To the editor: Polyreactive antibodies are a major component of the natural antibody repertoire and are capable of binding multiple structurally unrelated antigens (e.g. proteins, peptides, carbohydrates, DNA) [1–4]. About 15% of the B cells in human peripheral blood and up to 50% of the B cells in cord blood make polyreactive antibodies. Sequence analysis revealed that many of these antibodies are germline or near germline in configuration and have low antigen-binding affinities as compared to monoreactive antibodies [4]. The ability of polyreactive antibodies to bind multiple unrelated antigens is thought to be owing to the greater flexibility of their antigen-binding pockets as compared to the more rigid antigen-binding pockets of monoreactive antibodies. Although not as well studied, previously we showed by FACS analysis that receptors on the surface of the B cells that make polyreactive antibodies also can bind structurally unrelated antigens [4]. In the present experiments, using fluorochrome-labelled antigens and ImageStream, we demonstrate pictorially, for the first time, that a single polyreactive B cell can bind simultaneously multiple structurally unrelated antigens. Approximately 1.0 × 106, 8- to 12-week-old, C57BL/6 mouse peritoneal cells in a volume of 100 μl were incubated with one, two or three fluorochrome-labelled antigens along with FITC-conjugated antibody to B-220 (BD Biosciences, San Jose, CA, USA) and the death marker 7-AAD (BD Biosciences) for 30 min at 4°C and then analysed by FACS Calibur (BD, San Jose, CA, USA) or by ImageStream (Amins Corporation, Seattle, WA). Beta-galactosidase (beta-gal) and thyroglobulin (Tg) were conjugated with either R-phycoerythrin (R-PE) or allophycocyanin (APC), and insulin-biotin (Sigma-Aldrich, St Louis, MO, USA) was labelled with streptavidin-FITC, streptavidin-PE or streptavidin-APC. Only live B-220 cells that failed to take up 7-AAD were used to evaluate the binding of fluorochrome-labelled antigens. All experiments were carried out in compliance with institutional guideline and approved by the NIDCR ACUC (Bethesda, MD). FACS analysis revealed that 17.0% of the peritoneal B-220 cells bound both Tg and insulin, 11.0% insulin and beta-gal and 14.1% Tg and beta-gal. The binding of multiple fluorochrome-labelled antigens to individual polyreactive B cells then was pictorially evaluated by ImageStream technology. Approximately 50,000 images were saved using the same samples as in the flow cytometry experiments. Fig. 1A–D show individual B-220 positive cells to which (A) none of the antigens bound; (B) one antigen bounds; (C) two antigens bound; and (D) three antigens bound. Binding of multiple unrelated antigens to individual peritoneal B-220 cells as demonstrated by ImageStream. (A) Cells that failed to bind any of the three antigens (i.e. Tg, beta-gal and insulin); cells that bound (B) one antigen, (C) two antigens and (D) three antigens. Representative pictures from 50,000 images. It has long been thought, based on the clonal selection theory, that the receptors on the surface of antibody-producing B cells bind only their cognate antigen or a closely related antigen. However, with the discovery of polyreactive antibodies and with our previous FACS findings [4], it has become clear that the B cells that make these antibodies can bind to their surface multiple unrelated antigens. The experiments reported here confirm and extend these findings and show pictorially, for the first time, that at least three different and unrelated antigens can simultaneously bind to the same B cell. These findings have broad implications in terms of the role of polyreactive B cells in natural defence against bacterial [5] and viral infections [6]. In addition, these findings support the idea that polyreactive B cells may be involved in the presentation of antigens to T cells and could transport self-antigens to the thymus for initiation and/or maintenance of immunological tolerance [4]. Still unanswered is the question as to which of the many endogenous host antigens stimulate polyreactive B cells to secrete polyreactive antibodies and to keep the level of these antibodies relatively constant over time. Therefore, it is not unreasonable to speculate that polyreactive B cell stimulation and antibody secretion may depend on the number, density and affinity of endogenous host antigens that randomly bind to polyreactive B cell receptors. If this proves to be the case, it would provide new insight into the nature of the antigens that trigger and perpetuate the proliferation of polyreactive B cells in both natural immune defence and in diseases, such as chronic lymphocytic leukaemia in which many of the leukaemic cells produce polyreactive antibodies [7]. This work was supported by the Intramural Research Program of the National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, Maryland. The content of this publication is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. YX, ZZ and ALN designed the experiments. YX performed the experiments. ALN and YX wrote the paper. The authors report no conflict of interest.