Mildly oxidized low density lipoprotein rapidly stimulates via activation of the lysophosphatidic acid receptor Src family and Syk tyrosine kinases and Ca2+ influx in human platelets.

In contrast to native low density lipoprotein (LDL), mildly oxidized LDL (mox-LDL) induced platelet shape change and stimulated during shape change the tyrosine phosphorylation of specific proteins including Syk; the translocation of Src, Fyn, and Syk to the cytoskeleton; and the increase of cytosolic Ca(2+) due to mainly Ca(2+) entry. The stimulation of these early signal pathways by mox-LDL was inhibited by desensitization of the lysophosphatidic acid (LPA) receptor and specific LPA receptor antagonists, was independent of the alpha(IIb)beta(3)-integrin, and was mimicked by LPA. Stimulation of tyrosine phosphorylation and Syk activation were independent of the increase of cytosolic Ca(2+) and were suppressed by genistein and two specific inhibitors of the Src family tyrosine kinases, PP1 and PD173956. In contrast to PP1 and PD 173956, genistein prevented shape change by mox-LDL. The results indicate that mox-LDL, through activation of the LPA receptor, stimulates two separate early signal pathways, (a) Src family and Syk tyrosine kinases, and (b) Ca(2+) entry. The activation of these early signaling pathways by mox-LDL probably plays a role in platelet responses subsequent to shape change. The inhibition of mox-LDL-induced platelet activation by LPA receptor antagonists or dietary isoflavonoids such as genistein could have implications in the prevention and therapy of cardiovascular diseases.

Oxidative modification of LDL 1 plays an important role in the pathogenesis of atherosclerosis. Soft, lipid-rich plaques containing LDL and oxidized LDL (1,2) are vulnerable and upon rupture may expose thrombogenic LDL particles that activate circulating platelets, causing them to aggregate and to form an intravascular plug that eventually leads to stroke and myocardial infarction (3,4). Indeed, previous studies indicate that oxidatively modified LDL stimulates platelets, mildly oxidized LDL (mox-LDL) being more active than heavily oxidized LDL (ox-LDL) (5,6). mox-LDL or minimally modified LDL (mm-LDL) could also be present in the circulation. A subfraction of LDL that consists of electronegative, small dense, slightly oxidized LDL particles has been found in peripheral blood (7). An interaction of platelets with circulating mm-LDL may explain the enhanced platelet aggregation often observed in patients with cardiovascular disease. Indeed, a higher prevalence of small dense circulating LDL particles in coronary heart disease and in conditions commonly associated with atherogenesis has been described (8). The interaction of circulating mm-LDL with platelets may favor intravascular thrombus formation at hemodynamically critical sites of stenotic coronary and carotid arteries.
Whereas ox-LDL has often toxic effects on cells (9,10), mm-LDL and mox-LDL seem to alter the function of cells of the vessel wall and platelets through stimulation of specific signal transduction pathways (1,6,11). The elucidation of signal transduction mechanisms of mox-LDL and mm-LDL may provide a basis for new preventive and therapeutic strategies in atherosclerotic cardiovascular disorders. In platelets, LDL and ox-LDL bind with high affinity to specific sites that are different from the classic LDL receptor (12,13). The glycoproteins IIb and IIIa (␣ IIb ␤ 3 integrin) have been identified as binding sites for LDL on platelet membranes but not intact platelets (14,15). Although the ␣ IIb ␤ 3 integrin is probably not a receptor for LDL, a recent study shows that it is important in mediating platelet sensitization by LDL (16).
