Identification And Analysis Of Critical Amino Acids Matations In Epitopes In Hemagglutinin And Neuraminidase Of H9N2 Influenza Virus In The Pressure Of Antibody

Influenza virus is a negative-sense, single-stranded, segmented RNA virus of Orthomyxoviridae family. Multiple subtypes of influenza virus have been identified. Among these, H9N2 virus was first detected in the USA in 1966. After decades of evolution and spread, H9N2 virus is now prevalent all over the world. The infection of H9N2 virus is usually mild in nature, with respiratory syndrome and egg-dropping, but may lead to higher mortality if it is associated with secondary infection. In recent years, H9N2 virus has expanded its host range to infect mammals, including cats, pigs and humans. Moreover, H9N2 virus provided internal genes for the highly pathogenic H5N1 and H7N9 viruses, which have caused severe diseases and high mortality in human. For these reasons, H9N2 virus is put high in the list of influenza viruses with pandemic potential.Hemagglutinin (HA) and neuraminidase (NA) are two of the most important surface antigens of influenza virus. HA is responsible for virus binding to cell receptors while NA cleaves sialic acids from the cell surface, facilitating the release and spread of progeny virions. Antigenic sites on several HA (e.g., H1 and H3) and NA (e.g., N1) subtypes have been elucidated. For H9N2 virus, the crystal structure of H9 has been solved; however, no details for antigenic epitopes in these two proteins have been reported.In our studies, we generated mouse monoclonal antibodies (mAbs) against HA or NA of an H9N2 virus A/chicken/Jiangsu/x1/2004 (X1), and selected escape mutants that are resistant to neutralization/inhibition by these mAbs. We identified amino acids in HA and NA that are responsible for the resistance, thus mapping out a panel of key residues in the antigenic epitopes in HA and NA. Our data are of significance to the surveillance on the evolution of H9N2 virus, as well as the development and optimization of H9N2 influenza vaccines.1. Generation of mAbs against the HA or NA of H9N2 virusTo generate antibody reagents for antigenic mapping and surveillance on the evolution of HA and NA of H9N2 influenza virus, we immunized BALB/c mice with X1 virus, and performed a routine fusion with splenocytes from the immunized mice and mouse myeloma SP2/0 cells. Hybridomas secreting specific mAbs were screened with immunofluorescence assay (IFA). After two rounds of subcloning with limiting dilution,16 hybridomas were obtained. MAbs secreted by these hybridomas were used in subsequent assays. Eight of these mAbs, designated as 1C3,2G4,3B10,5B4, 6A5,6A10,6B6 and 6E6, inhibited the agglutination of red blood virus by X1 virus in hemagglutination inhibition (HI) assay, suggesting that these mAbs are against the HA of X1 virus. By performing IF A with 293 T cells that were transfected with a pcDNA plasmid containing the HA gene of X1 virus, we found mAb 2A8 is also specific to HA. Interestingly, this antibody neutralized X1 virus in microneutralization assay, although it was negative in HI assay, implying it recognizes an epitope that is far away from the receptor binding site of the HA. By performing NA inhibition assays with X1 virus and IFA with MDCK cells infected with H1N1, H1N2, or H9N1 viruses, we also identified two mAbs,1D1, and 1G8, which are specific to the NA of X1 virus. We characterized biological properties of these mAbs, and used these mAbs in antigenic mapping of the HA and NA of H9N2 virus.2. Identification of critical amino acids mutations in epitopes in the HA of H9N2 influenza virus in the pressure of the antibodiesHA is one of the most important surface antigens of influenza virus, serving as the target for the majority of neutralizing antibodies elicited against influenza virus infection. As H9N2 virus continues to evolve and reassert with other subtypes of influenza viruses, a further antigenic investigation on its HA of H9N2 virus is certainly necessary for effective control of this virus.Besides crystal structural analysis, an alternative way to map the antigenic sites of influenza virus proteins is to identify amino acid mutations present in these proteins of mutant viruses selected with specific mAbs. In this study, we chose 8 mAbs to investigate antigenic epitopes in the HA of a recent H9N2 virus X1. These 8 mAbs inhibited X1 virus with high titers in both HI and microneutralization (MN) assays, the HI titer was 211-219 and MN tieter was 1:5120-1:655360, suggesting that these