Introduction

Many economically important traits of domestic animals, such as growth, egg production, and body composition are controlled by quantitative trait loci. Classical quantitative genetics can not independently decompose individual gene effects from the multiple genes associated with the variation of complex and quantitative traits. With the advances of the cutting edge molecular biology, sequencing of the entire genome, and comparative genomics, candidate gene approach, facilitated by results of the comparative genomic study, has been proven to be powerful for studying the genetic architecture of complex traits and is a far more effective and economical method for direct gene discovery [1]. It has been widely applied in identification of genes responsible for economically important traits in animals [2–7].

Myocyte-specific enhancer-binding factor 2 (MEF2) genes belong to the MADS (MCM1, Agamous, Deficiens, Serum Response Factor) box family of transcription factors and have been shown to play a pivotal role in morphogenesis and myogenesis of skeletal, cardiac, and smooth muscle cells [8]. In vertebrates, there are four MEF2 genes, viz. MEF2A, MEF2B, MEF2C and MEF2D. Members of the MEF2 family share homology within the MADS box domain (56 amino acids), which mediates DNA binding and dimerization. An additional MEF2 domain (29 amino acids), adjacent to the MADS box, is also highly conserved among the MEF2 proteins but is absent from other MADS box proteins [9–11]. MEF2 is selectively expressed in differentiated myocytes and activates nearly all skeletal and cardiac muscle genes by binding a conserved A/T-rich DNA sequence in the control region of them [12–17]. Different muscle cell types share a common myogenic differentiation program controlled by MEF2. Somatic, cardiac, and visceral muscle cells fail to differentiate in Drosophila embryos with a loss of MEF2 function [18, 19]. Moreover, MEF2 controls skeletal muscle formation after terminal differentiation in nascent fibers and drives expression of genes encoding thick filament proteins [20]. Until now, more and more data provide evidence that MEF2 is an essential factor for vertebrate skeletal muscle development and differentiation [21–28].

Chicken is an important domestic animal for meat consumption, egg production, and entertainment (e.g. cockfighting). It is also a good animal model for disease study because the population has undergone intensive artificial selection during the breeding. In this study, we used the PCR-SSCP (single strand conformation polymorphism) method to identify the polymorphisms of the MEF2A gene, a member of the MEF2 gene family, in four Chinese chicken populations. Our aims are (1) to discern the potential association between the MEF2A polymorphisms and muscle traits, and (2) to test whether the single nucleotide polymorphisms of MEF2A can be used in the marker-assisted selection of chicken carcass traits.

Materials and methods

Resource populations

Four chicken populations, including Daheng quality meat chicken (DH) (n = 240), Sanhuang chicken (SH) (n = 105), Mountainous black-bone chicken (SD) (n = 66), and Abor Acre chicken (AA) (n = 60) were used in this study. The DH chicken was developed by the Sichuan Daheng Poultry Breeding Company and is characteristic of spotty feather and yellow or white skin. The SH chicken is an indigenous breed in Guangdong Province and was named by its yellow plumage, skin and shank. The SD is an indigenous breed in Sichuan Province and has black feather and skin. These three Chinese native chicken breeds have favorable meat quality, but grow slowly. The AA chicken is an introduced broiler with fast growth speed and white feather. All birds were hatched on the same day, housed on the deep-litter bedding, and transferred to the growing pens at the age of 7 weeks. Birds had access to feed (commercial corn-soybean diets meeting the National Research Council’s [NRC] requirements) and water ad libitum.

Before slaughter, blood samples were collected from all 471 individuals of the four chicken populations. The genomic DNA was isolated by the standard phenol/chloroform method.

Measurement for carcass traits

At the age of 90 day, live body weight (BW) was measured on live birds after 12 h with no access to feed. After slaughter at the same day of age, the carcass traits, including carcass weight (CW), eviscerated weight (EW), semi-eviscerated weight (SEW), breast muscle weight (BMW), leg muscle weight (LMW), were measured. The CW was measured on the chilled carcass after removal of the feather. Semi-eviscerated weight was measured on the carcass after removal of the trachea, esophagus, gastrointestinal tract, spleen, pancreas, and gonad. Eviscerated weight was measured on the semi-eviscerated weight after removal of the head, claws, heart, liver, gizzard, glandular stomach, and abdominal fat. The ratios of these traits to CW were calculated as eviscerated percentage (EP), semi-eviscerated percentage (SEP), breast muscle percentage (BMP), leg muscle percentage (LMP). The ratio of carcass weight to live body weight was calculated as carcass percentage (CP). All the experiments were complied with the requirements of the Directory Proposals on the Ethical Treatment of Experimental Animals of China.

