INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

REACTIVE OXYGEN SPECIES (ROS) or oxidants [superoxide anion, hydroxyl radical, hydrogen peroxide (H₂O₂), hypochlorous acid, and peroxynitrite] have been implicated in cell and tissue damage initiated by a number of agents, including inflammatory mediators, hyperbaric oxygen, and ischemic tissue reperfusion (15). Several studies have provided clear evidence that vascular disorders, including atherosclerosis, pulmonary hypertension, and adult respiratory distress syndrome, result, at least in part, from the activity of these highly reactive, short-lived oxidative agents (5). When generated at relatively high levels by activated leukocytes, for example, ROS can cause cell injury through oxidative damage to cellular macromolecules (15). Recent studies suggested that low levels of ROS modulate signal transduction pathways in mammalian cells (41). For example, H₂O₂ stimulates protein phosphorylation (34), activates protein kinases (PKs) (1, 3, 13, 19), inhibits tyrosine phosphatases (7), alters intracellular Ca²⁺ (Ca²⁺_i) (35), stimulates phospholipases (24), and regulates transcription factors such as nuclear factor-B and activator protein-1 (36). H₂O₂ activated several members of the mitogen-activated protein kinase (MAPK) family in vascular endothelial cells (ECs), fibroblasts, and smooth muscle cells (1, 13, 19; Scribner, Vepa, and Natarajan, unpublished data) through a tyrosine kinase-dependent mechanism. H₂O₂ alone or in combination with vanadate, an inhibitor of protein tyrosine phosphatases, mimics several of the metabolic and growth-promoting effects of growth factors and insulin (18). The biochemical and molecular pathways involved in the oxidant-mediated activation of PKs and the contribution of these changes to oxidative stress are yet to be clearly understood.

Focal adhesion kinase (FAK; p125^FAK) is a non-receptor tyrosine kinase involved in the structure and function of focal adhesions (12), which are critical for cellular integrity by maintaining cell-cell and cell-matrix interactions. Tyrosine phosphorylation of FAK and paxillin (p68) has been implicated as an early event in signal transduction initiated by integrins (12), growth factors (2), bioactive lipids such as lysophosphatidic acid (LPA), sphingosine and sphingosylphosphorylcholine (8, 32, 38), and extracellular matrix proteins (4, 16). Tyrosine phosphorylation of focal adhesion-associated proteins induced by these ligands is accompanied by alterations in the actin cytoskeletal network and relocation of focal adhesion-associated proteins (12, 27). Although ROS are known to alter cytoskeletal organization (19), which is vital for cell migration, differentiation, and cellular permeability changes, the effects of these agents on the individual proteins involved in cytoskeletal reorganization and focal adhesion have not been well studied. A recent study (40) examined the role of FAK in H₂O₂-induced apoptosis in T98G cells and concluded that tyrosine phosphorylation of FAK by H₂O₂ acted as a suppressor of apoptosis. Gozin et al. (11) reported hydroxyl radical-induced tyrosine phosphorylation of FAK and paxillin in human umbilical vein endothelial cells. The present study was undertaken to examine the ability of H₂O₂ to stimulate tyrosine phosphorylation of focal adhesion proteins in vascular ECs. Our data indicate that H₂O₂ induces both FAK and paxillin tyrosine phosphorylation through tyrosine phosphatase inhibition more than through activation of tyrosine kinases.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Materials. Medium 199 (M199), fetal bovine serum, trypsin, Dulbecco's phosphate-buffered saline (PBS), H₂O₂, 12-O-tetradecanoylphorbol 13-acetate (TPA), genistein, myelin basic protein (MBP) fragment, and cytochalasin D were obtained from Sigma (St. Louis, MO). Herbimycin, tyrphostin 47, 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA)-AM, and bisindolylmaleimide I were purchased from Calbiochem (La Jolla, CA). Monoclonal antibodies against FAK and paxillin were procured from Transduction Laboratories (Lexington, KY), and anti-phosphotyrosine antibody 4G10 was purchased from Upstate Biotechnology (Lake Placid, NY). MAPK antibodies [extracellular signal-regulated kinase (ERK) 1 and ERK2], polyclonal FAK antibodies and protein A/G plus agarose were from Santa Cruz Biotechnology (Santa Cruz, CA). Rhodamine-phalloidin and fura 2-AM were obtained from Molecular Probes (Eugene, OR). Precast acrylamide gels were purchased from NOVEX (San Diego, CA), and all electrophoretic reagents were from Bio-Rad (Hercules, CA). Bovine pulmonary artery ECs (BPAECs; passage 16) were obtained from American Type Culture Collection (Manassas, VA).

