J Biol Chem, Vol. 274, Issue 40, 28293-28300, October 1, 1999

Platelet Shape Change Is Mediated by both Calcium-dependent and -independent Signaling Pathways
ROLE OF p160 Rho-ASSOCIATED COILED-COIL-CONTAINING PROTEIN KINASE IN PLATELET SHAPE CHANGE^*

Benjamin Z. S. Paul, James L. Daniel^§, and Satya P. Kunapuli^§¶

From the Departments of  Pharmacology and ^¶ Physiology and the ^§ Sol Sherry Thrombosis Research Center, Temple University Medical School, Philadelphia, Pennsylvania 19140

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Platelets undergo shape change upon activation with agonists. During shape change, disc-shaped platelets turn into spiculated spheres with protruding filopodia. When agonist-induced cytosolic Ca²⁺ increases were prevented using the cytosolic Ca²⁺ chelator, 5,5'-dimethyl-bis-(o-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (5,5'-dimethyl-BAPTA), platelets still underwent shape change, although the onset was delayed and the initial rate was dramatically decreased. In the absence of cytosolic Ca²⁺, agonist-stimulated myosin light chain phosphorylation was significantly inhibited. The myosin light chain was maximally phosphorylated at 2 s in control platelets compared with 30 s in 5,5'-dimethyl-BAPTA-treated platelets. ADP, thrombin, or U46619-induced Ca²⁺-independent platelet shape change was significantly reduced by staurosporine, a nonselective kinase inhibitor, by the selective p160 Rho-associated coiled-coil-containing protein kinase inhibitor Y-27632, or by HA 1077. Both Y-27632 and HA 1077 reduced peak levels of ADP-induced platelet shape change and myosin light chain phosphorylation in control platelets. In 5,5'-dimethyl-BAPTA-treated platelets, Y-27632 and HA 1077 completely abolished both ADP-induced platelet shape change and myosin light chain phosphorylation. Our results indicate that Ca²⁺/calmodulin-stimulated myosin light chain kinase and p160 Rho-associated coiled-coil-containing protein kinase independently contribute to myosin light chain phosphorylation and platelet shape change, through Ca²⁺-sensitive and Ca²⁺-insensitive pathways, respectively.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Platelets are anucleate cells that mediate hemostasis through amplifying an initial stimulus and aggregating at a site of injury. Several agents including thrombin, ADP, and thromboxane A2 activate platelets. Activated platelets change shape, secrete -granules and dense granules, and release positive feedback mediators (1). When platelets are initially stimulated, the first event is a rearrangement of the cytoskeletal proteins (actin and myosin), and the normally disc-shaped cells change into spheres with filopodia (1, 2). Activation of phospholipase A₂ releases arachidonic acid from membrane phospholipids, which is converted into thromboxane A₂. Serotonin and ADP, released from dense granules, and thromboxane A₂ function as positive feedback mediators, which recruit more platelets into a primary hemostatic plug (3).

We have been investigating the intracellular events involved in agonist-induced platelet activation. Recently, we proposed a three-receptor model to explain the signaling events in platelet activation by ADP (for reviews, see Refs. 4 and 5). Through the use of receptor-specific antagonists, we have provided evidence for different functional roles for different P2 receptor subtypes present on platelets (6, 7). Platelets express two G protein-coupled P2 receptor subtypes: a P2T_AC receptor subtype, coupled to the inhibition of adenylyl cyclase via the heterotrimeric inhibitory G-protein, G_i, and P2Y1 receptor, coupled to the heterotrimeric protein G_q. Activation of the P2Y1 receptor results in the activation of phospholipase C, production of diacylglycerol, and mobilization of cytosolic Ca²⁺ in response to IP₃ production. ATP acts as an antagonist at both the P2Y1 and P2T_AC receptors while acting as an agonist at the ionotropic P2X₁ receptor, the third ADP receptor on platelets (7, 8). While ADP-induced platelet aggregation requires coactivation of both the P2Y1 and the P2T_AC receptors (9, 10), activation of the P2Y1 receptor is sufficient to cause ADP-induced platelet shape change (6). Similarly, activation of G_q-coupled 5HT_2A receptors by serotonin is also sufficient to induce shape change (9, 10).

