Surveillance for avian influenza viruses in wild birds at live bird markets, Egypt, 2014‐2016

Aim Egypt is the habitat for a large number of bird species and serves as a vital stopover for millions of migratory birds during their annual migration between the Palearctic and Afrotropical ecozones. Surveillance for avian influenza viruses (AIVs) is critical to assessing risks for potential spreading of these viruses among domestic poultry. Surveillance for AIV among hunted and captured wild birds in Egypt was conducted in order to understand the characteristics of circulating viruses. Methods Sampling of wild bird species occurred in two locations along the Mediterranean Coast of Egypt in the period from 2014 to 2016. A total of 1316 samples (cloacal and oropharyngeal swabs) were collected from 20 different species of hunted or captured resident and migratory birds sold at live bird markets. Viruses were propagated then sequenced. Phylogenetic analysis and receptor binding affinities were studied. Results Eighteen AIVs (1.37%) were isolated from migratory Anseriformes at live bird markets. Further characterization of the viral isolates identified five hemagglutinin (H3, H5, H7, H9, and H10) and five neuraminidase (N1, N2, N3, N6, and N9) subtypes, which were related to isolates reported in the Eurasian region. Two of the 18 isolates were highly pathogenic H5N1 viruses related to clade 2.2.1, while three isolates were G1‐like H9N2 viruses. Conclusions Our data show significant diversity of AIVs in Anserifromes sold at live bird markets in Egypt. This allows for genetic exchanges between imported and enzootic viruses and put the exposed humans at a higher risk of infection.


| INTRODUC TI ON
Wild birds, particularly waterfowl, are the natural reservoir of many subtypes of influenza A viruses and play an important role in the evolution and spread of avian influenza viruses (AIV). 1 Birds of the orders Anseriformes and Charadriiformes are thought to be the most common reservoirs of subtypes H1-H16 of influenza A viruses. 2 AIVs are classified into highly pathogenic avian influenza virus (HPAIV) and low pathogenic avian influenza virus (LPAIV).
HPAIV H5 Goose/Guangdong lineage can be transmitted to domestic poultry by infected wild birds and spread rapidly, causing serious disease with high mortality. 3 Other subtypes such as LPAIV H5 and H7 can become highly pathogenic within the domestic species population. 4 LPAIVs of subtypes H1 to H16 show mild or no disease in wild birds but can infect other hosts sharing their habitat and transmit the virus to highly susceptible poultry species such as chickens, turkeys, and other bird species. 5 Since several human infections with HPAIVs (eg, H5N1 and H7N7) and LPAIVs (eg, H7N9, H9N2, H10N8) 6,7 occurred over the last two decades, surveillance has been intensified in both poultry and wild avian life in order to understand disease evolution, virus spread, and risk factors associated with infection. [8][9][10] Egypt has habitats to a large number of bird species. The wetlands of the northern Nile Delta are a vital stopover for millions of migratory birds during their annual migration between the Palearctic and Afrotropical ecozones. 11 Two migratory birds flyways, the Black Sea-Mediterranean and East African-West Asian flyway, overlap in Egypt. 12 Therefore, the Egyptian environment is an important site on the wild birds migration network through the old world. 13 Even though surveillance of wild birds for AIVs has increased substantially worldwide in the last years, few studies have been conducted in Egypt. 14,15 Here, we conducted active surveillance of AIVs in resident and migratory wild birds either hunted or captured to be sold at live bird markets in Egypt from 2014 to 2016. from El-Arish in 2015 only or captured migratory or resident wild birds being sold at live bird markets in Damietta. All samples were collected from apparently healthy birds. The specimens were stored on ice and transported rapidly for laboratory processing.

