Four Tyrosine Residues in Phospholipase C-γ2, Identified as Btk-dependent Phosphorylation Sites, Are Required for B Cell Antigen Receptor-coupled Calcium Signaling*

Activation of phospholipase C-γ2 (PLCγ2) is the critical step in B cell antigen receptor (BCR)-coupled calcium signaling. Although genetic dissection experiments on B cells have demonstrated that Bruton's tyrosine kinase (Btk) and Syk are required for activating PLCγ2, the exact activation mechanism of PLCγ2 by these kinases has not been established. We identify the tyrosine residues 753, 759, 1197, and 1217 in rat PLCγ2 as Btk-dependent phosphorylation sites by using an in vitro kinase assay. To evaluate the role of these tyrosine residues in phosphorylation-dependent activation of PLCγ2, PLCγ2-deficient DT40 cells were reconstituted with a series of mutant PLCγ2s in which the phenylalanine was substituted for tyrosine. Substitution of all four tyrosine residues almost completely eliminated the BCR-induced PLCγ2 phosphorylation, indicating that these residues include the major phosphorylation sites upon BCR engagement. Cells expressing PLCγ2 with a single substitution exhibited some extent of reduction in calcium mobilization, whereas those expressing quadruple mutant PLCγ2 showed greatly reduced calcium response. These findings indicate that the phosphorylations of the tyrosine residues 753, 759, 1197, and 1217, which have been identified as Btk-dependent phosphorylation sites in vitro, coordinately contribute to BCR-induced activation of PLCγ2.

B cell antigen receptor (BCR) 1 engagement on B-lymphocytes initiates a diverse set of intracellular signaling cascades (1)(2)(3). The earliest known event is the activation of the following three distinct types of protein tyrosine kinases (PTKs): Src-PTK (such as Lyn), Syk, and Bruton's tyrosine kinase (Btk). The activation of these PTKs subsequently leads to the initiation of calcium signaling, i.e. the intracellular calcium mobilization and entry of extracellular calcium that regulate the selectivity of a variety of transcription factors (4,5). In B cells, phospholipase C-␥2 (PLC␥2) plays a dominant role in triggering the BCR-induced calcium signaling by producing the second messenger inositol 1,4,5-trisphosphate (IP 3 ) which binds to IP 3 receptors on the endoplasmic reticulum and subsequently makes the release of the internal store of calcium possible (6 -10).
The activation of PLC␥ isoforms has been shown to correlate with their tyrosine phosphorylation (11)(12)(13)(14)(15). Several studies have indicated that Btk and Syk are the most likely candidates for the PTKs that directly phosphorylate PLC␥2 in B cells (16 -21). An experiment using genetic dissection of Syk in DT40 chicken B-lymphoma cells (16) demonstrated that the BCRinduced tyrosine phosphorylation of PLC␥2, as well as IP 3 production and calcium mobilization, is completely eliminated in these cells, which suggests that Syk plays a critical role in PLC␥2 phosphorylation and activation. Btk-deficient DT40 cells (17) also exhibit a reduced level (almost three times lower than that of wild-type DT40 cells) of PLC␥2 phosphorylation. The recent discovery of BLNK (B cell linker protein (19,22), alternatively SLP-65 (23) or BASH (24)) has clarified the molecular mechanism whereby Btk and Syk cooperate to phosphorylate and activate PLC␥2. The proposed molecular connection (25) is that, after BCR cross-linking, Syk is activated and subsequently phosphorylates BLNK, which allows for the association of Btk with BLNK (21,26) as well as for the association of PLC␥2 with BLNK (19,27,28) via their respective SH2 domains. As Syk also colocalizes on phosphorylated BLNK (19,24), BLNK can nucleate an activation complex including Btk, Syk, and PLC␥2, for which PLC␥2 seems to be phosphorylated and subsequently activated by Btk and Syk.
