The tree generated from the amino acid sequences of 15 species showed that a positive selection was possible for the IGF1R gene in the early evolution of the Japanese firebelly newt and Africa clawed frog, and in the middle evolution of the zebrafish, turbot, and common carp (ω > 1). But the IGF1R gene was conservative in the whole evolution between the mammals and chickens (Figure 1).

In the present study, 10,818 bp of sequence in the IGF1R gene, in which exon regions were preferred but segmental introns were also included (Table 1 in Supplementary Materials File 1), was scanned, and 70 SNPs were identified between XH and WRR chickens. Among the 70 SNPs, 7 SNPs were located in the 5' regulated region, 15 in the coding regions, 57 in the intron regions, and 1 in the 3' regulated region. Average density of SNPs was one SNP per 173 bp (70/12,038) in the whole region studied, with one SNP per 273 bp (15/4092 bp) in the coding regions, and one SNP per 140 bp (48/6726 bp) in the intron regions. Among the 70 SNPs, 51 SNPs are transition, 17 transversion, and 1 one base insertion/deletion. Fifteen SNPs were found in the coding regions, but only one SNP was non-synonymous mutation (Asn → Ser in the exon 3).

The nucleotide diversity of the chicken IGF1R gene was (2.87 ± 0.28) × 10^-3, the nucleotide polymorphism was (2.62 ± 1.06) × 10^-3, and all parameters of the neutral test were negative but not significant. For the 5' flanking region, the nucleotide diversity of the chicken IGF1R gene was (3.83 ± 1.62) × 10^-3, the nucleotide polymorphism is (3.01 ± 1.0) × 10^-3, all parameters of the neutral test were positive, and the values of Fu and Li's D, and F test were significant (P < 0.05). For the intron region, the nucleotide diversity of the chicken IGF1R gene was (3.52 ± 1.49) × 10^-3, the nucleotide polymorphism was (3.98 ± 0.43) × 10^-3, all parameters of the neutral test were negative but not significant. For the exon region, the nucleotide diversity of the chicken IGF1R gene was (1.59 ± 0.57) × 10^-3, the nucleotide polymorphism was (1.43 ± 0.32) × 10^-3, and all parameters of the neutral test were positive but not significant (Table 2 in Supplementary Materials File 1).

Between the XH and WRR chickens, the allelic frequencies of C17393427T and A17323673G were significantly different at the P < 0.05 level and the allelic frequencies of the 7 SNPs, C17445985T, A17417734G, C17417042G, T17334342C, A17327275C, A17307494G, and A17299834G, were significantly different at the P < 0.01 level. No significant differences for mean heterozygosity of the 18 SNPs were found, but significant differences of A17313488G, A17307750G, A17307494G, A17299834G, and C17293932T were observed in the XH and WRR chickens (Table 3 in Supplementary Materials File 1).

Eight TagSNP, C17293932T, A17299834G, A17307750G, A17313488G, C17393427T, T17416994C, G17445596A, and C17445985T, were identified in the XH chickens by HapBLOCK software. Another eight TagSNP, C17293932T, A17299834G, A17307494G, A17307750G, T17416994C, C17417042G, A17417734G and G17445596A, were also identified as TagSNPs in the WRR chickens.

Average values of r² showed that the linkage disequilibria declined with increasing physical distance between SNP pairs in the XH and WRR chickens (Figure 1 in Supplementary Materials File 1). The effective extent of linkage disequilibrium was 27, 441 bp in the WRR chickens, but not found in the XH chickens. Possible regions of strong linkage disequilibrium were found between exon 6 and the 3' untranslating region (between A17327275C and C17293932T) in the WRR chickens (Table 4 in Supplementary Materials File 1).

Associations of the 6 SNP with chicken early growth and carcass traits were analyzed, but only the A17299834G of the chicken IGF1R gene was significantly associated with some growth and carcass traits. The A17299834G of the chicken IGF1R gene was significantly associated with chicken carcass weight, eviscerated weight with giblets, eviscerated weight, body weights at 28, 35, and 56 d of age, leg length at 56 d of age, and daily weight gain at 0–4 weeks (P < 0.05). Significantly and suggestively dominant effects of AG genotype were observed for chicken carcass weight, eviscerated weight with giblets, breast muscle weight, eviscerated weight, fat thickness under skin, fat width, body weights at 14, 21, 28, 35, 42, 49, 56, and 77 d of age, and leg length at 42 and 56 d of age (Table 1). In other words, the A17299834G SNP affected the chickens' early growth.