In the present study, we studied the platelet signal transduction mechanisms of mox-LDL activated during shape change, the earliest platelet response. Shape change can be induced by mox-LDL selectively without the formation of thromboxane A 2 and the release of substances such as ADP or von Willebrand factor, which on their own activate platelets and elicit intracellular signal pathways (6,17). Shape change also occurs without ␣ IIb ␤ 3 integrin outside-in signaling induced by bound fibrinogen (17). Therefore, the signal pathways ob-served during shape change are a direct consequence of the interaction of mox-LDL with its receptor and not mediated by secondary activation of other receptors. Since we have observed recently that mox-LDL contains lysophosphatidic acid (LPA) and stimulates platelet shape change through activation of LPA receptor(s) (18), we investigated whether the early signal transduction mechanisms elicited by mox-LDL were also mediated by activation of the LPA receptor. We found that mox-LDL stimulates through activation of the LPA receptor and independently of the ␣ IIb ␤ 3 integrin two separate signal transduction pathways in human platelets, the Src family kinasemediated stimulation of protein tyrosine phosphorylation and Syk as well as the stimulation of Ca 2ϩ influx.
Preparation of Native LDL and mox-LDL-LDL was isolated in the continuous presence of EDTA as described (6). LDL was dialyzed at 4°C using a N 2 -saturated buffer (pH 7.4) containing NaCl (150 mM), chloramphenicol (0.15 mM), and EDTA (1 mM) and then stored at 4°C in darkness under N 2 . All LDL concentrations are given in terms of their protein content. mox-LDL, which has been chemically characterized previously, was prepared from EDTA-free LDL by Cu 2ϩ -triggered oxidation exactly as described (6). LDL lipid peroxidation was determined as described (6). Platelet Isolation and Measurement of Platelet Shape Change-Human platelets were treated with acetylsalicylic acid and isolated as described (20). Shape change was measured by recording the light transmission in a LABOR aggregometer (Fresenius, Bad Homburg, Germany) (20). RGDS (2 mM) was added to the platelet samples to prevent the possible formation of platelet microaggregates during platelet stirring. In some experiments, platelets were pretreated with EGTA (1 mM) and BAPTA-AM (2-20 M, dependent on the platelet concentration) for 20 min at 37°C to chelate any increase of cytosolic Ca 2ϩ in stimulated platelets.
Isolation of the Platelet Cytoskeleton-Cytoskeletal fractions were prepared by using a modification of the method described by Jennings et al. (21). Briefly, platelet suspensions (0.6 ml) were lysed before or at various intervals after agonist addition by adding equal volumes of ice-cold 2ϫ Triton lysis buffer (pH 7.5) containing 100 mM Tris-HCI, 2% Triton X-100, 10 mM EGTA, 10 mM EDTA, 2 mM sodium orthovanadate, 20 g/ml leupeptin, 2 mM phenylmethylsulfonyl fluoride, 25 M pepstatin A, and 20 g/ml aprotinin. Samples (1.2 ml) were left on ice for 30 min and spun at 15,600 ϫ g for 15 min at 4°C. The pellets were washed once in 1 ϫ Triton X-100 (1%) lysis buffer. Pellets were resuspended in 100 l of gel electrophoresis sample buffer.
Immunoprecipitation of Syk, Separation of Platelet Proteins, and Immunoblotting-Syk was immunoprecipitated by using anti-Syk IgG 2a bound to protein A-Sepharose or covalently conjugated to agarose as described (22). Proteins from platelet suspensions, cytoskeletal fractions, or immunoprecipitates were separated by SDS-polyacrylamide (8%) gel electrophoresis on 1.5 mm ϫ 20 cm gels (Bio-Rad). Proteins were blotted to nitrocellulose membranes and detected as described (20). The dilutions of the mixed primary phosphotyrosine antibodies PY20 and 4G01 were each 1:2.000; the dilutions of the monoclonal anti-Fyn, anti-Src, and polyclonal anti-Syk antibodies were 1:500, FIG. 1. mox-LDL induces platelet shape change. a, suspensions of washed human platelets were placed into aggregometer cuvettes and exposed while being stirred to increasing concentrations of mox-LDL. The decrease in light transmission together with the disappearance of rapid oscillations is indicative of shape change. b, scanning electron micrographs of human platelets. Aliquots of the platelet suspension were taken from the aggregometer cuvettes 30 s after the addition of mox-LDL and fixed with 2.5% glutaraldehyde as described under "Experimental Procedures." Left, control platelets; right, platelets exposed to mox-LDL (0.5 mg/ml). 1:1000, and 1:1000, respectively. The dilutions of the horseradish peroxidase-linked secondary anti-mouse and anti-rabbit antibodies were 1:15,000 and 1:2500, respectively. The specificity of the anti-phosphotyrosine antibodies was tested by competition with phosphotyrosine (1 mM). Anti-phosphotyrosine immunoblots were stripped by incubating the nitrocellulose membrane in buffer (2% SDS, 62.5 mM Tris⅐HCl, and 100 mM mercaptoethanol, pH 6.8) for 30 min at 70°C and reprobed with specific antiprotein tyrosine kinase antibodies. Blots were detected by enhanced chemiluminescence (Amersham Pharmacia Biotech), exposed to Kodak films, and analyzed by densitometry.
