Activin receptor 2A and activin receptor 2B genes in chicken: effect on carcass traits

ABSTRACT

Carcass traits are of immense importance in chicken as some parts of the carcass particularly breast muscle and legs are mostly preferred by the consumers with good market price while the remaining parts have a lower price. Carcass traits primarily depend on the growth of the birds, which are controlled by the candidate genes. Activin receptor type 2A and 2B act as receptors for binding with the members of transforming growth factor superfamily like myostatin to expedite its biological functions. We analysed exon2 and exon4 of activin receptor type 2A and 2B genes in six chicken populations. Both the genes revealed the presence of four haplotypes in these chicken populations. Association studies revealed a significant effect of genotypes and haplotypes on certain carcass traits such as carcass weight, dressing %, back and neck weight, giblet weights, etc. It is concluded that the exons of ACVR2A and ACVR2B genes were polymorphic and potentially associated with certain carcass traits in chicken.

KEYWORDS:

• ACVR2A
• ACVR2B
• effect
• carcass traits

1. Introduction

Growth is a complex mechanism of metabolic and physiological effects derived from cross-talk among genetic, nutritional and environmental factors. Growth of an animal or its organs/tissues is directly under the control of a coordinated and integrated system, on which understanding of the fact is very limited (Conlon & Raff 1999). The size of an organ/tissue depends on the number and size of its cells and extracellular components. Several factors are directly/indirectly involved in deciding the number and size of muscular cells. Myostatin (MSTN) also known as growth differentiation factor-8, a member of the decapentaplegic-Vg-related subfamily belonging to transforming growth factor beta (TGF-β) superfamily, is one of them. It is demonstrated that MSTN plays a critical role in regulating muscle homeostasis postnatally by suppressing muscle growth (Costelli et al. 2008). During embryogenesis, MSTN is exclusively expressed in the skeletal muscle to control differentiation and proliferation of the myoblast, but in adulthood, it is found in the skeletal muscle and heart, adipose tissue, etc. to have a negative effect on tissue growth (McPherron et al. 1997; Allen et al. 2008).

MSTN is an extracellular cytokine that mediates signal by binding a cell-bound receptor called the activin type II receptor (Lee et al. 2005). Once activated, MSTN has high affinity for the activin IIB receptor (ACVR2B, also known as ActRIIB) and weak affinity for activin IIA receptor (ACVR2A also known as ActRIIA), both of which, like other receptors for TGF-β family members, bind multiple ligands (Lee et al. 2005). Activin receptors are transmembrane threonine/serine kinases classified into two groups such as activin type 1 receptors and activin type 2 receptors. The activin type 2 receptors modulate signals for ligands belonging to the TGF-β superfamily, which includes activin (or inhibin), bone morphogenetic proteins, MSTN, Nodal, etc. There are two activin type 2 receptors, that is, ACVR2A and ACVR2B. Several ligands that signal through the activin type II receptors regulate muscle growth. The natural defect in ActRIIB sensitivity in humans leads to a significant increase in muscle mass. The same effect was observed in the experiments that used blockade of the murine receptor (Lee et al. 2005). In an experiment, healthy mice treated with sActRIIB showed a 60% increase in muscle mass in 2 weeks after treatment. Surprisingly, in mice treated with antiMSTN antibodies or MSTN propeptide known to bind circulating MSTN, the hypertrophy of muscle was lower compared with sActRIIB (Hill et al. 2002, 2003). Moreover, in MSTN knockout mice treated with the sActRIIB, an increase in muscle mass by 15–25% was observed (Lee et al. 2005). This suggests the presence of at least one more ligand that binds to the receptor thereby regulating the mass of skeletal muscle. Souza et al. proposed that such ligands could embrace BMP11, activin A, B and AB, but this hypothesis requires more evidence (Souza et al. 2008). However, it could explain why MSTN blockade in clinical trials had no significant effect on muscle weight (Wagner et al. 2008). Keeping all these facts in view, the objectives of the study were designed to explore single nucleotide polymorphisms (SNPs) in the exon2 and exon4 of these two genes and to estimate the effect of the polymorphism on carcass traits in chicken.

2. Materials and methods

2.1. Experimental birds and husbandry practices

The study was conducted in six chicken population, namely PB-1, PB-2, CB, PD-1, layer control and Aseel maintained at the Institute farm, Hyderabad, India. The PB-1, PB-2, CB and PD-1 lines were fast-growing lines, while layer control and Aseel are slow-growing lines. The fast-growing lines reveal higher body weight, while slow-growing lines depict low body weight. To obtain higher variability in the growth and carcass traits for analysing the haplotype–trait relationship, both groups of birds were included in the study.