During platelet shape change, the discoid cells undergo cytoskeletal changes including the disassembly of a microtubule ring that results in an intermediate spherical shape. This is followed by actin polymerization and the slower extension of filopodia (11-13). Previous reports have shown that a strong correlation exists between phosphorylation of the regulatory myosin light chain and the initiation of shape change (14). Agonist-dependent phosphorylation of platelet myosin correlates with its polymerization and association with actin filaments (14-17). The concentration-response curve of ADP-induced myosin light chain phosphorylation closely parallels that of shape change, while both responses have the same half-maximal inhibitory concentration (IC₅₀) toward ATP (14). Myosin light chain kinase is present in platelets (18) and is activated in vitro by Ca²⁺ and calmodulin (19).

Small GTP-binding proteins have been implicated in rearrangement and activation of cytoskeletal proteins (20). The superfamily of small GTP-binding proteins is divided into subfamilies including Rho, Rac, and Cdc42. There are three forms of Rho proteins (Rho stands for Ras homologous) including RhoA, RhoB, and RhoC (for a review, see Ref. 20) that control the assembly and disassembly of the actin cytoskeleton in many cell types in response to extracellular signals (21). The activated GTP-bound form of Rho associates specifically with five protein kinases designated as p120 protein kinase N (p120^PKN), RhoA-binding kinase  (p150^ROK),¹ RhoA-binding kinase  (p150^ROK, p160 Rho-associated coiled-coil-containing protein kinase (p160^ROCK), and p164 Rho kinase (22-26). The pyridine derivative, Y-27632, has been shown to selectively inhibit p160^ROCK with an IC₅₀ of ~1 µM. This compound has a higher specificity for p160^ROCK (200-fold) than PKA or the PKC isoforms present in rat brain; furthermore, its specificity for myosin light chain kinase is 2000-fold lower than p160^ROCK (27). Y-27632 has been shown to selectively inhibit both the activity of p160^ROCK immunoprecipitated from human platelets and the involvement of this specific kinase in smooth-muscle contraction (24, 27). The homopiperazine derivative, HA 1077, has a slightly lower binding affinity for p160^ROCK than Y-27632, but it is also more selective for this kinase than PKC, protein kinase A, and myosin light chain kinase (27, 28).

Following the observation that shape change depends upon stimulation of a G_q-coupled receptor (6, 7, 9), our investigation has focused on the role of intracellular signaling events mediating shape change. Here, we provide evidence that platelet shape change incorporates both Ca²⁺-dependent and -independent mechanisms for cytoskeletal rearrangement. We have used 5,5'-dimethyl-BAPTA to prevent the increase in cytosolic Ca²⁺ that occurs following G_q activation. Previous studies (e.g. Jen et al. (29)) have successfully used 5,5'-dimethyl-BAPTA to prevent increases in cytosolic Ca²⁺ concentration without deleterious effects on either cell viability or morphology. Through the use of p160^ROCK-selective inhibitors, Y-27632 and HA 1077, we investigated the role of the RhoA/p160^ROCK pathway in platelet response. We show that both Ca²⁺-calmodulin-dependent myosin light chain kinase and the RhoA/p160^ROCK pathways contribute to ADP-induced platelet shape change and regulation of myosin light chain phosphorylation.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Apyrase (type V), ADP, fibrinogen, and bovine serum albumin (fraction V) were from Sigma. The acetoxymethyl ester of Fura PE-3 was from Teflabs (Austin, TX). The acetoxymethyl ester of 5,5'-dimethyl-BAPTA, U46619 (a stable analog of thromboxane A₂), Staurosporine, HA 1077, GF 109203X (bisindolylmaleimide I), Ro 31-8220 (bisindolylmaleimide IX), and phorbol-12-myristyl-13-acetate were from BioMol (Plymouth Meeting, PA). Bovine thrombin was from Parke, Davis and Co. (Detroit, MI). Y-27632 was a gift from Yoshitomi Pharmaceutical Industries, Ltd. (Osaka, Japan). SC-57101 was a gift from Searle Research and Development (Skokie, IL). Ultrapure acrylamide gel reagents were from ICN (Costa Mesa, CA) except for Tris base, dithiothreitol, and glycine, which were purchased from Fisher. All other chemicals were reagent grade, and deionized water was used throughout.

