Genetics of H5N1 and H5N8 High-Pathogenicity Avian Influenza Viruses Isolated in Japan in Winter 2021–2022

In winter 2021–2022, H5N1 and H5N8 high-pathogenicity avian influenza (HPAI) viruses (HPAIVs) caused serious outbreaks in Japan: 25 outbreaks of HPAI at poultry farms and 107 cases in wild birds or in the environment. Phylogenetic analyses divided H5 HPAIVs isolated in Japan in the winter of 2021–2022 into three groups—G2a, G2b, and G2d—which were disseminated at different locations and times. Full-genome sequencing analyses of these HPAIVs revealed a strong relationship of multiple genes between Japan and Siberia, suggesting that they arose from reassortment events with avian influenza viruses (AIVs) in Siberia. The results emphasize the complex of dissemination and reassortment events with the movement of migratory birds, and the importance of continual monitoring of AIVs in Japan and Siberia for early alerts to the intrusion of HPAIVs.

Here, we phylogenetically compared the full genomes of H5 HPAIVs that caused all outbreaks and some of the cases in wild birds in Japan in 2021-2022.We inferred the background of the dissemination of H5 HPAIVs in Japan, considering both the origins of all genes of the Japanese H5 HPAIVs in the winter of 2021-2022 and the movement of migratory birds.

Virus Isolation and Whole-Genome Sequencing
Tracheal or cloacal swabs collected at Japanese poultry farms with suspected HPAIV infections were inoculated into embryonated chicken eggs for virus isolation at the diagnostic laboratories of the livestock hygiene service centers in each farm's prefecture.Allantoic fluid inoculated with swab samples that showed HA activity against chicken red blood cells and swab samples from dead wild birds in Hokkaido Prefecture were submitted to the National Institute of Animal Health (Tsukuba, Japan) for diagnosis.
The whole genomes of the isolated viruses were obtained [17].In brief, cDNA libraries for next-generation sequencing were prepared with a NEBNext Ultra II RNA Library Prep Kit for Illumina (New England Biolabs, Ipswich, MA, USA) and sequenced with a MiSeq Reagent Kit v. 2 (Illumina, San Diego, CA, USA).Consensus sequences were generated in CLC Genomics Workbench (v.9.5.3,Qiagen, Hilden, Germany) and FluGAS software (v.2.2.5, World Fusion, Tokyo, Japan) [23].The newly determined viral sequences have been deposited in the GISAID database (http://platform.gisaid.org(accessed on 15 January 2024)) with the accession numbers listed in Table 1.

Phylogenetic Analysis
In Japan in the winter of 2021-2022, a total of 132 HPAI reports were made, comprising 25 outbreaks at poultry farms and 107 positive identifications in wild birds or in the environment.We used strains isolated from all 25 outbreaks at poultry farms and 32 detections in wild birds in Hokkaido, whose diagnosis is responsible in the National Institute of Animal Health, for phylogenetic analysis.We downloaded the sequences of the avian influenza viruses (AIVs) from the NCBI (Influenza Virus Resource; https://www.ncbi.nlm.nih.gov/genomes/FLU/Database/nph-select.cgi?go=database) and GISAID databases (accessed on 13 June 2022).The GISAID sequences were aligned with those of AIVs stored at the National Institute of Animal Health in Japan in BioEdit v. 7.2.5 [24] and MAFFT v. 7.490 [25] software.Removal of sequences with ambiguous nucleotide bases left 100,361 sequences for genes for polymerase basic protein 2 (PB2), 98,773 for PB1, 103,136 for polymerase acidic protein (PA), 11,028 for H5, 105,489 for nucleoprotein (NP), 60,075 for N1, 5549 for N8, 108,555 for matrix protein (MP), and 109,352 for nonstructural protein (NS) for the phylogenetic analyses.Maximum-likelihood trees based on the aligned sequences were constructed in FastTree v. 2.1.10software [26].