Cell culture. BPAECs were cultured as previously described (24). Subcultured BPAECs (from passages 18 to 21), used in all the experiments, showed cobblestone morphology and stained positive for factor VIII-related antigen. In our experiments, we have not observed any differences in EC response to oxidants whether the cells are serum starved or not. Thus, in all our experiments, we used cells grown in serum-containing medium.

Immunoprecipitation and Western blotting. BPAECs grown on T-75cm² flasks were washed once with serum-free M199 and stimulated with H₂O₂ or other agents in serum-free M199 for specified time periods. The cells were washed once in ice-cold PBS and again in ice-cold PBS containing 1 mM sodium orthovanadate. The cells (5× 10⁶) were scraped into 1 ml of lysis buffer (20 mM Tris · HCl, pH 7.4, containing 0.5% deoxycholic acid, 0.5% SDS, 1% Triton X-100, 1% Nonidet P-40, 0.5 mM phenylmethylsulfonyl fluoride, 5 µg/ml of leupeptin, 5 µg/ml of aprotinin, and 1 mM sodium orthovanadate). The samples were centrifuged at 14,000 rpm for 10 min at 4°C, and the supernatants were used for immunoprecipitation with either anti-FAK, anti-paxillin, or anti-MAPK antibody (2-4 µg) at 4°C for 4 h. Protein A/G plus agarose (20 µl) was then added and incubated for an additional 3 h at 4°C. The antigen-antibody complex was pelleted, washed three times with ice-cold lysis buffer, and either dissociated by boiling in 1× SDS sample buffer for 5 min or washed once with the nondenaturing assay buffer used for kinase assays [50 mM PIPES, pH 7.0, 10 mM MgCl₂, 3 mM MnCl₂, and 0.1 mM dithiothreitol (DTT)]. The samples were then analyzed on 8 or 10% SDS-PAGE gels. After SDS-PAGE, the proteins were transferred to Immobilon-P membranes by electroblotting, blocked with blocking buffer (GIBCO BRL, Life Technologies), and incubated for 18-24 h at 4°C with either anti-FAK (1:1,000 dilution), anti-paxillin (1:5,000 dilution), anti-phosphotyrosine (1 µg/ml), or anti-MAPK (1 µg/ml) antibody.The membranes were washed four times with PBS containing 0.1% Tween 20 followed by incubation with goat anti-rabbit or anti-mouse IgG conjugated to horseradish peroxidase (1:3,000 dilution) for 1 h at room temperature, and the blots were developed with enhanced chemiluminescence. Densitometric scanning of the blots was carried out with a Bio-Rad model GS-700 densitometer and quantified with Molecular Analyst software.

Rhodamine-phalloidin staining for visualization of actin stress fibers. BPAECs grown on glass coverslips were washed twice with M199 containing HEPES and incubated for the indicated times at 37°C in medium with and without H₂O₂. Thereafter, for actin staining, the cells were washed twice with ice-cold PBS and fixed in 3.0% formaldehyde in PBS for 10 min at room temperature. The cells were rinsed three times with PBS and permeabilized with PBS containing 0.1% Triton X-100 for 5 min at 22°C. The permeabilized cells were then washed three times in PBS and subsequently incubated with rhodamine-phalloidin (1 U/coverslip) for 30 min in the absence of light. Unbound rhodamine-phalloidin was removed by washing with PBS three times. The coverslips were mounted onto glass slides and sealed with nail polish. Fluorescent images of actin fibers were obtained with confocal microscopy at a magnification of ×100 (Bio-Rad model MRC1024 with a krypton/argon laser system) as described elsewhere by Hart et al. (17) and Schaphorst et al. (33).

