Epitope specificity plays a critical role in regulating antibody-dependent cell-mediated cytotoxicity against influenza A virus

Abstract

The generation of strain-specific neutralizing antibodies against influenza A virus is known to confer potent protection against homologous infections. The majority of these antibodies bind to the hemagglutinin (HA) head domain and function by blocking the receptor binding site, preventing infection of host cells. Recently, elicitation of broadly neutralizing antibodies which target the conserved HA stalk domain has become a promising "universal" influenza virus vaccine strategy. The ability of these antibodies to elicit Fc-dependent effector functions has emerged as an important mechanism through which protection is achieved in vivo. However, the way in which Fc-dependent effector functions are regulated by polyclonal influenza virus-binding antibody mixtures in vivo has never been defined. Here, we demonstrate that interactions among viral glycoprotein-binding antibodies of varying specificities regulate the magnitude of antibody-dependent cell-mediated cytotoxicity induction. We show that the mechanism responsible for this phenotype relies upon competition for binding to HA on the surface of infected cells and virus particles. Nonneutralizing antibodies were poor inducers and did not inhibit antibody-dependent cell-mediated cytotoxicity. Interestingly, anti-neuraminidase antibodies weakly induced antibody-dependent cell-mediated cytotoxicity and enhanced induction in the presence of HA stalk-binding antibodies in an additive manner. Our data demonstrate that antibody specificity plays an important role in the regulation of ADCC, and that cross-talk among antibodies of varying specificities determines the magnitude of Fc receptor-mediated effector functions.

Keywords: ADCC; Fc receptor; NK cell; antibody; influenza virus.

Fig. 1.

HAI⁺ antibodies inhibit induction of ADCC by bnAbs that bind to the HA stalk domain. (A and B) The ADCC assay was performed using primary NK cells. ADCC was measured as a percentage of activation above HK/68 H3-expressing target cells without antibody. The data shown represent the mean of technical replicates from each independent donor. (C and D) ADCC activation of NK cells grown in continuous culture was also assessed after incubation on infected A549 target cells incubated with the indicated antibodies. Data represent the mean and SEM of four independent experiments performed in duplicate. *P < 0.05, Student’s t test. (E–G) ADCC reporter assays were performed on X-31 (H3N2) cells (E and G) and on Cal/09-infected A549 cells (F) using combinations of murine mAbs (E and F) or human mAbs (G). The mean and SEM values for biological experiments performed in triplicate are shown.

Fig. S1.

Primary NK cell ADCC assay. (A) Primary human NK cells were grown in continuous culture on K562 cells. Gates were set on the CD56⁺CD3^- subset (NK cells) for downstream analysis. (B) IFNγ and CD107a expression was monitored on uninfected A549 target cells and on infected A549 cells not treated with antibody. (C) On incubation of infected target cells with C05, CR8020, or C05 and CR8020, NK cell activation was assessed by measuring IFNγ and CD107a expression. (D) Primary NK cells freshly isolated from PBMCs using a CD56⁺ selection kit were incubated with 293T cells expressing HK/68 H3. Cells were incubated with C05, CR8020, or C05 and CR8020. NK cell activation was assessed by measuring the IFNγ and CD107a expression.

Fig. 2.

Nonneutralizing HA-binding antibodies neither induce nor inhibit ADCC induced by HA stalk-binding bnAbs. (A) Western blot analysis was performed against 293T cells transfected with plasmids expressing His-tagged HK/68 H3 or N2 (negative control) using anti-His, FEE8, or FBE9. (B) The neutralization activities of 9H10, FEE8, and FBE9 were tested against X-31 by a microneutralization assay on MDCK cells. (C and D) The ability of FEE8 (Neut⁻, HA head-binding) and (C) FBE9 (Neut⁻, HA stalk-binding) (D) to induce ADCC or inhibit ADCC induced by 9H10 was evaluated on X-31–infected A549 cells. Because these assays were performed in parallel, the curves for 9H10, XY102 + 9H10, and control IgG + 9H10 are duplicated in C and D for clarity and ease of comparison. Data represent the mean and SEM values of biological experiments performed in triplicate.

Fig. S2.

