Negative Signaling via FcγRIIB1 in B Cells Blocks Phospholipase Cγ2 Tyrosine Phosphorylation but Not Syk or Lyn Activation

Crosslinking of the B cell antigen receptor surface immunoglobulin induces tyrosine phosphorylation and activation of the Src family and Syk tyrosine protein kinases, tyrosine phosphorylation of phospholipase Cγ2 (PLCγ2) and increases in intracellular second messengers inositol phosphates and Ca2+. These activation events, in conjunction with other pathways, culminate in the induction of B cell proliferation and differentiation. In contrast, co-crosslinking surface Ig with the B cell IgG Fc receptor prevents many of these activation events, including B cell proliferation and differentiation. The precise nature of the negative signal(s) derived from Fc receptors that prevent B cell activation is not known. Here, early activation events were examined in B cells stimulated via the antigen receptor alone or under co-crosslinking conditions. The data indicated a selective block in the tyrosine phosphorylation and activation of PLCγ2 but not in activation of the upstream kinases, Syk and Lyn, under co-crosslinking conditions. We conclude that the negative signal acts directly on PLCγ2 and is consistent with recent studies describing an activation-induced association of a phosphotyrosine phosphatase with tyrosine-phosphorylated B cell Fc receptor.

Stimulation of the B cell antigen receptor, surface immunoglobulin (sIg), 1 initiates a cascade of biochemical events that culminate in proliferation (reviewed in Refs. 1 and 2). B cell proliferation is regulated by several means, among which is the suppressive effect of secreted Ig (3). This effect represents a feedback inhibitory mechanism that limits further antibody secretion. To inhibit B cell proliferation, the suppressive antibody must bind specific antigen and bear an intact Fc domain (3), indicating that Fc␥ receptor (Fc␥RIIB1)-sIg co-crosslinking generates a dominant negative signal that prevents B cell proliferation (4 -6). It has been proposed that blocking of antibody-mediated Fc␥RIIB1-sIg crosslinking by Fc␥RIIB1-binding rheumatoid factors may play an etiological role in autoimmune diseases by disrupting this inhibitory mechanism, thereby facilitating continuous autoantibody production (3).
Studies by Phillips and Parker (4, 5) established a polyclonal model to study this mode of inhibition by using whole anti-Ig antibodies which co-crosslink sIg and Fc␥RIIB1 via their Fab and Fc domains, respectively. Such antibodies inhibit B cell proliferation (5,6), production of phosphatidylinositol-derived second messengers (diacylglycerol and inositol 1,4,5-trisphosphate (IP 3 ); Ref. 7), and increases in intracellular Ca 2ϩ (8). In contrast, F(abЈ) 2 fragments of anti-Ig antibodies are stimulatory toward B cells in regards to all these events. The molecular basis of the Fc␥RIIB1-mediated inhibition of PLC activation and the generation of the intracellular mediators diacylglycerol and IP 3 are poorly understood. Protein-tyrosine kinases (PTKs) of the Src family, namely, p56 blk , p58 lyn , and p59 fyn (9,10), and the unrelated p72 syk kinase (11,12) play an obligatory (13,14) role in B cell activation. These kinases are associated with sIg, and their activity is enhanced upon sIg triggering. PTK activation leads to, among other events, tyrosine phosphorylation and activation of phospholipase C␥2 (PLC␥2; Refs. 15 and 16), which is involved in the production of IP 3 and diacylglycerol second messengers (17,18). B cells deficient in p72 syk expression demonstrated an inability to undergo receptor-mediated PLC␥2 tyrosine phosphorylation, IP 3 formation, and calcium mobilization (19), suggesting that p72 syk mediates its action via tyrosine phosphorylation of PLC␥2.