In vitro kinase assays. For MAPK assays, an MBP fragment was used as a substrate (13). Immunoprecipitates obtained with anti-ERK1 and anti-ERK2 antibodies were washed once with kinase assay buffer (50 mM PIPES, pH 7.0, 10 mM MgCl₂, 3 mM MnCl₂, and 0.1 mM DTT) and incubated with a reaction mixture (50 µl) that contained 5 µg of MBP, 10 µCi of [-³²P]ATP (10 µM), and kinase assay buffer. The reaction was carried out at 37°C for 30 min and terminated by adding 10 µl of 6× Laemmli electrophoresis sample buffer and boiling for 5 min. MBP phosphorylation was measured after SDS-PAGE (14% gels) and autoradiography. In vitro kinase assays to determine the autophosphorylation of FAK were performed according to the protocol described by Rankin et al. (28). Briefly, immunoprecipitates of FAK prepared from BPAEC lysates after various treatments were washed extensively and suspended in kinase assay buffer (50 mM PIPES, pH 7.0, 10 mM MnCl₂, and 1 mM DTT). The reaction mixture was incubated for 10 min at 30°C in the presence of 0.25 µCi [-³²P]ATP/µl reaction mixture. The reactions were stopped by the addition of 6× Laemmli electrophoresis sample buffer and boiling for 5 min. In vitro phosphorylation of FAK was measured after SDS-PAGE (8% gels) and autoradiography.

Measurement of Ca²⁺_i concentration. BPAECs grown on glass coverslips were loaded with fura 2-AM as described earlier (23). Fura 2-loaded BPAECs were challenged with different agents and studied in a thermostat-regulated sample compartment of an SLM spectrofluorometer controlled by a PC program to excite the fura 2-loaded cells alternately at 340 and 380 nm, and fluorescence was monitored at 510 nm. The ratios of fluorescence intensities at 340 and 380 nm were calculated. In all experiments, measurements were corrected for autofluorescence from cells not loaded with fura 2.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

H₂O₂ increases tyrosine phosphorylation in BPAECs. To assess the effect of H₂O₂ on protein tyrosine phosphorylation, BPAECs were incubated with varying concentrations of H₂O₂ for 10 min, and the cell lysates were subjected to SDS-PAGE and Western blotting with anti-phosphotyrosine antibody. As shown in Fig.1, H₂O₂ in a time- and dose-dependent manner stimulated tyrosine phosphorylation of several proteins between 50 and 200 kDa, including heavily phosphorylated bands at 110-130 and 65-80 kDa. The increase in tyrosine phosphorylation of 110- to 130- and 65- to 80-kDa proteins reached a maximum (164 and 457% over basal levels, respectively) at 10 min of treatment with 1.0 mM H₂O₂ (Fig. 1A) and was linear from 0.05 to 1 mM H₂O₂ (Fig. 1B). However, significant increases were observed from 0.1 mM H₂O₂. The concentrations and time periods of H₂O₂ employed in this present study have been found to be noncytotoxic (24), and although the exact physiological relevance of these doses has not been determined, it has been speculated that under certain conditions oxidants are released into a relatively sequestered microenvironment, thus creating a high-oxidant concentration (43).

View larger version (41K): [in this window] [in a new window]   Fig. 1.   Time course and dose response of H2O2-stimulated tyrosine phosphorylation. Lysates (20 µg of protein) prepared from bovine pulmonary artery endothelial cells (BPAECs) treated with either 1 mM H2O2 for various time intervals (A) or different concentrations of H2O2 for 10 min (B) were subjected to SDS-PAGE and membrane transfer and analyzed by Western blotting with anti-phosphotyrosine antibody 4G10 followed by enhanced chemiluminescence as described in METHODS. Brackets, range of proteins in which tyrosine phosphorylation was increased; arrows, protein bands used in quantitation. Blots are representative of 3 independent experiments. Nos. on left, molecular mass.

