Analysis of Amino Acid Changes in the Fusion Protein of Virulent Newcastle Disease Virus from Vaccinated Poultry in Nigerian Isolates

The roles of fusion gene in the virulence of Newcastle disease virus are well established, but the extent of its variation among the XIV, XVII, and XVIII genotypes reported in Central Africa and West Africa has until recently been understudied. In this study, virulent Newcastle disease virus (vNDV) was isolated from dead chickens among vaccinated flocks between March and April 2020. Fusion (F) gene was sequenced and analysed for characterization and information about genetic changes. Many substitutions were observed along the region and some of their functions are yet to be determined. Results showed that all study isolates have virulent cleavage site sequence 112-RRRKR-116/F117 and clustered within genotype XIVb. Sequence analysis showed K78R mutation in the A2 antigenic epitope in all isolates and more along the F-gene which varied in some instances within the isolates. Mutation in this A2 antigenic epitope has been reported to induce escape mutation to monoclonal antibodies generated using the NDV LaSota strain. The range of percentage nucleotide and amino acid homology between the study isolates and commercially available vaccine strains is 81.14%–84.39% and 0.175–0.211, respectively. This report provides evidence of vNDV among vaccinated chicken flock and molecular information about circulating vNDV strains in Kano State, Nigeria, which is useful for the development of virus matched vaccines. Newcastle disease (ND) surveillance and molecular analysis of circulating strains in this region should be encouraged and reported. Furthermore, ND outbreaks or cases among vaccinated poultry presented to veterinary clinics should be reported to the state epidemiologist. Nucleotide sequences were assigned accession numbers OK491971–OK491977.


Introduction
Newcastle disease (ND) is a globally reported viral disease afecting over 200 species of birds [1] primarily controlled by vaccination [2]. It is an Ofce International desÉpizooties (OIE) notifable disease [1]; however, only few countries report its incidence to OIE, especially in developing countries where the disease is enzootic [3] among vaccinated and unvaccinated poultry. ND has received extensive attention because of its ability to spread, high mortality, vaccine failure, and other economic losses associated. Te aetiology of ND is virulent strain of Avian paramyxovirus type-1 (APMV-1) also known as Newcastle disease virus (NDV) of the genus Orthoavulavirus belonging to the family Paramyxoviridae and order Mononegavirales [1]. It is a negative sense, nonsegmented, and single-stranded enveloped RNA virus [4].
NDV fusion (F) glycoprotein mediates fusion between the viral and host cellular membranes [5,6]. It is synthesized as an unreactive F0 precursor, containing 1662 nucleotides (nt) coding for 553 amino acids (aa) with an approximate 55 kDa weight [7,8]. F0 is proteolytically cleaved by specifc host cellular proteases at the peptide bond between residues 116 and 117 forming two disulphide linked polypeptides, F1 and F2 which are 48-54 kDa and 10-16 kDa, respectively [9,10]. Tere are several domains that have been identifed throughout the length of these polypeptides important for viral fusion activity [11].
Although Newcastle disease (ND) is said to be enzootic in Nigeria, little information exists on the molecular epidemiology and the lineage distribution of the Newcastle disease virus (NDV) in the country [3] and there is paucity on reports of virulent Newcastle disease virus (vNDV) strains obtained from dead/sick vaccinated animals in the outpatient veterinary clinic. Te importance of detection and pathotyping of NDV in understanding the epizootiology of the virus in any region cannot be overemphasized [12] especially with the growing need for evaluation of the efcacy of existing ND vaccines [13]. Presently, two live attenuated monovalent vaccines-LaSota and Komarov-are commercially available for the control of ND in intensively reared poultry in Nigeria, but there are reports of ND outbreak among vaccinated focks. Knowledge of the increasingly evolving genetic variation of vNDV is important for developing genetically matched vaccines which can prevent ND vaccine failure and maybe future outbreaks in the country.
Te aim of this study was to isolate and characterize Newcastle disease virus (NDV) full fusion (F) gene detected from dead chicken of vaccinated fock presented to a veterinary clinic for post-mortem examination in Kano State, Nigeria, during March and April, 2020. Details of amino acid mutation were noted, documented, and compared with previously reported vNDV isolates from Nigeria, West Africa, and commercially available NDV vaccine strains.

Animal Ethics Declaration.
Tis study did not include the use of live chickens. International and national guidelines for the care and use of animals were followed by experts at the veterinary clinic during post-mortem examinations and sample collection.