The PB-1 was a synthetic colour broiler line selected initially for body weight at 6 weeks and later on body weight at 5 weeks for last 18 generations. The overall body weight of PB-1 at 5 weeks of age was 924.3 ± 0.12 g (PDP Annual Report, 2009). The PB-2 was a synthetic colour broiler female line. The control broiler (CB) line was a synthetic colour broiler line which is random-bred pedigreed over last nine generations. The body weight of CB line at 5 weeks of age was 625.5 ± 0.13 g (PDP Annual Report, 2010). The layer control was a random-bred control population was used as control to estimate genetic progress in the selected layer lines. The PD-1 line was a broiler line evaluated for growth traits over last seven generations. The body weight of the birds at 6 weeks of age was 667 g. The Aseel is a native chicken breed of India. The body weight of the birds at 8 and 16 weeks of age were 456 and 1188 g. These birds were well adapted to Indian agro-climatic conditions.

The broiler birds were kept in the brooder house on the deep litter system till the age of six weeks providing ad-lib feeding and watering and then shifted to the grower house. All the layer birds were reared in the deep litter system upto 18 weeks by providing ad lib feeding and watering. All the birds were hatched at the same time and housed all along in the shed at the age of 18 weeks. The PB-1, PB-2, CB and PD-1 chicks upto 3 weeks were fed with 2800 kcal ME and 21% crude protein while these birds from 3 to 6 weeks along with layer control and Aseel chicks upto 6 weeks were fed with 2600 kcal ME and 16% crude protein. During the brooding stage, the temperature was maintained at 90°F during the firstt week with a weekly gradual decrease of 5°F up to the fifth week. A proper vaccination schedule was followed for all the birds.

2.2. Sample collection and isolation

Blood samples were collected from 20 birds each from each of 6 chicken lines. A volume of 50 μl blood was aseptically collected from which genomic DNA was isolated following the standard protocol (Sambrook & Russel 2001). The quality of DNA was verified by 0.8% agarose gel electrophoresis, while spectrophotometric analysis was done to quantify the genomic DNA of each sample.

2.3. Polymerase chain reaction and single-stranded conformation polymorphism (SSCP)

Activin receptor type 2A comprises 512 amino acids of which the first 19 amino acids are signal peptide and the remaining 493 amino acids act as mature peptide, which is the functional part of the protein (http://genatlas.medecine.univ-paris5.fr) that bind with the ligands. The exon2 and exon4 of the gene are involved in the formation of the intra-cellular domain on which the ligand works to exert biological functions. Activin receptor 2B consisting of the first 24 amino acids is the signal peptide and the remaining 487 amino acids is the mature peptide and contains a functional putative 23 residue membrane-spanning region between 140 and 162 residues formed by exon2 and exon4 of the gene (Funkenstein et al. 2012). The primers used for the amplification of 2 exons (exon2 and exon4) of ACVR2A gene were designed from the chicken ACVR2A sequence (GenBank accession no. NC_006094) using DNASTAR software. The primers were designed from the chicken sequence available at NCBI (GenBank accession no. NC_006089) to amplify exon2 and exon4 of the ACVR2B gene (Table 1). The PCR reaction was set up with 50 µg of DNA template, 10 ng of each primer, 1.5 mM of MgCl₂, 100 µM of each dNTP, 1× assay buffer and 0.25 U of Taq DNA polymerase (Fermentas, Germany). The PCR was performed with initial denaturation at 94°C for 5 min, 30 cycles of denaturation at 94°C for 30 s, annealing at specific temperature (50°C for exon2 of ACVR2A, 48°C for exon4 of ACVR2A, 54°C for exon2 of ACVR2B and 48°C for exon4 of ACVR2B genes) and extension at 72°C for 30 s with a final extension at 72°C for 10 min (Table 2).

Table 1. Primer details used for amplification of ACVR2A and ACVR2B genes.

Oligo name	5′<-----Sequence----->3′	Size (bp)	Fragment	Tm (°C)
	ACVR2A
ACVR2AE2F	GTGCCATTCTTGGCAGATCAG	208	Exon 2	50
ACVR2AE2R	CTGTCGTAACAGTTGATG
ACVR2AE4F	CAACTTCCAATCCAGTCAC	155	Exon 4	48
ACVR2AE4R	TTGGGTTGGAACAAGTACTG
	ACVR2B
ACVR2BE2F	GCTTCAGGACCAGGTCAC	215	Exon 2	54
ACVR2BE2R	CTGTCATAACAGTTGAAGTC
ACVR2BE4F	TCATATATGAGCCGCCAC	152	Exon 4	48
ACVR2BE4R	CTCATTGATGTCAACGTGC

Table 2. Haplogroups and haplotype frequency of ACVR2A and ACVR2B genes in different lines of chicken.

