Highly pathogenic avian influenza virus of the A/H5N8 subtype, clade 2.3.4.4b, caused outbreaks in Kazakhstan in 2020

Abstract

Background: Large poultry die-offs happened in Kazakhstan during autumn of 2020. The birds' disease appeared to be avian influenza. Northern Kazakhstan was hit first and then the disease propagated across the country affecting eleven provinces. This study reports the results of full-genome sequencing of viruses collected during the outbreaks and investigation of their relationship to avian influenza virus isolates in the contemporary circulation in Eurasia.

Methods: Samples were collected from diseased birds during the 2020 outbreaks in Kazakhstan. Initial virus detection and subtyping was done using RT-PCR. Ten samples collected during expeditions to Northern and Southern Kazakhstan were used for full-genome sequencing of avian influenza viruses. Phylogenetic analysis was used to compare viruses from Kazakhstan to viral isolates from other world regions.

Results: Phylogenetic trees for hemagglutinin and neuraminidase show that viruses from Kazakhstan belong to the A/H5N8 subtype and to the hemagglutinin H5 clade 2.3.4.4b. Deduced hemagglutinin amino acid sequences in all Kazakhstan's viruses in this study contain the polybasic cleavage site (KRRKR-G) indicative of the highly pathogenic phenotype. Building phylogenetic trees with the Bayesian phylogenetics results in higher statistical support for clusters than using distance methods. The Kazakhstan's viruses cluster with isolates from Southern Russia, the Russian Caucasus, the Ural region, and southwestern Siberia. Other closely related prototypes are from Eastern Europe. The Central Asia Migratory Flyway passes over Kazakhstan and birds have intermediate stops in Northern Kazakhstan. It is postulated that the A/H5N8 subtype was introduced with migrating birds.

Conclusion: The findings confirm the introduction of the highly pathogenic avian influenza viruses of the A/Goose/Guangdong/96 (Gs/GD) H5 lineage in Kazakhstan. This virus poses a tangible threat to public health. Considering the results of this study, it looks justifiable to undertake measures in preparation, such as install sentinel surveillance for human cases of avian influenza in the largest pulmonary units, develop a human A/H5N8 vaccine and human diagnostics capable of HPAI discrimination.

Keywords: A/H5N8; Bayesian MCMC; Birds migration; Central Asia Migratory Flyway; Gs/GD lineage; Highly pathogenic avian influenza virus; Kazakhstan; Phylogenetic analysis; Whole genome sequencing.

Figure 1. Map of Kazakhstan (with adjacent territories of neighbor countries) showing locations where samples were collected for full-genome sequencing.

Figure 2. Phylogenetic tree for the hemagglutinin (HA) gene.

The tree is split in two parts, each part in a separate panel. The consensus tree was built using the Bayesian inference as described in the Materials and Methods. Branch tips are labeled with the GISAID isolate name (EPI_ISL), accession number (EPI) and additional markings. Numbers at tree nodes show the statistical support (posterior probability percent). A multiple alignment used to build this tree is provided in the Supplementary Information. (A) HPAI A/H5N8 viruses from Kazakhstan year-2020 outbreaks cluster together with contemporary isolates from Eurasia and belong to the H5 clade 2.3.4.4b. Hemagglutinin clades 2.3.4.4a–2.3.4.4 h as well as clades 2.3.4.1–2.3.4.3 (outgroup) are present in the tree. (B) Expanded look of the collapsed branch from (A). Viruses from Kazakhstan cluster together with isolates from Russia (Siberia, Urals, Caucasus) and Eastern Europe. Labels: viruses sequenced by our group are labeled with filled circles next to GISAID accession numbers. Isolates sequenced in other laboratories and downloaded from GISAID for this study are labeled with filled squares. Clusters with statistical support 99–100% are labeled B1–B5 for describing the results.

Figure 3. Phylogenetic tree for the neuraminidase (NA) gene.

The tree was built and labeled similarly to the tree in Fig. 2. A multiple alignment for the NA tree is presented in the Supplementary Information. NA clusters labeled B1–B4 correspond to the eponymous HA clusters in Fig. 2. Members of one HA cluster B5 in the NA tree are present as separate branches B5-1–B5-3.