Sphingosylphosphorylcholine rapidly induces tyrosine phosphorylation of p125FAK and paxillin, rearrangement of the actin cytoskeleton and focal contact assembly. Requirement of p21rho in the signaling pathway.

Sphingosylphosphorylcholine (SPC), a potent mitogen for Swiss 3T3 cells, rapidly induced tyrosine phosphorylation of multiple substrates including bands of Mr 110,000-130,000 and Mr 70,000-80,000 in Swiss 3T3 cells. Focal adhesion kinase (p125FAK) and paxillin were identified as prominent substrates for SPC-stimulated tyrosine phosphorylation. An increase in tyrosine phosphorylation of p125FAK was detected as soon as 30 s after SPC stimulation, reaching a maximum after 2.5 min. SPC induced tyrosine phosphorylation of p125FAK in a concentration-dependent fashion; a half-maximum effect occurred at 250 nM. Tyrosine phosphorylation of p125FAK induced by SPC could be dissociated from both protein kinase C activation and Ca2+ mobilization from intracellular stores. SPC induced a unique pattern of reorganization of the actin cytoskeleton with a rapid appearance of actin microspikes at the plasma membrane that was followed by the formation of actin stress fibers. This pattern of cytoskeletal changes was clearly distinguishable from that induced by bombesin and 1oleoyl-lysophosphatidic acid. Formation of microspikes and actin stress fibers were accompanied by striking assembly of focal adhesion plaques. Cytochalasin D, which disrupts the network of actin microfilaments, completely prevented SPC-induced tyrosine phosphorylation of p125FAK. In addition, tyrosine phosphorylation of p125FAK was markedly inhibited in the presence of platelet-derived growth factor at a concentration (30 ng/ml) that disrupts actin stress fibers. Finally, microinjection of Clostridium botulinum C3 exoenzyme, which inactivates p21rho, prevented SPC-induced formation of actin stress fibers, focal adhesion assembly, and tyrosine phosphorylation. Thus, p21rho is upstream of both cytoskeletal reorganization and tyrosine phosphorylation in SPC-treated cells.


Sphingosylphosphorylcholine (SPC), a potent mitogen
for Swiss 3T3 cells, rapidly induced tyrosine phosphorylation of multiple substrates including bands of M r 110,000 -130,000 and M r 70,000 -80,000 in Swiss 3T3 cells. Focal adhesion kinase (p125 FAK ) and paxillin were identified as prominent substrates for SPC-stimulated tyrosine phosphorylation. An increase in tyrosine phosphorylation of p125 FAK was detected as soon as 30 s after SPC stimulation, reaching a maximum after 2.5 min. SPC induced tyrosine phosphorylation of p125 FAK in a concentration-dependent fashion; a half-maximum effect occurred at 250 nM. Tyrosine phosphorylation of p125 FAK induced by SPC could be dissociated from both protein kinase C activation and Ca 2؉ mobilization from intracellular stores. SPC induced a unique pattern of reorganization of the actin cytoskeleton with a rapid appearance of actin microspikes at the plasma membrane that was followed by the formation of actin stress fibers. This pattern of cytoskeletal changes was clearly distinguishable from that induced by bombesin and 1oleoyl-lysophosphatidic acid. Formation of microspikes and actin stress fibers were accompanied by striking assembly of focal adhesion plaques. Cytochalasin D, which disrupts the network of actin microfilaments, completely prevented SPC-induced tyrosine phosphorylation of p125 FAK . In addition, tyrosine phosphorylation of p125 FAK was markedly inhibited in the presence of platelet-derived growth factor at a concentration (30 ng/ml) that disrupts actin stress fibers. Finally, microinjection of Clostridium botulinum C3 exoenzyme, which inactivates p21 rho , prevented SPC-induced formation of actin stress fibers, focal adhesion assembly, and tyrosine phosphorylation. Thus, p21 rho is upstream of both cytoskeletal reorganization and tyrosine phosphorylation in SPC-treated cells.
Sphingosylphosphorylcholine (SPC) 1 acts as a potent mitogen for Swiss 3T3 cells in the absence of any other exogenously added growth factor (1,2), but the molecular basis for this effect is largely unknown. The best characterized signaling event in response to SPC is the release of Ca 2ϩ from intracellular stores in a variety of cell lines (2)(3)(4)(5). However, it appeared unlikely that all cellular responses elicited by SPC are mediated by Ca 2ϩ mobilization. In the accompanying paper (6), we demonstrated that SPC induces transient activation of MAPK and p90 rsk by a pathway dependent on the activity of PKC and a pertussis toxin-sensitive G i protein. In this paper, we identify further cellular and molecular responses to exogenously added SPC to elucidate the mechanisms by which lysosphingolipids modulate cell function.