Measurement of Cytosolic [Ca 2ϩ ]-For measurement of the cytosolic Ca 2ϩ -concentration, suspensions of washed platelets (1 ϫ 10 9 /ml) were incubated for 1 h at 37°C with Fura-2/AM (4 M) in the presence of bovine albumin (1 mg/ml) (23). After washing, cells were resuspended at a concentration of 1 ϫ 10 9 /ml in the presence of bovine albumin (1 mg/ml). Aliquots of platelets were diluted to 0.2 ϫ 10 9 /ml and incubated for 5 min at 37°C before CaCl 2 (1 mM) or EGTA (5 mM) was added. Fluorescence measurement was then performed in a thermostatically controlled chamber (37°C) while stirring using a Delta-Scan 1 double beam fluorescence spectrophotometer (Photon Technology International) as described previously (24). Cuvettes were thoroughly rinsed between measurements with 70% EtOH and double distilled H 2 O to remove residual amounts of agonists. Calculations of cytosolic Ca 2ϩ concentrations were performed as described (24).
Scanning Electron Microscopy-Platelet suspensions were incubated at 37°C in aggregometer cuvettes while stirring and exposed to native or mox-LDL. Aliquots (0.05 ml) were transferred to 0.5 ml of fixation solution containing 2.5% glutaraldehyde and 75 mM sodium cacodylate. Platelets were centrifuged onto glass plates and dehydrated in a graded series of ethanol. After being critical point-dried with carbon dioxide, the specimens were sputter-coated with gold.
Analysis of Results-The chemiluminescence signals of the immunoblotted protein bands were measured by densitometry (Sharp JX 325 and Ultroscan XL; Amersham Pharmacia Biotech). In order to take into account variations of the amount of Syk immunoprecipitated, the densitometric values of tyrosine-phosphorylated Syk were divided by the corresponding values of Syk. Absorption of proteins in unstimulated control samples was set to 100%. Data are presented as mean Ϯ S.E. of individual experiments from different blood donors. The range of linearity of the densitometric measurements of tyrosine-phosphorylated Syk was determined by a dilution series of protein samples of mox-LDLstimulated platelets.

mox-LDL Induces Platelet Shape
Change-Human platelets were pretreated with aspirin and resuspended in buffer containing apyrase and RGDS in order to inhibit aggregation and secretion by mox-LDL (6) and to eliminate positive feedback amplification mechanisms (17), thereby enabling us to study the direct consequences of interaction of mox-LDL with its platelet receptor(s). Platelets exposed to mox-LDL showed a rapid shape change (Fig. 1a). Concentration-response curves indicated that shape change could be induced by concentrations of mox-LDL as low as 10 -15 g/ml. Shape change upon exposure of platelets to mox-LDL was also observed by scanning electron microscopy (Fig. 1b). Platelets changed their shape from discoechinocytes in control samples to spheroid cells with an irregularly folded and deeply invaginated plasma membrane and many pseudopods in samples stimulated by mox-LDL. The results show that mox-LDL induces shape change of platelets similar to activation by physiological agonists. In contrast to mox-LDL, native LDL (0.5 mg/ml) did not induce shape change (data not shown).
