Expression Analysis of the PITX2 Gene and Associations between Its Polymorphisms and Body Size and Carcass Traits in Chickens

Simple Summary The Wuliang Mountain Black-bone chicken is a Chinese indigenous breed with good meat quality and strong resistance to disease. Like most of the other Chinese domestic breeds, it has a much slower early growth rate compared with foreign chicken breeds. Therefore, the genetic selection of body size and carcass traits is still the focus of Chinese indigenous chicken breeding. The paired-like homeodomain transcription factor 2 (PITX2) gene, an important transcription factor, plays an important role during the development of the eye, heart, skeletal muscle and other tissues in mammals. In chicken, the PITX2 gene affects the late myogenic differentiation of the limb. The objectives of this study were to detect the expression of the PITX2 gene and analyze the associations between the polymorphisms in the exons of the PITX2 gene and body size as well as carcass traits in chickens. The results could contribute to Chinese chicken breeding based on marker assisted-selection. Abstract PITX2 is expressed in and plays an important role in myocytes of mice, and it has effects on late myogenic differentiation in chickens. However, the expression profile and polymorphisms of PITX2 remain unclear in chickens. Therefore, the aim of the present study was to detect its expression and investigate single nucleotide polymorphisms (SNPs) within its exons and then to evaluate whether these polymorphisms affect body size as well as carcass traits in chickens. The expression analysis showed that the expression level of chicken PITX2 mRNA in the leg muscle and hypophysis was significantly higher (p < 0.01) than those in other tissues. The results of polymorphisms analysis identified two SNPs (i.e., g.9830C > T and g.10073C > T) in exon 1 and 10 SNPs (i.e., g.12713C > T, g.12755C > T, g.12938G > A, g. 3164C > T, g.13019G > A, g.13079G > A, g.13285G > A, g.13335G > A, g.13726A > G and g.13856C > T) in exon 3, including four novel SNPs (i.e., g.9830C > T, g.12713C > T, g.12938G > A and g.13856C > T). In the loci of g.10073C > T and g.12713C > T, chickens with the CT genotype had the highest (p < 0.05 or p < 0.01) breast depth and breast angle, respectively. For the locus of g.13335G > A, chickens with the GG genotype had the highest (p < 0.05 or p < 0.01) breast angle and shank circumference. For the locus of g.13726A > G, chickens with the GG genotype had the highest breast width, fossil keel bone length and shank circumference. The locus of g.12713A > G had significant effects on the PITX2 mRNA expression level in leg muscle. The H1H7 diplotype showed the highest shank circumference, and the H2H8 diplotype showed the highest breast muscle rate. The present research suggested that polymorphisms of the exons of the PITX2 gene were significantly associated with the body size and carcass traits of Wuliang Mountain Black-bone chickens and the PITX2 gene could be a potential candidate gene for molecular marker-aided selection in Wuliang Mountain Black-bone chickens and other chicken breeds.


