INTRODUCTION

The aggregation of the FcRI¹ on basophils or mast cells initiates a cascade of biochemical events that results in degranulation and the release of inflammatory mediators (1, 2). Tyrosine phosphorylation of proteins plays a critical role in this signal transduction pathway (3-8). Among the signal transduction molecules that are tyrosine-phosphorylated are the  and  subunits of the receptor, Lyn, Syk, phospholipase C-1 and -2, Vav, Btk, and the focal adhesion kinase FAK (9-15). One of the earliest events after aggregation of FcRI is the activation of protein-tyrosine kinases, probably Lyn or another Src family kinase, that results in the tyrosine phosphorylation of the  and  subunits of the receptor (9, 16). The protein-tyrosine kinase Syk is then recruited by the tyrosine-phosphorylated receptor and is critical for the downstream activation signals (11, 17-19). For example, FcRI aggregation in a Syk-deficient mast cell line does not mobilize Ca²⁺ from intracellular and extracellular sources and fails to propagate downstream signaling events (20).

Previously we observed that FcRI aggregation results in the tyrosine phosphorylation of ~115-kDa proteins in the rat basophilic leukemia RBL-2H3 mast cell line (21, 22). Two of these proteins were identified as FAK and the cell surface adhesion molecule CD31 (15, 23). Recently Pyk2 was identified as another member of the FAK family of protein-tyrosine kinases (24-26). Pyk2 (also called RAFTK for related adhesion focal tyrosine kinase, CAK for cell adhesion kinase , CADTK and FAK2) is a cytoplasmic protein-tyrosine kinase that, like FAK, lacks a transmembrane region, myristoylation sites, and Src homology 2 and 3 domains. Both FAK and Pyk2 have a central kinase region flanked by large N-terminal and C-terminal domains. Pyk2 is expressed in neuronal cells, CD34⁺ bone marrow cells, primary bone marrow megakaryocytes, platelets, and T and B cells (24, 26, 27).

The stimulation of many different cell surface receptors results in the tyrosine phosphorylation and activation of Pyk2 (24, 26, 28-31). These stimuli include carbachol acting through nicotinic acetycholine receptors, stress signals, membrane depolarization, cytokines, and molecules that activate G-protein-coupled receptors. Recently, Pyk2 was found to be tyrosine-phosphorylated after integrin or immune receptor activation (27, 32-35). Pyk2 is also activated by addition of the calcium ionophore and PMA, suggesting that the activation of Pyk2 is downstream of the increase in intracellular calcium and the activation of protein kinase C (24, 36).

Here we report that Pyk2 was present in RBL-2H3 cells. Stimulation of the cells with different stimuli induced the tyrosine phosphorylation of Pyk2. In a Syk-deficient cell line, G-protein-coupled receptors, but not FcRI aggregation, still induced tyrosine phosphorylation of Pyk2. Therefore, there are Syk-dependent and -independent pathways for Pyk2 activation.

EXPERIMENTAL PROCEDURES

Mouse monoclonal anti-Pyk2 was from Transduction Laboratories (Lexington, KY). The horseradish peroxidase-conjugated anti-phosphotyrosine monoclonal antibody, 4G10, was from Upstate Biotechnology Inc. (Lake Placid, NY). Affinity-purified rabbit anti-mouse Igs were obtained from Jackson ImmunoResearch (West Grove, PA). All other antibodies have been described previously (11). The materials for electrophoresis were purchased from Novex (San Diego, CA), and the source of other materials was as described previously (11). N⁶-2-(4-Aminophenyl)ethyladenosine (APNEA) was from Research Biochemicals International (Natick, MA), and fibronectin was from Calbiochem (La Jolla, CA).

