Different Protein Tyrosine Kinases Are Required for B Cell Antigen Receptor–mediated Activation of Extracellular Signal–Regulated kinase, c-Jun NH₂-terminal Kinase 1, and p38 Mitogen-activated Protein Kinase

The mouse anti–chicken IgM mAb M4 was prepared as previously described (22); PMA, ionomycin, and 4α-phorbol were purchased from Calbiochem, Inc. (San Diego, CA). Thapsigargin was purchased from Sigma Chemical Co. (St. Louis, MO). Polyclonal rabbit antisera against JNK1 (C-17), ERK2 (C-14), ERK1 (C-16), and p38 MAPK (C-20) were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Polyclonal rabbit antiphospho-MAPK (ERK2) was purchased from Promega Corp. (Madison, WI). Purified mouse anti–human JNK1 monoclonal antibody was purchased from PharMingen (San Diego, CA). Myelin basic protein (MBP) was purchased from Upstate Biotechnology, Inc. (Lake Placid, NY). Glutathione S-transferase (GST)-c-jun (5–89) and GST-activating transcription factor (ATF)2 were expressed in Escherichia coli, and the fusion proteins were purified as previously described (23). Protein A–sepharose was obtained from Amersham Pharmacia Biotech, Inc. (Piscataway, NJ). BAPTA-AM [the acetoxymethyl ester of bis-(o-aminophenoxy) ethane-N,N,N′,N′-tetraacetic acid] and Pluronic F-127 were purchased from Molecular Probes, Inc. (Eugene, OR).

The generation of Syk-, Lyn-, and Btk-deficient, and Lyn/Syk double deficient, DT40 cell lines was carried out as previously described (20, 21). DT40 wild-type and PTK-deficient cell lines were cultured in RPMI 1640 medium supplemented with 10% heat-inactivated FCS, 2 mM glutamine, 10 U/ml penicillin, 10 μg/ml streptomycin, 1 mM pyruvate and nonessential amino acids, 50 μM mercaptoethanol, and 1% heat-inactivated chicken serum. The cells were maintained below a density of 3 × 10⁶ cells/ml.

Cells (2.5 × 10⁷ cell per sample) were collected and resuspended to 10⁷ cells/ml in fresh complete RPMI 1640 medium. After incubation at 37°C for 15 min, indo-1/AM (7 μg/ml final concentration) was added, and the cells were incubated for a further 45 min with resuspension every 15 min. The cells were subsequently centrifuged and resuspended in complete RPMI 1640 medium at 2 × 10⁶ cells/ml. The fluorescence intensity of intracellular indo-1 was monitored and analyzed using a FACStar^® plus flow cytometer (Becton Dickinson, Mountain View, CA) as previously described (24).

Unstimulated cells or anti-IgM stimulated cells (5 × 10⁶ cells per sample) were lysed in 100 μl RIPA lysis buffer (50 mM Tris pH 7.4, 150 mM NaCl, 10 mM Na₄P₂O₇, 25 mM sodium β-glycerophosphate, 1 mM EDTA, 1% (wt/vol) NP-40, 0.5% (wt/vol) sodium deoxycholate, 0.1% (wt/ vol) SDS, 1 mM PMSF, 1 mM Na₃VO₄, 10 μM E-64, 1 μg/ml pepstatin, 10 μg/ml aprotinin, 10 μg/ml leupeptin). After incubation on ice for 15 min, cell lysates were centrifuged at 16,000 g for 15 min at 4°C. Lysates were subsequently denatured in an equal volume of 2× SDS sample buffer, resolved by 10% SDS-PAGE and electrotransferred to nitrocellulose membranes in non-SDS-containing transfer buffer (25 mM Tris, 0.2 M glycine, 20% methanol, pH 8.5). Western blotting was performed with the following primary antibodies and accompanying dilutions: antiphospho ERK2 (1:20,000), anti-MAPK (ERK1 or ERK2, 1:2,000), anti-p38 (1:1,000), and mouse monoclonal anti-JNK1 (1:500) followed by 1:2,000 anti–rabbit or anti–mouse horseradish peroxidase–conjugated IgG (Santa Cruz Biotechnology, Inc.) and were developed with an ECL Western blotting kit (Amersham Corp., Arlington Heights, IL).

