Requirement for phosphatidylinositol 3'-kinase activity in platelet-derived growth factor-stimulated tyrosine phosphorylation of p125 focal adhesion kinase and paxillin.

The role of phosphatidylinositol 3'-kinase (PI 3'-kinase) activity in platelet-derived growth factor (PDGF)-stimulated tyrosine phosphorylation of focal adhesion kinase (p125FAK) and paxillin has been examined. The tyrosine phosphorylation of p125FAK and paxillin in response to PDGF was markedly inhibited by wortmannin in a dose-dependent manner. PDGF-stimulated PI 3'-kinase activity, membrane ruffle formation, and tyrosine phosphorylation of p125FAK and paxillin were all inhibited by the same low concentrations of wortmannin (>90% inhibition at 40nM). In contrast, tyrosine phosphorylation of p125FAK and paxillin in response to bombesin, endothelin, and phorbol 12,13-dibutyrate was not inhibited by wortmannin in these cells. Furthermore, LY294002, an inhibitor of PI 3'-kinase structurally unrelated to wortmannin, also inhibited PDGF-stimulated p125FAK tyrosine phosphorylation. PDGF was shown to stimulate the tyrosine phosphorylation of p125FAK in porcine aortic endothelial (PAE) cells transfected with the wild type PDGF-beta receptors, but not in PAE cells transfected with PDGF-beta receptors in which the PI 3'-kinase binding sites (Tyr-740/751) were replaced by phenylalanine. PDGF-stimulated, PI 3'-kinase-dependent tyrosine phosphorylation of p125FAK was not inhibited by rapamycin, and thus it was dissociated from the activation of p70 S6 kinase, previously identified as a molecular downstream target of PI 3'-kinase. Thus, we have identified a PI 3'-kinase-dependent signal transduction pathway in the action of PDGF, which leads to the phosphorylation of p125FAK and paxillin.

The role of phosphatidylinositol 3-kinase (PI 3-kinase) activity in platelet-derived growth factor (PDGF)stimulated tyrosine phosphorylation of focal adhesion kinase (p125 FAK ) and paxillin has been examined. The tyrosine phosphorylation of p125 FAK and paxillin in response to PDGF was markedly inhibited by wortmannin in a dose-dependent manner. PDGF-stimulated PI 3kinase activity, membrane ruffle formation, and tyrosine phosphorylation of p125 FAK and paxillin were all inhibited by the same low concentrations of wortmannin (>90% inhibition at 40 nM). In contrast, tyrosine phosphorylation of p125 FAK and paxillin in response to bombesin, endothelin, and phorbol 12,13-dibutyrate was not inhibited by wortmannin in these cells. Furthermore, LY294002, an inhibitor of PI 3-kinase structurally unrelated to wortmannin, also inhibited PDGF-stimulated p125 FAK tyrosine phosphorylation. PDGF was shown to stimulate the tyrosine phosphorylation of p125 FAK in porcine aortic endothelial (PAE) cells transfected with the wild type PDGF-␤ receptors, but not in PAE cells transfected with PDGF-␤ receptors in which the PI 3-kinase binding sites (Tyr-740/751) were replaced by phenylalanine. PDGF-stimulated, PI 3-kinasedependent tyrosine phosphorylation of p125 FAK was not inhibited by rapamycin, and thus it was dissociated from the activation of p70 S6 kinase, previously identified as a molecular downstream target of PI 3-kinase.
Thus, we have identified a PI 3-kinase-dependent signal transduction pathway in the action of PDGF, which leads to the phosphorylation of p125 FAK and paxillin.
PI 3Ј-kinase phosphorylates inositol phospholipids on the D3 position. In vivo this enzyme is thought to phosphorylate the head group of PtdIns (4,5)P 2 to yield PtdIns (3,4,5)P 3 , and this lipid has been postulated to act as a second messenger (27)(28)(29)(30)(31). p70 S6K has been identified as one of the putative molecular downstream targets of PI 3Ј-kinase activity (32-35) (but see also Ref. 36). There is evidence, however, that PI 3Ј-kinaseregulated signaling bifurcates upstream of p70 S6K , implying that there must be other molecular downstream targets of PI 3Ј-kinase activity (32,33). PDGF stimulates the recruitment of polymerized actin into membrane ruffles and this cytoskeletal response is a PI 3Ј-kinase-dependent event (37)(38)(39). Likewise, PDGF-stimulated chemotaxis is dependent on PI 3Ј-kinase activity (37). It is not known, however, whether p70 S6K or other molecular downstream targets of PI 3Ј-kinase are involved in these PDGF-stimulated processes.
