INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

RAS plays a key role in signaling pathways leading to cell proliferation, differentiation, and transformation (5, 28). Ras is activated through conversion of an inactive GDP-bound form to an active GTP-bound form. GTP-bound active form of Ras can associate with other signaling proteins including Raf-1. The interaction of Ras with Raf-1 leads phosphorylation and activation of downstream protein kinase mitogen-activated protein (MAP) kinase kinase and extracellular signal-regulated protein kinase (ERK). It is known that Ras is activated by recruitment of son of sevenless (SOS) to a complex with adaptor Grb2 in response to growth factors and G protein-coupled receptor agonists (7, 28). Once phosphorylated by tyrosine, Shc creates a binding site for Grb2, and Grb2 subsequently associates with Shc via its src homology 2 (SH2) (26).

Recently identified nonreceptor proline-rich tyrosine kinase (PYK2) (17), also known as cell adhesion kinase- (27), related adhesion focal tyrosine kinase (2), or calcium-dependent tyrosine kinase (38), has been shown to be one of the upstream regulators of Ras through Shc/Grb2 association. PYK2 contains high sequence homology to focal adhesion kinase (FAK) and has a similar structural organization as FAK with a central kinase domain flanked by noncatalytic domains at both NH₂- and COOH-terminal domains. Both PYK2 and FAK lack SH2 and SH3 domains. PYK2 is reported to exist in neuronal, hematopoietic, epithelial, or vascular smooth muscle cells and is demonstrated to be rapidly activated and tyrosine phosphorylated by G protein-coupled receptor agonists, growth factors, cytokines, and stress signals that increase intracellular calcium concentration or activate protein kinase C (PKC). PYK2 has been implicated in the regulation of MAP kinases, ERK, c-Jun NH₂-terminal kinase (JNK), and p38 MAP kinase (17, 24, 33, 38).

Lysophosphatidylcholine (LPC), a phospholipid component of oxidized low-density lipoprotein, plays an important role for pathophysiological vasculature changes associated with atherosclerosis. We (18) and others (13, 14, 37) have reported that LPC impairs endothelium-dependent vasorelaxation of isolated arterial strips. LPC upregulates expression of platelet-derived growth factor (16), intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (15), nitric oxide synthase III (10, 40), and cyclooxygenase-2 (39), and activates transcriptional factors such as AP-1 (8) and nuclear factor (NF)-B (30) in cultured endothelial cells. However, the intracellular signal transduction mechanisms of LPC are poorly understood. It has been reported that LPC increases intracellular calcium in a wide variety of cells. Inoue and colleagues (12) demonstrated that LPC rapidly induces calcium mobilization in bovine aortic endothelial cells (BAECs) in a biphasic manner with calcium release from intracellular stores and calcium influx from extracellular space. Although the precise mechanism of LPC-induced calcium transient remains unclear, LPC induces calcium mobilization without phosphoinositide hydrolysis, and LPC-induced calcium influx may result from the alteration of membrane permeability and fluidity. Another major mediator of LPC-mediated signal transduction is PKC. It was demonstrated that effects of LPC on impairment of endothelium-dependent vasorelaxation (14), upregulation of ICAM-1 expression (31), NF-B activation (30), and superoxide production (20) are prevented by inhibitors of PKC. It was recently reported that LPC activates ERK and JNK in endothelial cells (8, 23). However, it remains unknown whether PYK2 is activated by calcium mobilization or PKC activation in response to LPC and whether its critical downstream protein Ras is activated by LPC in vascular endothelial cells.

