Molecular characterization and computational structure prediction of activin receptor type IIB in aseel and broiler chicken

Research in Veterinary Science 126 (2019) 139–149

Contents lists available at ScienceDirect

Research in Veterinary Science journal homepage: www.elsevier.com/locate/rvsc

Molecular characterization and computational structure prediction of activin receptor type IIB in aseel and broiler chicken

T

P. Guru Vishnua, , T.K. Bhattacharyab, Bharat Bhushanc, Pushpendra Kumarc, R.N. Chatterjeed, Chandan Paswand, A. Rajendra Prasadc, D. Divyae, K. Dushyanthe ⁎

a

Animal Biotechnology, Sri Venkateswara Veterinary University, Tirupathi, A.P., India Molecular Genetics & Breeding, ICAR-Directorate of Poultry Research, Hyderabad, India c Division of Animal Genetics & Breeding, Indian Veterinary Research Institute, Izatnagar, U.P., India d ICAR-Directorate of Poultry Research, Hyderabad, India e SRF, ICAR-Directorate of Poultry Research, Hyderabad, India b

ARTICLE INFO

ABSTRACT

Keywords: Characterization ACTRIIB Three dimensional structure Ramachandran plot Hydrogen bond plot Contact map

The present study was formulated to characterize and comprehend the molecular structural characteristics of ACTRIIB receptor in Aseel and control broiler (CB) populations. The full length coding sequence (1539 bp) of the receptor was amplified, cloned, sequenced and analyzed using bioinformatic tools. The physico chemical properties of protein and structural features like secondary structure, solvent accessibility and disorder regions were computed. The 3D structure was predicted by I-TASSER and evaluated by Ramachandran Plot and tools under Structural Analysis and Verification Server. The nucleotides differences between CB and Aseel were c. [156G > A; 210 T > C; 493C > T; c.520G > C; 665A > C; 686G > A; 937C > G; 1011A > C; 1130A > G; 1208 T > A; 1326 T > C; 1433 T > C]. The amino acid substitutions between CB and Aseel were p. [(Pro165Ser; Glu174Gln; Gln222Pro; Ser229Asn; His313Asp; Gln377Arg; Val403Asp; and Ile478Thr)]. While, the silent changes includes p. [(Lys53=; Glu71=; Leu337=; Asp442=)]. The molecular weight of mature protein was predicted to be 55.51 kDa and 57.80 kDa in Aseel and CB, respectively. The higher rank 3D model had a C-score of −1.60 in Aseel and − 1.41 in CB, while the estimated TM-score (0.54 ± 0.14) and RMSD (5.8 ± 1.2 Å) were found to be similar in Aseel and CB. Among the 512 residues, > 90% were in favored region, 4.7% in allowed region and < 1.5% in disallowed region in both Aseel and CB. The pattern of contact map was comparable in Aseel and CB. The Hydrogen bond plots of the Aseel and CB shared similar secondary structure pattern. The ACTRIIB protein was predicted to interact with ACVR1B, ACVR1C, INHBA, SMAD 1,2,5,7 & 9 and BMPR1A&B. Clustal and phylogenetic analysis implied that both the lines were closely related and formed a sub cluster with in avian cluster. The current research provides insights about structural and functional aspects of the receptor and also aids in understanding the evolutionary history of ACTRIIB.

1. Introduction Proteins constitute an important molecules in the biological systems and have diverse functions than other class of biological macromolecules. The function of protein is majorly governed by its threedimensional structure at the molecular level and can provide a sensitive probe of distant evolutionary relationships (Zhang, 2008; Lopez et al., 2007). One of the major challenges in the structural genomics and proteomics era is to predict the three-dimensional structure of proteins using either experimental or computational modeling methods (Sleator and Walsh, 2010; Moult et al., 2016). Protein structure prediction through either NMR spectroscopy or X⁎

ray crystallography will solve only limited proteins among the numerous proteins deposited in the protein database, due to requirement of more time and expensive processes for determining structures experimentally (Emerson and Amala, 2017; Dubey et al., 2018). Additionally, the genome sequencing projects are generating protein sequences at a very faster rate. Consequently, there exists a huge gap between the number of generated protein sequences and number of solved protein 3D structures. Computational prediction of 3D structures of proteins from sequences of amino acids is a promising and dynamic approach to bridge this gap (Yang et al., 2015). These predictions methods have great potential to predict the structures of newly deposited sequences based on homology modeling or threading by using

Corresponding author. E-mail addresses: [email protected] (P.G. Vishnu), [email protected] (P. Kumar), [email protected] (R.N. Chatterjee).

https://doi.org/10.1016/j.rvsc.2019.08.025 Received 30 April 2019; Received in revised form 19 July 2019; Accepted 27 August 2019 0034-5288/ © 2019 Elsevier Ltd. All rights reserved.

Research in Veterinary Science 126 (2019) 139–149

P.G. Vishnu, et al.

already existing protein sequences or structures, respectively. Muscle growth is considered as the most important breeding criteria among the conventional and commercial poultry breeders. Hence, molecular characterization of muscle growth regulating genes has become main research focus in post genomic era. A plethora of experiments carried out during the recent times have identified many ligands belonging to beta-transforming growth factor (TGF-β) group, having pivotal role in controlling muscle mass (Lee et al., 2005). The activin receptor type IIB categorize under transmembrane receptor group is considered to be important muscle growth regulator. The ACTRIIB regulate these pathways with varied affinity for different ligands such as high affinity for activin A, growth differentiation factor 11 (GDF11) and myostatin, while low affinity for BMP- 2, 7 and nodal (Sako et al., 2010). The transmembrane proteins (TMPs) have been reported to contribute about 20–30% of native proteins in proteomes of multicellular organisms (Krogh et al., 2001; Stevens and Arkin, 2000). Nevertheless, the number of three-dimensional (3D) structures of TMPs is far below one hundred at present, in contrast to > 18,000 3D structures available for soluble proteins registered in Protein Data Bank (Westbrook et al., 2002). Moreover, the literature emphasizing the structural characteristics of the ACTRIIB receptor in avian species is meagre, as far as our knowledge is concerned. Therefore, keeping in view of the intrigue role of the ACTRIIB receptor in muscle growth regulation, the present study is formulated to characterize and predict the three-dimensional (3D) structure of the receptor in Aseel and Control Broiler (CB) chicken population.