Preparation of Fura PE-3 and 5,5'-Dimethyl-BAPTA-loaded Platelets-- Human blood was collected from a pool of informed healthy volunteers, all of whom are students or staff at Temple University School of Medicine. The donated blood was collected into a one-sixth volume of ACD (2.5 g of sodium citrate, 1.5 g of citric acid, and 2.0 g of glucose in 100 ml of deionized H₂O). Platelet-rich plasma was isolated by centrifugation of citrated blood at 180 × g for 15 min at room temperature. Platelet-rich plasma was incubated at 37 °C with 3 µM Fura PE-3 acetoxymethyl ester and 1 mM acetylsalicylic acid for 15 min followed by the addition of either 50 µM 5,5'-dimethyl-BAPTA acetoxymethyl ester or a corresponding volume of the vehicle, dimethyl sulfoxide, and further incubation for 30 min. After 15 min at room temperature, the platelet-rich plasma was centrifuged at 1000 × g for 10 min at room temperature. The platelet pellet was resuspended in calcium-free HEPES-buffered Tyrode's solution (7) supplemented with 0.2% bovine serum albumin and 20 µg/ml apyrase. The platelet count was adjusted to 2 × 10⁸ cells/ml. All experiments were repeated at least three times using platelets from different donors.

Measurement of Ca²⁺ with Fura PE-3 and Platelet Activation-- Aliquots (1.0 ml) of the platelet suspension were stirred in a water-jacketed cuvette maintained at 37 °C during activation. Fluorescence was constantly measured using a Perkin-Elmer LS-5 spectrofluorimeter with settings of 340 nm (excitation) and 510 nm (emission). Fura PE-3 fluorescence signals were calibrated as described previously (30). F_min was determined by the addition of 2 mM EGTA and 20 mM Tris base. F_max was determined by lysing the cells with 40 µM digitonin in the presence of saturating CaCl₂.

Platelet Aggregation and Analysis of Shape Change-- Agonist-induced platelet aggregation was determined by measuring the transmission of light through a 0.5-ml sample of aspirinated washed platelets (2 × 10⁸ cells/ml) with stirring in a lumiaggregometer at 37 °C (Chrono-Log, Havertown, PA). The base line was set using 0.5 ml of Tyrode's solution as a blank. Aggregation of washed platelets required the addition of fibrinogen (1 mg/ml) prior to the addition of an agonist, with the recorder output speed set to 0.2 mm/s. Platelet shape change was observed by the addition of 10 µM SC-57101 before agonist stimulation as described earlier (6, 14). SC-57101 is a known inhibitor of platelet aggregation through blocking fibrinogen binding to its receptor (31). Three shape change curve characteristics were measured. First, the time from the addition of agonist to initiation of shape change. Second, the time from initiation of shape change to the point where shape change is half complete (or time to reach half-maximal light absorbance). Third, the initial rate constant of the shape change curve. In order to better resolve these three shape change curve characteristics, results were printed out using a Kipp and Zonen BD-41 (Fisher) analog recorder with the output speed set to 1 mm/s. The initial rate constant was determined using the program Kaleidagraph (Synergy Software, Reading, PA) by graphing the fraction of shape change complete (fraction of maximal light absorbance) versus time, and the resulting points were fit with the exponential equation, y = 1  e^kt, where k represents the initial rate constant of platelet shape change and t is the time.

