A critical HA1 neutralizing domain of H5N1 influenza in an optimal conformation induces strong cross-protection

Abstract

The highly pathogenic avian influenza (HPAI) H5N1 viruses, especially the laboratory-generated H5N1 mutants, have demonstrated the potential to cross the species barrier and infect mammals and humans. Consequently, the design of an effective and safe anti-H5N1 vaccine is essential. We previously demonstrated that the full-length hemagglutinin 1 (HA1) could induce significant neutralizing antibody response and protection. Here, we intended to identify the critical neutralizing domain (CND) in an optimal conformation that can elicit strong cross-neutralizing antibodies and protection against divergent H5N1 strains. We thus constructed six recombinant proteins covering different regions of HA1 of A/Anhui/1/2005(H5N1), each of which was fused with foldon (Fd) and Fc of human IgG. We found that the critical fragment fused with Fd/Fc (HA-13-263-Fdc, H5 numbering) that could elicit the strongest neutralizing antibody response is located in the N-terminal region of HA1 (residues 13-263), which covers the receptor-binding domain (RBD, residues 112-263). We then constructed three additional recombinants fused with Fd plus His tag (HA-13-263-Fd-His), Fc only (HA-13-263-Fc), and His tag only (HA-13-263-His), respectively. We found that the HA-13-263-Fdc, which formed an oligomeric conformation, induced the strongest neutralizing antibody response and cross-protection against challenges of two tested H5N1 virus strains covering clade 1: A/VietNam/1194/2004 (VN/1194) or clade 2.3.4: A/Shenzhen/406H/06 (SZ/406H), while HA-13-263-Fc dimer and HA-13-263-Fd-His trimer elicited higher neutralizing antibody response and protection than HA-13-263-His monomer. These results suggest that the oligomeric form of the CND containing the RBD can be further developed as an effective and safe vaccine for cross-protection against divergent strains of H5N1 viruses.

Figure 1. Construction of recombinant truncated HA1 protein fragments fused with Fd and Fc.

(A) Schematic structure of HA protein of A/Anhui/1/2005 (H5N1 HA). SP, signal peptide. FP, fusion peptide. HA1-Fdc protein was previously expressed covering HA residues +3–322 , corresponding to residues 13–325 of H5 numbering as described in the published literature , , . (B) Truncated HA fragments covering various lengths of HA1 were constructed by fusing HA1 fragments with Fd and Fc. Each fragment contains IL2ss signal peptide at the 5′ terminus for the purpose of leading expressed proteins to the culture supernatant. The predicted RBD region was indicated in black.

Figure 2. SDS-PAGE and Western blot analysis of expressed proteins containing truncated HA1 fragments.

Protein samples were either boiled at 95°C for 5 min, or not boiled, followed by addition of 6× SDS-PAGE sample buffer. SDS-PAGE with Coomassie Blue staining was then run (A), as well as Western blot (B) by using a HA1-specific mAb. The protein molecular weight marker (kDa) (Invitrogen) is indicated on the left.

Figure 3. Detection of antibody responses and evaluation of neutralizing activity for sera of mice vaccinated with truncated HA1.

Sera from 10 days post-last vaccination were used for the detection. (A) Detection of sera IgG antibody responses by ELISA respectively coated with truncated HA1 protein fragments with the serum dilution at 1∶12800. The data are presented as mean A450± standard deviation (SD) of five mice per group. (B) ELISA detection of IgG binding ability to the truncated HA1 protein fragments. The data are presented as mean A450± SD of five mice per group at various dilution points. (C) Neutralizing antibody titers against HA of homologous strain A/Anhui/1/2005(H5N1) (AH-HA, clade 2.3.4) of H5N1 pseudovirus. The data are presented as mean 50% neutralizing antibody titer (NT₅₀) ± SD from five mice per group. * indicates significant difference (P<0.05) between HA-13–263-Fdc vaccination group and other groups.

Figure 4. Construction of protein fragments containing HA-13–263 with or without Fd or Fc and analysis of protein expression.

(A) Proteins containing residues of HA-13–263 with Fc (HA-13–263-Fc), with Fd (HA-13–263-Fd-His), or without fusion with Fd and Fc (HA-13–263-His) were constructed. Purified proteins were either boiled at 95°C for 5 min, or not boiled, followed by detection of the expression by SDS-PAGE and Coomassie Blue staining (B), and Western blot (C) by using a HA1-specific mAb. The protein molecular weight marker (kDa) (Invitrogen) is indicated on the left.