Primer-Blast (http://www.ncbi.nlm.nih.gov/tools/primer-blast). The amplification was carried out in 0.2 ml PCR tubes containing 2.5 μl of 10× PCR buffer, 1 μl of dNTP mix (2.5 mM), 1.5 μl (30 ng) each of forward and reverse primers, 0.3 μl (1.5 U) of Taq DNA polymerase, 1 μl of genomic DNA and nuclease free water to make the volume up to 25 μl. The amplified and gel eluted cDNA was cloned into pTZ57R/T vector with the aid of InsTAclone PCR Cloning Kit (Thermo Scientific, USA). The isolated plasmids of all fragments were sequenced with universal primers (M13/pUC (−40) forward and M13/pUC (−26) reverse) using sanger sequencing chemistry. Sequencing was done in ABI 3730xl 96 capillary system using Big Dye Terminator v3.1 kit by Xcelris laboratory (Ahmedabad, Gujarat). 2.3. SDS-PAGE analysis and Western blotting The pectoralis major muscle tissue extract was blended with equal volume of Laemmli 2× sample loading buffer (10% SDS, 0.025% Bromophenol blue and 1% DTT) and boiled for 3 min at 1000C. The digested samples (20 μl) containing approximately 15–20 μg of protein were electrophoresed in discontinuous buffer system Tris-Glycine-SDS buffer, pH 8.3 to separate the protein mixture. After completion of electrophoresis, protein in gel was transferred to polyvinyledene fluoride (PVDF) membrane using semi dry transfer apparatus (Bio-rad, USA) in the presence of Tris-Glycine–Methanol Buffer. Following transfer, the blotted PVDF was initially immersed in 3% BSA blocking buffer with primary antibody (1:1000 dilution in TBST) and incubated at 4 °C for overnight. After a series of washing steps, the membrane was incubated with anti-rat IgG HRP conjugate (Sigma, USA) diluted to 1:1000 in TBS Tween 20 buffer for 1.5 h with constant agitation. After few washings, the PVDF membrane was finally incubated in DAB substrate solution for 5–30 min until the colour development.

2. Materials and methods 2.1. Experimental birds The current research was carried out on Aseel (indigenous chicken line) and Control broiler (CB) maintained at the ICAR - Directorate of Poultry Research (DPR), Hyderabad. The Control Broiler (CB) line was a synthetic colour broiler line developed by random mating over last 9 generations. Aseel was an improved pure line variety developed by random mating over the last 4 generations. The pedigree had been maintained for all the birds to avoid any inbreeding while mating. Besides the random mating was practiced between sires and dams to maintain genetic variability in a population. All the birds were reared on deep litter system in the same shed under intensive management of farming, providing identical environmental conditions throughout the experimental period to nullify the effect of environment.

2.4. Bioinformatic prediction of amino acid sequence The different software tools under ExPASy (Expert Protein Analysis System), the SIB Bioinformatics Resource Portal were deployed to characterize the protein (Artimo et al., 2012). The amino acid sequences of Aseel and Control broiler (CB) were predicted using Translate tool (https://web.expasy.org/translate/) from the deduced nucleotide sequences. The sequenced nucleotide and predicted amino acid sequences were searched for similar sequences in the GenBank sequence database using BLASTn and BLASTp, respectively (https://blast. ncbi.nlm. nih.gov/Blast.cgi). A multiple-sequence alignment was performed using ClustalW2 (https://www.ebi.ac.uk/Tools/ msa/clustalw2/) for comparing sequences between lines and reference population. The parameters like the molecular weight, theoretical isoelectric point (pI), extinction coefficient, grand average of hydropathicity (GRAVY), estimated half-life, instability index, amino acid composition, atomic composition and aliphatic index were computed by ProtParam (Gasteiger et al., 2005). Hydropathy was calculated using the Kyte and Doolittle (1982) method (https://web.expasy.org/protscale/). The membrane-spanning regions and orientation of receptor was predicted using TMpred server (https://embnet.vital-it.ch/software/TMPRED_

2.2. Amplification, cloning and sequencing of ACTRIIB cDNA A total of 5 tissue samples (Pectoralis major muscle) from each line were used to isolate the RNA using Trizol (Invitrogen, USA) according to the manufacturer's protocol. The cDNA was synthesized from RNA using Verso cDNA Synthesis Kit (Thermo Scientific, USA). The full length ACTRIIB cDNA (1539 bp) was amplified using a pair of primers (Table 1) designed from the chicken cDNA sequence of activin A receptor, type IIB gene (GenBank accession no. NM_204317) using

Table 1 Primer sequence and thermal cycling conditions for amplification of ACTRIIB gene in chicken. Primer Sequence 5′-3′

Amplicon Size (bp)

Cycle step

Temperature

Time

Cycles

AR2BF: ATGAGCGCTTC GTGGCTGAC AR2BR: TTAGATACTGG ACTCTTTGGGC

1539

Initial denaturation

95 °C

5 min

1

Denaturation Annealing Extension Final Extension

95 °C 55 °C 72 °C 72 °C

30 s 30 s 1 min 30 s 10 min

35

140

1

Research in Veterinary Science 126 (2019) 139–149

P.G. Vishnu, et al.

Fig. 1. Analysis and confirmation of ACTRIIB cDNA and protein in chicken lines by A). Agarose gel electrophoresis, B) SDS-PAGE analysis and C) Western blot analysis. Lane 1–6: Amplified cDNA, Lane M: 1 kb bp DNA ladder – 250 bp to 10,000 bp (Thermo scientific).

form.html; Hofmann, 1993). The putative signal peptide cleavage site presence and location was envisaged by SignalP 4.1 Server (http:// www.cbs.dtu.dk/services/SignalP/; Petersen et al., 2011). The protein domain architecture was explored by Simple Modular Architecture Research Analysis (SMART) (Letunic and Bork, 2017) and Protein family (Pfam) database 32.0 (El-Gebali et al., 2018). The disulfide bond connectivity pattern was predicted by DiANNA 1.1 server (http:// clavius.bc.edu/~clotelab/DiANNA/) for given input of a protein sequence. The three-state description of the secondary structure was predicted by the Self-Optimized Prediction method with Alignment (SOPMA) (https://npsa-prabi.ibcp.fr/cgi-bin/npsa_automat.pl?page=/NPSA/ npsa_ sopma.html). RaptorX structure prediction (http://raptorx. uchicago.edu/; Wang et al., 2016), a web portal prediction program was used to predict the eight state secondary structure (SS8), solvent accessibility (ACC) and disordered regions in the protein sequence. The Protein–protein interaction networks of ActRIIB receptor were visualized by the biological database STRING (Search Tool for the Retrieval of Interacting Genes/Proteins Version 11.0 (https://string-db.org/; Szklarczyk et al., 2018). The predicted 3D structures of ActRIIB receptor of Aseel and CB were aligned by RaptorX structural alignment server (http://raptorx.uchicago.edu/Deep Align/submit/; Wang et al., 2016).

Volume Evaluation) program (Pontius et al., 1996). 2.7. Ramachandran plot Ramachandran plot displays the ψ (phi) and φ (psi) backbone conformational angles for each amino acid residue in a protein. The best 3D model obtained was analyzed by protein analysis program RAMPAGE (http://mordred.bioc.cam.ac.uk/~rapper/rampage.php; Lovell et al., 2003) for the evaluation of Ramachandran plot quality and WHATIF program (Vriend, 1990) for the calculation of package quality. 2.8. Protein description approaches Raptor X-Contact, a web server (http://raptorx.uchicago.edu/ ContactMap/; Wang et al., 2016) was used to predict the contact map of ACTRIIB protein sequence. The HB plot was visualized by a webbased application (http://hbplot.organic.hu/bin/infopage.php; Bikadi et al., 2007). 2.9. Phylogenetic tree construction The ACTRIIB phylogenetic tree was constructed by Mega V.7.0 (Kumar et al., 2016) employing the neighbor-joining (NJ) method (1000 bootstrap replicates) with ClustalW and MUSCLE as sequence alignment programs. The avian ACTRIIB sequence was compared with mammals, birds, amphibian and fish whose sequences were retrieved from Genbank database.