PI 3Ј-kinase activity, membrane ruffle formation, and tyrosine phosphorylation of p125 FAK and paxillin are stimulated by the same low concentrations of PDGF in Swiss 3T3 cells (18). Furthermore, the tyrosine phosphorylation of p125 FAK and paxillin by PDGF is critically dependent on the integrity of the actin cytoskeleton (18). Cytochalasin D, an agent that prevents actin polymerization, inhibits PDGF-stimulated tyrosine phosphorylation of these cytoskeletal-associated proteins (18). We reasoned, therefore, that PI 3Ј-kinase may lie upstream in a common signal transduction pathway stimulating the formation of membrane ruffles and the tyrosine phosphorylation of p125 FAK and paxillin. To test this hypothesis, we utilized two different experimental approaches. First, PI 3Ј-kinase activity was directly inhibited by pretreatment of Swiss 3T3 cells with two structurally unrelated inhibitors, wortmannin and LY294002. Second, PAE cells were transfected with either wild type PDGF-␤ receptors or PDGF ␤ receptors in which the PI 3Ј-kinase binding sites (Tyr-740/751) were replaced by phenylalanine. Using these two complementary approaches, we demonstrate here, for the first time, that the inhibition of PDGF-stimulated PI 3Ј-kinase activity, prevents the tyrosine phosphorylation of p125 FAK and paxillin in response to PDGF in a selective manner.

EXPERIMENTAL PROCEDURES
Cell Culture-Cell cultures of Swiss 3T3 fibroblasts and PAE cells were maintained and propagated in DMEM and Ham's F-12 medium, respectively, as described previously (40 -42). PAE cells were serumstarved by incubation in Ham's F-12 medium containing 1 mg/ml bovine serum albumin for 12 h.
Construction of Stable PAE Cell Lines Expressing Wild Type and Y740F/Y751F PDGF ␤ Receptors-cDNA encoding the PDGF ␤ receptor (43) was subcloned into the pAlter vector (Promega Corp.) and site directed mutagenesis was performed to substitute phenylalanine residues for the tyrosine residues 740 and 751 using the Altered Sites in vitro mutagenesis system (Promega Corp.). Wild type and mutated cDNAs were inserted into the expression vector pcDNA1 neo (Invitrogen) and PAE cells transfected with the constructs by electroporation. Stable cell lines expressing wild type (48 ϫ 10 3 receptors/cell) and Y740F/Y751F receptors (30 ϫ 10 3 receptors/cell) were selected with neomycin (G-418 sulfate).
Immunoprecipitations-Quiescent cultures of Swiss 3T3 cells or serum-starved PAE cells were washed twice with DMEM and Ham's F-12 medium, respectively, treated with peptide factors in 1 ml of the corresponding medium as indicated and lysed at 4°C in 1 ml of a solution containing 10 mM Tris/HCl, 5 mM EDTA, 50 mM NaCl, 30 mM sodium pyrophosphate, 50 mM NaF, 100 M Na 3 VO 4 , 1% Triton X-100, and 1 mM phenylmethylsulfonyl fluoride, pH 7.6 (lysis buffer). Lysates were clarified by centrifugation at 15,000 ϫ g for 10 min and precleared by incubation with albumin-agarose for 1 h at 4°C. After removal of albumin-agarose by brief (10 s) centrifugation, the supernatants were transferred to fresh tubes and proteins were immunoprecipitated for 12 h at 4°C with agarose-coupled monoclonal antibodies either directed against phosphotyrosine or specific cellular proteins, as described previously (44,45). Immunoprecipitates were washed three times with lysis buffer, extracted in 2 ϫ SDS-PAGE sample buffer (200 mM Tris-HCl, 6% SDS, 2 mM EDTA, 4% 2-mercaptoethanol, 10% glycerol, pH 6.8), and then fractionated by one-dimensional SDS-PAGE and analyzed as described under "Results" and in the figure legends.
p70 S6K Mobility Shift Assay-Activation of p70 S6K was determined by the appearance of slower migrating forms in SDS-PAGE as a result of phosphorylation on several clustered serine and threonine residues (46). Immunoblot analysis on cell lysates was performed using a rabbit polyclonal antibody which recognized both ␣I and ␣II isoforms of p70/ 85 S6K (47) as described under "Western Blotting." Western Blotting-After SDS-PAGE, proteins were transferred to Immobilon transfer membranes. Membranes were blocked using 5% nonfat dried milk in PBS, pH 7.2, and incubated for 2 h with either the anti-Tyr(P) mAbs (a mixture of Py20 and 4G10, 1 g/ml antibody), GAP antiserum (1:1000), p70 S6K rabbit polyclonal Ab (1 g/ml), or anti-p125 FAK Affiniti mAb (1:1000) as indicated, in PBS containing 0.05% Tween-20 and 3% nonfat dried milk. Immunoreactive bands were visualized using either 125 I-labeled sheep anti-mouse IgG for anti-Tyr(P) and p125 FAK or 125 I-labeled protein A for the GAP antiserum and p70 S6K antibody.