Here we demonstrate the regulation of PYK2 in vascular endothelial cells. We found that LPC stimulates PYK2 tyrosine phosphorylation in a calcium- and PKC-dependent manner. We also showed that the role of the intact cytoskeleton in PYK2 phosphorylation, but Rho, is not necessary. Importantly, LPC increases the association of Shc with PYK2, Shc tyrosine phosphorylation, and Grb2 binding to Shc, and results in activation of Ras. These results suggest that LPC activates Ras via the PYK2/Shc/Grb2 signaling pathway in vascular endothelial cells.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Materials. Monoclonal antibodies against PYK2, FAK, Shc, Grb2, Ras, and phosphotyrosine-20 were obtained from Transduction Laboratories (San Diego, CA). Anti-Shc polyclonal antibody was from Upstate Biotechnology (Lake Placid, NY). Anti-Rho A polyclonal antibody was from Santa Cruz Biotechnologies (Santa Cruz, CA). Synthetic palmitoyl-LPC, phorbol 12-myristate 13-acetate (PMA), calcium ionophore A-23187, and cytochalasin D were from Sigma (St. Louis, MO). 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid-acetoxymethyl ester (BAPTA-AM) and GF-109203X were from Carbiochem-Novabiochem (San Diego, CA). Botulinum C3 transferase was from Wako Pure Chemical Industries (Osaka, Japan).

Cell cultures. BAECs were isolated from bovine thoracic aortas as described previously (10, 12) and identified by their typical cobblestone appearance and positive immunofluorescence for factor VIII. Cells were cultured in Dulbecco's modified Eagle's medium (GIBCO; Rockville, MD) supplemented with 15% fetal calf serum, 10 µg/ml sulbenicillin, and 15 µg/ml gentamicin at 37°C in a humidified atmosphere of 95% air-5% CO₂ and used between passages 5 and 13.

Adenoviral transfection. Transient transfection was carried out by using replication-defective recombinant adenoviral vectors prepared as described previously (25). In brief, a dominant negative mutant of Rho A (Thr¹⁹ to Asn, Rho A T19N, generously provided by Dr. Y. Takai, Osaka University, Osaka, Japan) was placed in pAdex1CAwt, a cassette cosmid vector (kindly provided by Dr. I. Saito, University of Tokyo, Tokyo, Japan) under a CA promoter composed of a cytomegalovirus enhancer and a chicken -actin promoter (pAdex Rho A T19N). A recombinant adenovirus was constructed by in vitro homologous recombination in 293 cells using pAdex Rho A T19N and the adenovirus DNA-terminal protein complex. Cells were grown on 60-mm culture dishes. After reaching confluence, cells were infected with recombinant adenovirus expressing Rho A T19N or LacZ (Ad.Rho A T19N or Ad.LacZ) for 1 h at 37°C in a 95% air-5% CO₂ atmosphere. The viral suspension was removed and cells were cultured for 48 h.

Immunofluorescence staining. BAECs were cultured on eight-well chamber slide (Lab-Tek, Nunc; Naperville, IL). Cells were treated with ice-cold acetone for 10 min and then washed three times in PBS. After being blocked, cells were incubated with anti-PYK2 antibody (1:500), washed three times in PBS, and then incubated with a FITC-conjugated secondary antibody. Cells were mounted with fluorescent mounting medium (DAKO; Carpinteria, CA) and photographed using an Axioscop fluorescence microscope (Carl Zeiss; Thornwood, NY) fitted with a ×100 objective lens.

Immunoprecipitation. Cells were grown to reach confluence on 100-mm culture dishes and incubated in a serum-free Dulbecco's modified Eagle's medium for 24 h. After cells were stimulated with LPC in the manner indicated in each experiment, the cells were rinsed twice with ice-cold phosphate-buffered saline and lysed in a lysis buffer of 10 mM Tris · HCl (pH 7.4), 50 mM NaCl, 1 mM EDTA, 1 mM Na₃VO₄, 10 mM sodium pyrophosphate, 50 mM NaF, 1% Triton X-100, 5 µg/ml leupeptin, 5 µg/ml pepstatin A, and 1 mM phenylmethylsulfonyl fluoride. Immunoprecipitation was carried out at 4°C by incubating cell lysates (1 mg) for 2 h with 2 µg of anti-PYK2, anti-FAK monoclonal antibody or anti-Shc polyclonal antibody followed by incubation for 1 h with protein G sepharose 4 fast flow or protein A sepharose CL-4B (Amersham Pharmacia Biotech; Uppsala, Sweden). After being washed three times with lysis buffer, the beads were resuspended in 2× Laemmli buffer.