mox-LDL Stimulates Ca 2ϩ Mobilization through LPA Receptor Activation: Predominant Stimulation of Ca 2ϩ Entry by mox-LDL and LPA-Since shape change is often associated with Ca 2ϩ mobilization and Ca 2ϩ -dependent myosin light chain phosphorylation (17) and since we had recently observed that shape change by mox-LDL was mediated by LPA-receptor activation (18), we investigated whether mox-LDL increased cytosolic Ca 2ϩ during shape change and whether it was dependent on LPA receptor activation. In the presence of extracellular Ca 2ϩ (1 mM), both LPA and mox-LDL induced a rapid revers-ible increase of cytosolic Ca 2ϩ during shape change (Fig. 2a). Ca 2ϩ mobilization induced by LPA and mox-LDL showed a rapid homologous desensitization (LPA/LPA and mox-LDL/ mox-LDL) and cross-desensitization (LPA/mox-LDL and, vice versa, mox-LDL/LPA) and was specifically inhibited by the LPA receptor antagonist NPTyrPA (Fig. 2a) and NPSerPA (data not shown) (19,25). NPTyrPA did not inhibit the cytosolic Ca 2ϩ increase evoked by ADP, PAF (Fig. 2a), and YFLLRNP (data not shown). These results demonstrate that mox-LDL stimulates Ca 2ϩ mobilization in platelets through the activation of one or more LPA receptors. The cytosolic Ca 2ϩ rise above basal level was after LPA (1 M), 153 Ϯ 43 nM, and after mox-LDL (0.2 mg/ml), 60 Ϯ 24 nM (mean Ϯ S.D., n ϭ 7). In the presence of EGTA, the cytosolic Ca 2ϩ increase elicited by LPA (1 M) was very small (25 Ϯ 11 nM; mean Ϯ S.D., n ϭ 5), indicating that LPA induced mainly Ca 2ϩ influx across the plasma membrane and relatively little Ca 2ϩ mobilization from intracellular stores (Fig. 2b). Similarly to LPA, the cytosolic Ca 2ϩ increase induced by mox-LDL (0.2 mg/ml) was almost completely abolished by EGTA (Fig. 2b).
Dissociation of Shape Change from the Increase of Cytosolic Ca 2ϩ Induced by mox-LDL and LPA-Concentration-response curves of LPA and stimulations by mox-LDL performed in the presence of EGTA showed that mox-LDL and low concentrations of LPA were able to induce shape change without an increase of cytosolic Ca 2ϩ (Fig. 3, a and b). The EC 50 of LPA for the induction of shape change and cytosolic Ca 2ϩ increase was 18 Ϯ 4 nM and 1.6 Ϯ 0.5 M, respectively. This accounts for an at least 90-fold difference in dose for the induction of these two responses. In order to investigate whether shape change induced by LPA and mox-LDL proceeds independently of Ca 2ϩ mobilization from intracellular stores, we next treated platelets for 20 min with (in addition to EGTA) BAPTA-AM, which enters platelets and inhibits any cytosolic Ca 2ϩ increase due to mobilization from intracellular stores (20,23). As shown in Fig.  3c, LPA (100 nM) induced no increase of cytosolic Ca 2ϩ but elicited in the same platelet suspension a maximal shape change response, and intracellular BAPTA, which completely abolished the increase of cytosolic Ca 2ϩ caused by the mobilization from intracellular stores in ADP-stimulated platelets, delayed the shape change but barely affected the maximal response after LPA stimulation. Similar observations were made when platelets were stimulated with mox-LDL; BAPTA-AM and EGTA treatment caused a delay of the shape change without significantly affecting the maximal response (Table I, right column). We conclude that shape change induced by LPA and mox-LDL proceeds through a pathway that does not involve an increase of cytosolic Ca 2ϩ mox-LDL Stimulates, through Activation of the LPA Receptor, Protein Tyrosine Phosphorylation and the Tyrosine Kinase Syk during Shape Change: Independence of the ␣ IIb ␤ 3 -Integrin-We investigated further whether the shape change induced by mox-LDL was associated with an increase of protein tyrosine phosphorylation, since we have previously observed that the Ca 2ϩ -independent shape change elicited by thrombin receptor activation of human platelets occurred with the tyrosine kinase-mediated phosphorylation of specific proteins (20). We found that both mox-LDL and LPA stimulated the tyrosine phosphorylation of proteins of 62-, 68-, and 130-kDa molecular mass during shape change (Fig. 4). In contrast, native LDL did not stimulate the tyrosine phosphorylation of these proteins (data not shown). Stimulation of protein tyrosine phosphorylation induced by LPA and mox-LDL showed a rapid homologous desensitization (LPA/LPA and mox-LDL/mox-LDL) and cross-desensitization (LPA/mox-LDL and, vice versa, mox-LDL/ LPA; data not shown) and was specifically inhibited by the LPA receptor antagonist NPTyrPA (Fig. 4). Thus, mox-LDL stimulated tyrosine phosphorylation during shape change through activation of LPA-receptor(s). Tyrosine phosphorylation of these proteins reached maximum levels (2-3-fold) 15-30 s after the addition of mox-LDL and decreased subsequently (see below).