| Virus detection, isolation and subtyping
A volume of 50 mL phosphate-buffered saline (PBS) containing antibiotic/antimycotic mixture of 10 000 U/mL penicillin, 10 000 μg/mL streptomycin, and 25 μg/mL amphotericin B was used as egg infection medium (Lonza, Walkersville, MD, USA). A volume of 200 μL of the viral inoculum (100 μL sample and 100 μL egg infection medium) was inoculated into the allantoic cavities of 3 10-day-old specific pathogen-free embryonated chicken eggs and incubated for two days at 37°C as per World Health Organization (WHO) guidelines. 16 The inoculated eggs were tested for influenza virus by hemagglutination assays using 0.5% chicken red blood cell. The positive samples were aliquoted and stored at −80°C.
RNA from each positive sample was extracted using QIAamp Viral RNA Mini Kit (Qiagen, Germany), then typed by M gene using realtime PCR (RT-PCR) according to WHO protocol. 17 The positive M gene samples were further subtyped by RT-PCR with specific primers for hemagglutinin and neuraminidase genes as previously described. 18,19

| Statistical analysis
Chi-square and Fisher's exact tests were used to compare positivity rates between sample types (cloacal vs oropharyngeal) and sites (El-Arish vs Damietta).

| Full genome sequencing and analysis of sequences
The first-strand cDNA was synthesized using Superscript III Reverse transcriptase (Invitrogen, Carlsbad, CA) and Uni-12 primer (5'AGCRAAAGCAGG3') as per manufacturer's protocol. Using Phusion Master Mix kit (Thermo Scientific, Wilmington, USA), the desired genes of the isolates from poultry were amplified using universal primers. 20 Briefly, using gene-specific primers, 2 µL of each RT-PCR product was subjected to PCR by an initial denaturation step (98°C for 30 seconds), followed by 40 cycles each consisting of 98°C for 10 seconds, 57°C for 30 seconds, 72°C for 3 minutes, and final elongation (72°C for 10 minutes). Amplicons of the appropriate sizes were subsequently gel purified using QIAGEN gel extraction kit. The purified PCR products were directly used for sequence reactions at Macrogen sequencing facility (Macrogen, South Korea). Sequences were assembled using SeqMan Lasergene 7 software (DNASTAR, Madison, WI, USA). Related sequences for the desired gene were obtained from the Global Initiative on Sharing All Influenza Data (GISAID) (http://platform.gisaid.org/epi3/) in July 2018. Representative sequences of North American and Eurasian lineage strains were obtained from GenBank. Sequence alignments were performed using BioEdit 7.0 software. The phylogenetic tree was constructed using MEGA7 program by applying the neighbor-joining method with Kimura's two-parameter substitution model and 1000 bootstrap replicates.

| Receptor specificity assay
Virus receptor specificity for six representative AIVs detected in this study was determined as previously described. 21 Figure 1; Figure S1). This was also the case for all the gene segments of these viruses ( Figure S2)

| H10N3 isolate
Phylogenetic analysis of A/teal/Egypt/MB.D148OP/2015(H10N3) showed that this was closely related to viruses of the Eurasian lineage ( Figure 1; Figures S1 and S2). The receptor binding pocket area of this isolate had amino acids 95Y, 151 W, 183H, 190E, 191 K, 194L, 226Q, 227S, 228G, and 229R. None of these amino acid residues have been reported to be involved in the recognition of human-like receptors, 22 suggesting that the Egyptian H10N3 isolate likely binds to avian-like receptors (a2,3-SL) (Figure 2). The HA cleavage site sequence of the H10N3 isolate was 330 RGLF 333 (according to H3 numbering), which is the signature of low pathogenicity of influenza A viruses.

| D ISCUSS I ON
Studying the ecology and evolution of AIVs in wild aquatic birds is key to elucidate their role in virus dissemination among different hosts in different geographic regions. Egypt, where H5 and H9 viruses are enzootic in domestic poultry, acts as a stopover for millions of wild aquatic birds during autumn and spring migration seasons annually providing excellent grounds for reassortment events between AIVs.
Several of those species are hunted or captured to be sold live at bird markets. This provides a niche where AIVs can exchange genetic material and where human exposure raises the potential for infection. Therefore, it is important to determine which AIVs are present in wild birds in Egypt and use this information in designing effective control strategies.
The isolated AIVs included five different HA subtypes (3, 5, 7, 9, and 10) and five NA subtypes (1, 2, 3, 6, and 9) indicating that wild birds in Egypt introduce a wide range of viruses into the Egyptian environment. Similar to previous surveillance studies, 1

E TH I C A L A PPROVA L
All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. This article does not contain any studies with human participants performed by any of the authors.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.