Measurement of Myosin Light Chain Phosphorylation-- The percentage of 20-kDa myosin light chain in the phosphorylated form was determined using a protocol (32) that was adapted from a modification (33) of the method described by Perrie and Perry (34). In brief, aspirinated platelets were resuspended in Tyrode's solution at a concentration of 2 × 10⁹ cells/ml. Aliquots (0.5 ml) were stirred at 37 °C during stimulation in the lumiaggregometer. At specific time points, 25 µl of 6.6 N HClO₄ was added, and the resulting acid precipitate was collected and chilled on ice. The pellets were centrifuged at 10,000 × g for 2 min followed first by rinsing and then resuspension in 1 ml of ice-cold deionized water. The protein was again pelleted by centrifugation at 10,000 × g for 2 min. Protein pellets were dissolved in 50 µl of sample buffer containing 8 M urea, 20 mM Tris, 122 mM glycine, 5 mM dithiothreitol, pH 8.6, with approximately 0.1% bromphenol blue dye. The suspended pellets were further dissolved by sonication in a Branson (Shelton, CT) sonication bath. Gel electrophoresis was performed using 10% polyacrylamide slab gels containing 40% (v/v) glycerol with a 3.6% polyacrylamide stacking gel containing 8 M urea in a Bio-Rad model 220 (100-mm) gel apparatus. The running buffer used in the top chamber was 20 mM Tris, 122 mM glycine at pH 8.6 containing 4 mM urea. The samples were loaded onto the gels and electrophoresed at 8-9 mA for each gel plate being used. The electrophoresis was stopped 1 h after the bromphenol blue marker dye had come off the bottom of the gel. Gels were stained for 1 h in 0.05% (w/v) Coomassie Brilliant Blue R-250, destained, and scanned using a Hoeffer (San Francisco, CA) scanning densitometer hooked up to a Macintosh II computer via a National Instruments Corporation (Austin, TX) DAQ conversion board. The density peaks correlating to the phosphorylated and nonphosphorylated myosin light chains as well as the 16-kDa band were analyzed using the program Kaleidagraph (Synergy Software, Reading, PA). The data points were fit using a three-peak gaussian equation. The amount of total myosin light chain in the phosphorylated form was determined by dividing the area of the phosphorylated peak by the combined areas of the phosphorylated and the nonphosphorylated myosin light chain peaks. Results were expressed as the percentage of total myosin light chain in the phosphorylated form.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Effects of 5,5'-Dimethyl-BAPTA on Agonist-induced Increase in Cytosolic Ca²⁺, Platelet Aggregation, and Platelet Shape Change-- To investigate the role of the increase in cytosolic Ca²⁺ in agonist-induced platelet shape change, we loaded the platelets with Fura PE-3, either alone or in combination with the cytosolic Ca²⁺ chelator 5,5'-dimethyl-BAPTA, and measured the fluorescence response. The normal increase in the cytosolic Ca²⁺ concentration, which occurs in response to ADP, thrombin, and U46619 (Fig. 1A), did not occur in the platelets loaded with 5,5'-dimethyl-BAPTA (Fig. 1B). These traces are representative of experiments performed to establish (and reconfirm) the absence of an increase in cytosolic Ca²⁺ due to Ca²⁺ chelation by 5,5'-dimethyl-BAPTA.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   The effect of 5,5'-dimethyl-BAPTA on agonist-induced increase in cytosolic Ca2+. Aspirinated platelets labeled with Fura PE-3 were previously treated either with vehicle (dimethyl sulfoxide) (A) or with 50 µM 5,5'-dimethyl-BAPTA acetoxymethyl ester (B) and then stimulated with the indicated agonist in a cuvette maintained at 37 °C with stirring (900 rpm). The concentrations of agonists used were 10 µM ADP, 0.1 unit/ml thrombin, and 1 µM U46619. The arrows indicate the addition of agonist.