antibodies are against the globular head region of H9. Escape mutants of X1 virus were selected with these antibodies, and a total of 10 amino acid mutations were detected in the HA of these mutant viruses, at positions 147,153,164,167,168,196, 198,200,201 and 207. Four of these mutations, at positions 147,153,200 and 207, had been identified previously by other groups, while the remaining 6, at positions 164, 167,168,196,198 and 207 have not been reported elsewhere. The 10 positions are within antigenic sites I and II in H9, and the 6 newly identified positions fall into antigenic area equivalent to antigenic site B in H3. Natural mutations at all of the 10 positions have occurred, indicative of immune pressure on H9N2 virus in the field. Our findings add more data to the composition of H9 antigenic sites, which are critical to the surveillance on H9N2 virus evolution.3. Identification of amino acid mutations in epitopes in NA of H9N2 influenza virus in pressure of the antibodiesNA is the second most abundant glycoprotein on the surface of influenza virus. It cleaves the linkage between HA and cell surface receptors, and thus helps the release and spread of influenza virus. It may also play a role in removing the decoy receptors in mucin, facilitating the entry of influenza virus into respiratory epithelium. NA is a tetramer, with an enzyme active center in each of its monomer.We generated 2 mAbs,1D1 and 1G8, which are against the NA of X1 virus. In the enzyme-linked lectin assay (ELLA), which uses a large molecule fetuin (molecular weight:50 kd) as substrate, both antibodies effectively inhibited the NA activity (IC50 were 1.8μg/mL and 0.4μg/mL, respectively). However, in Mu-NANA assay, which uses the small molecule Mu-NANA (molecular weight:489 d) as substrate, antibody 1G8 inhibited the NA activity of X1 virus (IC50 was 25.4μg/mL), while 1D1 did not. These data suggest that although both 1D1 and 1G8 mAbs can inhibit the NA activity,1D1 probably binds an epitope that is farther away from the enzyme active site than does 1G8, thus fails to prevent the access of the smaller substrate Mu-NANA to NA.We used 1D1 and 1G8 to select escape mutants of X1 virus. Antibody 1D1 selected an escape mutant, named m1D1, and 1G8 selected two escape mutants, named m1G8-1 and m1G8-2. Sequencing of the NA gene of these mutants revealed R338S mutation in the NA of m1D1, and D198N and K199E mutations in the NA of m1G8-1 and m1G8-2, respectively. An examination of the 3D structure of N2 revealed that residue 338 is located on the lateral side of the NA monomer, while residues 198 and 199 are on a loop in close proximity to the NA active center. This observation is consistent with the data from ELLA and Mu-NANA. We examined 1130 NA gene sequences available in GenBank (as of February 16,2015), and found that natural mutations have occurred at all of these 3 positions, indicative of immune pressure on the NA of H9N2 virus in the field. This is the first antigenic mapping of the NA of H9N2 virus, opening an avenue for the detailed investigation on its antigenic structure. Our findings are also helpful for developing NA antibody reagents against H9N2 virus.4. Identification of antigenic domain bound by an H9-specific mAb without HI functionIn recent years, there have been multiple reports about antibodies that recognize epitopes far away from the receptor-binding site of the HA. These antibodies can usually neutralize influenza viruses of various HA subtypes, and thus have the potential for serving as therapeutic reagents. Antibody 2A8 that we generated in our study has similar properties, e.g., it effectively neutralizes H9N2 virus in microneutralization assay but not in HI assay, it recognizes both H1 and H9 viruses.To localize the antigenic domain in HA bound by antibody 2A8, we selected escape mutants of X1 virus with this antibody. Four mutants were obtained, and 2 amino acid mutations, G270R and N282K, were detected in the HA of these mutant viruses. These 2 positions are within antigenic area IV in H9, or antigenic site C in H3. We examined 1563 full-length H9 sequences available in GenBank, and found the G270R and N282K mutations are already present in some of the natural H9N2 isolates, indicative of antibody selection pressure on this antigenic area of H9N2 viruses in the field. Our findings highlight the presence of antigenic domains on H9, which can elicit neutralizing antibodies reactive to viruses of multiple HA subtype, although these antibodies do not have ability to inhibit virus in HI assay.