RBL-2H3, the Syk-negative variant TB1A2, and the Syk-transfected cells were maintained as monolayer cultures (20, 37). Cells were stimulated either with antigen after overnight culture in the presence of antigen-specific IgE, or with different stimuli essentially as described previously (3). The concentration of the stimuli were: 30 ng/ml dinitrophenyl coupled to human serum albumin, 0.5 µM calcium ionophore A23187, 40 nM PMA, 1 unit/ml human plasma thrombin, or 10 µM APNEA. After stimulation for the indicated times, the medium was removed for histamine analysis. The monolayers were then rinsed with ice-cold phosphate-buffered saline containing protease inhibitors and sodium orthovanadate with the same concentrations as in the lysis buffer described below and solubilized in lysis buffer (50 mM Tris, pH 7.4, containing 1% Triton X-100, 1 mM Na₃VO₄, 150 mM NaCl, 5 µg/ml leupeptin, 45 milliunits/ml aprotinin, 1 µg/ml pepstatin A, 1 mM phenylmethylsulfonyl fluoride). The cells were scraped, and supernatants were collected after centrifugation for 30 min at 16,000 × g, 4 °C. In experiments to deplete extracellular Ca²⁺, the monolayers were washed twice either with Ca²⁺-containing (1.8 mM CaCl₂) or Ca²⁺-free (1 mM EGTA) medium 199. The cells were then stimulated either in Ca²⁺-containing (1.8 mM CaCl₂) or Ca²⁺-free (10 µM EGTA) medium. Stimulation of nonadherent and fibronectin-adherent cells was done as described previously (22).

Lysates were precleared by mixing for 1 h at 4 °C with protein A-agarose beads. The lysates were then incubated with 3 µg of mouse IgG or anti-Pyk2 antibody that had been preincubated with 10 µg of rabbit anti-mouse Ig and 25 µl of protein A-agarose beads. After gentle rotation at 4 °C for 1 h, the beads were washed three times with 1 ml of wash buffer (cell lysis buffer with detergent concentration decreased by 50%), once with 150 mM NaCl, 50 mM Tris, pH 7.4, and the proteins eluted by boiling for 5 min with Laemmli's sample buffer as described previously (17).

Samples from the immunoprecipitations were separated by SDS-polyacrylamide gel electrophoresis under reducing conditions, electrotransferred to polyvinylidene difluoride membranes (Millipore), and tyrosine-phosphorylated proteins were detected with monoclonal antibody 4G10 conjugated to horseradish peroxidase as described previously (4). Proteins were visualized using the enhanced chemiluminescence kit from DuPont and Kodak X-Omat radiographic film (Eastman Kodak Co.). Antibodies were stripped from the membranes, and then membranes were reprobed with anti-Pyk2 antibodies.

Pyk2 immunoprecipitated as described above was further washed with kinase buffer (30 mM HEPES, pH 7.5, 10 mM MgCl₂, and 2 mM MnCl₂), and resuspended in 30 µl of kinase buffer. The reactions were started by the addition of 5 µCi of [-³²P]ATP and 5 µM ATP. After 30 min of incubation at room temperature, the reactions were stopped by the addition of 50 µl of 2 × Laemmli's sample buffer and boiling for 5 min. Following centrifugation, the eluted proteins were separated under reducing conditions by SDS-polyacrylamide gel electrophoresis (4-20% gels), electrotransferred to membranes, and visualized by autoradiography. In some experiments, the kinase reaction buffer contained as substrate 20 µg of poly(Glu-Tyr) (4:1) (20-50 kDa). Scanning densitometry was with a Pharmacia LKB Imagemaster.

RESULTS

Pyk2 Is Tyrosine-phosphorylated and Activated after FcRI Aggregation

We and others have reported that several ~115-kDa proteins including FAK are tyrosine-phosphorylated after stimulation of RBL-2H3 cells (6, 15). Since Pyk2 is another ~115-kDa molecule with homology to FAK, we examined whether Pyk2 was tyrosine-phosphorylated. By immunoblotting, the Pyk2 protein-tyrosine kinase was detectable in RBL-2H3 cells (see below). There was constitutive low level tyrosine phosphorylation of Pyk2 (Fig. 1A). FcRI aggregation induced a dramatic increase in the tyrosine phosphorylation of Pyk2 that was dependent on the extent of the stimulation with different concentrations of antigen. In time-course experiments, the tyrosine phosphorylation of Pyk2 was apparent at 1 min after stimulation and peaked by 10 min (Fig. 1B). This tyrosine phosphorylation paralleled the release of histamine from the cells. These results demonstrate that there is rapid tyrosine phosphorylation of Pyk2 after FcRI aggregation.