Sample Collection.
Pooled organ samples (proventriculus, spleen, and small intestine) and swabs (cloacal and tracheal) were collected in viral transport medium (VTM) from one hundred chicken cadavers with history of vaccination against NDV that was reported to a veterinary clinic located in Kano State metropolis during March-April 2020 from focks presenting with respiratory discomfort, weakness, greenish diarrhoea, anorexia, high mortality and morbidity, and drop in egg production in layers characteristic of ND. Te samples were labelled and transported immediately on ice and stored at −4°C until analysis was conducted.
Post-mortem examination was conducted on all specimens, and characteristic lesions were noted.

NDV Total Viral RNA Extraction, Reverse Transcription Polymerase Chain Reaction, and F-Gene Sequencing.
All experiments were carried out according to standard protocol. Viral RNA was extracted directly from pooled organ samples and swabs using the Quick-RNA TM Viral Kit (Zymo Research, USA) as specifed in the product manual. Fifty samples were processed successfully.
NDV M-gene was detected using protocols described [14]. Two overlapping fragments covering 1662 bp of the full F-gene were amplifed using two pairs of primers (Table 1). One-step RT-PCR was performed using One Taq one-step RT-PCR Kit (New England BioLabs Inc ). cDNA synthesis was achieved at 50°C for 30 minutes followed by incubation at 94°C for 15 minutes. RT-PCR was performed with 40 cycles of denaturing at 94°C for 30 seconds, annealing at 55°C for 1 minute, and extension at 68°C for 2 minutes with fnal extension at 68°C for 10 minutes, and PCR products were maintained at 4°C. Electrophoresis of PCR products was done on 1% ethidium bromide stained with 1.5% agarose gel at 90 volts and 120 Amps for 35 minutes and compared with a 100 bp DNA ladder, and amplifed products were visualized under ultraviolet (UV) illumination using gel documentation system-image capture (Biometra, Germany).
Te amplicons were sequenced by Macrogen Ltd (Netherlands), and sequences obtained were submitted on the NCBI database and assigned accession numbers OK491971-OK491977.

Phylogenetic Tree Construction and Evolutionary Distance
Analysis. Nucleotide sequence alignment, editing, and analysis were done using the Bioedit software (7.2.5). Nucleotide sequence similarity and molecular phylogenetic analysis was computed on MEGA (11.0.). Inferred evolutionary F-gene sequence of the study isolates, some reported NDV isolates, and known vaccine strains was conducted by the maximum likelihood method based on JJT matrix-based method using 1000 bootstrap replicates.

Characteristic Pathological Lesions on Organs Seen during Post-Mortem Examination. 3.2. Molecular Detection of NDV.
Te overall M-gene detection rate by RT-PCR was 54% (27/50). A 121-bp fragment was amplifed, and products of electrophoresis were visualized by ultraviolet (UV) trans-illumination.
F-gene amplifcation was successfully carried out in seven samples, and products of electrophoresis were visualized by UV trans-illumination ( Figure 1). 1662 bp nucleotide sequences obtained were deposited in the GenBank (details are given in Table 2).

Phylogenetic Tree and Evolutionary Distance Analysis.
Phylogenetic tree was constructed (Figure 2) using MEGA (11.0). Te genetic relatedness of the study isolates, reference KU665482.1 LaSota.71.IR/2016, eight commercially available vaccine strains and NDV F-gene sequences obtained from GenBank database was inferred by phylogenetic analysis. All study isolates clustered around the newly classifed genotype XIV (sub-genotype XIVb) in class II which is widely reported in Nigeria.
Nucleotide blast analysis shows a 99% nucleotide identity to virulent NDV MT543153 isolated in 2019 from backyard poultry in Niger (a country to the north border of Nigeria). Te evolutionary divergence as nucleotide (nt) homology and amino acid (aa) homology between study isolates and commercially available vaccines is shown Tables 3 and 4.