For JNK1, ERK2, and p38 MAPK in vitro kinase assays, cells (5–10 × 10⁶ cells per sample) were resuspended in complete RPMI 1640 medium to a density of 2 × 10⁶ cells/ml and stimulated with either anti-IgM (10 μg/ml) or PMA (50 ng/ml) with or without ionomycin (250 ng/ml) for the indicated times. Incubations were terminated by addition of 10 vol of ice-cold PBS and were subsequently centrifuged at 500 g for 8 min at 4°C. The supernatants were aspirated and cells were lysed by resuspension in 0.5 ml RIPA lysis buffer. After incubation on ice for 15 min, the cell lysates were centrifuged at 16,000 g for 15 min at 4°C. For JNK1 and ERK2 in vitro kinase assays, supernatants were added to 20 μl packed protein A–sepharose beads and 0.5 μg polyclonal rabbit anti-JNK or anti-ERK2 antibody, followed by overnight incubation at 4°C. For p38 MAPK assays, the lysates were first added to 1 μg polyclonal rabbit anti-p38 MAPK antiserum with an additional 0.1% SDS. After constant mixing by inversion for 3 h or overnight at 4°C, the mixtures were added to 20 μl packed protein A-sepharose beads for another 1 h. The immune complexes were pelleted at 4°C by centrifugation at 16,000 g for 10 min. The beads were washed twice each with (a) 1 ml of RIPA buffer with 0.1 mM Na₃VO₄ and (b) 1 ml of p38 MAPK assay buffer (25 mM Hepes, pH 7.4, 25 mM sodium β-glycerophosphate, 25 mM MgCl₂, 2 mM dithiothreitol, and 0.1 mM Na₃VO₄). The pellets were resuspended to 90 μl p38 MAPK assay buffer with 20 μM ATP containing 10 μCi γ-[³²P]ATP and 50 μg/ml substrate (MBP for ERK2, GST–c-jun for JNK1, or GST-ATF2 for p38 MAPK). After incubation for 20 min at 30°C with mixing every 2 min, the reactions were terminated by adding 45 μl 2× SDS sample buffer. After boiling for 5 min followed by brief centrifugation, samples were resolved by SDS-PAGE (12% for JNK1 and p38 MAPK assays and 14 or 12% for ERK2 assays). The gels were stained for 30 min with 40% methanol, 10% acetic acid, and 0.005% Coomassie brilliant blue G-250, destained for 30 min with 40% methanol, 10% acetic acid, and 5% glycerol, and dried before autoradiography at −70°C for various times.

Since engagement of the BCR complex strongly activates p42 ERK2 in other B cell lines and in normal B cells (25, 26), we tested whether BCR cross-linking in DT40 B cells could also lead to ERK2 activation. In vitro kinase assays were performed with anti-ERK2 immunoprecipitates to assess induction of ERK2 activation. Consistent with previous studies, ERK2 activation was detected shortly after BCR cross-linking, with a maximal fourfold activation compared with unstimulated cells evident at 2 min and sustained for at least 15 min (Fig. 1,A). To determine the PTK requirement for BCR-mediated ERK2 activation, we compared ERK2 activation in wild-type versus mutant DT40 cell lines deficient in Lyn, Syk, or Btk that expressed comparable levels of surface IgM (20, 21). BCR-mediated ERK2 activation was completely abolished in Syk-deficient cells even though ERK2 was activated by PMA to a similar extent compared with wild-type DT40 cells (Fig. 1,A); thus, Syk is required for BCR-mediated ERK2 activation. In contrast, BCR-mediated ERK2 activation was largely maintained in Lyn-deficient cells with a maximal threefold activation at 5 min (Fig. 1,A), indicating that ERK2 can be activated via the BCR in the absence of Lyn. In contrast to either Syk- or Lyn-deficient cell lines, BCR ligation in Btk-deficient cells led to a transient and weak ERK2 activation (less than twofold), suggesting that a Btk-independent ERK2 activation pathway exists and that Btk is required for sustained ERK2 activation (Fig. 1 A).

