Although traditional egg-based inactivated influenza vaccines can protect against infection, there have been significant efforts to develop improved formats to overcome disadvantages of this platform. Here, we have assessed human CD4 T cell responses to a traditional egg-based influenza vaccine with recently available cell-derived vaccines and recombinant baculovirus-derived vaccines. Adults were administered either egg-derived Fluzone ® , mammalian cell-derived Flucelvax ® or recombinant HA (Flublok ® ). CD4 T cell responses to each HA protein were assessed by cytokine EliSpot and intracellular staining assays. The specificity and magnitude of antibody responses were quantified by ELISA and HAI assays. By all criteria, Flublok vaccine exhibited superior performance in eliciting both CD4 T cell responses and HA-specific antibody responses, whether measured by mean response magnitude or percent of responders. Although the mechanism(s) underlying this advantage is not yet clear, it is likely that both qualitative and quantitative features of the vaccines impact the response.

Currently, licensed influenza virus vaccines are designed and tested only for their ability to elicit hemagglutinin (HA)-reactive, neutralizing antibodies. Despite this, the purification process in vaccine manufacturing often does not completely remove other virion components. In the studies reported here, we have examined the viral protein composition of a panel of licensed vaccines from different manufacturers and licensed in different years. Using western blotting, we found that, beyond HA proteins, there are detectable quantities of neuraminidase (NA), nucleoprotein (NP), and matrix proteins (M1) from both influenza A and influenza B viruses in the vaccines but that the composition differed by source and method of vaccine preparation. We also found that disparities in viral protein composition were associated with distinct patterns of elicited antibody specificities. Strikingly, our studies also revealed that many viral proteins contained in the vaccine form heterologous complexes. When H1 proteins were isolated by immunoprecipitation, NA (N1), M1 (M1-A), H3, and HA-B proteins were co-isolated with the H1. Further biochemical studies suggest that these interactions persist for at least 4 h at 37°C and that the membrane/intracytoplasmic domains in the intact HA proteins are important for the intermolecular interactions detected. These studies indicate that, if such interactions persist after vaccines reach the draining lymph node, both dendritic cells and HA-specific B cells may take up multiple viral proteins simultaneously. Whether these interactions are beneficial or harmful to the developing immune response will depend on the functional potential of the elicited virus-specific CD4 T cells.

Antibodies specific for the hemagglutinin (HA) protein of influenza virus are critical for protective immunity to infection. Our studies show that CD4 T cells specific for epitopes derived from HA are the most effective in providing help for the HA-specific B cell responses to infection and vaccination. In this study, we asked whether HA epitopes recognized by CD4 T cells in the primary response to infection are equally distributed across the HA protein or if certain segments are enriched in CD4 T cell epitopes. Mice that collectively expressed eight alternative MHC (Major Histocompatibility Complex) class II molecules, that would each have different peptide binding specificities, were infected with an H1N1 influenza virus. CD4 T cell peptide epitope specificities were identified by cytokine EliSpots. These studies revealed that the HA-specific CD4 T cell epitopes cluster in two distinct regions of HA and that some segments of HA are completely devoid of CD4 T cell epitopes. When located on the HA structure, it appears that the regions that most poorly recruit CD4 T cells are sequestered within the interior of the HA trimer, perhaps inaccessible to the proteolytic machinery inside the endosomal compartments of antigen presenting cells.

The most effective measure to induce protection from influenza is vaccination. Thus, yearly vaccination is recommended, which, together with infections, establishes diverse repertoires of B cells, antibodies and T cells. Here, we examined the impact of this accumulated immunity on human responses of adults to split, subunit and recombinant protein-based influenza vaccines. ELISA assays, used to quantify serum antibodies and peptide-stimulated CD4 T cell cytokine ELISpots, revealed that pre-existing levels of HA-specific antibodies were negatively associated with gains in antibody post-vaccination, while pre-existing levels of CD4 T cells were negatively correlated with vaccine-induced expansion of CD4 T cells. These patterns were seen independently of the vaccine formulation administered and the subjects’ influenza vaccine history. Thus, although memory CD4 T cells and serum antibodies consist of components that can enhance vaccine responses, on balance, the accumulated immunity specific for influenza A H1 and H3 proteins is associated with diminished future responses.

The first exposures to influenza have been shown to bias the immune system towards a certain strain which is then boosted by subsequent infection and vaccination throughout life. The impact of repeated stimulation of hemagglutinin (HA) epitopes of the CD4 T cell repertoire is unknown. We are interested in the CD4 T cells specific to conserved regions of HA as the conserved stalk region is a target for new vaccine designs and antibodies to HA are a correlate of protection from infection. We are also interested in how CD4 T cell phenotypes after repeated antigen exposure from conserved HA epitopes could change the CD4 T cell repertoire. In order to probe these questions, we analyzed PBMCs collected from healthy human donors using cytokine ELISpots and flow cytometry. We created overlapping peptide pools containing conserved or unique regions of HA, to compare differences in cytokine secretion, abundance, and phenotypes of CD4 T cells elicited to different regions of HA. Our hypothesis is that CD4 T cells specific to repeatedly stimulated epitopes will have a higher frequency and different phenotypes compared to unique epitopes. We have found that there is a robust circulating CD4 T cell response to influenza antigens. The reactivity to conserved regions of HA varies between individuals as well as between strains of influenza in healthy human donors and in some donors, there are CD4 T cells that are biased towards conserved regions of HA. Those CD4 T cells to conserved regions of the stalk could be recalled after vaccination with stalk-based vaccine strategies. We plan to further examine the phenotypes of those CD4 T cells and to determine if decade of birth and vaccination history will impact the abundance of CD4 T cells to conserved HA regions.