For the IGF1R gene, two haplotype blocks were observed in the F₂ resource population. Block 1 comprised A17307750G and A17307494G, located between intron 17 and 18, and block 2 comprised A17299834G and C17293932T, located between exon 20 and the 3'untranslation region (Figure 2 in Supplementary Materials File 1).

In block 1, four haplotypes were observed in the F₂ individuals of the resource population. Three distinct haplotypes, H1, H2, and H3, accounted for 95.4% of the total number of the four haplotypes. Among the four haplotypes, allele H4 had the lowest allelic frequency of 0.46%, and H1 had the highest allelic frequency of 51.60% (Table 5 in Supplementary Materials File 1). In block 2, four haplotypes were also observed in the F₂ individuals of the resource population. Three distinct haplotypes, E1, E2, and E3, accounted for 96.6% of the four observed haplotypes. Among the four haplotypes, the allelic frequency of E4 was the lowest at 0.34%, the highest allelic frequency was 57.88% for haplotype E2.

Significant associations of the haplotypes of A17307750G and A17307494G with chicken growth and carcass traits were observed. The haplotypes of A17307750G and A17307494G were significantly associated with body weights at 28 and 49 d of age, and with daily weight gain at 0–4 weeks at the P < 0.05 level, and significantly associated with body weight at 35 d of age, and leg length at 42 and 49 d of age at the P < 0.01 level. The haplotypes composed of A17299834G and C17293932T affected the chickens' early growth. Significantly and suggestively dominant effects of H1H3 diplotype were observed for body weights at 7, 14, 21, 28 d of age, and daily weight gain at 0–4 weeks. The H2H4 diplotype was dominant for body weights at 35 and 49 d of age (Table 2). Significant associations of the haplotypes of A17299834G and C17293932T with chicken growth and carcass traits were observed. The haplotypes of A17299834G and C17293932T were significantly associated with breast angle width, eviscerated weight with giblets, breast muscle weight, body weights at 28 and 35 d of age, leg length at 56 d, and daily weight gain at 0–4 weeks (P < 0.05), and significantly associated with body weight at 35 d of age, and leg length at 42 and 49 d of age at the P < 0.01 level. In other words, the haplotypes affected the chickens' early growth. The E1E4 diplotype was dominant for body weights at 7, 14, 21, 28, 35, 42, and 49 d of age, daily weight gain at 0–4 weeks, carcass weight, eviscerated weight with giblets, eviscerated weight, breast muscle weight, brain neck weight, breast angle width, small intestines length at 90 d of age, and leg length at 56 d of age (Table 3). Multiple comparisons showed that the E1E4 diplotype had a dominant effect in the all traits.

The initial SNP discovery was carried out on the DNA pools of 7 breeds, XH chickens, Taihe Silkie chickens, Beijing Fatty chickens, Yangshan chickens, Dwarf chickens, White Leghorn chickens, and WRR chickens. Ten individuals for each breed generated one pooled DNA sample. An equal amount of DNA was taken from each individual and was pooled to generate the 7 pooled DNA samples. DNA samples were diluted and reassessed to obtain an equal amount of DNA from each individual.

A F₂ resource population was constructed by reciprocal crossing the XH with WRR chickens 26. The F₂ individuals were raised in floor pens and fed commercial corn-soybean diets that met NRC requirements. The birds from six batches were kept in different pens, and the sizes of all pens were the same. The body weight was measured in grams at hatch, 7, 14, 21, 28, 35, 42, 49, 56, and 90 d of age. The 434 individuals from the F₂ generations (221 male and 213 female) were slaughtered at 90 d of age. Shank length (mm), head width (mm), breast width (mm), breast depth (mm), body length (cm), breast angle width (degree), carcass weight (g), fat thickness under skin (mm), fat width (mm), eviscerated weight with giblets (g), eviscerated weight (g), breast muscle weight (g), leg muscle weight (g), wing weight (g), abdominal fat pad weight (g), head and neck weight (g), weights of heart, liver, and gizzard (g), and small intestine length (cm) were recorded.

Two unrelated populations, consisting of 112 XH individuals and 86 WRR individuals, respectively, were sampled for genetic diversity investigation in the present study. The XH and WRR chickens were parents of the F₂ resource population, both from Guangdong Wens Foodstuff Corporation Ltd. (Guangdong, China). The XH chicken is a Chinese native breed with slow growth rate, and the WRR chicken is of fast growth rate. There is significant difference in growth and carcass traits between the XH and WRR chickens.

Available sequences of the chicken IGF1R gene were used as templates for designing specific primers by the Genetool software http://www.biologysoft.com. Twenty-two primers were obtained and an optimal length of the PCR product was set between 450 and 800 bp. Exon regions were preferred, but segmental intron sequences were also included. Details were listed in Table 1 in Supplementary Materials File 1.