It is notable that the activation events blocked by Fc␥RIIB1-sIg co-crosslinking, i.e. production of IP 3 and increased intracellular Ca 2ϩ , are directly or indirectly dependent on the prior activation of PLC and that PLC activation is dependent on prior activation of PTKs. This fact raises the possibility that the Fc␥RIIB1-derived negative signal may be due to the abrogation of sIg-mediated PTK activation and/or of tyrosine phosphorylation and activation of PLC␥2. Here, we examined these possibilities in murine ex vivo B cells by immunoprecipitation of the various PTKs or PLC␥2 and in vitro assays of their enzymatic activity or tyrosine phosphorylation status. The data indicated that, under stimulatory conditions using F(abЈ) 2 fragments of anti-Ig, Syk and Lyn were activated and PLC␥2 was tyrosine-phosphorylated. Under inhibitory conditions of Fc␥RIIB1-sIg co-crosslinking, the kinase activity of Syk and Lyn were likewise activated; however, and in contrast to activating conditions, PLC␥2 tyrosine phosphorylation was abrogated. These findings are able to account for previous reports (7,8) of inhibition of PLC activation under Fc␥RIIB1-sIg cocrosslinking conditions and are discussed in the context of current knowledge of lymphocyte antigen receptor-mediated activation events.

MATERIALS AND METHODS
Animals and Reagents-Mice were obtained from The Jackson Laboratory (Bar Harbor, ME) and used at 4 -8 weeks of age. F(abЈ) 2 fragment and whole molecule of rabbit anti-mouse IgG antibody were obtained from Organon Technika, Westchester, PA. Anti-Thy 1.2 monoclonal antibody and guinea pig complement were purchased from Sigma. Immunoprecipitating and immunoblotting antibodies were purchased from Upstate Biotechnology, Inc. (Lake Placid, NY); protein * This work was supported by a grant from the National Science Foundation MCB9317027. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Purification of Murine Splenic B Cells-B cells were derived from murine spleen as described previously (16). Briefly, splenocytes were treated with anti-Thy1.2 monoclonal antibody, and T cells were lysed with the addition of guinea pig complement. The resulting cell population, hereafter referred to as B cells, were Ͼ95% sIg ϩ and Ͻ1% Thy1.2 ϩ , as determined by fluorescence-activated cell sorter analysis (16). The B cells were washed with Hanks' balanced salt solution before use.
Lysis, Immunoprecipitating for Tyrosine-phosphorylated PLC␥2 and p72 syk , and Immunoblotting-All lysis, immunoprecipitation, and immunoblotting procedures were performed as described previously (16). Briefly, B cells (10 ϫ 10 6 cells) were stimulated with 10 g/ml F(abЈ) 2 fragment or whole molecule of rabbit anti-mouse IgG for the indicated times at 37°C and lysed with 500 l of ice-cold modified radioimmune precipitation buffer (100 mM NaC, 50 mM Tris, pH 8.0, 0.5% sodium deoxycholate, 0.1% SDS, 10% glycerol, 1% Nonidet P-40, 1 mM sodium orthovanadate, 10 g/ml each of aprotinin and leupeptin). Total cell lysates were prepared by lysis in 50 l of the same buffer. The postnuclear extracts were collected and used as total cell lysates following addition of SDS sample buffer (60 mM Tris, pH 6.8, 2.3% SDS, 10% glycerol, 0.01% bromphenol blue; final concentrations) or were subjected to immunoprecipitation with 10 l of polyclonal PLC␥2 antisera (16); 10 g of 4G10 (Upstate Biotechnology, Inc.); 10 l rabbit polyclonal anti-Lyn (Upstate Biotechnology, Inc.) or 10 l of rabbit polyclonal anti-Syk antibody (20); 10 l of normal rabbit serum was used as a negative control. We have previously established (16) that the anti-PLC␥2 antisera specifically immunoprecipitates PLC␥2, but not PLC␥1. Furthermore, both immunoprecipitation and immunoblotting were inhibited by inclusion of the immunizing peptide derived from the sequence of PLC␥2 but were not inhibited by the inclusion of an irrelevant peptide (16). Samples were collected by addition of protein Gagarose, washed with lysis buffer, resuspended in SDS sample buffer, and held at 95°C for 5 min. The material was separated by SDS-PAGE, transferred to nitrocellulose, and filters were probed with the antibody of interest (anti-PLC␥2, anti-Syk, 4G10). The immunoblots were developed by enhanced chemiluminescence. For some experiments, membranes were stripped of primary antibody with 100 mM glycine, pH 2.5, 1 M NaCl, washed, and reprobed.