Introduction
Wuliang Mountain Black-bone (WLMB) chicken is a Chinese indigenous breed with good meat quality, strong resistance to disease and medicinal and health-promoting values [1]. Like most of the other Chinese domestic breeds, it has a much slower early growth rate than foreign chicken breeds in which the selection for a fast growth rate has led to major welfare problems, such as musculoskeletal disorders, myopathies and organ failures [2][3][4]. Therefore, the selection of body size and carcass traits is still the focus of Chinese indigenous chicken breeding. Recently, the identification of DNA molecular markers related to quantitative traits based on candidate genes has become an important means of marker-assisted breeding to improve body size and carcass traits [5]. Paired-like homeodomain transcription factor 2 (PITX2) is a member of the bicoid-like homeobox transcription factor family [6], which has a homeobox-2 domain and an OAR domain (15-aa) [7]. The homeobox-2 domain can combine specifically with DNA through its helix-turn-helix (HTH) structure, and the OAR domain is where PITX2 interacts with pituitary homeobox 1.
In the past two decades, many researchers have demonstrated that the PITX2 gene is expressed in, and plays an important role during, the development of various tissues in mice, such as the heart, lung and dental germ [8][9][10][11]. In the myocytes of mice, the expression of myogenic regulatory factors (MRFs) might be regulated by the PITX2 gene [12]. PITX2-overexpression causes significant down-regulation of MyoD (an MRF) and an up-regulation of paired box factor 3 gene [13]. A subset of microRNA regulated by PITX2 also has profound effects on myoblast proliferation [14]. Moreover, it has been reported that the PITX2 protein can activate the Wnt signaling pathway and induce cell proliferation [15][16][17].
In chicken, the expression pattern of PITX2 is highly similar to that of mouse during pituitary development [18][19][20], and the PITX2 gene affects the late myogenic differentiation of the limb in the chick embryo [21]. These results suggested that the PITX2 gene is associated with the development of myocytes.
Recently, polymorphism studies of the PITX2 gene were carried out on humans and mainly focused on its associations with diseases [22][23][24]. In livestock, polymorphisms of the PITX2 gene have been reported to be related to the milk traits of dairy goats and the meat quality and growth of pigs [25][26][27]. However, the expression profile and polymorphisms of PITX2 remain unclear in poultry. Therefore, the objectives of this study were to detect the expression of the PITX2 gene and analyze the associations between the polymorphisms in the exons of the PITX2 gene and body size as well as carcass traits in chickens. The PITX2 gene of chickens is 4112bp long with three exons and is located on chromosome 4. The results of this study could contribute to functional research on the PITX2 gene and chicken breeding based on marker-assisted selection.

Materials and Methods
The present experiment was conducted following Chinese guidelines for animal welfare and was approved by the animal welfare committee of Zhejiang University (Approval Number: 12969).

Experimental Animals and Design
In the present study, Wuliang Mountain Black-bone (WLMB) chickens were used. These chickens have feathered feet, green ear lobes and black bones, along with the presence of massive dermal and visceral pigmentation. According to the Standards of Agricultural Industry of the People's Republic of China (NY/T 828-2004), the age of 300 days is a typical period to reflect the breed characteristics of poultry. Four hundred (300-day-old) female WLMB chickens, randomly selected from Wuliang Mountain Black-bone Chicken Professional Cooperatives of Long Street (Jingdon, China), were measured for body size and carcass traits for the association analysis with polymorphisms of the PITX2 gene. A total of 11 different tissues (heart, liver, spleen, lung, kidney, breast muscle, leg muscle, muscular stomach, glandular stomach, hypothalamus and hypophysis) were separately isolated from 10 chickens (randomly selected from 400 WLMB chickens) which were used for analyzing PITX2 mRNA expression in different tissues. After single nucleotide polymorphisms (SNPs)-traits association and tissues relative expression analysis, 10 individuals for each genotype of each SNP (a total of 120 WLMB chickens were randomly selected from 400 WLMB chickens) associated with body size or carcass traits were selected, analyzing the PITX2 mRNA relative expression in their leg muscles.
The body size traits, including body slope length (BSL), breast width (BW), breast depth (BD), breast angle (BA), fossil bone length (FBL), pelvis width (PW), shank length (SL) and shank circumference (SC) were measured before slaughter. Live weight (LW) was also measured before slaughter, and then carcass weight (CW), semi-eviscerated weight (SEW), eviscerated weight (EW), breast muscle weight (BMW) and abdominal fat weight (AFW) were measured on the carcass. The ratio of CW, SEW and EW to LW was calculated as slaughter rate (SR), semi-eviscerated rate (SER), eviscerated rate (ER), respectively. The ratios of BMW to EW were calculated as breast muscle rate (BMR). The ratios of AFW to the sum of EW and AFW were calculated as abdominal fat rate (AFR). All of the 8 body size traits and 11 carcass traits were evaluated in accordance with the Standards of the agricultural industry of the People's Republic of China (NY/T 823-2004).