Haplogroups/haplotypes	Lines
	CB	Layer control	PB-1	PB-2	PD-1	Aseel
	ACTVNR2A
h1h1	0.57	0.464	0.184	0.333	0.095	0.333
h1h2	0.107	0.230	0.136	0.085	0.095	0.041
h1h3	0.250	0.230	0.590	0.166	0.572	0.500
h1h4	0.071	0.076	0.090	0.416	0.238	0.126
h1	0.786	0.732	0.592	0.667	0.548	0.667
h2	0.054	0.115	0.068	0.043	0.048	0.021
h3	0.125	0.115	0.295	0.083	0.286	0.250
h4	0.036	0.038	0.045	0.208	0.119	0.063
	ACTVNR2B
h1h1	0.666	0.518	0.572	0.519	0.333	0.571
h1h2	0.034	0.297	0.250	0.259	0.043	0.108
h1h3	0.200	0.148	0.107	0.111	0.166	0.250
h1h4	0.100	0.037	0.071	0.111	0.458	0.071
h1	0.833	0.759	0.786	0.760	0.667	0.786
h2	0.017	0.149	0.125	0.130	0.022	0.054
h3	0.100	0.074	0.054	0.056	0.083	0.125
h4	0.050	0.019	0.036	0.056	0.229	0.036

A 12% native poly acrylamide gel electrophoresis (PAGE) (50:1, acrylamide and bis-acrylamide) with 5% glycerol was prepared to resolve the SSCP pattern following the standard protocol (Vohra et al. 2006). Furthermore, 3 µl of PCR product was mixed with 15 µl formamide dye [95% formamide, 0.025% xylene cyanol, 0.025% bromophenol blue, 0.5 M ethylene diamine tetra-acetic acid (EDTA)] was denatured at 95°C for 5 min followed by snap cooling on ice for 15 min. Then, the product was loaded in the gel and electrophoresis was performed at 4°C for 12 h at 200 V followed by staining with silver nitrate to visualize banding patterns.

2.4. Sequencing and haplotypes

Three PCR products amplified from each SSCP variant of both exons of ACVR2A and ACVR2B genes, derived with HotStar HiFidelity DNA Polymerase (Farmentas, USA), were sequenced with the fragment-specific primers from both ends by the automated dye-terminator cycle sequencing method in ABI PRIZM 377 DNA Sequencer (Perkin-Elmer, USA).

Haplotypes were constructed by combining SSCP patterns of all the fragments of both ACVR2A and ACVR2B genes of each individual bird. Haplotypes in the diploid state of an individual reveal the haplotype combinations of that animal. Haplotype sequences were analysed with DNASTAR Software (Lasergene Inc.). Frequencies of haplotype and its combinations were calculated by the gene counting method.

2.5. RT-PCR and qPCR

The total RNA was isolated from the pectoral muscle by following the TRIZOL method (Invitrogen). First-strand cDNA was synthesized from total RNA with reverse transcriptase enzyme. The qPCR was performed for ACVR2A and ACVR2B genes along with the GAPDH gene as the internal control with the cDNA templates in the thermal cycler Stratagene Mx3000P machine with Platinum® SYBR® Green qPCR UDG supermix (Invitrogen). A pair of primers for ACVR2A gene, namely ACVR2AE4F: 5′-CAACTTCCAATCCAGTCAC-3′ and ACVR2AE4R:5′-TTGGGTTGGAACAAGTACTG-3′, was designed from the chicken cDNA sequences of the ACVR2A gene (GenBank accession no. NC_006094) with DNASTAR software (LasergeneInc) for the qPCR study. The 155 bp fragment of this gene was amplified at 57°C annealing temperature. For amplification of the ACVR2B gene, a pair of primers, namely ACVR2BE4F:5′-TCATATATGAGCCGCCAC-3′ and ACVR2BE4R: 5′-CTCATTGATGTCAACGTGC-3′, was designed from the NCBI sequence available (GenBank accession no. NC_006089) where the fragment size was 152 with the optimum annealing temperature of 57°C. A fragment of 119 bp of GAPDH gene was amplified at 57°C using a pair of primers, namely QGAPDHF: 5′-CTGCCGTCCTCTCTGGC-3′ and QGAPDHR: 5′-GACAGTGCCCTTGAAGTGT-3′ designed from the chicken GAPDH sequence (accession no. AF047874) with DNASTAR software (Lasergene Inc.). Reactions were prepared in triplicates with a final volume of 25 µl containing 12.5 µl of Platinum® SYBR® Green qPCRsupermix, 0.5 µl of ROX reference dye, 0.2 µM of each primer and 2 µl of cDNA. The qPCR conditions were: initial denaturation at 95°C for 10 min, 40 cycles of denaturation at 95°C for 30 s, annealing at 57°C for 1 min and extension at 72°C for 30 s. Following amplification, a dissociation melting curve analysis was conducted by programming the PCR machine from 55°C to 95°C to detect possible non-specific products. Fluorescence threshold was determined by the default method at 32.5% with Stratagene software for Mx3000P real-time PCR machine. Ct values of each sample were noted and average Ct values of each sample generated in duplicate qPCR reactions were used in calculating delta Ct and fold change (fold change = 2^−ΔΔCt) of gene expression at different ages of birds.

