Shifting immunodominance of humoral immunity against influenza viruses

PROJECT SUMMARY Influenza viruses remain a major global threat to human health, which is exacerbated by the lack of an effective vaccine. Humoral immunity does not target all parts of the virus evenly, a phenomenon known as immunodominance. Following seasonal vaccination, humans preferentially generate antibody responses against evolving epitopes of the hemagglutinin (HA) head domain rather than against broadly protective epitopes of the conserved stalk domain. In the absence of preexisting immunity against the HA head, as occurred with the 2009 pandemic H1N1 virus, humans can preferentially recall memory B cells against the stalk domain. However, naïve B cells against novel epitopes of the HA head are induced, populate the memory B cell pool, and are preferentially recalled following exposure to an antigenically similar virus. Therefore, immunodominance of the head domain and the preferential recruitment of naïve B cells against new epitopes remains a major obstacle for the generation of a broadly protective influenza vaccine. In order to generate broadly protective humoral immunity, a vaccine needs to be designed that induces robust and durable antibody responses against the stalk domain but prevents the recruitment of naïve B cells against neoepitopes. Immunodominance is in part dictated by which B cells can acquire antigen and efficiently present antigen to CD4 T cells. B cell avidity, the simultaneous binding of both binding-sites of an antibody with its epitope and the cross- linking of multiple B cell receptors on a B cells, plays a critical role in which B cell specificities are selected. B cells against the head domain have an avidity advantage, as these epitopes are more accessible than those of the stalk domain. We hypothesize that reducing B cell avidity for head epitopes will reduce competition for antigen, limit the induction of anti-head B cells, and preferentially select for B cells against the stalk domain. Using combinatorial mutagenesis and yeast-display, we will generate and select for a library of stable HAs with diverse head epitopes but an identical stalk domain, which we refer to as scrambled HA (Aim 1). By titrating the diversity of scrambled HAs, we will determine the effect of increasing head epitope diversity on which B and T follicular helper cell specificities are induced. We will perform these experiments in naïve and H1N1 immune mice to mimic immune histories in infants and adults, respectively (Aim 2). Moreover, we will immunize human ex vivo lymph node organoids generated from individuals of diverse immune histories to determine if our approach recalls memory B cells against the stalk domain (Aim 3). Together, this study will generate a vaccine that shifts immunodominance towards the stalk domain, which will be proven using multiple models of preexisting immunity. Although this proposal focuses on shifting immunodominance of anti-influenza humoral immunity, the approach taken provides a proof-of-concept that can be applied to other rapidly evolving pathogens for which immunodominance is a key barrier for successful vaccine generation, including HIV and coronaviruses.