DNA and RNA Extraction, cDNA Synthesis
Blood samples were collected from the jugular vein and stored at −20 • C (with EDTA) until genomic DNA extraction was performed. Genomic DNA was extracted using a Genome DNA Extraction Kit (TIANGEN, Beijing, China) under the guidance of the instruction manual.
Tissues were frozen immediately in liquid nitrogen after being separated and stored at −80 • C. Total RNA was extracted using TRIzol A + Total RNA reagent (TIANGEN, Beijing, China). The amount and purity (OD260/OD280 ratio > 1.9) of the extracted RNA were measured by the NanoDropND2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), and the quality of the obtained RNA was checked by electrophoresis. The PrimeScript RT Reagent Kit with gDNA Eraser (Takara, Japan) was used to convert RNA to cDNA following the manufacturer's protocol.
The PCR amplifications were performed with an initial denaturation cycle at 94 • C for 5 min followed by 35 cycles of 94 • C for 30 s, 45 s at the annealing temperature and extension at 72 • C for 90 s, and ending with an extension cycle at 72 • C for 10 min. The PCR products were checked by 0.8% agarose gel electrophoresis. Next, the PCR samples obtained were sequenced using Sanger sequencing by Tsingke Biotech Co., Ltd. (Hangzhou, China).
Real-time PCR was performed using a StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) with a 20 µL PCR mixture including 2 µL of cDNA, 10 µL of 2 × SYBR Premix Ex TaqII, 0.4 µL ROX Reference Dye (50×) 0.8 µL of forward primer, 0.8 µL of reverse primer and 6 µL of nuclease-free water. The 2× SYBR Premix Ex TaqII, ROX Reference Dye (50×) and nuclease-free water were contained in the TB Green Premix Ex Taq TM II (Takara, Dalian, China). Reactions were incubated in a 96-well optical plate (Applied Biosystems, American) at 95 • C for 30 s, followed by 40 cycles of 95 • C for 5 s, and 60 • C for 30 s. Each sample was analyzed in triplicate. The expression levels of the mRNA were normalized to the mRNA expression of chicken β-actin and were calculated using the 2 −∆∆Ct method [28].  [29].

Statistical
Sequences were analyzed with SeqMan II version 5.01 (DNAStar Inc., Madison, WI, USA). Genotype frequency, allele frequency and diversity parameters were calculated and the deviations from Hardy-Weinberg equilibrium (HWE) were checked. HaploView was used to analyze linkage disequilibrium (LD). Haplotypes and diplotypes were analyzed using PHASE 2.1, which implements a Bayesian statistical method for reconstructing haplotypes from population genotype data.
The general linear model (GLM) was separately implemented on each SNP and diplotype using statistical software SPSS 20.0 (SPSS, Chicago, IL, USA). The linear model used was as follows: where, Y ijk = the phenotypic value of traits, µ = the least squares mean of the population, G i = effect of genotypes (i = 1, 2 or 3) or diplotypes (i = 1, 2, 3, 4, 5, 6, 7 or 8), B j = effect of birthdate (j = 1, 2, or 3) and e ijk = random residual error. Multiple comparison results of the genotypes and diplotypes were corrected by Bonferroni correction.
The additive and dominance effects of the PITX2 gene were estimated as follows [30]: where AA, AB and BB were the least-squares means of the genotype AA, AB and BB groups, respectively. The AA genotype was a homozygous genotype for the two bases located before the '>' of the SNPs, while BB was a homozygous genotype for the two bases situated after the '>' of the SNPs. For example, the AA, AB and BB genotypes of the g.9830C > T locus in the PITX2 gene means the genotypes of CC, CT and TT. The relative expression data were checked for normality by the Kolmogorov-Smirnov test in statistical software SPSS v20.0 (IBM, Chicago, IL, USA), and logarithmic transformation (log2) was used to correct the non-normal distribution. Then, one-way ANOVA was used to analyze the relative expression of the PITX2 gene among groups. Multiple comparison results of the groups were corrected by Bonferroni correction.