Immunoblot analysis. Immunoprecipitated proteins or whole cell lysates were resolved by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to membranes (Immobilon-P, Millipore; Bedford, MA). Membranes were immunoblotted with appropriate antibodies (1:1,000) followed by incubation with the secondary antibody (1:1,000) and developed by using the ECL detection assay (Amersham Pharmacia Biotech).

Evaluation of Ras activation. The active GTP-bound form of Ras was detected by using glutathione-S-transferase (GST) fusion protein corresponding to the Ras-binding domain (RBD) of Raf-1 bound to glutathione-agarose (Upstate Biotechnology; Waltham, MA) according to the manufacturer's protocol. After each stimulation, cells were lysed in a magnesium-containing lysis buffer of 25 mM HEPES (pH 7.5), 150 mM NaCl, 1% NP-40, 0.25% sodium deoxycholate, 10% glycerol, 25 mM NaF, 10 mM MgCl₂, 1 mM EDTA, 1 mM Na₃VO₄, 10 µg/ml leupeptin, and 10 µg/ml aprotinin. Lysates were incubated with 5 µl of Raf-1(RBD)-GST-agarose beads at 4°C for 30 min. Beads then were washed three times with the lysis buffer and resuspended in 2× Laemmli buffer. The supernatant was boiled for 5 min and then collected by centrifugation. The protein samples were resolved by SDS-PAGE and transferred onto membranes. Ras was detected by immunoblot analysis with anti-Ras antibody.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

LPC stimulates tyrosine phosphorylation of PYK2 in BAECs. The expression and distribution of PYK2 in BAECs were examined by immunostaining with anti-PYK2 antibody. As shown in Fig. 1, PYK2 immunostaining was mainly located within the perinuclear domain and was not localized to the focal adhesion contacts. PYK2 immunostaining did not show distinct change after treatment with LPC or PMA (Fig. 1).

View larger version (51K): [in this window] [in a new window]   Fig. 1.   Distribution of proline-rich tyrosine kinase (PYK2) in bovine aortic endothelial cells (BAECs). BAECs, which were untreated (Control), treated for 10 min with lysophosphatidylcholine (LPC, 20 µM), or treated with phorbol 12-myristate 13-acetate (PMA, 1 µM) were immunolabeled with anti-PYK2 antibody.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Tyrosine phosphorylation of PYK2 by LPC treatment in BAECs. A: serum-starved BAECs were treated with LPC (20 µM) for the indicated periods. Cell lysates were immunoprecipitated with monoclonal anti-PYK2 antibody, subjected to 7.5% SDS-PAGE, immunoblotted with monoclonal antiphosphotyrosine antibody (top), and reprobed with anti-PYK2 antibody (bottom). Left, molecular mass markers. Representative results and densitometric analysis of PYK2 phosphorylation are shown. Results represent means ± SE of 3 independent experiments. PY-20, phosphotyrosine-20. B: serum-starved BAECs were treated for 10 min with LPC at the indicated concentrations. Representative results and densitometric analysis of PYK2 phosphorylation are shown. Results represent means ± SE of 3 independent experiments. C: serum-starved BAECs were treated with LPC (20 µM) for the indicated periods. Cell lysates were immunoprecipitated with monoclonal anti-focal adhesion kinase (FAK) antibody, subjected to 7.5% SDS-PAGE, and immunoblotted with monoclonal antiphosphotyrosine antibody (top). Membranes were reprobed with monoclonal anti-FAK antibody (bottom). Representative results and densitometric analysis of FAK phosphorylation are shown. Results represent means ± SE of 3 independent experiments. IP, immunoprecipitates.