Phylogenetic Analysis of Japanese H5 Isolates from the 2021-2022 Winter
Maximum likelihood analysis using the HA gene of the Japanese H5 HPAIVs isolated in the winter of 2021-2022 placed the viruses in three clusters in clade 2.3.4.4b-G2a,G2b, and G2d (Figure 1a)-which were previously named 20A, 20E, and 21E, respectively [21].The average nucleotide sequence identity of the HPAIVs within a clade was >99.7%.That between the HPAIVs in G2a and G2b was 98.6%, and that between both of those and G2d was 97.9%.
G2a H5N8 HPAIVs isolated in November 2021 shared a common ancestor with viruses that were isolated in Japan, Korea, and China in the 2020-2021 winter (December 2020-February 2021) (Figure 1b).While almost all strains in this cluster were identified as H5N8, H5N6 isolates were also detected in humans in China during July to September 2021, implying the possibility of transmission from birds.G2a H5N8 HPAIVs were not detected in the nine months following February 2020 and might thus have been retained in the Asian continent and reintroduced into Japan in the 2021-2022 winter.
G2b H5N1 HPAIVs caused outbreaks from November 2021 to May 2022 (Figure 1c).The H5 HA genes shared a common ancestor with European H5N8 HPAIVs in the 2020-2021 winter, and those of Japanese isolates were closely related to the H5N8 HPAIVs isolated in the Novosibirsk region of Russia in August 2020 and May 2021.
G2d H5N1 HPAIVs caused outbreaks from February to May 2022 (Figure 1d).The H5 HA genes shared a common ancestor with European H5N1 HPAIVs in the 2021-2022 winter.The H5 HA genes of Japanese and European G2d H5N1 HPAIVs branched from those of H5N1 HPAIVs isolated in the Chelyabinsk, Novosibirsk, and Omsk regions of Russia during August and September 2021 as well as an isolate from Bangladesh in December 2021.

Characteristics of the Outbreaks in Japan during 2021-2022
In Japan, 25 outbreaks due to H5N1 and H5N8 HPAIVs were reported across 12 prefectures between 10 November 2021 and 13 May 2022 (Table 1; Figure 2a).Around the same time, 107 positive identifications in wild birds or in the environment were reported across seven prefectures (Table S1).The three clusters of H5 HPAIVs isolated during the 2021-2022 winter had different detection periods.G2a H5N8 HPAIVs caused only two outbreaks, in Akita and Kagoshima, on 10 and 15 November, respectively, and none have been detected in Japan since (Figure 2b).G2b H5N1 HPAIVs were first detected in Kagoshima on 13 November and caused outbreaks throughout Japan (Table 1; Figure 2c).With the exception of the one in Hokkaido, outbreaks caused by G2b H5N1 HPAIVs were reported from November 2021 to January 2022.It is worth noting that the last case was reported 3½ months after the others (Table 1).Outbreaks caused by G2d H5N1 HPAIVs were reported in northern Honshu (Iwate, Miyagi, Aomori, and Akita) and Hokkaido during February to May 2022 (Figure 2d).Out of nine cases caused by G2d H5N1 HPAIVs, six were found in chickens and three in emus.

Characteristics of the Outbreaks in Japan during 2021-2022
In Japan, 25 outbreaks due to H5N1 and H5N8 HPAIVs were reported across 12 fectures between 10 November 2021 and 13 May 2022 (Table 1; Figure 2a).Around same time, 107 positive identifications in wild birds or in the environment were repo across seven prefectures (Table S1).The three clusters of H5 HPAIVs isolated during 2021-2022 winter had different detection periods.G2a H5N8 HPAIVs caused only outbreaks, in Akita and Kagoshima, on 10 and 15 November, respectively, and none been detected in Japan since (Figure 2b).G2b H5N1 HPAIVs were first detected in K shima on 13 November and caused outbreaks throughout Japan (Table 1; Figure 2c).W the exception of the one in Hokkaido, outbreaks caused by G2b H5N1 HPAIVs were reported from November 2021 to January 2022.It is worth noting that the last case was reported 3½ months after the others (Table 1).Outbreaks caused by G2d H5N1 HPAIVs were reported in northern Honshu (Iwate, Miyagi, Aomori, and Akita) and Hokkaido during February to May 2022 (Figure 2d).Out of nine cases caused by G2d H5N1 HPAIVs, six were found in chickens and three in emus.

Gene Constellations of H5 HPAIVs Isolated in Japan in the 2021-2022 Winter
Phylogenetic analyses based on all RNA segments of the Japanese H5 HPAIVs revealed that almost all HPAIVs had internal genes with the same relationship as the HA genes: internal genes of the Japanese H5N8 HPAIVs formed a cluster with those of Asian and Russian isolates in 2020-2021, those of G2b H5N1 HPAIVs formed a cluster with the H5N8 HPAIVs in the 2020-2021 winter, and those of G2d H5N1 HPAIVs formed a cluster with the isolates that circulated in Europe in the 2021-2022 winter (Figures S1-S15, Table S2).Exceptionally, G2b H5N1 HPAIVs that caused outbreaks in Ehime during late December 2021 to early January 2022 had PB1 genes that were phylogenetically distant from