PI 3Ј-Kinase Assay-Phosphotyrosyl proteins were immunoprecipitated from cells as described above. The immunoprecipitates were assayed for phosphatidylinositol phosphorylation activity as described by Whitman et al. (48).
Immunostaining of Cells-Quiescent Swiss 3T3 cells were washed in DMEM and then incubated for 10 min at 37°C with endothelin (10 nM) and EGF (20 ng/ml). For staining of actin, cells were washed once with PBS, fixed in 3.7% paraformaldehyde in PBS for 30 min at 4°C, and then permeabilized with PBS containing 0.2% Triton X-100 for 8 min at room temperature. The cells were then incubated with tetramethyl rhodamine B isothiocyanate-conjugated phalloidin (0.25 mg/ml) in PBS for 10 min at room temperature and visualized utilizing a light microscope.
Materials-Bombesin, wortmannin, albumin-agarose, agaroselinked anti-mouse IgG and phalloidin were obtained from Sigma. Ham's F-12 medium was from Life Technologies, Inc. Recombinant PDGF (BB homodimer), [␥-32 P]ATP (5000 Ci/mmol), 125 I-labeled sheep anti-mouse IgG (15 Ci/g), and 125 I-labeled protein A (15 Ci/g) were from Amersham. Py20 anti-Tyr(P) mAb and the mAb directed against paxillin (mAb 165) were from ICN, Buckinghamshire, United Kingdom (UK). The 4G10 anti-Tyr(P) mAb was from Tissue Culture Supplies, Buckinghamshire, UK. Anti-p125 FAK mAb was obtained from Affiniti Research Products Ltd., Nottingham, UK. p70 S6K rabbit polyclonal antibody was from Santa Cruz Biotechnology, Santa Cruz, CA. The anti-Tyr(P) mAb PY72 was obtained from the hybridoma development unit of the Imperial Cancer Research Fund. GAP antiserum was a gift of Dr. J. Downward, Imperial Cancer Research Fund. LY294002 was provided by Dr S. Cartlidge, Zeneca, UK.

RESULTS
The Effect of Wortmannin on PDGF-and Bombesin-stimulated Tyrosine Phosphorylation-Wortmannin directly binds to and inhibits the catalytic (110 kDa) subunit of PI 3Ј-kinase (49 -52). To establish concentrations of wortmannin that effectively inhibited PI 3Ј-kinase activity in intact Swiss 3T3 cells stimulated with PDGF (3 ng/ml), we initially examined the effect of wortmannin on both PDGF-stimulated PI 3Ј-kinase activity and membrane ruffle formation. Fig. 1A shows that wortmannin caused a 90% inhibition of PDGF-stimulated PI 3Ј-kinase activity when cells were pretreated with 40 nM wortmannin for 10 min. The lowest concentration of wortmannin that produced an effective inhibition of PDGF-stimulated membrane ruffle formation was found to be 30 nM (data not shown).
To determine the effect of wortmannin on the tyrosine phosphorylation of proteins in response to PDGF and bombesin, quiescent Swiss 3T3 cells were preincubated with 30 nM wortmannin for 10 min and then incubated for another 10 min in the presence of either 3 or 30 ng/ml PDGF or 10 nM bombesin. The cell lysates were immunoprecipitated with the anti-Tyr(P) mAb PY72 and then analyzed by SDS-PAGE, followed by immunoblotting with a mixture of the anti-Tyr(P) mAb PY20 and 4G10. In agreement with our previous results (18), PDGF stimulated the tyrosine phosphorylation of distinct substrates at different concentrations (Fig. 1B). Bands with M r 110,000 -130,000 and 70,000 -75,000 were phosphorylated on tyrosine in response 3 ng/ml PDGF, while their intensity was markedly reduced at 30 ng/ml PDGF, a concentration of PDGF that stimulated the phosphorylation of numerous other proteins (Fig. 1B). Preincubation of cells with 30 nM wortmannin caused FIG. 1. Effect of wortmannin on PDGF-stimulated PI 3-kinase activity and the tyrosine phosphorylation of proteins in response to PDGF and bombesin. A, quiescent Swiss 3T3 cells were preincubated for 10 min at 37°C with 0, 20, or 40 nM wortmannin. Cells were subsequently incubated with 3 ng/ml PDGF for 10 min and then lysed. The lysates were immunoprecipitated with anti-Tyr(P) mAb, and the PI 3Ј-kinase activity in the immunoprecipitates was assayed. The autoradiogram obtained was scanned with an LKB Ultrascan XL densitometer to quantify phospholipids in terms of peak area. Values correspond to the phosphorylation of PI expressed as a percentage of the maximal response. The results shown are representative of at least three experiments. B, cells were preincubated in the presence (ϩ) or absence (Ϫ) of 30 nM wortmannin for 10 min, subsequently incubated for 10 min with either 3 or 30 ng/ml PDGF or 10 nM bombesin and then lysed. The lysates were immunoprecipitated with the anti-Tyr(P) mAb PY72 and then analyzed by immunoblotting with either anti-Tyr(P) mAbs (PY) (a mixture of PY20 and 