Assessment of human tetanus toxoid-specific memory B cell subsets by flow cytometry using t-distributed stochastic neighbour embedding for data analysis reveals multiple subsets (#387)
Generation of antibodies that can prevent and/or eliminate infection underlies the success of most vaccines. Antibody memory is retained by memory B cells (mBC), which express antibody on their cell surface as part of the B cell receptor (BCR). In humans, multiple subsets of mBC exist and can be defined based upon differential surface expression of IgG, IgM and IgD. The contribution of these subsets to vaccine-specific antibody responses is poorly understood. Therefore, we sought to study antigen-specific peripheral blood mBC subsets that expand following vaccination, using Tetanus Toxoid (TT) as a model antigen, and analysis of flow cytometry data by t-distributed stochastic neighbour embedding (t-SNE) to identify phenotypes of TT+ mBC subsets. Ten healthy individuals donated blood before, and at days 7, 14, 21, and 28 after, TT vaccination, and TT-specific mBC were identified by binding of fluorochrome-labelled TT multimers. Seven days post-vaccination, there was a transient expansion of IgG+ mBC subsets that were CD20hiCD27lo and exhibited various degrees of surface binding to TT multimers. At days 14 and 21 post-vaccination, phenotypes of IgG+ mBC subsets changed to express CD20loCD27hi. CD20loCD27hi TT+IgG+ mBC frequencies correlated with levels of plasma TT-specific IgG antibodies. In contrast, frequencies of IgM+/IgD+ TT+ mBC subsets exhibited either no change, or inconsistent changes, following vaccination, suggesting that the majority of TT+ IgM+/IgD+ cells harbour polyreactive BCRs against TT. At day 14 post-vaccination, a population of IgMloIgDhi cells emerged, followed by emergence of IgM+IgD+ and IgMhiIgDloCD27hi subsets at day 21. These data demonstrate that the TT-specific mBC response following vaccination is multilayered and comprised of both IgG+ and IgM+/IgD+ subsets. Further work is warranted to determine clonal relationships between these TT-specific subsets. Data analysis using t-SNE may be of benefit for this.