Introduction

Top Abstract Introduction Methods Results Discussion References

Many platelet proteins become phosphorylated at tyrosine residues during agonist stimulation (Ferrel and Martin, 1988; Golden and Brugge, 1989; Nakamura and Yamamura, 1989; Oda et al., 1992; Bachelot et al., 1992), although the functional significance of this is not clear. Tyrosine phosphorylation might play a pivotal role in platelet signal transduction (Pumiglia et al., 1992) where signaling enzymes, such as PLC-, phosphatidylinositol-3 kinase and nonreceptor tyrosine kinase including the src gene products, may cluster with interactions between these molecules stabilized by their SH2 domains (Sadowski et al., 1986; Ullrich and Schlessinger, 1990; Koch et al., 1991). Tyrosine phosphorylation may also be induced by [Ca⁺⁺]_i elevation or by Ca⁺⁺ depletion in intracellular Ca⁺⁺ storage site (Vostal et al., 1991).

MAPK is a family of serine/threonine kinase that appears to be a component common to signaling pathway initiated by a wide range of factors including hormones, differentiation factors and mitogens (Tomas, 1992; Nishida and Gotoh, 1993; Davis, 1993). Phosphorylation of tyrosine and threonine residues is required for the activation of MAPK (Chatani et al., 1992), which is mediated by a dual phosphorylating specificity of MAPK kinase (Nishida and Gotoh, 1993; Davis, 1993). The MAPK pathway is not a single pathway, but represents a common mechanism of signal transduction that has been adapted to couple different stimuli to distinct physiological responses. Ligand binding to a receptor protein tyrosine kinase triggers the activation and autophosphorylation of the kinase, which leads to activation of ras, a small GTP-binding protein. Activated ras then initiates a cascade of sequential MAPK (Avruch et al., 1994; Cobb et al., 1994). MAPK is well characterized as one of intracellular signaling enzyme involved in proliferation of cells. Because platelets are nonproliferative cells, the signal transduction pathway, including MAPK, cannot lead to a mitogenic signal and instead may regulate cytoskeletal or secretory changes during platelet activation (Papkoff et al., 1994).

Upon stimulation, PIP₂ is cleaved by PLC to form IP₃ and diacylglycerol. IP₃ releases Ca⁺⁺ from the intracellular stores and rises the cytosolic-free Ca⁺⁺ level (O'Rourke et al., 1985), although diacylglycerol activates PKC, leading to protein phosphorylation, granule secretion and fibrinogen-receptor expression (Rink et al., 1983). PLC-2 is immunologically distinct from PLC-1, but does contain the SH2 regions (Kumjian et al., 1991). When PLC-2 is transfected and overexpressed in rat-2 fibroblasts, PLC-2 is transiently tyrosine-phosphorylated by PDGF, with a similar kinetics to those for PLC-1 (Sultzman et al., 1991). However, several isoforms of PLC, e.g., PLC-, -1, -2 and -, exist in platelets (Banno et al., 1992) and are capable of contributing to increase in diacylglycerol and IP₃. However, the physiological function and control of each isoform in platelets is still relatively undefined.

Recent studies have revealed that the regulation of cPLA₂ is complex and involves both changes in [Ca⁺⁺]_i mediated by PLC activation and phosphorylation of cPLA₂ (Clark et al., 1991; Lin et al., 1992; Nemenoff et al., 1993). It has also been shown that the p42 MAPK phosphorylates cPLA₂ that results in a 3- to 4-fold increase in specific activity in the presence of a high Ca⁺⁺ concentration (Nemenoff et al., 1993; Lin et al., 1993), and that arachidonic acid liberation may be regulated by activated MAPK and increase in [Ca⁺⁺]_i in human platelets (Nakashima et al., 1994).

ZT-A was recently isolated from a symbiotic marine alga Symbiodinium sp. as potent vasoconstrictor compound (Nakamura et al., 1993, 1995), and characterized as large molecules (2872 Da) containing a large number of oxygen and olefinic carbons. The chemical characters are different from those of other marine toxins such as palytoxin and maitotoxin that exhibit potent vasoconstrictor activities (Asari et al., 1993; Nakamura et al., 1993). We previously reported that ZT-A was capable of eliciting aggregation accompanied with an increase in [Ca⁺⁺]_i and TXA₂ release in rabbit washed platelets (Rho et al., 1995). These stimulatory effects of ZT-A were inhibited by indomethacin, a cyclooxygenase inhibitor, and genistein, a protein tyrosine kinase inhibitor. Our study is designed to determine if the platelet activation elicited by ZT-A in rabbit washed platelets involves the activation of tyrosine kinase, PLC-2, MAPK and PLA₂.