2.6. Traits

Twenty birds each (10 males and 10 females) of each line were slaughtered at the age of 6 weeks and different parts along with giblets were subjected to weighing for recording the weight of the slaughter parts. For this study, depending on the available resources, 80 birds from the fast-growing group and 40 birds from the slow-growing groups (20 birds from layer control and 20 birds from Aseel) were randomly taken for the study. Consequently, carcass traits (dressing % and slaughter parameters) were estimated. Slaughter of birds was performed following the standard protocol of cervical dislocation approved by the Institutional Animal Ethics Committee (IAEC). The animal welfare measures like adlib feeding, watering and management on the floor by providing required space during the experiment were followed.

2.7. Statistical analysis

The frequencies of haplotype and haplogroup/haplogroups were estimated by the gene counting method. The association of haplotype combinations with carcass traits was explored following the general linear model technique using SPSS software where genotype, haplotype, sex and line were used as fixed effects and sire as random effect in the following model:where µ is the overall mean, S_i is the ith sire effect, L_j is the jth line effect, HPL_k is the kth haplogroup effect, X_m is the mth sex effect, L_jX HPL_k is the interaction effect between jth line and kth haplogroup, and e_ijklm is the random error with NID (0, σ²e). The effects of gene expressions on carcass traits were also estimated by the linear regression technique using SPSS software.

3. Results

3.1. Polymorphism

The exon2 of the ACVR2A gene revealed the presence of 2 genotypes (11 and 12) in different chicken lines. Likewise, exon4 of the gene had 2 genotypes (11 and 12) in all the chicken lines. These genotypes were used for preparing haplotypes and haplogroups. Accordingly a total of four haplogroups were found in all the lines in which the h1h1 haplogroup was the most frequent one in CB and layer control lines. The h1h3 haplogroup was found with the highest frequency in PB-1, PD-1 and Aseel lines, whereas h1h4 was the most frequent in the PB-2 line (Table 2). Consequently, there were four haplotypes present in the lines in which h1 was the most predominant haplotype.

The exon2 and exon4 of ACVR2B gene revealed two genotypes in different chicken lines. In all the populations, a total of four haplogroups were found of which h1h1 was the most predominant one in all the lines except PD-1 where h1h4 had the highest frequency. Consequently, a total of four haplotypes were found in all the populations. The haplotype h1 was the most frequent one in all the lines having frequency in the range of 0.67–0.83.

3.2. Haplotype variability

The sequences of exon2 and exon4 of the ACVR2A gene were combined to prepare haplotypes and haplogrpoups. For this gene, a total length of 363 bp was considered where nucleotide differences were studied. In the exon2 fragment, total nine changes of nucleotides were observed at different locations. The nucleotide substitutions were found in position 6:T>T>A>A, 28:A>A>C>C, 29:A>A>C>C, 62:G>G>A>A, 146:T>T>C>C, 171:C>C>A>A, 183:G>G>C>C, 205:C>C>A>A and 206:G>G>A>A of h1, h2, h3 and h4 haplotypes, respectively. In the exon4 region of the haplotype, there was only one nucleotide substitution at position 219:G>A>G>A in h1, h2, h3 and h4 haplotypes, respectively, which was of transtitonal type. As far as sequence similarities are concerned, h1 haplotype had 99.7%, 97.5% and 97.2% similarities with h2, h3 and h4 haplotypes whereas h2 haplotype had 97.2% and 97.5% similarities with h3 and h4 haplotypes. The h3 haplotype had 99.7% similarity with h4 haplotype. As far as the protein sequence is concerned, amino acid change was found at 9th, 21st, 57th, 61st, 65th and 105th positions of the haplotype protein sequence, which entails differential structures of the functional part of the protein. However, all the haplotype sequences have been submitted to the NCBI GenBank and the accession numbers were obtained as KF583559, KF583560, KF583561 and KF583562 for h1, h2, h3 and h4 haplotypes, respectively.