In the absence of an increase in cytosolic Ca²⁺, platelets did not aggregate in response to ADP, thrombin, or U46619 (Fig. 2), indicating that an increase in cytosolic Ca²⁺ is essential for fibrinogen receptor activation. However, agonist-induced platelet shape change still occurred. Close examination revealed that shape change in platelets treated with 5,5'-dimethyl-BAPTA possess different characteristics from shape change occurring in control platelets. It is first apparent that the rate of increase in light absorbance (indicating the change in platelet shape from disc to spiny sphere) was significantly slower. Moreover, a delay in the initiation of shape change following the addition of an agonist can be observed. After repeating each of these conditions three or more times in three different donors, we quantitated and compared these three features of platelet shape changes in normal platelets and 5,5'-dimethyl-BAPTA-treated platelets. In order to do so, the recorder's printout speed was increased to give greater resolution. In the absence of an increase in cytosolic Ca²⁺ concentration, the time to initiate shape change increased substantially for all agonists examined (Fig. 3A). The time for half-completion of shape change was also dramatically increased in 5,5'-dimethyl-BAPTA-treated platelets (Fig. 3B). A significant decrease in the initial rate of shape change was observed for U46619 (~40% of control) and both ADP and thrombin (~60% of control) (Fig. 3C).

View larger version (13K): [in this window] [in a new window]   Fig. 2.   The effect of 5,5'-dimethyl-BAPTA on agonist-induced platelet aggregation. Platelet aggregation was measured as described. The ordinate represents the observed changes in light absorbance (optical density) due to light scattering by the platelets. Aspirinated platelets were previously treated with either vehicle (dimethyl sulfoxide) and are labeled control or with 50 µM 5,5'-dimethyl-BAPTA acetoxymethyl ester as indicated. The arrow indicates the addition of 10 µM ADP, 0.1 unit/ml thrombin, or 1 µM U46619 into a cuvette maintained at 37 °C with stirring (900 rpm). The traces are representative of three experiments.

View larger version (35K): [in this window] [in a new window]   Fig. 3.   The effect of 5,5'-dimethyl-BAPTA on three characteristics of agonist-induced platelet shape change. Platelet shape change was induced by 10 µM ADP, 0.1 unit/ml thrombin, or 1 µM U46619 in aspirinated platelets that were previously treated with either vehicle (dimethyl sulfoxide, labeled control) or with 50 µM 5,5'-dimethyl-BAPTA acetoxymethyl ester as indicated. Agonist-induced platelet shape change characteristics were measured as described under "Experimental Procedures." The time from the addition of agonist until the initiation of platelet shape change (A), the time to half-complete shape change (point at which half-maximal light absorbance was reached) (B), and the initial rate of light absorbance during platelet shape change (change in absorbance/s) (C) as compared with control values are presented. Results are presented as mean values ± S.E. (n = 6). The effect of 5,5'-dimethyl-BAPTA as compared with control is significant in each case (p < 0.01, Student's unpaired t test).

Effect of 5,5'-Dimethyl-BAPTA on Agonist-induced Myosin Light Chain Phosphorylation-- In normal platelets following stimulation by ADP, thrombin, and U46619, there was a large increase in the percentage of phosphorylated myosin light chain (Fig. 4). When the increase in cytosolic Ca²⁺ concentration was prevented with 5,5'-dimethyl-BAPTA, the levels of phosphorylated myosin light chain were dramatically reduced in platelets stimulated with ADP, thrombin, and U46619 compared with vehicle-loaded platelets. The difference in levels of phosphorylated myosin light chain in 5,5'-dimethyl-BAPTA-treated platelets following agonist stimulation was significant (p < 0.05; n = 3). In the absence of agonist stimulation, there was a very minor increase in the level of phosphorylated myosin light chain in 5,5'-dimethyl-BAPTA-treated platelets; however, the difference between the level of phosphorylation in unstimulated treated platelets and unstimulated control platelets was not significant (Fig. 4B).