Calcium- and PKC-dependent tyrosine phosphorylation of PYK2 by LPC. It has been demonstrated (12, 14, 20, 30, 31) that LPC induces calcium mobilization and PKC activation in endothelial cells. Calcium mobilization and PKC activity have been shown to regulate tyrosine phosphorylation of PYK2. We investigated the role of calcium and PKC in LPC-induced PYK2 phosphorylation. Cells treated with calcium ionophore A-23187 (10 µM) showed an increase in PYK2 phosphorylation (Fig. 3A). Treatment with calcium chelators BAPTA-AM (25 µM) or EGTA (3 mM) almost completely abolished LPC-induced increase in PYK2 phosphorylation, suggesting that calcium is required for LPC-induced PYK2 phosphorylation.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Calcium- and PKC-dependent tyrosine phosphorylation of PYK2. A: serum-starved BAECs were pretreated with 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid-acetoxymethyl ester (BAPTA-AM, 25 µM), EGTA (3 mM), or vehicle (0.06% dimethyl sulfoxide) for 30 min followed by stimulation with LPC (20 µM) or A-23187 (10 µM) for 10 min. Representative results and densitometric analysis of PYK2 phosphorylation are shown. Results represent means ± SE of 3 independent experiments. B: serum-starved BAECs were pretreated with GF-109203X (5 µM) or vehicle (0.1% dimethyl sulfoxide) for 30 min followed by stimulation with LPC (20 µM) or PMA (1 µM) for 10 min. Representative results and densitometric analysis of PYK2 phosphorylation are shown. Results represent means ± SE of 3 independent experiments. C: BAECs were pretreated with PMA (1 µM, 24 h) or vehicle (0.05% dimethyl sulfoxide) for 24 h followed by stimulation with LPC (20 µM) or PMA (1 µM) for 10 min. Representative results and densitometric analysis of PYK2 phosphorylation are shown. Results represent means ± SE of 3 independent experiments. dep(), Not depleted; dep(+), depleted.

Effects of cytochalasin D, C3 transferase, and Rho A T19N on PYK2 phosphorylation. Recently, it has been demonstrated that PYK2 localizes in the actin cytoskeleton and that dependency of PYK2 tyrosine phosphorylation on the actin cytoskeleton is also suggested in hematopoietic and vascular smooth muscle cells (6, 9, 11). To investigate whether the integrity of the actin cytoskeleton is necessary for PYK2 phosphorylation in BAECs, cells were treated for 2 h with cytochalasin D (3 µM) to disrupt the actin cytoskeleton. As shown in Fig. 4A, treatment with cytochalasin D significantly inhibited both LPC- and PMA-stimulated PYK2 phosphorylation, suggesting that the integrity of the actin cytoskeleton is required for PYK2 phosphorylation.

View larger version (24K): [in this window] [in a new window]   Fig. 4.   Effects of cytochalasin D (CytoD) and C3 transferase on PYK2 tyrosine phosphorylation. A: serum-starved BAECs were pretreated with cytochalasin D (3 µM) or vehicle (0.1% dimethyl sulfoxide) for 15 min followed by stimulation with LPC (20 µM) or PMA (1 µM) for 10 min. Representative results and densitometric analysis of PYK2 phosphorylation are shown. Results represent means ± SE of 3 independent experiments. B: BAECs were untreated [C3()] or pretreated with C3 transferase [C3(+), 10 µg/ml, 48 h] followed by stimulation with LPC (20 µM) or PMA (1 µM) for 10 min. Representative results and densitometric analysis of PYK2 phosphorylation are shown. Results represent means ± SE of 3 independent experiments. C: BAECs were infected with Ad.Rho A T19N or Ad.LacZ at a multiplicity of infection of 50. Forty-eight hours after infection, cells were stimulated with LPC (20 µM) or PMA (1 µM) for 10 min. Representative results and densitometric analysis of PYK2 phosphorylation are shown. Results represent means ± SE of 3 independent experiments. To confirm overexpression of Rho A T19N, whole cells lysates (WCL) from the same samples were used for immunoblot analysis with anti-Rho A antibody.

