Rapid profiling of RSV antibody repertoires from the memory B cells of naturally infected adult donors

Abstract

Respiratory syncytial virus (RSV) causes substantial morbidity and mortality in young children and the elderly. There are currently no licensed RSV vaccines, and passive prophylaxis with the monoclonal antibody palivizumab is restricted to high-risk infants in part due to its modest efficacy. Although it is widely agreed that an effective RSV vaccine will require the induction of a potent neutralizing antibody response against the RSV fusion (F) glycoprotein, little is known about the specificities and functional activities of RSV F-specific antibodies induced by natural infection. Here, we have comprehensively profiled the human antibody response to RSV F by isolating and characterizing 364 RSV F-specific monoclonal antibodies from the memory B cells of three healthy adult donors. In all donors, the antibody response to RSV F is comprised of a broad diversity of clones that target several antigenic sites. Nearly half of the most potent antibodies target a previously undefined site of vulnerability near the apex of the prefusion conformation of RSV F (preF), providing strong support for the development of RSV vaccine candidates that preserve the membrane-distal hemisphere of the preF protein. Additionally, the antibodies targeting this new site display convergent sequence features, thus providing a future means to rapidly detect the presence of these antibodies in human vaccine samples. Many of the antibodies that bind preF-specific surfaces are over 100 times more potent than palivizumab, and several cross-neutralize human metapneumovirus (HMPV). Taken together, the results have implications for the design and evaluation of RSV vaccine candidates and offer new options for passive prophylaxis.

Fig. 1

Anti-RSV F repertoire cloning. (A) RSV F-specific B cell sorting. FACS plots show RSV F reactivity of IgG⁺ and IgA⁺ B cells from three healthy adult donors. B cells in quadrant 2 (Q2) were single cell sorted. (B) Isotype analysis. Index sort plots show the percentage of RSV F-specific B cells that express IgG or IgA. (C) Clonal lineage analysis. Each slice represents one clonal lineage; the size of the slice is proportional to the number of clones in the lineage. The total number of clones is shown in the center of the pie. Lineage numbering can be found in Data file S1. Clonal lineages were assigned based on the following criteria: 1) matching of variable and joining gene segments; 2) identical CDR3 lengths; and 3) >80% homology in CDR3 nucleotide sequences. (D) VH repertoire analysis. VH germline genes were considered to be enriched in RSV repertoires if at least two out of three donors showed >3-fold enrichment over non-RSV-specific repertoires (33). (E) CDRH3 length distribution. (F) Somatic hypermutation in VH (excluding CDRH3). Red bars indicate the average number of nucleotide substitutions. Each clonal lineage is only represented once in (D) and (E). Data for non-RSV reactive IgGs were derived from published sequences obtained by high-throughput sequencing of re-arranged antibody variable gene repertoires from healthy individuals (33).

Fig. 2

Antibody preferences for conformational state and subtype of RSV F are similar across three adult donor repertoires. (A) IgG affinities for preF and postF are plotted for each donor. PreF-specific antibodies are colored pink, preF/postF-cross reactive antibodies are orange, and postF-specific antibodies are white. N.B., non-binder (B) Percentage of antibodies within each donor repertoire that are preF-specific, preF/postF-cross reactive or postF-specific. (C) Percentage of antibodies within each donor repertoire that bind specifically to RSV F derived from subtype A (black), subtype B (white), or both subtypes A and B (grey).

Fig. 3

Antibodies isolated from adult donors recognize six antigenic sites spanning the surface of preF. (A) Structure of RSV preF with one protomer shown as a ribbon colored by antigenic site. (B) Percentage of antibodies targeting each antigenic site within each donor repertoire. The antigenic sites recognized by 215 antibodies with higher than 2 nM affinity for preF were mapped using a combination of antibody binning and patch variant mutational analysis. (C) Percentage of preF-specific antibodies targeting each antigenic site.

Fig. 4

The majority of potent neutralizing antibodies recognize preF-specific sites. (A) Neutralization IC₅₀s for the antibodies isolated from each donor repertoire. Data points are colored based on neutralization potency. Red and blue dotted lines depict motavizumab and D25 IC₅₀s, respectively. N.N., non-neutralizing. (B) Percentage of neutralizing antibodies in each donor repertoire, grouped by potency. (C) Apparent binding affinities for preF and postF are plotted for each antibody and colored according to neutralization potency. (D) Neutralization IC₅₀s are plotted for preF-specific, postF-specific, and cross-reactive antibodies. Red and blue dotted lines depict motavizumab and D25 IC₅₀s, respectively. Black bar depicts median IC₅₀.

Fig. 5

The most potent neutralizing antibodies bind with high affinity to preF and recognize antigenic sites Ø and V. (A) Neutralization IC₅₀ is plotted against apparent preF K_D and colored according to antigenic site. N.N., non-neutralizing; N.B., non-binder. (B) Neutralization IC₅₀ is plotted against apparent postF K_D and colored by antigenic site. (C) Antibodies are grouped according to neutralization potency and colored by antigenic site.