FcRI aggregation results in tyrosine phosphorylation of Pyk2.

[View Larger Version of this Image (33K GIF file)]

Since tyrosine kinase activity is important for signal transduction, we next determined whether Pyk2 was activated by stimulation of FcRI. By in vitro kinase reaction there was increased autophosphorylation of Pyk2 within 2 min after FcRI aggregation (Fig. 2). There was also increased kinase activity as determined by phosphorylation of the poly(Glu-Tyr) substrate. Therefore, aggregation of FcRI induced increased tyrosine phosphorylation and kinase activity of Pyk2.

FcRI aggregation stimulates the in vitro kinase activity of Pyk2.

[View Larger Version of this Image (38K GIF file)]

Following FcRI aggregation, some proteins are tyrosine-phosphorylated early, whereas others are phosphorylated at later stages after a rise in intracellular calcium or the activation of protein kinase C (3, 4, 8, 21). Furthermore, activation and tyrosine phosphorylation of Pyk2 by stimulation of several cell-surface receptors may be due to the increase in intracellular calcium (24). In the RBL-2H3 cells, addition of calcium ionophore A23187 to directly increase intracellular calcium induced an increase in the tyrosine phosphorylation of Pyk2, which was similar to that by FcRI activation (Fig. 3). There was also strong tyrosine phosphorylation of Pyk2 when cells were stimulated with PMA, an activator of protein kinase C. Under these conditions, there was histamine release with the calcium ionophore A23187 but no degranulation with PMA. These experiments suggested that the increased tyrosine phosphorylation of Pyk2 could be due to the rise in intracellular calcium and/or the activation of protein kinase C. 

[View Larger Version of this Image (44K GIF file)]

FcRI aggregation results in the release of Ca²⁺ from intracellular stores followed by the influx of Ca²⁺ from the medium. Since calcium ionophore induced the tyrosine phosphorylation of Pyk2, we examined the role of extracellular Ca²⁺ in the FcRI-mediated tyrosine phosphorylation of Pyk2. RBL-2H3 cells were washed and then stimulated in a Ca²⁺-free medium (Fig. 4). The tyrosine phosphorylation of Pyk2 was decreased when the cells were stimulated in the absence of Ca²⁺ in the medium. Therefore, at least part of the tyrosine phosphorylation of Pyk2 was independent of the large increase in intracellular Ca²⁺ that occurs by influx of Ca²⁺ from the medium.

Depletion of extracellular calcium reduces but does not eliminate the FcRI-induced tyrosine phosphorylation of Pyk2.

[View Larger Version of this Image (45K GIF file)]

Previously we observed that integrin-mediated adherence of RBL-2H3 cells to fibronectin resulted in the tyrosine phosphorylation of FAK (15). Cell adhesion also enhanced the FcRI-induced secretion and protein-tyrosine phosphorylation of FAK (15, 38). For the experiments described so far, the cells were adherent as monolayers and there was always some constitutive tyrosine phosphorylation of Pyk2. We therefore investigated whether adherence by integrins regulated the tyrosine phosphorylation of Pyk2 (Fig. 5). When RBL-2H3 cells were added to either BSA- or fibronectin-coated surfaces, more than 90% of the cells adhered to fibronectin-coated surfaces, but none attached to BSA-coated surfaces. After plating the cells for 20 min at 37 °C, there was a slight increase in the tyrosine phosphorylation of Pyk2 in adherent compared with nonadherent cells (data not shown). Cell stimulation had dramatically different effects in nonadherent as compared with adherent cells. In the nonadherent cells, there was a very slight increase in the tyrosine phosphorylation of Pyk2 after FcRI aggregation but no detectable change with the calcium ionophore or with PMA. In contrast, adherence dramatically enhanced the tyrosine phosphorylation of Pyk2 by stimulation with antigen, calcium ionophore, and PMA. Therefore, cell activation by adherence and other receptors synergistically regulate the tyrosine phosphorylation of Pyk2.