4G10) or a GAP antiserum (GAP) as indicated. In this and subsequent figures, representative autoradiograms are shown of experiments repeated at least three times. a marked decrease in the tyrosine phosphorylation of the M r 110,000 -130,000 and 70,000 -75,000 bands stimulated by 3 ng/ml PDGF. Several lines of evidence indicate that this effect of wortmannin was selective. (a) At both 3 and 30 ng/ml PDGF, the most prominent band phosphorylated on tyrosine has a M r 170,000 -190,000 and corresponds to the autophosphorylated ␣ and ␤ PDGF receptor chains (2,3). Pretreatment of cells with wortmannin had no effect on the degree of tyrosine phosphorylation of the PDGF receptor chains stimulated by either 3 or 30 ng/ml PDGF, implying that wortmannin did not interfere with PDGF receptor autophosphorylation in Swiss 3T3 cells (Fig. 1B). (b) Wortmannin had no apparent effect on the tyrosine phosphorylation of multiple proteins stimulated by 30 ng/ml PDGF. (c) Accordingly, wortmannin pretreatment of cells had no effect on PDGF-stimulated tyrosine phosphorylation of GAP (Fig. 1B) or the mobilization of intracellular Ca 2ϩ that occurs as a result of PDGF activation of phospholipase C␥ (results not shown). These results suggest that the inhibitory effect of wortmannin on PDGF-stimulated tyrosine phosphorylation is selective for a subset of proteins.
The neuropeptide, bombesin, stimulates the tyrosine phosphorylation of multiple proteins including the broad bands of M r 110,000 -130,000 and 70,000 -80,000 (45). In contrast to PDGF, the effects of neuropeptides are mediated through Gprotein-coupled seven transmembrane receptors (53) that do not stimulate PI 3Ј-kinase activity in Swiss 3T3 cells (45,54,55). Hence, bombesin-stimulated tyrosine phosphorylation of proteins should not be inhibited by wortmannin. As shown in Fig. 1B, preincubation of cells with 30 nM wortmannin had no effect on the bombesin-stimulated tyrosine phosphorylation of the broad bands of M r 110,000 -130,000 and 70,000 -80,000. Indeed wortmannin up to 100 nM had no effect on bombesinstimulated tyrosine phosphorylation of proteins (results not shown). In addition, bombesin-stimulated formation of actin stress fibers is not affected by wortmannin pretreatment of cells (results not shown). Thus wortmannin distinguishes between PDGF and bombesin-stimulated tyrosine phosphorylation of specific proteins and cytoskeletal changes.
Effect of Wortmannin on PDGF-, Bombesin-, EGF-, Endothelin-, and PDB-stimulated Tyrosine Phosphorylation of p125 FAK -The cytosolic tyrosine kinase p125 FAK (20,21) has recently been identified as a prominent component of the M r 110,000 -130,000 band, which is tyrosine-phosphorylated in response to either PDGF at low concentrations or bombesin (18,19,44,56). The effect of wortmannin on the tyrosine phosphorylation of p125 FAK in response to either PDGF or bombesin was therefore determined.
Cells were preincubated with wortmannin (0 -40 nM) and then stimulated with either 3 ng/ml PDGF or 10 nM bombesin. The cell lysates were immunoprecipitated with an anti-Tyr(P) mAb and the immunoprecipitates analyzed by immunoblotting with an anti-p125 FAK mAb. The results shown in Fig. 2 (A and  B) indicate that wortmannin induced a dramatic dose-dependent inhibition of the tyrosine phosphorylation of p125 FAK in response to 3 ng/ml PDGF. In five independent experiments, pretreatment of cells with 30 nM wortmannin for 10 min produced an 85 Ϯ 6% inhibition of the tyrosine phosphorylation of p125 FAK in response to subsequent stimulation with PDGF (3 ng/ml). In contrast the tyrosine phosphorylation of p125 FAK stimulated with 10 nM bombesin was not inhibited by preincubation of the cells with wortmannin up to 40 nM.
To extend the results presented above, we examined the wortmannin sensitivity of p125 FAK tyrosine phosphorylation in response to EGF, endothelin, and PDB in Swiss 3T3 cells. Here we demonstrate, for the first time, that EGF stimulated an increase in the tyrosine phosphorylation of p125 FAK (3.6 Ϯ 1-fold, n ϭ 4) in Swiss 3T3 cells. EGF-stimulated phosphorylation of p125 FAK was inhibited by 85 Ϯ 12% when the cells were pretreated with 30 nM wortmannin (Fig. 2C), consistent with an inhibition of EGF-stimulated PI 3Ј-kinase activity by wortmannin (data not shown). In accord with the data obtained with PDGF, EGF stimulated the rapid formation of membrane ruffles in Swiss 3T3 cells, a response that was markedly inhibited by pretreatment of the cells with 30 nM wortmannin for 10 min (data not shown).