Gene Constellations of H5 HPAIVs Isolated in Japan in the 2021-2022 Winter
Phylogenetic analyses based on all RNA segments of the Japanese H5 HPAIVs revealed that almost all HPAIVs had internal genes with the same relationship as the HA genes: internal genes of the Japanese H5N8 HPAIVs formed a cluster with those of Asian and Russian isolates in 2020-2021, those of G2b H5N1 HPAIVs formed a cluster with the H5N8 HPAIVs in the 2020-2021 winter, and those of G2d H5N1 HPAIVs formed a cluster with the isolates that circulated in Europe in the 2021-2022 winter (Figures S1-S15, Table S2).Exceptionally, G2b H5N1 HPAIVs that caused outbreaks in Ehime during late December 2021 to early January 2022 had PB1 genes that were phylogenetically distant from other Japanese G2b H5N1 HPAIVs (Figures 3 and 4) but close to those of AIVs isolated in Siberia.In addition, Japanese H5 HPAIVs had NA and internal genes that shared a common ancestor with some non-H5 AIVs.N1 genes of Japanese G2b H5N1 HPAIVs were closely related to that of the H1N1 AIV isolated in Novosibirsk in August 2020 (Figure 5), and AIVs in Siberia had internal genes that were phylogenetically related to those of Japanese G2b H5N1 HPAIVs (Figure 6).
other Japanese G2b H5N1 HPAIVs (Figures 3 and 4) but close to those of AIVs isolated in Siberia.In addition, Japanese H5 HPAIVs had NA and internal genes that shared a common ancestor with some non-H5 AIVs.N1 genes of Japanese G2b H5N1 HPAIVs were closely related to that of the H1N1 AIV isolated in Novosibirsk in August 2020 (Figure 5), and AIVs in Siberia had internal genes that were phylogenetically related to those of Japanese G2b H5N1 HPAIVs (Figure 6).

Discussion
In the winter of 2021-2022, three phylogenetically distinct types of H5 HPAIVs (G2a, G2b, and G2d) intruded into Japan and caused serious outbreaks.Since the first case of HPAI in Japan in 2004 [6], that of 2021-2022 was the longest (November 2021-May 2022).The migration of multiple groups of wild birds might have contributed to the extended detection of H5 HPAIVs in 2021-2022.At the same time, at least two intrusions of H5 HPAIVs contributed to the extended outbreak: G2a and G2b H5 HPAIVs intruded around the start of November, and outbreaks caused by both were reported in northern and southern Japan.G2d H5 HPAIVs started to be detected in February 2022, and they caused outbreaks in limited areas in northern Japan.Along the East-Asia-Australian flyway lap in Japan [27], wild birds are considered to migrate to Japan along four routes [28]: the Kamchatka Peninsula-Kuril Islands, Sakhalin, Sea of Japan, and Korean Peninsula.Thus, the dissemination of G2d H5 HPAIVs might correspond to the flyways through the Kamchatka Peninsula-Kuril Islands and Sakhalin, resulting in their limited distribution.G2b H5 HPAIV caused an outbreak in Hokkaido in May 2022 after no detection for 3½ months.Spring movement northward via Hokkaido to Siberia might also contribute to the dissemination of viruses, resulting in the outbreak in spring.
Detection periods differed among clades: G2a H5 HPAIVs were detected only in November, but the other two were detected over 3 months.These differences might be due to the number of wild birds that are immunologically susceptible to G2a H5 HPAIVs.Phylogenetic analyses revealed that G2a H5 HPAIVs caused outbreaks in Asia in 2020-2021 [17,[29][30][31] and were reintroduced in 2021-2022, while G2b and G2d H5 HPAIVs were detected for the first time in 2021-2022.These results imply that the migratory birds coming into Japan may have become more resistant to G2a H5 HPAIVs through previous exposure, resulting in the prevalent circulation of G2b and G2d H5 HPAIVs in 2021-2022.Further studies on antibodies and the higher prevalence of individual H5 HPAIVs among wild birds are needed to assess this explanation.

