This study was conducted on Bengal ducks under organized farming system. This investigation was carried out at a livestock farm of West Bengal University of Animal and Fishery Sciences, Belgachia, Kolkata, West Bengal. The farm data and samples from ducks were collected from March to May, 2020. Later, samples were processed for appropriate molecular biology work. The annual mean temperature is 26.8°C (80°F); monthly mean temperatures range from 19°C to 30°C (67°F to 86°F), and maximum temperatures can often exceed 40°C (104°F) during May–June.

In total, 60 healthy indigenous ducklings (Anas platyrhynchos) were used for this study. The ducklings were properly maintained under standard feeding and management. Duck eggs were collected from the field and subjected to hatching for an average of 28 days and reared in our farm from day-old stage. The ducklings (day old to 2 weeks of age) were given a starter ration containing 20% crude protein and 2,750 Kcal ME per kg. The first growing phases (3–8 weeks) received the feed composed of 18% crude protein and 2,750 kcal ME per kg, and the second growing phases (9–20 weeks) received the feed composed of 15% crude protein and 2,700 kcal ME per kg (22). Studies were conducted with ethical approvals obtained from the Institutional Animal Ethics Committee, West Bengal University of Animal and Fishery Sciences.

At the end of each week, the ducks were weighed using a digital weighing balance to determine the body weight in grams. Other morphometric parameters, i.e., shank length, breast width, keel length, drumstick length, neck length, body length, chest girth, and wing length, were measured with measuring tape for 8 weeks and expressed in centimeters (cm). Daily body weight gain was assessed through a computational approach.

Based on the available data, the entire duckling population data are classified into high body weight (better growing) as values above mean + standard deviation, and low body weight (less growing) as values less than mean – standard deviation. Trends for average daily body weight gain have been depicted. Similar groups were also observed for average daily body weight gain and biomorphometric traits of the ducks.

Duodenum and cecum were collected from the duck belonging to the two groups, namely, higher and lower body weights. As a routine farm operation, at the laying stage, we maintain male:female as 1:10, and we sell the surplus males. It was clearly mentioned that based on the available data, the entire duckling population data were classified into high body weight (better growing) as values above mean + standard deviation and low body weight (less growing) as values less than mean – standard deviation.

Six ducks (males) from each group, high body weight and low body weight, were sold as a routine farm operation. Individual birds were already tagged at a very early age of their life. Samples of duodenum and cecum were collected in TRIzol maintaining cold chain and other norms from the slaughterhouse from the same birds (already tagged) sold to them.

All experiments were conducted in accordance with relevant guidelines and regulations of the Institutional Animal Ethics Committee, and all experimental protocols were approved by the Institutional Biosafety Committee, West Bengal University of Animal and Fishery Sciences, Kolkata.

The total RNA was isolated from the duodenum and cecum of Duck by TRIzol method (15, 23–26) and was further used for cDNA synthesis.

Taq DNA polymerase, 10× buffer, and dNTP were purchased from Invitrogen, and SYBR Green qPCR Master Mix (2× ) was obtained from Thermo Fisher Scientific Inc. (PA, USA). L-Glutamine (Glutamax 100× ) was purchased from Invitrogen Corp. (Carlsbad, CA, USA). Penicillin-G and streptomycin were obtained from Amresco (Solon, OH, USA). Filters (Millex GV. 0.22 μm) were purchased from Millipore Pvt. Ltd. (Billerica, MA, USA). All other reagents were of analytical grade.

The 20 μl reaction mixture contained 5 μg of total RNA, 0.5 μg of oligo dT primer (16–18 mer), 40 U of ribonuclease inhibitor, 10 M of dNTP mix, 10 mM of DTT, and 5 U of MuMLV reverse transcriptase in reverse transcriptase buffer. The reaction mixture was gently mixed and incubated at 37°C for 1 h. The reaction was stopped by heating the mixture at 70°C for 10 min and chilled on ice. The integrity of the cDNA was checked by PCR. To amplify the full-length open reading frame (ORF) of gene sequence, a specific primers pair was designed based on the mRNA sequences of Gallus gallus by DNASTAR software. The primers have been listed in Table 1; 25 μl of the reaction mixture contained 80–100 ng cDNA, 3.0 μl of 10× PCR assay buffer, 0.5 μl of 10 mM dNTP, 1 U Taq DNA polymerase, 60 ng of each primer (as in Table 1), and 2 mM MgCl₂. PCRs were carried out in a thermocycler (PTC-200, MJ Research, USA) with cycling conditions such as initial denaturation at 94°C for 3 min, denaturation at 94°C for 30 s, varying annealing temperature (as mentioned in Table 1) for 35 s, and extension at 72°C for 3 min for 35 cycles followed by a final extension at 72°C for 10 min.

