Molecular Characterization of Velogenic Avian Avulavirus Type 1 Isolated from Apparently Healthy Emu Birds: Implications to Viral Maintenance and Spread

Newcastle disease (ND) though identified nearly a century ago, still continues to be a major challenge to the poultry producers. Though several vaccination strategies are being adopted at the field level to control ND, outbreaks appear regularly either due to involvement of a new genotype, asymptomatic birds acting as reservoirs of virulent Avian Avula virus-1 (AAvV-1) or due to inadequate vaccination. AAvV-1 belongs to genus Avian orthoavulavirus 1 (AOAV-1) within the subfamily Avulavirinae of the family Paramyxoviridae (ICTV, 2019). AAvV-1 causes infections in a wide range of domestic and wild birds worldwide. As many as 241 species from 27 orders of the avian family are reported to be susceptible to ND (Kaleta and Baldauf, 1988). Fusion protein cleavage site (FPCS) is one of the most important virulence factors that distinguishes an isolate to be lentogenic or velogenic (Office International des Epizooties, 2012). Further, based on the full - length fusion gene sequences, AAvV-1 has been classified into two major classes, I and II. All the avirulent isolates were classified under Class I, which includes a single genotype with three subgenotypes. Class II contains most of the virulent viruses circulating worldwide, classified into 21 genotypes with multiple subgenotypes (Dimitrov et al., 2019) and this diversity continues to increase as the surveillance studies improves.

Presence of virulent viruses in healthy birds and their constant evolution over time suggests existence of a high environmental over load with continuous replication of AAvV-1 strains in endemic countries (Wajid et al., 2017). Indian poultry farming systems presents a unique opportunity to study viral maintenance and spread. India has a wide variety of poultry species; however limited information is available regarding the potential of these avian species as asymptomatic reservoirs of AAvV-1. To understand the role of emu birds in the dissemination of potential pathogens, a surveillance study was undertaken in organized emu flocks of Andhra Pradesh and Telangana, India. Here we report isolation of velogenic AAvV-1 from apparently healthy emu birds which was proven to be highly pathogenic to chicken by infection studies.

Ethical statement
 
Animal experimental protocols performed in the study are approved by Institutional animal ethics committee of NTR College of Veterinary Science, Gannavaram, Andhra Pradesh, India (Approval number: 5/IAEC/NTRCVSC/2018).
 
Virus isolation and confirmation
 
During the years 2017 to 2019, both oropharyngeal and cloacal swabs (n=175) were collected from apparently healthy emu flocks of Andhra Pradesh and Telangana states. Swabs were collected in sterile containers with sterile phosphate buffered saline (PBS) pH 7.0 -7.4 containing antibiotics (100 units of penicillin G, 100 µg of streptomycin and 0.25 µg of amphotericin B/ ml), transported on ice to the lab of Department of Veterinary Microbiology, NTR College of Veterinary Science and stored at -80°C until processing. The samples were processed following standard protocols of virus isolation as specified in Terrestrial manual of OIE (Office International des Epizooties, 2012). In brief, the supernatant fluids of oropharyngeal and cloacal swabs were centrifuged at 5000 g for 10 minutes at 4°C and filtered. The filtrates were added with antibiotic- antimycotic solution and used for inoculation of 9-11 day old chicken embryos through allantoic cavities as previously described (Office International des Epizooties, 2012). The presence of AAvV-1 in allantoic fluid was confirmed by hemagglutination inhibition (HI) assay using polyclonal serum raised against LaSota and RT- PCR targeting fusion gene encompassing FPCS (Office International des Epizooties, 2012).
 
RNA extraction, cDNA synthesis, PCR and fusion gene sequencing
 
Viral RNA was extracted from the allantoic fluid using TRIzol LS reagent (Genei, Bangalore). iScript TM cDNA synthesis kit (Biorad, USA) was used for synthesis of first strand cDNA using random hexamers. Presence of virus was confirmed by RT-PCR using primers targeting FPCS of fusion gene (Nantha Kumar et al., 2000). Complete fusion gene was amplified using consensus primers (Gowthaman et al., 2018) and used for phylogenetic analysis and genotype determination. Services of commercial sequencing firm (Barcode Biosciences, India) were utilized for obtaining the sequence of complete fusion gene.
 
Pathogenicity indices and infection of CEF
 
The pathogenicity indices, including the mean death time (MDT) and the intra-cerebral pathogenicity index (ICPI) were determined as described previously (Office International des Epizooties, 2012). Virus infection of CEF cells was performed according to the methods described previously (Freshney, 2015).
 
Phylogenetic analysis
 
Homology searches were conducted using the NCBI program BLAST and AAvV-1 reference serotype sequences were retrieved from the GenBank data base. Multiple alignments of the sequences were performed using Clustal W in MEGA version X (Kumar et al., 2018). Following multiple alignment, maximum likelihood values were determined to calculate percent identity and diversity, to generate phylogenetic trees and to determine the genotype of the isolate under study. The significance of all deduced phylogenetic trees were verified by bootstrap analysis of 1000 replicates. The maximum likelihood method based on the General Time Reversible (GTR) model with a discrete gamma distribution (five categories [+G]) was utilised for tree construction. All positions containing gaps and missing data were eliminated. There were a total of 1662 positions in the final dataset. The tree with the highest log likelihood (-15133.69) is shown. The evolutionary relationship between the Emu/5 isolate and 41 representative isolates of 21 genotypes of class II AAvV-1 were estimated by calculating the number of base differences per site from averaging over all sequence pairs between groups using MEGAX (Kumar et al., 2018). Details of the accession numbers used for phylogenetic analysis were mentioned in the supplementary material (S1).