Endothelin, a neuropeptide that acts through a different G-protein-coupled receptor to bombesin also stimulates tyrosine phosphorylation of p125 FAK and formation of actin stress fibers in Swiss 3T3 cells (44,57). In contrast to the results obtained with PDGF and EGF, endothelin-stimulated p125 FAK tyrosine phosphorylation and actin reorganization were not affected by pretreatment of the cells with 30 nM wortmannin for 10 min (Fig. 2C and results not shown). In addition, the tyrosine phosphorylation of p125 FAK in response to direct activation of protein kinase C by PDB (56) was not affected by pretreatment of the cells with 30 nM wortmannin (Fig. 2C).
Effect of LY294002 on PDGF-and Bombesin-stimulated p125 FAK Tyrosine Phosphorylation-In order to substantiate the results obtained with wortmannin, we examined if a structurally unrelated compound, LY294002 (a flavonoid related to quercetin), which has been identified as a specific inhibitor of PI 3Ј-kinase with an IC 50 of 1-10 M (58), also inhibits PDGFstimulated p125 FAK tyrosine phosphorylation in a selective

FIG. 2. Effect of wortmannin on PDGF-, EGF-, bombesin-, endothelin-and PDB-stimulated tyrosine phosphorylation of p125 FAK .
A, quiescent Swiss 3T3 cells were preincubated for 10 min at 37°C with wortmannin (0 -40 nM). Cells were subsequently incubated with either 10 nM bombesin or 3 ng/ml PDGF for 10 min, lysed, and then immunoprecipitated with the anti-Tyr(P) mAb PY72. Immunoprecipitates were analyzed by immunoblotting with anti-p125 FAK mAb. B, the autoradiogram obtained in A was scanned with an LKB Ultrascan XL densitometer to quantify phosphoproteins in terms of peak area. Values correspond to the phosphorylation of p125 FAK expressed as a percentage of the maximal response in response to 10 nM bombesin (open circles) or 3 ng/ml PDGF (closed circles). C, quiescent Swiss 3T3 cells were preincubated for 10 min at 37°C with or without wortmannin (30 nM). Cells were subsequently incubated with either 3 ng/ml PDGF, 10 ng/ml EGF, 10 nM bombesin, 10 nM endothelin (End), or 100 nM PDB for 10 min and lysed. The cell lysates were then immunoprecipitated with the anti-Tyr(P) mAb PY72. Immunoprecipitates were analyzed by immunoblotting with anti-p125 FAK mAb.
manner. Fig. 3 shows that pretreatment with LY294002 inhibited p125 FAK tyrosine phosphorylation induced by PDGF in a dose-dependent fashion. At 10 M LY294002 inhibited PDGFstimulated p125 FAK tyrosine phosphorylation by 65%. In contrast, LY294002 had only a slight effect on p125 FAK tyrosine phosphorylation in response to bombesin (Fig. 3). In addition, anti-Tyr(P) immunoblots of anti-Tyr(P) immunoprecipitates of lysates of cells stimulated with 3 ng/ml PDGF showed that 10 M LY294002 inhibited the tyrosine phosphorylation of the M r 110,000 -130,000 and 70,000 -75,000 bands, but did not interfere with the autophosphorylation of the PDGF receptor or with the tyrosine phosphorylation of other substrates induced by 30 ng/ml PDGF (results not shown). In contrast, 10 M LY294002 had no effect on the tyrosine phosphorylation of the M r 110,000 -130,000 and 70,000 -75,000 bands stimulated by bombesin (results not shown). Thus LY294002, like wortmannin, inhibits PDGF-stimulated tyrosine phosphorylation of a subset of proteins including p125 FAK in a selective manner. The results presented in Figs. 1-3 suggest that PDGF stimulates tyrosine phosphorylation of p125 FAK through a PI 3Ј-kinasedependent pathway.
PDGF Receptors Lacking the Major PI 3Ј-Kinase Binding Sites Do Not Stimulate Tyrosine Phosphorylation of p125 FAK -Next we used an alternative system to examine further the role of PI 3Ј-kinase in PDGF-stimulated tyrosine phosphorylation of p125 FAK . Wild type PDGF ␤ receptors and mutated receptors with the tyrosine residues at positions 740 and 751 replaced by phenylalanine residues (Y740F/Y751F) were expressed in PAE cells that had been transfected with the appropriate cDNAs. As previously reported, both PDGF-stimulated recruitment of PI 3Ј-kinase to the receptor and accumulation of PtdIns (3,4,5)-P 3 were markedly reduced in PAE cells transfected with mutated receptors (38). Furthermore, PDGF does not stimulate either membrane ruffle formation or chemotaxis in PAE cells expressing Y740F/Y751F receptors (37,38). Here we examined whether PDGF stimulates p125 FAK tyrosine phosphorylation in PAE cells transfected with wild type and Y740F/Y751 PDGF receptors.
