Molecular characterization, receptor binding property, and replication in chickens and mice of H9N2 avian influenza viruses isolated from chickens, peafowls, and wild birds in eastern China

Abstract

H9N2 avian influenza viruses are widely prevalent in birds and pose an increasing threat to humans because of their enhanced virulence and transmissibility in mammals. Active surveillance on the prevalence and evolution of H9N2 viruses in different avian hosts will help develop eradication measures. We isolated 16 H9N2 viruses from chickens, green peafowls, and wild birds in eastern China from 2017 to 2019 and characterized their comparative genetic evolution, receptor-binding specificity, antigenic diversity, replication, and transmission in chickens and mice. The phylogenetic analysis indicated that the green peafowl viruses and swan reassortant shared the same ancestor with the poultry H9N2 viruses prevalent in eastern China, while the seven wild bird viruses belonged to wild bird lineage. The chicken, peafowl, and swan H9N2 viruses that belonged to the poultry lineage preferentially recognized α-2, 6-linked sialic acids (human-like receptor), but the wild bird lineage viruses can bind both α-2, 3 (avian-like receptor) and human-like receptor similarly. Interestingly, the H9N2 viruses of poultry lineage replicated well and transmitted efficiently, but the viruses of wild bird lineage replicated and transmitted with low efficiency. Importantly, the H9N2 viruses of poultry lineage replicated in higher titer in mammal cells and mice than the viruses of wild birds lineage. Altogether, our study indicates that co-circulation of the H9N2 viruses in poultry, wild birds, and ornamental birds increased their cross-transmission risk in different birds because of their widespread dissemination.

Keywords: Avian influenza virus; H9N2; chicken; peafowl; wild birds.

Figure 1.

Genetic relationships among the HA gene and genotypes of H9N2 influenza viruses isolated from chickens, peafowls, and wild birds. (A) The genotype of the 16 H9N2 viruses. Groups were divided in each phylogenetic tree as shown in Figure 1B, Figure 2B, Figure 3, and Figure S1 according to their nucleotide homology, then genotypes were confirmed according to the group combination of eight gene segments of each virus. The genetic lineage indicates the genetic lineage (poultry-originated lineage, P; or wild bird originated lineage, W). (B) Bayesian time-measured phylogenetic tree of HA genes of H9N2 viruses. The HA MCC tree was constructed with the BEAST software package (v1.10.4) and then visualized by FigTree (v1.4.4). Branches are coloured by posterior probability, and the node bars indicate 95% highest posterior density of the node height. Tip labels in black are reference sequences downloaded from the GISAID database, and the left 16 sequences in this study are coloured according to their hosts.

Figure 2.

Genetic relationships among the NA genes and genotype evolution of H9N2 viruses. (A) Genetic evolution and reassortment of H9N2 viruses. (B) Bayesian time-measured phylogenetic trees of NA genes of H9N2 viruses. The NA MCC tree was constructed with the same software package and method as the Figure 1 legend. The 16 sequences in this study are coloured according to their hosts, as shown in the figure.

Figure 3.

Phylogenetic diagram of PB2 (A), PB1 (B), PA (C), NP (D), M (E), and NS (F) genes of H9N2 avian influenza viruses. The phylogenetic tree was constructed by MEGA 7.0 with the Neighbor-Joining method. The virus name was not shown in the tree. The branches and the sequence name with colour (red, blue, green, pink) were the viruses isolated in this study, and the sequences with black were download from the database. Larger versions of the images of the phylogenetic trees, with more detailed sequence information, were provided in Figure S1 A-F in the supplemental part.

Figure 4.

Characterization of the receptor-binding specificity of H9N2 viruses. The binding affinity of the test viruses to two different glycopolymers was tested. The data shown are the mean of three replicates; the error bar indicates the standard deviation.

Figure 5.

Replication of H9N2 viruses in chickens. Viral titers in the organs of chickens on day 3 p.i. with 10⁶EID₅₀ of test virus. Data shown are the mean with SD titers from three chickens. The dashed line indicates the lower limit of detection. The virus titers of CK/932/18 and GP/1656/19 in the trachea and lung of chickens were compared with the SW/10429/19. Statistical analysis was performed by using one-way ANOVA with GraphPad Prism 8 software. *, P < 0.05; ns, not significant.

Figure 6.

Virulence and replication of H9N2 viruses in mice. (A) Bodyweight change of the mice infected with H9N2 viruses isolated from chickens. (B) Bodyweight change of the mice infected with H9N2 viruses isolated from green peafowls and wild birds. Viral titers were detected in eggs (C-L). Data shown are the mean titer from three mice; the error bar indicates the standard deviation. The dashed line indicates the lower limit of detection. Viral titers in the nasal turbinates and lungs of mice at 3 dpi were compared with viral titers at 5 dpi, and significance was tested with a Student's t-test. *, P < 0.05; **, P < 0.01.

Figure 7.

Virus growth curves and polymerase activity of two H9N2 viruses. Virus growth curves in MDCK (A), A549 (B), and CEF (C) cells. The dashed lines indicate the lower limit of detection. Data shown are mean ± SD for three independent experiments, and significance was assessed with a two-tailed unpaired Student’s t-test. (D) Polymerase activity of WD/11452/19 and CK/98/18 in 293 T cells. All of the data were normalized to the activity of the WD/11452/19 sample. Data shown are the mean polymerase activity ± standard deviation (n = 3). Statistical analysis was performed by using one-way ANOVA with GraphPad Prism 8 software. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, not significant.