For genetic diversity investigation and linkage disequilibrium analyses, 18 SNPs (Table 3 in Supplementary Materials File 1) were selected based on the following criterions. (1) Positions: SNP in coding regions were selected preferably over those from non-coding regions. (2) Functional domain: SNP in the important structural and conservative functional domains such as extracellular domain, joint of α subunit and β subunit, and tyrosine kinase domain, were preferred to those in the other domains. (3) Potential regulating units: SNP located at the potential regulatory sites of the un-translating region were preferred to those located at the other sites. (4) Density: An average density of 1 SNP per 8.3 kb was determined, and a total of 18 SNPs were selected in a 150 kb full sequence of the chicken IGF1R gene.

The ω value was calculated using codeml program of PAML software 27, and formula was followed,

ω = dN/dS,

Where dN and dS are the number of non-synonymous substitutions per non-synonymous site, and the number of synonymous substitutions per synonymous site, respectively. In the present study, the selection function of the DNA sequences was analyzed in species using Branch model of codeml program (M0 and M1), and the selection function of the amino acid was analyzed using site model (M7 and M8) in the evolution http://taxonomy.zoology.gla.ac.uk/rod/treeview.html.

SNPs having different allelic frequencies in the 7 pooled DNA samples were validated by sequencing the PCR product from both ends and by re-sequencing PCR products. SNPs were identified by alignment of sequences using the BioEdit program http://www.biologysoft.com. SNP calls were made on sequences of high quality.

The PCR was performed in a final volume of 25 μL containing 1 μL genomic DNA (2.5 ng/μL), 0.25 μL each primer (25 μM), 0.5 μL deoxynucleotide triphosphates (10 μM) mixture, 1.5 μL MgCl₂ (25 mM), 0.2 μL DNA polymerase (5 U/μL) (TaKaRa, Japan) and 2.5 μL 10 × reaction buffer on an ABI 2700 thermal cycle1 with the following profile, initial denaturation at 94°C for 4 min; 35 cycles of 94°C for 30 s, 60°C for 30 s, and 72°C for 30 s, and a final elongation at 72°C for 5 min.

We used DNASP4.10 to perform tests of neutrality on the basis of the allelic frequency spectrum 28. These included Tajima's D 29, and Fu and Li's D and F test statistics 30.

In the dynamic programming algorithm for haplotype block partitioning and tag SNP selection based on haplotype data, Zhang et al. used the following recursive formula 1931,

Sj = min{Si_ 1 + f(i,..., j),

if block(i,..., j) = 1}(1 _ j < n),

Where f(i,..., j) is the number of tag SNPs in this block, block(i,..., j) is a Boolean function, and block(i,..., j) = 1 if and only if SNP (i,..., j) can form a block, Sj is the minimum number of tag SNPs for the optimal haplotype block partition of the first j SNPs, and S0 = 0 http://www.cmb.usc.edu/msms/HapBlock.

The difference of allelic frequencies between the two unrelated chicken populations was tested using Mantel-Haenszel ChiSquare (SAS 8.1 FREQ).

The linkage disequilibrium r² value between each pair of SNPs and the haplotype structure of SNP within the gene were estimated by Haploview 32.

Haplotypes were constructed based on haplotype structure of the 18 SNPs in all 434 experimental animals by use of the PHASE 2.0 programme 33, whose function was to reconstruct haplotypes from the population data.

Data were analyzed by the GLM procedure of SAS 8.0 (Statistical Analysis Systems Institute Inc., Cary, NC) and the genetic effects were analyzed by a mixed procedure according to the following model,

Y= μ +S_i+B_j+g_k+f_x+e_ijkx

Where Y represented the dependent variable, μ, S_i, B_j, g_k, f_x, and e_ijkx represented the population mean, fixed effects of sex, fixed effects of hatch, genotype effect, family effect, and random error, respectively. Multiple comparisons were analyzed with least squares means, followed by the multiple comparison procedure, the multiple comparison procedures was followed:

Y i − Y ¯ i = ( Y i − Y ^ i ) + ( Y ^ i − Y ¯ i ) Where  ∑ ( Y i − Y ^ i ) 2  was least value ,  and  ∑ ( Y i − Y ^ i ) 2 = 0 MathType@MTEF@5@5@+=feaagaart1ev2aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacPC6xNi=xI8qiVKYPFjYdHaVhbbf9v8qqaqFr0xc9vqFj0dXdbba91qpepeI8k8fiI+fsY=rqGqVepae9pg0db9vqaiVgFr0xfr=xfr=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@802D@