LPC increases Shc tyrosine phosphorylation and formation of PYK2/Shc/Grb2 complex. To determine whether PYK2 phosphorylation by LPC facilitates signals via the Shc/Grb2 pathway, we examined the effects of LPC on formation of PYK2/Shc/Grb2 complex. For this purpose, cell lysates were immunoprecipitated with anti-PYK2 antibody, followed by immunoblotting with anti-Shc antibody. As shown in Fig. 5, p52 isoform of Shc was associated with PYK2 under unstimulated condition, and LPC time dependently increased coimmunoprecipitation of p52 and p66 isoforms of Shc with anti-PYK2 antibody, which peaked at 10 min and declined thereafter. As shown in Fig. 6C, in whole cell lysates or in Shc immunoprecipitates, the three isoforms p46, p52, and p66 were expressed in BAECs. Antiphosphotyrosine immunoblotting showed that p52 Shc exhibited the highest tyrosine phosphorylation level under unstimulated condition and showed a slight increase in tyrosine phosphorylation by LPC (Fig. 6A). It is clearly evident that LPC treatment increased tyrosine phosphorylation of p66 Shc (Fig. 6B). By anti-Grb2 immunoblotting of the Shc immunoprecipitates, Grb2 associated with Shc was time dependently increased by LPC (Fig. 6D). Increased Shc association with PYK2, Shc tyrosine phosphorylation, and Grb2 association with Shc in response to LPC paralleled the time course of increased PYK2 tyrosine phosphorylation (Fig. 2A). Thus PYK2 may transmit LPC-mediated signals through formation of PYK2/Shc/Grb2 complex.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   LPC stimulates PYK2-Shc complex formation. Serum-starved BAECs were treated with LPC (20 µM) for the indicated periods. Cell lysates were immunoprecipitated with monoclonal anti-PYK2 antibody and immunoblotted with polyclonal anti-Shc antibody (top). Membrane was reprobed with anti-PYK2 antibody (bottom). Left, molecular mass markers. Representative results of 3 independent experiments are shown.

View larger version (64K): [in this window] [in a new window]   Fig. 6.   LPC stimulates Shc tyrosine phosphorylation and Grb2 binding to Shc. Serum-starved BAECs were treated with LPC (20 µM) for the indicated periods. Cell lysates were immunoprecipitated with polyclonal anti-Shc antibody, then subjected to 12% SDS-PAGE, and immunoblotted with antiphosphotyrosine (A, B) or anti-Grb2 antibody (D), then reprobed with monoclonal anti-Shc antibody (C). Representative results of 3 independent experiments are shown.

Stimulation of Ras by LPC. PYK2 facilitates Ras activation via the Shc/Grb2-dependent pathway (17). We determined whether LPC induces Ras activation. Ras is activated through conversion of an inactive GDP-bound form to an active GTP-bound form. To estimate formation of an active GTP-bound form of Ras, lysates from cells stimulated with LPC for various periods were incubated with glutathione-agarose beads coupled to GST-RBD, and the precipitated proteins were then analyzed by immunoblotting with anti-Ras antibody. As shown in Fig. 7, A and B, LPC induced a transient increase in the formation of an active GTP-bound form of Ras with a peak at 10 min. Calcium chelators BAPTA-AM and EGTA inhibited LPC-induced Ras activation, whereas GF-109203X did not show any effect (Fig. 7C).