 
Virus infection of chicken
 
The AAvV-1 isolate was further characterized to understand the early host immune response induced by the virus in chicken and to assess the risk of transmission to naïve chicken. A total number of thirty-one, 3-week-old broiler chicks were split into three groups viz., uninfected control group (n=14), Emu/5 inoculated group (n=14) and sentinel group (n=3). Control group was inoculated with PBS, while the experimental group was inoculated with a single dose of 100 µl allantoic fluid (HA titre of 1024 per 50 µl) through subcutaneous route as described in (Office International des Epizooties, 2012). Control group was placed in separate facility. Sentinel group was placed in a cage immediately next to the infected group to assess viral transmission from infected to un- infected birds.  Infected birds were observed regularly for clinical signs and mortality. Three birds from each group were euthanized at 1, 3 and 5 day post infection (dpi) and spleen was collected in RNA later® (Sigma), for evaluating cytokine responses. Relative changes in RNA expression of pro- inflammatory cytokine genes (TNF-α, IL-1β and IL-6), chemokine (IL-8) and apoptotic gene (Cas-8) and IL-10, an anti-inflammatory cytokine and β- actin for infected and mock- infected controls were measured as changes in SYBR green fluorescence as described previously (Rajasekaran et al., 2018). The remaining birds were monitored twice daily for clinical signs/ mortality upto 7 days. Necropsies were performed on the dead birds to examine for gross lesions. Tissues viz., spleen, intestines, trachea, lung, kidney, liver and heart were collected in 10% neutral buffered formalin for histopatho- logical analysis by routine paraffin embedding technique and microtomy. Four-micron thick sections were stained by hematoxylin and eosin method (Culling, 1968).

In this study, we obtained AAvV-1 isolate (Emu/5, Genbank: MT178234) from an apparently healthy emu flock. Previous reports confirmed isolation of vNDV from an infected emu flock raised in a poultry dense area (Gowthaman et al., 2016). Shinde et al., (2012) demonstrated that 15.3% of emus are seropositive in some regions of India. In general, emus are housed in open pens in Indian field conditions providing opportunity for interaction with other wild and feral birds.

The virus could be isolated during second passage in embryonated chicken eggs. Harvested embryos showed typical lesions such as sub-occipital haemorrhages and the allantoic fluid had a haemagglutination titre of 256-512 per 50 µl. HI titre with polyclonal chicken antiserum to LaSota was 1:128 thus confirming the isolate as AAvV-1. The AAvV-1 infective dose for a chicken is reported to range from 103.0 (King, 1996) to 104.0 EID50 (Alexander et al.,  2006). Hence, it can be assumed that virus load harboured by the asymptomatic birds might be very low thus necessitating repeated passages in embryonated chicken eggs for isolation of the virus.

In addition, Mean death time score in 9-11 day old embryonated chicken eggs was 48 h and ICPI score in day- old chicks was 1.82, suggesting that the isolate was highly virulent to chicken. The isolate was further characterized by propagating in chicken embryo fibroblasts (CEF) cultures. Characteristic cytopathic changes were observed in the first passage which included cellular granulation, syncytia formation, vacoulation and cytoplasmic process formation.

Amplification and sequence analysis of the fusion gene revealed 112R-R-R-K-R116-F motif at the fusion protein cleavage site, typically found in velogenic AAvV-1. Full- length fusion gene was found to have a cds of 1662 nucleotides coding for 553 amino acids. Phylogenetic analysis based on the full- length fusion gene classified the virus as belonging to subgenotype XIII 2.2 (Fig 1). The Emu/5 isolate clustered closely with Avian avulavirus 1 isolate 410/16A (GenBank ID MF422129.1) and Avian avulavirus 1 isolate 96-15 (GenBank ID MF422125.1) isolated from Andhra Pradesh. The nucleotide and deduced amino acid sequence of fusion gene of Emu/5 isolate was aligned with representative sequences of all 21 genotypes and with the current vaccine strains and used in determination of pairwise comparisons and homologies. The percentage of nucleotide and deduced amino acid sequence identity of full- length F gene of the isolate with subgenotype XIII 2.2 isolates was 98.56% and 94.76%, while the nucleotide and deduced amino acid sequence identity with currently used vaccine strains was only 81.2% and 87.3% respectively. During 2002-2003, Kinde et al., (2005) reported isolation of exotic newcastle disease (END) virus from emus. Similarly, Gowthaman et al., (2016) reported isolation of velogenic AAvV-1 of subgenotype XIII 2.2 from an infected emu flock with a similar motif at FPCS. ND outbreaks in commercial chicken flocks caused by subgenotype XIII 2.2 viruses were also reported in different parts of India (Jakhesara et al., 2016; Gowthaman et al., 2018).