The exon2 fragment of the ACVR2B gene in the haplotype alignment study revealed nucleotide change at 2 positions, that is, on 187:T>T>C>C and 189:G>G>A>A in h1, h2, h3 and h4 haplotypes. At both the locations, the mutations were of transitional type. In exon4, there was presence of nucleotide substitutions at 20 locations. The nucleotide substitutions were observed in 222:A>TA>T, 230:C>T>C>T, 242:A>C>A>C, 251:G>C>G>C, 280:C>G>C>G, 283:T>C>T>C, 308:T>G>T>G, 318:G>A>G>A, 321:T>A>T>A, 335:G>A>G>A, 338:C>G>C>G, 340:T>G>T>G, 341:C>G>C>G, 342: C>G>C>G, 351:A>G>A>G, 353:G>T>G>T, 356:G>A>G>A, 358:C>T>C>T, 363:T>A>T>A and 364:T>C>T>C of h1, h2, h3 and h4 haplotypes, respectively. The amino acid changes were observed at 63rd, 73rd, 80th, 83rd, 102nd, 106th, 111st, 112nd, 113rd, 116th, 117th, 118th and 120th position of the haplotype (protein composition). However, all the haplotype sequences of the gene have been submitted to the NCBI GenBank and the accession numbers were obtained as KF583564, KF583565, KF583566 and KF583567 for h1, h2, h3 and h4 haplotypes, respectively.

3.3. mRNA expression

The expression of ACVR2A gene in the skeletal muscle varied among the haplogroups irrespective of lines. The highest expression was observed in the h1h3 haplogroup, whereas the lowest expression was found in the h1h2 group (Table 3). The trend of expression was h1h3>h1h4>h1h1>h1h2. The h1h3 haplogroup showed 11.4 fold higher expression of the ACVR2A gene in muscle as compared with the h1h2 haplogroup (Table 4). This group also had 3.4 and 4 fold increase in expression as compared with h1h4 and h1h1 haplogroups. In the case of the ACVR2B gene, the highest expression of this gene was found in the h1h2 haplogroup while the lowest expression was observed in the h1h3 haplogroup (Table 3). The trend of gene expression was h1h2>h1h4>h1h1>h1h3. The h1h2 haplogroup determined 4.9 fold higher expression than the h1h3 haplgroup while with h1h4 and h1h1 haplogroups, the h1h2 haplogroup revealed 0.7 and 2.5 fold higher expression in muscle (Table 4). Line-wise expression profile revealed that broiler populations showed the highest expression of both ACVR2A and 2B gene in the skeletal muscle, whereas layer line showed the lowest expression of ACVR2A gene and indigenous line had the lowest expression of the ACVR2B gene. The levels of expression of the ACVR2A gene among lines were not significantly different, but that of the ACVR2B gene differed significantly (P < .05) among those lines.

Table 3. Expression profile (40-Delta ct) among haplogroups and populations in ACVR2A and ACVR2B genes (n = sample size).

Haplogroup	n	ACVR2A	ACVR2B
h1h1	23	28.70 ± 0.85^ab	27.83 ± 1.04^ab
h1h2	21	27.20 ± 0.91^b	29.10 ± 1.46^a
h1h3	53	30.72 ± 0.64^a	26.83 ± 1.17^b
h1h4	23	28.98 ± 0.89^ab	28.71 ± 1.35^ab
Population
Broiler	80	29.8 ± 0.39	29.9 ± 0.34^a
Layer	20	27.7 ± 0.76	25.8 ± 0.53^b
Indigenous	20	29.4 ± 0.48	22.8 ± 0.66^c

Note: Different superscripts indicate significance at P < .05.

Table 4. Fold change among the haplogroups of ACVR2A and ACVR2B genes.

	ACVR2A				ACVR2B
ACVR2A Haplogroups	h1h3	h1h4	h1h2	ACVR2B Haplogroups	h1h1	h1h4	h1h2
h1h2	11.4	3.4	–	h1h3	2.0	3.7	4.9
h1h1	4.0	1.2	0.3	h1h1	–	1.8	2.5
h1h3	–	0.3	0.1	h1h2	–	0.7	–

3.4. Association of ACVR2A and 2B polymorphism with cut of parts

The haplogroups of ACVR2A gene showed significant weight of leg and back cut (Tables 5, 6). The h1h4 haplogroup had the highest leg weight while h1h2 had the lowest weight. The h1h4 haplogroup revealed 76% higher leg weight than h1h2 haplogroup. The haplogroup-wise performance trend was h1h4>h1h3>h1h1>h1h2. In the case of back cut, the h1h2 haplogroup showed the highest weight, whereas the h1h4 haplogroup had the lowest weight. The h1h2 haplogroup determined 207% higher weight of back cut than the h1h4 haplogroup.

