A unique influenza A (H5N1) virus causing a focal poultry outbreak in 2007 in Manipur, India

Background A focal H5N1 outbreak in poultry was reported from Manipur, a north-eastern state, of India, in 2007. The aim of this study was to genetically characterize the Manipur isolate to understand the relationship with other H5N1 isolates and to trace the possible source of introduction of the virus into the country. Results Characterization of the complete genome revealed that the virus belonged to clade 2.2. It was distinctly different from viruses of the three EMA sublineages of clade 2.2 but related to isolates from wild migratory waterfowl from Russia, China and Mongolia. The HA gene, had the cleavage site GERRRRKR, earlier reported in whooper swan isolates from Mongolia in 2005. A stop codon at position 29 in the PB1-F2 protein could have implications on the replication efficiency. The acquisition of polymorphisms as seen in recent isolates of 2005–07 from distinct geographical regions suggests the possibility of transportation of H5N1 viruses through migratory birds. Conclusion Considering that all eight genes of the earlier Indian isolates belonged to the EMA3 sublineage and similar strains have not been reported from neighbouring countries of the subcontinent, it appears that the virus may have been introduced independently.


Background
Highly pathogenic avian influenza (HPAI) A -H5N1 viruses have now appeared in about 60 countries causing devastating outbreaks in poultry with continued capacity to impact humans [1]. The virus was initially isolated from geese in Guangdong, China in 1996 [2]. The Hong Kong reassortant viruses that infected human in 1997 [3] were eliminated due to massive culling of poultry, but the ancestors remained and generated various new genotypes [4]. The virus that re-emerged in South Korea in late 2003 [5] spread to south-east Asian countries [6]. Another major emergence was noticed after an outbreak in migratory birds in Qinghai lake, western China, in 2005, [7] causing outbreaks in many countries in Europe, Middle-East, Africa and Asia [8,9]. The virus is continuously evolving and diversifying into different clades. All the viruses that caused outbreaks in China, Europe, Middle-Eastern and African regions grouped into genotype Z, clade 2.2 [10,11]. The isolates from India and Bangladesh perhaps form the south-eastern geographical boundary for this clade. The clade includes 3 sublineages namely EMA 1 to 3 [12] and some unassigned viruses.
The first outbreak of the H5N1 virus in India was reported from Maharashtra in January 2006 [13]. Seven episodes in poultry were recorded up to April 2006 in the western states of Maharashtra, Gujarat and central Madhya Pradesh. Complete genome sequencing of the H5N1 isolates of 2006 revealed that all eight genes belonged to the sublineage EMA3 of the clade 2.2. The close similarity of the virus to geographical regions of the East Africa/West-Asian and Central Asian migratory bird flyways, suggested that the virus in India might have been introduced through migratory birds [14]. In July 2007, a small outbreak of H5N1 in poultry was reported from Manipur, a north-eastern state, of India. The outbreak was controlled and no spread was noted in the neighbouring areas.
The aim of this study was to genetically characterize the Manipur isolate of 2007 to understand the relationship with other H5N1 isolates and to trace the possible source of introduction of the virus into the country.

Materials and methods
The state of Manipur (latitude 23°83'N -25°68'N and longitude 93°03'E -94°78'E) is known for some animal sanctuaries that are home to many exotic flora and fauna ( Figure 1). A large number of migratory birds visit Loktak, an ecologically rich freshwater lake dotted with floating islands, about 45 km away from the capital city, Imphal. The city lies in a valley of ~700 sq. miles surrounded by mountains at an elevation of 790 metres above sea level. The outbreak was reported in Chingmeirong, East Imphal. The capital is well connected to Myanmar in the east and to Bangladesh in the west both of which reported Avian Influenza outbreaks in 2006-07.

Virus Isolation
Six clinical samples from different organs (trachea, lung, spleen, liver, heart and kidney) of a sick bird were received from Manipur. Specimens were processed for virus isolation in specific-pathogen-free (SPF) embryonated chicken eggs and Madin Darby Canine Kidney (MDCK) cell lines as described earlier [14]. Inoculated eggs were observed for 24-48 hours before harvesting the allantoic fluid. All experiments using infectious virus were conducted in a biosafety level 3 (enhanced) laboratory.