As an alternative assay for ERK2 activation, we measured the phosphorylation of ERK2 using Western blotting and a phospho-MAPK antiserum, which only detected phosphorylated ERK2 in DT40 cells (Fig. 1,B). Consistent with our in vitro kinase assays, phosphorylated ERK2 was detected in DT40 cells between 2 and 15 min after anti-IgM stimulation (Fig. 1,B). No BCR-mediated phosphorylation of ERK2 was detected in Syk-deficient cells; in Lyn-deficient cells phosphorylation of ERK2 was comparable with wild-type DT40 cells, whereas weak and transient ERK2 phosphorylation was observed in Btk-deficient cells (Fig. 1 B). In summary, BCR-mediated ERK2 activation requires Syk, whereas sustained ERK2 activation requires Btk.

JNKs are activated preferentially by a variety of extracellular stresses, and in most circumstances they are activated through a different pathway from ERKs (10, 13– 15). BCR cross-linking also led to JNK1 activation in DT40 cells as determined by its ability to phosphorylate GST–c-jun. JNK1 was maximally activated (fourfold) between 10 and 15 min and then decreased by 30 min (Fig. 2,A). To test whether BCR-mediated JNK1 activation has the same PTK requirement as BCR-mediated ERK2 activation, we compared the ability of the BCR to activate JNK1 in Syk-, Lyn-, and Btk-deficient DT40 cells. The expression levels of JNK1 were not affected either by the deletion of various PTK genes or anti-IgM stimulation (Fig. 2,B). BCR-mediated JNK1 activation was maintained in Lyn-deficient cells with a maximal fivefold JNK1 activation at 15 min (Fig. 2,A), suggesting that Lyn was not essential for BCR-mediated JNK1 activation. However, in both Syk- and Btk-deficient cells, the anti-IgM–induced JNK1 activation was completely abolished (Fig. 2 A), indicating that both Syk and Btk were essential for JNK1 activation. Thus, BCR-mediated JNK1 activation requires both Syk and Btk but not Lyn.

As a relatively new member of the mammalian MAPKs, p38 MAPK has both a different substrate specificity and different upstream activating kinases (MKKs) and appears to have a distinct transducing pathway and function (9, 10, 16, 17). p38 MAPK activation was examined by the ability of anti-p38 MAPK immunoprecipitates to phosphorylate GST-ATF2. p38 MAPK reached a maximal fourfold activation at 15 min after anti-IgM stimulation, and, distinct from both ERK2 and JNK1 activation, persisted for at least 12 h (Fig. 3,A and data not shown). To investigate whether p38 MAPK has different requirements for upstream PTKs, we examined p38 MAPK activation in various PTK-deficient DT40 cells. There was no significant difference in p38 MAPK expression levels between the parental DT40 cells and various PTK-deficient DT40 cells including the Lyn/Syk double mutant cells (Fig. 3,B). In addition, the recovery of p38 MAPK after immunoprecipitation was also not significantly different in DT40 wild-type and mutant cell lines (data not shown). To our surprise, BCR-mediated p38 MAPK activation in Lyn-deficient cells (sixfold maximal activation), Syk-deficient cells (threefold), and Btk-deficient cells (threefold) was similar to that in wild-type DT40 cells (Fig. 3,A). The fold activations of p38 MAPK in Syk- and Btk-deficient cells were lower in part because these cells had a higher basal level of p38 MAPK activation (Fig. 3,A). Thus, these data suggest that none of these PTKs alone is essential for the BCR to activate p38 MAPK. Previous studies have shown that both Lyn and Syk associate with the BCR and account for most of the initial tyrosine phosphorylation including activation of Btk (21). Our data suggested either that there was another PTK required for p38 MAPK activation, or alternatively that Syk and Lyn had a redundant role in BCR-mediated p38 MAPK activation. To test these possibilities, a Lyn/Syk double deficient DT40 cell line was examined. BCR-mediated p38 MAPK activation was lost in Lyn/Syk double deficient cells (Fig. 3 A), indicating that either Lyn or Syk was sufficient for BCR-mediated p38 MAPK activation.