Phylogenetic Relationships of PITX2 Proteins among 12 Species
As shown in Figure 1, the PITX2 proteins of all 12 species fell into two subgroups. All of the mammal sequences clustered into one subgroup and the sequences from the four avian species (Gallus gallus, Nothoproctaperdicaria, Parus major and Anas platyrhynchos) clustered into another subgroup. The additive and dominance effects of the PITX2 gene were estimated as follows [30]: where AA, AB and BB were the least-squares means of the genotype AA, AB and BB groups, respectively. The AA genotype was a homozygous genotype for the two bases located before the '>' of the SNPs, while BB was a homozygous genotype for the two bases situated after the '>' of the SNPs. For example, the AA, AB and BB genotypes of the g.9830C > T locus in the PITX2 gene means the genotypes of CC, CT and TT. The relative expression data were checked for normality by the Kolmogorov-Smirnov test in statistical software SPSS v20.0 (IBM, Chicago, IL, USA), and logarithmic transformation (log2) was used to correct the non-normal distribution. Then, one-way ANOVA was used to analyze the relative expression of the PITX2 gene among groups. Multiple comparison results of the groups were corrected by Bonferroni correction.

Phylogenetic Relationships of PITX2 Proteins among 12 Species
As shown in Figure 1, the PITX2 proteins of all 12 species fell into two subgroups. All of the mammal sequences clustered into one subgroup and the sequences from the four avian species (Gallus gallus, Nothoproctaperdicaria, Parus major and Anas platyrhynchos) clustered into another subgroup.

Relative Expression Levels of Chicken PITX2 mRNA in 11 Tissues
The relative expression results showed that chicken PITX2 mRNA was widely expressed in the heart, liver, spleen, lung, kidney, breast muscle, leg muscle, muscular stomach, glandular stomach, hypothalamus and hypophysis, but the level of expression was regulated in a tissue-specific fashion. The expression level of chicken PITX2 mRNA was significantly higher (p < 0.01) in the leg muscle and hypophysis than in other tissues ( Figure 2).

Relative Expression Levels of Chicken PITX2 mRNA in 11 Tissues
The relative expression results showed that chicken PITX2 mRNA was widely expressed in the heart, liver, spleen, lung, kidney, breast muscle, leg muscle, muscular stomach, glandular stomach, hypothalamus and hypophysis, but the level of expression was regulated in a tissue-specific fashion.
The expression level of chicken PITX2 mRNA was significantly higher (p < 0.01) in the leg muscle and hypophysis than in other tissues ( Figure 2). . Figure 2. Relative expression levels of chicken PITX2 mRNA in 11 tissues with the average ΔCt value of heart as the calibrator. Data were shown as the mean ± SEM for 10 chickens. Columns "**" showed highly significant difference (p < 0.01).

Genetic Polymorphisms
The results of sequence analysis showed that g.9830C > T and g.10073C > T in exon 1 and g.12713C > T, g.12755C > T, g.12938G > A, g.12961C > T, g.13019G > A, g.13079G > A, g.13285G > A, g.13335G > A, g.13726A > G and g.13856C > T in exon 3 were found in the chicken PITX2 gene, including four novel SNPs (i.e., g.9830C > T in exon 1, and g.12713C > T, g.12938G > A and g.13856C > T in exon 3). There was no variation found in exon 2. As shown in Table 2, no mutations led to amino acid changes.

Genotypic Frequencies, Allelic Frequencies and Diversity Parameter
Genotypic frequency, allele frequency and diversity parameters were displayed in Table 3. All SNPs exhibited three genotypes, except g.13856C > T. The dominant alleles of all SNPs were nonmutation alleles. The χ2-test showed that the genotypic distributions of different variants were Relative expression levels of chicken PITX2 mRNA in 11 tissues with the average ∆Ct value of heart as the calibrator. Data were shown as the mean ± SEM for 10 chickens. Columns "**" showed highly significant difference (p < 0.01).