The results presented in Fig. 4 demonstrate that PDGF stimulates the tyrosine phosphorylation of p125 FAK in PAE cells expressing wild type PDGF-␤ receptors. In contrast, PDGF failed to stimulate an increase in the tyrosine phosphorylation of p125 FAK in PAE cells expressing Y740F/Y751F PDGF-␤ receptors (Fig. 4). This result suggests that PDGFstimulated tyrosine phosphorylation of p125 FAK is dependent on the interaction of the p85 regulatory subunit of PI 3Ј-kinase with the PDGF receptor chains. It is interesting to note that PAE cells expressing the Y740F/Y751F mutated PDGF-␤ receptors exhibit a higher basal level of p125 FAK tyrosine phosphorylation and that these cells have previously been demonstrated to contain higher levels of PI 3Ј-lipids (59). This result provides an independent line of evidence supporting the conclusion that PDGF-stimulated tyrosine phosphorylation of p125 FAK is dependent on the activation of PI 3Ј-kinase.
The Effect of Wortmannin on PDGF-and Bombesin-stimulated Tyrosine Phosphorylation of Paxillin-The focal adhesion-associated protein paxillin (25,26) has recently been identified as a major component of the M r 70,000 -80,000 band, tyrosine-phosphorylated in response to either PDGF at low concentrations or bombesin (18,60). As tyrosine phosphorylation of paxillin and p125 FAK has been shown to be coordinately regulated (61), we examined the effect of wortmannin on the tyrosine phosphorylation of paxillin in response to bombesin and PDGF. Quiescent Swiss 3T3 cells were preincubated with wortmannin (0 -30 nM) and then stimulated with either PDGF (3 ng/ml) or 10 nM bombesin. In other experiments the cells were preincubated with 30 nM wortmannin and then treated with increasing concentrations of PDGF. The cell lysates were immunoprecipitated with the anti-paxillin mAb 165 and the immunoprecipitates analyzed by immunoblotting with anti-Tyr(P) mAbs. The results shown in Fig. 5 clearly demonstrate that wortmannin inhibited the tyrosine phosphorylation of paxillin in response to various concentrations of PDGF. In contrast, wortmannin had no effect on bombesin-stimulated tyrosine phosphorylation of paxillin.
Effect of Rapamycin on PDGF-and Bombesin-stimulated Tyrosine Phosphorylation of p125 FAK -p70 S6K is a ubiquitous serine/threonine kinase that is activated by many mitogens through distinct signaling pathways (62). p70 S6K has been reported to be a molecular downstream target of PI 3Ј-kinase (32)(33)(34)(35). In view of the results presented here, it was important to clarify whether p70 S6K lies in the signaling pathway that mediates p125 FAK tyrosine phosphorylation in response to PDGF. Bombesin, in contrast, activates p70 S6K , presumably by a protein kinase C-dependent signaling pathway (62) and bombesin-stimulated tyrosine phosphorylation of p125 FAK has been shown to be protein kinase C-independent (56).
The immunosuppressant rapamycin is a selective inhibitor of p70 S6K activation in many cell types, including Swiss 3T3 cells (47). We therefore examined the effect of rapamycin on the PDGF-stimulated tyrosine phosphorylation of p125 FAK . Quiescent cultures of Swiss 3T3 cells were pretreated with either rapamycin (20 nM for 20 min) or wortmannin (30 nM for 10 min) and then stimulated with PDGF (3 ng/ml) or bombesin (10 nM) for 10 min. Cell lysates were analyzed by immunoblotting with an anti-p70 S6K polyclonal Ab and the phosphorylated form of p70 S6K (pp70 S6K ) was identified by its retarded mobility. The results shown in Fig. 6 demonstrate that pretreatment of cells with rapamycin completely inhibited both the PDGF-and bombesin-stimulated phoshorylation of p70 S6K . Cell lysates derived from parallel cultures of Swiss 3T3 cells were immunoprecipitated with an anti-Tyr(P) mAb and the immunoprecipitates analyzed by immunoblotting with an anti-p125 FAK mAb. As shown in Fig. 6, rapamycin had no effect on either PDGF-or bombesin-stimulated tyrosine phosphorylation of p125 FAK . Furthermore, pretreatment of Swiss 3T3 cells with rapamycin (20 nM) had no effect on PDGF-stimulated formation of mem-  (1-15 M). Cells were subsequently incubated with either 10 nM bombesin or 3 ng/ml PDGF for 10 min, lysed, and then immunoprecipitated with the anti-Tyr(P) mAb PY72. Immunoprecipitates were analyzed by immunoblotting with anti-p125 FAK mAb.