Platelets contain several protein tyrosine kinases that may be stimulated by mox-LDL. We studied the tyrosine kinase Syk, because previous studies have shown that this kinase is rapidly activated in human platelets stimulated by thrombin and other agonists including high concentrations of LPA (26 -28). As shown in Fig. 5, mox-LDL rapidly enhanced the tyrosine phosphorylation of Syk, which is known to increase its enzymatic activity. Tyrosine phosphorylation reached maximum levels 15-30 s after the addition of mox-LDL and declined subsequently. In 21 experiments using platelets from different donors, the tyrosine phosphorylation of Syk was measured before and 30 s after exposure to mox-LDL (0.5 mg/ml) (i.e. the time point of maximal shape change) and was found to increase 2.5 Ϯ 1.0-fold over basal level (mean Ϯ S.D.). Dose-response curves of mox-LDL showed that the extent of stimulation of protein tyrosine phosphorylation of p62, p68, p130, and the kinase Syk correlated with the degree of platelet shape change ( Fig. 5c and data not shown).
Because the ␣ IIb ␤ 3 -integrin is a possible binding protein and a signaling molecule for LDL on platelets (14, 16), we investi- gated whether the ␣ IIb ␤ 3 -integrin is involved in the platelet response to mox-LDL. Studies were performed using (a) thrombasthenic platelets deficient in ␣ IIb ␤ 3 -integrin (Ͻ5% of control), (b) EGTA pretreatment that dissociates ␣ IIb ␤ 3 -integrin, and (c) RGDS that inhibits ligand (fibrinogen, von Willebrand factor, fibronectin) binding to the ␣ IIb ␤ 3 -integrin. The stimulation of tyrosine phosphorylation of p62, p68, p130, and Syk during shape change induced by mox-LDL was in thrombasthenic platelets not different from that seen in control platelets (Fig.  6). Also, concentration-response curves of mox-LDL showed that the magnitude of shape change responses was similar in thrombasthenic and control platelets (data not shown). Furthermore, EGTA pretreatment did not inhibit the stimulation of protein tyrosine phosphorylation during shape change induced by mox-LDL (data not shown). These data demonstrate that the ␣ IIb ␤ 3 -integrin does not participate in the early platelet response induced by mox-LDL and that outside-in signaling through the ␣ IIb ␤ 3 -integrin is not involved in the stimulation of protein tyrosine kinases by mox-LDL.
Increase of Tyrosine-phosphorylated p62, p68, and p130 and the Tyrosine Kinases Src, Fyn, and Syk in the Cytoskeleton during Shape Change Induced by mox-LDL-Shape change is associated with a profound reorganization of the cytoskeleton (17). Resting discoid platelets contained 3-5% of tyrosine-phosphorylated p62, 6 -9% of p68, and 10 -20% of p130 in the actin-rich cytoskeleton. During platelet shape change induced by mox-LDL, the content of tyrosine-phosphorylated p62, p68, and p130 increased maximally 2.9-, 7.2-, and 2.4-fold, respectively (Table I). One possible mechanism for the increased tyrosine phosphorylation of cytoskeletal proteins is the translocation of protein-tyrosine kinases to the cytoskeleton. We therefore studied the subcellular distribution of several tyrosine kinases in resting platelets and during shape change induced by mox-LDL. Unstimulated platelets contained 4 -7% of Fyn, 3-6% of Src, and 0.7-1.4% of Syk in the cytoskeleton, indicating that these tyrosine kinases are associated to a different extent with the cytoskeleton in resting platelets. Interestingly, the association of Fyn with the cytoskeleton increased more (4.8-fold) than the association of Src (1.5-fold) and Syk (3.1-fold) with the cytoskeleton during shape change induced by mox-LDL. The translocation of active protein-tyrosine kinases to the cytoskeleton may be important for the phosphorylation of specific cytoskeletal target proteins.