Genetic Polymorphisms
The results of sequence analysis showed that g.9830C > T and g.10073C > T in exon 1 and g.12713C > T, g.12755C > T, g.12938G > A, g.12961C > T, g.13019G > A, g.13079G > A, g.13285G > A, g.13335G > A, g.13726A > G and g.13856C > T in exon 3 were found in the chicken PITX2 gene, including four novel SNPs (i.e., g.9830C > T in exon 1, and g.12713C > T, g.12938G > A and g.13856C > T in exon 3). There was no variation found in exon 2. As shown in Table 2, no mutations led to amino acid changes.

Genotypic Frequencies, Allelic Frequencies and Diversity Parameter
Genotypic frequency, allele frequency and diversity parameters were displayed in Table 3. All SNPs exhibited three genotypes, except g.13856C > T. The dominant alleles of all SNPs were non-mutation alleles. The χ2-test showed that the genotypic distributions of different variants were consistent with HWE (p > 0.05), except g.9830C > T (p < 0.05). Moreover, the data of polymorphism information content (PIC) indicated that g.13335G > A and g.13726A > G were moderately polymorphic while the rest were lowly polymorphic. The locus of g.9830C > T was not used in further analysis since it deviated from the HWE. After excluding the genotypes with less than 10 individuals, the SNPs with three genotypes (g.10073C > T, g.12713C > T, g.13335G > A and g.13726A > G) were used for further analysis of the association between the SNPs and body size as well as carcass traits.

Analysis of LD Coefficient, Haplotype and Diplotype
The results of LD analysis indicated that g.12961C > T, g.13019G > A, g.13079G > A, g.13285G > A, and g.13335G > A showed strong LD, constructing a 374 bp block 1 ( Figure 3). As shown in Table 4, nine haplotypes (H1-H9) were formed based on block 1, where the H1 haplotype showed the highest frequency, followed by H2 and H7. In the diplotype study, eight diplotypes were formed based on the nine haplotypes, where the frequency of H1H2 was the highest followed by H1H1. The diplotypes with fewer than 10 individuals were excluded.

Association of Gene Polymorphisms with Body Size and Carcass Traits
The association of PITX2 gene SNPs with body size and carcass traits of WLMB chickens are shown in Table 5 and Table 6. For the locus of g.10073C > T, BD in chickens with the CT genotype was significantly higher than those with the CC (p < 0.05) and TT genotypes (p < 0.01). For the locus of g.12713C > T, chickens with the CT genotype had a higher BA than those with the CC genotype (p < 0.01). For the locus of g.13335G > A, chickens with the GG genotype had a higher BA than those with the GA genotype (p < 0.01), and a higher SC than those with GA (p < 0.05) and AA genotypes (p < 0.01). For the locus of g.13726A > G, chickens with the GG genotype had higher FBL and SC than those with the AA genotype (p < 0.05), and a higher BW than those with AA (p < 0.01) and AG genotypes (p < 0.05). Moreover, g.10073C > T had an over-dominant effect on BD, and g.12713C > T and g.13335G > A had an over-dominant effect on BA. In contrast, for g.13335G > A on SC, and