Tyrosine phosphorylation of the cytosolic protein kinase p125 FAK and of the cytoskeletal associated protein paxillin has recently been identified as an early event in the action of diverse signaling molecules that mediate cell growth and differentiation, including mitogenic neuropeptides (7)(8)(9), growth factors such as PDGF (10), the bioactive lipid LPA (11)(12)(13), sphingosine (14), extracellular matrix proteins (15)(16)(17)(18)(19), and transforming variants of pp60 src (16,20). The increases in p125 FAK and paxillin tyrosine phosphorylation are accompanied by profound alterations in the organization of the actin cytoskeleton and in the assembly of focal adhesions (10,13,14,21,22), the distinct areas of the plasma membrane where p125 FAK and paxillin are localized (23)(24)(25)(26). Recently, the small G protein p21 rho , a member of the Ras superfamily of small GTP binding proteins, has been implicated in mitogen-stimulated formation of focal adhesions and actin stress fibers as well as tyrosine phosphorylation of p125 FAK and paxillin (21,(27)(28)(29). The effects, if any, of exogenously added SPC on tyrosine phosphorylation, the organization of the actin cytoskeleton and focal adhesion assembly, and the potential involvement of small G proteins such as p21 rho in this signaling pathway remain unknown.
In this paper, we demonstrate that SPC rapidly stimulates tyrosine phosphorylation of multiple proteins including p125 FAK and paxillin in Swiss 3T3 cells and induces a unique pattern of actin organization, which was accompanied by the formation of focal contacts. Microinjection of C. botulinum C3 exoenzyme, which ADP-ribosylates and inactivates p21 rho , blocked the changes in the organization of the actin cytoskeleton, the assembly of focal contacts, and tyrosine phosphorylation in response to SPC.

EXPERIMENTAL PROCEDURES
Cell Culture and Down-regulation of PKC-Cell culture of Swiss 3T3 cells and down-regulation of PKC by prolonged pretreatment with PDB were performed as described in the accompanying paper (6).
Immunoprecipitation-Quiescent cultures of Swiss 3T3 cells (1-2 ϫ 10 6 ) were washed twice with DMEM, treated with SPC or other factors in 10 ml of DMEM for the times indicated, and lysed at 4°C in 1 ml of a solution containing 10 mM Tris/HCl, pH 7.6, 5 mM EDTA, 50 mM NaCl, * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 30 mM sodium pyrophosphate, 50 mM NaF, 100 M Na 3 VO 4 , and 1% Triton X-100 (lysis buffer). Proteins were immunoprecipitated at 4°C for 14 h with anti-mouse IgG-agarose-linked mAbs directed against phosphotyrosine, paxillin, or p125 FAK as indicated. Immunoprecipitates were washed three times with lysis buffer and extracted for 10 min at 95°C 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 analyzed by SDS-PAGE.
Western Blotting-Treatment of quiescent cultures of cells with factors, cell lysis, and immunoprecipitations were performed as described above. After SDS-PAGE, proteins were transferred to Immobilon membranes. Membranes were blocked using 5% non-fat dried milk in PBS, pH 7.2, and incubated for 2 h at 22°C with a mixture of Py20 and 4G10 anti-Tyr(P) mAbs (1 g/ml of each). Immunoreactive bands were visualized by autoradiography using 125 I-labeled sheep anti-mouse IgG (1:1000) followed by autoradiography. Autoradiograms were scanned using an LKB Ultrascan XL densitometer, and labeled bands were quantified using the Ultrascan XL internal integrator. The values expressed represent percentages of the maximum increase in tyrosine phosphorylation above control values.
Measurement of [Ca 2ϩ ] i -Confluent and quiescent Swiss 3T3 cells in 100-mm dishes were washed twice with DMEM and incubated for 10 min in DMEM containing 1 M fura-2-tetraacetoxymethyl ester. After this time, the cells were washed twice with PBS at 37°C and once with electrolyte solution, which contained 120 mM NaCl, 5 mM KCl, 1.8 mM CaCl 2 , 0.9 mM MgCl 2 , 25 mM glucose, 16 mM HEPES, 6 mM Tris/HCl, and a mixture of amino acids at the same concentrations as are present in DMEM, pH 7.2. The cells were then suspended in 2 ml of electrolyte solution by gentle scraping and transferred to a 1-cm 2 quartz cuvette, which was placed in a Perkin-Elmer LS-5 luminescence spectrophotometer. The cell suspension was stirred continuously at 37°C, and fluorescence was monitored at an excitation wavelength of 336 nm and emission wavelength of 510 nm. Factors were added to the cell suspension while fluorescence was being monitored as indicated. [Ca 2ϩ ] i was determined as described previously (30).
Immunostaining of Cells-Quiescent Swiss 3T3 cells were washed twice with DMEM and incubated for the indicated time at 37°C with the indicated concentration of SPC or other factors. Thereafter, for staining of actin, cells were washed twice with PBS, fixed in 4% paraformaldehyde in PBS for 30 min at 4°C, and permeabilized with PBS containing 0.2% Triton X-100 for 8 min at room temperature. The cells were then incubated with TRITC-conjugated phalloidin (0.25 g/ ml) in PBS for 10 min at room temperature and visualized utilizing a Zeiss Axiophot immunofluorescence microscope. In experiments in which quiescent Swiss 3T3 cells were labeled with TRITC-conjugated phalloidin and anti-vinculin or anti-Tyr(P) antibody, after fixing and permeabilizing the cells as described above, TRITC-conjugated phalloidin (0.25 g/ml) and anti-vinculin antibody (dilution 1:100) or anti-Tyr(P) mAb 4G10 (dilution 1:300) were added together to the cells for 30 min at room temperature. Cells were subsequently washed three times in PBS and then incubated with FITC-labeled anti-mouse IgG as a second antibody at a dilution of 1:100 for another 30 min at room temperature. In the microinjection experiments shown in Fig. 8, triple labeling was used. Cells injected with C3 exoenzyme and rabbit IgG were fixed and permeabilized as described above. TRITC-conjugated phalloidin (0.25 g/ml) and anti-vinculin antibody (dilution 1:100) or anti-Tyr(P) mAb 4G10 (dilution 1:300) were added together to the cells for 30 min at room temperature. Cells were subsequently washed three times in PBS and then incubated with FITC-labeled anti-rabbit IgG to control for microinjection efficiency together with Cy5-conjugated affinity pure sheep anti-mouse IgG to label either vinculin or Tyr(P) at a dilution of 1:100 each for another 30 min at room temperature.