In Vitro Kinase Assay for Syk and Lyn-These experiments were performed as described (19) with minor modifications. Cells were stimulated, lysed with modified RIPA buffer, and the lysates were subjected to immunoprecipitation as described above. The samples were washed three times with lysis buffer and twice with kinase buffer (50 mM Hepes, pH 7.0, 150 mM NaCl, 10 mM MnCl 2 , 10 mM magnesium acetate) before resuspending in 40 l of kinase buffer. 15 Ci of [␥-32 P]ATP was added to the samples and the mixture was incubated at 30°C for 10 min; Lyn assays were supplemented with 2 g of acid-denatured enolase (Sigma). Reactions were terminated with the addition of 1 ml of kinase buffer, followed by centrifugation, removal of supernatant, and resuspending the proteins in 30 l of SDS sample buffer. Samples were heated at 95°C for 5 min, separated by SDS-PAGE, dried, and exposed to film. To quantitate changes in activity, bands corresponding to the autophosphorylated target (Lyn or Syk) or enolase (Lyn assays) were scanned by laser densitometry or cut from the gel and counted by liquid scintillation counting. 2 and Whole Anti-Ig-B cells triggered by F(abЈ) 2 fragments of anti-Ig antibodies undergo tyrosine phosphorylation of multiple proteins (1,2). In order to determine whether inhibition of B cell proliferation by intact anti-Ig results from defective PTK activation, profiles of PTK substrates in total extracts of B cells stimulated with F(abЈ) 2 or whole anti-Ig antibodies were compared (Fig. 1). Both stimuli induced an increase in tyrosine phosphorylation of cellular proteins; the overall profile and time course of tyrosinephosphorylated substrates induced by the two stimuli were similar. Dominant band(s) at 56 -60, 72, and 110 kDa are indicated with arrowheads and may represent activated Src family PTK p56 blk , p58 lyn , and/or p59 fyn (9, 10); p72 syk (11,12) and p110 subunit of phosphatidylinositol 3-kinase (21). Slight reduction in the tyrosine phosphorylation of a series of low molecular weight bands were observed in B cell lysates derived from cells stimulated with whole anti-Ig antibodies. Especially apparent is a band migrating at approximately 35 kDa in lysates derived from lysates of F(abЈ) 2 -stimulated B cells and reduced in lysates of whole anti-Ig antibody-stimulated B cells. Thus, despite the failure of whole anti-Ig antibodies to induce a proliferative response in ex vivo B cells (4,5), both forms of anti-Ig antibodies trigger the activation of cellular PTK(s) that are responsive to sIg ligation.

Cellular Tyrosine Phosphorylation by F(abЈ)
Fc␥RIIB1-sIg Co-crosslinking Reduces Tyrosine Phosphorylation of PLC␥2-The apparently intact activation of PTKs in B cells treated with whole anti-Ig antibodies ( Fig. 1) does not rule out a selective defect in tyrosine phosphorylation of a critical PTK substrate. Indeed, the reduced tyrosine phosphorylation of the 35-kDa band (Fig. 1) indicates that some PTK substrates may be differentially phosphorylated following stimulation with the two forms of anti-Ig. Our previous studies defined PLC␥2 as the major B cell isoform of PLC␥, which responded to PTK activation and which acquired phosphorylated tyrosine residues upon sIg triggering (15,16). PLC␥2 tyrosine phosphorylation was examined after stimulation of B cells with intact or F(abЈ) 2 fragments of anti-Ig by immunoprecipitation of tyrosine-phosphorylated proteins with 4G10 anti-phosphotyrosine followed by anti-PLC␥2 immunoblotting. The results, shown in Fig. 2, demonstrated rapid tyrosine phosphorylation of PLC␥2 following F(abЈ) 2 anti-Ig treatment and no detectable PLC␥2 phosphorylation with intact anti-Ig treatment. This finding indicates PLC␥2 tyrosine phosphorylation is reduced when sIg is co-crosslinked with Fc␥RIIB1.