2.5. Protein threading and structural assembly The structural templates for amino acid sequence of our interest was identified utilizing I-TASSER (Iterative Threading ASSEmbly Refinement) server (Yang et al., 2015). The ITASSER models generated were submitted to COACH and COFACTOR algorithms for prediction of Enzyme Commission (EC) Number and Gene Ontology Terms (Yang et al., 2015).

3. Results

2.6. Protein structural models refinement and assessment

3.1. Cloning and confirmation of ACTRIIB gene

The prediction accuracy of each model was quantitatively measured by C-score (confidence score), TM-score (template modeling score) and RMSD (root-mean-square deviation) score. The predicted models were evaluated further for geometry of protein by Structural Analysis and Verification Server (SAVES) tools i.e., VERIFY3D (Eisenberg et al., 1997), ERRAT (Colovos and Yeates, 1993), WHATCHECK (Vriend, 1990), ProSA (Wiederstein and Sippl, 2007) and PROVE (PROtein

The PCR amplification of ACTRIIB cDNA for both the lines had resulted in specific bands of 1539 bp (Fig. 1A). The full length cDNA (1539 bp) encoding ACTRIIB receptor in Aseel and Control broiler (CB) was cloned, sequenced and submitted successfully in the NCBI database with GenBank ID; KX714223 for Aseel and KX865073 for CB. The SDSPAGE gel had demonstrated a 55 k dalton ACTRIIB protein (Fig. 1B) in CB and Aseel, which was also confirmed by western blot (Fig. 1C). 141

Research in Veterinary Science 126 (2019) 139–149

P.G. Vishnu, et al.

Fig. 2. Summary of prediction result of Secondary structures (SS3 and SS8), Solvent accessibility (ACC) and disordered regions (DISO) in Aseel and Broiler. SEQ- Amino acid sequence; SS3- Three state secondary structures; SS8 – Eight state secondary structures. Abbreviations in SS3: Helix (H), Beta sheet (E) and Coil (C) and Abbreviations in SS8: α-helix (H), 3–10 helix (G), π-helix (I), β-strand (E), isolated beta bridge (B), hydrogen bonded turn (T), bend (S) and Coil (C). * indicates disordered portion of sequence and ● represents ordered portion of amino acid sequence

3.2. Nucleotide sequence analysis

clustal analysis of predicted amino acid sequences demonstrated a score of 99.2% (CB vs. reference genome), 98.44% (CB vs. Aseel) and 99.2% (Aseel vs. reference genome). Sequence analysis of cDNA had revealed several sequence variants between Aseel and CB, when compared with the reference sequence of Gallus gallus. The nucleotides differences between reference genome and CB were c. [520G > C; 686G > A; 937C > G; 1130A > G; 1326 T > C]. The differences of nucleotides between reference genome and Aseel were c. [156G > A; 210 T > C; 493C > T; 665A > C; 1011A > C; 1208 T > A; 1433 T > C]. The nucleotide substitutions between CB and Aseel were c. [156G > A; 210 T > C; 493C > T; c.520G > C; 665A > C; 686G > A; 937C > G; 1011A > C; 1130A > G; 1208 T > A; 1326 T > C; 1433 T > C].

In Aseel breed, the composition (%) of different nucleotides in ACTRIIB were 25.28 (A), 27.62(G), 21.05 (T) and 26.06(C), and the Davis Botstein Roth melting temperature and Wallace temperature were 86.58 and 5532 °C, respectively. While in CB birds, the makeup of the A, G, T and C were 25.28, 27.68, 21.12 and 25.93%, respectively. The Davis Botstein Roth melting temperature and Wallace temperature were 86.55 and 5526 °C, respectively. The ACTRIIB open reading frame (ORF) contained 1539 nucleotides coding for 512 amino acids. Clustal analysis of the nucleotide sequences revealed an alignment score of 99.5% similarity between Aseel and reference genome, 99.6% between CB and reference genome and 99.2% between CB and Aseel. Similarly,

142

Research in Veterinary Science 126 (2019) 139–149

P.G. Vishnu, et al.

[(Lys53=; Glu71=; Leu337=)]. The amino acid substitutions between CB and Aseel were p. [(Pro165Ser; Glu174Gln; Gln222Pro; Ser229Asn; His313Asp; Gln377Arg; Val403Asp; and Ile478Thr)]. While the silent changes positioned at p. [(Lys53=; Glu71=; Leu337=; Asp442=)].

Table 2 Evaluation of amino acid sequences of ACTRIIB gene in chicken. Control broiler MSASWLTLAVLCATLGAGPGHGEAETRECIYYNANWELEKTNQSGVERCEGEKDKRLHCY ASWRNNSGSIELVKKGCWLDDFNCYDRQECVATEENPQVFFCCCEGNYCNEKFTHLPEVT GPEVIYEPPPPTPSLLNILVYSLLPIAVLSVAILLAFWMYRHRKPPYGHVDINQDPGPPP PSPLVGLKPLQLLEIKARGRFGCVWKAQLMNDYVAVKIFPIQDKQSWQNEREIFNTPGMK HENLLQFIAAEKRGTNLETELWLITAFHDKGSLTDYLKGNIISWNELCHVAETMARGLSY LHEDVPWCKGEGDKPAIAHRDFKSKNVLLKNDLTAVLADFGLAVRFEPGKPPGDTHGQVG TRRYMAPEVLEGAINFRRDAFLRIDMYAMGLVLWELVSRCRAVDGPVDEYMLPFEEEIGQ HPSLEDLQEVVVHKKMRPVFKDHWLKHPGLAQLCVTIEECWDHDAEARLSAGCVEERIAQ IRKSVNGTTSDCLVSIVTSVTNVDLPPKESSI

3.4. Secondary structure analysis The three state secondary structure types i.e., Helix (H), Beta sheet (E) and Coil (C) and eight state secondary structure types viz., α-helix (H), 3–10 helix (G), π-helix (I), β-strand (E), isolated beta bridge (B), hydrogen bonded turn (T), bend (S) and Coil (C) were depicted in the Fig. 2. The secondary structure analysis predicted 27.18% helix (H), 15.81% Beta sheet (E) and 57.01% random coils (C) for three state model in Aseel, while in CB it was 27.68% helix (H), 16.56% Beta sheet (E) and 55.76% random coils (C). 3.5. Solvent accessibility and disorder prediction of amino acid residues