Confocal Microscopy-Confocal imaging was performed using a Bio-Rad MRC-1000 laser scanning head fitted onto a Diaphot 200 microscope. A 60 ϫ N. A. 1.4 planapochromat oil immersion lens (Nikon) and the dicroic filter blocks T1 and E2 were used. Images were collected sequentially using a krypton/argon mixed gas laser (Bio-Rad) with excitation filters at 488, 568, and 647 nm for FITC, TRITC, and Cy5 fluorochromes, respectively, and emission filters at 622, 605, and 680 nm, respectively. Correction of images for bleed through and other processing was carried out using the COMOS program (Bio-Rad) run on a QE 486 33C computer. Optical sections were recorded at 0.5 m. Final images were photographed directly from the VDU screen or processed on a Silicon Graphics Indigo XS 24 workstation using a Sony digital color printer.
Microinjection-For microinjection experiments, Swiss 3T3 cells were plated in 35-mm Nunc Petri dishes at 10 5 cells/dish in DMEM containing 10% fetal bovine serum and used after 6 -8 days when the cells were confluent and quiescent. To facilitate localization of microinjected cells, a circle was scored on the bottom of the dishes, and in each experiment approximately 60 cells in the circle were microinjected. The efficiency of injection was determined by coinjecting rabbit IgG at 0.5 mg/ml followed by staining of the cells with FITC-linked anti-rabbit IgG. The same batch of C3 exoenzyme was used in all experiments described.
Materials-SPC, LPA, PDB, bombesin, cytochalasin D, TRITC-conjugated phalloidin, monoclonal anti-vinculin antibody, and FITC-linked anti-rabbit IgG were obtained from Sigma. Cy5-conjugated affinity pure sheep anti-mouse IgG and chrome pure rabbit IgG were from Jackson ImmunoResearch Laboratories Inc. The specific PKC inhibitor GF 109203X and thapsigargin were obtained from Calbiochem-Novabiochem Ltd. (Nottingham, UK). Anti-mouse IgG-agarose-linked anti-Tyr(P) mAb clone Py72 was obtained from the hybridoma development unit (Imperial Cancer Research Fund, London, UK). Py20 anti-Tyr(P) mAb and the mAb directed against paxillin (mAb 165) were from ICN (High Wycombe, UK). 4G10 anti-Tyr(P) mAb and mAb 2A7 directed against p125 FAK were from TCS Biologicals Ltd. (Buckingham, UK). Anti-p125 FAK mAb for Western blotting was obtained from AFFINITI Research Products Ltd. (Nottingham, UK). 125 I-Sheep anti-mouse IgG (15 mCi/mg) and recombinant BB homodimer PDGF were from Amersham. Recombinant C. botulinum exoenzyme C3 was the kind gift of Prof. Shuh Narumiya (Dept. of Pharmacology, Faculty of Medicine, Kyoto University, Japan). All other reagents used were of the purest grade available.

SPC Induces Tyrosine Phosphorylation of Several Proteins
Including p125 FAK and Paxillin-To examine the effect of SPC on protein tyrosine phosphorylation, quiescent Swiss 3T3 cells were incubated with 5 M SPC for 5 min and lysed. Lysates of the treated cells were incubated with anti-Tyr(P) mAb, and the immunoprecipitates were further analyzed by Western blotting with a mixture of anti-Tyr(P) mAbs. As shown in Fig. 1, upper panel, left, SPC stimulated tyrosine phosphorylation of multiple components including bands migrating with an apparent M r of 110,000 -130,000 and 70,000 -80,000. The pattern of tyrosine-phosphorylated proteins induced by SPC was comparable to that obtained with 10 nM bombesin ( The cytosolic tyrosine kinase p125 FAK (23,24) and the cytoskeleton-associated protein paxillin (25,26) have been identified as prominent tyrosine-phosphorylated proteins in bombesin-treated Swiss 3T3 cells (7)(8)(9). As SPC elicited a pattern of tyrosine-phosphorylated bands similar to that induced by bombesin ( Fig. 1, upper panel, left), we determined whether p125 FAK and paxillin were also substrates for SPC-mediated tyrosine phosphorylation. Lysates of quiescent Swiss 3T3 cells incubated with 5 M SPC for 5 min were immunoprecipitated with mAb 2A7 directed against p125 FAK or mAb 165 directed against paxillin and further analyzed by immunoblotting with a mixture of anti-Tyr(P) mAbs. Fig. 2, top panel, left and right, shows that SPC markedly induced tyrosine phosphorylation of p125 FAK and paxillin, respectively. Thus, p125 FAK is a component of the broad M r 110,000 -130,000 band, and paxillin is a component of the diffuse M r 70,000 -80,000 band of tyrosinephosphorylated proteins in SPC-treated Swiss 3T3 cells. The degree of tyrosine phosphorylation of these specific substrates in response to 5 M SPC was comparable to that induced by 10 nM bombesin in parallel cultures (Fig. 2, top panel, left and right). As shown in Fig. 2, middle panel, SPC induced tyrosine phosphorylation of p125 FAK in a concentration-dependent man-ner; half-maximum and maximum effects were achieved at 0.25 and 1 M SPC, respectively. The kinetics of SPC-induced tyrosine phosphorylation of p125 FAK is shown in Fig. 2, lower panel. An increase in tyrosine phosphorylation of p125 FAK was first detectable after 0.5 min and reached its maximum after 2.5 min of incubation in the presence of SPC.