We also studied the effect of BAPTA-AM/EGTA treatment, which inhibits cytosolic Ca 2ϩ fluxes in activated platelets (23). This treatment barely affected the increase of the cytoskeletal association of the protein-tyrosine kinases Src, Fyn, and Syk after stimulation with mox-LDL. BAPTA-AM/EGTA reduced the tyrosine phosphorylation of p62, p68, and p130 in resting platelets (data not shown) as observed previously (20), thereby leading to a decrease of these tyrosine-phosphorylated proteins in the cytoskeleton of unstimulated cells. Protein tyrosine phosphorylation in resting platelets is kept in balance by the activity of protein tyrosine kinases and protein-tyrosine phosphatases, and its decrease by BAPTA-AM treatment may be caused by the reduction of the basal cytosolic Ca 2ϩ concentration under 80 nM as observed previously (23). Stimulation of platelets by mox-LDL increased in BAPTA-AM/EGTA-pretreated cells the protein tyrosine phosphorylation in whole cells (data not shown) and in the cytoskeleton (Table I). Therefore, it seems that normal basal levels of cytosolic Ca 2ϩ (80 -100 nM) are required for the tyrosine phosphorylation of p62, p68, and p130 in resting platelets. However, in platelets stimulated by mox-LDL (present study) or thrombin-receptor activation (20), the stimulation of protein tyrosine phosphorylation, which is mediated by Src family tyrosine kinases (see below), is not secondary to an increase of cytosolic Ca 2ϩ .

Effect of Genistein and the Src Family Kinase Inhibitors PP1 and PD173956 on Protein Tyrosine Phosphorylation and Shape
Change Induced by LPA and mox-LDL-Further studies were performed to investigate the functional significance of the stimulation of protein-tyrosine kinases for shape change. Preincubation of platelets with the protein-tyrosine kinase inhibitor genistein inhibited platelet shape change evoked by mox-LDL,  4. Stimulation of tyrosine phosphorylation of p62, p68, and p130 during shape change by mox-LDL is mediated by activation of the LPA receptor. Protein tyrosine phosphorylation was measured in parallel to shape change (20). Platelet suspensions were incubated with ethanol (0.2%; lanes 1, 2, 5, and 7) or the LPA receptor antagonist NPTyrPA (10 M) dissolved in ethanol (lanes 3, 4, 6, and 8) for 5 min before exposure to buffer (lanes 1 and 3) indicating that protein-tyrosine kinases may be functionally involved in this early platelet response (Fig. 7a). In parallel, genistein also inhibited the increase of tyrosine phosphorylation of p62, p68, and p130 during platelet shape change (Fig. 7,  b and c). Another general tyrosine kinase inhibitor, tyrphostin A23 had similar effects (data not shown).
Also, the specific inhibitors of Src family kinases PP1 and PD173956 completely prevented the stimulation of protein tyrosine phosphorylation of p62, p68, p130, and Syk induced by LPA and mox-LDL (Fig. 8, a and b). In contrast to genistein and tyrphostin A23, they did not inhibit shape change (data not shown). These results demonstrate that activation of Src kinases leads to the increase of the tyrosine phosphorylation of p62, p68, p130, and Syk during shape change but that these biochemical events are not functionally involved in the induction of shape change by mox-LDL and LPA.

