Human CD43+ B cells are closely related not only to memory B cells phenotypically but also to plasmablasts developmentally in healthy individuals

Abstract

CD20(+)CD27(+)CD43(+) B (CD43(+) B) cells have been newly defined among PBMCs and proposed to be human B1 cells. However, it is controversial as to whether they are orthologs of murine B1 cells and how they are related to other B-cell populations, particularly CD20(+)CD27(+)CD43(-) memory B cells and CD20(low)CD27(high)CD43(high) plasmablasts. Our objective is to identify phenotypically the position of CD43(+) B cells among peripheral B-lineage cell compartments in healthy donors, with reference to B-cell subsets from patients with systemic lupus erythematosus (SLE). We found that CD43(+) B cells among PBMCs from healthy subjects were indistinguishable phenotypically from memory B cells in terms of surface markers, and spontaneous in vitro Ig and IL-10 secretion capability, but quite different from plasmablasts. However, a moderate correlation was found in the frequency of CD43(+) B cells with that of plasmablasts in healthy donors but not in SLE patients. An in vitro differentiation experiment indicated that CD43(+) B cells give rise to plasmablasts more efficiently than do memory B cells, suggesting that they are more closely related to plasmablasts developmentally than are memory B cells, which is also supported by quantitative PCR analysis of mRNA expression of B-cell and plasma cell signature genes. Thus, we conclude that, in healthy individuals, CD43(+) B cells are closely related not only to memory B cells phenotypically but also to plasmablasts developmentally, although the developmental origin of CD43(+) B cells is not necessarily the same as that of plasmablasts.

Keywords: ILT3/LILRB4; auto-antibody; natural antibody; plasmablast/plasma cell.