Characterization of a non-pathogenic H5N1 influenza virus isolated from a migratory duck flying from Siberia in Hokkaido, Japan, in October 2009

A total of 711 fecal samples were collected from migratory water birds at lakeside of Ono Pond on the campus of Hokkaido University, Sapporo and Lake Ohnuma in Wakkanai, Hokkaido, Japan, between September and November 2009. Each sample was mixed with Minimum Essential Medium (Nissui) containing antibiotics and inoculated into the allantoic cavities of ten-day-old chicken embryos. The subtypes of influenza viruses were identified by hemagglutination inhibition (HI) and neuraminidase inhibition (NI) tests with antisera to the reference influenza virus strains[14].

Viral RNA was extracted from the allantoic fluid of chicken embryos infected with the isolates by TRIzol LS Reagent (Invitrogen) and reverse-transcribed with the Uni12 primer [15] and SuperScript Reverse Transcriptase III (Invitrogen) or M-MLV Reverse Transcriptase (Invitrogen). The full-length of each gene segment was amplified by polymerase chain reaction with gene-specific primer sets [15]. Direct sequencing of each gene segment was performed using an auto-sequencer CEQ 2000XL (Beckman Coulter) or 3500 Genetic Analyzer (Applied Biosystems).

The nucleotide sequences were phylogenically analyzed based on those of the H5 HA and N1 NA genes of influenza viruses by the neighbor-joining method [3, 16]. Sequence data of the viral genes were compared with those from GenBank/EMBL/DDBJ.

To determine the intravenous pathogenicity index (IVPI), 0.2 ml of the 1:10 dilution of infectious allantoic fluid of embryonated eggs was inoculated intravenously into ten seven-week-old chickens (White Leghorn). The score for IVPI was calculated according to the manual of World Organisation for Animal Health (OIE) [17].

To assess the intranasal pathogenicity for poultry, Mal/Hok/24/09 (H5N1) of 10^6.0 50% egg infectious dose (EID₅₀) was inoculated intranasally into eight four-week-old chickens (Boris Brown), domestic ducks (Chelly Valley), and quails (Japanese Quail). Four of eight birds were euthanized three days post-inoculation (dpi), and the trachea and cloaca swabs, brain, trachea, lungs, kidneys, and colon were collected aseptically for virus recovery. The birds were observed daily for disease signs for 14 days after inoculation. Sera were collected from them on the day of inoculation and 14 dpi to test for antibodies against H5N1 virus. The swabs and tissue homogenates were inoculated into ten-day-old chicken embryos and the infectivity titers of virus were calculated and expressed as the EID₅₀ per milliliter of swab or gram of tissue samples. Sera were examined for the presence of antibodies against H5N1 virus by enzyme-linked immunosorbent assay (ELISA) [18]. The purified A/duck/Hokkaido/Vac-1/04 (H5N1) generated from H5N2 and H7N1 viruses isolated from migratory water birds by genetic reassortment in embryonated chicken eggs [19] was used as antigen for ELISA. Each of the birds was housed in a self-contained isolator unit (Tokiwa Kagaku) at a BSL-3 facility at the Graduate School of Veterinary Medicine, Hokkaido University, Japan.

The antigenic properties of H5 viruses, A/duck/Hokkaido/WZ21/08 (H5N2), A/duck/Hokkaido/WZ75/09 (H5N2), Mal/Hok/24/09 (H5N1), A/whooper swan/Hokkaido/1/08 (H5N1), and A/peregrine falcon/Hong Kong/810/09 (H5N1), were determined by the fluorescent antibody method with monoclonal antibodies (MAbs) against H5 HA produced previously [20]. MDCK cells infected with H5 influenza viruses were fixed with cold 100% acetone for eight hours post-inoculation. The reactivity patterns of the MAbs to H5 viruses were investigated by the immunofluorescent method with a FITC-conjugated goat IgG to mouse IgG (ICN Biomedicals). Fluorescence was visualized with the Axiovert 200 (Carl Zeiss).

In 2009, a total of 19 viruses were isolated from 711 fecal samples of migratory water birds. Those were 1 H1N3, 1 H1N5, 5 H4N6, 1 H5N1, 1 H5N2, 4 H6N1, 2 H6N8, 3 H11N9, and 1 H12N5 viruses. In our previous surveillance until 2008, H5N1 virus had not been isolated (Table 1) [6, 11‐13]. In autumn 2009, Mal/Hok/24/09 (H5N1) was isolated from a fecal sample of a mallard in Hokkaido, Japan.

The pathogenicity of Mal/Hok/24/09 (H5N1) was evaluated by IVPI test using chickens. None of the ten birds intravenously inoculated with Mal/Hok/24/09 (H5N1) showed clinical signs during ten days of observation (IVPI = 0.00). None of the chickens, domestic ducks, and quails intranasally inoculated with 10^6.0 EID₅₀ of Mal/Hok/24/09 (H5N1) showed clinical signs during 14 days of observation (Table 3). The virus was not recovered from the tracheal and cloacal swabs and tissues of chickens intranasally inoculated with Mal/Hok/24/09 (H5N1) on three dpi, and there were no antibodies to H5N1 virus detected by ELISA on 14 dpi (Table 3), indicating that chickens were not infected with the isolate. Although virus was not recovered from the swabs and tissues of domestic ducks inoculated with the virus on three dpi, antibodies against H5N1 virus were detected in the sera of the birds, indicating that domestic ducks were infected with the isolate. Viruses of 10^3.3 and 10^3.6 EID₅₀/ml were recovered from tracheal swabs of two of four quails inoculated with the virus on three dpi, respectively. Antibodies against H5N1 virus were detected in the sera of the birds on 14 dpi. These findings indicate that quails are susceptible to infection with the isolate.

Each gene of Mal/Hok/24/09 (H5N1) was phylogenetically analyzed. The HA and NA genes of Mal/Hok/24/09 (H5N1) were classified into the Eurasian lineage, and were different from HA and NA genes of H5N1 HPAIVs, respectively (Figure 1). In addition, the other six genes of Mal/Hok/24/09 (H5N1) were not closely related to those of HPAIVs, but related to those of NPAIVs isolated from migratory water birds (data not shown). The eight segments of Mal/Hok/24/09 (H5N1) were analyzed by the Basic Local Alignment Search Tool (BLAST) available from the DDBJ/EMBL/GenBank (Table 4). It was found that all genes of Mal/Hok/24/09 (H5N1) were derived from those of the viruses circulating in water birds in nature. M gene of the virus was classified into North American lineage, and the other genes were classified into Eurasian lineage (Table 4), indicating that genetic reassortment occurs between the viruses whose genes classified into North American and Eurasian lineages. The amino acid sequence of the HA cleavage site of Mal/Hok/24/09 (H5N1) was RETR/GLF, and insertion or substitution of multiple basic amino acids found in the HAs of HPAIVs [21] was not observed.

The HA of Mal/Hok/24/09 (H5N1) was antigenically analyzed using a panel of MAbs recognizing six different epitopes on the HA of A/duck/Pennsylvania/10218/84 (H5N2) [20]. Each of the MAb bound to the antigen of Mal/Hok/24/09 (H5N1) as well as those of the other non-pathogenic H5 viruses, and few MAbs bound to the antigen of H5N1 HPAIVs recently isolated in Mongolia, Japan, and Hong Kong (Table 5), indicating that the HA of Mal/Hok/24/09 (H5N1) is antigenically closely related to the H5 HA of the viruses circulating in migratory water bird.