First Outbreak of an H5N8 Highly Pathogenic Avian Influenza Virus on a Chicken Farm in Japan in 2020

On 5 November 2020, a confirmed outbreak due to an H5N8 highly pathogenic avian influenza virus (HPAIV) occurred at an egg-hen farm in Kagawa prefecture (western Japan). This virus, A/chicken/Kagawa/11C/2020 (Kagawa11C2020), was the first HPAI poultry isolate in Japan in 2020 and had multiple basic amino acids—a motif conferring high pathogenicity to chickens—at the hemagglutinin cleavage site. Mortality of chickens was 100% through intravenous inoculation tests performed according to World Organization for Animal Health criteria. Phylogenetic analysis showed that the hemagglutinin of Kagawa11C2020 belongs to clade 2.3.4.4B of the H5 Goose/Guangdong lineage and clusters with H5N8 HPAIVs isolated from wild bird feces collected in Hokkaido (Japan) and Korea in October 2020. These H5N8 HPAIVs are closely related to H5N8 HPAIVs isolated in European countries during the winter of 2019–2020. Intranasal inoculation of chickens with 106 fifty-percent egg infectious doses of Kagawa11C2020 revealed that the 50% chicken lethal dose was 104.63 and the mean time to death was 134.4 h. All infected chickens demonstrated viral shedding beginning on 2 dpi—before clinical signs were observed. These results suggest that affected chickens could transmit Kagawa11C2020 to surrounding chickens in the absence of clinical signs for several days before they died.


Introduction
Since 2003, highly pathogenic avian influenza viruses (HPAIVs) of the H5 subtype in which the hemagglutinin (HA) genes are derived from A/goose/Guangdong/1/1996 [1] -the so-called Goose/Guangdong (Gs/Gd) lineage-have been spreading to poultry and wild bird species worldwide, except in countries in South America and Oceania [2,3]. On 5 November 2020, an outbreak due to an H5N8 HPAIV was confirmed at an egg-hen farm where 300,000 of layer were raised in cages in Kagawa prefecture (western Japan). Eleven days before the outbreak (i.e., 24 October 2020), another H5N8 isolate, A/northern pintail/Hokkaido/M13/2020 [4], was detected in feces collected in Hokkaido prefecture (northern Japan). In addition, since the end of August 2020, H5N8 HPAIVs have been detected in wild bird and poultry species in Russia, Kazakhstan, Israel, European Union countries, and South Korea [3].
In the current study, we characterized the H5N8 HPAIV A/chicken/Kagawa/11C/2020 (Kagawa11C2020: H5N8), isolated from the first H5N8 poultry case in Japan in 2020, by using genetic analysis and experiments in chickens.

Virus Isolation
A cloacal swab from a dead chicken at the affected farm was inoculated into embryonated eggs; infectious allantoic fluid was examined for hemagglutination activity at the Kagawa Municipal Animal Hygiene Center and then was submitted for further study at the National Institute of Animal Health of Japan.

Sequencing
Viral RNA was extracted by using an RNeasy Mini Kit (QIAGEN, Benelux B.V., Amsterdam, The Netherlands), followed by cDNA library preparation by the NEBNext Ultra II RNA Library Prep Kit for Illumina (NEB, Ipswich, MA, USA). The cDNA was sequenced with a Miseq sequencer by using Miseq Reagent Kit version 2 (lllumina, San Diego, CA, USA) as previously described [13]. Consensus sequences were generated by using FluGAS software (version 2.25; World Fusion, Tokyo, Japan).

Phylogenetic Analysis
Among all the H5 HA sequences downloaded from GISAID and NCBI on 10 December in 2020, the H5 HA genes of Gs/Gd isolates were extracted and subjected to CD-Hit analysis with a homology threshold of 99.5%, although the sequences of viruses closely related to Kagawa11C2020 were not reduced. The HA gene sequences of Kagawa11C2020 and closely related viruses were aligned by using MAFFT version 7 [14] and analyzed according to a generalized time-reversible (GTR) model by using Fasttree version 2.1.10 [15]. The phylogenetic tree was visualized by using FigTree (version 1.4.4, The University of Edinburgh, Edinburgh, UK) [16].