	Methods

Top Abstract Introduction Methods Results Discussion References

Materials. ZT-A was isolated as described previously (Nakamura et al., 1993). Indomethacin was obtained from Merck Company Inc. (Rahway, NJ). Ionomycin, genistein, staurosporine and calphostin C were obtained from Wako Pure Chemical Industries (Osaka, Japan). BSA were from Sigma Chemical CO. (St. Louis, MO). [³H]TXB₂ (110.6 Ci/mmol), [¹⁴C]Arachidonic acid (50 mCi/mmol) and [³H]myo-inositol (23.4 Ci/mmol) were obtained from Du Pont/NEN (Boston, MA). [-³²P]ATP was from Amersham International plc. (Buckinghamshire, England). [³H]Arachidonic acid (220 Ci/mmol) was from Moravek Biochemicals (Brea, CA). Anti-TXB₂ serum and TXB₂ were given to us by ONO Pharmaceutical (Osaka, Japan). Antiphosphotyrosine antibody (4G10) was obtained from Upstate Biotechnology Incorporate (Lake Placid, NY). Anti-MAPK antibody (ERK2, C-14) and anti-PLC-2 antibody (Q-20) were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Protein A-Sepharose was obtained from ZYMED Laboratories, Inc. (South San Francisco, CA). The reagents required for the MAPK assay, including a specific peptide substrate related to a sequence of the epidermal growth factor receptor, were provided in the form of a BIOTRAK MAPK assay kit, currently available from Amersham International plc. PD98059 was obtained from New England Biolabs, Inc. (Bevely, MA). All other chemicals were of analytical grade.

Preparation of washed platelets. Fresh blood was obtained from male rabbits (Japanese white rabbits weighing about 2-3 kg), collected into plastic tubes containing acid citrate dextrose solution (1/6 volume of blood) composed of citric acid (65 mM), trisodium citrate (85 mM), and dextrose (2%) at pH 4.5, subsequently centrifuged at 250 × g for 10 min to obtain platelet-rich plasma. Platelet-rich plasma was centrifuged at 650 × g for 10 min at room temperature (20-25°C). The pellet was washed twice with Tyrode/HEPES solution (pH 6.35). The resultant pellet was resuspended in the second Tyrode/HEPES solution (pH 7.35) with a final density of approximately 5 × 10⁸ platelets/ml (Ardlie et al., 1970; Vickers et al., 1982). The Tyrode/HEPES solution was composed of NaCl 138.3 mM, KCl 2.68 mM, MgCl₂ · 6H₂O 1.0 mM, NaHCO₃ 4.0 mM, HEPES 10 mM, glucose 0.1% (w/v) and albumin 0.35% (w/v) at pH 6.35 or 7.35.

Determination of platelet aggregation. Platelet aggregation was determined by a standard turbidometric method (Born, 1962) using an aggregometer (PAM-6C, Merbanix, Tokyo, Japan). Platelet aggregation was expressed as an increase in light transmission. The levels of light transmission were calibrated as 0% for a platelet suspension and 100% for the Tyrode/HEPES solution. Platelet suspension (0.3 ml) in the aggregometer cuvette was preincubated for 2 min at 37°C under continuous stirring at 1000 r.p.m. and then CaCl₂ or EGTA was added at the final concentration of 1 mM. After 2 min, ZT-A was added and platelet aggregation was monitored for 10 min. Various blockers were preincubated for 5 min before the addition of ZT-A.