DISCUSSION
In the present study, we found that mox-LDL stimulates two early signal transduction pathways in platelets: (a) activation and cytoskeletal translocation of Src family tyrosine kinases and the tyrosine kinase Syk and (b) Ca 2ϩ entry. By using specific LPA receptor antagonists and LPA receptor desensitization, we show that the stimulation of both Ca 2ϩ entry and Src kinase-dependent protein tyrosine phosphorylation induced by mox-LDL was dependent on activation of LPA receptor(s). Stimulated platelets can produce and release LPA (29,30), which may mediate the action of mox-LDL. However, based on the following reasons we conclude that LPA present in mox-LDL, and not LPA released from platelets is responsible for the observed effects: (a) we have previously shown that LPA is formed during mild oxidation of LDL and is the active compound in mox-LDL responsible for platelet shape change (18); (b) only mox-LDL and not native LDL induced in the present study the stimulation of Ca 2ϩ entry and protein tyrosine phos-FIG. 5. mox-LDL activates Syk during platelet shape change. a, Western blot of Syk immunoprecipitates from platelets. Platelets were stimulated for various times with mox-LDL (0.5 mg/ml) to induce shape change, or platelets were aggregated with thrombin (1 unit/ml) as indicated. Platelet lysates were incubated with specific anti-Syk antibody (lanes 1-10) or rabbit IgG (lane 11). Syk immunoprecipitates were immunoblotted with anti-phosphotyrosine (anti-PY) and anti-Syk antibodies. b, graphic representation of results of stimulation of Syk tyrosine phsophorylation by mox-LDL. Anti-phosphotyrosine immunoblots of Syk-immunoprecipitates were evaluated by densitometry from six experiments with platelets from different donors (mean Ϯ S.E.). c, concentration-response curves of mox-LDL for shape change and tyrosine phosphorylation of p62, p68, and Syk. Suspensions of washed platelets were exposed to various concentrations of mox-LDL for 30 s. Aliquots (0.05 ml) of the platelet suspensions were transferred to sample buffer for measurement of tyrosine phosphorylation of p62 and p68. The remaining samples (0.6 ml) were lysed and processed for Syk immunoprecipitation. Tyrosine-phosphorylated protein bands were measured by densitometry. The experiment is representative of two others. phorylation; (c) the stimulation of Ca 2ϩ entry and Src kinasedependent protein tyrosine phosphorylation and Syk activation was similarly elicited by exogenous LPA; (d) under our experimental conditions, it is unlikely that platelets produced LPA, since the platelet stimulation was weak, and the extracellular Ca 2ϩ concentration necessary for LPA formation by platelets was too low (31).
mox-LDL and LPA induced mainly Ca 2ϩ influx across the plasma membrane and relatively little Ca 2ϩ mobilization from intracellular stores, a finding unique for these stimuli and not observed for other platelet agonists (17,32). In concordance with the small Ca 2ϩ mobilization by LPA from intracellular stores, LPA has been reported to induce only little if any phosphoinositide breakdown by phospholipase C and production of inositol 1,4,5-trisphosphate in indomethacin-treated platelets (31,33). Interestingly, we observed the Ca 2ϩ response only upon stimulation with high concentrations of LPA and mox-LDL, whereas low concentrations of LPA and mox-LDL in the presence of EGTA could induce full shape change and stimulate protein tyrosine phosphorylation in the absence of an increase in cytosolic Ca 2ϩ . These results were further supported by experiments with the cytosolic Ca 2ϩ chelator BAPTA. After pretreatment of platelets with BAPTA-AM, which inhibited any increase of cytosolic Ca 2ϩ due to Ca 2ϩ mobilization in ADP-stimulated platelets, shape change after LPA and mox-LDL stimulation was delayed, but the maximal response was barely reduced (Fig. 3c). Moreover, mox-LDL still induced the translocation of Src, Fyn, and Syk to the cytoskeleton and increased the tyrosine phosphorylation of cytoskeletal proteins after BAPTA-AM/EGTA treatment (Table I). Together, these results indicate that shape change, translocation of tyrosine kinases to the cytoskeleton, and the increase of protein-tyrosine phosphorylation are not secondary to an increase of cytosolic Ca 2ϩ after platelet stimulation with mox-LDL and LPA.