The next question to be clarified is how PLC␥2 is activated through its tyrosine phosphorylation by these two PTKs. Although the number of molecules identified as the substrate of Btk is very limited (29), heterogeneous transfection experiments have demonstrated that coexpression of Btk and PLC␥2 in NIH3T3 cells or 293T cells results in the tyrosine phosphorylation of PLC␥2 (20,21), which indicates that PLC␥2 is the direct substrate of Btk. Although the reduction of PLC␥2 phosphorylation in Btk-deficient B cells is relatively small as compared with that in Syk-deficient B cells, the former also show almost complete loss of IP 3 production and calcium mobilization (16,17). These findings suggest that the Btk-dependent PLC␥2 phosphorylation is critical for initiating calcium signaling in B cells and that the residual PLC␥2 phosphorylation in Btk-deficient cells, which is presumably caused by Syk or other PTK(s), is not sufficient for PLC␥2 activation. In the study presented here, we investigated the tyrosine phosphorylation of PLC␥2 by Btk. Determination of the Btk-dependent phosphorylation sites in PLC␥2 enabled us to evaluate the role of these tyrosine residues in BCR-induced PLC␥2 activation, which results in the initiation of calcium signaling.
Expression Constructs and Transfection-Wild-type human Btk and kinase-inactive human Btk cDNA (K430R, lysine 430 to arginine substitution, Btk(KϪ)) in the pEF-BOS vector were described previously (29), as were T7-tagged rat PLC␥2 expression vectors (27). Mutations in PLC␥2 (substitution of phenylalanine for tyrosine) constructs were created by means of polymerase chain reaction-based mutagenesis and inserted into the pApuro vector. DNA transfection into 293T cells has been described elsewhere (29). cDNAs of T7-tagged rat wild-type or mutant PLC␥2 were transfected into DT40 cells by electroporation at 260 V, 960 microfarads and selected in the presence of 0.5 g/ml puromycin.
Purification of GST Protein and in Vitro Kinase Assay-The cDNA fragments of rat wild-type or mutant PLC␥2 were in-frame inserted into pGEX-5X-3 vectors (Amersham Pharmacia Biotech). GST expression vectors were transfected into NM522 strain. GST protein was purified with glutathione-Sepharose 4B (Amersham Pharmacia Biotech), according to the manufacturer's protocol, and used as the substrate for the in vitro kinase assay. Immunoprecipitates with anti-Btk mAb 48-2H from transfected 293T cell lysates were washed twice with kinase buffer (20 mM PIPES, pH 7.0, 20 mM MnCl 2 ). 50 l of kinase buffer, 1 g of indicated GST proteins, and [␥-32 P]ATP (10 Ci) were added to each sample. The reaction was performed at 30°C for 10 min and then terminated by adding the sample buffer and heating at 100°C for 5 min. Autophosphorylation of Btk and transphosphorylation of GST proteins were detected with autoradiography (29).
Calcium Measurement-Cells were suspended in phosphate-buffered saline containing 20 mM Hepes, pH 7.2, 5 mM glucose, 0.025% bovine serum albumin, and 1 mM CaCl 2 and loaded with 3 M Fura-2AM at 37°C for 45 min. Cells were washed twice and adjusted to 1 ϫ 10 6 cells/ml. Emission at 510 nm was monitored under excitation of the cell suspension at two different wavelengths (340 and 380 nm) with a fluorescence spectrophotometer (model RF-1500; Shimadzu, Kyoto, Japan) at 37°C. Calibration and calculation of calcium levels were performed as described elsewhere (33).
In Vivo IP 3 Production Analysis-Cells (2 ϫ 10 6 ) suspended in 50 l of culture medium were stimulated with mAb M4 (2 g/50 l) at 37°C for the indicated times. The reaction was stopped by adding 50 l of ice-cold 15% (w/v) trichloroacetic acid. After centrifugation, superna-tants were washed four times with 1 ml of water-saturated diethyl ether. Kinetic analysis of IP 3 production from cells (2 ϫ 10 6 ) was performed with the BIOTRAK IP 3 assay system (Amersham Pharmacia Biotech) according to the manufacturer's protocol.