Anti-IgM stimulation leads to a rapid increase of intracellular calcium [Ca²⁺]_i composed of both Ca²⁺ release from intracellular pools and extracellular calcium influx (27). Previous studies have demonstrated that loss of either Btk or Syk but not Lyn resulted in abrogation of calcium mobilization upon BCR cross-linking of DT40 cells (20, 21). Since the pattern of BCR-mediated JNK1 activation correlated precisely with that of BCR-mediated calcium mobilization, we hypothesized that a calcium signal may modulate BCR-mediated JNK1 activation. Furthermore, we could activate JNK1 only with a combination of both ionomycin and PMA (Fig. 2,A and data not shown), also supporting the idea that a calcium signal was required for BCR-mediated JNK1 activation. To define the requirement for calcium in BCR-mediated JNK1 activation and specifically determine whether intracellular or extracellular calcium was required, cells were treated with either BAPTA-AM, a cell permeable calcium chelator, or EGTA, which chelates extracellular calcium before anti-IgM stimulation. The concentrations of BAPTA-AM and EGTA were titrated to prevent calcium release and extracellular calcium influx, respectively (data not shown). BAPTA-AM pretreatment completely blocked BCR-induced JNK1 activation, whereas EGTA pretreatment did not (Fig. 4,A), suggesting that BCR-mediated JNK1 activation is dependent on intracellular rather than extracellular calcium. We also asked whether a calcium signal alone was sufficient for BCR- mediated JNK1 activation. To directly address this question, we pretreated DT40 mutant cells with thapsigargin, a pharmacological reagent that prevents reuptake of intracellular calcium into storage vesicles and thus mimics the receptor-dependent increase of [Ca²⁺]_i while bypassing phosphatidylinositol 4,5-bisphosphate hydrolysis (28). The concentrations of thapsigargin were carefully titrated to give a similar peak calcium signal as anti-IgM stimulation (data not shown). JNK1 activation after BCR ligation was not reconstituted in either Syk- or Btk-deficient DT40 cells (Fig. 4 B), indicating that a calcium signal alone was not sufficient to activate JNK1. Other BCR-mediated signals, which require Syk and/or Btk, are also needed for JNK1 activation.

Since partial BCR-mediated ERK2 activation was detectable in Btk-deficient DT40 cells (Fig. 1), which are incapable of calcium mobilization after BCR ligation (21), it seemed unlikely that BCR-mediated ERK2 activation was calcium dependent. Indeed, pretreatment of DT40 cells with either BAPTA-AM or EGTA did not block BCR-mediated ERK2 activation (Fig. 4 A), indicating that a BCR-mediated calcium signal was not required for ERK2 activation.

Both Syk- and Btk-deficient DT40 cells maintained the ability to activate p38 MAPK via the BCR, suggesting that BCR-mediated p38 MAPK activation was independent of calcium. Indeed, neither BAPTA-AM nor EGTA pretreatment blocked BCR-mediated p38 MAPK activation, showing that BCR-mediated p38 MAPK activation was independent of calcium (Fig. 4,A). Furthermore, BAPTA-AM alone activated p38 MAPK (Fig. 4,A). Together with the observation that the Btk- and Syk-deficient DT40 cells, two mutants that lose their ability to mobilize calcium upon BCR ligation, also exhibited higher basal levels of p38 MAPK activation (Fig. 3), our results suggest that free calcium may play a negative role in p38 MAPK activation.