Stimulation of FAK and paxillin tyrosine phosphorylation by H₂O₂. Because ROS treatment of BPAECs produced a pattern of tyrosine phosphorylation very similar to that induced by bombesin, sphingosine, and sphingophosphorylcholine in Swiss 3T3 cells (28, 38), we investigated whether FAK and paxillin were tyrosine phosphorylated in response to H₂O₂ in BPAECs. The cell lysates from control and H₂O₂-treated BPAECs were incubated with either anti-FAK or anti-paxillin monoclonal antibodies, and the immunoprecipitates were probed with anti-phosphotyrosine, anti-FAK, or anti-paxillin antibody. As shown in Fig.2, A and C, at 1.0 mM H₂O₂, tyrosine phosphorylation of FAK was rapid, was detected as early as 30 s of exposure, was linear up to 10 min (290% of basal levels), and declined thereafter, reaching near basal levels at 60 min. H₂O₂-induced FAK tyrosine phosphorylation was concentration dependent (330% of control value with 1 mM H₂O₂) and was observed with concentrations of H₂O₂ as low as 125 µM at 10 min (Fig. 2, B and D). Generation of ROS by xanthine (100 µM)/xanthine oxidase (10 U/ml) also resulted in a time-dependent increase in FAK phosphorylation (data not shown). In addition to FAK, we also investigated the effect of H₂O₂ on tyrosine phosphorylation of paxillin. Similar to FAK, treatment of BPAECs with H₂O₂ (1 mM) also enhanced tyrosine phosphorylation of paxillin as evidenced by Western blotting analysis of anti-paxillin immunoprecipitates with anti-phosphotyrosine (Fig. 3). Although H₂O₂-induced tyrosine phosphorylation of paxillin also showed a time and concentration dependence (Fig. 3), the time course of tyrosine phosphorylation of paxillin was different compared with that of FAK. These results indicate that H₂O₂ induces tyrosine phosphorylation of FAK and paxillin in BPAECs, and on the basis of these results, for all further experiments, ECs were incubated with 1 mM H₂O₂ for 10 min.

View larger version (68K): [in this window] [in a new window]   Fig. 2.   H2O2-induced tyrosine phosphorylation of focal adhesion kinase (FAK). A: confluent BPAECs were treated with 1 mM H2O2 for indicated time periods, and cell lysates (1 mg of protein) were subjected to immunoprecipitation with anti-FAK monoclonal antibody followed by immunoblotting (I.B.) of equal amount of immunoprecipitates with anti-phosphotyrosine (A-PTyr) and anti-FAK (A-FAK) antibodies as described in METHODS. B: confluent BPAECs were exposed to indicated concentrations of H2O2 for 10 min. Cell lysates (1 mg of protein) were immunoprecipitated with A-FAK, and equal aliquots of immunoprecipitates were further analyzed by Western blotting with A-PTyr and A-FAK. C and D: densitometric analysis of blots in A and B, respectively, was performed as described in METHODS, and tyrosine phosphorylation was normalized for amount of FAK and is presented as percent increase over control levels. Blots are representative of 3 independent experiments.

View larger version (87K): [in this window] [in a new window]   Fig. 3.   H2O2-induced tyrosine phosphorylation of paxillin. A: confluent BPAECs were treated with 1 mM H2O2 for indicated time periods, and cell lysates (1 mg of protein) were subjected to immunoprecipitation with anti-paxillin monoclonal antibody (A-Paxillin) followed by immunoblotting of equal amount of immunoprecipitates with A-PTyr and A-Paxillin as described in METHODS. B: confluent BPAECs were exposed to indicated concentrations of H2O2 for 10 min. Cell lysates (1 mg of protein) were immunoprecipitated with A-Paxillin, and equal aliquots of immunoprecipitates were further analyzed by Western blotting with A-PTyr and A-Paxillin. C and D: densitometric analysis of blots in A and B, respectively, was performed as described in METHODS, and tyrosine phosphorylation was normalized for amount of paxillin and is presented as percent increase over control levels. Blots are representative of 3 independent experiments.