Our results show that stimulation of protein tyrosine phosphorylation and Ca 2ϩ entry are two separate early signal pathways in platelets triggered by LPA receptor activation. It is unclear at present if the platelet LPA receptor belongs to one of the three putative seven-transmembrane domain receptors for LPA that have been cloned recently (for a review, see Ref. 34). It is tempting to speculate that the activation of Src family and Syk tyrosine kinases and the stimulation of Ca 2ϩ mobilization are mediated by the activation of two separate receptors. Al- Platelets were then stirred for 3 min and exposed to mox-LDL (100 g/ml) or LPA (80 nM). Aliquots of the platelet suspensions before (C) and 30 s after exposure to mox-LDL or LPA were transferred to sample buffer for measurement of protein phosphorylation (a) or lysed for Syk immunoprecipitation (b). The experiment is representative of two others. ternatively, the same LPA receptor may activate different signal transduction pathways. LPA receptor activation has been shown to couple to several heterotrimeric G proteins such as G i and G q (35). In platelets, the LPA receptor through coupling to G i may activate Src and Syk tyrosine kinases, whereas G q activation may lead to the Ca 2ϩ response. Indeed, G i -mediated Src-kinase and Syk activation has been observed in other cell types (36,37).
We further found that the stimulation of these two early signal transduction pathways of mox-LDL and LPA is induced during, but not functionally related to, shape change. The dissociation of these two early signal pathways from shape change is remarkable and at first sight surprising. However, it has been observed previously that epinephrine, which elicits on its own a small, phospholipase C-independent increase of cytosolic Ca 2ϩ and enhances Ca 2ϩ influx to other agonists, does not induce shape change (38 -40). Also, protein tyrosine phosphorylation and Syk kinase can be stimulated in platelets without leading to shape change (22). These studies and our present work demonstrate that Ca 2ϩ influx, stimulation of protein tyrosine phosphorylation, and Syk activation can be evoked independently of platelet shape change. The rapid stimulation of these early signal pathways by mox-LDL or LPA possibly regulates platelet responses subsequent to shape change such as aggregation and secretion. Cytosolic Ca 2ϩ is known to regulate cytosolic phospholipase A 2 and the receptor-mediated release of arachidonate from phospholipids in platelets (41), and we have observed previously that the activation of phospholipase A 2 , leading to the release of arachidonate from platelet phospholipids, is the triggering step for platelet aggregation and secretion by mox-LDL (6). The early activation of Src family kinases and Syk by mox-LDL observed in our study may be involved in ␣ IIb ␤ 3 -integrin activation, which is required for platelet aggregation (42). Recent studies have shown that tyrosine kinases, particularly Syk, participate in integrin ␣ IIb ␤ 3 exposure (43,44).
The inhibition of shape change by the broad tyrosine kinase inhibitors genistein and tyrphostin A23 upon platelet stimulation with mox-LDL corroborates our previous study, which reported the inhibition of shape change by these substances after thrombin receptor activation (20). We conclude that either unidentified tyrosine kinases are involved in shape change or that these substances also inhibit nontyrosine kinases such as Rho kinase and myosin light chain kinase, which regulate shape change (23,45). The observation that genistein inhibited the mox-LDL-induced shape change is of potential medical interest, since this natural isoflavonoid is present in certain foods and is increased in plasma after a diet rich in soy proteins (46).
In conclusion, we have shown that mox-LDL stimulates through LPA receptor activation two early separate signal transduction pathways in human platelets. The rapid activation of Src family kinases and Syk kinase and the stimulation of Ca 2ϩ influx by mox-LDL is likely to play a role in platelet aggregation occurring subsequent to shape change. We suggest that oxidatively modified LDL, particularly mm-LDL or mox-LDL, could use similar mechanisms to stimulate other cells involved in the pathogenesis of atherosclerosis such as monocytes, endothelial cells, and vascular smooth muscle cells. Antagonists of the LPA receptor(s) and inhibitors of tyrosine kinases such as the dietary isoflavonoid genistein may be beneficial by dampening the pathological stimulation of platelets and vascular cells by oxidatively modified LDL.