Report

A Distinct Signaling Pathway Used by the IgG-Containing B Cell Antigen Receptor

See allHide authors and affiliations

Science  20 Dec 2002:
Vol. 298, Issue 5602, pp. 2392-2395
DOI: 10.1126/science.1076963

Abstract

The immunoglobulin G (IgG)–containing B lymphocyte antigen receptor (IgG-BCR) transmits a signal distinct from that of IgM-BCR or IgD-BCR, although all three use the same signal-transducing component, Igα/Igβ. Here we demonstrate that the inhibitory coreceptor CD22 down-modulates signaling through IgM-BCR and IgD-BCR, but not that through IgG-BCR, because of the IgG cytoplasmic tail, which prevents CD22 phosphorylation. These results suggest that the cytoplasmic tail of IgG specifically enhances IgG-BCR signaling by preventing CD22-mediated signal inhibition. Enhanced signaling through IgG-BCR may be involved in efficient IgG production, which is crucial for immunity to pathogens.

B cells express the membrane-bound form of immunoglobulin (mIg) on the surface as a component of the B cell antigen receptor (BCR) (1,2), and distinct isotypes of mIg are expressed by B cells, depending on their developmental stage. Naı̈ve B cells in the peripheral lymphoid organs express both mIgM and mIgD, whereas memory B cells in spleen and lymph nodes express mostly mIgG. B cells expressing mIgG show an enhanced response to antigen stimulation compared with those expressing mIgM and/or mIgD (3), suggesting that IgG-BCR transmits a signal distinct from IgM-BCR or IgD-BCR. However, all mIg isotypes associate with the common BCR signaling component Igα/Igβ, indicating that all BCRs activate the same signaling pathways (1, 2).

IgM-BCR signaling is negatively regulated by inhibitory coreceptors such as CD22 and CD72 (4–15), and these coreceptors are suggested to set a signaling threshold for ligation of IgM-BCR. However, little is known about whether these coreceptors regulate IgD-BCR and/or IgG-BCR.

K46μmλ, K46δmλ, and K46γ2amλ are transfectants of the B lymphoma line K46 that express mIgM, mIgD, and mIgG, respectively (16, 17). These mIgs of different isotypes contain identical antigen-binding variable (V) regions, specific for the hapten nitrophenol (NP). We first tested whether inhibitory coreceptors could regulate BCR signaling depending on mIg isotypes. Because K46 cells express CD22 but not CD72, the CD22-negative variants K46μv, K46δv, and K46γv were isolated by repeated cell sorting. Subsequently, expression of CD22 or CD72 was reconstituted by transfecting expression plasmids for these molecules (fig. S1). In the transfectants, expression of CD72 reduced antigen-induced phosphorylation of both extracellular signal–regulated kinase 1 and 2 (ERK1 and ERK2), regardless of mIg isotypes (Fig. 1A and fig. S2, A and B). In contrast, CD22 expression reduced ERK phosphorylation in both K46μv and K46δv, but not that in K46γv, suggesting that CD22 fails to down-modulate ERK activation induced by IgG-BCR ligation. The isotype-specific regulation of ERK activity by CD22 was also confirmed by in vitro kinase assay (Fig. 1B). Additionally, antigen-induced Ca2+ mobilization was also regulated by CD22 in the same isotype-dependent manner (Fig. 1C and fig. S2C). Indeed, expression of CD22 reduced BCR-mediated Ca2+ mobilization in both K46μv and K46δv, but not that in K46γv, whereas CD72 reduced the Ca2+ mobilization in K46γv, as well as K46μv and K46δv. Taken together, CD22 but not CD72 regulates BCR signaling in an mIg isotype–specific manner, and signaling through IgG-BCR, but not that through IgM-BCR or IgD-BCR, is resistant to CD22-mediated signal inhibition in K46 cells.

