A New B Cell With a New B-Cell Function

The colony-stimulating factors, granulocyte/macrophage colony-stimulating factor (GM-CSF), macrophage colony-stimulating factor (M-CSF), and granulocyte colonystimulating factor (G-CSF), were originally defined as hematopoietic cell growth factors. Subsequently, CSFs have been found to have complex, pleiotropic effects in inflammation and in other states.¹  In most murine models of inflammation and autoimmunity (e.g., experimental autoimmune encephalomyelitis, nephritis, and arthritis), CSF depletion results in the suppression of the disease state. These findings, which are consistent with a proinflammatory function of CSFs, have led to clinical trials of anti-GM-CSF and anti-M-CSF antibodies in rheumatoid arthritis.

Prior to the present study by Rauch et al. from the laboratory of Filip Swirski, GM-CSF was thought to be produced by non-hematopoietic cells, macrophages, or T cells. However, the authors examined murine GM-CSF expression by flow cytometry and made the surprising observation that in response to lipopolysaccharide (LPS), GM-CSF was found mainly in a splenic B-cell subpopulation. Consistent with this interpretation, immunofluorescence of spleen sections from LPS recipients co-localized GM-CSF expression to red pulp cells that coexpressed the B-cell markers, IgM, B220, and PAX5. Subsequent extensive characterization by flow cytometry revealed that GM-CSF expressing cells were IgM^high, CD43^high, CD93⁺, CD138⁺, VLA4^high, LFA1^high, CD284⁺, CD23^low, IgD^low, and CD21^low. In addition to GM-CSF, the cells under investigation secreted IgM and IL-3, but not pro–IL-1β, IL-6, or tumor necrosis factor–α. This set of phenotypic markers defined these cells as a unique mature B-cell population. The authors named the cells “innate response activator B cells” (IRA B cells) because of the known role of GM-CSF in activating innate leukocytes.

After maturation, B lymphocytes either recirculate through secondary lymphoid organs as part of a long-lived pool (follicular cells) or join more static compartments in the marginal zone of the spleen (MZ B cells) or peritoneal and pleural cavities (B1 B cells).²  The authors performed transcriptional analysis on isolated IRA B cells and using hierarchical clustering and principal component analysis found that IRA B cells fit into a population separate from transitional, follicular, marginal zone, B1, or plasma cells. Adoptive transfer and parabiosis experiments demonstrated that peritoneal B1 cells give rise to IRA B cells (Figure), of which B1a cells were the dominant precursor. B1a cells in culture produced multiple cell types, including IRA B cells. MZ B cells use the adhesive ligands VLA-4 and LFA-1 to remain lodged in the spleen. Similarly, IRA B cells, which express VLA-4 and LFA-1, are dislodged from the spleen upon injection of anti-VLA-4 and anti-LFA-1 antibodies. These and other experiments described by the authors are consistent with a model in which IRA B cells are the progeny of peritoneal B1a B cells and migrate to and take up residence in the spleen.

To assess the functional significance of IRA B cells, the authors employed a murine cecal ligation and puncture (CLP) sepsis model. They produced mixed chimeras by reconstituting lethally irradiated mice with bone marrow from a B-cell-deficient mouse called μMT and from a GM-CSF– deficient (Csf2^–/–) mouse. The B cells of these mice (called GM/μMT chimeras) cannot produce GM-CSF. The mortality of GM/μMT mice was significantly greater than that of normal mice in the CLP model. Additionally, the peritoneal cavities of GM/μMT chimeras had more neutrophils, and the mice developed an IL-1β, IL-6, and TNFα cytokine storm. This cytokine profile is associated with a defect in bacterial clearance. Neutrophils from the GM/μMT chimeras also phagocytosed bacteria poorly and had higher levels of bacteremia.