Association of Gene Polymorphisms with Body Size and Carcass Traits
The association of PITX2 gene SNPs with body size and carcass traits of WLMB chickens are shown in Tables 5 and 6. For the locus of g.10073C > T, BD in chickens with the CT genotype was significantly higher than those with the CC (p < 0.05) and TT genotypes (p < 0.01). For the locus of g.12713C > T, chickens with the CT genotype had a higher BA than those with the CC genotype (p < 0.01). For the locus of g.13335G > A, chickens with the GG genotype had a higher BA than those with the GA genotype (p < 0.01), and a higher SC than those with GA (p < 0.05) and AA genotypes (p < 0.01). For the locus of g.13726A > G, chickens with the GG genotype had higher FBL and SC than those with the AA genotype (p < 0.05), and a higher BW than those with AA (p < 0.01) and AG genotypes (p < 0.05). Moreover, g.10073C > T had an over-dominant effect on BD, and g.12713C > T and g.13335G > A had an over-dominant effect on BA. In contrast, for g.13335G > A on SC, and g.13726A > G on BW, FBL and SC, the additive effect was bigger than the dominance effect, whereas they had no association with carcass traits (Table 6). Tables 7 and 8 show the association of diplotypes with body size and carcass traits. SC in chickens with the H1H7 diplotype was significantly higher than those with the H1H4 diplotype (p < 0.01) and chickens with the H2H8 diplotype had a higher BMR than those with the H1H2 and H1H8 diplotypes.

PITX2 mRNA Relative Expression with Different Genotypes in Leg Muscle
Since the highest PITX2 mRNA relative expression was in the leg muscle, we selected it to detect expression levels for different genotypes of the loci (g.10073C > T, g.12713C > T, g.13335G > A and g.13726A > G) that were significantly associated with body size traits. The results showed that only the genotype of locus g.12713C > T was significantly associated with the expression level of the PITX2 gene. As shown in Figure 4, for the locus of g.12713C > T, leg muscles with the CT genotype had higher (p < 0.05) PITX2 mRNA expression level than those with CC and TT genotypes.

PITX2 mRNA Relative Expression with Different Genotypes in Leg Muscle
Since the highest PITX2 mRNA relative expression was in the leg muscle, we selected it to detect expression levels for different genotypes of the loci (g.10073C > T, g.12713C > T, g.13335G > A and g.13726A > G) that were significantly associated with body size traits. The results showed that only the genotype of locus g.12713C > T was significantly associated with the expression level of the PITX2 gene. As shown in Figure 4, for the locus of g.12713C > T, leg muscles with the CT genotype had higher (p < 0.05) PITX2 mRNA expression level than those with CC and TT genotypes.
. Figure 4. Relative expression levels of the chicken PITX2 mRNA in leg muscle with different genotypes in the loci of g.10073C > T, g.12713C > T, g.13335G > A and g.13726A > G with the average ΔCt value of CC, CC, GG and AA genotype as the calibrator, respectively. Data were shown as the mean ± SEM for 10 chickens of each genotype. The column with "*" showed a significant difference (p < 0.05).

Discussion
Based on the construction of a neighbor-joining phylogenetic tree in which the PITX2 amino acid sequences of 12 species were aligned, the result showed that the PITX2 amino acid sequences were relatively conserved and the SNPs detected in the present study occurred at conserved regions, which indicated the function of the PITX2 gene might be similar across the 12 species. As shown in the phylogenetic tree, chicken (Gallus gallus) and mallard (Anas platyrhynchos) were grouped, indicating that the function of the PITX2 gene in chicken was most similar to that in mallard.
In our study, PITX2 mRNA was expressed widely in 11 tissues of WLMB chickens which was the same as in other animals [32]. Although the study of Abu-Elmagd et al. did not detect an effect of PITX2 on early limb myogenesis in the chick embryo, they found the late myogenic differentiation in the limb was affected by PITX2 [21]. Moreover, several studies have confirmed that the expression of the PITX2 gene can activate the LIM, POU and SIX families to regulate the development and function of the hypophysis [33]. PITX2 mRNA was also detected during the early development of the hypophysis in the chick [34]. These studies showed that the PITX2 gene might play an important role in the development of chicken limb muscles and the hypophysis. In the present study, the expression level of the PITX2 gene in the leg muscle and hypophysis was the highest, which was consistent with these prior reports.
In the current research, 12 SNPs were found in the exons of the chicken PITX2 gene, including four novel SNPs that have not been previously recorded by NCBI. According to the results of the χ2 Figure 4. Relative expression levels of the chicken PITX2 mRNA in leg muscle with different genotypes in the loci of g.10073C > T, g.12713C > T, g.13335G > A and g.13726A > G with the average ∆Ct value of CC, CC, GG and AA genotype as the calibrator, respectively. Data were shown as the mean ± SEM for 10 chickens of each genotype. The column with "*" showed a significant difference (p < 0.05).

