Intersubtype Reassortments of H5N1 Highly Pathogenic Avian Influenza Viruses Isolated from Quail

H5N1 highly pathogenic avian influenza (HPAI) viruses are considered a threat to national animal industries, causing production losses and high mortality in domestic poultry. In recent years, quail has become a popular terrestrial poultry species raised for production of meat and eggs in Asia. In this study, to better understand the roles of quail in H5N1 viral evolution, two H5N1-positive samples, designated A/quail/Vietnam/CVVI-49/2010 (CVVI-49/2010) and A/quail/Vietnam/CVVI-50/2014 (CVVI-50/2014), were isolated from quail during H5N1 outbreaks in Vietnam, and their whole genome were analyzed. The phylogenetic analysis reveals new evolutionary variation in the worldwide H5N1 viruses. The quail HA genes were clustered into clades 1.1.1 (CVVI-49/2010) and clade 2.3.2.1c (CVVI-50/2014), which may have evolved from viruses circulating from chickens and/or ducks in Cambodia, mainland of China, Taiwan, Indonesia, and South Korea in recent years. Interestingly, the M2 gene of the CVVI-49/2010 strain contained amino acid substitutions at position 26L-I and 31S-N that are related to amantadine-resistance. In particular, the CVVI-50/2014 strain revealed evidence of multiple intersubtype reassortment events between virus clades 2.3.2.1c, 2.3.2.1b, and 2.3.2.1a. Data from this study supports the possible role of quail as an important intermediate host in avian influenza virus evolution. Therefore, additional surveillance is needed to monitor these HPAI viruses both serologically and virologically in quail.


Introduction
Avian influenza viruses (AIV) belong to the Orthomyxoviridae family. The viral genome consists of eight segments of single-stranded negative RNA, encoding at least 10 well-described functional proteins (PB1, PB2, PA, HA, NP, NA, M1, M2, NS1, and NS2), and five recently identified functional proteins (PB1-F2, PB1-N40, PA-X, PA-N155, and PA-N182) [1]. Hemagglutinin (HA) and neuraminidase (NA) are surface antigenic proteins that play a major role in the host humoral immune response against these viruses [2]. Based on the presence of HA and genetically or phylogenetically characterized. Interestingly, in Vietnam, quails are usually maintained by natural poultry farming practices (e.g., backyard or grazing frame methods) that may contribute to the persistence and spread of influenza viruses throughout the country. In this study, the full genomes of two HPAI H5N1 virus strains isolated from H5N1 outbreaks on quail farms in Khanh Hoa province in 2010 and 2014, a south-central region of Vietnam were characterized in order to determine the genetic relatedness of these strains to reference AIV strains and to describe the possible role of quail in the evolution of AIVs in Vietnam.

Ethics statement
Data in this study were obtained from case-investigation reports at the Central Vietnam Veterinary Institute (CVVI), Nha Trang, Vietnam and being reported to the authorities. The accurate poultry farm information was supplied by the owners with none of the field studies involve endangered or protected species. Ethics approval for the animal experiments was obtained from the National Institute of Veterinary Research (NIVR) Ethics Committee, Hanoi, Vietnam.

Sample collection
Two HPAI H5N1 virus strains, designated A/quail/Vietnam/CVVI-49/2010 (CVVI-49/2010) and A/quail/Vietnam/CVVI-50/2014 (CVVI-50/2014), were isolated from H5N1 outbreaks in quail in a passive surveillance program in Khanh Hoa province, a south-central region of Vietnam. These outbreaks occurred on quail farms in January 2010 and March 2014, and killed 476/2300 (20.7%) and 250/900 (27.8%) of the quail in each farm. The clinical signs in infected quail included depression, seizures, ataxia, neurological signs, and green, diarrheal feces. In most cases, two sick or dead quail from each farm were collected and transported to the Central Vietnam Veterinary Institute. Tissue samples from the brain, lung, spleen, bronchus, and intestine were collected and stored at −80°C for further examination. All handling tissue samples and viral cultures were carried in the Biosafety Level 3 laboratory.