Btk Phosphorylates Four Distinct Tyrosine Residues in PLC␥2 in
Vitro-To identify the sites of Btk-mediated tyrosine phosphorylation in PLC␥2, GST fusion protein series that covers the whole sequence of rat PLC␥2 (Fig. 1A) was constructed. As the source of purified Btk protein, Btk cDNA was transfected into 293T cells, and expressed Btk protein was immunoprecipitated with anti-Btk mAb. In vitro kinase assay was then carried out with each of the GST-PLC␥2 fusion proteins as the potential substrate ( Fig. 1B shows representative cases). Prominent transphosphorylation of the GST fusion proteins by Btk was observed only when the GST fusion proteins carrying the residues in the linker region between the SH2 and SH3 domains (the minimum sequence is 639 -770) or in the carboxylterminal region (the minimum sequences are 1141-1212 and 1200 -1244) of PLC␥2 were used for the assay (Fig. 1A). Because phosphorylation of the GST fusion proteins was not observed when kinase-inactive Btk (Btk(KϪ)) was used for the in vitro kinase assay (Fig. 2, 2nd lane), phosphorylation did not depend on the activities of other kinases that might be contaminated in the immunoprecipitate. The sites of tyrosine phosphorylation in PLC␥2 were further determined by using GST constructs with phenylalanine substituted for tyrosine. The sequence 639 -770 in PLC␥2 contains 11 tyrosine residues. The sequence in this region of PLC␥2 exhibits a certain homology with that of PLC␥1 (14, 15,55). Based on the previous finding that the tyrosine phosphorylation sites in PLC␥1 are residues 771 and 783 (34), we decided to mutate the tyrosine residues 753 and 759 in PLC␥2. The GST fusion protein with a double mutant, Y753F/Y759F, in which tyrosine residues 753 and 759 were simultaneously changed to phenylalanine, was also examined. The two sequences in the carboxyl-terminal region (1141-1212 and 1200 -1244) of PLC␥2 have no homology with that of PLC␥1 (14, 15). The sequence 1141-1212 contains three tyrosine residues (1153, 1171, and 1197), whereas only one tyrosine residue (1217) is found in the sequence 1200 -1244. These tyrosine residues were sequentially substituted by phenylalanine. As shown in Fig. 2A, the Y753F substitution in GST-PLC␥2-(639 -770) greatly reduced the phosphorylation when compared with that of the non-substituted GST fusion protein, whereas the extent of phosphorylation in the GST protein carrying the Y759F substitution was hardly distinguishable from that in the non-substituted protein. However, in contrast to the observation that phosphorylation in the Y753F mutant was still detectable, the phosphorylation was completely eliminated when the Y753F/Y759F double mutant was subjected to the same assay. These findings indicated that both the tyrosine residues 753 and 759 are Btk-dependent phosphorylation sites. With the same procedure, the residues 1197 in the GST-PLC␥2-(1141-1212) protein (Fig. 2B) and 1217 in the GST-PLC␥2-(1200 -1244) protein (Fig. 2C) were identified as Btk-dependent phosphorylation sites. The confirmed sites of Btk-dependent phosphorylation in PLC␥2 are thus tyrosine residues 753 and 759, which fall in the linker region between the SH2 and SH3 domains, and 1197 and 1217, which are located in the carboxyl-terminal region (Fig. 3). These four tyrosine residues in rat PLC␥2 are also conserved in human PLC␥2.
Btk-dependent Phosphorylation of PLC␥2 in Reconstituted Cells-We next examined the Btk-dependent phosphorylation of PLC␥2 in reconstituted cells. Wild-type or a series of mutant PLC␥2s tagged with the T7 sequence ( Fig. 3) were expressed with Btk in 293T cells (Fig. 4). The quadruple mutant, Y-4F, in which tyrosine residues 753, 759, 1197, and 1217 were simultaneously substituted by phenylalanine, was also examined. Cell lysates were immunoprecipitated with anti-T7 mAb, and the tyrosine phosphorylation of each type of PLC␥2 was evaluated. As shown in Fig. 4, PLC␥2 with Y753F, Y1197F, or Y1217F substitution exhibited a lesser extent of tyrosine phosphorylation than did wild-type PLC␥2. Strikingly, the Btk-dependent phosphorylation of the Y-4F mutant was greatly reduced despite a similar level of protein expression. Although the phosphorylation level of single mutant PLC␥2 in the reconstituted cells exhibited a different pattern from that in DT40 cells (see below), these results indicated that the major sites of Btk-dependent phosphorylation in PLC␥2 as seen in the in vitro kinase assay and in the coexpression systems were identical.
