We conducted passive surveillance of HPAIV infections in wild birds in a public garden in Sapporo, the prefectural capital of Hokkaido, Japan; ≈2,000 crows flock together during winter and are observed by garden staff. We isolated viruses from tracheal and cloacal swab samples collected from dead crows in the garden by inoculating 10-day-old embryonated eggs; we confirmed results by using reverse transcription PCR (Appendix). On November 23 and 24, 2023, we isolated H5N1 HPAIVs from 2 dead large-billed crows (Corvus macrorhynchos), designated as A/large-billed crow/Hokkaido/B067/2023 (H5N1) and A/large-billed crow/Hokkaido/B068/2023 (H5N1). The hemagglutinin (HA) gene sequences from those 2 H5N1 HPAIVs indicated they clustered with the G2d subgroup of HPAIVs found in Hokkaido during the winter seasons 2021–22 and 2022–23. In contrast, HA genes of 3 H5 HPAIVs isolated from dead crows on January 8–11, 2024, were closely related to the G2a subgroup of H5N5 HPAIVs found in northern Europe and North America. Subsequent whole-genome sequencing analysis of the 3 G2a HPAIVs confirmed their subtype was H5N5; we named them A/large-billed crow/Hokkaido/B073/2024 (H5N5), A/large-billed crow/Hokkaido/B074/2024 (H5N5), and A/crow/Hokkaido/B075/2024 (H5N5) (Table 1).

Figure

Figure. Phylogenetic analysis of genetically distinct highly pathogenic avian influenza H5N5 and H5N1 viruses isolated in Japan in winter 2023–24. H5 hemagglutinin (A) and N5 neuraminidase (B) gene segments of H5N5...

We phylogenetically analyzed virus isolates along with reference sequences obtained from GISAID (https://www.gisaid.org); the HA genes of H5N5 HPAIVs isolated in Hokkaido diverged considerably from HPAIVs isolated in Japan during winter 2020–21 (10), forming a distinct branch within the G2a subgroup (Figure). In addition, the other gene segments of H5N5 HPAIVs from Hokkaido were genetically distant from those in HPAIV strains isolated in Japan during winter 2021–22 (Appendix Figures 1–6). BLAST (https://blast.ncbi.nlm.nih.gov) analysis of sequences from GISAID revealed that all 8 gene segments of H5N5 HPAIVs from Hokkaido were very close (genetic similarity >99%) to H5N5 HPAIVs detected in northern Europe since 2022, in contrast to those from North America (Table 2), suggesting a low possibility of virus transmission from North America. H5N5 HPAIVs from Hokkaido shared a common ancestor with H5N5 HPAIV from Europe assigned the genotype EA-2021-I by the European Food Safety Authority (11). Parent strains of H5N5 HPAIVs from Europe, represented by A/swan/Rostov/2299-2/2020 (H5N5), were proposed to originate in western Russia during autumn 2020. Those viruses underwent genetic evolution via reassortment events involving H5N8 HPAIVs circulating in Europe since 2018 (12) and the N5 NA gene derived from concurrently circulating LPAIVs (13). H5N5 HPAIVs reported in northern Europe during 2022–2023 exhibited specific genetic differences compared with H5N5 HPAIVs detected in Europe during autumn 2020, particularly in the N5 NA gene. Those differences included a 66-bp nucleotide deletion within the N5 NA gene, which we also observed in the H5N5 HPAIVs from Hokkaido. Truncation of the NA stalk has been attributed to the adaptation of those viruses from wild birds to Galliformes spp. birds (14). However, most H5N5 HPAIV infections in Europe were detected in wild birds, and no cases have been detected in Galliformes spp. birds since 2022 (15). Further investigation is needed to clarify whether NA stalk truncation affects pathogenesis of H5N5 HPAIVs.

During winter 2023–24, we confirmed H5N5 HPAIV infections in wild birds, especially in crows, in Erimo (December 19, 2023, in south-central Hokkaido) and in Kushiro (January 18, 2024, in eastern Hokkaido); we also confirmed infection in a peregrine falcon (Falco peregrinus) in Tamana, Kumamoto Prefecture, Kyushu Island, on January 16, 2024. We classified the isolate from Tamana, A/peregrine falcon/Kumamoto/4301C001/2024 (H5N5), into the G2d subgroup according to its HA gene sequence, whereas its NA gene sequence was similar to that of LPAIVs isolated in East Asia (Table 2). Although this combination had not been observed in Japan, reassortment events between the HPAIV H5N1 G2d subgroup and LPAIVs have been documented (9). We detected H5N5 HPAIVs in Hokkaido in January 2024; a total of 85 crows were found dead in the Sapporo garden, 80 of which we diagnosed as HPAIV positive by the end of April. No HPAIVs were detected in birds within the garden after April 2024. The continuous detection of H5N5 HPAIVs in the Sapporo garden during January–April without unusual deaths of birds other than crows and multiple isolations of H5N5 HPAIVs in other areas of Hokkaido suggest the potential for widespread dissemination of H5N5 HPAIVs within the Hokkaido region.

H5N1 G2d HPAIVs persisted in crows residing in the Sapporo garden even after the introduction of H5N5 G2a viruses, indicating concurrent circulation of genetically distinct viruses within a single crow population. Indeed, the average nanopore sequencing coverage for A/large-billed crow/Hokkaido/B080/2024 (H5N1) was 5497.4 reads for the N1 NA gene (G2d subgroup) and 1943.7 reads for the N5 NA gene (G2a subgroup) (Appendix Table 1). This observation suggests single hosts were co-infected with 2 viruses and reassortment occurred between viruses originating from geographically distant areas. Antigenic characterization of H5N5 HPAIVs suggested that the antigenicity of A/large-billed crow/Hokkaido/B073/2024 (H5N5) was close (2–4-fold differences in hemagglutination inhibition titers) to that of other H5 HPAIVs in the G2d subgroup (Table 3) despite their genetic diversity (Appendix Table 2).

We found that H5N5 HPAIVs consisting of unique gene constellations were likely introduced into Japan through a step-by-step bird migration through northern Eurasia. We confirmed the cocirculation of 2 genetically distinct viruses in a single flock of crows. The presence of H5N5 HPAIV infections in waterfowl in Japan is relatively unknown, and the lack of reports from neighboring countries on the presence of H5N5 HPAIVs from Europe has hampered the reconstruction of this genotype’s spread to eastern Asia. Continuous monitoring and rapid information sharing between countries are needed to determine the global dynamics of HPAIVs and prevent their spread.