Role of Ca 2ϩ , G i , and PKC in SPC-stimulated Tyrosine Phosphorylation of p125 FAK -SPC is known to induce rapid mobilization of Ca 2ϩ from intracellular stores leading to an increase in [Ca 2ϩ ] i (2)(3)(4)(5). Changes in [Ca 2ϩ ] i have been implicated in tyrosine phosphorylation of a M r 125,000 band induced by angiotensin II or GP IIbIIIa (31). To examine whether SPCinduced tyrosine phosphorylation of p125 FAK could also be dependent on [Ca 2ϩ ] i , quiescent Swiss 3T3 cells were pretreated with the tumor promoter thapsigargin, an agent that specifically inhibits the endoplasmic reticulum Ca 2ϩ -ATPase and thereby depletes Ca 2ϩ from intracellular stores (32). Treatment of cells with 30 nM thapsigargin for 30 min did not inhibit tyrosine phosphorylation of p125 FAK stimulated by SPC (Fig. 3, upper panel) but blocked the increase in [Ca 2ϩ ] i by SPC (data not shown). To examine whether the extracellular [Ca 2ϩ ] could be important for SPC-induced tyrosine phosphorylation of p125 FAK , extracellular Ca 2ϩ was chelated by addition of 3 mM EGTA 30 min before stimulation with SPC. This procedure also failed to attenuate tyrosine phosphorylation of p125 FAK in response to SPC. Thus, SPC stimulates tyrosine phosphorylation of p125 FAK independently of either intracellular Ca 2ϩ release or Ca 2ϩ influx. Treatment of Swiss 3T3 cells with 30 nM thapsigargin for 30 min also did not inhibit SPC-induced tyrosine phosphorylation of the M r 70,000 -80,000 band, which corresponds to paxillin (data not shown).
SPC induces activation of MAPK by a pertussis toxin-sensitive mechanism, suggesting the involvement of a G protein of the G i /G o subfamily in the signaling pathway (6). To examine whether SPC stimulates p125 FAK tyrosine phosphorylation via G i , quiescent Swiss 3T3 cells were pretreated with various concentrations of pertussis toxin for 3 h and subsequently challenged with 5 M SPC for 5 min. Pertussis toxin at concentrations between 0.1 and 100 ng/ml did not affect tyrosine phosphorylation of p125 FAK in response to SPC (data not shown).
In the accompanying paper (6), we show that SPC stimulates activation of PKC in Swiss 3T3 cells. As activation of PKC is also a potential pathway leading to tyrosine phosphorylation of p125 FAK and paxillin (8), we examined the role of PKC in SPC-induced p125 FAK and paxillin tyrosine phosphorylation. PKC was either selectively inhibited by pretreatment of quiescent Swiss 3T3 cells for 1 h with the bisindolylmaleimide GF 109203X at 3.5 M (33) or down-regulated by prolonged exposure (48 h) to 800 nM PDB (34). As shown in Fig. 3, middle panel, both treatments completely blocked stimulation of p125 FAK tyrosine phosphorylation by 200 nM PDB. In contrast, inhibition or down-regulation of PKC only slightly reduced p125 FAK tyrosine phosphorylation in response to 5 M SPC (Fig. 3, middle panel). Thus, SPC induces p125 FAK tyrosine phosphorylation largely through a PKC-independent pathway.
Paxillin contains several potential target sites for phosphorylation by PKC (35). Tyrosine phosphorylation of paxillin in response to bombesin, LPA, or phorbol esters is accompanied by a mobility shift that results in the appearance of slower migrating forms of this protein, a process mediated by PKC (9). SPC at 5 M also induced this mobility shift of paxillin (Fig. 3,  lower panel). Treatment of Swiss 3T3 cells with GF 109203X did not affect the degree of tyrosine phosphorylation of paxillin induced by SPC but prevented the appearance of the slower migrating forms of paxillin. Interestingly, the sphingolipid breakdown product sphingosine at 12.5 M induced tyrosine phosphorylation of paxillin to a lesser degree than SPC, and no mobility shift of paxillin could be detected (Fig. 3, lower panel). This is probably due to the fact that sphingosine, in contrast to SPC, does not strongly activate PKC (14).