Immunoblotting. For analysis of total platelet proteins, the reaction was terminated by addition of Laemmli sample buffer, and the mixture was then boiled for 5 min and analyzed on an 8% SDS-PAGE. Immunoblot assays were performed as described by Papkoff et al. (1994), with slight modifications. For immunoblotting, proteins were electrically transferred to the polyvinylidene difluoride membrane for 80 min at 120 mA. Blots were incubated for 2 hr with 1% (w/v) BSA in TBS to block residual protein binding sites. Immunodetection of tyrosine phosphorylation was achieved by using a specific antiphosphotyrosine monoclonal murine antibody (4G10, 2 µg/ml) in TBS containing 1% BSA for 2 hr. MAPK was detected by rabbit anti-MAPK (ERK2, C-14) antibody (1 µg/ml) in TBS containing 1% BSA for 2 hr. PLC-2 was detected by rabbit anti-PLC-2 antibody (Q-20, 1 µg/ml) in TBS containing 1% BSA for 2 hr. The primary antibody was removed and blots were washed in TBS with 0.05% Tween-20 five times. To detect the primary antibody, blots were incubated with alkaline phosphatase-conjugated anti-mouse or anti-rabbit antibody (Bio-Rad, Hercules, CA) diluted to 1:3000 in TBS containing 1% BSA for 2 hr, and then washed five times in TBS with 0.05% Tween-20. After the blots were exposed to enhanced chemiluminescence reagents (Bio-Rad) for 30 min, they were then exposed to Hyper film-enhanced chemiluminescence (Amersham) for 10 to 30 min.

Immunoprecipitation. For immunoprecipitation of MAPK, platelet suspensions (0.3 ml) were harvested by addition of 20 µl denaturing buffer (10% SDS, 10 mM dithiothreitol, 20 mM HEPES, pH 7.4) (Papkoff et al., 1994). The sample was heated at 95°C for 5 min and diluted with 0.8 ml of immunoprecipitation buffer (150 mM NaCl, 10% glycerol, 1% triton X-100, 1.5 mM MgCl₂, 2 mM EGTA, 10 µg/ml leupeptin, 10 µg/ml phenylmethanesulfonyl fluoride, 0.5 mM sodium orthovanadate and 50 mM HEPES, pH 7.4). The samples were centrifuged at 15,000 × g × 20 min. The supernatants (0.8 ml) were then incubated for 3 hr at 4°C with 1 µg/ml of the anti-MAPK (ERK2, C-14) antibody, and were further incubated for 1 hr at 4°C after addition of protein A-Sepharose (20 µg). Immune complexes were collected by centrifugation in a microcentrifuge and then washed three times with washing buffer (150 mM NaCl, 1% Nonidet P-40, 0.1% SDS, 1% aprotinin, 10 µg/ml leupeptin, 10 µg/ml phenylmethanesulfonyl fluoride, 0.5 mM sodium orthovanadate and 50 mM HEPES, pH 7.4). The precipitate was solubilized by Laemmli sample buffer, and the mixture was then boiled for 10 min and applied to 10% SDS-PAGE and immunoblotting described above. For immunoprecipitation of PLC-2, platelet suspensions (0.3 ml) were harvested by the addition of an equal volume of ice-cold Nonidet P-40 lysis buffer containing 2% (w/v) Nonidet P-40, 100 mM Tris-acetate (pH 8.0), 2.0 mM EDTA, 2.0 mM EGTA, 100 mM NaF, 200 mM NaCl, 6.0 mM sodium orthovanadate, 20 mM sodium pyrophosphate, 2.0 mM phenylmethylsulfonyl fluoride, 0.4 mM leupeptin and 20 µg/ml aprotinin (Tate and Rittenhouse, 1993). The samples were then incubated for 15 min on ice and spun at 15,000 × g 10 min. Each sample (1 ml) was incubated with 1 µg/ml of the anti-PLC-2 antibody (Q-20) for 2 hr, followed by an 1-hr incubation with protein A-sepharose (20 µg) at 4°C. Immune complexes were collected by centrifugation in a microcentrifuge, and then washed once with lysis buffer, twice with phosphate-buffered saline containing 150 mM NaCl, 10 mM NaH₂PO₄ (pH 7.6) and 3.0 mM sodium orthovanadate, and twice with buffer containing 100 mM Tris-HCl (pH 7.4), 500 mM LiCl and 3.0 mM sodium orthovanadate. The precipitate was solubilized by Laemmli sample buffer, and the mixture was applied to 7% SDS-PAGE and immunoblotting described above, after boiling for 10 min.