Figure 1

CD22 fails to down-modulate ERK activation and intracellular Ca2+ mobilization induced by ligation of IgG-BCR. (A) Phosphorylation of ERK. Indicated transfectants were treated with 0.2 μg/ml NP–bovine serum albumin (NP-BSA) for the indicated times at 37°C. Cells were subsequently lysed and subjected to Western blot analysis with antibody to phospho-ERK (anti–phospho-ERK). The same blots were reprobed with anti–β-tubulin to ensure equal loading. Representative data from at least three experiments are shown. Two independent lines were examined for each transfectant, but data on only one of them are shown because essentially the same results were obtained for the other lines. Dose-response analysis is shown in fig. S2A. (B) In vitro kinase assay for ERK2. Indicated transfectants were treated with medium alone or with 0.2 μg/ml NP-BSA for the indicated times at 37°C. Cells were lysed, and ERK2 was immunoprecipitated. Half of the immunoprecipitates were subjected to in vitro kinase assay, using myelin basic protein (MBP) as a substrate. Relative ERK2 activity is indicated. The other half of the ERK2 immunoprecipitates were subjected to Western blot analysis for ERK2 to ensure the presence of ERK2 in the immunoprecipitates. Representative data from three experiments are shown. (C) Ca2+ mobilization. K46μv and K46γv transfected with the empty vector (black curve), the vector containing CD22 (red curve), or the vector containing CD72 (blue curve) were loaded with Fluo-4/AM, and intracellular free Ca2+ was measured by FACSCalibur (Becton Dickinson, Franklin Lakes, NJ). Cells were added with 0.2 μg/ml NP-BSA at 30 s (indicated by arrows), and measurement of free Ca2+ was continued for 300 s. Representative data from three experiments are shown.

Both CD22 and CD72 contain immunoreceptor tyrosine-based inhibition motifs (ITIMs) in the cytoplasmic region and negatively regulate IgM-BCR signaling by recruiting Src homology 2 domain–containing tyrosine phosphatase–1 (SHP-1) to their phosphorylated ITIMs upon IgM-BCR ligation (18, 19). Consistent with this, both CD22 and CD72 were phosphorylated (Fig. 2), and these phosphorylated coreceptors coprecipitated with SHP-1 upon ligation of IgM-BCR in K46μv transfectants (Fig. 3). When IgG-BCR was ligated on K46γv transfectants, CD72 was efficiently phosphorylated and coprecipitated with SHP-1. In contrast, CD22 was weakly phosphorylated (Fig. 2), and only a small amount of phosphorylated CD22 was associated with SHP-1 (Fig. 3). This is not due to defects in CD22 in K46γvCD22 cells, because retrovirus-mediated expression of NP-specific IgM-BCR (20) restored BCR ligation–induced CD22 phosphorylation and SHP-1 recruitment (fig. S3). Rather, phosphorylation of CD22, essential for SHP-1–mediated signal inhibition, is specifically prevented upon ligation of IgG-BCR on K46 cells. To determine whether this observation is restricted to K46 cells, we reconstituted expression of NP-specific IgM-BCR and IgG-BCR using retroviral vectors (20) in other B cell lines such as WEHI-279, BAL17, and A20 (fig. S4A)—all of which express endogenous CD22—and primary mouse spleen B cells (fig. S4B). In these cells, both CD22 phosphorylation and SHP-1 recruitment were strongly induced by IgM-BCR ligation, but poorly by IgG-BCR ligation (Fig. 4 and fig. S5), in agreement with the results on K46 cells. Moreover, IgG-BCR ligation induced enhanced ERK phosphorylation compared with that induced by IgM-BCR ligation. These results strongly suggest that ligation of IgG-BCR fails to phosphorylate CD22, thereby silencing CD22-mediated signal inhibition by keeping SHP-1 inactive.

Figure 2

Ligation of IgG-BCR fails to induce phosphorylation of CD22. CD22 transfectants and CD72 transfectants of K46μv, K46δv, and K46γv were stimulated either with the indicated amounts of NP-BSA for 3 min or with 0.2 μg/ml NP-BSA for the indicated times at 37°C. Cells were lysed, and the lysates were immunoprecipitated with either anti-CD22 or anti-CD72. Immunoprecipitates were subjected to Western blot analysis with anti–phospho-tyrosine (4G10). The same blots were reprobed with anti-CD22 or anti-CD72 to ensure equal loading (lower panels). Representative data from at least three experiments are shown.

Figure 3

Ligation of IgG-BCR fails to recruit SHP-1 to CD22. K46μv, K46δv, and K46γv transfected with either empty vector (vector) or vector containing CD22 or CD72 were treated with 0.2 μg/ml NP-BSA for 3 min at 37°C. Cells were lysed, and the lysates were immunoprecipitated with anti–SHP-1. The immunoprecipitates were analyzed by Western blotting with anti–phospho-tyrosine (4G10). The positions for CD22 and CD72 are indicated. The positions are determined by the size and presence of the bands specifically in the CD22 or CD72 transfectants. The blot was reprobed with anti–SHP-1 to ensure the presence of SHP-1 in the immunoprecipitates (lower panels). Representative data from at least three experiments are shown.