Discussion
In the winter of 2021-2022, three phylogenetically distinct types of H5 HPAIVs (G2a, G2b, and G2d) intruded into Japan and caused serious outbreaks.Since the first case of HPAI in Japan in 2004 [6], that of 2021-2022 was the longest (November 2021-May 2022).The migration of multiple groups of wild birds might have contributed to the extended detection of H5 HPAIVs in 2021-2022.At the same time, at least two intrusions of H5 HPAIVs contributed to the extended outbreak: G2a and G2b H5 HPAIVs intruded around the start of November, and outbreaks caused by both were reported in northern and southern Japan.G2d H5 HPAIVs started to be detected in February 2022, and they caused outbreaks in limited areas in northern Japan.Along the East-Asia-Australian flyway lap in Japan [27], wild birds are considered to migrate to Japan along four routes [28]: the Kamchatka Peninsula-Kuril Islands, Sakhalin, Sea of Japan, and Korean Peninsula.Thus, the dissemination of G2d H5 HPAIVs might correspond to the flyways through the Kamchatka Peninsula-Kuril Islands and Sakhalin, resulting in their limited distribution.G2b H5 HPAIV caused an outbreak in Hokkaido in May 2022 after no detection for 3½ months.Spring movement northward via Hokkaido to Siberia might also contribute to the dissemination of viruses, resulting in the outbreak in spring.
Detection periods differed among clades: G2a H5 HPAIVs were detected only in November, but the other two were detected over 3 months.These differences might be due to the number of wild birds that are immunologically susceptible to G2a H5 HPAIVs.Phylogenetic analyses revealed that G2a H5 HPAIVs caused outbreaks in Asia in 2020-2021 [17,[29][30][31] and were reintroduced in 2021-2022, while G2b and G2d H5 HPAIVs were detected for the first time in 2021-2022.These results imply that the migratory birds coming into Japan may have become more resistant to G2a H5 HPAIVs through previous exposure, resulting in the prevalent circulation of G2b and G2d H5 HPAIVs in 2021-2022.Further studies on antibodies and the higher prevalence of individual H5 HPAIVs among wild birds are needed to assess this explanation.
Several genes of H5 HPAIVs detected in Japan in 2021-2022 were phylogenetically related to those of the non-H5 AIVs in wild birds, indicating that they emerged through reassortment with AIVs.Notably, AIVs detected in Siberia had genes that were phylogenetically closely related to those of Japanese isolates (Figures 4 and 5).These results suggest that the reassortment events happen in migratory birds in Siberia, the summer breeding range, and that the H5 HPAIVs are disseminated through southward migration, as reported previously [32,33].
G2d H5 HPAIVs caused outbreaks on poultry farms in northern Japan, including three emu farms.Full-genome analyses revealed that the G2d H5 HPAIVs isolated from emus had lysine (K) at position 627 in the PB2 segment while isolates from chickens had glutamic acid (E) at that position in the present study.The substitution E627K enhances replication and the pathogenicity of H5 HPAIVs in mammalian hosts [34,35], and the amino acid at position 627 in PB2 is reportedly exclusively glutamic acid in bird isolates and lysine in human isolates [36,37].Internal tissues and organs of emus co-express α-2,3 and α-2,6 sialic acid receptors [38].In the present study, emus are sensitive to avian-origin H5 HPAIVs and selected viruses with the substitution E627K in PB2, which can enhance replication in mammalian hosts, whereas our emu isolates had no substitutions that are deduced to increase the preference for α-2,6 sialic acid receptors.These results suggest that emus contribute to the emergence of new viruses with pandemic potential in humans.Further experiments to test host range and replication activity in cells are needed.
In summary, phylogenetic analyses revealed that H5 HPAIVs of three H5 HA genotypes caused influenza outbreaks in Japan during the 2021-2022 winter, each with a different mode of dissemination.Full-genome sequencing analyses revealed a strong relationship between viruses from Japan and Siberia, suggesting that the HPAIVs arose from reassortment events with AIVs in Siberia.Our results emphasize the complex of dissemination and reassortment events of HPAIVs with the movement of migratory birds and the importance of continual monitoring of AIVs in Japan and Siberia for early alerts to the intrusion of HPAIVs.

Figure 3 .
Figure 3. Section of the maximum-likelihood tree based on PB1 genes.G2b H5 HPAIVs isolated in Ehime Prefecture are in green.Fast-global bootstrap values of ≥0.60 are shown.

Figure 3 .Figure 4 .
Figure 3. Section of the maximum-likelihood tree based on PB1 genes.G2b H5 HPAIVs isolated in Ehime Prefecture are in green.Fast-global bootstrap values of ≥0.60 are shown.Viruses 2024, 16, x FOR PEER REVIEW 8 of 13

Figure 4 .Figure 5 .
Figure 4. Section of the maximum-likelihood tree based on PB1 genes.G2b H5 HPAIVs isolated in this study are green.Fast-global bootstrap values of ≥0.60 are shown.

Figure 5 .
Figure 5. Section of the maximum-likelihood tree based on NA genes.G2b H5 HPAIVs isolated in this study are in green.Fast-global bootstrap values of ≥0.60 are shown.The arrow indicates the isolate in Siberia.

Figure 6 .
Figure 6.Section of the maximum-likelihood tree based on NP genes.G2b H5 HPAIVs isolated in this study are in green.Fast-global bootstrap values of ≥0.60 are shown.

Figure 6 .
Figure 6.Section of the maximum-likelihood tree based on NP genes.G2b H5 HPAIVs isolated in this study are in green.Fast-global bootstrap values of ≥0.60 are shown.

Table 1 .
Information on H5 HPAIVs isolated from poultries in this study.