To confirm this finding, lysates of B cells, unstimulated or stimulated as described above for various times, were subjected to immunoprecipitation with anti-PLC␥2 antibodies and analyzed by 4G10 immunoblotting. The results (Fig. 3, upper panel) demonstrated transient PLC␥2 tyrosine phosphorylation when B cells were stimulated under inhibitory sIg-Fc␥RIIB1 co-crosslinking conditions as compared with activating (F(abЈ) 2 anti-Ig stimulation) conditions. This difference was not due to differing amounts of PLC␥2 in the immunoprecipitations as stripping the above immunoblot and re-probing with anti-PLC␥2 antibodies revealed similar amounts in PLC␥2 immunoprecipitates but not in normal rabbit Ig immunoprecipitates (Fig. 3, lower panel).
Tyrosine Phosphorylation and Activation of p72 syk and p58 lyn following sIg Crosslinking or sIg and Fc␥RIIB1 Co-crosslink- ing-Previous experiments in avian B cells (19) suggest that Syk directly phosphorylates and activates PLC␥2. The reduced tyrosine phosphorylation of PLC␥2 observed under sIg-Fc␥RIIB1 co-crosslinking conditions may therefore be the re-sult of reduced tyrosine phosphorylation and/or activation of Syk. To examine this hypothesis, tyrosine phosphorylation of p72 syk following co-crosslinking of sIg with or without cocrosslinking with Fc␥RIIB1 was studied. Tyrosine phosphorylation status of p72 syk was analyzed by immunoprecipitating p72 syk from stimulated or unstimulated murine splenic B cell lysates using rabbit polyclonal anti-Syk antibody and immunoblotting with 4G10 following SDS-PAGE and membrane transfer. Results (Fig. 4, upper panel) show that p72 syk is tyrosinephosphorylated following stimulation of sIg with F(abЈ) 2 anti-Ig (lanes 2 and 3), as described previously (12). Likewise, no detectable difference in Syk tyrosine phosphorylation was observed following co-crosslinking of sIg with Fc␥RIIB1 using intact anti-Ig (lanes 4 and 5). Lane 1 shows Syk immunoprecipitated from unstimulated cells; lane 6 shows material immunoprecipitated with normal rabbit IgG. The presence of essentially equivalent amounts of p72 syk in the samples was confirmed by stripping and reprobing the immunoblot with anti-Syk antibody (Fig. 4, lower panel).
The activation status of p72 syk following crosslinking sIg or co-crosslinking sIg-Fc␥RIIB1 was studied by measuring Syk autophosphorylation using [␥-32 P]ATP in Syk immunoprecipitates derived from lysates of stimulated or unstimulated murine splenic B cells. Results (Fig. 5) showed that stimulation of B cells with F(abЈ) 2 anti-Ig antibody activated p72 syk (lanes 2 and 3), which was detectable after 2 min of stimulation. Likewise, treatment of B cells with intact anti-Ig to induce cocrosslinking of sIg and Fc␥RIIB1 (lanes 4 and 5) induced p72 syk activation to a similar extent and with similar kinetics. Lane 1 represents the unstimulated sample, and lane 6 shows the FIG. 2. Differential tyrosine phosphorylation of PLC␥2 following sIg crosslinking and co-crosslinking with Fc␥RIIB1. Murine splenic B cells at 10 ϫ 10 6 cells/100 l were stimulated with 10 g/ml of F(abЈ) 2 fragment or whole molecule (WM) of rabbit anti-mouse IgG for the indicated times in minutes, and lysates were immunoprecipitated with 10 g of 4G10 anti-phosphotyrosine antibody. The immunoprecipitates were separated by SDS-PAGE, transferred to nitrocellulose, immunoblotted with anti-PLC␥2 antibody, and developed with horseradish peroxidase-conjugated protein G by chemiluminescence. Immunoblotting of PLC␥2 was blocked by inclusion of immunizing PLC␥2, but not PLC␥1 peptide, as reported previously (Ref. 16  normal rabbit IgG immunoprecipitate control. Thus, p72 syk tyrosine phosphorylation and activation of kinase activity is not distinguishable between stimulation of sIg alone or upon co-crosslinking of sIg-Fc␥RIIB1. Quantitation of the autophosphorylated Syk band from the samples stimulated for 2 min with F(abЈ) 2 anti-Ig revealed a 6-fold increase over unstimulated controls; the corresponding sample of intact anti-Ig-stimulated B cells revealed a 5-fold increase over unstimulated controls.
