Leveraging vaccination-induced protective antibodies to define conserved epitopes on influenza N2 neuraminidase

Abstract

There is growing appreciation for neuraminidase (NA) as an influenza vaccine target; however, its antigenicity remains poorly characterized. In this study, we isolated three broadly reactive N2 antibodies from the plasmablasts of a single vaccinee, including one that cross-reacts with NAs from seasonal H3N2 strains spanning five decades. Although these three antibodies have diverse germline usages, they recognize similar epitopes that are distant from the NA active site and instead involve the highly conserved underside of NA head domain. We also showed that all three antibodies confer prophylactic and therapeutic protection in vivo, due to both Fc effector functions and NA inhibition through steric hindrance. Additionally, the contribution of Fc effector functions to protection in vivo inversely correlates with viral growth inhibition activity in vitro. Overall, our findings advance the understanding of NA antibody response and provide important insights into the development of a broadly protective influenza vaccine.

Keywords: antibody; cross-reactive; cryo-EM; influenza virus; neuraminidase; protective; structures; vaccination.

Figure 1.. Structural characterization of the three antibodies to the underside of NA head domain.

(A-C) Cryo-EM structures of Mos99 NA in complex with (A) 3C08 Fab, (B) 3A10 Fab, and (C) 1F04 Fab. For clarity, only the variable region of one Fab is displayed in each tetrameric structure. The NA is in surface representation with one protomer colored in white and the other three in dark gray. The active site of NA is highlighted in yellow. The heavy chain of each antibody is colored in cyan, and light chain in pink. Glycans are shown as sticks. (D-F) The epitopes of (D) 3C08, (E) 3A10, and (F) 1F04 are highlighted in blue on the NA structure. (G) The locations of classical antigenic regions of N2 NA (blue) as well as known antibody escape mutations in N2 NA from eight studies that are outside of the classical antigenic regions (purple)^,–. (H) The epitopes of NA antibodies (any influenza A subtype or type B) with structural information available in PDB. (I) The epitopes of all N2 NA antibodies in PDB. In all panels, Mos99 NA structure (PDB 7U4F) is used. See also Figure S4, Table S1, Table S2, and Table S3.

Figure 2.. Atomic interactions between NA and the three antibodies.

(A) Interactions between 3C08 and NA. (B-C) Two pockets on NA surface that are occupied by the CDR H3 of 3C08 are indicated. Key CDR H3 residues of 3C08 that interact with NA are shown. (D) Interactions between 3A10 and NA. (E) Key CDR H2 residues of 3A10 that interact with NA are shown. (F) Interactions between 1F04 and NA. (G) Key CDR H3 residues of 1F04 that interact with NA are shown. Light chain residues that interact with CDR H3 are also shown. (H) Key CDR L1 residues of 1F04 that interact with NA are shown. In panels A, D and F, epitope residues interacting with the heavy chain are shown in orange, whereas those interacting with the light chain are shown in yellow. Other settings are the same as Figure 1A–C. In all other panels, black dashed lines represent H-bonds. Residues labeled in red represent somatic hypermutations. See also Figure S2, Figure S5, and Figure S6.

Figure 3.. Sequence conservation of the underside epitopes.

(A-D) Sequence conservation score for each residue is shown on the NA structure (PDB 7U4F). A high conservation score indicates high sequence conservation. The epitopes of (A) 3C08, (B) 3A10, (C) 1F04, as well as (D) the classical antigenic regions are indicated by the cyan outline. (E-G) The bar charts indicate the buried surface area (BSA) of each residue in the epitopes of (E) 3C08, (F) 3A10, and (G) 1F04 upon binding. The sequence logos represent the sequence diversity of each epitope residue in natural human H3N2 strains from 1968 to 2020. Residues with side chains that H-bond with the indicated antibody are labeled by a black triangle underneath the sequence logo. (H) Sequence conservation scores of individual residues in the antibody epitopes as well as the classical antigenic regions are compared. The distributions of sequence conservation scores are shown as boxplots. P-values are calculated by two-sided Wilcoxon rank sum test. See also Table S4.

Figure 4.. In vitro functional activity of the three antibodies.

(A) Virus growth inhibition assay against recombinant viruses that carried NA from Mos99 (H3N2). The other seven segments of the recombinant viruses were from A/Puerto Rico/8/1934 (PR8). Zanamivir is used as a positive control. Minimum inhibitory concentration (MIC) is shown. (B) NA inhibition activity of 3C08, 3A10, and 1F04 was measured by ELLA assays, using recombinant Mos99 NA protein (C) NA inhibition activity of 3C08, 3A10, 1F04 at 100 μg/mL, and zanamivir (positive control) at 100 ng/mL was tested by MUNANA assays using Mos99 NA protein. Relative NA activity was computed by normalizing the NA activity to a negative control with no antibody.

Figure 5.. In vivo protection activity of the three antibodies.

(A-C) Female BALB/c mice at 6 weeks old were injected intraperitoneally with 5 mg/kg of the indicated antibody 2 hours prior to challenge with 5 LD₅₀ of a recombinant virus that carried NA from H3N2 Mos99 with the other seven segments from A/Puerto Rico/8/1934 (PR8). A representative experiment from two independent replicates with similar results is shown. (A) The mean percentage of body weight change post-infection is shown (n = 5). The humane endpoint, which was defined as a weight loss of 25% from initial weight on day 0, is shown as a dotted line. (B) Kaplan-Meier survival curves are shown (n = 5). (C) Lung viral titers on day 3 after infection are shown (n = 3). Solid black lines indicate means ± SD. The dotted line represents the lower detection limit. (D-I) Same as panel A and B, except (D, G) 1 mg/kg of the indicated antibody was injected intraperitoneally 2 hours prior to challenge, (E, H) 0.3 mg/kg of the indicated antibody was injected intraperitoneally 2 hours prior to challenge, and (F, I) 5 mg/kg of the indicated antibody was injected intraperitoneally 72 hours after challenge. (J-K) Same as panel A and B, except a recombinant virus that carried NA from H3N2 A/Singapore/INFIMH-16-0019/2016 (Sing16) with the other seven segments from A/Puerto Rico/8/1934 (PR8) was used. 2B04, which is a SARS-CoV-2 spike antibody, was used as a control. See also Figure S7.

Figure 6.. Fc effector functions contributes to in vivo protection.

(A-B) Antibody-dependent cellular cytotoxicity (ADCC) activity of the three antibodies was measured. FI6v3, which is an antibody to hemagglutinin (HA) and has known ADCC reporter assay activity against H3N2, was used as a positive control. IgG from human serum (Sigma, catalog #: I4506-50MG) was used as a negative control. ADCC activity of each sample was normalized to that of 2B04, which is a SARS-CoV-2 spike antibody. (A) Titration curve, and (B) area under the curve (AUC) are shown. (C-E) Same as Figure 5A–C, except antibodies with LALA-PG mutations, which eliminate Fc effector functions, were used. (F) P-value of the difference in weight loss between mice treated with the indicated antibody and its corresponding LALA-PG mutant at each day post-infection is computed by two-tailed Student’s t-test. (G) P-value of the difference in lung viral titer at day 3 post-infection between mice treated with the indicated antibody and its corresponding LALA-PG mutant is computed by two-tailed Student’s t-test. See also Figure S7.