Aseel MSASWLTLAVLCATLGAGPGHGEAETRECIYYNANWELEKTNQSGVERCEGEKDKRLHCY ASWRNNSGSIELVKKGCWLDDFNCYDRQECVATEENPQVFFCCCEGNYCNEKFTHLPEVT GPEVIYEPPPPTPSLLNILVYSLLPIAVLSVAILLAFWMYRHRKSPYGHVDINEDPGPPP PSPLVGLKPLQLLEIKARGRFGCVWKAQLMNDYVAVKIFPIPDKQSWQSEREIFNTPGMK HENLLQFIAAEKRGTNLETELWLITAFHDKGSLTDYLKGNIISWNELCHVAETMARGLSY LHEDVPWCKGEGHKPAIAHRDFKSKNVLLKNDLTAVLADFGLAVRFEPGKPPGDTHGQVG TRRYMAPEVLEGAINFQRDAFLRIDMYAMGLVLWELVSRCRADDGPVDEYMLPFEEEIGQ HPSLEDLQEVVVHKKMRPVFKDHWLKHPGLAQLCVTIEECWDHDAEARLSAGCVEERTAQ IRKSVNGTTSDCLVSIVTSVTNVDLPPKESSI

The predicted solvent accessibility uncovered that 24% of amino acid residues were not accessible to solvent (buried), 25% of amino acid residues were accessible to solvent (exposed) and 50% of amino acid residues were in between (medium) in Aseel and CB (Fig. 2). Protein disorder prediction system predicted that among total 512 amino acid residues, 94 residues (18%) and 82 residues (16%) were predicted as disordered in Aseel and CB respectively. In both the populations the disordered region was mainly observed around regions 1–25, 165–187 and 503–512 amino acid residues of protein sequence (Fig. 2). 3.6. Domain organisation and composition

Predicted signal peptide is underlined, predicted transmembrane region is indicated by box. Bold C represents the conserved cysteine residues. NQS, NNS and NGT represents N-glycosylation sites. Proline rich residues juxtaposed on the TMD were presented by dotted line box.

Domain architecture analysis of ACTRIIB amino acid sequence by exploiting SMART tool predicted the following domains: signal peptide (1–24), intra cellular domain (ICD)/Activin receptor domain (27–117), transmembrane domain (138–160) and extra cellular domain (ECD)/ serine threonine protein kinase catalytic domain (190–479). Hydropathy analysis utilizing the TMpred program and Kyte & Doolittle hydrophobic plot revealed two hydrophobic regions: a 19 amino acid stretch (inside to outside orientation) at the N-terminal assumed to be a signal peptide and a single putative 23 residue membrane-spanning region (outside to inside orientation) between residues 138 and 160. The proline rich residues, transmembrane residues and other conserved residues of ACTRIIB were highlighted in the Table 2. Bioinformatic analyses envisaged the N-linked glycosylation sites at amino acid positions 42, 65 and 486 (Table 2). It also disclosed 19 cysteine residues at positions 12th, 29th, 49th, 59th, 77th, 84th, 90th, 102nd, 103rd, 104th, 109th, 203rd, 288th, 308th, 400th, 454th, 460th, 473th and 492th position in the amino acid sequence (Table 2). The disulfide bonds were identified between the following residues: 12–492, 29–84, 49–59, 77–102, 90–473, 103–203, 104–308, 288–454 and 400–460.

3.3. Amino acid sequence analysis Sequence analysis of ACTRIIB protein indicated that it comprised 512 amino acid residues, of which first 24 residues were signal peptide and the remaining 488 residues (25–512) were mature peptide in Aseel and CB. In Aseel, the molecular weight of pre-protein was 57.73 kDa, while the mature protein was 55.51 kDa. It encompassed 52 strongly basic amino acids, 71 strongly acidic amino acids, 175 hydrophobic amino acids and 119 polar amino acids. The theoretical isoelectric point and charge at PH 7.0 were 5.460 and − 16.756, respectively. The molecular weight of pre protein in CB was 57.80 kDa, while the mature protein was 55.58 kDa. It comprised 53 strongly basic amino acids, 70 strongly acidic amino acids, 177 hydrophobic amino acids and 118 polar amino acids. The theoretical isoelectric point and charge at PH 7.0 were 5.544 and − 14.925, respectively. The total number of negatively (Asp+Glu) charged residues were 71 in Aseel and 70 in CB, while the number of positively charged residues (Arg + Lys) were 52 and 53 in Aseel and CB, respectively. The computed value of extinction coefficient was similar in Aseel and CB and was found to be 94,975 (Cystine bridges) and 93,850 (cysteine residues), respectively. The estimated half-life was 30 h in both Aseel and CB. The physico-chemical properties like Instability index, Aliphatic index and Grand average of hydropathicity (GRAVY) were 43.92, 84.16 and - 0.302 in Aseel, the corresponding values in CB were 42.85, 85.49 and − 0.29. The amino acid substitutions between reference genome and CB were p. [(Glu174Gln; Ser229Asn; His313Asp; Gln377Arg)]. The silent changes between these two genomes was p. [(Asp442=)]. The amino acid substitutions between reference and Aseel were p. [(Pro165Ser; Gln222Pro; Val403Asp; Ile478Thr)]. The silent changes included p.

3.7. Three dimensional structure modeling The Activin Receptor Type II B 3D structure was submitted successfully in Protein Model Database (PMDB) with the PMDB ID – PM0081578 for Aseel and PM0081579 for CB. Among the top 5 models predicted by I-TASSER, the C-score for higher rank model was −1.60 in Aseel and − 1.41 in CB, while the estimated TM-score (0.54 ± 0.14) and RMSD (5.8 ± 1.2 Å) were found to be similar in Aseel and CB (Fig. 3). The PDB hit 2qluA (Activin receptor type II kinase domain from human) ranked top among all the 10 threading templates used by I-TASSER, with a normalized Z-score of 2.50 in Aseel and CB. The top structural analog identified from the PDB library was 3q4tB (Activin receptor type-IIA (ACVR2A) kinase domain), whose TM-score and RMSD was 0.584 and 1.64 for both Aseel and CB. The best PDB hit with respect to ligand binding site prediction was 143

Research in Veterinary Science 126 (2019) 139–149

P.G. Vishnu, et al.