PCR amplicons verified by 1% agarose gel electrophoresis (Supplementary Figures) were purified from gel using a gel extraction kit (Qiagen GmbH, Hilden, Germany) and ligated into a pGEM-T easy cloning vector (Promega, Madison, WI, USA) following the manufacturer's instructions. The 10 μl of the ligated product was directly added to 200 μl competent cells, and heat shock was given at 42°C for 45 s in a water bath, and cells were then immediately transferred on chilled ice for 5 min, and SOC was added. The bacterial culture was pelleted and plated on an LB agar plate containing ampicillin (100 mg/ml) added to the agar plate 1:1,000, IPTG (200 mg/ml), and X-Gal (20 mg/ml) for blue-white screening. Plasmid isolation from overnight-grown culture was done by the small-scale alkaline lysis method. Recombinant plasmids were characterized by PCR using gene-specific primers and restriction enzyme digestion based on the reported nucleotide sequence for chicken. The enzyme EcoRI (MBI Fermentas, USA) is used for fragment release. Gene fragment inserts in the recombinant plasmid were sequenced by an automated sequencer (ABI prism) using the dideoxy chain termination method with T7 and SP6 primers (12, 18, 20, 27–32).

The nucleotide sequence so obtained was analyzed for protein translation, sequence alignments, and contigs comparisons by DNASTAR Version 4.0, Inc., USA. The novel sequence was submitted to the NCBI Genbank, and an accession number was obtained which is available in the public domain now.

The predicted peptide sequence of Mucin 2 of indigenous duck was derived by Edit sequence (Lasergene Software, DNASTAR) and then aligned with the peptide of other chicken breeds and avian species using Megalign Sequence Programme of Lasergene Software, DNASTAR (12, 14, 18, 20, 27–29, 31–35).

Prediction of the signal peptide of the Mucin 2 gene was conducted using the software (Signal P 3.0 Sewer-prediction results, Technical University of Denmark). Estimation of leucine percentage was conducted manually from the predicted peptide sequence. Disulfide bonds were predicted using suitable software (http://bioinformatics.bc.edu/clotelab/DiANNA/) and by homology search with other species.

A protein sequence-level analysis study was carried out with specific software (http://www.expasy.org./tools/blast/) for the determination of leucine-rich repeats (LRR), leucine zipper, N-linked glycosylation sites, detection of leucine-rich nuclear export signals (NES), and detection of the position of GPI anchor. Detection of leucine-rich nuclear export signals (NES) was carried out with the NetNES 1.1 server, Technical University of Denmark (35). Analysis of O-linked glycosylation sites was carried out using the NetOGlyc 4 server (http://www.expassy.org/), whereas the N-linked glycosylation site was detected by the NetNGlyc 1.0 software (http://www.expassy.org/) (36). Detection of leucine-zipper was conducted through the Expassy software, Technical University of Denmark (37). Regions for alpha-helix and beta-sheet were predicted using NetSurfP (38). Protein Surface Accessibility and Secondary Structure Predictions, were analyzed as per the Technical University of Denmark. Domain linker prediction was done according to the software developed (39). LPS-binding sites (40) and LPS-signaling sites (41) were predicted based on homology studies with polypeptides of other species.

The templates that possessed the highest sequence identity with our target template were identified by using PSI-BLAST (http://blast.ncbi.nlm.nih.gov/Blast). The homology modeling was used to build a 3D structure based on homologous template structures using the PHYRE2 server (42). The 3D structures were visualized by PyMOL (http://www.pymol.org/), which is an open-source molecular visualization tool. Subsequently, the mutant model was generated using the PyMoL tool. The Swiss PDB Viewer was employed for controlling energy minimization. The structural evaluation along with a stereochemical quality assessment of the predicted model was carried out by using SAVES (Structural Analysis and Verification Server), which is an integrated server (http://nihserver.mbi.ucla.edu/SAVES/). The ProSA (Protein Structure Analysis) webserver (https://prosa.services.came.sbg.ac.at/prosa) was used for the refinement and validation of protein structure (43). The ProSA was used for checking model structural quality with potential errors, and the program shows a plot of its residue energies and Z-scores, which determine the overall quality of the model. The solvent accessibility surface area of the mucin 2 gene was generated by using the NetSurfP server [http://www.cbs.dtu.dk/services/NetSurfP/; (38)]. It calculates the relative surface accessibility, Z-fit score, the probability for alpha-helix, the probability for beta-strand and coil score, etc. TM align software was used for the alignment of 3D structure of IR protein for different species and RMSD estimation to assess the structural differentiation (44). PDB structure for 3D structural prediction of the mucin 2 gene for the duck was carried out through the PHYRE software. Protein–protein interactions have been studied through the String analysis (45).