Lack of potent ADCC induction by nonneutralizing antibodies is independent of affinity. (A) ELISAs were performed using recombinant HK/68 H3 HA to compare the binding affinities of CR8020, 1G05, 3B01, and 5E02. (B) The neutralization activities of CR8020, 1G05, 3B01, and 5E02 were measured against X-31 by a microneutralization assay on MDCK cells. (C–E) In vitro ADCC assays were performed on X-31–infected A549 cells using the human HA stalk-binding bnAb CR8020 either alone or in combination with human nonneutralizing mAb 1G05 (C), 3B01 (D), or 5E02 (E). Because the experiments were performed in parallel, the curves for CR8020, C05 + CR8020, and control IgG + CR8020 are shown in triplicate in C, D, and E for clarity and ease of comparison. Data represent the mean and SEM of biological experiments run in triplicate.

Fig. 3.

NAI⁺ antibodies are capable of cooperating with HA stalk-binding bnAbs to boost ADCC induction. (A) ELISAs were performed using recombinant HK/68 N2 protein to test the binding affinities of 3C2, 4D6, and 8C10. (B–D) In vitro ADCC assays were performed on X-31–infected A549 cells using the HA stalk-binding antibody 9H10 either alone or in combination with the NAI⁺ antibodies 3C2 (B), 8C10 (C), or 4D6 (D). Data represent the mean and SEM values of biological experiments performed in triplicate.

Fig. 4.

The induction of ADCC by antibodies in human polyclonal serum is regulated by the relative proportion of ADCC activating and inhibiting antibodies. Polyclonal IgG was purified from the serum of human donors over protein G-Sepharose columns. (A) Quantities of H3 stalk-binding antibodies were determined by ELISA against recombinant cH5/3 HA. (B and C) Quantities of total HA-binding antibodies (B) and N2-binding antibodies (C) were determined by ELISA against recombinant HK/68 HA or NA. (D–F) In vitro ADCC assays were performed on X-31–infected A549 cells using polyclonal IgG isolated from healthy adult donors either alone or in combination with the HAI⁺ human mAb antibody C05. (D) Donor 1. (E) Donor 2. (F) Donor 3. Data represent the mean and SEM values of biological experiments performed in triplicate.

Fig. 5.

ADCC inhibition by HAI⁺ antibodies is mediated by competition for binding HA on the surface of infected cells and virus particles. (A) In vitro ADCC assays were performed on cH4/3 virus-infected A549 cells using XY102 (HAI⁺), 9H10 (stalk-binding bnAb), or combinations thereof. (B and C) ELISA-based binding competition assays were performed on X-31–infected A549 cells (B) or on purified X-31 (C) using XY102 (HAI⁺), biotinylated 9H10 (HA stalk-binding bnAb), or both. The limit of detection of these assays is shown as a dashed line (background + 3 SD). (D) Another binding competition assay was performed on purified X-31 using biotinylated 9H10 and FEE8 (HAI⁻, head-binding). Data represent the mean and SEM values of biological experiments performed in triplicate.

Fig. S3.

The Fc region of the HAI⁺ antibody is not required for blocking of the ADCC mediated by HA stalk-binding bnAbs. (A) Binding of 9H10, XY102, or both was assayed by biolayer interferometry against immobilized recombinant HK/68 H3 protein. (B) Western blot analysis was performed to test the purity of XY102 F(ab)₂ using anti-murine F(ab)₂ antibody. (C) ELISAs were performed using recombinant HK/68 H3 to compare the binding affinities of XY102 IgG and XY102 F(ab)₂. XY102 IgG and XY102 F(ab)₂ were tested at equal molarity. (D) In vitro ADCC assays were performed on X-31–infected A549 cells using XY102, XY102 F(ab)₂, 9H10, or combinations thereof. XY102 IgG and XY102 F(ab)₂ were tested at equal molarity. Data represent the mean and SEM of biological experiments run in triplicate.

Fig. S4.

Model of ADCC activation by IAV-specific antibody specificity. (A and B) HA stalk-binding bnAbs potently stimulate the activation of ADCC (A), whereas HAI⁺ antibodies that bind to the HA head domain do not (B). (C) However, HAI⁺ antibodies are able to inhibit ADCC induction by stalk-binding bnAbs through direct competition for binding to HA on the surface of viral particles and infected cells. (D) HA-reactive nonneutralizing antibodies do not inhibit ADCC mediated by HA stalk-binding bnAbs. (E) NAI⁺ antibodies do not potently activate ADCC on their own. (F) However, they can cooperate with HA stalk-binding bnAbs to boost ADCC induction.