Fig. 3. Top Five 3D models predicted by I-TASSER for ACTRIIB protein in Aseel and Broiler. C-score refers to the confidence of each model, where a C-score of higher value signifies a model with a high confidence and vice-versa.

predicted to be 5E8XA (Aseel) and 5E8Y (CB) (TGF-BETA RECEPTOR TYPE 1 & 2 KINASE DOMAIN). It had a C-score and cluster size of 0.93 and 758 for Aseel, while in CB it was 0.79 and 652. The COACH algorithm deduced the ligand binding site residues at 196,197,198,199,204,215,217,230,245,265, 266, 267,268,269, 271,272,325, 328,338 and 339th position in both populations. The COFACTOR predicted the best PDB hit as 3g2fA (Kinase domain of bone morphogenetic protein receptor type II) among all the templates pertaining to Enzyme commission prediction. The C-score, TM and RMSD score were 0.191, 0.559 and 1.97 in Aseel, correspondingly, the values in CB were 0.156, 0.560 and 1.97. The active site residues for Aseel and CB were predicted at 202,204,271 and 202,204,271,328,330, respectively. The consensus prediction of Gene Ontology (GO) terms for all the templates illustrated that both populations possessed receptor signaling protein activity in molecular function category. Whereas, the biological process category showed protein phosphorylation and transmembrane receptor protein serine/ threonine kinase signaling pathway activities and the membrane activity under cellular component category. The GO score for all these functions ranged from 0.54 to 0.71.

predicted that the buried outlier protein atoms was 7.2% in Aseel (120 atoms) and CB (146 atoms). The Z score RMS of buried atoms for resolutions of 1.6 A° or better was essentially constant and around 1.0 in both the populations. ProSA (Protein Structure Analysis) displayed quite similar overall model quality (Z-score) for 3D structures of Aseel (−5.73) and CB (−6.80). ProSA black dot representing z-score of the input structures was much closer to the middle region of scores observed for experimentally determined protein structures in the PDB and fall in NMR based reliability scale in Aseel and CB (Fig. 5B). Different parameters like 1st generation packing quality, Ramachandran plot appearance, chi-1/chi-2 rotamer normality and backbone conformation were analyzed by WHATCHECK programme and predicted the overall summary report as pass for all the parameters in Aseel and CB. 3.9. Black and white protein contact map The pattern of contact map (matrix representation of ACTRIIB protein residue–residue contacts) was comparable in Aseel and CB (Fig. 6A). The cluster of contacts in the map primarily denoted secondary structures. The alpha helices formed contact patterns that line along the main diagonal of the contact matrix, while contact patterns that were either parallel (parallel beta sheets) or perpendicular (antiparallel beta sheets) to the main diagonal were indicative of betasheets. Contrastingly, the contact patterns of tertiary structure (interactions between two secondary structural components) were less dense and distant from the main diagonal. Furthermore, the contact patterns in tertiary structure did not manifest definitive contact shapes compared to the well-defined secondary structure groups.

3.8. Ramachandran plot and model evaluation The Ramachandran plot analysis using RAMPAGE software exposed that among the 512 residues, 94.1% were in favored region, 4.7% in allowed region and 1.2% in disallowed region in Aseel. Whereas in case of CB, 94.5% were in favored region, 4.7% in allowed region and 0.8% in disallowed region. In both the lines > 90% residues had allowed conformations (Fig. 4). Further, > 2/3rd amino acid residues of the protein fall in the upper left quadrant i.e., the ϕ angles of amino residues were between -180o and 0o and ψ angles of amino residues were between 0o and + 180o. Ramachandran plots for general, glycine, preproline and proline were also done and all fall under favored regions in Aseel and CB (Fig. 4). VERIFY3D assessment of the predicted structures displayed that > 80% of the residues had an averaged 3D-1D score ≥ 0.2 in Aseel (84.38%) and CB (85.74%) indicating the result as pass in both populations (Fig. 5A). The overall quality factor computed by ERRAT program showed that, for about 88.86% of protein in case of Aseel and 64.67% of protein in CB, the calculated error value falls below the 95% statistical rejection limit. The analysis of entire structure by PROVE

3.10. Hydrogen bond plot The HB plots of the Aseel and CB demonstrated similar secondary structure pattern. Main diagonal corresponded to the secondary structure elements, while off-diagonal points represented the tertiary structure elements (Fig. 6B). The overall shape of the plot was typical representative of an alpha-class protein with only few off diagonal points. It was observed that, about 90–95% of hydrogen bonds accounted for stabilizing secondary structure elements, while 5–10% of hydrogen bonds were responsible for forming tertiary hydrogen bonds, pointing out that the dominating elements of a HB plot were the hydrogen bonds stabilizing the secondary structures. The tertiary 144

Research in Veterinary Science 126 (2019) 139–149

P.G. Vishnu, et al.

Fig. 4. Rampage analysis of Ramachandran plot in Aseel and Broiler. φ (psi) and ψ (phi) backbone dihedral angles of each amino acid residue in a protein are plotted on the X-axis and Y- axis respectively. Dark sky blue refers to core regions, pale sky blue refers to allowed regions and white areas corresponds to unfavorable regions. Square, Triangle and Cross symbols in the plot represents amino acid residues. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

4. Discussion

hydrogen bonding network was mostly observed in the residues around 25–150 and 250–450. Visual inspection of the HB plots revealed both dense (numerous hydrogen bonds) and empty regions (few hydrogen bonds), explicitly, the scattered gaps were observed only in the initial one third portion of plot and not in the remaining portion of plot.

In the present study, we observed that the ACTRIIB gene coding DNA sequence length was 1539 bp, which translated in to 512 amino acid polypeptide chain. The ORF length was consistent in CB and Aseel, besides a comparison with other avian species also revealed that the ORF length was almost similar. However, sheep, yak, buffalo and camel had demonstrated slight deviation from the chicken ORF length, due to the divergence between species. Clustal analysis between CB and Aseel showed an alignment score of > 95%, suggesting that the differences at the nucleotide level were probably result of point mutations.

3.11. Protein – protein interaction The structural alignment score i.e., TM score was recorded to be 0.613 (> 0.6) between Aseel and CB and accordingly, the 3D structures of ACTRIIB protein in Aseel and CB aligned perfectly, indicating similar fold nature of 3D structures (Fig. 7). Analysis of protein-protein interactions predicted that the ACTRIIB protein interacted with ACVR1B, ACVR1C, INHBA, SMAD 1,2,5,7 & 9 and BMPR1A&B. The protein kinase catalytic domain of ACTRIIB receptor interacted with ACVR1B at amino acid position 170–511(pfam domain), ACVR1C at position 104–444 (pfam domain), SMAD 1 & 7 at position 2–142 (Domain A in dwarfin family proteins - DWA domain), SMAD 2, 5 & 9 at position 240–465 (DWB domain), BMPR1A & B at position 233–567 (pfam domain) and INHBA at position 309–424 (TGFB domain).