Animal Experiments
Chickens were inoculated with H5N8 HPAIVs in the Biosafety Level 3 facilities at the National Institute of Animal Health, Japan. Four or five chickens were kept in each isolator with not restricted food and water intake. Infectious allantoic fluid of Kagawa11C2020 contained 10 9.45 EID 50/mL titer of the virus in each virus stock tube and was kept in −80 degree until used. The pathogenicity of the H5N8 Japanese strains, Kagawa11C2020, was evaluated by inoculation intravenously in 7-week-old chickens according to the Manual of Diagnostic Tests and Vaccines for Terrestrial Animals 2019 [17]. Eight of seven-week-old chickens (L-M-6 strain; Nisseiken Co., Ltd., Tokyo, Japan) were intravenously inoculated with 200 µL of a 1/10 dilution of the infectious allantoic fluid of Kagawa11C2020 containing 10 7.75 EID 50 titer of the virus in 200 µL. As a natural infection model, twenty of 4-week-old chickens were used for intranasal inoculation of Kagawa11C2020. One of four doses (10 2 of fifty-percent egg infectious dose (EID 50 ), 10 4 EID 50 , 10 5 EID 50 , and 10 6 EID 50 ) of virus was inoculated intranasally into five 4-week-old healthy White Leghorn chickens that had never been exposed to influenza virus(L-M-6 strain; Nisseiken Co., Ltd., Tokyo, Japan). Tracheal and cloacal swabs were collected when the chickens died or at 1, 2, 3, 5, 7, 10, and 14 days post-inoculation (dpi). The sample collection from chickens and viral titration of samples were conducted as previously described [18]. Influenza A Virus Antibody Test Kit (IDEXX Laboratories, Westbrook, ME, USA) was used to detect antibodies against type A influenza viruses in sera collected from surviving chickens at 14 dpi. Survival analysis according to log-rank testing was conducted by using SYSTAT 13.2 software (HULINKS, Tokyo, Japan) with Holm correction. The mean time to death (MTD) of infected chickens and the viral titers in tracheal and cloacal swabs obtained during animal experiments were compared between Kagawa11C2020 group and Japan isolate experimental groups by using Dunnett's test.