Effect of H₂O₂ on FAK autophosphorylation activity. An increase in FAK phosphorylation by growth factors or sphingosine derivatives has been associated with increased FAK activity (30, 31). To examine whether the H₂O₂-mediated increase in FAK tyrosine phosphorylation was associated with a concomitant increase in FAK activity, we tested the FAK immunoprecipitates obtained from H₂O₂-treated EC lysates. As shown in Fig. 4, no increase in in vitro autophosphorylation activity of FAK was observed in anti-FAK immunoprecipitates from H₂O₂-treated ECs. Similar results were also obtained when we used a protocol described by Rodriguez-Fernandez and Rozengurt (30) to carry out autophosphorylation with a polyclonal FAK antibody (data not shown). These data suggest that the sites that are being tyrosine phosphorylated in response to H₂O₂ probably are not involved in regulating the autophosphorylating activity of FAK.

View larger version (36K): [in this window] [in a new window]   Fig. 4.   Effect of H2O2 on autophosphorylating activity of FAK. Confluent BPAECs were treated with H2O2 for 10 min, and cell lysates (1 mg of protein) were subjected to immunoprecipitation (I.P.) and I.B. with A-FAK. Equal amounts of immunoprecipitate were subjected to autophosphorylation reaction with [-32P]ATP (10 µCi, 10 µM), separated by SDS-PAGE, and subjected to autoradiography as described in METHODS. This experiment is representative of 3 independent determinations. Nos. on right, molecular mass.

H₂O₂ stimulates actin stress-fiber formation in BPAECs. Tyrosine phosphorylation of FAK has been shown to be an early event in cell adhesion that is crucial to the formation and assembly of focal adhesion structures (12, 27) and is invariably preceded by actin reorganization. We therefore examined whether ROS modulated actin stress-fiber orientation and cytoskeletal organization in ECs. As shown in Fig. 5, resting BPAECs contain very few actin stress fibers, and treatment of BPAECs with H₂O₂ (100 µM and 1 mM for 2 min) caused changes in the organization and orientation of actin stress fibers. Within 2 min, H₂O₂ caused reorganization of actin fibers throughout the cytoplasm (data not shown), and by 10 min, the cells were packed with numerous stress fibers (Fig. 5). We examined whether disruption of the actin stress fibers could affect the ROS-mediated increase in FAK tyrosine phosphorylation. Preincubation of BPAECs for 30 min with cytochalasin D (5-10 µM), a selective disruptor of the intracellular actin cytoskeletal network, attenuated H₂O₂-induced tyrosine phosphorylation of FAK (Fig. 6), suggesting a link between FAK activation and actin stress-fiber formation.

View larger version (78K): [in this window] [in a new window]   Fig. 5.   Effect of H2O2 on actin stress-fiber formation. Confluent BPAECs grown on glass coverslips were washed twice with medium 199 and incubated with 100 µM and 1 mM H2O2 for 10 min. Cells were subsequently fixed in formaldehyde and stained for F-actin with rhodamine-phalloidin. Changes in actin cytoskeleton were visualized with a confocal microscope at a magnification of ×100 as described in METHODS.

View larger version (48K): [in this window] [in a new window]   Fig. 6.   Effect of cytochalasin D on H2O2-induced tyrosine phosphorylation of FAK. Confluent BPAECs were treated for 45 min with cytochalasin D and then incubated with (+) and without () H2O2 for 10 min. Cell lysates (1 mg of protein) were subjected to immunoprecipitation with A-FAK, and immunoprecipitates were divided into equal aliquots and analyzed by Western blotting with A-FAK or A-PTyr. Blot is representative of 3 independent experiments.

H₂O₂-mediated increases in FAK tyrosine phosphorylation is independent of PKC. We examined the role of PKC on tyrosine phosphorylation of FAK by incubating cells with the PKC activator TPA. Treatment of BPAECs with TPA (100 nM) for 10 min had no effect on the tyrosine phosphorylation of FAK (Fig. 7), and the same concentration of TPA also had no effect on H₂O₂-induced FAK tyrosine phosphorylation (Fig. 7). Preincubation of the cells for 1 h with the specific PKC inhibitor bisindolylmaleimide I followed by treatment with H₂O₂ (1 mM, 10 min) did not have any effect on H₂O₂-induced FAK phosphorylation (data not shown).