PMA treatment of wild-type DT40 cells leads to p38 MAPK and ERK2 activation, whereas both PMA and ionomycin are required for JNK1 activation (Fig. 1–3). Since both BCR cross-linking and PMA treatment lead to PKC activation, it was interesting to examine the role of PKCs in BCR-mediated MAPK signaling.

DT40 cells were pretreated for 24 h with PMA (which leads to depletion of PMA-dependent PKCs), an inactive derivative of PMA, 4α-phorbol, or medium, before stimulation with either anti-IgM or PMA. ERK2, JNK1, or p38 MAPK activation of the cells was then analyzed by in vitro kinase assays. Neither PMA nor 4α-phorbol pretreatment of DT40 cells had a significant effect on the expression levels of JNK1, p38 MAPK, and ERK2 (Figs. 2,B and 3,B and data not shown). BCR ligation did not significantly increase ERK2 activation above basal levels in PKC-depleted cells (Fig. 5,A), suggesting that BCR-mediated ERK2 activation is dependent on PMA-depletable PKCs. JNK1 activation was also strongly decreased in PKC-depleted cells (Fig. 5,B), suggesting that PMA-depletable PKCs are required in addition to a calcium signal (Fig. 4) for BCR-mediated JNK1 activation. The requirement of both a calcium signal and PKC activation is consistent with the observation that only a combination of PMA and ionomycin, but not PMA or ionomycin alone, could lead to significant JNK1 activation in DT40 cells (Fig. 5,B and data not shown). For p38 MAPK, significant BCR-dependent activation (greater than threefold increase) was still observed in PKC-depleted cells (Fig. 5 C), suggesting that BCR-mediated p38 MAPK activation is largely independent of PMA-depletable PKCs. However, our results cannot rule out the participation of other PMA-insensitive PKCs such as the atypical PKCζ in BCR-mediated p38 MAPK activation.

In this report, we show that each of the three major mammalian MAPK family members activated through the BCR requires a different set of upstream protein tyrosine kinases (Figs. 1–3 and 6).

Although it has been reported that ERKs are activated upon anti-IgM stimulation via a Ras/Raf-1/MEK signaling pathway (26, 29), the participation of proximal PTKs in BCR-mediated MAPK activation is not well understood. Our data demonstrate that Syk but not Lyn is absolutely essential for BCR-mediated activation of ERK2 (Fig. 1). Supporting this notion, recent studies using the Lck-negative JCaM1.6 variant of the Jurkat T cell line have shown that Syk enabled the TCR to cause a full activation of ERKs in the absence of the Src family kinase Lck (30). In contrast, BCR-mediated ERK2 activation is only partially dependent on Btk, suggesting that Btk has a regulatory role rather than being an essential component in the ERK2 signaling pathway. This possibility is also supported by the fact that Syk is essential for phosphorylation of PLC-γ2, whereas the loss of Btk only reduces PLC-γ2 phosphorylation (20, 21). Previous studies suggest that BCR-mediated activation of Ras and PKCs may converge at the level of Raf-1 for activation of ERKs (26, 29, 31). In a parallel study (32), we found that BCR-mediated activation of ERK2 was incompletely blocked in PLC-γ2–deficient DT40 cells or in DT40 cells transfected with a dominant negative form of Ras (Ras N17). A complete block of ERK2 activation was seen in PLC-γ2–deficient cells transfected with Ras N17, indicating that both Ras- and PLC-γ2–regulated PKCs are involved in BCR-mediated ERK2 activation. Taken together, our data suggest that Syk is essential for both Ras-dependent activation of ERK2 and the PLC-γ2– and PKC-dependent, Btk-regulated pathway leading to ERK2 activation (Fig. 6). Lyn was reported to play a negative role in ERK activation (33), and Lyn-deficient DT40 cells showed augmented BCR-mediated PKC activation (34), leading to the speculation that Lyn-deficient DT40 cells have augmented ERK2 activation. The fact that we did not detect augmented ERK2 activation in Lyn-deficient cells may reflect different roles of Lyn and/or involvement of different PKCs for ERK activation at different stages of B cell development. Interestingly, a novel PKC isoform, PKCμ, which coprecipitates with both Syk and PLC-γ1/2 (35), showed the same pattern of BCR- mediated activation as did ERK2. PKCμ was activated in wild-type and Lyn-deficient DT40 cells but not in Syk- deficient DT40 cells, and was only partially activated in Btk-deficient cells. Furthermore, PKCμ is both calcium independent and PMA inducible, suggesting that PKCμ may participate directly in the activation of ERK2 upon BCR ligation. A possible role of PKCμ in MAPK signaling would not be surprising since other novel PKC isoforms are also capable of activating ERKs (36). Further studies are required to elucidate the precise role of PKCμ and other PKC isoforms in BCR-mediated ERK2 activation.