In conclusion, the results depicted in Fig. 3 indicate that SPC induces tyrosine phosphorylation of p125 FAK and paxillin through a pathway largely independent of Ca 2ϩ mobilization, a G i protein, or PKC activation.
SPC Induces a Unique Pattern of Organization of the Actin Cytoskeleton-As p125 FAK and paxillin are localized at focal adhesion plaques, the distinct sites on the plasma membrane where actin stress fibers emanate, we examined whether SPC modulates actin cytoskeletal organization and focal adhesion assembly. As shown in Fig. 4A, quiescent Swiss 3T3 cells contain only very few actin stress fibers. Incubation of cells with SPC for various times induced dramatic changes in the organization of the actin cytoskeleton. Within 1 min, SPC induced actin reorganization with loss of punctate actin and an increase in polymerized actin as cell surface protrusions. These protrusions formed microspikes. The reorganization of actin into microspikes peaked within 3-5 min of addition of SPC. Interestingly, 7-10 min after addition of SPC the pattern of actin organization changed, and a diffuse increase in actin fibers throughout the cytoplasm was visible. By 20 min, the cells contained numerous densely packed stress fibers, whereas the microspikes were markedly reduced (Fig. 4A). All of these effects were independent of an increase in [Ca 2ϩ ] i as treatment of Swiss 3T3 cells with 30 nM thapsigargin for 30 min did not prevent the subsequent changes in the actin cytoskeleton in response to SPC (data not shown).
Both the neuropeptide bombesin and the bioactive lipid LPA stimulate the reorganization of actin with a time course comparable to that of SPC. However, these compounds primarily induce organization of actin into stress fibers visible as early as 1 min after addition of the factors but not into microspikes (21). 2 Therefore, the pattern of the changes in actin organization induced by SPC seemed to be unique. To further substantiate this observation, we compared the early effects of 5 M SPC, 10 nM bombesin, and 2 M LPA on the actin cytoskeleton in parallel cultures. As shown in Fig. 4B, SPC induced the typical microspike pattern of actin arrangement. In contrast, bombesin induced actin stress fiber formation and membrane ruffling, whereas LPA predominantly induced actin stress fiber formation in accordance with previous observations (13,21). Therefore, the pattern of actin rearrangements in response to SPC showing first microspikes and later stress fibers is clearly distinguishable from that induced by other factors that cause rapid changes in the organization of the actin cytoskeleton in Swiss 3T3 cells.

SPC-induced Assembly of Focal Contacts Is Related to the Formation of Both Actin Microspikes and Actin Stress Fibers-
Focal adhesions are subcellular structures that are formed at regions of close contacts between cells and their underlying substratum. Several proteins are specifically localized in focal adhesions including vinculin, paxillin, talin, and ␣-actinin (36). To assess the effect of SPC on the assembly of focal adhesions, vinculin, a major focal adhesion-associated protein, was localized by immunofluorescence with monoclonal anti-vinculin antibody. Surprisingly, both the reorganization of actin into microspikes and the later occurrence of actin stress fibers were accompanied by the formation of focal contacts as seen by the time-dependent aggregation of vinculin into focal adhesions starting as early as 1 min after addition of SPC to the cells (Fig. 5A).
To corroborate these findings and clarify the spatial relationship between microspikes and focal contacts in response to SPC, we used double labeling to visualize actin and vinculin in the same cells. Addition of 5 M SPC to quiescent Swiss 3T3 2 T. Seufferlein and E. Rozengurt, unpublished observations.

FIG. 3. Role of Ca 2؉ and PKC activity in SPC-induced p125 FAK tyrosine phosphorylation.
Upper panel, quiescent Swiss 3T3 cells were incubated with 30 nM thapsigargin (Thapsigargin, ϩ) or 3 mM EGTA (EGTA, ϩ) for 30 min as indicated. Control cells received an equivalent amount of solvent (Ϫ). Cells were subsequently stimulated with 5 M SPC for 5 min and lysed, and the lysates were immunoprecipitated with anti-Tyr(P) mAb followed by Western blotting with anti-p125 FAK mAb. Shown is a representative experiment of three independent experiments. Middle panel, quiescent Swiss 3T3 cells were pretreated for 48 h with 800 nM PDB (PDB pre, ϩ), incubated for 1 h with GF 109203X (GF, ϩ), or received an equivalent amount of solvent (Ϫ) and then were stimulated with 200 nM PDB or 5 M SPC. Cells were lysed, and lysates were immunoprecipitated with mAb 2A7 directed against p125 FAK and further analyzed by Western blotting with anti-Tyr(P) mAbs. Shown is a representative experiment of three independent experiments. The position of p125 FAK is indicated by an arrow. The increase in tyrosine phosphorylation of p125 FAK was quantified by scanning densitometry. Values shown are the mean Ϯ S.E. of three independent experiments and are expressed as percentages of the maximum increase in tyrosine phosphorylation of p125 FAK above unstimulated control values. PDB or GF 109203X pretreatment of Swiss 3T3 cells reduced tyrosine phosphorylation of p125 FAK in response to PDB to the same level obtained in PDB or GF 109203X pretreated but unstimulated cells. These controls were therefore omitted for clarity. Lower panel, quiescent Swiss 3T3 cells were pretreated for 1 h with of 3.5 M GF 109203X (GF, ϩ) or received an equivalent amount of solvent (Ϫ). Cells were then treated for 30 min with 5 M SPC, 10 nM bombesin (Bom) or 12.5 M sphingosine (Sph) and were subsequently lysed; the lysates were immunoprecipitated with mAb 165 directed against paxillin and further analyzed by Western blotting using anti-Tyr(P) mAbs. Shown is a representative experiment of two independent experiments. cells for 5 min led to recruitment of actin into microspikes at the plasma membrane, giving the cells a "hedgehog"-like appearance (Fig. 5B, left picture). Vinculin staining of focal contacts was clearly visible at the ends of these microspikes (Fig.  5B, right picture). Thus, in SPC-stimulated cells, focal adhesion assembly with localization of vinculin to focal contacts clearly occurred before formation of actin stress fibers and was also related to a different form of actin organization, namely the formation of actin microspikes.