Results and Discussion
The deduced amino-acid sequence of the HA cleavage site of Kagawa11C2020 is PLREKRRKR/GLFG, which includes the run of consecutive basic amino acids that is the characteristic motif of HPAIVs [19], suggesting that Kagawa11C2020 is highly pathogenic to chickens. High pathogenicity to chickens was confirmed through testing according to OIE criteria, because all of the seven-week-old chickens inoculated intravenously died within 48 h, resulting in a mortality rate of 100%.
Blast analysis to the GISAID database of each segment of Kagawa11C2020 revealed that they possessed high similarity to the viruses isolated in Europe from January to March, 2020 along with A/Mandarin duck/Korea/H242/2020 (GISAID isolate ID, EPI_ISL_631824) which was isolated in October 2020 in Korea. They were also highly homologous (99.0%; NS gene to 99.8%; HA gene) to A/northern pintail/Hokkaido/M13/2020 described by Isoda et al. [4].
Phylogenetic analysis showed that the HA gene of Kagawa11C2020 belongs to clade 2.3.4.4B and clusters with those of A/northern pintail/Hokkaido/M13/2020 [4], A/mandarin duck/Korea/K20-551-4/2020 [20], and A/Mandarin duck/Korea/H242/2020 (GISAID isolate ID, EPI_ISL_631824) ( Figure 1). The HA of Kagawa11C2020 also is closely related to those of H5N8 subtype European HPAIVs detected in wild-bird and poultry species during January-March 2020, such as A/chicken/Germany-BW/AI00049/2020. The phylogenetic trees of the other seven gene segments showed the same clustering pattern as that for the HA gene (data not shown). The branch including Kagawa11C2020 as well as the European strains stems from the HPAIVs isolated in South Africa in 2017. In contrast, the viruses in the Kagawa11C2020-containing branch are not closely related to the H5N1, H5N5, and H5N8 HPAIVs isolated from wild birds and poultry in Europe during September-November 2020, although they also belong to clade 2.3.4.4B. Chickens were inoculated intranasally with Kagawa11C2020 to estimate the 50% chicken lethal dose (CLD50) and mean time to death (MTD). All of the chickens inoculated with 10 6 EID50 and four of the five chickens inoculated with 10 5 EID50 died, whereas all of the chickens inoculated with the 10 2 or 10 4 EID50 dose survived. All chickens that died  Chickens were inoculated intranasally with Kagawa11C2020 to estimate the 50% chicken lethal dose (CLD 50 ) and mean time to death (MTD). All of the chickens inoculated with 10 6 EID 50 and four of the five chickens inoculated with 10 5 EID 50 died, whereas all of the chickens inoculated with the 10 2 or 10 4 EID 50 dose survived. All chickens that died from infection showed depression beginning on 3 dpi at the earliest (one chicken), but comb cyanosis occurred in only one chicken in each of the 10 5 and10 6 EID 50 groups. The MTD of chickens inoculated with 10 6 EID 50 was 134.4 h, and the CLD 50 of Kagawa11C2020 was 10 4.63 (Table). None of the chickens that survived in 10 2 , 10 4 and 10 5 EID 50 inoculation groups was ELISA-positive for antibody against type A influenza virus; they had therefore not been infected with the virus.
In our examination of virus shedding from intranasally inoculated chickens, virus was detected at 1 dpi in one of the five tracheal swabs from chickens inoculated with either 10 6 EID 50 or 10 5 EID 50 virus. All infected chickens in the 10 6 EID 50 and 10 5 EID 50 inoculation groups shed virus on 2 dpi (Figure 2). These results indicated that virus was shed from the trachea and in feces before clinical signs caused by the Kagawa11C2020 HPAIV became apparent. Therefore, chickens infected with Kagawa11C2020 might cause greater pollution of the farm environment before being noticed than birds infected with HPAIVs with shorter MTDs.
Infections of migratory wild birds, such as white-fronted goose, with those viruses in Europe that were prevalent during last winter from 2019 to 2020 were evident [23]. These species of migratory wild birds tend to winter in Europe and breed in Siberia during the summer, where they may co-mingle with migratory birds that winter in Asia. Therefore, migratory birds might have carried H5N8 HPAIVs that were isolated in early 2020 from Europe to Siberia, where the viruses could have been passed to birds that migrate to far-east Asia, thus providing a mechanism for the poultry outbreaks that occurred in Korea and Japan [12]. Although the intermediate ancestor of Kagawa11C2020 was not found in Siberia during summer 2020, intense surveillance for AIVs in Siberia during various birds' breeding seasons is important, so that the poultry industry in this region might have advance warning of potential HPAI. To this end, it is important to identify where co-mingling of European and Asian migratory birds occurs. *: Mean time to death was significantly shorter than for Kagawa11C2020. **: Significantly different from Kagawa11C2020 according to survival analysis. N.A represents "not applicable". ***: Mean of maximum viral titer was significantly lower than for Kagawa11C2020. †: These values were retrieved from reference [5].
Viruses 2021, 13, 489 6 of 7 10 6 EID50 or 10 5 EID50 virus. All infected chickens in the 10 6 EID50 and 10 5 EID50 inoculation groups shed virus on 2 dpi (Figure 2). These results indicated that virus was shed from the trachea and in feces before clinical signs caused by the Kagawa11C2020 HPAIV became apparent. Therefore, chickens infected with Kagawa11C2020 might cause greater pollution of the farm environment before being noticed than birds infected with HPAIVs with shorter MTDs. The maximum viral titer in tracheal and cloacal swabs collected from chickens experimentally infected with Kagawa11C2020 occurred on the day of death or 1 day before death (Figure 2). In addition, the mean maximum viral titers of tracheal and cloacal swabs from chickens experimentally inoculated with 10 6 EID50 of Kagawa11C2020 were similar to, or significantly higher than, those from birds inoculated with other Japan isolates, A/Mandarin duck/Miyazaki/22M-765/2011 (MiyazakiMD2011, H5N8), A/chicken/Shimane/1/2010 (Shimane2010C, H5N1), Miyazaki2014 and A/northern goshawk/Tokyo/1301B003T/2018 (Tokyo2018, H5N6) ( Table 1).
Infections of migratory wild birds, such as white-fronted goose, with those viruses in Europe that were prevalent during last winter from 2019 to 2020 were evident [23]. These species of migratory wild birds tend to winter in Europe and breed in Siberia during the summer, where they may co-mingle with migratory birds that winter in Asia. Therefore, migratory birds might have carried H5N8 HPAIVs that were isolated in early 2020 from Europe to Siberia, where the viruses could have been passed to birds that migrate to fareast Asia, thus providing a mechanism for the poultry outbreaks that occurred in Korea and Japan [12]. Although the intermediate ancestor of Kagawa11C2020 was not found in Siberia during summer 2020, intense surveillance for AIVs in Siberia during various birds' breeding seasons is important, so that the poultry industry in this region might have advance warning of potential HPAI. To this end, it is important to identify where co-mingling of European and Asian migratory birds occurs.

Institutional Review Board Statement:
Animal experiments in this study were conducted in compliance with the institutional protocol, which was reviewed and approved by the Institutional Animal Care and Use Committee of National Agriculture and Food Research Organization, NARO (approval number 20-066, 17 November 2020, approval number 10-115, 31 March 2011). Animal experiments referred in this paper [5,11,18,21,22] were conducted under the ethical approval as mentioned in each paper (approval number 14-

Data Availability Statement:
The data that support the findings of this study are available from the corresponding author upon reasonable request.