View larger version (34K): [in this window] [in a new window]   Fig. 4.   Effect of 5,5'-dimethyl-BAPTA on levels of agonist-stimulated myosin light chain phosphorylation. A, alkaline-urea-PAGE of HClO4 pellets from platelets treated with either dimethyl sulfoxide (labeled Control) or 50 µM 5,5'-dimethyl-BAPTA acetoxymethyl ester (labeled 5,5'-dimethyl BAPTA). A 0.5-ml volume of washed and aspirinated platelets (2 × 109 cells/ml) was stimulated with one of the following: 10 µM ADP, 0.5 unit/ml thrombin, or 1 µM U46619 before the addition of HClO4 at the indicated times. Each sample was treated at 37 °C with stirring (900 rpm). The top band is the nonphosphorylated 20-kDa myosin light chain (indicated by MLC, and the band in the middle position is the phosphorylated 20-kDa myosin light chain (indicated by MLC-P). The bottom band is the 16-kDa myosin light chain. The above results are representative of three experiments each performed using platelets from different donors. B, densitometric analysis of the above experiment. These results are representative of the results from three experiments.

Measuring the amount of phosphorylated myosin light chain at the above time points (2, 20, and 45 s) gave only a partial indication of the signaling events following agonist stimulation. Shape change begins 2 s after the addition of ADP to control platelets. This is in contrast to 5,5'-dimethyl-BAPTA-treated platelets, in which ADP-induced shape change did not begin until after 7.5 s following the addition of agonist. Therefore, we analyzed the changes in myosin light chain phosphorylation over time in vehicle-treated control platelets and compared our findings to changes in myosin light chain phosphorylation over time in 5,5'-dimethyl-BAPTA-treated platelets (Fig. 5). Both the extent and rate of myosin light chain phosphorylation were dramatically inhibited in the absence of an increase in cytosolic Ca²⁺. A peak in myosin light chain phosphorylation occurred at 2 s in control platelets in contrast to a lesser peak in myosin light chain phosphorylation occurring at 30 s in 5,5'-dimethyl-BAPTA-treated platelets.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   The effect of 5,5'-dimethyl-BAPTA on the time course of myosin light chain phosphorylation induced by ADP. ADP (10 µM) was added at time 0 to samples (0.5 ml) of washed and aspirinated platelets (2 × 109 cells/ml) with stirring at 37 °C. Samples of platelets were either untreated or previously treated with 5,5'-dimethyl-BAPTA as described under "Experimental Procedures." Each data point is the mean ± S.E. percentage value (n = 3) of the amount of myosin light chain in the phosphorylated form. The samples that make up each of these data points are from the platelets of different donors. The reaction was stopped by the addition of 25 µl of 6.6 N HClO4 at indicated times directly into the cuvette. All agonist-induced increases in the level of myosin light chain phosphorylation are significantly greater than unstimulated levels (p < 0.05, Student's unpaired t test) except for the 45-s time point in 5,5'-dimethyl-BAPTA-treated platelets.

Effect of Protein Kinase C Inhibitors on Ca²⁺-independent Shape Change-- ADP-induced shape change is mediated through stimulation of the G_q-coupled P2Y1 receptor (6). G_q-coupled receptors mediate the activation of serine/threonine PKC isoforms (35). Using two cell-permeable inhibitors of PKC, Ro 31-8220 and GF 109203X, we investigated whether PKC stimulated by ADP, thrombin, or U46619 mediates Ca²⁺-independent platelet shape change. The -, -, and -isoforms of PKC are potently inhibited by Ro 31-8220 and GF 109203X with in vitro IC₅₀ values of 5-27 nM and 16-20 nM, respectively (36). The compound GF 109203X inhibits the -isoform of PKC with an in vitro IC₅₀ value of 210 nM and inhibits the -isoform with an in vitro IC₅₀ value of 132 nM (37). The compound Ro 31-8220 inhibits the -isoform of PKC with an in vitro IC₅₀ of 24 nM (38). The activity of these compounds was determined by blocking aggregation induced by 100 µM phorbol-12-myristyl-13-acetate (not shown). Neither of these compounds had an effect on agonist-induced platelet shape change in either the presence or absence of an increase in cytosolic Ca²⁺ (data not shown).