Molecular Characterization and Mutational
Analysis of the Functional and Antigenic Domains. Te complete translated 553 fusion protein amino acid sequences obtained from the study isolates were used to compare their functional and antigenic domains relative to nine vaccine strains using KU665482.1 LaSota.71.IR/2016 as reference, six vNDV strains previously reported from Nigeria, and two vNDV strains isolated from West Africa (Tables 5-10). Notable substitutions around these regions were observed. Interestingly, among the research isolates, these substitutions were sometimes observed diferently. Numbering system of amino acid (aa) was used to name the detected aa substitutions with respect to observed genetic variations.
All seven isolates share the characteristic virulent motif 112 R-R-R-K-R/F 117 at the F0 cleavage site indicating that they are velogenic NDV strains (Figure 3(a)). G112R, Q114R, and G115K have been observed in study isolates (Tables 5, 7, and 8). Along the fusion peptide region (117-142), fve aa substitutions, L117F, I118V, I121V,  International Journal of Microbiology G124S, and I135V, are seen (Tables 5, 8, and 9). L117F and I118V are expected in the virulent furin-like molecule. In addition, the F protein has six highly conserved potential N-linked glycosylation sites Ng 1 -Ng 6 [18] with sequence Asn (Asparagine)-X-Ser(Serine)/Tr(Treonine) (N-X-S/T) where X is any aa except proline and aspartate [19,20]. Amino acids at these sites were used and conserved in all NDV isolates of this research at residues 85NRT, 191NNT, 366NTS, 447NIS, 471NNS, and 541NNT. Hence, there was no loss of glycosylation site though there was one substitution compared to the LaSota strain at residue 191NKT ( Figure 3(a)). However, nt substitutions occurred in diferent patterns at these regions among study isolates which resulted in same sense mutation. Cysteine residues are important in the connection between F1 and F2 sub-units to maintain the F protein structure. Cysteine residues are conserved at positions 25,27,76,199,338,347,362,370,394,399,401,424,514, and 523 of the F protein in most NDV isolates [20]. Amino acids are used and conserved in all the isolates except for a unique point cysteine (C) to Serine substitution at residue 394 in OK491977 leading to loss of one cysteine residue. Several nt substitutions were observed in the region which resulted in same sense mutation (Figure 3(b), Tables 6 and 9).   Te number of amino acid substitutions per site between sequences is shown. Analyses were conducted using the Poisson correction model [17]. Tis analysis involved 24 amino acid sequences. All positions with less than 95% site coverage were eliminated, i.e., fewer than 5% alignment gaps, missing data, and ambiguous bases were allowed at any position (partial deletion option). Tere were a total of 1662 positions in the fnal dataset. Evolutionary analyses were conducted in MEGA11 [16].
Te efective B-cell epitope regions 157-171 involved in virus neutralization are surface-exposed amino acids which may cause antigenic diference between the vaccine and wild strains [19]. Amino acid sequence analysis of the isolates shows no aa substitutions within this region (Figure 3(a))