BCR-induced Phosphorylation Sites in PLC␥2 Are Involved in the Four Tyrosine Residues-To evaluate the contribution of these four tyrosine residues (753, 759, 1197, and 1217) to BCR-mediated signaling, wild-type or mutant PLC␥2 cDNAs were stably transfected into PLC␥2-deficient DT40 cells. The expression level of mutant PLC␥2 was approximately the same as that of the transfectant with wild-type PLC␥2, which was about three times higher than that of the endogenous chicken PLC␥2 in wild-type DT40 cells (Fig. 5A). The in vitro PLC assay for wild-type and mutant PLC␥2 resulted in similar levels of IP 3 production (Fig. 5B), indicating that these mutants did not significantly affect the in vitro lipid hydrolyzing activity of PLC␥2 in resting cells. The cell surface expressions of BCR on these cells measured by flow cytometric analysis were essentially the same levels as that on the parental PLC␥2-deficient DT40 cells (data not shown). The global patterns of BCRinduced tyrosine phosphorylation of these transfectants, which were assessed by anti-phosphotyrosine blot of the whole cell lysate, were also indistinguishable (data not shown). By using these transfectant cells, the contribution of the four tyrosine residues to BCR-induced PLC␥2 phosphorylation was evaluated (Fig. 6A, top panel, arrow). Although the extent of BCRinduced tyrosine phosphorylation in single mutants (Y753F, Y759F, Y1197F, or Y1217F) was indistinguishable from that in wild-type PLC␥2, the double mutant, Y753F/Y759F, exhibited a significant loss of PLC␥2 phosphorylation, whereas the quadruple mutant, Y-4F, showed almost complete loss of BCR-induced PLC␥2 phosphorylation. Although the phosphorylation of Tyr-759 in vitro ( Fig. 2A) and in reconstituted cells (Fig.  4) appeared to be weak, Tyr-759 proved to be phosphorylated in DT40 cells upon BCR stimulation because Y753F/Y759F showed a significant reduction in PLC␥2 phosphorylation as compared with Y753F. These results indicated that the major sites of BCR-induced PLC␥2 phosphorylation are included in the four tyrosine residues, 753, 759, 1197, and 1217.
Since it has been demonstrated that the association of PLC␥2 and phosphorylated BLNK is one of the prerequisites for PLC␥2 activation upon BCR engagement (27,28), the association of mutant PLC␥2 and BLNK was evaluated by assessing the coprecipitation of BLNK with PLC␥2. As shown in Fig.  6A, the association of BLNK with wild-type or mutant PLC␥2 remained unaltered, indicating that it was not affected by the tyrosine phosphorylation of PLC␥2. 3 Production-It has been convincingly demonstrated that the activation of PLC␥2, which generates IP 3 , is essential for both the intracellular calcium flux and the entry of extracellular calcium following BCR engagement (8). As biochemical studies have indicated that the activation of PLC␥ isoforms mainly depends on its tyrosine phosphorylation (11,12), we next evaluated the contribution of the phosphorylation of the four tyrosine residues to PLC␥2 activation. PLC␥2-deficient DT40 cells that ectopically expressed wild-type or mutant PLC␥2 were stimulated with anti-chicken IgM mAb (M4) followed by evaluation of the calcium mobilization (Fig. 6B). Compared with that of the ectopic expression of the wild-type PLC␥2, expressions of the single mutants (Y753F, Y759F, Y1197F, or Y1217F) resulted in, respectively, 55, 66, 32, or 30% of the calcium mobilization at peak level. Furthermore, the expression of the double mutant Y753F/Y759F resulted in 45% of the calcium mobilization and that of the quadruple mutant Y-4F in only 16% of the cells expressing the wild-type PLC␥2. PLC␥2 activity as measured by calcium mobilization was thus greatly diminished by substituting the four tyrosine residues, which indicated that the phosphorylations of the four tyrosine residues coordinately activate PLC␥2 in B cells.

Phosphorylation of the Four Tyrosine Residues in PLC␥2 Is Required for Calcium Mobilization and IP
The effect of phenylalanine mutations on the activity of PLC␥2 was also evaluated by measuring in vivo and in vitro IP 3 production upon BCR stimulation. The amounts of in vivo IP 3 production by the cells transfected with the quadruple mutant were greatly reduced as compared with that produced by the cells transfected with wild-type PLC␥2 (Fig. 6C). The BCRinduced enhancement of PLC␥2 activity of the quadruple mutant Y-4F measured by in vitro IP 3 production was also reduced as compared with that of wild-type PLC␥2 (Fig. 6D), although the reduction was less severe than those seen in calcium mobilization and IP 3 production in vivo. The similar reduction of in vitro phospholipase activity was also observed with the stimulation of 0.5 mM pervanadate (35,36).
Finally, the quadruple mutant Y-4F was expressed in wildtype DT40 cells, and the BCR-induced calcium mobilization was compared with that in parental cells. The expression of Y-4F PLC␥2 was approximately five times greater than that of the endogenous PLC␥2 in wild-type DT40 cells (Fig. 7A). As shown in Fig. 7B, expression of Y-4F resulted in a 35% reduction in calcium mobilization and demonstrated that the expression of Y-4F in DT40 cells has a dominant-negative effect on BCR-induced calcium mobilization.