Identification
Hemagglutination (HA) and Hemagglutination inhibition (HAI) tests were performed as described by Kendal et al. [15]. Horse Red blood cells (1.0% suspension) were used for the HA and HAI test. The reference antisera used were influenza A(H5N1)-NIV/Navapur, H5N1-WHO, H5N2, H9N2, H7N3, and Newcastle disease virus (obtained from the OIE reference laboratory, Venice, Italy).
RNA was extracted using QIAamp Viral RNA Minikit (QIAGEN, Germany) following manufacturers instruction. One-Step reverse transcription-PCR (RT-PCR) was performed using the QIAGEN one-step RT-PCR kit and WHO recommended diagnostic primer sets specific for influenza A HA (H5) and NA (N1) genes [16]. RNA isolated from the specimens was tested by Real Time RT-PCR. Applied Biosystems' TaqMan Influenza A/H5 Detection Kit Version 1.0 was used on Applied Biosystems' 7300 Real-Time PCR platform. Negative controls were processed along with the specimens to rule out cross contamination.
Map showing the location of the H5N1 outbreak in the state of Manipur, India Whole genome sequencing RNA isolated from the specimens was reverse transcribed as mentioned earlier [14]. cDNA was used to amplify all the eight gene segments using segment specific primers [17]. The PCR products were purified from an agarose gel using gel extraction kit (QIAGEN,) and amplicons were directly sequenced using an automated 3130 XL Genetic analyzer (Applied Biosystems).

Phylogenetic analysis
For phylogenetic analysis, representative sequences of the H5N1 viruses belonging to the Z genotype were selected from the GenBank based on sequence identity (100% identical sequences were excluded) and geographical representation. In this process of selection, the sequences whose whole genome was available were preferred. Phylogenetic analysis was performed using the Bayesian approach for tree construction as implemented in Mr Bayes 3.2 [18]. The GTR (General Time Reversible) + I (Invariable sites) model with gamma-distributed rate variation across sites and a proportion of invariable sites, was specifically used and other parameters were kept as default.
The GenBank accession numbers for the PB2, PB1, PA, HA, NP, M and NS gene segments of the isolate, A/Ck/ India/NIV9743/07, herein referred to as the Manipur isolate, is from FJ719831-FJ719838. The percent nucleotide identity (PNI) and percent amino acid identity (PAI) values were calculated as pairwise p-distances, for a dataset of about 80 representative sequences in each gene. For clarity, limited representative sequences are shown in the phylogenetic trees (Figures 2, 3 and 4).

Results
In Chingmeirong, Manipur, mortality was noticed in a poultry flock on 7 th July 2007. The birds were housed in 2 separate backyard coops, 132 in one and 12 in another, separated by just about 5 meters. All the birds of the first coop died in about 4 days. After one week, one bird in the second coop fell sick and was culled. The clinical samples were collected and sent to our laboratory for investigations. All the other birds were culled. Subsequent investigations revealed that the birds were moved from a village farm, located at the interface of forest and agricultural land, about 40 kms away from Imphal city, just a few days before mortality was noticed. It is likely that the birds might have contracted the infection in that place which is frequented by many wild birds. Samples collected from other parts of the city and in adjoining localities were all negative for the virus.

Isolation and Identification
Virus was isolated from all six samples and in all the cases chick embryos died within 24 hours post-infection. Con-fluent monolayers of MDCK cell lines infected with the specimen showed high cytopathic effect. Allantoic fluid from dead eggs and tissue culture supernatants from infected cultures tested positive for avian influenza (H5N1) virus in HA and HAI.
Samples tested in one step RT-PCR showed amplification of influenza A, H5 and N1 gene specific bands. A 219 base pair band appeared showing presence of H5 and 668 base pair for the presence of N1. Real Time RT-PCR analysis showed the presence of Avian influenza (H5N1) in all the specimens (Data not shown).

Phylogenetic analysis
Phylogenetic analysis of 41 whole genomes, all the eight gene segments concatenated, showed that the Manipur isolate was unique in the clade 2.2, as it did not cluster with the majority of the isolates in the EMA sublineages 1 to 3 ( Figure 2). It was close to A/Dk/Novosibirsk/56/05 and other ungrouped isolates such as A/Gs/Suzdalka/10/ 05, A/Gf/Shantou/1341/06 and A/Ck/Tula/4/05.
The NA gene of the Manipur isolate did not group with any of the EMA sublineages and maintained its position with the other unassigned viruses ( Figure 4) 9). Again, in all these genes the Manipur isolate was distinct from the others and also did not cluster with any of the EMA sublineages.

Mutations
In the HA gene, among amino acid mutations, the Manipur isolate had a substitution K328R (H5 numbering and substitution mentioned with reference to A/Vietnam/ 1203/04, belonging to clade 1). This corresponds to the novel HA cleavage site, GERRRRKR that was originally found in three whooper swan isolates of Mongolia in 2005 (GenBank accession numbers AB233320-AB233322). Subsequently, the same pattern was noted in several isolates of 2007 from Egypt (EF535822-EF535825, EU496393) and Nigeria (EU148396). Another non synonymous mutation, S155N, at the N154 glycosylation site, observed in the Manipur isolate was also noted in the 2005 isolates of Novosibirsk, Qinghai, Tam A synonymous substitution, A91C, in the NP gene, was unique and not observed in any of the available H5N1 sequences. In the M gene, V68I of the Manipur isolate was shared with A/Ck/Nigeria/104754/06 and RostovonDon/ 07 isolates. The non synonymous mutations included, A408T shared with A/Dk/Novosibirsk/56/05; G480A with A/Ck/Nigeria/10719/07; G890A with A/Ck/Nigeria/ 104754/06 and Rostovon/07 isolates. In the NS1 gene, two unique mutations, V111M and L212P, were observed while a unique mutation, F55L, was observed in the NS2 gene.
The gene coding for the PB1-F2 protein in the Manipur isolate, was observed to have two nucleotide polymorphisms, A85T and G86A. These polymorphisms resulted in the introduction of a stop codon at position 29 in the PB1-F2 that was unique to the Manipur isolate. Similar to this, a stop codon was noted in A/Gf/Shantou/1342/06 at position 25. Another nucleotide substitution, C123T in PB1-F2 of the Manipur isolate was observed only in clade 2.1 isolates. An amino acid substitution, G110A was shared with A/Co/Croatia/1/05 and A/Ck/Nigeria/10719/ 07.
The amino acid variations observed in the Indian isolates, for all the eight gene segments and the PB1-F2 protein are indicated in Table 1.