MAPK assay. Platelets were lysed by sonication in 10 mM Tris, 150 mM NaCl, 2 mM EGTA, 2 mM dithiothreitol, 1 mM sodium orthovanadate, 1 mM phenylmethansulfonyl fluoride, 10 µg/ml leupeptin, 10 µg/ml aprotinin, pH 7.4 measured at 4°C and centrifuged at 15,000 × g 20 min. Supernatants (15 µl) were then incubated for 30 min at 30°C with 10 µl of the substrate buffer containing 6 mM substrate peptide (NH₂-K R E L V E P L T P A G E A P N Q A L L R-OH), 75 mM HEPES, 300 µM sodium orthovanadate, and 0.05% sodium azide, pH 7.4, and 5 µl of ATP buffer containing 0.3 mM [-³²P]ATP (0.67 Ci/mmol) and 90 mM MgCl₂. After incubation, 10 µl of 300 mM orthophosphoric acid were added to terminate the reaction. After 30 µl of each sample were spotted onto phosphocellulose discs, the discs were washed twice for 2 min in 1% acetic acid, and then washed twice for 2 min in distilled water. The radioactivity on each disc was determined by scintillation counting. To examine the effect of indomethacin and genistein, platelets were preincubated with the drugs for 5 min at 37°C before addition of ZT-A.

Radioimmunoassay of TXB₂. The release of TXB₂ from rabbit platelets was determined by a radioimmunoassay (Matsuoka et al., 1989). In brief, the reaction of platelets was terminated by addition of equi-volume of ice-cold 50 µM indomethacin/50 mM EDTA solution, and the sample solution was centrifuged at 1700 × g for 10 min at 4°C. The assay mixture of 0.1 ml sample (1700 × g supernatant) or standard (0.1-3,000 ng/ml), 0.1 ml antiserum against TXB₂ (20,000 times dilution) and 0.1 ml [³H]TXB₂ (10 nCi) was incubated overnight at 4°C. [³H]TXB₂ bound to antiserum was assayed in a liquid scintillation spectrophotometer after the sedimentation of free [³H]TXB₂ with dextran-coated charcoal.

Analysis of arachidonic acid metabolites. Platelets were labeled with [¹⁴C]arachidonic acid (0.8 µCi/ml) at 37°C for 1 hr. Platelets were then washed twice with Tyrode/HEPES-albumin solution (pH 7.35). Platelets (0.3 ml) were incubated for 5 min with ZT-A in aggregometer cuvettes with constant stirring at 37°C. The reaction was terminated by addition of equi-volume of ice-cold 50 µM indomethacin/50 mM EDTA solution. After centrifugation, the supernatant was adjusted pH to 3.0 with 1 N HCl, and the released [¹⁴C]arachidonic acid metabolites were extracted with ethyl acetate. The ethyl acetate solubles were dried by stream of nitrogen gas and applied to thin-layer chromatography plate (LK5D, Whatman, Clifton, NJ). The developer used was benzene/isooctane/acetic acid (60:30:3, v/v). [¹⁴C]Arachidonic acid metabolites were visualized in radioluminogram with a molecular imager (GS363, Bio-Rad, Hercules, CA).

Determination of arachidonic acid liberation. For the determination of arachidonic acid liberation, platelets were labeled with [³H]arachidonic acid (10 µCi/ml) at 37°C for 1 hr. Then, platelets were washed twice with Tyrode/HEPES-albumin solution (pH 7.35). Platelets were incubated with drugs in aggregometer cuvettes with constant stirring at 37°C. The reaction was terminated by addition of equi-volume of ice-cold 50 µM indomethacin/50 mM EDTA solution (Matsuoka et al., 1989). After centrifugation, the released [³H]radioactivity in the supernatant was estimated as arachidonic acid liberation.

Measurement of inositol phosphates. Washed platelets suspended in albumin-free Tyrode/HEPES solution (pH 7.35) were labeled with 25 µCi/ml [³H]myo-inositol at 37°C for 1 hr. Platelets were washed with Tyrode/HEPES-albumin solution (pH 7.35), and resuspended at 5 × 10⁸ platelets/ml. After platelets were preincubated for 2 min, they were incubated with drugs in the presence or absence of 10 mM LiCl for 5 min. Reaction was terminated by addition of equal volume of ice-cold 10% TCA. The TCA extracts were washed three times with diethyl ether to remove TCA. Diethyl ether was removed by keeping the samples at 47°C for 30 min. [³H]IP₃ or total [³H]IPs were separated by anion exchange column (Bio-Rad AG 1X-8, 100-200 mesh, formate form) previously described (Nakahata et al., 1989). The elute was counted with a liquid scintillation spectrophotometer.