Both JNK and p38 MAPK are preferentially activated by inflammatory cytokines and cellular stresses instead of classic mitogenic stimuli (37, 38). The JNK activation pathway is well studied, and Rac1/cdc42, MEKK1, and MKK4/ MKK7 contribute to JNK activation (39–41). Consistent with our finding that BCR-mediated JNK1 activation requires both a calcium signal and PMA-sensitive PKCs, JNK activation in a mouse B cell line also requires both PMA and ionomycin (42). In T lymphocytes, JNK activation requires engagement of both the TCR and CD28 or both PMA and ionomycin in contrast to the full activation of ERK1 and ERK2 by either PMA or TCR engagement alone (43). Our data also reveal that BCR-mediated JNK1 activation is dependent on Syk and Btk, which regulate PLC-γ2, but is not dependent on the Src family kinase Lyn (Fig. 6). Consistent with our data, a recent study showed that Syk but not Src family kinases in cooperation with Rac led to enhanced JNK activation in Jurkat T cells (44). Similarly, both a calcium signal and PKCs seem to account for the requirement of Syk for JNK activation. Since Syk- induced JNK activation is potentiated by anti-CD3, anti-CD28, or PMA, it is likely that the requirement of Syk could be at least partly due to a requirement for calcium (44); the fact that JNK activation is not solely activated by ionomycin or calcineurin, and that calcineurin does not synergize with Rac to enhance JNK activation, suggests that PKCs are the second component involved in JNK activation by Syk (45). We also found that PLC-γ2–deficient cells exhibited impaired JNK1 activation and, interestingly, a dominant negative Rac1 (Rac1 N17) blocked activation of JNK1, suggesting a model where cooperation of PKCs and Rac1 leads to JNK1 activation (32). Evidence from studies of Jurkat cells also supports this model. A recent study revealed that a particular PKC isoform, PKCθ, but not PKCα or PKCε, participated in JNK activation; moreover, specific cooperation between calcineurin and PKCθ converged on Rac leading to potent JNK activation (45).

Btk is of interest since Btk is essential for BCR-mediated JNK1 activation but not for p38 MAPK or ERK2 activation. A study using mast cells from Btk-defective mice also had a similar result: the lack of Btk only led to a loss of FcεRI-mediated JNK1 activation but not ERKs or p38 MAPK activation (46). Recently, a novel MKK kinase, MEKK4, which specifically activates the JNK pathway but not ERKs or p38 MAPK, was identified (47); MEKK4 has a Rac/cdc42 interacting binding motif, a putative pleckstrin homology (PH) domain, and a proline-rich region that could provide the binding site for the Src homology (SH)3 domain. Since Btk has a PH domain, a proline-rich region, and SH2 and SH3 domains (48, 49), and since PH domains can interact with PKCs (50), one attractive model is that the interaction of Btk and MEKK4 defines the specificity of JNK signaling pathway (Fig. 6).