The six neutralizing epitopes critical for structure and function of fusion protein (D72, E74, A75, K78, A79, L343) were found to be unaltered in the Emu/5 isolate when compared with the LaSota vaccine strain. Six potential N-glycosylation sites (Asn-X-Ser/Thr or N-X-S/T, where X present any amino acid except aspartic acid or proline) at positions 85NRT87, 191NNT193, 366NTS368, 447NIS449, 471NNS473 and 541NNT543 were also found to be conserved. of the eight trans membrane domains of AAvV-1 isolates located at positions 14-27, 15-25, 118-131, 120-128, 266-269, 429-432, 499-525 and 501-523 in the F protein, one substitution was observed at M14R. Further, several amino acid substitutions were observed in the heptad repeat regions (HRa, HRb, HRc and HRd) and hypervariable region of the fusion protein (Table 1 and 2). Heptad repeats have been shown to be critical for fusion and mutations in these regions decrease the fusion activity (MC Ginnes et al., 2001). Significance of the mutations observed in the fusion gene and influence on pathogenicity needs further investigations. It is worth to note that an amino acid substitution (G271A) in HRb region was unique to Emu/5 isolate and not found in other subgenotype XIII 2.2 viruses.



​

Experimental inoculation of three- week old chicks with the Emu/5 isolate resulted in virulent ND with 100% mortality in the infected and sentinel groups. Infected birds were active and apparently healthy on 1dpi. Birds started exhibiting mild clinical signs such as lack of alertness and depression by 2 dpi, ruffled feathers, marked depression and drooping eyelids by 3dpi. One bird was found dead on 4dpi. By 5dpi, marked clinical signs such as labored breathing, lateral recumbency and ataxia were seen in all birds. Gross lesions in tissues until 3dpi, included diffuse mild haemorrhages throughout the carcass. Necropsy of the bird dead on 4dpi, showed haemorrhages in the proventriculus, necrotic lesions in the intestines, subcutaneous edema and tracheal haemorrhages. Similar lesions were more pronounced in birds which succumbed later (Fig 2).



Tissues collected on 1dpi such as spleen, intestines, lungs, kidneys, trachea and heart did not show any pathological lesions. Mild haemorrhages were observed in tissues collected on 3dpi. Tissues collected on 5dpi revealed severe necrosis and lymphocytolysis of spleen, denudation of tracheal epithelium, vacoulation, mononuclear cell infiltration in trachea, hyperplasia and hypertrophy of bronchiolar epithelium and severe submucosal infiltration in lungs, severe necrosis and fusion of villi in the intestines, severe tubular necrosis in kidneys, pericarditis with infiltration of fibrin, fibroblasts and mononuclear cells in the heart (Fig 3).



Relative expression of cytokine genes in response to Emu/5 isolate are represented graphically (Fig 4). The expression levels of pro inflammatory cytokines, IL-1β and IL-6 induced were upregulated at 1 dpi and further upregulation was observed at 3dpi by several folds and reduced by 5 dpi. Step- wise upregulation was observed in the expression of TNF-α from 1 to 5 dpi. The expression of Cas-8 is markedly upregulated and reached its peak by 5 dpi. The expression of anti- inflammatory cytokine, IL-10 gene was not detected during 1 and 3 dpi, but reached its peak and became significantly higher on 5 dpi. Gradual upregulation of IL-8 expression was observed from 1dpi to 3 dpi, later down regulated by 5 dpi (0.57- fold). Pro inflammatory cytokines and chemokines were found to be upregulated during acute infection and responsible for tissue damage caused due to necrotic lesions in intestines and cytolytic changes in spleen. Il-10, an anti- inflammatory cytokine was upregulated during the later part of infection that has a role in dampening the inflammation and tissue repair. Maximum upregulation of proinflammatory cytokines and apoptotic gene was observed in spleen of chicken infected by a velogenic AAvV-1, while minimal upregulation was observed by lentogenic isolates (Liu et al., 2012; Rajasekaran et al., 2018).



Previous studies reported isolation of virulent NDV from healthy village chicken (Ananth et al., 2008; Munir et al., 2012). In this study, for the first time, we confirm circulation of virulent AAvV-1 in apparently healthy emu birds. Emus exhibit certain level of indigenous resistance which might have helped the birds to withstand ND infection without exhibiting clinical signs, but they may act as carriers and transmit infection to poultry. AAvVs that spill-over from wild/feral birds or other species into poultry continue to evolve as they are rapidly passaged in chickens. Hence, extensive studies are required to estimate the prevalence of AAvV-1 in emu and their role in epidemiology and transmission of AAvV-1 in domestic poultry.

Better understanding on the epidemiological relations among the circulating AAvV-1, their genetic diversity, role of domesticated birds other than chicken in the maintenance and transmission of virus is quite essential and crucial for planning novel control strategies for ND.

We are very thankful to the Sri Venkateswara Veterinary University (SVVU), Tirupati for providing facilities to carry out the research.

The authors declare that they have no conflict of interest.