Molecular markers
Of the 14 residues reported to form the receptor-binding pocket in HA1 [19], two substitutions K193R and R216K were observed in the Manipur isolate. The residues that preferentially binds the avian-like α2,3-NeuAcGal linkages, G225, S227, G228, Q226, S221 and R216 were conserved. Host specific amino acid residues [20] in the PB2, PA, NP, M1 and M2 proteins, all suggested avian specificity except V28 in the M2 protein, known for adaptation to human hosts. The isolate possessed Lys at position 627 of the PB2 gene, which is associated with increased virulence in mammals [21]. Of the known markers for enhanced polymerase activity [22], only one substitution, L13P in PB1 protein is observed in the Manipur isolate. The PB1-F2 protein, that has been recognized as a virulence contributor and functions in reducing the immune responses as well as increasing the cytotoxicity [23,24], was observed to possess a stop codon at position 29. Similar to this, a stop codon was noted in Gf/Shantou/1342/ 06 at position 25. The C-terminal protein was conserved when compared to other PB1-F2 proteins except for a P48Q mutation (Table 1). E92 in the NS1 protein, implicated in human adaptation, was observed in the Manipur isolate. The C terminal four amino acid stretch corresponding to the PDZ-domain ligand motif in the NS1 protein was ESKV and is reported as being specific to avian isolates [25].
The NA sequence did not show any mutation conferring resistance to oseltamivir [26]. Similarly, no mutation was observed in the M2 ion channel that conferred resistance to amantadine [27].

Discussion
The EMA 1-3 sublineages represent the influenza (H5N1) viruses isolated in Europe, the Middle-Eastern region and Africa beyond 2005 [12]. A minority of isolates, such as those of Qinghai, Novosibirsk region, Shantou and Omsk did not group with either of these sublineages and had been left unassigned. On the basis of the Bayesian phylogenetic analysis in this study, in all the 8 gene segments, the Manipur isolate grouped with the unassigned viruses in the EMA nomenclature.
The molecular characterization of the viral proteins of the Manipur isolate was carried out for pathogenicity markers and sensitivity to antivirals. The HA of the Manipur isolate had the novel cleavage site (GERRRRKR) that was first reported in three whooper swan isolates in Mongolia, 2005 and also in the recent Nigeria/07 and Egypt/07 isolates. Though variations in the cleavage site such as RERRRKKR, RERRKKR and RERRRR have been linked to human H5N1 cases [28], no such report exists for GERRRRKR motif. Hence the significance of this cleavage site remains to be understood. D54N mutation in the Manipur isolate, results in an additional putative glycosylation site, the implication of which is not yet known. The two substitutions in the HA1 receptor binding domain were also noted in other clade 2.2 viruses including our 2006 isolates and may not have implications in modifying avian associated α2-3 linked sialic acid specificities. Similarly, no marked mutations have been observed in either of the PB2, PA, NP, M1 and M2 proteins with respect to enhanced polymerase activity or host specificity. Sensitivity to antivirals was also noted in the Manipur isolate. Notably there was a stop codon in the PB1-F2 protein of the Manipur isolate. On one hand, pro-  [24], implying that the protein of the Manipur isolate may still get expressed from a downstream initiation codon. Further, though the C-terminal region of the PB1-F2 protein, known to be responsible for major functional roles such as mitochondrial localization and PB1-F2 induced apoptosis [29,30]  Myanmar, it appears that there may have been a independent introduction into the country. Existing reports, that, migratory waterfowl from the Qinghai lake migrate southwards to Myanmar and over the Himalayas to India, around September returning to the Qinghai lake around April [41], support such an assumption.

Conclusion
Overall, our findings suggest that the Manipur 2007 virus isolate is a unique variant and not related to the 2006 Indian isolate. The introduction of such a virus, either directly or indirectly, into India calls for improved surveillance in the country and subcontinent. The appearance of such variants is also serious concern for the emergence of even more highly pathogenic strains and a pandemic threat.
Publish with Bio Med Central and every scientist can read your work free of charge