[View Larger Version of this Image (43K GIF file)]

Syk Dependence of the FcRI-induced Tyrosine Phosphorylation of Pyk2

In a Syk-deficient variant of the RBL-2H3 cell line, some proteins including the  and  subunits of FcRI are still tyrosine-phosphorylated, but there is no release of Ca²⁺ from intracellular sources and no influx of Ca²⁺ (20). We therefore used these Syk-deficient cells to evaluate the role of Syk in tyrosine phosphorylation of Pyk2 (Fig. 6). Although there was less Pyk2 in the Syk-negative cells, its constitutive tyrosine phosphorylation was similar to that in the wild type RBL-2H3 cells. FcRI aggregation did not induce an increase in the tyrosine phosphorylation of Pyk2 in the Syk-negative cells. However, in the cells that had been stably transfected with Syk, there was reconstitution of the FcRI-induced tyrosine phosphorylation of Pyk2. Therefore, the FcRI-induced tyrosine phosphorylation of Pyk2 requires Syk.

The FcRI-induced tyrosine phosphorylation of Pyk2 requires Syk.

[View Larger Version of this Image (46K GIF file)]

In different cell types, stimulation of G-protein-coupled receptors such as those for bradykinin, thrombin, or lysophospatidic acid results in the tyrosine phosphorylation of Pyk2 (24, 26, 31). Stimulation of thrombin and adenosine G-protein-coupled receptors in RBL-2H3 cells results in transient mobilization of intracellular calcium (39, 40). Thrombin stimulation is mediated by a pertussis toxin-insensitive G-protein, whereas adenosine is inhibited by this toxin, suggesting that it is probably mediated by G_i (39, 41). Stimulation with thrombin of both the RBL-2H3 and the Syk-negative TB1A2 cells induced the rapid tyrosine phosphorylation of Pyk2 (Fig. 7A). This tyrosine phosphorylation in both the wild type and in the Syk-negative cells was detectable within 30 s of stimulation. Similarly the stimulation of adenosine receptors induced Pyk2 tyrosine phosphorylation in both the RBL-2H3 and the Syk-negative cells (Fig. 7B). Therefore, Syk is not required for tyrosine phosphorylation of Pyk2 induced by G-protein-coupled receptors.

[View Larger Version of this Image (28K GIF file)]

DISCUSSION

These studies indicate that Pyk2 was present in mast cells and was tyrosine-phosphorylated and activated after cell stimulation. There were at least two different pathways that led to Pyk2 activation. The pathway from FcRI aggregation required Syk, whereas that from G-protein-coupled receptors was Syk-independent. Pyk2 was also tyrosine-phosphorylated either by the addition of calcium ionophore A23187 to raise intracellular calcium or when protein kinase C was activated with PMA. These results strongly suggest that, similar to results in other cells, the tyrosine phosphorylation and activation of Pyk2 in mast cells was due to the rise in intracellular calcium (24, 28, 31).

Syk plays a major role in FcRI-mediated activation of mast cells (20). In this pathway, receptor aggregation activates a protein-tyrosine kinase, probably Lyn, which results in tyrosine phosphorylation of the receptor subunits. Syk then binds to the tyrosine-phosphorylated receptor subunits and is activated to propagate downstream signals including the tyrosine phosphorylation of phospholipase C-, the release of calcium from intracellular sources, and the influx of calcium from the extracellular medium. Although FcRI aggregation in the Syk-deficient cells results in the tyrosine phosphorylation of the  and  subunits of the receptor and of several proteins (20, 42), it did not induce the activation or tyrosine phosphorylation of Pyk2. In contrast, activation of the T cell receptor, another member of the immune receptor family, results in tyrosine phosphorylation of Pyk2 that is selectively mediated by the Src family kinase Fyn (32, 33). These seemingly contradictory observations can be explained if, in mast cells, FcRI aggregation by a Syk-dependent pathway results in an increase in intracellular calcium, which then utilizes a Src family kinase to tyrosine-phosphorylate Pyk2.