Genotyping for MEF2A gene polymorphisms

Ten pairs of primers were designed to investigate the potential SNPs in the MEF2A gene, according to the Gallus gallus MEF2A sequence (GenBank Accession No: NC_006097.2) (Table 1). The PCR reaction was performed in a final volume of 10 μl containing 0.8 μl of genomic DNA (2.5 ng/μl), 0.3 μl of each primer (10 pmol/μl), 3.6 μl ddH2O, 5 μl of 2 × MasterMix (Tiangen, Beijing, China). The following PCR cycle condition was used: an initial denaturation at 94°C for 5 min; 35 cycles of 94°C for 30 s, 51–55°C (depends on the primer pair used) for 30 s, and 72°C for 60 s; and a final elongation at 72°C for 10 min. The PCR products of the MEF2A gene were screened by using the SSCP method and were separated by 12% polyacrylamide gel electrophoresis. Genotypes were recorded according to the band patterns. PCR products that had polymorphism as revealed by SSCP were further amplified, purified, and sequenced by a commercial sequencing company (Invitrogen, Shanghai, China).

Table 1 Forward (F) and reverse (R) primers for amplification of the chicken MEF2A gene
Full size table

Statistical analysis

Because the data for some carcass traits were not normally distributed according to the Shapiro-Wilks test in SAS 8.0 (SAS Institute Inc., Cary, NC), we had some transformations for the data before further processing. The parameters of BW, CW, SEW, EW, BMW, and LMW were analyzed under the linear model with parameters estimated on the Square Root scale. The CP, SEP, LMP were shifted and rescaled to give approximate normality and equality of variance, with parameters converted by arcsine transformation. The transformed data were analyzed by the GLM procedures of SAS (SAS Inst. Inc., Cary NC). The genetic effects were analyzed by mixed procedure according to the following model:

$$ Y = \mu + B + S + G + (B \times G) + bX + e $$

where Y, the dependent variable; μ, the population mean; B, fixed effects of breed; S, fixed effects of sex; G, fixed effects of genotype or haplotype, B × G, the interaction between the genotype or haplotype and breed, X, carcass weight (covariance), b, coefficient of regression, and e, random error. The interaction of G and B was included in the model if its effect was P < 0.05 for a given trait, otherwise, they were excluded from the final model. Multiple comparisons were performed with the least squares means. Values are considered significant at P < 0.05 and are presented as least square means ± standard error means.

Haplotype reconstruction

Haplotypes were constructed based on the SNPs identified in all 471 experimental birds using the PHASE 2.0 programme [29]. The function of this program is to reconstruct haplotypes from the population data. The genetic status of the subjects were expressed as the combination of two haplotypes (diplotype configuration). Genetic effects of the diplotypes were performed with the mixed model mentioned above.

Results

SNPs of the chicken MEF2A gene

Three MEF2A SNPs were detected by the PCR-SSCP method, which defined nine genotypes (Table 2). The homozygous genotype for each locus was further confirmed by sequencing. For locus A, the C → T mutation at position 46023 nucleotide (relative to GenBank accession No: NC_006097.2) was located on the 5′-untranslated region 5′ UTR. For locus B, the G → A mutation at position 72626 nucleotide was located on the fourth exon and this SNP didn’t cause amino acid change. For locus C, the G → T mutation at position 89232 nucleotide was located on seventh intron of the chicken MEF2A gene.

Table 2 Genotypes and allele frequencies of the chicken MEF2A gene in four populations
Full size table

Association of the MEF2A alleles and genotypes with chicken carass traits

The frequencies of the MEF2A alleles and genotypes are presented in Table 2. In locus A, the allele A2 was predominant due to its higher frequency than allele A1 in all the 4 populations. In locus B, homozygous genotype B1B1 had the highest frequency in all the populations. In locus C, the frequency of genotype C2C2 was the lowest in all four populations and this genotype was not detected in SD and AA chickens. The χ2-test was performed to examine the Hardy–Weinberg equilibrium (HWE). The three SNPs significantly deviated from HWE, except for SNP2 in population AA and SNP3 in population SH and SD. The χ2-test for the genotype frequency showed that there were statistically significant differences among all the four chicken breeds (P < 0.05 or 0.01).

We also estimated the potential association between genotypes of the MEF2A gene and 11 carcass traits (BW, CW, SEW, EW, BMW, LMW, CP, SEP, EP, BMP, LMP). The results of the GLM analysis were summarized in Table 3. In locus A, genotypes had great significantly effect on BW (P = 0.0122), CW (P = 0.0115), SEW (P = 0.0185), EW (P = 0.0150), and LMW (P = 0.0208), but no association was observed for the other six carcass traits (P > 0.05). In particular, the chickens harboring genotype A2A2 had significantly higher BW, CW, EW (P < 0.0.1), SEW, LMW (P < 0.05) than those of A1A1 chickens. In locus B, genotypes had great significantly effect on BW (P = 0.0002), CW (P = 0.0003), BMW (P = 0.0114), LMW (P = 0.0007) and LMP (P = 0.0350). The BW, CW, BMW and LMW of chicken with genotype B1B1 were significantly higher than those of genotypes B2B2 (P < 0.01). The LMP of B1B1 was significantly higher than that of genotypes B1B2 and B2B2 (P < 0.05). In locus C, genotypes had great significantly effect on CP (P = 0.0109) and SEP (P = 0.0238). The chickens with genotype C1C2 had higher CP than chickens with genotype C1C1 and C2C2 (P < 0.01). The SEP of genotype C2C2 was lower than C1C1 (P < 0.05), but no difference with C1C2.

