Influenza Anti-Stalk Antibodies: Development of a New Method for the Evaluation of the Immune Responses to Universal Vaccine

Abstract

Growing interest in universal influenza vaccines and novel administration routes has led to the development of alternative serological assays that are able to detect antibodies against conserved epitopes. We present a competitive ELISA method that is able to accurately determine the ratio of serum immunoglobulin G directed against the different domains of the hemagglutinin, the head and the stalk. Human serum samples were treated with two variants of the hemagglutinin protein from the A/California/7/2009 influenza virus. The signals detected were assigned to different groups of antibodies and presented as a ratio between head and stalk domains. A subset of selected sera was also tested by hemagglutination inhibition, single radial hemolysis, microneutralization, and enzyme-linked lectin assays. Pre-vaccination samples from adults showed a quite high presence of anti-stalk antibodies, and the results were substantially in line with those of the classical serological assays. By contrast, pre-vaccination samples from children did not present anti-stalk antibodies, and the majority of the anti-hemagglutinin antibodies that were detected after vaccination were directed against the head domain. The presented approach, when supported by further assays, can be used to assess the presence of specific anti-stalk antibodies and the potential boost of broadly protective antibodies, especially in the case of novel universal influenza vaccine approaches.

Keywords: HA2-antibody; competitive ELISA; hemagglutinin; stalk domain; universal influenza vaccine.

Figure 1

Schematic overview of the competitive ELISA method. (A) ELISA plates were coated with a purified hemagglutinin (HA) recombinant protein from A/California/7/2009 (H1N1) influenza strain. (B) A 1:250 pre-diluted serum sample is treated and incubated with different HA and HA1 concentrations. (C) The resulting OD difference between the highest head domain (HA1)-treated and the HA-treated sample can be attributed to the stalk domain (HA2) response. Two examples of treatment are reported, with appreciable pre-vaccination differences in the HA2 response between adults and children. $\overline{\mathrm{OD}}\mathrm{HA}1$ is the average optical density (OD) of the samples that were incubated with the head protein (HA1); $\overline{\mathrm{OD}}\mathrm{HA}$is the average OD of the samples that were incubated with the HA protein; and $\overline{\mathrm{OD}}\mathrm{NT}$is the average OD of the samples that were incubated with the buffer (non-treated samples).

Figure 2

(A) Serum samples from adult subjects tested by ELISA, hemagglutination inhibition (HI), single radial hemolysis (SRH), micro-neutralization (MN) and enzyme-linked lectin (ELLA) assays; (B) Serum samples from adults tested by ELISA and HI assays. The HI titer of each sample is indicated below the x-axis. Blue bars = head signal, and green bars = stalk signal. Asterisks indicate statistical significance; a black asterisk indicates a significant increase in the HA signal; a blue asterisk indicates a significant increase in the head signal; a green asterisk indicates a significant increase in the stalk signal. Error standard bars are reported both for the head and stalk signals.

Figure 3

(A) Serum samples from children tested by ELISA, HI, SRH, MN and ELLA assays; (B) Serum samples from children tested by ELISA and HI. The HI titer of each sample is indicated below the x-axis. Blue bars = head signal, and green bars = stalk signal. Asterisks indicate statistical significance; a black asterisk indicates a significant increase in the HA signal; a blue asterisk indicates a significant increase in the head signal; a green asterisk indicates a significant increase in the stalk signal. Error standard bars are reported both for head and stalk signal.