DISCUSSION
Antigen receptor-mediated calcium signaling is mainly triggered by the activation of PLC␥ isoforms (13-15). B cells predominantly express PLC␥2 (7), whereas T cells mainly express

FIG. 5. Expression of mutant PLC␥2 in PLC␥2-deficient DT40 cells (A) and PIP 2 hydrolysis activity of mutant PLC␥2 in vitro (B).
A, PLC␥2-deficient DT40 cells were transfected with pApuro vector (mock) or the indicated mutant PLC␥2 constructs. Whole cell lysates (WCL) prepared from these transfected cells (1st to 8th lanes) and wild-type DT40 cells (DT40 WT, 9th lane) were separated on SDS-PAGE gels and immunoblotted with anti-T7 mAb (top panel) or the anti-rat PLC␥2 Ab (2nd panel). Expression levels of PLC␥2 were measured with a densitometer. B, PIP 2 hydrolysis activity of mutant PLC␥2 in these cells was measured by means of an in vitro phospholipase activity assay. Open bars indicate the relative in vitro PLC activity of mutant PLC␥2 with the value for WT PLC␥2 set at 100. PLC␥1 (37). In B cells, the critical role of BCR-coupled cytoplasmic PTKs, especially Btk, in PLC␥2 activation has been corroborated by several biochemical and biological findings. Consistent with the observation that Btk-deficient DT40 cells exhibit a complete loss of BCR-coupled calcium mobilization (17), B cells from X-linked agammaglobulinemia patients (20) or from X-linked immunodeficiency mice (38) were also found to show significant reductions of BCR-coupled calcium signaling. Recent biochemical studies have demonstrated that PLC␥2 is one of the major substrates of Btk (20,21) and that the connection of Btk and PLC␥2 is mediated by BLNK (21). In addition, the similarity of the B cell defect seen in Btk- (39,40), PLC␥2-(9, 10), and BLNK (41-44)-deficient mice also suggests that the calcium signaling mediated by PLC␥2 is one of the major downstream steps in the Btk-signaling pathway via BLNK. However, despite the importance of PLC␥2 for BCR signaling and B cell development, its exact activation mechanism has remained unclear. Our current finding that mutations of the four tyrosine residues in PLC␥2 virtually eliminated BCR-induced calcium signaling complements previous observations regarding PTK-deficient B cells (16, 17), indicating that the tyrosine phosphorylation by PTKs such as Btk plays a major role in the activation of PLC␥2 in B cells.
As PLC␥1 and PLC␥2 share common structural modules and their amino acid sequences are ϳ50% identical (13), it is reasonable to postulate that the activation mechanism of the two isoforms is similar. It has been reported that stimulation of platelet-derived growth factor receptor (PDGF) (34) or T cell receptor (45,46) triggers the phosphorylation of PLC␥1 in at least three tyrosine residues (tyrosine 771, 783, and 1254). A study performed with PDGF-responsive NIH3T3 cells (34) indicated that substituting phenylalanine for tyrosine 783 (Y783F) completely blocked the activation of PLC␥1 as measured by IP 3 production, whereas Y1254F only partially inhibited this activation. Conversely, the third substitution Y771F increased the PDGF-dependent PLC␥1 activation. This observation indicated that multiple tyrosine residues in PLC␥1 are phosphorylated following the activation of tyrosine kinases and that each of them has a different impact on PLC␥1 activation. In our study, we demonstrated that tyrosines 753, 759, 1197, and 1217 in PLC␥2 are involved in the BCR-induced phosphorylation and activation of PLC␥2. Tyr-753 and Tyr-759 in PLC␥2, which fall in the linker region between the SH2 and SH3 domains, appear to be homologous to Tyr-771 and Tyr-783 in PLC␥1 (Fig. 8). The location of Tyr-1197 and Tyr-1217 of PLC␥2 in the carboxyl-terminal regions is similar to that of Tyr-1254 of PLC␥1, although the amino acid sequences in these regions exhibit no homology between PLC␥1 and -␥2. Despite this similarity in localization, the observed contributions of these tyrosine residues to the enzymatic activation of PLC␥1 and -␥2 appear to be different. Our results indicate that the phosphorylation of all four tyrosine residues is necessary for PLC␥2 activation and that the tyrosine phosphorylation in the linker region between the SH2 and SH3 domains appears to be less effective than that in the carboxyl-terminal region. It may thus be necessary to confirm this observed difference in the effects of phenylalanine mutations by means of other experimental systems and to clarify this discrepancy.