View larger version (47K): [in this window] [in a new window]   Fig. 7.   Effect of 12-O-tetradecanoylphorbol-13-acetate (TPA) on H2O2-induced tyrosine phosphorylation of FAK. BPAECs were preincubated with either medium 199 or medium 199 containing TPA for 30 min and challenged with H2O2 for 10 min, and cell lysates were prepared as described in METHODS. Cell lysates (1 mg of protein) were subjected to immunoprecipitation with A-FAK, and equal amounts of immunoprecipitates were analyzed by Western blotting with A-PTyr or A-FAK. Blot is representative of 2 separate experiments.

Effect of Ca²⁺ on H₂O₂-mediated increases in FAK tyrosine phosphorylation. Changes in Ca²⁺_i have been shown to modulate signaling pathways involving protein tyrosine phosphorylation (20). We therefore examined the role of Ca²⁺_i in H₂O₂-induced FAK phosphorylation. H₂O₂ treatment of BPAECs grown on coverslips increased Ca²⁺_i (Fig. 8A) as measured by the emitted fluorescence at 510 nm of fura 2-loaded cells excited at 340 and 380 nm but was quenched by BAPTA-AM, a chelator of free Ca²⁺_i (24). As shown in Fig. 8B, treatment of BPAECs with BAPTA showed no attenuation of H₂O₂-induced FAK tyrosine phosphorylation. Although we have observed an increase of tyrosine phosphorylation with high concentrations of BAPTA, this increase was not reproducible. These data indicate that the H₂O₂-stimulated tyrosine phosphorylation of FAK is independent of changes in Ca²⁺_i in ECs.

View larger version (19K): [in this window] [in a new window]   Fig. 8.   Effect of Ca2+ on H2O2-induced tyrosine phosphorylation of FAK. A: BPAECs grown on glass coverslips and loaded with fura 2 as described in METHODS were treated with Hanks' balanced salt solution (HBSS) or HBSS containing 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA)-AM for 30 min (30') and challenged with H2O2. Intracellular Ca2+ was measured with a luminescence spectrometer as described in METHODS. B: confluent BPAECs were incubated with medium 199 or medium 199 containing BAPTA-AM for 30 min followed by a 10-min exposure to H2O2. Cell lysates (1 mg of protein) were subjected to immunoprecipitation with A-FAK and analyzed by Western blotting for phosphotyrosine or FAK with appropriate antibodies. Blot is representative of 3 independent experiments. Densitometric analysis of blot was performed, and relative intensities were calculated.

H₂O₂-mediated increases in FAK tyrosine phosphorylation is resistant to tyrosine kinase inhibitors. Because protein tyrosine phosphorylation is a balance between protein tyrosine kinases and tyrosine phosphatases, we investigated the effect of tyrosine kinase inhibitors on H₂O₂-induced FAK tyrosine phosphorylation. BPAECs pretreated with genistein (100 µM, 3 h), tyrphostin (100 µM, 3 h), or herbimycin (5 µM, 5 h) before being challenged with H₂O₂ showed no effect on H₂O₂-induced FAK phosphorylation (Fig. 9). However, under similar experimental conditions, these tyrosine kinase inhibitors attenuated H₂O₂-induced MAPK activity as measured by MBP phosphorylation (Fig. 9). These results suggest that H₂O₂ differentially regulates MAPK activation and increases in tyrosine phosphorylation of FAK.