Less is known concerning the p38 MAPK signaling pathway. p38 MAPK can be turned on independently of JNKs since both MKK3 and MKK6 activate p38 MAPK specifically (51, 52); however, the upstream activators of MKK3 and MKK6 are still not fully defined. One potential upstream activator may be a STE20/PAK homologue such as Mst1 (53, 54). It is also not clear at what point p38 MAPK and JNK pathways diverge and what the mechanisms are that lead to specific activation of p38 MAPK versus JNKs. Although our data do not entirely solve the puzzle, they do indicate that in DT40 cells p38 MAPK is activated by BCR ligation via a distinct pathway from JNK1: either Syk or Lyn is sufficient and Btk is not required for BCR-mediated p38 MAPK activation; in contrast, anti-IgM–induced JNK activation requires both Syk and Btk but not Lyn (Fig. 6). BCR-mediated p38 MAPK was only partially affected in Syk-deficient (Fig. 3,A) and PKC-depleted cells (Fig. 5,C), again suggesting that both Syk- and Lyn-dependent signals may contribute to p38 MAPK activation independently (Fig. 6). Just how the two p38 MAPK signaling pathways interact and are regulated remains to be determined.

Activation of MAPK family members may contribute to the induction of programmed cell death or apoptosis. Apoptosis plays an important role in B cell development and the generation of immune responses (55), and DT40 cells undergo apoptosis upon BCR cross-linking. Our studies using the B104 B cell line show that p38 MAPK plays a role in BCR-mediated apoptosis (56) and that activation of JNK1 correlates with induction of cell death (18). In DT40 cell lines, BCR-mediated apoptosis is blocked in both Syk- and Btk-deficient cells (57, 58), which also do not activate JNK1; cells in which JNK1 is activated via the BCR (wild-type and Lyn-deficient DT40 cells) also undergo apoptosis (53); in addition, BAPTA-AM blocks both apoptosis and JNK1 activation but not BCR-mediated ERK2 or p38 MAPK activation (Fig. 4 and data not shown). Thus, activation of JNK1 correlated best with the induction of apoptosis in DT40 cells. We also tested the role of p38 MAPK in apoptosis and found that only a relatively high dose of SB203580 (20 μM), a highly specific and cell-permeable inhibitor of p38 MAPK, completely blocked BCR-mediated apoptosis in DT40 cells (Jiang, A., A. Craxton, and E.A. Clark, unpublished data). Thus, it is not presently clear whether JNK1 alone or in cooperation with p38 MAPK may be needed to activate a death program in DT40 cells. Further work with knockout mice and primary B cells is needed to assess the function of distinct MAPKs in B cell development and activation.

We would like to thank M. Domenowske for preparation of figures, G. Shu and K.E. Draves for technical assistance, Y. Sun for reading the manuscript, and other members of the Clark laboratory for helpful discussions.

This work was supported by National Institutes of Health grants GM-42508 and GM-37905 (to E.A. Clark) and by grants from the Japanese Ministry of Education, Science, Sports and Culture (to T. Kurosaki).

ATF

activating transcription factor

BAPTA-AM

the acetoxymethyl ester of bis-(o-aminophenoxy) ethane-N,N,N′, N′-tetraacetic acid

BCR

B cell antigen receptor

[Ca²⁺]_i

intracellular cytoplasmic free Ca²⁺ concentration

ERK

extracellular signal-regulated kinase

GST

glutathione S-transferase

JNK

c-jun NH₂-terminal kinase

MAPK

mitogen-activated protein kinase

MBP

myelin basic protein

PKC

protein kinase C

PLC-γ2

phospholipase C-γ2

PTK

protein tyrosine kinase