Figure 5. Detection of conformation of HA-13–263 proteins by native PAGE (N-PAGE) and cross-linker analysis.

(A) N-PAGE analysis of expressed HA-13–263 proteins, followed by Coomassie Blue staining and Western blot (B), by using a HA1-specific mAb. The native protein molecular weight marker (kDa) (Invitrogen) is indicated on the left. (C) Cross-linker analysis followed by SDS-PAGE and Coomassie Blue staining of expressed HA-13–263 proteins (with cross-linker). The proteins without cross-linking (w/o cross-linker) were used as the controls. The protein molecular weight marker (kDa) (Invitrogen) is indicated on the left.

Figure 6. Detection of IgG antibody responses and subtypes by ELISA in HA-13–263 protein-vaccinated mouse sera.

PBS was used as the negative control. Ability of IgG binding to HA1 (HA-13–263-His) protein (A) and full-length HA protein (B) was detected using mouse sera from 10 days post-last vaccination. The data are presented as mean A450± SD of five mice per group at various dilution points. Ability of IgG1 (C) and IgG2a (D) antibodies to bind to HA-13–263-His protein was detected using sera from 10 days post-last vaccination. The data are presented as mean A450± SD of five mice per group at various dilution points.

Figure 7. Comparison of neutralizing antibodies in sera of mice vaccinated with HA-13–263 proteins.

Sera were collected at 10 days post-last vaccination. PBS was used as the negative control. (A) Neutralizing antibody titers against HA of homologous AH-HA strain (clade 2.3.4) and heterologous A/Hong Kong/156/97 (HK-HA, clade 0) and A/VietNam/1194/2004 (1194-HA, clade 1) strains of H5N1 pseudovirus. The data are presented as mean NT₅₀± SD from five mice per group. * indicates significant difference (P<0.05), respectively, between HA-13–263-Fdc, HA-13–263-Fc, or HA-13–263-Fd-His vaccination groups and HA-13–263-His group or PBS control for the three H5N1 pseudoviruses. (B) Neutralizing antibody titers against heterologous strains of A/VietNam/1194/2004 (VN/1194, clade 1) and A/Shenzhen/406H/06 (SZ/406H, clade 2.3.4) H5N1 live viruses. The titers of neutralizing antibodies are presented as mean ± SD of five mice per group. * indicates significant difference (P<0.05), respectively, between HA-13–263-Fdc, HA-13–263-Fc, or HA-13–263-Fd-His vaccination groups and HA-13–263-His group or PBS control for both VN/1194 and SZ/406H viruses. The dotted line shows the limit of detection.

Figure 8. Comparison of cross-protective activity of HA-13–263 proteins against lethal H5N1 virus challenge.

PBS was used as the control. Survival rate (%) of mice vaccinated with HA-13–263 proteins challenged with lethal dose of H5N1 virus from VN/1194 (clade 1) (A) and SZ/406H (clade 2.3.4) (B). * indicates significant difference (P<0.05), respectively, between HA-13–263-Fdc and HA-13–263-His group, or PBS control, for VN/1194 virus, and between HA-13–263-Fdc, HA-13–263-Fc, or HA-13–263-Fd-His vaccination groups and HA-13–263-His group, or PBS control for SZ/406H virus. Percentage of body weight change (%) of HA-13–263-vaccianted mice after challenge with VN/1194 (C) and SZ/406H (D) H5N1 virus was shown.

Figure 9. Comparison of viral titers in virus-challenged lung tissues of mice vaccinated with HA-13–263 proteins.

Viral titers of VN/1194 (clade 1) and SZ/406H (clade 2.3.4) H5N1 viruses were determined in the lung tissues of the mice vaccinated with HA-13–263 proteins, respectively, collected on 5 day post-challenge. Mice injected with PBS were used as the negative control. * indicates significant difference (P<0.05), respectively, between HA-13–263-Fdc, HA-13–263-Fc, or HA-13–263-Fd-His vaccination groups and HA-13–263-His or PBS control group. The data are expressed as mean ± SD of viral titers 50% tissue culture infective dose (Log₁₀TCID₅₀/g) of lung tissues from five mice per group. The limit of detection is 1.5 Log₁₀TCID₅₀/g of tissues.