FIG. 4. PDGF-stimulated tyrosine phosphorylation of p125 FAK in PAE cell lines expressing wild type or mutant PDGFR-␤ receptors. PAE cells expressing wild type PDGFR-␤ or PDGFR-␤ Y740F/ Y751F mutant receptors molecules were incubated with PDGF (0 -30 ng/ml) for 10 min at 37°C. The cells were lysed, and the lysates were immunoprecipitated with the anti-Tyr(P) mAb PY72 and then analyzed by immunoblotting with anti-p125 FAK mAb. brane ruffles (data not shown). In contrast, wortmannin, which only slightly inhibited the PDGF-stimulated phoshorylation of p70 S6K at 30 nM, dramatically inhibited the PDGF-stimulated tyrosine phosphorylation of p125 FAK at this concentration (Fig.  6). Wortmannin had little effect on either the bombesin-stimulated phosphorylation of p70 S6K or tyrosine phosphorylation of p125 FAK . These results therefore dissociate p70 S6K activation from both PDGF-and bombesin-mediated p125 FAK tyrosine phosphorylation.

DISCUSSION
The signal transduction pathways implicated in PDGF-induced tyrosine phosphorylation of the focal adhesion proteins p125 FAK and paxillin had not been elucidated. The results presented here demonstrate that wortmannin at nanomolar concentrations dramatically inhibits PDGF-stimulated tyrosine phosphorylation of a subset of proteins, including p125 FAK and paxillin in Swiss 3T3 cells. We verified that at these concentrations wortmannin inhibits PDGF-stimulated PI 3Јkinase activity and the reorganization of actin into membrane ruffles in Swiss 3T3 cells. In contrast, wortmannin at nanomolar concentrations does not affect p125 FAK tyrosine phosphorylation induced by bombesin, which does not stimulate PI 3Ј-kinase in Swiss 3T3 cells (45,54,55). Furthermore, the PI 3Ј-kinase inhibitor LY294002, which is structurally unrelated to wortmannin, also inhibited PDGF-stimulated p125 FAK tyrosine phosphorylation in a selective manner. Utilizing PAE cells transfected with wild type and mutant PDGF ␤ receptors, we have shown that PDGF stimulates an increase in the tyrosine phosphorylation of p125 FAK in PAE cells transfected with wild type PDGF ␤ receptors, but not in PAE cells that were transfected with PDGF ␤ receptors lacking the PI 3Ј-kinase binding sites. Therefore, employing different experimental approaches, we were able to demonstrate that the activation of PI 3Ј-kinase is necessary for PDGF-stimulated tyrosine phosphorylation of p125 FAK . These results suggest that PI 3Ј-kinase lies upstream in the signal transduction pathway linking the PDGF receptor to tyrosine phosphorylation of p125 FAK and paxillin.
The results presented in this study should be distinguished from those recently published by Chen and Guan (19). These authors reported that a small fraction (5-6%) of total cellular PI 3Ј-kinase activity can be recovered from p125 FAK immunoprecipitates derived from lysates of NIH 3T3 cells treated with PDGF (19). 2 This association was maximal when the cells were stimulated with 25 ng/ml PDGF, a concentration of PDGF that did not stimulate the tyrosine phosphorylation of p125 FAK in these cells. Furthermore, cytochalasin D, a potent inhibitor of PDGF-stimulated p125 FAK tyrosine phosphorylation, did not affect the association (19). Thus, the association between p125 FAK and PI 3Ј-kinase seen in NIH 3T3 cells treated with high concentrations of PDGF 2 cannot account for the results presented in this study in which PI 3Ј-kinase was identified as an upstream element in a signal transduction pathway leading to p125 FAK and paxillin tyrosine phosphorylation in cells treated with low concentrations of PDGF.
p70 S6K has been identified as a molecular downstream target of PI 3Ј-kinase (32)(33)(34)(35). It was important, therefore to determine whether p70 S6K lies along a linear signaling pathway leading to p125 FAK tyrosine phosphorylation in PDGF-treated cells. Here we have demonstrated that inhibition of the phosphorylation and activation p70 S6K , with the immunosuppressant rapamycin (47), did not inhibit the PDGF-stimulated tyrosine phosphorylation of p125 FAK . In addition we found that rapamycin does not affect PDGF-stimulated membrane ruffling. We can therefore conclude that p70 S6K activation and p125 FAK tyrosine phosphorylation constitute two independent molecular downstream targets of PI 3Ј-kinase in PDGF-treated cells.