Despite several research projects conducted in this area, it is still unclear how tyrosine phosphorylation affects the activation of PLC␥1 and -␥2. The phosphorylation of these tyrosine residues may enhance the PLC enzymatic activity by altering the intramolecular conformation (47), recognition of the PIP 2 substrate (48), or protein/protein interaction (49). Our experiments indicate that the tyrosine to phenylalanine mutations reduce the enzymatic activity of PLC␥2 without the change of association with BLNK (Fig. 6A) or of the overall pattern of the molecules coprecipitating with PLC␥2 (data not shown). This observation seems to support the idea that the phosphorylation of the four tyrosine residues enhances the PLC␥2 activity, at least in part, by altering the intramolecular conformation or the substrate recognition. In addition to the phosphorylation of critical tyrosine residues, recent studies have suggested the presence of an activation mechanism independent of its tyrosine phosphorylation (13). It has been reported that phospholipids such as phosphatidylinositol 3,4,5-trisphosphate contribute to the PLC␥ activation in some cell systems (50 -52). The cells transfected with the quadruple mutant Y-4F, which displayed almost complete loss of BCR-induced tyrosine phosphorylation, still show minor calcium mobilization and IP 3 production. This residual PLC␥2 activation seen in the Y-4F transfectant may reflect a tyrosine phosphorylation-independent activation.
It should be noted that the reduction in BCR-induced calcium flux (Fig. 6B) and in vivo IP 3 production (Fig. 6C) of Y-4F transfectant was more pronounced as compared with that in in vitro PLC␥2 activity (Fig. 6D). This observation may indicate that BCR stimulation not only induces the phosphorylation of PLC␥2 but also alters the interaction of PLC␥2 with unidentified co-factor(s) that affect the PLC␥2 activity in vivo and is missing in immunoprecipitates used for in vitro PLC assay (48).
An important question is which PTKs are directly involved in the phosphorylation of PLC␥2 in B cells. Our findings that Btk phosphorylates Tyr-753, Tyr-759, Tyr-1197, and Tyr-1217 in PLC␥2 in vitro and that the same residues are involved in BCR-induced PLC␥2 phosphorylation suggest that Btk is one of the PTKs that phosphorylate these residues upon BCR engagement. However, in contrast to the previously reported finding (17) that Btk-deficient DT40 cells exhibit a relatively small reduction of BCR-induced PLC␥2 phosphorylation (only three times lower than that of wild-type cells), the quadruple mutant Y-4F used by us exhibited almost complete loss of BCR-induced tyrosine phosphorylation. This suggests that, in addition to Btk, other PTK(s) such as Syk are also involved in the phosphorylation of these four tyrosine residues in PLC␥2. Although it has been demonstrated that Btk and Syk cooperate to activate PLC␥2 (16,17), no report has precisely characterized the Syk-dependent PLC␥2 phosphorylation. The observation that Syk-deficient DT40 cells exhibit a complete loss of PLC␥2 phosphorylation (16) does not necessarily mean that Syk directly phosphorylates PLC␥2. Previous studies have demonstrated that Btk-dependent PLC␥2 phosphorylation also requires the activity of Syk because the phosphorylation of BLNK by Syk is necessary for providing docking sites for both Btk (21,26) and PLC␥2 (19,27). Furthermore, the kinase activity of Btk to some extent depends on Syk (53,54). It is currently not known whether the residual PLC␥2 phosphorylation in Btk-deficient cells directly depends on Syk or, alternatively, on unidentified PTK(s) that require Syk for their activation. However, we observed that Syk phosphorylates to a minor extent the GST fusion proteins that contain the linker region between the SH2 and SH3 domains but not those that contain the carboxylterminal region (data not shown). It is therefore possible that Syk is another PTK that phosphorylates Tyr-753 and/or Tyr-759 in PLC␥2. Nevertheless, the complete loss of calcium mobilization and IP 3 production seen in Btk-deficient DT40 cells (17) indicates that PLC␥2 phosphorylation by other PTKs such as Syk, even if it phosphorylates some of the four tyrosine residues, is not sufficient to trigger the activation of PLC␥2.