View larger version (38K): [in this window] [in a new window]   Fig. 9.   Effect of tyrosine kinase inhibition on mitogen-activated protein kinase (MAPK) activity (A) and FAK phosphorylation (B). BPAECs were treated with either medium 199 or medium 199 containing genistein (3 h), herbimycin (5 h), or tyrphostin (3 h) followed by exposure to H2O2 for 10 min. Cell lysates (1 mg of protein) were immunoprecipitated with anti-extracellular signal-regulated kinases 1 and 2. Immunoprecipitates were assayed for MAPK activity with myelin basic protein (MBP) and [-32P]ATP as described in METHODS. Arrow, position of [-32P]MBP on 14% SDS-PAGE. Similarly, cell lysates (1 mg of protein) prepared as described above were immunoprecipitated with A-FAK and subsequently analyzed by Western blotting with A-PTyr. Data are representative of 3 experiments. Nos. on right, molecular mass.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

ROS-mediated alterations in cellular signaling pathways play an important role in vascular dysfunction (5). Natarajan et al. (25) previously reported a functional link between oxidant-induced phospholipase D activation and protein tyrosine phosphorylation. The present work is an attempt to further identify individual proteins in which tyrosine phosphorylation is altered on oxidant exposure and to elucidate the contribution of these to the oxidant injury. The tyrosine phosphorylation profile observed in the present study was very similar to those described with LPA, bombesin, and sphingophosphorylcholine (8, 28, 31, 32, 38, 39). Because previous studies (2, 26, 28, 32, 38, 39) have shown that growth factors, hormones, and bioactive lipids stimulate tyrosine phosphorylation of FAK and paxillin in mammalian cells, we investigated the possible effect of H₂O₂ in modulating the tyrosine phosphorylation of FAK and paxillin in ECs. The results of the present study clearly show a dose- and time-dependent increase in the tyrosine phosphorylation of FAK and paxillin. However, our results suggest differences in the time course of H₂O₂-induced tyrosine phosphorylation of FAK and paxillin. Paxillin phosphorylation appears to be delayed compared with that of FAK. Recently, H₂O₂-induced increases in tyrosine phosphorylation of FAK in the human glioblastoma cell line T98G were reported (40). In T98G cells, the increase in tyrosine phosphorylation of FAK was apparent only after 30 min and continued up to 5 h, whereas in ECs with a relatively high concentration of H₂O₂ (1 mM), we observed increases as early as 30 s of oxidant exposure that reached a plateau by 10-15 min. In addition, we observed a decline in both tyrosine phosphorylation and cell viability after 2 h of H₂O₂ exposure (data not shown). These results indicate clear differences between cell types in response to H₂O₂-induced tyrosine phosphorylation of FAK. Gozin et al. (11) have shown increased tyrosine phosphorylation of FAK and paxillin by ROS produced through the xanthine/xanthine oxidase system in human umbilical vein ECs. This study has identified the hydroxyl radical as the reactive species responsible for the observed increases. However, the precise biochemical mechanism by which ROS increase tyrosine phosphorylation of these proteins is not clear. Because ROS are known to alter the redox status of the cells by depleting thiols (37), it will be interesting to examine the role of thiols in ROS-induced increases in tyrosine phosphorylation of focal adhesion proteins. FAK is a non-receptor tyrosine kinase and possesses autophosphorylating activity (6). An earlier study (31) with LPA, endothelin, bombesin, and sphingophosphorylcholine showed a correlation between increases in FAK tyrosine phosphorylation and autophosphorylating activity. However, in our study, we were unable to demonstrate any increase in the autophosphorylating activity of FAK in the immunoprecipitates obtained after oxidant exposure of ECs. This may be due to differences in the sites of FAK phosphorylation by oxidants compared with other agents. Inhibition of the autophosphorylating activity of FAK by 1 mM H₂O₂ is intriguing and may suggest that H₂O₂-induced phosphorylation of certain tyrosine residues is inhibitory to the autophosphorylating activity. In addition, FAK immunoprecipitates obtained from oxidant-treated cells have not revealed any increases in tyrosine kinase activity compared with FAK immunoprecipitates obtained from untreated cells (data not shown). These results suggest that treatment of ECs with oxidants did not increase the association of PKs with FAK. Earlier studies (12, 38, 39) have demonstrated a correlation between FAK activation and cytoskeletal rearrangement as measured by the formation of actin stress fibers. Cytochalasin D, an inhibitor of the microfilament network, attenuated tyrosine phosphorylation of FAK and associated proteins. The results from the present study are consistent with these earlier observations. Although the physiological relevance of actin stress fibers is not firmly established, the presence and formation of these fibers under in vivo situations were demonstrated in ECs (10). The functional significance of ROS-induced actin stress-fiber formation requires further investigation.