Total RNA was estimated from the duodenum and cecum of duck from high-body weight and low-body weight groups by the TRIzol method, and the quantitative analysis of total RNA was performed using formaldehyde gel electrophoresis. The 28S rRNA and 18S rRNA demarcated the quality of RNA. First-strand cDNA was synthesized by the process of reverse transcriptase polymerase chain reaction (RT-PCR) in the automated temperature-maintained thermocycler machine. M-MLVRT (200 u/μl) was used as a reverse transcriptase enzyme. All the primers were designed using the primer 3 software (v. 0. 4.0) as per the recommended criteria. The primers used are listed in Table 1. An equal amount of RNA (quantified by Qubit fluorometer, Invitrogen), wherever applicable, was used for cDNA preparation (Superscript III cDNA Synthesis Kit; Invitrogen). All qRT-PCRs were conducted on ABI 7500 fast system. Each reaction consisted of 2 μl cDNA template, 5 μl of 2× SYBR Green PCR Master Mix, 0.25 μl each of forward and reverse primers (10 pmol/μl), and nuclease-free water for a final volume of 10 μl. Each sample was run in duplicate. Analysis of real-time PCR (qRT-PCR) was performed by the delta–delta–Ct (ΔΔCt) method (32–34).

The entire reactions were performed in triplicate (as per MIQE Guidelines), and the experiment was repeated two times, in 20 μl reaction volume, using FastStart Essential DNA Green Master (Himedia) on ABI 7500 system.

The duodenum and cecum samples were fixed in formalin (10%), embedded in paraffin, and processed for histological examination and stereology. The tissues were submerged in Lillie fixative for 1 week at room temperature and then were processed and embedded vertically in paraffin wax. Then, each sample was exhaustively sectioned into 4 μm-thick sections by a fully automated rotary microtome (Leica RM2255, Germany). Each of these sections was stained with hematoxylin and eosin and mounted. From each sample, 10–15 sections were chosen by the systematic uniform random sampling (SURS) method.

Paraffin tissue blocks were prepared by standard manual alcohol-acetone protocol. The 5–6 μm thick paraffin sections obtained from the duodenum of both the high-growth and low-growth ducks were taken on the Millennia 2.0 adhesion slides (Cat. No. 71863-01, Abcam). The tissue sections were de-paraffinized and hydrated in distilled water. The tissue slides were covered with trypsin enzymatic antigen-retrieval solution (Cat. No. ab970, Abcam) and kept in an incubator in a humid environment at 37°C for 5–10 min. The sections were then incubated for 60 min in peroxidase blocking solution (Lot. No. 00065614, Dako) at room temperature to block non-specific antibody-binding activity. After subsequent washing with phosphate-buffered saline (PBS), the sections were incubated at 37°C for 2 h in a humid environment with mouse monoclonal anti-mucin antibody in 1: 200 dilution. Immunoreactivity was detected after 1-h incubation at 37°C with a secondary antibody, Rabbit anti-mouse IgG H&L (HRP Conjugated, Cat. No. ab6728; Abcam) in dilution 1:200. Slides were then rinsed three times in PBS for 5 min each, followed by treatment with freshly prepared DAB solution for 3 min (DAB substrate, Cat. No. 34001, Thermo Fisher Scientific). The sections were counter-stained with Mayer's hematoxylin, hydrated in ethanol, cleared in xylene, and then mounted in DPX.

Descriptive statistics with mean and standard error were estimated through the SYSTAT package for the expression level analyzed through real-time PCR and presented accordingly in the graph.

The model employed:

Where, Y eijk = kth observation of the target trait, μ= overall mean, Gi = fixed effect of the ith genotype corresponding to the expression level, Aj = fixed effect of the jth bird, eijk = random error. Expression level with real-time PCR was estimated as 2^−ΔΔCt.