Data analysis. The results obtained were expressed as mean ± S.E. and the statistical difference was determined with unpaired Student's t test.

	Results

Top Abstract Introduction Methods Results Discussion References

Effects of indomethacin and genistein on ZT-A- and ionomycin-induced platelet aggregation. ZT-A and ionomycin caused aggregation in rabbit platelets (fig. 1). Though ZT-A- (2 µM) induced platelet aggregation was inhibited by indomethacin (10 µM) or genistein (50 µM), ionomycin- (5 µM) induced aggregation persisted after the treatment of platelets with above drugs (fig. 1). These results suggest that ZT-A activates platelets through cyclooxygenase products and tyrosine phosphorylation, not due to Ca⁺⁺ ionophore-like action.

View larger version (K): [in this window] [in a new window]   Fig. 1.   Effects of indomethacin and genistein on ZT-A- or ionomycin-induced platelet aggregation. Aggregation was measured by turbidometric method. Drugs were added 5 min before addition of ZT-A (2 µM) (A) or ionomycin (5 µM) (B) in the presence of 1 mM CaCl2: 1, indomethacin (10 µM); 2, genistein (50 µM); 3, vehicle (dimethyl sulfoxide, final 0.05%). A typical experiment representative of three is shown.

Effects of ZT-A on tyrosine phosphorylation. The stimulation of platelets by ZT-A (2 µM) resulted in an increase in the amount of tyrosine phosphorylation of 135-, 110-, 105- and 42-kDa proteins (fig. 2A). To determine whether 42-kDa protein is MAPK or not, platelet lysates were immunoprecipitated with anti-MAPK (ERK2, C-14) antibody and MAPK was analyzed by using of antiphosphotyrosine antibody or anti-MAPK (ERK2, C-14) antibody (fig. 2B and C). ZT-A (2 µM) increased phosphorylation of the tyrosine residue of p42 MAPK but not p44 MAPK (fig. 2B). In these analyses, the amount of immunoprecipitated p42 MAPK was unchanged by ZT-A (fig. 2C) and p44 MAPK was not immunoprecipitated with anti-MAPK (ERK2, C-14) antibody.

View larger version (K): [in this window] [in a new window]   Fig. 2.   Protein tyrosine phosphorylation upon ZT-A stimulation. Platelets were incubated with ZT-A (2 µM) for 5 min in the presence of 1 mM CaCl2. Platelet lysates were fractionated on a 8% SDS-PAGE following by immunoblotting with the antiphosphotyrosine antibody (A), and immunoprecipitated with the anti-MAPK antibody (B and C). Immunoprecipitates were fractionated on a 10% SDS-PAGE following by immunoblotting with the antiphosphotyrosine antibody (B) or anti-MAPK antibody (C) as described in "Materials and Methods."

Elevation of MAPK activity during ZT-A-induced platelet activation. MAPK activity was measured by incorporation of ³²P to the specific substrate for MAPK. ZT-A increased MAPK activity in a concentration-dependent manner with a maximum concentration of 2 µM (fig. 3A). Time course analysis revealed that ZT-A activated MAPK by two phases. Within 60 sec, ZT-A increased MAPK activity 2-fold, and then it increased 10-fold at 5 min after addition of ZT-A (fig. 3B).

View larger version (K): [in this window] [in a new window]   Fig. 3.   Effect of ZT-A on MAPK activity. Platelets were incubated with the indicated concentration of ZT-A for 5 min in the presence of 1 mM CaCl2 (A) and were incubated with ZT-A (2 µM) in the presence of 1 mM CaCl2 for the indicated time (B). MAPK activity (A, 566.0 ± 26.4 cpm/tube, B, 588.5 ± 54.8 cpm/tube) that was not treated with ZT-A in the presence of 1 mM CaCl2 was taken as 1 (control). Data are expressed as fold of control. Values represent the mean ± S.E. (n = 3) and are representative of three independent experiments.