Table 5. Effect of ACVR2A and ACVR2B haplogroups on cut of parts in chicken.

Variants	Parameter	Dressed carcass wt (g)	Dressing %	Neck wt (g)	Leg wt (g)	Wing wt (g)	Breast wt (g)	Back cut (g)
	ACVR2A
h1h1	Mean	600.17 ± 321.49	66.47 ± 8.50	24.47 ± 13.36	99.57 ± 73.93^a	57.93 ± 29.41	92.37 ± 57.50	79.28 ± 70.46^c
h1h2	Mean	608.72 ± 295.50	67.96 ± 8.00	24.38 ± 11.94	88.40 ± 65.15^a	60.22 ± 29.47	102.48 ± 61.70	101.68 ± 78.00^d
h1h3	Mean	593.68 ± 292.52	69.15 ± 8.47	26.46 ± 14.71	134.78 ± 81.37^b	56.66 ± 27.22	89.34 ± 51.59	69.92 ± 71.27^b
h1h4	Mean	680.03 ± 251.68	65.73 ± 7.49	28.61 ± 14.47	155.63 ± 70.13^c	61.71 ± 22.66	103.79 ± 45.17	33.07 ± 51.21^a
	ACVR2B
h1h1	Mean	642.25 ± 296.87^c	67.32 ± 7.45^ab	26.22 ± 14.23^b	120.87 ± 86.03	61.67 ± 28.30	98.29 ± 54.45	82.84 ± 72.47
h1h2	Mean	600.99 ± 317.19^b	72.30 ± 6.39^b	28.14 ± 17.65^bc	148.55 ± 86.61	57.67 ± 28.54	93.25 ± 56.68	63.72 ± 64.47
h1h3	Mean	545.12 ± 292.15^a	68.51 ± 8.66^ab	23.11 ± 10.77^a	98.54 ± 67.02	52.21 ± 26.94	87.83 ± 57.56	72.18 ± 72.59
h1h4	Mean	679.19 ± 219.62^c	63.09 ± 7.84^a	28.00 ± 10.97^bc	134.84 ± 53.24	62.88 ± 21.20	102.92 ± 40.03	52.43 ± 77.30

Note: Column-wise different superscripts indicate significance at P < .05.

The ACVR2B haplogroups showed a significant effect on dressed carcass weight, dressing % and neck weight (Table 5). The highest dressed carcass weight was found in birds with the h1h4 haplogroup. The h1h3 haplogroup had the lowest dressed carcass weight. The h1h4 haplogroup revealed 24.5% higher dressed weight than the h1h3 haplogroup. The highest dressing % was observed in animals having the h1h2 haplogroup, while the h1h4 haplogroup showed the lowest dressing %. The h1h2 haplogroup revealed 13.4% higher dressing % than the h1h4 haplogroup. The highest neck weight was found in animals possessing h1h2 and h1h4 haplogroups, while the lowest performance was found in animals with the h1h3 haplogroup. The h1h2 and h1h4 haplogroups determined 21.1% higher neck weight than the h1h3 haplogroup.

3.5. Association of ACVR2A and 2B polymorphism with giblets

The haplogroups of ACVR2A gene had a significant (P < .05) effect on the weight of giblets (Table 7). The h1h4 haplogroup showed the highest weight of head (32.37 ± 9.21 g), while the h1h1 haplogroup had the lowest head weight. The h1h4 group revealed 15.27% higher head weight than the h1h1 group in chicken. The haplogroup also showed a significant effect (P < .05) on weight of gizzard where h1h4 and h1h3 haplogroups had the highest and lowest weight, respectively. The h1h4 haplogroup showed 23.7% higher gizzard weight than the h1h3 group.

Table 6. Effect of expressions of ACVR2A and ACVR2B genes on carcass traits.

	ACVR2A			ACVR2B
Trait	Intercept	Regression coefficient	P	Intercept	Regression coefficient	P
6 weeks body wt	736 ± 75.9	−3.8 ± 6.5	.561	921 ± 56.6	−18.81 ± 4.13*	.000
Dressed carcass wt	654 ± 67.2	−3.7 ± 5.8	.519	814 ± 50.1	−16.71 ± 3.66*	.000
Dressing %	73.7 ± 1.8	−0.56 ± 0.15*	.001	72.5 ± 1.46	−0.39 ± 0.10*	.000
Neck wt	25.2 ± 3.2	0.07 ± 0.27	.778	31.5 ± 2.54	−0.45 ± 0.18*	.000
Leg wt	193.7 ± 16.7	−6.57 ± 1.45*	.000	211 ± 11.4	−7.28 ± 0.83*	.000
Wing wt	57.4 ± 6.2	0.09 ± 0.54*	.000	74.13 ± 4.79	−1.31 ± 0.35*	.000
Breast wt	105 ± 12.3	−0.97 ± 1.06*	.000	135 ± 9.04	−3.37 ± 0.66*	.000
Back cut wt	85.1 ± 16.5	−1.4 ± 1.43*	.000	11.34 ± 12.6	−3.60 ± 0.92*	.000

Note: *Significant.