4.1. Protein physico-chemical properties and structure analysis The physicochemical properties are unique for each amino acid and hence predicting the different characteristics based on amino acid properties, is prerequisite for functional and comparative studies on protein. All the physicochemical properties computed were analogous with minor differences between the two populations. The computation of different parameters for the amino acid sequence aids in determining the topology of protein, which can be useful for further studies. The higher instability index (> 40) indicates that the protein may be unstable. The higher aliphatic index signifies the greater structural stability of ACTRIIB protein. Negative GRAVY value indicates the nonpolar and hydrophilic nature of ACTRIIB protein. Secondary structure analysis revealed a certain variation in composition of alpha helices and beta sheets between CB and Aseel. Many researchers reported that secondary structures play an important role in protein folding and its activity (Ji and Li, 2010). Therefore, alteration in the proportion of alpha helices and beta sheets may be considered as a check point for the differential magnitude of ACTRIIB protein function in CB and Aseel. The solvent accessibility prediction had immense role in refining the prediction of protein secondary and tertiary structures (MomenRoknabadi et al., 2008). A transmembrane residue can be coarsely classified as exposed to the lipid environment or buried inside the protein. Complying with the transmembrane proteins class, the majority of amino acid residues (approximately 75%) were not accessible

3.12. Phylogenetic relationship Phylogenetic analysis of ACTRIIB amino acid sequence among different species divulged that all avian species were closely related and formed many sub clusters within avian cluster (Fig. 8). Both CB and Aseel clustered into a clade together with Gallus gallus, however Aseel formed a sub cluster. The Aseel and CB strains were closely related to turkey and quail. On the other hand duck, penguin and ostrich formed separate sub clusters. The avian cluster was very closely related to zebra fish and closely related to rat and mouse cluster. While it was distantly related to sheep, goat, cow and buffalo. Among the livestock species both large and small ruminants formed a separate cluster. Similarly, non-ruminants (pig, horse and camel) and canines also formed separated clusters. Likewise, Human and chimpanzee also formed separate clusters, which were closely related to avian cluster than other clusters. 145

Research in Veterinary Science 126 (2019) 139–149

P.G. Vishnu, et al.

Fig. 5. Assessment of ACTRIIB three dimensional structure by A) Verify 3D and B) ProSA in Aseel and Broiler. A) Residue number and 3D-1D profile score are plotted on the X-axis and Y- axis respectively, B) ProSA black dot represents z-score of the input structures and fall in NMR based reliability scale.

to solvent in Aseel and CB. Tusnády et al. (2015)) asserted that intrinsically disordered regions (IDRs) tend to be in the N- or C-terminal regions of TMPs. Accordingly, we also observed disordered regions majorly in the N- or C-terminal regions of protein. It was noticed that position and placement of cysteine residues in ACTRIIB receptor were comparable to TGF-β type II receptor family ligands. Equivalently, Funkenstein et al. (2012) also identified ten highly conserved cysteine residues in ECD of ACTRIIB in fish. Considering the conservative behavior of cysteine residues in homologous proteins, the topology pattern of disulfide bridges between cysteine residues observed in ACTRIIB protein can be utilized as diagnostic tool for identification of proteins of similar 3D structure.

90 residue ECD, a 23 residue hydrophobic TMD and a 290 residue larger ICD, consistent with the findings of Hinck (2012), Funkenstein et al. (2012) and Han et al. (2007). Earlier investigations conducted in mouse demonstrated that seven amino residues (Tyr60, Val73, Trp78, Leu79, Phe82, Val99 and Phe101) of ACTRIIB receptor played major role in binding of ACTRIIB receptor with activin and inhibin (Thompson et al., 2003). These residues were also conserved in the ECD of both CB and Aseel in their position and spacing. In majority of the TGF-β type II ligands, the proline rich motifs were positioned on either side of TMD (Hinck, 2012). Comparably, the TMD of ACTRIIB in both the lines was also flanked by proline-rich sequences making the domain biologically more active. A structural analysis study about ACTRIIB kinase domain conducted in humans had identified four kinase domain residues (Phe234, Leu245, Phe247 and Thr265) in ICD, crucial for the activity of ACTRIIB active site (Han et al., 2007). Congruent with these findings, ICD in CB and Aseel had presented all these residues, indicating the conserved nature of ICD among protein kinase family in different populations. Additionally, Mathews and Vale (1991) identified two subdomains containing serine kinase consensus sequences (DFKSKN and GTRRYMAPE) in ACTRIIB responsible for predicting tyrosine vs serine/threonine

4.2. Bioinformatic analysis of the predicted amino acid domains Proteins generally comprised of one or more functional regions, usually defined as domains. The occurrence of distinct domains in various permutations results in diverse repertoire of proteins available in biological systems. Hence, ascertaining the domains in the protein may provide further insights into the function of any protein (Schulz and Schirmer, 2013). ACTRIIB possessed the hallmark regions of TGFβ type 2 receptor superfamily, including a 24 residue signal peptide, short 146

Research in Veterinary Science 126 (2019) 139–149

P.G. Vishnu, et al.

Fig. 6. Description of ACTRIIB protein by (A) Contact Map and (B)Hydrogen Bond plot in Aseel and Broiler. A) The Cα atom of each amino acid is considered as vertices of the corresponding protein contact network. A cut-off value of 8 Å distance was used to determine whether any two residues were connected. If any two residues were connected, then the matrix cell values were set to 1 (black colour) or else 0 (white colour) if they were not connected. B) The HB plot is obtained by plotting the amino acid residues involved in hydrogen bonding horizontally and vertically. Briefly, maximum distances of donor and acceptor atoms are 3.9 A°; minimum angles of donor–hydrogen–acceptor are 90.0°.

active sites and gene ontology terms were within appreciable range, implying more confident predictions for all the features at the structural and functional level. 4.4. Model assessment Verify 3D profile score for the model structures infers that predicted 3D model is accurate and consistent in both populations. The proportion of correctly determined regions of protein structures predicted by ERRAT program was more, signifying the reliability of predicted model. The ProSA Z- score in both structures was within the range of scores typically found for native proteins of similar size, implying good quality of homology modelled protein and plausible accuracy in dihedral angles distribution and stereochemical characteristics. The PROVE Z-score RMS for both the populations was below 3, inferring that all the atoms in majority of regions is having acceptable standards of stereo chemical parameters and there is no volume irregularity of a protein structure. The Whatcheck programme predicted the result as pass, denoting that packing quality is within in normal range.

Fig. 7. Structural alignment of ACTRIIB three dimensional structure of Aseel and Broiler.

4.5. Analysis of Ramachandran plot

substrate specificity, which were also deduced in ACTRIIB of CB and Aseel.

An important intermediate step in modeling the three dimensional structure of a protein is accurate prediction of the secondary structures. In general, the protein secondary structure is more conserved than its amino acid sequence (Illergard et al., 2009). The various secondary structures can be differentiated by their range of φ and ψ values, with the values of different secondary structures mapping to distinct regions of the Ramachandran plot. The ideal values of ϕ/ψ were determined to be as follows: right-handed α-helix (−57o/−47o); left-handed α-helix (+57o/+47o); right-handed 310 helix (−74o/−4o); right-handed πhelix (−57o/−70o); parallel β-sheet (uncommon) (−119o/+113o); antiparallel β-sheet (common) (−139o/+135o) (Hollingsworth and Karplus, 2010). Assessment of the Ramachandran plot illustrated that the majority of the amino acid residues were mapped in the favored region, indicating that most amino acid conformations of phi and psi angles in

4.3. 3D structure prediction accuracy The quality of structure models predicted by I-TASSER I is evaluated by the value of C-score. It is typically in the range between −5 and 2, where a C-score of higher value signifies a model with a high confidence and vice-versa (Yang et al., 2015). The predicted C– score for both the populations signifies high confident models. The TM and RMSD score are known standards for measuring the accuracy of structure modeling when the native structure is known. A TM-score of > 0.5 and low RMSD score in our study suggests a model of correct topology and similar nature of protein folding in two populations. The different scores like Z- score, C-score, TM-score, RMSD score and GO score pertaining to prediction of structural analogs, ligand binding sites, 147

Research in Veterinary Science 126 (2019) 139–149

P.G. Vishnu, et al.