The Integrity of the Actin Cytoskeleton Is Necessary for SPCstimulated Tyrosine Phosphorylation-The effects of SPC on actin stress fiber formation and focal adhesion assembly depicted in Figs. 4 and 5 and the strong correlation of the kinetics of cytoskeletal changes and tyrosine phosphorylation in response to SPC prompted us to examine whether the integrity of the actin cytoskeleton was necessary for SPC-induced tyrosine phosphorylation. As shown in Fig. 6A, cytochalasin D inhibited SPC-induced tyrosine phosphorylation in a concentration-dependent manner, completely preventing tyrosine phosphorylation of p125 FAK at a concentration of 1.2 M.
The inhibitory effect of cytochalasin D on SPC-induced tyrosine phosphorylation was specific, as cytochalasin D at the concentrations used profoundly disrupted the network of actin filaments and focal contacts in Swiss 3T3 cells (data not shown) but did not interfere with SPC-induced Ca 2ϩ mobilization (Fig.   6B) or with SPC-stimulated MAPK activation as shown in the accompanying paper (6). This suggests that SPC induces tyrosine phosphorylation of p125 FAK and MAPK activation by clearly distinct mechanisms.

Effect of High Concentrations of PDGF on SPC-induced Actin Stress Fiber Formation and p125 FAK Tyrosine
Phosphorylation-Recent data from our laboratory have shown that PDGF, at a high concentration (30 ng/ml), completely abolishes bombesin-, LPA-, and sphingosine-induced actin stress fiber formation (10,13,14). As SPC induced a unique pattern of cytoskeletal changes (Figs. 4 and 5), we examined the effect of low (5 ng/ml) and high (30 ng/ml) concentrations of PDGF on actin microspikes and stress fibers induced by SPC. As shown in Fig.  7A, the marked increase in actin stress fiber formation induced by 5 M SPC was unaffected by PDGF at 5 ng/ml, but it was completely prevented when cells were incubated with both 5 M SPC and 30 ng/ml PDGF for 20 min. Assembly of focal contacts in response to SPC was also reduced in the presence of PDGF at 30 ng/ml (data not shown). However, organization of actin into microspikes was still visible in these cells, resembling the pattern of actin organization at earlier times (e.g. 3 min) of incubation with SPC.
The results shown in Fig. 6 demonstrated that SPC stimulated p125 FAK tyrosine phosphorylation by a mechanism dependent on the integrity of the actin cytoskeleton. In view of the results depicted in Fig. 7A, we examined whether SPCstimulated tyrosine phosphorylation of p125 FAK could also be affected by high concentrations of PDGF. As shown in Fig. 7B, PDGF at 30 ng/ml markedly reduced tyrosine phosphorylation of p125 FAK induced by 5 M SPC in agreement with its effects on SPC-induced actin reorganization.

Effects of C3 Exoenzyme on SPC-induced Actin Stress Fiber Formation, Focal Contact Assembly, and Protein Tyrosine
Phosphorylation-Recently, the small G protein p21 rho has been implicated in the formation of actin stress fibers and focal contacts in response to bombesin and LPA in Swiss 3T3 cells (21). The function of p21 rho is specifically impaired by treatment with C. botulinum C3 exoenzyme, which ADP-ribosylates the Asn 41 of p21 rho and thereby prevents its interaction with downstream targets (37). To examine whether SPC-induced changes in actin cytoskeleton and focal contact assembly were dependent on p21 rho , confluent and quiescent Swiss 3T3 cells were microinjected with 100 g/ml C3 exoenzyme and subsequently stimulated with 5 M SPC for various times. The efficiency of injection was determined by coinjection of rabbit IgG at 0.5 mg/ml together with C3 exoenzyme followed by immunostaining with FITC-linked anti-rabbit IgG. Fig. 8, C and F, shows the cells injected with C3 exoenzyme and rabbit IgG. The formation of actin stress fibers after 20 min of incubation with 5 M SPC was completely prevented in cells microinjected with C3 exoenzyme. The injected cells appeared contracted, suggesting a complete breakdown of the actin cytoskeleton as compared to the surrounding cells not injected (Fig. 8, A and D). C3 exoenzyme also prevented the formation of focal adhesions in response to SPC as shown by additional staining of the cells with anti-vinculin mAb (Fig. 8B). C3 exoenzyme was equally efficient at concentrations between 20 and 400 g/ml. Thus, p21 rho is required for the organization of the actin stress fibers and focal adhesion assembly induced by SPC. Swiss 3T3 cells injected with rabbit IgG alone exhibited the typical pattern of actin organization and focal contact assembly in response to SPC, demonstrating that microinjection itself did not attenuate SPC-induced changes in the actin cytoskeleton and focal adhesion formation (data not shown).
