IgM memory B cells: origin, mutation, and mobilization by polysaccharide vaccination

Comment on Weller et al, page 3647

Weller and colleagues show that human peripheral blood IgM⁺CD27⁺ B lymphocytes, known as IgM memory B cells, derive from the splenic marginal zone, mutate their immunoglobulin genes in early ontogeny, and respond to polysaccharide antigen, a T-cell–independent antigenic stimulus.

In humans, B lymphocytes expressing cell surface immunoglobulin M (IgM) and CD27 comprise a significant proportion of the B-cell population in the peripheral circulation. These cells have been termed IgM memory B cells because, like conventional class-switched memory cells, they possess somatically hypermutated variable region genes and express CD27.¹  The splenic marginal zone (MZ) has been implicated as a source of circulating IgM memory B cells, as their representation in the peripheral circulation is reduced in asplenic subjects, and their progressive appearance in the peripheral circulation during the first and second year of life correlates with the development of the splenic MZ.²  Weller and colleagues provide further evidence that the IgM memory population can derive from the splenic MZ. First, they show that blood IgM memory B cells phenotypically resemble splenic MZ B cells in being IgM^hiIgD^loCD21^hiCD1c^hi. Second, they use gene expression profiling to show that the set of genes expressed in splenic MZ B cells closely parallels that of IgM memory peripheral B cells and differs from the set of genes expressed in naive and class-switched memory B cells. Third, using CDR-3 spectratyping, they demonstrate that a particular B-cell clone is mobilized in both splenic and blood populations following vaccination with pneumococcal and meningococcal polysaccharides. Interestingly, this clone had mutations prior to vaccination and acquired further mutations after vaccination. This result is consistent with previous studies indicating that MZ B cells participate in responses to T-cell–independent polysaccharide antigens and to encapsulated bacterial pathogens.^2,3  In keeping with the findings of Kreutzmann et al,²  Weller and colleagues observed a diminution of blood IgM memory B cells in asplenic adults. However, unlike Kreutzmann et al,²  they did not observe this diminution in asplenic children, and therefore IgM memory B cells in children can be generated in extrasplenic sites.FIG1 

Weller et al confirm that the immunoglobulin genes of IgM memory B cells are heavily mutated, and importantly, they show that this mutation occurs as early as 12 months of age. Previous work by Weller et al⁴  has shown that hyper IgM patients having defective CD40L lacked germinal centers and did not have class-switched memory B cells. Nonetheless, these patients possessed IgM memory B cells with mutated immunoglobulin variable region genes. Thus, they proposed a second pathway of B-cell hypermutation that occurs in the absence of classical cognate T-cell–B-cell interaction, and they suggested this mutation may occur as part of an intrinsic developmental pathway as is known to occur in sheep.

MZ B cells (and B1 B cells) are rapidly mobilized in response to blood-borne encapsulated bacteria,³  and Kearney and colleagues (Martin et al,³  and Lopes-Carvalho and Kearney⁵ ) have suggested that this represents an innate-like immune response capable of producing IgM antibodies at a time when the conventional adaptive response is still developing. It is intriguing to consider the possibility that in humans, 2 distinct B-cell developmental programs exist: one that relies upon antigen-driven cognate T–B-cell interactions and germinal center formation to generate a mutated, high-affinity adaptive antibody response and memory, and another that uses an antigen-independent hypermutation process to generate a prediversified set of B-cell receptors that mediate rapid innate IgM responses to bacterial and viral pathogens.