Fig. 8. Phylogenetic analysis of ACTRIIB gene based on amino acid sequence. The amino acid sequences were aligned using Clustal W and the dendogram was constructed Neighbor joining method. Numbers at the tree nodes represent bootstrap values after 1000 replicates. Aseel and CB were highlighted in Square and Rhombus shapes.

the protein do not have any steric hindrance/clashes between atoms. Laskowski et al. (1993) proclaimed that good model generally had 90% of its residues in the allowable regions of a Ramachandran plot and accordingly, the predicted 3D model in our study is acceptable and can be considered as a good model. The detailed inspection of the Ramachandran plot indicates that the propensity of amino acid residues to be in beta helix is higher (more number of residues in upper left quadrant region) in both the lines, than other secondary structures (right handed and left handed alpha helix).

hydrogen bonding network plays major role in determining 3D structure, function and flexibility of protein (Amitai et al., 2004). Tertiary hydrogen bond network of connection points illustrates, how interactions between distant residues cooperate, to which part of the secondary structure of the protein the information flows and how they can influence the local interactions (Bikadi et al., 2007). Analyzing the network of tertiary hydrogen bond inferred that this network corresponded to two domains i.e., Activin receptor domain (27–117) and serine threonine protein kinase catalytic domain (190–479). The residues in these domains are crucial for maintaining fast communication between distant sites and plays key role in the spread of information within a protein (Bikadi et al., 2007). The secondary structure pattern and ordered network of tertiary hydrogen bonds (characteristic of the protein family) was similar in Aseel and CB, implying that the structure is more conserved within a protein family than the amino acid sequence.

4.6. Contact map and HB plot The protein contact maps are two-dimensional representation of the three-dimensional layout of protein structures (Wang et al., 2017). The conserved or similar pattern of inter residue contacts in both the populations indicates that Aseel and CB tend to share similar 3D structures. The hydrogen bonding network offer many advantages for protein description over more traditional approaches such as distance plots and contact maps, due to its consideration of specific secondary interactions within the protein (Bikadi et al., 2007). The hydrogen bonds can act within distant amino acid residues and therefore can be considered as one of driving forces for 3D structure stabilization and flexibility (Parker et al., 1996). Secondary hydrogen bonding network is imperative for stabilizing secondary structure elements, while tertiary

4.7. Phylogenetic analysis The close evolutionary relationship of CB and Aseel with other avian species denotes the orthologous nature of the ACTRIIB gene. The close phylogenetic relationship between CB and Aseel suggests that both CB and Aseel would have accumulated equal amount of genetic change over time during evolution history. 148

Research in Veterinary Science 126 (2019) 139–149

P.G. Vishnu, et al.

5. Conclusion

Gasteiger, E., Hoogland, C., Gattiker, A., Wilkins, M.R., Appel, R.D., Bairoch, A., 2005. Protein identification and analysis tools on the ExPASy server. In: The Proteomics Protocols Handbook, pp. 571–607. Han, S., Loulakis, P., Griffor, M., Xie, Z., 2007. Crystal structure of activin receptor type IIB kinase domain from human at 2.0 Å resolution. Protein Sci. 16, 2272–2277. Hinck, A.P., 2012. Structural studies of the TGF-βs and their receptors–insights into evolution of the TGF-β superfamily. F.E.B.S. Lett. 586, 1860–1870. Hofmann, K.A.W.S., 1993. TMbase-A database of membrane spanning proteins segments. Biol. Chem. Hoppe Seyler 374, 166. Hollingsworth, S.A., Karplus, P.A., 2010. A fresh look at the Ramachandran plot and the occurrence of standard structures in proteins. Biomol. Concepts. 1, 271–283. Illergard, K., Ardell, D.H., Elofsson, A., 2009. Structure is three to ten times more conserved than sequence—a study of structural response in protein cores. Proteins. 77, 499–508. Ji, Y.Y., Li, Y.Q., 2010. The role of secondary structure in protein structure selection. Eur. Phys. J. E. Soft. Matter. 32, 103–107. Krogh, A., Larsson, B., von Heijne, G., Sonnhammer, E.L., 2001. Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J. Biol. Chem. 305, 567–580. Kumar, S., Stecher, G., Tamura, K., 2016. MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol. Biol. Evol. 33, 1870–1874. Kyte, J., Doolittle, R.F., 1982. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157, 105–132. Lach-Trifilieff, E., Minetti, G.C., Sheppard, K., Ibebunjo, C., Feige, J.N., Hartmann, S., et al., 2014. An antibody blocking activin type II receptors induces strong skeletal muscle hypertrophy and protects from atrophy. Mol. Cell. Biol. 34 (4), 606–618. Laskowski, R.A., MacArthur, M.W., Moss, D.S., Thornton, J.M., 1993. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291. Lee, S.J., Reed, L.A., Davies, M.V., Girgenrath, S., Goad, M.E., Tomkinson, K.N., et al., 2005. Regulation of muscle growth by multiple ligands signaling through activin type II receptors. Proc. Natl. Acad. Sci. U. S. A. 102, 18117–18122. Letunic, I., Bork, P., 2017. 20 years of the SMART protein domain annotation resource. Nucleic Acids Res. 46, D493–D496. Lopez, G., Rojas, A., Tress, M., Valencia, A., 2007. Assessment of predictions submitted for the CASP7 function prediction category. Proteins. 69, 165–174. Lovell, S.C., Davis, I.W., Arendall III, W.B., De Bakker, P.I., Word, J.M., Prisant, M.G., Richardson, D.C., 2003. Structure validation by Cα geometry: ϕ, ψ and Cβ deviation. Proteins. 50, 437–450. Mathews, L.S., Vale, W.W., 1991. Expression cloning of an activin receptor, a predicted transmembrane serine kinase. Cell. 65, 973–982. Momen-Roknabadi, A., Sadeghi, M., Pezeshk, H., Marashi, S.A., 2008. Impact of residue accessible surface area on the prediction of protein secondary structures. B.M.C. Bioinforma. 9, 357. Moult, J., Fidelis, K., Kryshtafovych, A., Schwede, T., Tramontano, A., 2016. Critical assessment of methods of protein structure prediction: Progress and new directions in round XI. Proteins 84, 4–14. Parker, M.J., Sessions, R.B., Badcoe, I.G., Clarke, A.R., 1996. The development of tertiary interactions during the folding of a large protein. Fold. Des. 1, 145–156. Petersen, T.N., Brunak, S., von Heijne, G., Nielsen, H., 2011. SignalP 4.0: discriminating signal peptides from transmembrane regions. Nat. Methods 8, 785. Pontius, J., Richelle, J., Wodak, S.J., 1996. Deviations from standard atomic volumes as a quality measure for protein crystal structures. J. Mol. Biol. 264, 121–136. Relizani, K., Mouisel, E., Giannesini, B., Hourde, C., Patel, K., Gonzalez, S.M., et al., 2014. Blockade of ACTRIIB signaling triggers muscle fatigability and metabolic myopathy. Mol. Ther. 22 (8), 1423–1433. Sako, D., Grinberg, A.V., Liu, J., Davies, M.V., Castonguay, R., Maniatis, S., et al., 2010. Characterization of the ligand binding functionality of the extracellular domain of activin receptor type IIb. J. Biol. Chem. 285, 1037–1048. Schulz, G.E., Schirmer, R.H., 2013. Principles of Protein Structure. Springer Science & Business Media. Sleator, R.D., Walsh, P., 2010. An overview of in silico protein function prediction. Arch. Microbiol. 192, 151–155. Stevens, T.J., Arkin, I.T., 2000. Do more complex organisms have a greater proportion of membrane proteins in their genomes? Proteins. 39, 417–420. Szklarczyk, D., Gable, A.L., Lyon, D., Junge, A., Wyder, S., Huerta-Cepas, J., Jensen, L.J., 2018. STRING v11: protein–protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 47, D607–D613. Thompson, T.B., Woodruff, T.K., Jardetzky, T.S., 2003. Structures of an ActRIIB: activin a complex reveal a novel binding mode for TGF-β ligand: receptor interactions. EMBO J. 22, 1555–1566. Tusnády, G.E., Dobson, L., Tompa, P., 2015. Disordered regions in transmembrane proteins. Biochim. Biophys. Acta 1848, 2839–2848. Vriend, G., 1990. WHAT IF: a molecular modeling and drug design program. J. Mol. Graph. Model. 8, 52–56. Wang, S., Li, W., Liu, S., Xu, J., 2016. RaptorX-property: a web server for protein structure property prediction. Nucleic Acids Res. 44, W430–W435. Wang, S., Sun, S., Li, Z., Zhang, R., Xu, J., 2017. Accurate de novo prediction of protein contact map by ultra-deep learning model. PLoS Comput. Biol. 13, e1005324. Westbrook, J., Feng, Z., Jain, S., Bhat, T.N., Thanki, N., Ravichandran, V., Gilliland, G.L., Bluhm, W., Weissig, H., Greer, D.S., Bourne, P.E., Berman, H.M., 2002. The protein data Bank: unifying the archive. Nucleic Acids Res. 30, 245–248. Wiederstein, M., Sippl, M.J., 2007. ProSA-web: interactive web service for the recognition of errors in three-dimensional structures of proteins. Nucleic Acids Res. 35, W407–W410. Yang, J., Yan, R., Roy, A., Xu, D., Poisson, J., Zhang, Y., 2015. The I-TASSER suite: protein structure and function prediction. Nat. Methods 12, 7. Zhang, Y., 2008. Progress and challenges in protein structure prediction. Curr. Opin. Struct. Biol. 18, 342–348.