Table 7. Effect of ACVR2A and ACVR2B haplogroups on giblets in chicken.

Variants	Parameter	Head (g)	Gizzard (g)	Liver (g)	Spleen (g)	Bursa (g)	Heart (g)
	ACTVNR2A
h1h1	Mean	28.08 ± 10.20^a	30.17 ± 16.20^a	19.08 ± 12.27	1.33 ± 0.81	0.61 ± 0.50	4.44 ± 1.99
h1h2	Mean	29.98 ± 11.14^a	30.60 ± 13.0^a 5	18.06 ± 8.40	1.45 ± 0.84	0.63 ± 0.37	4.04 ± 2.14
h1h3	Mean	29.85 ± 11.63^a	28.66 ± 14.24^a	18.77 ± 11.19	1.48 ± 0.88	0.54 ± 0.37	4.37 ± 1.50
h1h4	Mean	32.37 ± 9.21^b	35.45 ± 13.62^b	25.98 ± 14.08	1.74 ± 0.78	0.56 ± 0.37	4.48 ± 1.26
	ACTVNR2B
h1h1	Mean	30.51 ± 11.08	30.86 ± 14.36	20.08 ± 12.38	1.57 ± 0.78	0.61 ± 0.39	4.43 ± 1.66
h1h2	Mean	29.97 ± 11.43	30.65 ± 16.13	21.95 ± 13.98	1.50 ± 1.08	0.53 ± 0.33	4.13 ± 1.85
h1h3	Mean	27.19 ± 11.13	27.22 ± 13.11	17.25 ± 10.49	1.43 ± 0.80	0.55 ± 0.43	4.34 ± 1.76
h1h4	Mean	33.29 ± 8.18	34.93 ± 13.16	21.98 ± 8.72	1.49 ± 0.68	0.60 ± 0.43	4.57 ± 1.36

Note: Column-wise different superscripts indicate significance at P < .05.

The haplogroups of ACVR2B gene did not show any significant effect on weight of giblets in chicken. However, haplogroups of this gene showed a positive trend of effect on gizzard weight where the h1h4 haplogroup had the highest weight and h1h3 had the lowest weight. The h1h4 group had 28.3% higher gizzard weight than h1h3 group.

4. Discussion

Activin receptor 2A and activin receptor 2B exons encode proteins forming the receptors for binding members of TGF-β superfamily. MSTN as one of the important members of TGF-β superfamily binds with both activin receptor 2A and 2B with different intensities for exerting its biological function. The binding intensity of MSTN is relatively greater with the 2B receptor as compared with the 2A receptor. These receptors are locally present on the cell membrane of various tissues including the skeletal muscle. The function of MSTN depends on the number and type of active receptor sites which is primarily controlled at the level of expression as well as nucleotide constitution of the genes. In this study, we explored the nucleotide constitution of the genes in the form of SNPs in different chicken population. We studied four broiler populations, namely PB-1, PB-2, CB and PD-1, one layer population, namely layer control and one indigenous population namely, Aseel. There was no clear trend of allelic and haplotypic distribution among broiler, layer and indigenous populations. The differences of the distribution of haplotypes among populations were statistically non-significant. The genetic background of the populations although were different like broiler, layer and indigenous types, the pattern of haplotypic distribution did not vary too much. Earlier reports revealed that most of the quantitative trait locus for growth traits are located on chromosomes1, 2, 4, 5, 7, 8, 13 and 14 (Zhou et al. 2006) and in our study, we confined analysis on chromosomes 2 and 7 where activin receptor type 2B and activin receptor type 2A were located, respectively.