View larger version (K): [in this window] [in a new window]   Fig. 4.   Effects of indomethacin and genistein on ZT-A-induced TXB2 release (A) and MAPK activation (B). Platelets were pretreated with indomethacin (10 µM) or genistein (50 µM) for 5 min, and then incubated with ZT-A (2 µM) for 5 min in the presence of 1 mM CaCl2 (A). After preincubation of indomethacin (10 µM) for 5 min, genistein (50 µM) was applied 5 min before addition of ZT-A (2 µM) (B). TXB2 release (0.8 ± 0.1 ng/5 × 108 platelets) and MAPK activity (674.0 ± 22.6 cpm/tube) that were not treated with ZT-A in the presence of 1 mM CaCl2 was taken as 1 (control). Values represent the mean ± S.E. (n = 4) and are representative of three independent experiments. Significant difference (*P < .05) by unpaired t test when compared to control.

Effects of PD98059 on ZT-A-induced TXB₂ release. We further investigated the effects of PD98059, a MAPK kinase 1 (MEK1) inhibitor, on ZT-A-induced TXB₂ release, because cPLA₂ can be activated by MAPK (Qui et al., 1993). PD98059 (50 µM) potently inhibited TXB₂ release in response to ZT-A (2 µM), indicating that MAPK is responsible for TXB₂ release probably via cPLA₂ (table 1).

Effect of ZT-A on arachidonic acid metabolites liberation. To confirm PLA₂ activation induced by ZT-A (2 µM), we examined the liberation of arachidonic acid and its subsequent metabolites. ZT-A (2 µM) stimulated liberation of arachidonic acid from membrane phospholipids, which was converted to 12-hydroperoxyeicosatetraenoic acid (12-HETE) and TXB₂ in platelets (fig. 5).

View larger version (K): [in this window] [in a new window]   Fig. 5.   Effect of ZT-A on liberation of arachidonic acid and its metabolites. Platelets were labeled with 0.8 µCi/ml [14C]arachidonic acid for 1 hr at 37°C. Platelets were incubated with ZT-A (2 µM) for 5 min in the presence of 1 mM CaCl2. [14C]arachidonic acid metabolites were visualized in radioluminogram with a molecular imager (GS363, Bio-Rad). The results are representative in three similar experiments.

Effect of ZT-A on phosphoinositide hydrolysis. ZT-A stimulated PI hydrolysis in a concentration-dependent manner with a maximum at 2 µM in the presence of external Ca⁺⁺ (fig. 6A). Figure 6B shows the time course of ZT-A-induced accumulation of IP₃ in the absence of LiCl. The peak of IP₃ accumulation was 30 sec after the addition of ZT-A. The time course analysis showed that ZT-A activated PLC before MAPK.

View larger version (K): [in this window] [in a new window]   Fig. 6.   Effect of ZT-A on PI hydrolysis. Platelets were labeled with 25 µCi/ml [3H]myo-inositol for 1 hr at 37°C. Platelets were incubated with various concentrations of ZT-A for 5 min (A) and incubated with ZT-A (2 µM) for the indicated time (B) in the presence of 1 mM CaCl2. The reaction was carried out in the presence (A) or absence (B) of 10 mM LiCl, and was terminated by the addition of 0.3 ml of 10% TCA. Total [3H]IPs and [3H]IP3 were determined as described in "Materials and Methods." Values represent the mean ± S.E. (n = 4) and are representative of three independent experiments.

View larger version (K): [in this window] [in a new window]   Fig. 7.   Effects of indomethacin and genistein on ZT-A-induced PI hydrolysis. Platelets were labeled with 25 µCi/ml [3H]myo-inositol for 1 hr at 37°C. Platelets were preincubated with indomethacin for 5 min (A) or with genistein for 5 min in the presence of indomethacin (10 µM) (B), then were incubated with ZT-A (2 µM) for 5 min. The reaction was carried out in the presence of 10 mM LiCl, and was terminated by the addition of 0.3 ml of 10% TCA. Total [3H]IPs were determined as described in "Materials and Methods." Values represent the mean ± S.E. (n = 3) and are representative of three independent experiments. Significant difference (*P < .05) by unpaired t test.