Virus isolation in embryonated chicken eggs
The tissue samples were homogenized in phosphate buffer saline (PBS; pH 7.4) with antibiotics and clarified by centrifugation at 400 × g for 10 min. Supernatants were collected and filtered using a 0.45-μm sterile syringe filter (Corning Costar, Corning, NY, USA). Viruses were isolated using embryonated chicken eggs according to the World Health Organization (WHO) manual [31]. Briefly, 100 μl of each sample was inoculated into the amniotic cavity of three 10-day-old specific pathogen-free embryonated chicken eggs, and incubated at 37°C for 48 hrs. Virus was then harvested and stored at −80°C for further examinations.

RNA extraction and real-time RT-PCR
The pooled tissue samples from each farm were homogenized in phosphate buffer saline (PBS; pH 7.4) and clarified by centrifugation at 400 × g for 10 min. Viral RNA was extracted using a QIAamp Viral RNA Mini Kit (Qiagen, Valencia, CA, USA), according to the manufacturer's instructions. Extracted RNA was resuspended in RNase-free water and stored at −80°C until analysis. Real-time RT-PCR was performed according to the WHO manual [32]. Briefly, the QuantiTech Probe RT-PCR Kit (Qiagen) was used with 12.5 μl of Master Mix, 1.5 μl each of forward and reverse primers (10 μM), 0.5 μl of probe (5 μM), 0.25 μl of QuantiTect 1 RT Mix, 3.75 μl of RNAase free water, and 5.0 μl of RNA template in a 25-μl total volume. Using an ABI 7500 real-time thermocycler (Applied Biosystems, Foster City, CA, USA), reverse transcription was carried out for 30 minutes at 50°C, followed by polymerase activation for 15 minutes at 95°C. Denaturation for 15 seconds at 94°C and annealing-extension for 1 minute at 56°C were performed for 45 cycles to obtain cycle threshold (Ct) values.

Reverse transcription-polymerase chain reaction (RT-PCR)
The SuperScript III First-Strand Synthesis Super Mix (Invitrogen, Carlsbad, CA, USA) was used to prepare cDNA from the extracted RNA with Uni12 primer (5 0 -AGCRAAAGCAGG-3 0 ). The reaction was carried out at 42°C for 60 min, followed by 72°C for 10 min. The full lengths of eight gene segments, HA, NA, PB2, PB1, PA, NP, M, and NS, were amplified as described previously [33,34] (S1 Table). Each PCR product was separated on a 1.2% SeaKem LE agarose gel (FMC, Rockland, ME, USA) and stained with ethidium bromide. The gels were viewed on a Gel Doc XR image analysis system (BioRad, Hercules, CA, USA).

Nucleotide sequencing and sequence analysis
The amplified PCR products were purified using a QIAquick Gel Extraction Kit (Qiagen). RT-PCR primers were used for direct sequencing of the HA, NA, PB2, PB1, PA, NP, M, and NS genes using a BigDye Terminator Cycle Sequencing Kit and an ABI 3730 DNA sequencer (Applied Biosystems). Walking sequencing primers were designed to obtain the sequences of the 5 0 and 3 0 ends of each gene (S1 Table). The resulting nucleotide and deduced amino acid sequences were aligned using the Clustal_X 2.1 program [35] and Lasergene software (DNAS-TAR; Madison, WI, USA), with parameters set based on H5N1 viral sequences in the NCBI GenBank database. The nucleotide sequences obtained in this study were deposited in the Gen-Bank database under the accession numbers KP872889-KP872904.