Activation of the Src family PTKs has been described previously (9,10) and, by analogy with the T cell antigen receptor (22), is likely upstream of Syk activation. The activation of Src family PTKs by F(abЈ) 2 and intact anti-Ig antibodies was assessed using Lyn as a representative B cell Src family PTK. Lyn activity was examined by immunoprecipitation of lysates derived from B cells unstimulated, F(abЈ) 2 -or intact anti-Igstimulated followed by in vitro kinase activity measurements. Results, shown in Fig. 6, demonstrated an equivalent activation of Lyn (as measured by Lyn autophosphorylation or by enolase phosphorylation) by both forms of anti-Ig, consistent with measurements of Syk activation shown above. Quantitation of the two characteristic (9) autophosphorylated Lyn bands (p56 and p58) revealed a 2-fold increase in F(abЈ) 2 anti-Igstimulated cells and a 3-fold increase in intact anti-Ig-stimulated cells; both of these were compared with Lyn activity obtained from unstimulated controls. Likewise, liquid scintillation counting of the enolase band revealed a 2-fold increase in both F(abЈ) 2 and intact anti-Ig-stimulated samples, as compared with enolase phosphorylation in unstimulated controls. DISCUSSION Negative signaling in B cells is important in the maintenance of immune regulation. Co-crosslinking sIg and Fc␥RIIB1 was proposed to represent a negative feedback mechanism by bridging sIg and Fc␥RIIB1 via surface-bound antigen or via anti-idiotype antibodies and thereby preventing continuous production of antibodies (3). Fc receptor function in B cells may therefore be important for immune regulation via antigenantibody complexes by inhibiting B cell proliferation and maturation into antibody secreting cells. Furthermore, rheumatoid factor, an IgM anti-Ig Fc antibody, was been proposed to prevent normal negative signaling by binding the Fc portion of secreted Ig, thereby potentially playing an etiological role in the development of rheumatoid arthritis (3).
Previous studies regarding the biochemical nature of the aborted activation process demonstrated transient formation of IP 3 when B cells are treated under sIg-Fc␥RIIB1 co-crosslinking conditions (7), indicating a short-lived PLC activation. In contrast, B cells stimulated with F(abЈ) 2 fragments of anti-Ig showed a more sustained IP 3 formation, as well as later changes in cell activation (7). Thus, cell cycle progression induced upon sIg crosslinking with F(abЈ) 2 anti-Ig antibodies may be at least in part the result of increased IP 3 and intracellular calcium, two important intracellular second messengers (7,8). The lack of these intracellular mediators under sIg-Fc␥RIIB1 co-crosslinking conditions likely contribute to the observed inhibitory effect. Recently, two reports (23,24) revealed decreased calcium flux upon treating A20 cells, a murine B cell lymphoma, with intact Ig antibodies. Additional experiments indicated that negative signaling required an intact immunoreceptor tyrosine-based Inhibitory motif (ITIM) within Fc␥RIIB1; furthermore, phosphorylation of the tyrosine residues within the ITIM was necessary for negative signaling (23). However, defective activation of A20 under sIg-Fc␥RIIB1 co-crosslinking conditions could not be accounted for by the decreased tyrosine phosphorylation of PLC␥2 and Syk; both these activation events were induced by intact anti-Ig (23,24).