Discussion
NDV management in Africa is complicated [21]. Te economic impact of ND among vaccinated and unvaccinated        However, despite mass administration of vaccines, there are reports of vaccine failure and high mortality among vaccinated focks [22], sub-optimal protection levels among vaccinated focks [23], viral evolution and identifcation of new genotypes [19,21], and signifcant antigenic distance between circulating and vaccine strains [24]. Despite huge patronization of veterinary clinics (for drugs and vaccines) and consultancy by poultry producers, to the best of our knowledge, this is the frst report of isolation of vNDV from dead chickens from vaccinated fock presented to a veterinary clinic in Nigeria for post-mortem examination. Tis research also provides details of F-gene sequence of seven clinical vNDV isolated from Kano State, Nigeria. Seven genotypes (I, II, IV, VI, XIV, XVII, and XVIII) have been reported to circulate in Nigeria suggesting high genetic diversity of NDV for one country [25]. A phylogenetic tree was constructed based on the full F gene of the research isolates to determine their genotype. Isolates clustered among genotype XIVb in class II similar to earlier reports of newly classifed genotypes XIV, XVII, and XVIII are identifed as the circulating strains in Central Africa and West Africa [2,21,24]. Te NDV isolates reported here originated from diferent poultry farms within and outside the metropolis of Kano State. Post-mortem lesions include haemorrhagic intestinal ulcers, haemorrhagic caecum tonsils, haemorrhagic and infamed proventriculus, and haemorrhagic trachea, among others (Figure 4). Tese lesions agree with overt clinical signs reported by poultry handlers such as difculty in breathing, greenish diarrhoea, weakness, and anorexia with wing and leg paralysis as the most common neurological symptoms among focks. Haemorrhage at the tip of the proventriculus is highly suggestive of ND [26,27]. Generally, distributed lesions suggest that the virus is able to infect and replicate in most organs, typical of vNDV as supported by the cleavage site motif of the isolates.
In a previous review report, phylogenetically, the Nigerian genotype XIVa isolates form a cluster with some strains in Niger Republic while genotype XIVb isolates tend to be more closely related to the 2009 isolates from Benin Republic [24]. Tey further stated that the isolates in genotype XIVa that share the highest nucleotide similarity with those from Niger Republic were all obtained from Sokoto State which shares a direct international border with Niger Republic. Isolates reported here show 99% homology with MT543153 NDV/ chicken/Niger/89/2019 (obtained on GenBank as on 15th February, 2022) isolated in 2019 from Niger Republic. It has been reported that importation of poultry in Niger Republic is an informal sector with porous borders between Nigeria and Burkina Faso that allow for the entry of live poultry (chicken and guinea fowl) without prior registration and customs clearance. Informal exportation of poultry from Niger is primarily to Nigeria [28]. Tis close phylogenetic relationship can be explained by the cross border movements between the two countries which may facilitate the spread of the virus. Live bird trading is commonly practised within West Africa, and birds can be moved over long distances and across porous borders. In Nigeria, environmental factors, high demand, and movement of birds tally with increased incidence of NDV [29,30]. Sub-genotype XIVb has been isolated in a commercial farm in Nigeria that has vaccinated with unspecifed vaccine strain against NDV [21]. With recent isolation of genotype XIVb among fock vaccinated with LaSota and     Figure 3: Molecular phylogenetic analysis based on the nucleotide sequences of the F-gene of study isolates, some reported NDV isolates and known vaccine strains using bootstrap consensus of 1000 replicates. Te evolutionary history was inferred by using the Maximum Likelihood method and JTT matrix-based model [15]. Te tree with the highest log likelihood (-20685.06) is shown. Te percentage of trees in which the associated taxa clustered together is shown next to the branches. Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the JTT model, and then selecting the topology with superior log likelihood value. Tis analysis involved 61 amino acid sequences. Tere were a total of 1662 positions in the fnal dataset. Evolutionary analyses were conducted in MEGA11 [16].
Komarov vaccine strains in Kano State, Nigeria, the spread of this genotype to other regions of the continent may spread vaccine resistance as well. Te F protein is capable of provoking host immune response, and it is necessary for producing neutralizing antibodies against NDV induced by vaccines [19]. Mutations along this gene will impact antibody production which will be heterologous even at the level of F-gene [31]. Percentages of nucleotide identity and aa homology between the study isolates and commercially available vaccines ranges between 81.14 and 84.39% and 0.175-0.211 respectively (Tables 3 and  4 [32]. Tis is similar to reports by [21], based on the analysis of the complete F coding sequences reported, that all the virulent NDV types circulating in Nigeria are shown to be distantly related to currently available vaccine strains in the country.
Te evolutionary distance between vaccine and the circulating feld strains is an important factor in efective disease control since it explains the continuous occurrence of ND outbreaks despite the extensive poultry vaccination programme which has also been reported in Nigeria [24]. Genomic divergence and antigenic divergence between wild infective strains and vaccine strains are among reported causes of ND vaccine failure [19]. Most commonly used ND vaccine strains including LaSota were developed in the 1950s and 1960s [32] and show considerable degree of genetic divergence from currently circulating vNDV wild strains [31]. LaSota is classifed under genotype II, while most commonly reported wild vNDV strains in the 21 st century are found among genotype VII [24]. Existing antigenic variations among West Africa strains and the LaSota vaccine may afect its protective efcacy to confer protection against all West African strains [33]; in cases where the LaSota vaccine provided protection against clinical disease, it did not prevent infection and viral shedding [33]. Although all NDV strains belong to one serotype, protection provided by genotype II vaccines against heterologous challenge has been recently under controversy [31] with several reports of LaSota vaccine failing to provide complete protection against morbidity and or mortality during experimental heterologous vNDV challenge necessitating the growing need for development of antigenically matched vaccines to circulating strains [12].
Te fusion protein is a major target for the immune response, and immunity raised against this protein is effective in the neutralization of NDV infectivity [34]. In this report, several nt substitutions occur at specifc and conserved antigenic sites which in some instances resulted in aa substitution (mutation) (Tables 7-10). Te neutralizing  epitopes are important in forming antigenic epitopes, and aa substitution in this region induces the formation of neutralizing escape variants [20,[35][36][37]. Te single point K78E mutation of the A2 antigenic epitope seen in all study isolates is due to AAA-AGA nt substitution at codon 232-234 (Table 7) which has been previously reported to induce escape mutation in NDV Beaudette-C clone against MAbs generated using LaSota strains [5]. Te role of this mutation in genotype XIVb requires more attention. Tough the scope of this research did not involve the generation of K78E mutants to test LaSota efcacy, it is notable from samples and data collected that though the focks were vaccinated, ND is still reported. Te fusion peptide domain between aa residues 117 to 142 is a conserved hydrophobic region located at the amino end of the F1 polypeptide and has been reported to insert into the target membrane to initiate membrane fusion [38,39]. Phenylalanine (F) residue at position 117 has been reported as a major contributor to neurological symptoms [40,41]. All study isolates show F117, and paralysis of leg and wings was highly reported among the focks. Previous report shows that mutation along the fusion peptide domain inhibits fusion afecting syncytia formation, content mixing, and hemifusion [42]. 128-Glycine residue afects folding of the molecule [43]. Among study isolates, I118V, I121V, G1124S, and I135V were observed (Table 5), and the efects of some of these mutations have not been established and require further study.
Tere is loss of one conserved cysteine residue in study isolate OK491977 Avian orthoavulavirus 1 isolate KN75 due to TGC to TGT nt substitution at codon 1180-1182 causing unique point cysteine (C) to serine substitution at residue (C394S mutation). Cysteine residues have been reported to enhance F1 and F2 disulphide linkages [44].
Te fusion gene has three heptad repeat (HR) regions, HRa, HRb, and HRc at positions 143-185, 268-299, and 471-500, respectively [19]. K145, N272Y, S278P, I285K, T288N, N297K, N476T, N479D, E482A, R486N, K494R, and T498S were observed among isolates (Table 6). Tere is loss of 285-isoleucine residue in isolates OK491973 and OK491975; however, the efect of substitution of hydrophobic isoleucine with hydrophilic lysine at this point on the leucine zipper motif [45] is unknown. Replacing glutamic acid with alanine at position 482 (E482A) has been reported to have little or no efect on the fusogenic activity of the protein [45]. Predicted glycosylation sites on the F gene are well conserved in all isolates, and hence no impairment is expected in viral structure, during viral replication or virulence mediated by diferential glycosylation.
Mutations occur over time along the F gene due to response to vaccine or drug pressure [46][47][48]. In Nigeria, there is report of sub-optimal vaccine dose administration, excess intake of vaccine by focks, improper handling and storage of vaccines, and indiscriminate and prolonged exposure of poultry to antibiotics for preventive or therapeutic purposes for vaccine preventable infection and disease with evidence of antibiotic residue in poultry products: meat, ofals, and eggs [49,50]. Due to degeneracy of aa, not all nt substitution resulted into aa mutation in this report. However, continuous vaccine and drug pressure may cause mutation at somewhat conserved regions as NDV continuously evolves. Furthermore, these results suggest that there should be improved reporting of ND outbreak among vaccinated fock in Kano State and Nigeria as a whole.

