Expression of the immunoregulatory molecule FcRH4 defines a distinctive tissue-based population of memory B cells

Abstract

The FcRH4 transmembrane molecule, a member of the Fc receptor homologue family, can potently inhibit B cell receptor (BCR) signaling. We show that cell surface expression of this immunoregulatory molecule is restricted to a subpopulation of memory B cells, most of which lack the classical CD27 marker for memory B cells in humans. The FcRH4+ and FcRH4- memory B cells have undergone comparable levels of immunoglobulin isotype switching and somatic hypermutation, while neither subpopulation expresses the transcription factors involved in plasma cell differentiation. The FcRH4+ memory cells are morphologically distinctive large lymphocytes that express the CD69, CD80, and CD86 cell activation markers. They are also shown to be poised to secrete high levels of immunoglobulins in response to stimulation with T cell cytokines, but they fail to proliferate in response either to BCR ligation or Staphylococcus aureus stimulation. A heightened expression of the CCR1 and CCR5 chemokine receptors may facilitate their preferential localization in lymphoid tissues near epithelial surfaces. Cell surface FcRH4 expression thus marks a unique population of memory B cells with distinctive morphology, functional capabilities, and tissue localization.

Figure 1.

Specificity analysis of the 2A6 anti-FcRH4 monoclonal antibody. (A) A20-IIA1.6 B cells lacking the FcγRIIB receptor were transiently transfected with expression constructs encoding a FcRH4-GFP fusion protein (filled histogram) or GFP-only control constructs (open histogram). Cells were stained with biotinylated F(ab′)2-fragments of anti-FcRH4 antibodies and streptavidin-PE. Analysis gates were set on the GFP⁺ cell fraction of transfected cells. (B) Lysates from 293T cells transiently transfected with the FcRH-GFP fusion constructs, HA-FcRH fusion constructs or empty vector control constructs were subjected to immunopreciptation with the 2A6 anti-FcRH4 antibody. The immunoprecipitates (top) and whole cell lysates (bottom) were separated via SDS-PAGE and probed by Western blotting with anti-GFP antibodies and anti-HA antibodies, respectively. The appearance of a faint band of immunopreciptated HA-FcRH5 on prolonged exposure of the blot (not shown), suggested weak FcRH5 cross-reactivity but the 2A6 antibody was found to be nonreactive with native FcRH5 bearing cell lines, SUDHL-6 and Namalwa (Table I) (reference 21).

Figure 2.

Analysis of FcRH4 expression on B cell subpopulations in human tonsils. (A) CD19-purified tonsillar B cells were stained with anti-CD38, anti-IgD, and anti-FcRH4. A subpopulation of ∼10% were stained positively for cell surface FcRH4 (M2 gate). M1 gate is the entire population of CD19 positive tonsillar B cells. (B) Analysis of the entire population of tonsillar B cells (M1 gate) and (C) FcRH4⁺ tonsillar B cells (M2 gate) for expression of CD38 and IgD. (D and E) CD19-purified tonsillar B cells were stained for CD38, IgD, CD27, and FcRH4. The gate was set on the CD38⁻/IgD⁻ memory B cell population and analyzed for forward scatter (FSC) versus FcRH4 and CD27 versus FcRH4 expression, respectively. Note the increased forward light scattering of the FcRH4⁺ subpopulation (D) and the discordant FcRH4 and CD27 expression (E). (F) Wright-GIEMSA staining and electron microscopy of FcRH4⁺ (top) and FcRH4⁻ (bottom) memory B cells reveals their striking morphological differences. Bar, 2 μm.

Figure 2.

Analysis of FcRH4 expression on B cell subpopulations in human tonsils. (A) CD19-purified tonsillar B cells were stained with anti-CD38, anti-IgD, and anti-FcRH4. A subpopulation of ∼10% were stained positively for cell surface FcRH4 (M2 gate). M1 gate is the entire population of CD19 positive tonsillar B cells. (B) Analysis of the entire population of tonsillar B cells (M1 gate) and (C) FcRH4⁺ tonsillar B cells (M2 gate) for expression of CD38 and IgD. (D and E) CD19-purified tonsillar B cells were stained for CD38, IgD, CD27, and FcRH4. The gate was set on the CD38⁻/IgD⁻ memory B cell population and analyzed for forward scatter (FSC) versus FcRH4 and CD27 versus FcRH4 expression, respectively. Note the increased forward light scattering of the FcRH4⁺ subpopulation (D) and the discordant FcRH4 and CD27 expression (E). (F) Wright-GIEMSA staining and electron microscopy of FcRH4⁺ (top) and FcRH4⁻ (bottom) memory B cells reveals their striking morphological differences. Bar, 2 μm.

Figure 3.

Profile analysis of cell surface markers expressed by FcRH4⁺ and FcRH4⁻ memory B cells. Purified tonsillar CD19⁺ B cells were stained for CD38, IgD, FcRH4, and the indicated cell surface antigens. The analysis gate was then set on the IgD⁻/CD38⁻ memory B cell population. Staining of FcRH4⁺ cells is indicated by a solid line; staining of FcRH4⁻ cells is indicated by a dashed line. The filled gray histogram indicates background staining with an irrelevant isotype-matched control antibody.

Figure 4.

Quantitative assessment of transcription factor and chemokine receptor mRNA expression by FcRH4⁺ and FcRH4⁻ memory B cells. (A) mRNA derived from FcRH4⁺ and FcRH4⁻ memory B cells, germinal center (GC) B cells and plasma cells (PC) and (B) from FcRH4⁺ and FcRH4⁻ memory B cells was used as template for real-time PCR analysis of the indicated genes. All values are normalized to expression levels of the large subunit of the RNA-polymerase 2. Values represent the mean ± SEM from three independent cDNA preparations from three different tonsil samples, with each PCR performed in duplicate.

Figure 5.

Response of memory B cell subpopulations to cytokine stimulation and BCR ligation. Analysis of the proliferative response and of immunoglobulins secreted by purified FcRH4⁺ (filled bars) and FcRH4⁻ (open bars) memory B cells. (A) Cells were cultured for 40 h with or without the indicated stimuli. After addition of ³H-thymidine for an additional 10 h, the cells were harvested and ³H-thymidine incorporation assessed. (B) Cells were cultured for 96 h before the supernatant of the cells was analyzed for secreted immunoglobulins via capture ELISA. Data shown are the mean ± SD one representative of four or more experiments performed with different tonsil samples. (C) Analysis of IgA, IgG, and IgM secretion by FcRH4⁺ and FcRH4⁻ memory B cells after stimulation. Cells isolated and stimulated as in panel B were plated onto ELISPOT plates coated with isotype-specific antibodies and after incubation for 5 h the number of antibody-secreting cells was assessed. Shown is a representative experiment out of at least three independently performed experiments.