of the THRSP gene to evaluate the effect of the THRSP haplotype on growth-related phenotypes in the F₁ chicken population. Haplotype and diplotype frequencies of the THRSP gene in the F₁ chickens are listed in the Table 2. The results of haplotype-based association analysis indicated that there were significant association between the diplotype and BW12. Ht1/ht2 diplotype was clearly favorable for BW12 (Table 3). Additionally, significant association of the diplotype with GR10–12 was detected in the F₁ animals. Ht1/ht2 diplotype was also highest in growth at 10 to 12 weeks of age. Ht4/ht4, where the most KNC population included in, was relatively moderate to high BW12 and GR10–12 traits (Table 3).