Phylogenetic analysis
The nucleotide sequences of the HA, NA, PB2, PB1, PA, NP, M, and NS genes obtained were compared against representative gene sequences from available HPAI H5N1 sequences in the GenBank database. The viral clades were defined follow the manual of the WHO⁄OIE⁄FAO H5N1 Evolution Working Group [7]. Phylogenetic trees were constructed using the neighborjoining algorithm in the PHYLIP suite and the Kimura two-parameter model using MEGA 6.06 software [36][37][38]. Evolutionary distances for the neighbor-joining analyses were based on the model described by Jukes and Cantor [39]. Tree topology was evaluated by a bootstrap resampling method, with 1000 replicates of the neighbor-joining dataset, using the SEQBOOT and CONSENSE programs in PHYLIP [38].

Identification of HPAI H5N1 cases
The quail was dissected at CVVI to observe the clinical and pathological features of AIV infection. Quail with suspected AIV infections all displayed signs of acute phase disease. Internal organs, including the lung, liver, pancreas, and intestines, exhibited extreme swelling and hemorrhage. Viral RNA was extracted from pooled tissues. AIVs were identified by real-time RT-PCR of the M gene, showing Ct values between 17.6 and 22.6. The HA and NA genes of these two AIVs belonged to the H5 and N1 subtypes, respectively. The viruses were propagated in embryonated chicken eggs. The two H5N1 viruses that were isolated were given the strains designations of A/quail/Vietnam/CVVI-49/2010 (CVVI-49/2010) and A/quail/Vietnam/ CVVI-50/2014 (CVVI-50/2014) (Fig 1).

Genetic analysis
Results obtained from molecular analysis of the HA genes are shown in Table 1 (Table 1).
Mutations at amino acid positions 42 P-S , 92 D-E , and 149 V-A of the NS1 protein have been reported to increase the virulence of the H5N1 viruses in different hosts [41][42][43][44]. The CVVI-49/2010 and CVVI-50/2014 strains both displayed 42 P-S and 149 v-A mutations, and, additionally, the CVVI-50/2014 strain contained a 92 D-E mutation ( Table 1). In addition, five amino acid deletions at positions 80-84, an ESEV sequence at the NS1 C-terminal region, and a PDZdomain motif were found that could indicate the study strains are highly virulent [44] ( Table 1). No genetic marker associated with virulence in the PB2 protein (627 E-K and 701 D-N ) in mammals was found in these two strains [45,46].   (Table 2). These deletions have been related to AIV virulence and the ability of AIVs to be transmitted from water fowl to terrestrial poultry [47]. No amino acid substitutions were found in the NA region of the study strains. Interestingly, two amino acid mutations, 26 L-I and 31 S-N , were found in the M2 sequences that are known to be related to amantadine resistance [48] (Table 2).