Mucin 2 coats the epithelia of the intestines, airways, and other mucus membrane-containing organs. Certain important domains for mucin 2 protein have been detected. The total sequence length of the mucin 2 gene in Anas platyrhynchos is 12.013 KB. In this study, we could amplify the partial mucin 2 gene of Bengal duck as 11,968 bp, with gene bank accession number as OQ789085. In this study, we have reported mucin 2 gene sequence for ducks for the first time.

It was attempted to explore the growth potential ability of the domains. The signal peptide was detected at amino acid position 1–18 for the predicted 3D protein structure of the mucin 2 gene. An important domain has been identified as the VWFD (von Willebrand factor domain) at amino acid sites of 33–236 (red sphere), 388–601 (forest green), 859–1,065 (orange sphere), and 2,945–3,164 (Figure 1). Other identified domains were VWFC-1 at sites of amino acids 3,293–3,343 and 3,402–3,448. VWFC-2 domains were predicted at the amino acid location of 3,275–3,344 and 3,382–3,449. Other two important sites were CTCK1 (C terminal cystine knot signature) and CTCK2 (C terminal cysteine knot domain profile). The sites for the domain of CTCK1 are 3,580–3,618 and CTCK2 at aa position 3,532–3,619. Sites for disulfide bonds were detected at aa positions 56–64, 411–419, 882–890, 2,968–2,976, 3,532–3,581, 3,557–3,611, and 3,561–3,613 (Figure 2).

In the current study, we are concerned with growth parameters for the mucin gene. We had identified an important domain involved with growth as EGF1 (EGF-like domain signature) at sites 3,296–3,307. Certain important observations were revealed. Since disulfide bonds were not included in this region, it is presumed to be open. EGF1 domain is inserted within VWFC-1 and VWFC-2.

The sites for glycosylation are very important. The sites for N-linked glycosylation and O-linked glycosylation have been listed in Supplementary material; 22 sites for N-linked glycosylation and O-linked glycosylation have been predicted.

The pictorial description for the secondary structure of the mucin 2 gene is described in Figure 3. It represents the site for the alpha helix, beta sheet, and loop structure for the mucin gene.

In this study, we studied the growth parameters of the indigenous duck of Bengal – Bengal duck. Body weight was studied of the ducklings just after hatching up to the 10th week of age (Table 2). Figure 4 represents the diagrammatic representation of body weight at successive ages of brooding of Bengal ducks up to the 10th week of age. Accordingly, the average daily body weight gain was also recorded at successive ages up to the 10th week (Table 3). Figure 5 shows the diagrammatic representation of the average daily body weight gain for both the groups of ducklings – Group I (higher growth) and Group II (lower growth).

Biomorphometric characteristics are useful parameters for assessing growth parameters. They also serve as an indirect way to assess the body weight of the birds. The average biometry characteristics for adult male and female ducks including pool average for both higher (Group I) and lower groups (Group II) have been listed in Table 4. Significant differences in growth parameters were observed for the body length of the Bengal duck. Figure 6 represents the morphometric characteristics of adult Bengal ducks from Group I (higher growth) and Group II (lower growth).

The growth curve for ducks at successive stages of life based on body weight is represented in Figure 7, whereas Figure 8 depicts the growth trend with respect to daily body weight gain.

Mucin is a protein expressed in the inner layer of the epithelium of some organs. It is commonly observed in the gut of livestock and poultry. In this study, we studied the expression profiling of two important organs – duodenum (part of small intestine) and cecum (part of large intestine) of Bengal ducks with respect to high- and low-body weights.

Figure 9 reflects the better expression profile for the mucin gene in Group I compared to that of Group II for duodenum and cecum, respectively. The expression for the mucin gene was observed to be more than 5-fold in Group I compared to that of Group II. It is evident from Figure 9 that although a similar trend of a better expression of mucin is observed for both the organs for the gut, mucin is better expressed in the duodenum compared to that of the cecum. The line graph represents a steep line in mucin expression in the duodenum, compared to that of the cecum. The expression of the mucin gene in Group I was observed to be more than 6-fold (6.29) in the duodenum compared to that of the caecum. Similarly, expression of mucin gene was observed to be more than 5-fold (5.3) in duodenum compared to that of cecum.

Light microscopic images of Mayer's hematoxylin-stained optical sections of the duodenum and cecum have been depicted in Figures 10A, B, respectively.

Immunohistochemistry for the expression of mucin 2 protein in the duodenum of both the high-growth group and the low-growth group has been represented in Figures 11A, B, respectively.