Our data indicated that the activating signal under cocrosslinking conditions is identical to that induced by sIg crosslinking alone, i.e. both Syk and Lyn are activated and the B cells display similar increases in a number of tyrosine-phosphorylated proteins. We observed the induction of a ϳ35-kDa phosphotyrosine-containing protein in cells stimulated with F(abЈ)2 fragments of anti-Ig that appeared to be absent from cells stimulated with intact anti-Ig. The precise nature of this band is not clear; however, preliminary studies established that Ig-␣ is tyrosine-phosphorylated under both conditions of stimulation. We have not examined the status of Ig-␤ under either stimulation condition. However, despite the essentially equivalent activation of upstream PTKs and increased tyrosine-phosphorylated proteins, PLC␥2 tyrosine phosphorylation appeared only transiently (Fig. 3) in samples stimulated under negative signaling conditions. This finding can account for the previous reports that IP 3 production and Ca 2ϩ influx are transient events under co-crosslinking conditions (7,23,24) as well as account for the obstruction of cell cycle progression.
Recently it has been shown that tyrosine 309 in the ITIM motif of Fc␥RIIB1 associates with a phosphotyrosine phosphatase PTP1C (25). PTP1C is an intracellular phosphotyrosine FIG. 5. Activation of p72 syk following crosslinking of sIg and co-crosslinking sIg with Fc␥RIIB1. 10 ϫ 10 6 murine splenic B cells per sample were treated with F(abЈ) 2 fragment or whole molecule of rabbit anti-mouse IgG for the indicated times at 37°C and immunoprecipitated as described in the legend to Fig. 5. The immunoprecipitates were washed in kinase buffer and incubated with [␥-32 P]ATP at 30°C for 10 min. The reactions were stopped by addition of excess kinase buffer, resuspended in SDS sample buffer, separated by SDS-PAGE and subjected to autoradiography at Ϫ70C for 2 h. Immunoblotting analysis (not shown) demonstrated essentially equal amounts of Syk in all lanes. The experiment is representative of three others.
FIG. 6. Activation of p58 lyn following crosslinking of sIg cocrosslinking sIg with Fc␥RIIB1. 5 ϫ 10 6 murine splenic B cells per sample were treated with F(abЈ) 2 fragment or whole molecule of rabbit anti-mouse IgG for 60 s (ϩ) or left unstimulated (Ϫ) and lysates were subjected to immunoprecipitation with 5 g of anti-Lyn polyclonal antisera (Upstate Biotechnology, Inc.). The immunoprecipitates were washed in kinase buffer and incubated with [␥-32 P]ATP and acid-denatured enolase at 30°C for 5 min. The reactions were stopped by addition of SDS sample buffer, separated by SDS-PAGE, and subjected to autoradiography at Ϫ70°C for 2 h. Immunoblotting analysis (not shown) demonstrated essentially equal amounts of Lyn in all lanes. The experiment is representative of two others. phosphatase with two tandem N-terminal SH2 domains (26 -28) and associates with tyrosine-phosphorylated c-Kit (29) and the ␤ chain of interleukin-3 receptor (30) via interaction of their SH2 domains. In B cells, the association and activation of PTP1C with Fc␥RIIB1 was detected upon co-crosslinking sIg with Fc␥RIIB1 and was dependent on the phosphorylation of tyrosine 309 on the ITIM (25). The decreased or abolished tyrosine phosphorylation of PLC␥2 observed upon sIg-Fc␥RIIB1 co-crosslinking and reported here can be accounted for by the phosphatase activity of PTP1C. We therefore propose that upon sIg-Fc␥RIIB1 co-crosslinking, tyrosine 309 in the ITIM of Fc␥RIIB1 is phosphorylated and PTP1C binds to Fc␥RIIB1 via its SH2 domain to phosphorylated tyrosine 309. This binding induces activation of PTP1C phosphatase activity, as reported (25). Activated PTP1C then selectively dephosphorylates PLC␥2, while tyrosine phosphorylation and activation of Syk and Lyn activation are not affected. This model is currently under investigation in our laboratory.