Conclusion
Virulent NDV genotype XIVb was detected among vaccinated poultry in Kano State, Nigeria. Fusion gene of seven vNDV strains was successfully sequenced, and details are available on NCBI GenBank (OK491971-OK491977). Based on amino acid analysis, several mutations were seen and reported along the F-gene compared to commercially available vaccines. Whether these mutations individually or combined afect antigenicity of the virus remains to be established. Based on nucleotide analysis, all isolates show high degree of antigenic variability from commonly used LaSota and Komarov ND vaccines in Kano State and others commercially available. Although caution is warranted in considering genetic distance between NDV vaccines and the challenge virus as the sole cause of the reported cases of vaccine escape in feld report, [51] the isolation and detection of vNDV genotype XIVb among vaccinated fock require further research and especially the role of the A2 antigenic epitope in inducing antibody escape mutation among that sub-genotype. Furthermore, the role of Genotype II NDV strain vaccine pressure on evolutionary change among Class XIVb vNDV require further evaluation. Te role of Genotype II NDV strain vaccine pressure on evolutionary change among Class XIVb vNDV require further evaluation. In order to generate genetically matched vaccines for this region, ND surveillance and molecular analysis of circulating strains should be encouraged and reported.

Data Availability
Te F-genes of all seven vNDV detected and characterized during this research are available with no restriction on NCBI database with their corresponding accession numbers.

Disclosure
Tis work was part of the PhD study of Olubukola O. Funsho-Sanni. Tis paper is available online as a preprint [52]. responsible for data sourcing, fgures, and tables. Olufunsho S. Sanni, Lawal D. Rogo, and Olubukola O. Funsho-Sanni were responsible for original draft preparation. Ismaila Shittu was responsible for F-gene primer design and protocol. Elijah E. Ella, Helen I. Inabo, and Sodangi A. Luka were responsible for supervision and validation. All authors were responsible for review and editing.