View larger version (24K): [in this window] [in a new window]   Fig. 7.   LPC activates Ras. Serum-starved BAECs were treated with LPC (20 µM) for the indicated periods. Ras activity was determined by immunoblot analysis of proteins precipitated with glutathione-S-transferase-Ras-binding domain (GST-RBD) as described in MATERIALS AND METHODS. Representative results and densitometric analysis are shown (A and B). Results represent means ± SE of 3 independent experiments. C: serum-starved BAECs were pretreated with BAPTA-AM (25 µM), EGTA (3 mM), GF-109203X (5 µM), or vehicle (0.1% dimethyl sulfoxide) for 30 min followed by stimulation with LPC (20 µM) for 10 min. Representative results of 2 independent experiments are shown.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Expression and regulatory mechanisms of PYK2 have been previously reported in a variety of cells such as neuronal cells (17, 33), hematopoietic cells (1, 2, 11, 33), liver epithelial cells (38), and vascular smooth muscle cells (6). However, little is known about the PYK2 regulation in vascular endothelial cells. The results of the present study demonstrate for the first time that treatment with LPC rapidly increases tyrosine phosphorylation of PYK2 in vascular endothelial cells. The increase in PYK2 phosphorylation by LPC was observed as low as 2.5 µM and was more evident at the concentrations over 5 µM. LPC is an amphiphilic molecule, and the critical micellar concentration is 40-50 µM (3). Because the concentrations of LPC used in the present study are below the critical micellar concentration, PYK2 phosphorylation by LPC is unlikely due to its nonspecific detergent effect.

LPC treatment leads to calcium mobilization and PKC activation in vascular endothelial cells (12, 14, 20, 30, 31). It was shown that calcium mobilization and PKC activation cause PYK2 tyrosine phosphorylation in some types of cells. However, depleting PMA-sensitive PKC does not block CD28- and CD3-induced PYK2 tyrosine phosphorylation in Jurkat T cells (35). In the present study, LPC-stimulated PYK2 phosphorylation was effectively inhibited by treatment with calcium chelators BAPTA-AM or EGTA, and calcium ionophore A-23187 increased PYK2 phosphorylation, suggesting the involvement of calcium in PYK2 phosphorylation in endothelial cells. In addition, a PKC activator PMA increased PYK2 phosphorylation, which was prevented by PKC inhibition with a specific PKC inhibitor GF-109203X or PKC depletion. These treatments also prevented PYK2 phosphorylation in response to LPC. Previous reports demonstrated that five PKC isoforms (, , , , and ) were identified in BAECs (36). Because GF-109203X, a PKC inhibitor used in the present study, is known to inhibit PKC-, -, and - (34), PKC-, or - are likely involved in LPC-induced PYK2 phosphorylation. Together, these results suggest that both calcium mobilization and PKC activation are involved in PYK2 phosphorylation in endothelial cells. However, PYK2 is not activated by calcium in vitro, indicating the possible existence of the candidate for intermediating signal(s) between calcium and PYK2 (17). It has been reported that PYK2 is downstream of Syk (22), Janus kinase (Jak)2 (32) and Jak3 (19) in FCRI receptor-, interferon- receptor-, and interleukin-2 receptor-signaling pathways, respectively, in some cell types. Identification of the candidate for intermediating signal(s) between PYK2 and either calcium or PKC requires further investigations.

PYK2 is related to FAK because of similarities of amino acid sequence and structural organization lacking in SH2 and SH3 domains. Both PYK2 and FAK play an essential role in activation of ERK pathway. Activation of PYK2 and FAK promotes Grb2 association with SOS through their common targets src-family tyrosine kinases and Shc and leads to activation of Ras/ERK cascade. Despite the high sequence homology, structural similarities, and common downstream effector molecules between PYK2 and FAK, recent studies provide evidence of their different subcellular localization and regulatory mechanisms (29, 41). FAK is mainly localized to the focal adhesion region, whereas PYK2 is located in perinuclear regions in mouse fibroblasts (29) or diffusely exists throughout the cytoplasm in rat vascular smooth muscle cells (41). Immunostaining using anti-PYK2 antibody showed diffuse staining throughout the cytoplasm in BAECs (Fig. 1). LPC does not increase tyrosine phosphorylation of FAK (Fig. 2C), and therefore, the activities of PYK2 and FAK are likely to be regulated by different upstream molecules in LPC-stimulated endothelial cells. Tyrosine phosphorylation of PYK2, but not of FAK, in response to stimuli other than LPC has been previously reported (11, 35, 38). Zheng et al. (41) demonstrated that the distinctive COOH-terminal domains of PYK2 and FAK are responsible for their different subcellular localization and regulation.

