Figure legends:
Figure 1. (A) The location of amino acid mutations in the three-dimensional structure of HA protein of H9N2 subtype avian influenza virus. Yellow color indicates the locations of the HA receptor binding sites including the positions 146-150, 109, 161, 163, 191, 198, 202, 203, and 232-237. Pink color (the positions 133, 145, 164, 168, 198 and 234) indicates mutations on antigenic sites that have been reported. Red color (the positions 131, 181, 264, 270, 274, 278, 386 and 399) indicates mutations that have not been previously reported as antigenic sites. (B) HI titers of F/98 immune sera from chickens (n=8) to each recombinant virus with single mutants in HA from the passaged viruses occurring antigenic drift in embryonated chicken eggs or chickens. (C) HI titers of F/98 immune sera from chickens (n=8) to each recombinant virus with multiple mutants in HA from the passaged viruses occurred in the 47th generation in embryonated chicken eggs under selective pressure on antibodies. A ≥4-fold change in HI titers of standard antiserum was considered as significant antigenic change.
Figure 2. (A) The location of amino acid mutations in the three-dimensional structure of HA protein of H9N2 subtype avian influenza virus. Blue color indicates the locations of the position 198 in the HA receptor binding site. Yellow color indicates the locations of the HA receptor binding sites including the positions 109, 161, 163, 191, 198 (blue color), 202, 203; orange color indicates the locations of the right edge of the receptor binding pocket 232-237 and the left edge of the receptor binding pocket 146-150. (B) MN titers of F/98 immune sera from chickens (n=8) to the viruses F/98 or rF/HAA198V. (C) MN titers of rF/HAA198Vimmune sera from chickens (n=8) to the viruses F/98 or rF/HAA198V. (D) MN titers of GD/SS/94 immune sera from chickens (n=6) to the viruses F/98, rF/HAA198V or GD/SS/94. (E) MN titers of JS/YZ618/12 immune sera from chickens (n=6) to the viruses F/98, rF/HAA198V or JS/YZ618/12. (F) MN titers of rF/HAA198V immune sera from chickens (n=6) to the viruses F/98, rF/HAA198V, GD/SS/94, or JS/YZ618/12. A ≥4-fold change in MN titers of standard antiserum was considered as significant antigenic change. These data indicated the fold change was relative to the virus rF/HAA198V.
Figure 3. (A) HA mutations A198V increase receptor binding avidity of H9N2 virus F/98, whereas S145N decreases receptor binding avidity. Relative viral receptor binding avidities were determined by hemagglutination of red blood cells pretreated with increasing amounts of α2-3,6,8 neuraminidase. Data are expressed as the maximal amount of neuraminidase that allowed full agglutination. The data are representative of three independent experiments. (B, C) Single A198V mutation does not affect Ab binding. Direct antibody binding to F/98 or rF/HAA198V viruses were determined by ELISA using sera collected from chickens vaccinated with inactivated F/98 (B) or rF/HAA198V (C). The AUC of ELISA was calculated for virus-Abs binding by GraphPad Prism 8 software above the value of the corresponding negative control, which was performed under the same conditions. Means and SD from three independent experiments. Statistical significance was based on student’s t test (**P < 0.01; *** P < 0.001). O.D., optical density.
Figure 4. (A) Neuraminidase activities of F/98 and rF/HAA198V viruses, which were determined in the fluorescence intensity of NA fluorogenic substrate. (B) Virus elution from chicken erythrocytes. Two-fold dilutions of F/98 and rF/HAA198V viruses were incubated with equal volumes of chicken erythrocytes at 4 ℃ for 30 min, and the HA titers at 37℃ representing virus elution from chicken erythrocytes was monitored for 6 hours. The result was presented as the percentage of the initial HA titer at 4℃. (C, D) Virus release assay. A198V HA mutation does not affect viral infection (C), and prevents virus released from MDCK cells surface significantly (D). The results were displayed as mean±SDof three independent experiments. The comparisons were performed with t test. **Indicates very significant difference between groups (P <0.01).
Figure 5. The cross-HI reactions between F/98 and rF/HAA198V viruses. Three-week-old SPF chickens were vaccinated once by subcutaneous injection of 0.3 mL of oil-emulsion of inactivated whole virus vaccines of the F/98 and rF/HAA198V viruses, which were inactivated by adding 0.2% formalin (v /v) for 24 h at 37 °C, respectively. At 21 d.p.v., 12 chickens from each group (the F/98 vaccine group and the rF/HAA198V vaccine group) were bled to analyze cross-HI titers against the F/98 and rF/HAA198V viruses. Statistical significance was based on student’s t test (*P< 0.05).