Many researchers observed that down regulation or abolishing the activity of ACTRIIB either by administration of anti-ACTRIIB antibody (human or murine) or soluble form of ACTRIIB (sACTRIIB) had increased muscle mass tissue in mouse and fish (Lach-Trifilieff et al., 2014; Relizani et al., 2014). The current research is the pioneer among various studies to predict the three dimensional structure of the ACTRIIB protein and it may provide further insights about structural and functional aspects of the gene and also aids in understanding the evolutionary history of ACTRIIB gene. Therefore, the molecular information pertaining to structural and functional aspects of ACTRIIB gene documented by our study is very essential to understand the molecular regulation of muscle growth by this gene. We observed substantial differences in cDNA of ACTRIIB between CB and Aseel at nucleotide and protein level, however dendogram analysis indicated both of them were analogous. Phylogenetic analysis showed strong similarity to other avian species indicating common ancestry of receptor for all avian species. Acknowledgements All the authors would like to thank to the Indian Council of Agricultural Research (ICAR) for giving financial assistance in carrying out this experiment under National Fellow project. Author contributions P. Guru Vishnu conducted wet lab experiment. T.K. Bhattacharya guided the whole experiment, designed oligos and prepared the manuscript. P. Kumar and Bharat Bhushan assisted to analyse data. R.N. Chatterjee and Chandan paswan prepared the tables and figs. K.Dushyanth, D.Divya and A.Rajendraprasad helped in compiling the manuscript. Ethical standards We declare that the whole experiment conducted was conforming to the ethical standards and was approved by Institutional Animal Ethics Committee (IAEC) of ICAR-Directorate of Poultry Research, Hyderabad. Declaration of Competing Interest The authors declare that there is no conflict of interest with the contents of this manuscript. References Amitai, G., Shemesh, A., Sitbon, E., Shklar, M., Netanely, D., Venger, I., Pietrokovski, S., 2004. Network analysis of protein structures identifies functional residues. J. Mol. Biol. 344, 1135–1146. Artimo, P., Jonnalagedda, M., Arnold, K., Baratin, D., Csardi, G., de Castro, E., et al., 2012. ExPASy: SIB bioinformatics resource portal. Nucleic Acids Res. 40, W597–W603. Bikadi, Z., Demko, L., Hazai, E., 2007. Functional and structural characterization of a protein based on analysis of its hydrogen bonding network by hydrogen bonding plot. Arch. Biochem. Biophys. 461, 225–234. Colovos, C., Yeates, T.O., 1993. Verification of protein structures: patterns of nonbonded atomic interactions. Protein Sci. 2, 1511–1519. Dubey, S.P., Kini, N.G., Balaji, S., Kumar, M.S., 2018. A review of protein structure prediction using lattice model. Crit. Rev. Biomed. Eng. 46, 147–162. Eisenberg, D., Lüthy, R., Bowie, J.U., 1997. [20] VERIFY3D: assessment of protein models with three-dimensional profiles. Methods Enzymol. 277, 396–404. El-Gebali, S., Mistry, J., Bateman, A., Eddy, S.R., Luciani, A., Potter, S.C., Sonnhammer, E.L.L., 2018. The Pfam protein families database in 2019. Nucleic Acids Res. 47, D427–D432. Emerson, I.A., Amala, A., 2017. Protein contact maps: a binary depiction of protein 3D structures. Phys. A. 465, 782–791. Funkenstein, B., Krol, E., Esterin, E., Kim, Y.S., 2012. Structural and functional characterizations of activin type 2B receptor (ACTRIIB) ortholog from the marine fish, gilthead sea bream, Sparusaurata: evidence for gene duplication of ACTRIIB in fish. J. Mol. Endocrinol. 49, 175–192.

149