We seek to answer a series of questions about the human immune response to influenza virus (and to rapidly evolving pathogens generally) and to integrate the answers into strategies for improved influenza vaccines.  Long-range goals are vaccination strategies that confer lasting immunity to influenza virus in the face of rapid antigenic variation of currently circulating subtypes and of potential introduction of new subtypes into the human population. Our premise is that analysis of B-cell ontogeny following influenza-virus vaccination or infection can facilitate design of immunogens to elicit a broadly neutralizing response. We will integrate results from experimental immunizations of rhesus macaques (RMs) and mice with data from human studies.


Hypotheses: Imprinting causes primary responses to influenza virus to bias later responses to antigenically drifted variants; secondary responses update B-cell memory by affinity maturation to the drifted strains. What are the dynamics of imprinting and updating when driven by immunization with HA (rather than by infection)?  Imprinting by vaccination may differ from imprinting by infection (with relevance to vaccination policy). 


Studies and experiments: Analyze rearranged Ig gene sequences of single B-cells in primary responses and in homologous and heterologous recall responses elicited by immunization with HA trimer antigens. Map the response across the HA molecular surface with a structurally calibrated reference panel of Fabs covering the HA surface with overlapping footprints.


Hypothesis. For a given epitope, development of breadth during a secondary response depends on the
differences between primary and challenge immunogens. We can quantify those differences, as seen by the immune system of the host in question, by the relative frequencies of recall and new memory mapped by position on the HA surface ("epitopic distance"). How distant (at a particular epitope) must two strains be for new memory to appear (rather than recall and further affinity maturation)?

Studies and experiments: Immunize RMs and mice with HA of one strain and challenge with HA of a second (later) strain from the same subtype or of a strain from a different subtype. Characterize the effects of increased epitopic distance on clonal selection and affinity maturation. Follow the infant human subjects in successive years, as they receive annual MIV immunizations, and analyze changes in B-cell memory.


Hypothesis. Systematic, controlled studies of imprinting and epitopic distance, coupled with structural

analyses, can lead to practical design strategies for vaccine immunogens.


Studies and experiments: (a) Antigen modification. Carry out systematic tests, in human JH6 knock-in mice, of three antigen modification approaches: (i) removing unwanted epitopes ("head only constructs"); (ii) masking unwanted epitopes by altered glycosylation; (iii) modifying epitopes by replacing surface structures on one HA with their homologs on another. (b) B-cell lineage-based immunogen design. We will examine independently the two components of this concept (use a primary immunogen with high affinity for the "desired" germline response; boost with immunogens designed to "coax" affinity maturation along a desired pathway). Test the validity of each of these notions in animal models for immunization, starting with a lineage of RBS-directed mouse antibodies we have already studied.