In spite of there being a number of vaccines, influenza remains a significant global cause of morbidity and mortality. Understanding more about natural and vaccine induced immune protection against influenza infection would help to develop better vaccines. Virus specific IgG is a known correlate of protection, but other factors may help to reduce viral load or disease severity, for example IgA. In the current study we measured influenza specific responses in a controlled human infection model using influenza A/California/2009 (H1N1) as the challenge agent. Volunteers were pre-selected with low haemagglutination inhibition (HAI) titres in order to ensure a higher proportion of infection; this allowed us to explore the role of other immune correlates. In spite of HAI being uniformly low, there were variable levels of H1N1 specific IgG and IgA prior to infection. There was also a range of disease severity in volunteers allowing us to compare whether differences in systemic and local H1N1 specific IgG and IgA prior to infection affected disease outcome. H1N1 specific IgG level before challenge did not correlate with protection, probably due to the pre-screening for individuals with low HAI. However, the length of time infectious virus was recovered from the nose was reduced in patients with higher pre-existing H1N1 influenza specific nasal IgA or serum IgA. Therefore, IgA contributes to protection against influenza and should be targeted in vaccines.

Traditionally, immunoglobulin A (IgA) was thought to neutralize virus by forming complexes with viral attachment proteins, blocking attachment of virions to host epithelial cells. Recently we have proposed an intracellular action for dimeric IgA, which is actively transported through epithelial cells by the polymeric immunoglobulin receptor (pIgR), in that it may be able to bind to newly synthesized viral proteins within the cell, preventing viral assembly. To this effect, we have previously demonstrated that IgA monoclonal antibodies against Sendai virus, a parainfluenza virus, colocalize with the viral hemagglutinin-neuraminidase protein within infected epithelial cells and reduce intracellular viral titers. Here we determine whether IgA can interact with influenza virus hemagglutinin (HA) protein within epithelial cells. Polarized monolayers of Madin-Darby canine kidney epithelial cells expressing the pIgR were infected on their apical surfaces with influenza virus A/Puerto Rico/8-Mount Sinai. Polymeric IgA anti-HA, but not IgG anti-HA, delivered to the basolateral surface colocalized with HA protein within the cell by immunofluorescence. Compared with those of controls, viral titers were reduced in the supernatants and cell lysates from monolayers treated with anti-HA IgA but not with anti-HA IgG. Furthermore, the addition of anti-IgA antibodies to supernatants did not interfere with the neutralizing activity of IgA placed in the basal chamber, indicating that IgA was acting within the cell and not in the extracellular medium to interrupt viral replication. Thus, these studies provide additional support for the concept that IgA can inhibit replication of microbial pathogens intracellularly.