Discussion
AIVs in land-based domestic poultry and mammalian species, including humans, can evolve rapidly [49]. The variety of viral reservoirs, especially aquatic birds and domestic poultry, may suggest why these viruses persist and are widespread in many countries. HPAI viruses, influenza virus subtypes H5 and H7, have caused high mortality in wild birds and domestic poultry around the world. In Vietnam, HPAI H5N1 viruses were first isolated in 2001; since then, these viruses have become endemic and caused devastating economic repercussions due to millions of poultry deaths, including those from culling [25,26]. The role of quail in the evolution of HPAI H5N1 viruses, and specifically, their ability to maintain and spread these viruses to difference species, was elucidated by several HPAI H5N1 surveillance studies, as well as in vivo and in vitro examinations [12,13,18,21,50]. Information on HPAI H5N1 outbreaks in quail in Vietnam is limited [30]. This study provides the first characterization of the viral genomes of two HPAI H5N1 viruses isolated from quail H5N1 outbreaks in 2010 and 2014 in Khanh Hoa   [51,52]. In 2010, viruses from clade 2.3.2.1 that further evolved into subclades 2.3. 2.1 (a,b,c), were identified in the north and have since rapidly spread to the entire country [53]. Detection of clades 1.1.1 and 2.3.2.1c in quail in Khanh Hoa province indicate that quail are susceptible hosts and that they may play important roles in maintaining and spreading HPAI H5N1 viruses among poultry and wild birds, with the potential for human infections [21].
Viral mutation and reassortment are the major evolutionary strategies of influenza viruses in response to immune and environmental pressures, resulting in genetic divergence, the potential for virulence enhancement, and new HPAI outbreaks [54]. Whole genome analyses of AIVs provide an excellent platform for determining the interspecies evolutionary relationships of these viruses [55]. Genetic analysis of HPAI H5N1 viruses circulating in Hong Kong in 1997 (A/Hong Kong/156/97) suggested that this virus strain probably originated from reassortment events between H9N2 (A/quail/HK/G1/97-like), H5N1 (A/goose/Guangdong/1/1996-like), and H6N1 (A/teal/HK/W312/97-like) viruses [56].  Quail H5N1 HPAI shared a common ancestor with the original clade 1.1.1 circulating in Cambodia and Vietnamese. Analysis of the amino acid (aa) sequence revealed that the two study strains shared high identity to the duck and chicken HPAI H5N1 viruses isolated in recently years, rather than with quail HPAI H5N1 viruses, suggesting that the study strains were spread from chickens or ducks to quail. These results support the idea that quail may play an important role in the evolution of HPAI H5N1 viruses in Vietnam [57,58]. Further analyses will be needed to understand whether or not these strains can be transmitted to other adapted species. Further analysis revealed that the quail were infected with HPAI H5N1 viruses with common motifs of multiple dibasic amino acids at cleavage sites that have been associated with high pathogenicity and that may be related to the high mortality rates seen at these farms.
Quail are terrestrial poultry that should be considered potential intermediate hosts for influenza viruses because of their susceptibility to both mammalian and AIV subtypes [13]. Substitutions at aa positions 42 A-S and 92 D-E in the NS1 protein that are associated with increased virulence of AIVs in chickens and mice as well as inhibition of host immune responses were found in the study strains [42]. In addition, quails are usually raised in close contact with humans and their population is increasing in Vietnam, posing a concern for public health in areas where HPAI H5N1 viruses are circulating.
Humans infected by HPAI H5N1 viruses are of global public health concern. Worldwide, approximately 667 laboratory-confirmed human cases of H5N1 virus infections were reported during the 2003-2014 transmission season, resulting in 393 deaths [59]. Vietnam is among the countries with the highest rates of human H5N1 infections, with a reported 127 laboratoryconfirmed human cases; of these cases, 62 were fatal [60]. Recent human HPAI pandemics have been associated with resistance to anti-influenza drugs, including adamantanes and neuraminidase inhibitors [61]. Resistance to adamantanes, including amantadine and rimantadine, are likely related to specific amino acid substitutions in the M2 protein at positions 26, 27, 30, 31, and 34, and resistance remains high among circulating influenza A viruses [62]. Four types of neuraminidase inhibitors have been developed, including oseltamivir (Tamiflu), zanamivir (Relenza), peramivir (Rapiacta), and laninamivir (Inavir) [63]. Resistance to neuraminidases are likely related to the amino acid substitutions at positions 275 (275 H-Y ) and 295 (295 N-S ) [63,64]. In strains from this study, no amino acid substitutions that are known to be related to oseltamivir resistance were found. However, the M2 gene of the CVVI-49/2010 strain contained amino acid substitutions at positions 26 L-I and 31 S-N , which may cause amantadine resistance [48]. To maintain treatment options, continued antiviral susceptibility monitoring in H5N1 viruses is needed [48].
In conclusion, this study provided the full genomic characterization of HPAI H5N1 subtype viruses in quail in Vietnam. The results indicate the possibility of multiple genetic reassortment events among HPAI H5N1 virus clades 2.3.2.1c, 2.3.2.1b, and 2.3.2.1a. In addition, markers of resistance to anti-influenza drugs, the adamantanes, were identified. These data provide support for the role of quail as an important intermediate host in AIV evolution. Therefore, additional surveillance is needed to monitor AIVs, serologically and virologically, in quail in Vietnam.
Supporting Information S1