PLC-2 activation upon ZT-A stimulation. The stimulation of platelets by ZT-A (2 µM) resulted in an increase in protein tyrosine phosphorylation of PLC-2, determined by immunoprecipitation (fig. 8). Protein tyrosine phosphorylation of PLC-2 was partially inhibited by indomethacin (10 µM), and completely inhibited by genistein (50 µM) plus indomethacin (10 µM) (fig. 8A). In these analyses, the amount of immunoprecipitated PLC-2 was not affected by indomethacin, genistein or ZT-A (fig. 8B).

View larger version (K): [in this window] [in a new window]   Fig. 8.   Effects of indomethacin and genistein on ZT-A-induced tyrosine phosphorylation of PLC-2. Platelets were preincubated with indomethacin (10 µM) for 5 min or with genistein (50 µM) for 5 min in the presence of indomethacin (10 µM), then were incubated with ZT-A (2 µM) for 5 min. The lysates were immunoprecipitated with an anti-PLC-2 antibody (Q-20). Immunoprecipitates were fractionated on a 7% SDS-PAGE following by immunoblotting with anti-phosphotyrosine antibody (A) or anti-PLC-2 antibody (B) as described in "Materials and Methods." The results are representative in three similar experiments.

Effects of staurosporine and calphostin C on ZT-A-induced MAPK activation. Because PLC activation was responsible for activation of PKC, leading to MAPK activation (Rink et al., 1983; Qiu and Leslie, 1994), we investigated the effects of staurosporine and calphostin C, PKC inhibitors, on MAPK activation induced by ZT-A in the presence of indomethacin (10 µM). ZT-A- (2 µM) induced MAPK activation was inhibited by above drugs in a concentration-dependent manner (fig. 9). Both staurosporine and calphostin C had no effect on the phosphorylation of PLC-2 (data not shown).

View larger version (K): [in this window] [in a new window]   Fig. 9.   Effects of staurosporine and calphostin C on ZT-A-induced MAPK activation in the presence of indomethacin. Platelets were preincubated with staurosporine (A) and calphostin C (B) for 5 min, and then incubated with ZT-A (2 µM) for 5 min in the presence of indomethacin (10 µM). MAPK activity (A, 603 ± 109 cpm/tube, B, 517 ± 46) that was not treated with ZT-A in the presence of 1 mM CaCl2 was taken as 1. Values represent the mean ± S.E. (n = 3) and are representative of three independent experiments.

Effects of depletion of external Ca⁺⁺ on ZT-A-induced platelet activation. Because ZT-A-induced platelet aggregation was dependent on the presence of external Ca⁺⁺ (Rho et al., 1995), we investigated the effect of depletion of external Ca⁺⁺ on ZT-A-induced MAPK activation, TXB₂ release, arachidonic acid liberation and PI hydrolysis. These ZT-A-induced activations were also strictly dependent on the presence of external Ca⁺⁺ (table 2).

	Discussion

Top Abstract Introduction Methods Results Discussion References

Platelet activation is accompanied by the phosphorylation of numerous proteins at tyrosine residue (Findik and Presec, 1988; Golden et al., 1990; Bading and Greenberg, 1991). In an effort to clarify the roles of tyrosine phosphorylation in platelet signal transduction, we have investigated the platelet activation induced by ZT-A, which is genistein sensitive (Rho et al., 1995). ZT-A increased the tyrosine phosphorylation of p42 MAPK in rabbit platelets. In human platelets, thrombin has been shown to stimulate the activation of p42 MAPK but not p44 MAPK (Papkoff et al., 1994). PMA and platelet-activating factor also activated p42 MAPK, but not p44 MAPK in sheep platelets (Samiei et al., 1993). In agreement with other reports, we also observed that ZT-A increased protein tyrosine phosphorylation of p42 MAPK but not p44 MAPK in rabbit platelets. ZT-A also increased the tyrosine phosphorylation of 110- and 105-kDa proteins. The contribution of these proteins to p42 MAPK activity and platelet aggregation in response to ZT-A is left to be studied.