The integrin-mediated adherence of RBL-2H3 cells to fibronectin results in cell spreading, reorganization of the cytoskeleton, and a redistribution of the granules to the periphery of the cells (38, 43). Adhesion also results in tyrosine phosphorylation of proteins such as FAK and the cytoskeletal protein paxillin (15, 44, 45). Adherence of platelets, megakaryocytes, and T and B cells, but not fibroblasts, results in the tyrosine phosphorylation of Pyk2 (25, 27, 35, 46). Similarly, in the present experiments, there was a slight increase in tyrosine phosphorylation of Pyk2 after adherence to fibronectin, a reaction that is probably mediated by ₁ integrins. The aggregation of ₁ integrins in B cells and ₃ integrins in T cells results in tyrosine phosphorylation of Pyk2 (34, 35). Pyk2 localizes to sites of cell-to-cell contact and to focal adhesion like structures (25, 27). At these sites, the cytoplasmic domains of the integrins form focal adhesion complexes that contain cytoskeletal proteins such as talin, vinculin, -actinin, filamin, FAK, and other phosphoproteins (27, 44, 47). Although RBL-2H3 cells do not form classical focal adhesion complexes, stimulation of the cells results in actin plaques and tyrosine phosphorylation of FAK (15, 48). Here we observed that the stimulated tyrosine phosphorylation of Pyk2, as was previously reported for FAK, was dramatically enhanced by the adhesion of the cells to fibronectin (15). Therefore, adhesion to fibronectin modulates the activation of both Pyk2 and FAK. There is also enhanced secretion from adherent cells. Therefore, adherence by regulating the level of the tyrosine phosphorylation of Pyk2, FAK, and other proteins could control the extent of degranulation in these cells.

Pyk2 interacts with signaling molecules and may therefore play a role in signal transduction in mast cells. Pyk2 associates through its C-terminal region with paxillin, a 68-kDa cytoskeletal protein that is tyrosine-phosphorylated after FcRI aggregation (36, 49). Paxillin also accumulates at focal adhesion sites and binds to pp60^src, Lyn, Crk, vinculin, and talin. Pyk2 also associates with Grb2, which by binding to Shc and Sos may activate the Ras pathway (24, 32). Although Src family kinases such as Fyn, Lck, and c-Src associate by their SH2 domains with Pyk2 (31, 32), we could not detect association of Pyk2 with the Src family kinase Lyn (data not shown). There is also binding of p130^cas to Pyk2, probably by the SH3 domain of p130^cas binding to the C-terminal proline-rich region of Pyk2 (35). Pyk2 tyrosine-phosphorylates the potassium channel and suppresses channel currents (24) and also acts as an upstream regulator for stress signal activation of c-Jun N-terminal kinase (28). As many of these pathways are activated in stimulated mast cells, tyrosine phosphorylation of Pyk2 may play a role in the signals that lead to generation of mediators.

Stimulation of cells by many G-protein-coupled receptors results in the tyrosine phosphorylation of Pyk2 (24, 26, 30, 31). Mast cells have thrombin receptors that couple to G_q, a pertussis toxin-insensitive G-protein -subunit and A₃ adenosine receptors that couple to a pertussis toxin-sensitive -subunit G_i. The stimulation of these receptors in RBL-2H3 cells results in a transient increase in intracellular Ca²⁺ (39, 40). Here we observed that stimulation of the cells with either thrombin or an adenosine analog induced the rapid tyrosine phosphorylation of Pyk2. Interestingly the extent of this phosphorylation was not dependent on the presence of Syk in the cells. These results demonstrate that there are Syk-independent pathways that link G-protein-coupled receptors to Pyk2 activation.

In summary, these experiments indicate that stimulation of mast cells with different stimuli induces the tyrosine phosphorylation of Pyk2. The FcRI-mediated phosphorylation was downstream of Syk and probably secondary to the mobilization of intracellular Ca²⁺. In contrast, G-protein-coupled receptors induced tyrosine phosphorylation of Pyk2, which was independent of Syk. Adherence of cells to fibronectin regulated the tyrosine phosphorylation of Pyk2, similar to that which we had observed previously for FAK (15). Since there are two different focal adhesion kinases in RBL-2H3 cells, it will be important to clarify the function of these two molecules in the signaling cascade.