Figure 4

CD22 regulates BCR signaling in an isotype-specific manner in the B cell line BAL17 and mouse spleen B cells. (A) NP-specific IgM-BCR, IgG-BCR, IgM/G chimera, or IgG/M chimera was reconstituted on the B cell lines BAL17 with retrovirus vectors. Infectants were stimulated with 0.2 μg/ml NP-BSA for the indicated times at 37°C. (B) Alternatively, NP-specific IgM-BCR or IgG-BCR was reconstituted on lipopolysaccharide-stimulated spleen B cells from C57BL/6 mice with retrovirus vectors and then stimulated with 10 μg/ml NP-BSA for 1 min at 37°C. Cells were lysed, and the indicated molecules were immunoprecipitated. Immunoprecipitates were subjected to Western blot analysis with anti–phospho-tyrosine (4G10). The blots were reprobed with anti-CD22, anti-CD72, or anti–SHP-1 to ensure equal loading. Alternatively, the phosphorylation level of ERK was examined by Western blotting of total cell lysates with anti–phospho-ERK. The same blot was reprobed with anti–β-tubulin to ensure equal loading. Representative data from at least three experiments are shown. Dose-response analysis on BAL17 cells is shown in fig. S5A.

IgG, but not IgM or IgD, contains a long cytoplasmic tail, which is conserved in sequence among IgG subtypes and among species (21). The cytoplasmic tail of IgG is crucial for efficient IgG production (22) and is responsible for the enhanced response of mIgG+ B cells to antigen stimulation (3). To assess the role of the cytoplasmic tail of IgG, we infected BAL17 cells with retrovirus to induce expression of chimeras of IgM and IgG (fig. S4A and fig. S5D) because of efficient retrovirus infection of this cell line. Ligation of the IgG/M-BCR containing the extracellular and transmembrane region of IgG and cytoplasmic region of IgM induced strong phosphorylation of CD22 and marked recruitment of SHP-1, as is the case for IgM-BCR ligation (Fig. 4A), suggesting that the cytoplasmic tail of IgG is responsible for preventing CD22-mediated SHP-1 activation. This is confirmed by the finding that ligation of IgM-BCR containing the cytoplasmic region of IgG (IgM/G-BCR) resulted in weak CD22 phosphorylation and poor SHP-1 recruitment, as is the case for IgG-BCR. IgG-BCR thus appears to be protected from CD22-mediated signal inhibition by containing the conserved cytoplasmic tail.

We have demonstrated here that CD22 negatively regulates signaling through IgM-BCR and IgD-BCR, but not that through IgG-BCR. B cells deficient in CD22 alone exhibit hyperreactivity to ligation of IgM-BCR (11–14), demonstrating that lack of CD22-mediated negative regulation alone can make B cells hyperreactive to antigen stimulation. Thus, the absence of CD22-mediated signal inhibition of IgG-BCR signaling may be involved in the enhanced response of mIgG+ B cells. Remarkably, the conserved cytoplasmic tail of IgG is responsible for both prevention of CD22-mediated signal inhibition and the enhanced response of IgG+ B cells. It is likely that the IgG tail prevents CD22 phosphorylation essential for signal inhibition, thereby causing IgG-BCR to become more excitable to antigen stimulation. However, the distinct function of IgG-BCR may also involve other, yet unknown properties specific to IgG-BCR. CD22 is expressed on activated B cells such as germinal center B cells and memory B cells (23), as well as naı̈ve B cells (4). Hyperresponsiveness of IgG-BCR may thus confer upon mIgG+ B cells a growth advantage over mIgM+mIgD+ or mIgM+ B cells in both activated and memory B cell pools. The growth advantage of mIgG+ B cells may be involved in the efficient switching from IgM to IgG production at the cellular level and the efficient response of mIgG+ memory B cells. However, the improved response of memory B cells may also be attributed to other factors such as their increased affinity to antigens as a result of accumulated somatic mutations of immunoglobulin during the generation of memory B cells (24).

Supporting Online Material

www.sciencemag.org/cgi/content/full/298/5602/2392/DC1

Materials and Methods

Figs. S1 to S5

References and Notes

  • * These authors contributed equally to this work.

  • Present address: Department of Biochemistry and Molecular Immunology, University of Bielefeld, D-33615 Bielefeld, Germany.

  • To whom correspondence should be addressed. E-mail: tsubata.imm{at}mri.tmd.ac.jp

REFERENCES AND NOTES

View Abstract

Navigate This Article