The haplotypes of the receptors vary in nucleotide and protein constitutions of the genes, which depict differential functional activities with respect to binding to the proteins. In the study, we found substitutions of the nucleotides at various positions of the alleles and haplotypes. Thus, haplogroups determining different protein structures and their concentrations on the cells have been explored through gene expression. In the past, candidate gene approach for polymorphism and association with carcass traits were studied by the workers who revealed SNPs in intron8 and exon10 of CAPN3 gene and the effect of polymorphism on carcass traits was explored in Chinese chicken population (Zhang et al. 2012). Likewise, researchers determined polymorphism of the TGFBR1 gene and its association with carcass traits in pig (Chen et al. 2012). Level of expression varied among the populations, thus clearly denoting differential concentrations of receptors in different lines. The activity of MSTN would vary among lines and ultimately affecting the growth patterns have become the pivotal factors for regulating growth rate in broiler, layer and indigenous lines. Earlier study on MSTN polymorphism and its association with growth traits were reported in broiler and layer chicken with the presence of total 13 haplotypes and significant effect on body weight at 14 and 49 days of age (Bhattacharya & Chatterjee 2013; Paswan et al. 2014). While looking at the expression of ACVR2A and ACVR2B genes in muscle, the trend of ACVR2A expression was h1h2>h1h4>h1h1>h1h3, while that of ACVR2B gene expression was h1h2>h1h4>h1h1>h1h3. Thus, differential expression of haplogroups might be one of major factor for controlling performances of carcass traits in chicken, which has been observed in association analysis of gene expression and haplogroups with carcass traits.

The expression patterns of activin receptor 2A and 2B were different in the different lines. The levels of expression for the ACVR2B gene were statistically different among broiler, layer and indigenous lines. For both ACVR2A and 2B genes, broiler expressed the lowest level while indigenous line was at the highest level for the ACVR2B gene, which reiterates the growth pattern in broiler, layer and indigenous lines. It is imperative that other factors also play a role for regulating growth, but MSTN being the negative regulator of growth limiting growth rate incurred by the positively regulatory molecules such as growth hormone and growth hormone receptor, IGF-1 and IGF-2, etc. However, MSTN working as a negative regulator directly controls growth rate through binding its receptors in broiler, layer and indigenous lines. Thus, activin receptor 2A and activin receptor 2B play a major role in exerting function of MSTN. Earlier studies revealed that the SNPs of TGF-β receptor1 and its association with growth (Guo et al. 2012). Other researchers also determined the SNPs of GH, GHR and calpastatin gene and their association with carcass traits in cattle (Schenkel et al. 2006; Tatsuda et al. 2008, Bahrami et al. 2012). The effect of haplogroups revealed the association pattern with carcass traits in chicken. Broiler meat in particular whole carcass is important for the consumers, but certain consumers prefer specific types of cut parts. Of whole carcass, breast meat is regarded as the principal component of chicken meat which reveals the highest meat and bone ratio while other parts contain lower meat and bone ratio. In this study, we explored the effect of polymorphism of activin receptor 2A and 2B on edible carcass such as cut of parts and giblets.

Haplogroups affect only back cut weight revealing the best performance by the h1h2 haplogroup. The presence of the h2 haplotype in the haplogroup performed best while other haplotype in collaboration with h1 haplotype did not perform well. If we look at the haplogroup performance, dressed weight, dressing % and neck weight were affected by the diplotypic variation. Haplogroup h1h4 performed the best for carcass weight, while h1h2 had the best dressing %. But, both the haplogroups showed the best neck weight. The presence of h2 and h4 haplotype in association with h1 haplotype performed better as compared with other combinations.

The association with giblets (head, gizzard, liver, spleen and bursa) showed a significant effect of haplogroups. The h1h4 haplogroup performed best for both head and gizzard weights, while other combinations performed with lesser potential. However, this haplogroup although affect most of the traits non-significantly, showed better performance than other haplogroups. The combination of h4 with h1 haplotypes remains the best performer than h1 haplotype combined with other haplotypes for giblet weights. Sometimes there was lack of significant association between traits and haplogroups while there was significant association between traits with genotypes. Haplotype is the combination of genotypes and hence analysing genotypes in a specific population shows higher variability among genotype groups due to the larger number of birds in each genotypic group. On the other hand, in the case of haplogroup, the number of birds within each haplogroup will be less with lesser variability. Hence, genotypes would show significant result while haplogroup would show non-significant association in the specific population. Our results suggest that the genotypes as well as haplogroups may be used as selection criteria for improvement of carcass traits. Before slaughtering the birds, by looking at haplogroups of the gene we may predict the probable performance of the carcass traits in birds. Thus, we can dispose the poor performing birds for carcass traits without killing them, which can economize the farming enterprise with better efficiency. We may suggest that polymorphism of genes may be used for characterization of populations and association study may be prioritized during planning and execution of breeding programme for chicken.

5. Conclusion

It may be concluded that both activin receptor type 2A and 2B genes were polymorphic having four haplotypes for each gene in different chicken populations. The polymorphisms had a significant effect on certain carcass traits in chicken.

Funding

Authors are thankful to Indian Council of Agricultural Research for providing financial support to carry out the research work under National Fellow project.