PDGF is a potent chemotactic agent for fibroblasts and other cell types (1,37). It has been shown that PI 3Ј-kinase activation is required for the formation of membrane ruffles and the stimulation of chemotaxis induced by growth factors (37,38,55). Recently it has been shown that the small G protein Rac lies downstream of PI 3Ј-kinase, and there is evidence that PI 3Ј-lipids may promote Rac-GTP formation (59). Furthermore, it has been demonstrated that there is a GTP-dependent and PDGF-stimulated association of Rac with PI 3Ј-kinase in Swiss 3T3 cells (63). Activated Rac has been demonstrated to direct the formation of membrane ruffles and the assembly of focal adhesions (64). In a previous study we demonstrated that the PDGF-stimulated tyrosine phosphorylation of p125 FAK and paxillin is dependent on the integrity of the actin cytoskeleton (18). The results presented here, demonstrating that PI 3Јkinase activation is required for PDGF-stimulated tyrosine phosphorylation of p125 FAK and paxillin and the formation of membrane ruffles, establish another link between the reorganization of the actin cytoskeleton and the tyrosine phosphorylation of these focal adhesion-associated proteins. Taken together, all these findings suggest that there is a linear signal 2 Using the anti-p125 mAb, 2A7, we have not been able to immunoprecipitate PI3Јkinase activity from Swiss 3T3 cells treated with PDGF (1-30 ng/ml). Furthermore, using agarose-bound GST-fusion proteins of the C-and N-terminal SH2 domains of the p85␣ subunit of PI 3Ј-kinase, we were unable to demonstrate any interaction with p125 FAK in control or PDGF-stimulated cells. These observations may reflect a difference between Swiss 3T3 cells and NIH 3T3 cells.
FIG. 5. Effect of wortmannin on PDGF-and bombesin-stimulated tyrosine phosphorylation of paxillin. Upper, quiescent Swiss 3T3 cells were preincubated for 10 min at 37°C with 0, 20 or 30 nM wortmannin and subsequently with 3 ng/ml PDGF or with 10 nM bombesin (Bom), as indicated. Lower, other cell cultures were preincubated for 10 min at 37°C with (ϩ) or without (Ϫ) 30 nM wortmannin, and subsequently incubated with PDGF (0 -10 ng/ml) for 10 min. Cell lysates were immunoprecipitated with the anti-paxillin mAb 165 and then analyzed by immunoblotting with a mixture of anti-Tyr(P) mAbs.
FIG. 6. Effect of rapamycin and wortmannin on PDGF-and bombesin-stimulated phosphorylation of p70 S6K and the tyrosine phosphorylation of p125 FAK . Two parallel sets of cells were preincubated in the presence (ϩ) or absence (Ϫ) of either 20 nM rapamycin (R) for 20 min or 30 nM wortmannin (W) for 10 min, the cells were subsequently incubated for 10 min with either 3 ng/ml PDGF or 10 nM bombesin and then lysed. The whole cell lysates from one set of cells were analyzed directly by immunoblotting with the anti-p70 S6K rabbit polyclonal Ab. The lysates from the other set of cells were immunoprecipitated with anti-Tyr(P) mAb and then analyzed by immunoblotting with anti-p125 FAK mAb. transduction pathway whereby ligation of the PDGF receptor activates PI 3Ј-kinase and thereby stimulates Rac-GTP formation. Activated Rac induces the formation of focal contacts, reorganization of the actin cytoskeleton, and the tyrosine phosphophorylation of p125 FAK and paxillin. This signal transduction pathway could function in the regulation of chemotaxis. The recent demonstration that p125 FAK -deficient cells do not display polar migratory shape and exhibit a striking reduction in motility (65) is in agreement with this interpretation.
Activation of PI 3Ј-kinase in response to PDGF is thought to occur as a result of the association of the p85 regulatory subunit of PI 3Ј-kinase with the tyrosine-phosphorylated PDGF receptor chains (12,13). The recent demonstration that Ras may interact with and activate PI 3Ј-kinase suggests another mechanism for PI 3Ј-kinase activation in response to either PDGF or EGF in at least some cell types (66,67). The relative contribution of these two pathways leading to PI 3Ј-kinase activation and hence p125 FAK tyrosine phosphorylation in response to PDGF or EGF warrants further experimental work.
The tyrosine phosphorylation of p125 FAK and paxillin induced by bombesin and endothelin is dependent on the integrity of the actin cytoskeleton (56,60) and the activity of the small GTP-binding protein Rho (57). Here we demonstrate that the tyrosine phosphorylation of p125 FAK in response to either bombesin or endothelin is not prevented by wortmannin at concentrations that virtually abolished the tyrosine phosphorylation of p125 FAK and paxillin in response to PDGF and EGF. Similar results were obtained when LY294002 was used instead of wortmannin. An important implication of these results, therefore, is that there is a PI 3Ј-kinase-dependent and PI 3Ј-kinase-independent signal transduction pathway stimulating the tyrosine phosphorylation of p125 FAK and paxillin in the same cells.