It has been demonstrated that the integrity of the cytoskeleton is required for PYK2 phosphorylation and that PYK2 associates with paxillin and p130Cas in other cell systems (1, 11). Rho is implicated in activation of the tyrosine phosphorylation of FAK, paxillin, and p130Cas. In addition, it was reported that PYK2 can bind Graf, a Rho GTPase-activating protein (21). We therefore examined whether the intact cytoskeleton and Rho are involved in LPC-induced PYK2 phosphorylation. We found that the intact cytoskeleton is required for PYK2 phosphorylation, but Rho is not upstream of PYK2. The differences in the subcellular distributions of PYK2 and other cytoskeletal molecules such as FAK, p130Cas, and paxillin may contribute to the differences in dependency on Rho.

PYK2 has been shown to function as an essential intermediate providing a link between extracellular stimuli and signaling pathways of ERK, JNK, and p38 MAP kinase (17, 24, 33, 38). Recent evidence (4, 17) suggested that PYK2 activates the Ras/ERK pathway by directly binding the Grb2/SOS complex or indirectly via Grb2 binding to tyrosine-phosphorylated Shc. Blaukat et al. (4) have demonstrated that Grb2 binding to tyrosine-phosphorylated Shc plays a major role in the PYK2-induced activation of the Ras/ERK pathway. In this study, we show that LPC stimulates Shc association with PYK2, Shc tyrosine phosphorylation, and Grb2 binding to Shc (Figs. 5 and 6). Although direct binding of the Grb2/SOS complex to PYK2 may be involved in the PYK2-dependent activation of Ras by LPC, formation of the Shc/Grb2 complex links PYK2 with Ras activation. The results demonstrating that LPC-induced activation of Ras was significantly inhibited by calcium chelators are consistent with this hypothesis.

It was reported that LPC-induced MAP kinase activation was PKC independent in BAECs (23). PYK2 phosphorylation was inhibited by either treatment with calcium chelators or PKC inhibition, whereas the formation of PYK2/Shc/Grb2 induced by LPC was inhibited by calcium chelators, but not by PKC inhibition (data not shown). In addition, although PMA stimulated PYK2 tyrosine phosphorylation, PMA did not stimulate formation of PYK2/Shc/Grb2. These results suggest that PKC activity is required for LPC-induced PYK2 phosphorylation, whereas LPC-induced formation of PYK2/Shc/Grb2 is independent of PKC activity. It appears dependent on the redundancy. LPC-induced calcium transient may be enough to stimulate formation of PYK2/Shc/Grb2. These are consistent with the data showing that an inhibitor of PKC failed to inhibit LPC activation of Ras (Fig. 7C).

The functional role of LPC-induced activation of PYK2/Ras pathway is not defined in this study. Ras activation mediates activation of several downstream signaling cascades including ERK. We have observed that LPC induces cyclooxygenase-2 expression partly via the ERK-dependent pathway in BAEC (unpublished observations). It is likely that LPC-activated PYK2/Ras pathway leads to activation of ERK and these signaling pathways are involved in the LPC-inducible genes including cyclooxygenase-2. In addition to a significant role in the LPC-induced PYK2-dependent activation of ERK, Ras activation may play a role in the PYK2-dependent activation of JNK because it has been shown that a dominant negative mutant of Ras inhibits the PYK2-dependent activation of JNK (33). On the other hand, the p130Cas/Crk pathway is also involved in the PYK2-induced activation of JNK (4). It remains to be determined whether the PYK2/p130Cas/Crk signaling pathway is involved in the LPC-induced JNK activation. We are currently investigating the role of p130Cas in the signal transduction by LPC.