To examine whether C3 exoenzyme could also prevent tyrosine phosphorylation in response to SPC, microinjected cells were stained for actin, Tyr(P), and rabbit IgG as described under "Experimental Procedures." In control cells stimulated with 5 M SPC for 20 min, anti-Tyr(P) staining was visible at the ends of actin stress fibers in focal contacts (Fig. 8, D and E). In microinjected cells, this pattern of staining was not detect- able, indicating that C3 exoenzyme inhibited tyrosine phosphorylation in response to SPC (Fig. 8, D-F). Thus, p21 rho is an upstream regulator of both cytoskeletal changes and tyrosine phosphorylation in response to SPC.

DISCUSSION
The results presented here demonstrate that SPC induces tyrosine phosphorylation of multiple substrates in Swiss 3T3 cells. We identified two substrates that were phosphorylated on tyrosine residues in response to SPC, p125 FAK , and paxillin. p125 FAK is a cytosolic tyrosine kinase that lacks SH2 and SH3 domains but associates with other proteins including v-Src and paxillin (38). The recent molecular cloning and sequencing of paxillin revealed that this protein contains multiple domains that can participate in protein-protein interactions (35,39). A coordinate increase in tyrosine phosphorylation of p125 FAK and paxillin is elicited by a variety of extracellular agents that modulate cell growth and differentiation (see Introduction for references). Our results suggest that p125 FAK and paxillin could also play a role in the signaling pathways induced by SPC.
SPC is known to release Ca 2ϩ from intracellular stores, and changes in [Ca 2ϩ ] i can lead to tyrosine phosphorylation of a M r 125,000 band (31). However, SPC-stimulated tyrosine phosphorylation of p125 FAK is completely independent of intracellular Ca 2ϩ release or Ca 2ϩ influx. As shown in the accompanying paper (6), SPC rapidly activates PKC. Since direct activation of PKC has been shown to stimulate tyrosine phosphorylation of p125 FAK , SPC could activate tyrosine phosphorylation of p125 FAK through a PKC-dependent mechanism. However, neither down-regulation of PKC with PDB nor the PKC inhibitor GF 109203X blocked SPC-stimulated tyrosine phosphorylation . Subsequently, cells were washed with PBS and fixed in 4% paraformaldehyde; actin was then labeled with TRITC-conjugated phalloidin (0.25 g/ml) in PBS for 10 min at room temperature. Panel B, effects of PDGF on SPC-induced tyrosine phosphorylation of p125 FAK . Quiescent Swiss 3T3 cells were incubated for 20 min at 37°C with 5 or 30 ng/ml PDGF, 5 M SPC (ϩ), or 5 M SPC in the presence of 5 or 30 ng/ml PDGF. Control cells received an equivalent amount of solvent (Ϫ). Tyrosine phosphorylation of p125 FAK was analyzed by immunoprecipitation with mAb 2A7 directed against p125 FAK and immunoblotting with anti-Tyr(P) mAbs. The position of p125 FAK is indicated by an arrow. SPC-induced tyrosine phosphorylation of p125 FAK in the presence of 5 or 30 ng/ml PDGF was quantified by scanning densitometry. The values shown are expressed as the percentage of the maximum increase in p125 FAK tyrosine phosphorylation above control, unstimulated values and are the mean Ϯ S.E. of three independent experiments. of p125 FAK . Thus, SPC induces tyrosine phosphorylation of p125 FAK by a pathway largely independent of both Ca 2ϩ mobilization and PKC activation.
The increases in tyrosine phosphorylation of p125 FAK and paxillin in response to SPC were accompanied by a novel and dramatic reorganization of the actin cytoskeleton. Specifically, SPC evoked a striking formation of peripheral actin microspikes followed by the development of actin stress fibers. Both events were associated with the assembly of focal adhesion plaques, which were localized at early times at the ends of actin microspikes and later at the ends of actin stress fibers. Thus, focal adhesion assembly in response to SPC occurs prior to actin stress fiber formation and is also linked to the formation of microspikes at the plasma membrane.