Because indomethacin and SQ-29548 inhibited ZT-A-induced platelet aggregation (Rho et al., 1995), the aggregation is mediated through TXA₂ that is known to induce platelet aggregation (Hamberg et al., 1975; Swayne et al., 1988; Ohkubo et al., 1996a). Our study demonstrates that ZT-A stimulates arachidonic acid liberation as well as MAPK in rabbit platelets. Because arachidonic acid liberation may be regulated by MAPK and increase in [Ca⁺⁺]_i in human platelets (Nakashima et al., 1994), ZT-A could activate the signaling pathway of MAPK-cPLA₂. The fact that PD98059, a MEK1 inhibitor, inhibits ZT-A-induced TXB₂ release (table 1) clearly shows that activation of MAPK is responsible for cPLA₂ activation.

Although indomethacin completely inhibited TXB₂ release and aggregation induced by ZT-A, it partially inhibited the MAPK activation induced by ZT-A. Therefore, TXA₂ released by ZT-A is able to activate MAPK and cPLA₂ in rabbit platelets, i.e., ZT-A activates cPLA₂ and releases TXA₂, which secondarily enhances cPLA₂ activation via a TXA₂ receptor. In fact, TXA₂ has been shown to activate MAPK and arachidonic acid liberation in rabbit platelets (Ohkubo et al., 1996b). Furthermore, it is assumed that the remained ZT-A-induced MAPK activation in the presence of indomethacin is independent of TXA₂. MAPK can be activated by extracellular stimuli via tyrosine kinase receptors (McCormic, 1994) or G protein-coupled receptors (Lange-Carter et al., 1993; Crespo et al., 1994). Because the remained ZT-A-induced MAPK activation in the presence of indomethacin was completely inhibited by genistein, a protein tyrosine kinase is responsible for ZT-A-induced MAPK activation independently of TXA₂. The fact that MAPK activity was increased 2-fold within 60 sec by ZT-A supports the idea that there is a primary MAPK activation in ZT-A action. In addition, ZT-A-induced platelet aggregation and p42 MAPK activation were also inhibited by tyrphostin 23, another tyrosine kinase inhibitor (Rho et al., 1997).

PKC has been shown to play a role in cPLA₂ activation (Qiu et al., 1993; Wijkander and Sundler, 1992). Diacylglycerol stimulates PKC synergistically with Ca⁺⁺, leading to activation of MAPK cascade in several cell types (Qiu and Leslie, 1994). To elucidate the primary signaling pathway of ZT-A to activate MAPK, the involvement of PLC and PKC was investigated in the presence of indomethacin. ZT-A caused genistein-sensitive PI hydrolysis in the presence of indomethacin, suggesting that tyrosine phosphorylation was involved in PLC activation. The involved PLC isoform is defined as PLC-2, based on the observation that ZT-A-induced tyrosine phosphorylation of PLC-2 in the presence of indomethacin was completely inhibited by genistein. The result is consistent with the case of thrombin or collagen in tyrosine phosphorylation of PLC-2 in human platelets (Tate and Rittenhouse, 1993; Daniel et al., 1994).

However, ZT-A-induced MAPK activation was concentration-dependently inhibited by staurosporine and calphostin C, PKC inhibitors, in the presence of indomethacin. These results imply that ZT-A primarily causes PLC-2 activation, leading to PKC and MAPK activation. In agreement with the results, ZT-A-induced platelet aggregation was also inhibited by staurosporine (Rho et al., 1995) and calphostin C (M.-C. Rho, N. Nakahata, H. Nakamura, A. Murai and Y. Ohizumi, unpublished observation).

ZT-A-induced platelet activation was strictly dependent on the presence of external Ca⁺⁺. A possibility raised that ZT-A acts as a Ca⁺⁺ ionophore. However, ZT-A- (2 µM) induced aggregation was quite different from ionomycin- (5 µM) induced one, i.e., indomethacin (10 µM) and genistein (100 µM) potently inhibited ZT-A- but not ionomycin-induced aggregation. Therefore, ZT-A does not act as a Ca⁺⁺ ionophore. Influxed Ca⁺⁺ by ZT-A may participate in its pharmacological action cooperatively with its activation of tyrosine kinase signaling pathway.

In conclusion, ZT-A may primarily activates a protein tyrosine kinase in the presence of external Ca⁺⁺. The activated protein tyrosine kinase subsequently stimulates the activation of MAPK probably via PLC-2 and PKC. The activated MAPK in turn stimulates cPLA₂ and liberates arachidonic acid that is rapidly converted to TXA₂. Released TXA₂ may enhance platelet function including MAPK activity.