H₂O₂ is a known activator of PKC in ECs (42), and the involvement of PKC in tyrosine phosphorylation of FAK was demonstrated in certain cell types (14, 21, 32, 39, 44). However, in ECs, we observed neither a direct influence of PKC on tyrosine phosphorylation of FAK nor an indirect influence of PKC on H₂O₂-induced tyrosine phosphorylation of FAK. In contrast to the present findings in ECs, in a previous study with platelets (14), bisindolylmaleimide, an inhibitor of PKC, was shown to attenuate tyrosine phosphorylation of FAK. Also, downregulation of PKC attenuated only phorbol 12,13-dibutyrate-induced, but not endothelin-mediated, FAK tyrosine phosphorylation in fibroblasts (39). However, in Rat-1 fibroblasts, Saville et al. (32) showed that endothelin- but not LPA-mediated FAK tyrosine phosphorylation was attenuated by inhibiting PKC. These results indicate that depending on the cell type, agonist-induced tyrosine phosphorylation of FAK is either PKC dependent or independent.

Our results with tyrosine kinase inhibitors are in contrast to those observed in Swiss 3T3 fibroblasts (8) and human neuroblastoma cells (SK-N-SH clone SY54) (4) in which tyrosine phosphorylation of FAK was found to be sensitive to one or more tyrosine kinase inhibitors (erbstatin, tyrphostin, herbimycin, and/or genistein). Our data show that under the current experimental conditions, the H₂O₂-mediated tyrosine phosphorylation of FAK in ECs is insensitive to tyrosine kinase inhibitors and thus suggest alternate pathways of increased tyrosine phosphorylation. However, the same inhibitors significantly attenuated H₂O₂-induced overall tyrosine phosphorylation as measured by immunoblotting of total cell lysates (data not shown) and also the tyrosine kinase-dependent activation of MAPK as measured by MBP phosphorylation (Fig. 9). It is possible that H₂O₂-induced tyrosine phosphorylation of FAK in ECs may be primarily through the inhibition of protein tyrosine phosphatases. Recent reports (9, 22) indicated that inhibitors of tyrosine phosphatase, such as vanadate and pervanadate, activate tyrosine phosphorylation of FAK and paxillin. In fact, a role for protein tyrosine phosphatase activity in focal adhesion and stress-fiber formation was demonstrated in Swiss 3T3 cells (29). In addition, many studies (7, 18) have demonstrated the inhibitory effect of H₂O₂ on tyrosine phosphatases. Based on the published data, it is possible that in ECs, ROS-mediated increases in FAK tyrosine phosphorylation result, at least in part, from the inhibition of a protein tyrosine phosphatase(s) specific to FAK. However, the present data do not exclude the involvement of tyrosine kinases, which were not inhibited under our experimental conditions, in the stimulation of FAK tyrosine phosphorylation. Because FAK is a known substrate for Src and ROS are known to activate Src in various cell types including ECs (3), one has to cautiously interpret the lack of inhibition of H₂O₂-induced FAK tyrosine phosphorylation by tyrosine kinase inhibitors.

In summary, the present results demonstrate that ROS stimulate tyrosine phosphorylation of FAK and paxillin in vascular ECs. However, in contrast to the effect of H₂O₂ on MAPK activation, the increase in tyrosine phosphorylation of FAK appears to be insensitive to tyrosine kinase inhibition. These findings suggest a novel and differential regulation of FAK and MAPK by ROS. Presently, the physiological significance of this differential regulation by ROS remains unclear; however, recent reports (1, 13, 34) demonstrate the participation of ROS in signaling cascades involving members of the MAPK and Syk tyrosine kinase families. Consistent with this postulate, Huot et al. (19) recently demonstrated a role for ROS-induced p38 kinase in the cytoskeletal rearrangement in ECs. Future efforts need to focus on the role of specific protein tyrosine phosphatases and protein tyrosine kinases in the regulation of ROS-mediated tyrosine phosphorylation of focal adhesion proteins and actin cytoskeleton assembly in vascular endothelium.