Structural Requirements for Broadly Protecting Antibodies to Influenza A &B

Marasco, Wayne

Abstract

Influenza remains a major medical problem and is a constant threat to human health. Since their first administration in the 1940s, influenza vaccines have provided global seasonal protection against influenza infections. However, because of influenza's propensity to continuously change its hemagglutinin (HA) protein by a process known as """"""""antigenic drift"""""""", and for pandemic strains by """"""""antigenic shift"""""""", current vaccines require near-annual reformulation and accurate prediction of circulating strains for the upcoming season. The newly discovered epitopes in influenza's hemagglutinin (HA) stem region to which rare broadly neutralizing antibodies (BnAbs) are directed has led to renewed optimism that the development of a """"""""universal'influenza vaccine that elicits these types of BnAbs will be possible. However, recent attempts to produce such a vaccine have shown only limited success. We now recognize from molecular studies of B cell responses during the 2009 H1N1 pandemic in which a high frequency of BnAbs was seen, that the limitation in eliciting protective levels of BnAbs is not due to an inherent lack of antigenicity, but rather to an immunodominant Ab response to the HA globular, a result of repeated annual memory B cell expansion to seasonal influenza vaccines which outcompete stem-directed memory B cells in numbers and for resources. Our preliminary data point to an effective approach to selecting or eliciting BnAbs, involving the rational engineering of glycosylation sites on HA that """"""""redirect"""""""" Abs to the stem region via """"""""glycan shielding"""""""" of head epitopes and unmasking of stem epitopes.
In aim 1, will we propose to investigate the broad applicability of glycan shielding and deshielding as a strategy to select broadly reactive stem-binding Abs (BrAbs) using our circa 50 billion member human-Ab phage display library. The recovered BrAbs will be tested for virologic and immune clearance properties (e.g ADCC, CDC).
In aim 2, we will investigate whether HA-glycan variants can enhance the isolation of stem- directed Br/BnAbs by FACS sorting of memory B cells/plasmablasts from influenza-experienced donors. We will narrow the glycan variants by comparing Ab germline gene useage, somatic mutations, relative numbers and diversity of Br/BnAbs recovered from aim 1/2 wt HA vs glycan variant screens.
In aim 3, we will perform structural studies to map the epitopes of the most promising Br/BnAbs. The rationale here is that under- standing the epitope structure/location and the contribution of germline vs somatic amino acid substitutions will provide important insights, at the atomic level, into what type(s) of protective, stem-directed BrAbs we want to elicit by vaccination.
In aim 4, we will test our lead HA-glycan variant(s) through in vivo vaccine studies in humanized &Balb/c mice, for their ability to induce Br/BnAb responses that are protective against heterosubtypic virus challenge, by preferentially stimulating the expansion of pre-existing stem-directed human memory B cells or priming of naive mouse B cells in vivo to secrete protective Br/BnAbs. These glycan variants should provide """"""""lead antigens"""""""" for the design of recombinant HA-based universal influenza vaccines.

Public Health Relevance

A universal vaccine that induces broad immunity against multiple subtypes of influenza viruses has been a long sought but elusive goal in medical research. Newly discovered broadly protective human antibodies that are directed to the stem region of influenza's hemagglutinin (HA) protein has led to the recognition that the development of a universal influenza vaccine may be possible. This proposal is focused on one promising strategy to develop a universal influenza vaccine.