Age and vaccine-dependent anti-NA antibody response against A(H1N1) and A(H1N1)pdm09 influenza viruses

Abstract

Antigenic drifts and shifts of hemagglutinin (HA) and neuraminidase (NA) shape influenza virus immunity, yet the current literature shows a knowledge gap on how infection and vaccination may shape our antibody responses for HA and NA over time. During primary exposure to the influenza virus, memory B cells reactive to a range of conserved and non-conserved epitopes found predominantly on HA and NA are generated, and these lead to an enduring “imprint”, a process named “immune imprinting”. The immune response to later infection or vaccination is thus influenced by the antigenic similarity of later strains to a person’s initial exposure. Immune imprinting has been identified as an important driver of the anti-HA antibody immune landscape. Parallel studies on the anti-NA antibody landscape have not been done. This is partly due to the limited knowledge of the antigenic drift of NA. Influenza vaccine strain selection is based on HA antigenic drifts, and anti-NA antibody response is not monitored for seasonal influenza vaccine strain selection. To investigate immune imprinting patterns of anti-NA antibodies, I first characterised the antigenicity of N1 proteins of A(H1N1) influenza viruses from 1977 to 1991 to extend the NA antigenic profile of A(H1N1) and A(H1N1)pdm09 viruses previously reported. I identified that the NA of A/USSR/90/77 (USSR/77), A/Singapore/06/86 (Singapore/86), and A/Texas/36/91 were antigenically distinct, showing dissonance of antigenic drift of NA from that of HA. The amino acid substitution N386K on the globular head domain affected the polyclonal antibody binding, potentially contributing to the drift between USSR/77 and Singapore/86. Then, using a comprehensive panel of antigenically distinct HAs and NAs from 1977-2015, I carried out an age-stratified cross-sectional sero-epidemiology to investigate the immune imprinting patterns of neuraminidase inhibition (NI) antibodies compared to hemagglutinin inhibition (HI) antibodies. The highest NI titers were consistently observed towards the strains circulated during the first decade of their life, similar to HI antibodies, showing evidence of childhood imprinting. However, NI antibodies showed a broader breadth of reactivity compared to HI antibody responses. I investigated the NI antibody response in individuals who received a single dose of standard-dose inactivated influenza vaccine (SIIV) or enhanced inactivated influenza vaccine (eIIV) and in individuals repeatedly vaccinated with homologous or heterologous vaccine combinations over two consecutive influenza seasons (2017/2018 and 2018/2019). In both influenza seasons, MF59 adjuvanted influenza vaccine (A-eIIV) recipients showed a higher magnitude of NI response post-vaccination. Finally, I investigated the back-boosting of NI antibodies with infection and vaccination. I observed that natural infection with A(H1N1)pdm09 virus and vaccination with A-eIIV back-boosted NI antibodies towards antigenically distinct NA of historical seasonal A(H1N1) viruses. These results emphasise the importance of routine monitoring of NA antigenic drift in parallel with HA for seasonal influenza vaccine strain selection. Owing to the slower antigenic drift of NA, standardisation of NA protein content and the use of adjuvants in seasonal influenza vaccines may contribute to NA-mediated protection elicited by influenza vaccines, even when there is a significant HA antigenic drift variant.