Table 3 The GLM analysis of association between the chicken carcass traits and MEF2A gene SNPs
Full size table

Construction of haplotypes and their association with chicken carcass traits

Table 4 shows all haplotypes that were reconstructed from the three SNPs identified in all 471 experimental birds. Eight haplotypes, with the minor frequencies above 2%, were identified. Two main haplotypes, H1 (CGG) and H2 (CGT), accounted for 71.27% of the observations. Table 5 shows the frequencies of haplotypes in each population. The haplotype H1 (CGG) was predominant due to its higher frequency than other haplotypes in all the four populations. Haplotype H8 (TAT) was not detected in SD chickens. Only haplotypes H1 (CGG), H2 (CGT), H4 (CAT) and H5 (TGG) were found in AA chickens, and the two main haplotypes H1and H2, accounted for 93.33% of the observations in AA chickens. Twenty-four diplotypes were obtained based on these 8 haplotypes. Among them, 17 diplotypes had frequency higher than 1% and totally accounted for 96.78% (Table 6). The mixed model analysis indicated that there was significant association of diplotypes with carcass traits (Table 7). Diplotypes had great significantly effect on were associated with BW (P < 0.0001), CW (P < 0.0001), SEW (P = 0.0216), BW (P = 0.0199), LW (P < 0.0001), and CP (P = 0.0155). Significantly dominant effects of diplotypes H1H2 were observed for traits BW, CW, SEW, BMW, and CP, whereas H5H5 had a negative effect on BW, CW, SEW, BMW and LMW.

Table 4 Haplotypes inferred based on the three single nucleotide polymorphisms in the MEF2A gene
Full size table
Table 5 The frequency of each haplotype in each strain in the MEF2A gene
Full size table
Table 6 Diplotypes of chicken MEF2A gene
Full size table
Table 7 Associations between diplotypes and the chicken carcass traits
Full size table

Discussion

Many important traits of domestic animals are controlled by multiple genes and complex gene interaction. The study of candidate genes can be useful to determine whether specific genes are related to the economic traits [30]. The MEF2 proteins are involved in regulation of many muscle specific genes and play active roles in myogenesis, proliferation, and differentiation [8, 31]. However, much of the available information about this gene was taken from studies on humans and might not be directly applicable to poultry.

The success of poultry meat production has been strongly related to improvements in growth and carcass yield, mainly by increasing breast proportion [32]. Uncovering the molecular mechanism of muscle traits will help with more efficient selection and breeding in broiler chickens. In the current study, we screened the SNP of MEF2A gene in four chicken populations and identified three SNPs. We can see that chickens with genotype A2A2 in locus A and B1B1 in locus B had higher carcass value than those with other genotypes, and although they were predominant in all the four strains, but the frequencies of them in AA chickens were highest. As we all know, AA chickens was a international well-known breed with fast growth speed, the frequencies of the favorable genotypes might be changed during the high-strength for carcass traits, and the result also indicated that the selection for carcass traits of DH, SD and SH chickens didn’t go far enough, and efforts should be strengthened. These polymorphisms were significantly associated with BW, CW, SEW, EW, BMW, LMW, CP, SEP, and LMP, respectively. This result indicated that MEF2A gene was a good candidate gene for selection to increase skeletal growth, breast muscle yield, and leg muscle yield in chickens, and supported the notion that MEF2 proteins play key roles in the regulation of muscle-specific transcription and control of myogenesis.

As a traditional approach for studying both trait association (marker vs. trait) and linkage disequilibrium (marker vs. marker), single-marker analysis has created many problems, such as noisy, unsatisfied, and obscured important localization information [33]. Haplotype or haplotype block reconstruction was much more powerful than marker-by-marker analysis and provided a practical solution to resolve these problems [33, 34]. In this study, the H1H2 and H1H1 diplotypes presented higher BW, CW, SEW, BMW, LMW and CP than all the other haplotype combinations. Therefore, H1 may be the most advantageous haplotype for carcass traits. The H5H5 diplotype was shown to produce less meat than the other haplotype combinations, which suggested that H5 haplotype may play a negative effect on muscle traits. The data also showed that haplotypes had not significantly effect on EW, LMP and SEP (P > 0.05), while genotypes of three SNPs were found have great significantly effect on them (P < 0.05), respectively. About BW, CW, SEW, BMW, LMW and CP, both the haplotypes and the genotypes of three SNPs had great significantly effect on them. So, that may indicate that associations of haplotypes with phenotypic traits were more accurate than those of single SNP. Thus, it was observed that haplotype diversity is preferred over one SNP, and the method of SNP selection based on maximizing haplotype diversity is preferred [35–37].

In summary, each of the three SNPs was significantly associated with two or five carcass traits in chickens. The results implied that the MEF2A gene may be important in muscle traits in chickens. The MEF2A gene, therefore, is a potential marker for molecular marker-assisted selection of carcass traits in chickens. The work presented here is just a preliminary to a long period of research into the effects of MEF2 proteins family for chicken body composition. So, further research for associations between other family members of MEF2 and carcass traits and function studies of defining the effect of SNP of cMEF2A gene should be carried out.