Tyrosine phosphorylation of p125 FAK and paxillin and the cytoskeletal changes in response to SPC can be clearly distinguished from effects induced by sphingosine. Tyrosine phosphorylation of p125 FAK and paxillin as well as actin rearrangement elicited by sphingosine developed gradually, reaching a maximum by 60 min (14). In sharp contrast, SPC induced a maximum increase in p125 FAK tyrosine phosphorylation and visible changes in the actin cytoskeleton and focal contact assembly after 1 min of its addition. Furthermore, tyrosine phosphorylation of paxillin by SPC is characterized by a mobility shift resulting in the appearance of slower migrating forms, a process mediated by PKC (9). Indeed, paxillin contains multiple potential sites for PKC phosphorylation (35,39). Sphingosine-induced tyrosine phosphorylation of paxillin is not accompanied by any measurable mobility shift in accord with the very small stimulation of PKC by sphingosine in intact Swiss 3T3 cells (14). Thus, SPC and sphingosine induce different signaling pathways in Swiss 3T3 cells. Interestingly, exog-enously added sphingosine 1-phosphate induces tyrosine phosphorylation of p125 FAK and paxillin with kinetics similar to that of SPC. 3 Tyrosine phosphorylation of p125 FAK and paxillin in response to bombesin and other agents is critically dependent on the integrity of the actin cytoskeleton (8). Given the similar kinetics of SPC-stimulated tyrosine phosphorylation and cytoskeletal changes, it was of interest to establish whether these events were also linked in SPC-treated Swiss 3T3 cells. Pretreatment of quiescent Swiss 3T3 cells with cytochalasin D led to a complete disruption of the actin cytoskeleton and abolished tyrosine phosphorylation of p125 FAK stimulated by SPC. Thus, the integrity of the actin cytoskeleton is essential for SPCinduced tyrosine phosphorylation. This conclusion was substantiated by experiments using PDGF at a high concentration (30 ng/ml), which disrupted actin stress fibers in response to SPC. At this concentration, PDGF also profoundly decreased SPC-induced tyrosine phosphorylation of p125 FAK , revealing a novel cross-talk between SPC and PDGF.
Recent findings indicate that both cytoskeletal changes and tyrosine phosphorylation of p125 FAK and paxillin induced by bombesin require functional p21 rho protein (21,27). These findings suggested the existence of a pathway activated by seven transmembrane domain receptors in which p21 rho is upstream of both cytoskeletal responses and tyrosine phosphorylation of specific proteins. In view of the results described above, it was plausible that the small G protein p21 rho could also be involved in SPC-induced cytoskeletal changes. Microinjection of C. botulinum C3 exoenzyme, which ADP-ribosylates and inactivates FIG. 8. SPC-induced changes in actin cytoskeleton, focal contacts, and tyrosine phosphorylation require the small G protein p21 rho . Confluent and quiescent Swiss 3T3 were microinjected with C. botulinum C3 exoenzyme at 100 g/ml and rabbit IgG at 0.5 mg/ml. 40 min after microinjection, cells were stimulated for 20 min with 5 M SPC, fixed, permeabilized, and stained using triple labeling as described under "Experimental Procedures." Cells were stained with TRITC phalloidin (A and D) to reveal actin filaments, and the same cells were stained with anti-vinculin mAb (B) to localize focal contacts or anti-Tyr(P) mAb 4G10 (E). Rabbit IgG was detected with FITC-conjugated anti-rabbit IgG to localize the microinjected cells (C and F). For comparison, a sector of the dish was chosen where C3 exoenzyme-injected cells were adjacent to non-injected cells. The cytoplasmic staining visible in Fig. 8, B and E, was due to unspecific labeling by the secondary Cy5-conjugated anti-mouse IgG antibody as shown in experiments where the cells were microinjected with C3 exoenzyme, treated with SPC as described above, and stained with Cy5-conjugated anti-mouse IgG without prior incubation with either anti-vinculin mAb or anti-Tyr(P) mAb (data not shown). p21 rho function, prevented actin stress fiber formation, focal contact assembly, and tyrosine phosphorylation in response to SPC. Thus, p21 rho is upstream of both cytoskeletal changes and tyrosine phosphorylation in response to SPC.
Recently, a possible link between p125 FAK tyrosine phosphorylation and the MAPK pathway has been suggested. Several groups demonstrated that integrin-mediated cell adhesion activates MAPK by a pathway critically dependent on the integrity of the actin cytoskeleton (40,41). Schlaepfer et al. (42) showed that integrin engagement stimulates an interaction between p125 FAK and the adapter protein Grb2, suggesting a possible link between p125 FAK and the p21 ras /Raf/MAPK pathway. In view of the data provided in the accompanying paper (6) on SPC-induced MAPK activation and here on p125 FAK tyrosine phosphorylation, it was important to define whether these events are causally related. Several lines of evidence demonstrate a dissociation of p125 FAK tyrosine phosphorylation from MAPK activation in response to SPC in Swiss 3T3 cells. SPC-induced MAPK activation is dependent on the activity of PKC and a pertussis toxin-sensitive G i protein. In contrast, tyrosine phosphorylation of p125 FAK was largely independent of PKC and not affected by treatment with pertussis toxin. Crucially, tyrosine phosphorylation of p125 FAK in response to SPC was completely prevented by disruption of the actin cytoskeleton with cytochalasin D, whereas MAPK activation by SPC was unaffected by an identical treatment with cytochalasin D. Thus, p125 FAK tyrosine phosphorylation and MAPK activation are independently regulated in SPC-treated cells.
In conclusion, our results demonstrate, for the first time, that SPC stimulates tyrosine phosphorylation of p125 FAK and paxillin. Furthermore, SPC induces a unique pattern of reorganization of the actin cytoskeleton and focal adhesion assembly in Swiss 3T3 cells. The integrity of the polymerized actin network and functional p21 rho are essential for SPC-induced tyrosine phosphorylation.