Plasticity of Amino Acid Residue 145 Near the Receptor Binding Site of H3 Swine Influenza A Viruses and Its Impact on Receptor Binding and Antibody Recognition

Abstract

The hemagglutinin (HA), a glycoprotein on the surface of influenza A virus (IAV), initiates the virus life cycle by binding to terminal sialic acid (SA) residues on host cells. The HA gradually accumulates amino acid substitutions that allow IAV to escape immunity through a mechanism known as antigenic drift. We recently confirmed that a small set of amino acid residues are largely responsible for driving antigenic drift in swine-origin H3 IAV. All identified residues are located adjacent to the HA receptor binding site (RBS), suggesting that substitutions associated with antigenic drift may also influence receptor binding. Among those substitutions, residue 145 was shown to be a major determinant of antigenic evolution. To determine whether there are functional constraints to substitutions near the RBS and their impact on receptor binding and antigenic properties, we carried out site-directed mutagenesis experiments at the single-amino-acid level. We generated a panel of viruses carrying substitutions at residue 145 representing all 20 amino acids. Despite limited amino acid usage in nature, most substitutions at residue 145 were well tolerated without having a major impact on virus replication in vitro All substitution mutants retained receptor binding specificity, but the substitutions frequently led to decreased receptor binding. Glycan microarray analysis showed that substitutions at residue 145 modulate binding to a broad range of glycans. Furthermore, antigenic characterization identified specific substitutions at residue 145 that altered antibody recognition. This work provides a better understanding of the functional effects of amino acid substitutions near the RBS and the interplay between receptor binding and antigenic drift.IMPORTANCE The complex and continuous antigenic evolution of IAVs remains a major hurdle for vaccine selection and effective vaccination. On the hemagglutinin (HA) of the H3N2 IAVs, the amino acid substitution N 145 K causes significant antigenic changes. We show that amino acid 145 displays remarkable amino acid plasticity in vitro, tolerating multiple amino acid substitutions, many of which have not yet been observed in nature. Mutant viruses carrying substitutions at residue 145 showed no major impairment in virus replication in the presence of lower receptor binding avidity. However, their antigenic characterization confirmed the impact of the 145 K substitution in antibody immunodominance. We provide a better understanding of the functional effects of amino acid substitutions implicated in antigenic drift and its consequences for receptor binding and antigenicity. The mutation analyses presented in this report represent a significant data set to aid and test the ability of computational approaches to predict binding of glycans and in antigenic cartography analyses.

Keywords: H3 subtype; hemagglutinin; influenza vaccines; swine influenza; virus evolution.

FIG 1

Amino acid plasticity at residue 145. (A) Publicly available H3 HA sequences were retrieved, and the relative frequencies of identified amino acids at residue 145 were calculated for swine, human, avian, canine, and equine AIVs. Amino acids present at frequency below 1% are not labeled in the figure. (B) Schematic representation of the HA gene segment of A/turkey/Ohio/313053/2004 (H3N2), depicting the codon corresponding to residue 145 and respective amino acid substitutions introduced by site-directed mutagenesis. (C) HA monomeric structure of A/turkey/Ohio/313053/2004 (H3N2), indicating the location of HA residue 145.

FIG 2

Substitutions at residue 145 show no major impact on virus growth but decrease receptor binding avidity. (A) Confluent monolayers of MDCK cells were inoculated with H3 viruses carrying amino acid substitutions at residue 145 at an MOI of 0.01 and incubated at 37°C. At 6, 12, 24, 48, and 72 hpi, tissue culture supernatants from inoculated cells were collected for virus RNA quantification by rRT-PCR, which was expressed as log₁₀ TCID₅₀/ml equivalents. Plotted data represent means ± standard deviations (SD). (B and C) Turkey red blood cells pretreated with different amounts of neuraminidase from either Clostridium perfringens (B) or Arthrobacter ureafaciens (C) were mixed with H3 viruses carrying amino acid substitutions at residue 145 to quantify virus agglutination as a measure of virus binding avidity. Data are expressed as the maximal amount of neuraminidase that allowed full agglutination. Panels B and C show representative data of one out of two and three independent experiments, respectively, with samples run in duplicates in each experiment. Plotted data represent means ± standard deviations (SD).

FIG 3

Mutants with substitutions at residue 145 retain binding to SAα2-6Gal. H3 viruses carrying amino acid substitutions at residue 145 were tested for receptor binding specificity with various concentrations of SAα2-3Gal (α-2,3-linked SA) or SAα2-6Gal (α-2,6-linked SA). (A to P) 145 N (wt) virus (A); 145 A virus (B); 145 F virus (C); 145 G virus (D); 145 H virus (E); 145 I virus (F); 145 K virus (G); 145 L virus (H); 145 M virus (I); 145 P virus (J); 145 Q virus (K); 145 R virus (L); 145 S virus (M); 145 T virus (N); 145 V virus (O); 145 Y virus (P). (Q and R) Human pH1N1 (Q) and avian ΔH5N1 (R) were used as controls for binding to SAα2-6Gal and SAα2-3Gal, respectively. Glycan concentration is expressed as arbitrary units (AU). Plotted data represent means ± standard deviations (SD). Data are representative of 2 independent experiments with 2 replicates per experiment.

FIG 4

Substitutions at residue 145 modulate binding to a broad range of SAα2-6Gal glycans. Glycan microarray analysis of H3 viruses carrying amino acid substitutions at residue 145 was performed. The array is comprised of nonsialoside control (1 to 10; gray), SAα2-3Gal (11 to 76; yellow), and SAα2-6Gal (77 to 128; green) glycans. Glycans are grouped by structure type: L, linear; O, O linked; N, N linked; and L^x, sialyl Le^x. (A) 145 N (wt) virus; (B) 145 A virus; (C) 145 F virus; (D) 145 G virus; (E) 145 H virus; (F) 145 I virus; (G) 145 K virus; (H) 145 L virus; (I) 145 M virus; (J) 145 P virus; (K) 145 Q virus; (L) 145 R virus; (M) 145 S virus; (N) 145 T virus; (O) 145 V virus; (P) 145 Y virus. Plotted data represent means ± standard deviations (SD) using four replicates per virus. RFU, relative fluorescence units.

FIG 5

Substitutions at residue 145 modulate serum reactivity. Antibody responses to H3 viruses carrying amino acid substitutions at residue 145 was determined by ELISA using swine antisera generated against OH/04 (A), NY/11 possessing HA residue 145 N (B), or IA/14 possessing HA residue 145 K (C). Two sets of sera were tested independently. The first set of sera was tested 2 times in duplicates. The second set of sera was tested only once in duplicate. Plotted data represent means ± standard deviations (SD). O.D., optical density.

FIG 6

Substitutions at residue 145 impact HI titers. HI titers were measured against H3 viruses carrying amino acid substitutions at residue 145 using swine antisera generated against OH/04 (cyan), NY/11 possessing HA residue 145 N (red) or IA/14 possessing HA residue 145 K (light green). For each panel, two sets of sera were tested independently. The first set of sera was tested 3 times in duplicates. Confirmatory tests were run on a second set of sera, which was tested once in duplicates. The serum reactivities of all H3 mutant viruses to the respective swine antisera are depicted as fold change (log₂ HI titer for mutant virus/log₂ HI titer for homologous virus). Plotted data represent means ± standard deviations (SD). Colors are based on the antigenic cluster designation for swine H3N2 IAVs.