Discussion
Based on the construction of a neighbor-joining phylogenetic tree in which the PITX2 amino acid sequences of 12 species were aligned, the result showed that the PITX2 amino acid sequences were relatively conserved and the SNPs detected in the present study occurred at conserved regions, which indicated the function of the PITX2 gene might be similar across the 12 species. As shown in the phylogenetic tree, chicken (Gallus gallus) and mallard (Anas platyrhynchos) were grouped, indicating that the function of the PITX2 gene in chicken was most similar to that in mallard.
In our study, PITX2 mRNA was expressed widely in 11 tissues of WLMB chickens which was the same as in other animals [32]. Although the study of Abu-Elmagd et al. did not detect an effect of PITX2 on early limb myogenesis in the chick embryo, they found the late myogenic differentiation in the limb was affected by PITX2 [21]. Moreover, several studies have confirmed that the expression of the PITX2 gene can activate the LIM, POU and SIX families to regulate the development and function of the hypophysis [33]. PITX2 mRNA was also detected during the early development of the hypophysis in the chick [34]. These studies showed that the PITX2 gene might play an important role in the development of chicken limb muscles and the hypophysis. In the present study, the expression level of the PITX2 gene in the leg muscle and hypophysis was the highest, which was consistent with these prior reports.
In the current research, 12 SNPs were found in the exons of the chicken PITX2 gene, including four novel SNPs that have not been previously recorded by NCBI. According to the results of the χ2 test, g.9830C > T was not in agreement with HWE, which might be explained by the artificial selection in the breed [27]. All mutations were synonymous, demonstrating that the sequence of the PITX2 amino acids seemed to be conserved [35]. It is worth noting that silent mutations can also affect gene function and phenotype by regulating the stability of the mRNA or inducing alternative splicing events [36][37][38]. Synonymous mutations change the tRNA that transports the same amino acid, and abundances of different tRNAs are different, which will influence the transcription efficiency [39]. Moreover, the function of microRNA is to regulate the mRNA level and protein expression by binding to the 3 UTR of mRNA [40]. Thus, silent mutations could not be ignored.
Polymorphisms of the PITX2 gene are significantly related to the milk traits and meat quality of dairy goats and pigs, respectively, and might contribute to the improvement of these traits [26,27]. Significant effects of the polymorphisms of the PITX2 gene on chicken body size and carcass traits were detected in the current study. The results showed that four SNPs (g.10073C > T, g.12713A > G, g.13335G > A and g.13726A > G) were associated with the body size traits of WLMB chickens. The locus of g.12713A > G had significant effects on the PITX2 mRNA expression level in leg muscle. Three SNPs of the PITX2 gene also had significant effects on goat growth traits and mRNA expression levels [41]. The four mutations associated with body size traits in this study were silent. The molecular mechanism of how the four silent mutations affect the phenotype still needs further research. Although there were no SNPs associated with the carcass traits, the diplotypes had a significant effect on BMR. All of the above demonstrate that polymorphisms of the chicken PITX2 gene are associated with the body size and carcass traits of WLMB chickens.

Conclusions
In summary, the present study showed that the expression level of chicken PITX2 mRNA in the leg muscle and hypophysis was the highest, and polymorphisms of the PITX2 gene were significantly associated with the body size and carcass traits of Wuliang Mountain Black-bone chickens, which indicated that the PITX2 gene could be a potential candidate gene for molecular marker-aided selection in chicken breeding.