Effect of Staurosporine on Ca²⁺-independent Shape Change-- Staurosporine is a potent and nonspecific inhibitor of both tyrosine and serine/threonine protein kinases. We used staurosporine to establish the potential role of a Ca²⁺-independent kinase pathway in platelet shape change. We observed that ADP-induced shape change was inhibited by ~56% in vehicle-treated platelets pretreated with 0.3 µM staurosporine. Inhibition of shape change by staurosporine also occurred in control platelets stimulated by thrombin or U46619 (data not shown). However, 0.3 µM staurosporine completely inhibited shape change induced by ADP, U46619, and thrombin in 5,5'-dimethyl-BAPTA-treated platelets (data not shown), suggesting a role for kinase activity in Ca²⁺-independent shape change.

Effect of p160^ROCK-selective Inhibitors on Ca²⁺-independent Shape Change-- The RhoA/p160^ROCK pathway has been shown to play a role in smooth muscle contraction (27, 39, 40) as well as a role in the contractile responses of fibroblasts (41), endothelial cells (42), and neuronal cell lines (43-45). Staurosporine has recently been reported to dramatically inhibit ROK at a concentration of 1 µM (46). Hence, we investigated the role of the RhoA/p160^ROCK pathway in Ca²⁺-independent platelet shape change. Both HA 1077 and Y-27632 show selective inhibition of p160^ROCK purified from human platelets with IC₅₀ values of ~2 µM and 1-1.5 µM, respectively (27). In vehicle-treated platelets, 10 µM HA 1077 (Fig. 6A) and 10 µM Y-27632 (Fig. 6B) inhibited the extent of platelet shape change by ~30% and ~35% each. In the absence of an increase in cytosolic Ca²⁺ concentration caused by 5,5'-dimethyl-BAPTA, both 10 µM HA 1077 and 10 µM Y-27632 completely abolished ADP-induced platelet shape change. Moreover, the IC₅₀ for the inhibition of shape change by HA 1077 was ~1.2 µM (Fig. 6A), and that for Y-27632 was ~1.1 µM (Fig. 6B), in excellent agreement with that for inhibition of purified platelet p160^ROCK (27). In control platelets, Y-27632 (10 µM) did not inhibit ADP-induced platelet aggregation (Fig. 7A); however, it inhibited the extent but not the rate of ADP-induced platelet shape change (Fig. 7B). We investigated whether the effects of Y-27632 on platelet shape change in both control and 5,5'-dimethyl-BAPTA-treated platelets are reversible. PRP was incubated with Y-27632 (10 µM) for 30 min at 37 °C. Upon resuspension in HEPES-buffered Tyrode's solution, these platelets behaved no differently than control platelets (Fig. 7). Thus, the effects of Y-27632 appear to be reversible.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Comparison of inhibition of ADP-induced platelet shape change by HA 1077 and Y-27632 in control and 5,5'-dimethyl-BAPTA-treated platelets. Platelet shape change was induced by 10 µM ADP in aspirinated platelets that were previously treated with either vehicle (dimethyl sulfoxide, labeled control) or with 50 µM 5,5'-dimethyl-BAPTA acetoxymethyl ester (labeled 5,5'-dimethyl BAPTA) as indicated. Platelets were incubated with increasing doses of either HA 1077 (A) or Y-27632 (B), and shape change was measured as described under "Experimental Procedures." Shape change (as measured by an increase in the light absorbance) in the presence of inhibitor was normalized to shape change in control platelets (maximal light absorbance possible and designated as 100% shape change). Each point is the mean ± S.E. of three experiments.

View larger version (12K): [in this window] [in a new window]   Fig. 7.   The reversible effect of Y-27632 on shape change in control and 5,5'-dimethyl-BAPTA-treated platelets in contrast to agonist-induced platelet aggregation. Aspirinated platelets were pretreated either with vehicle (dimethyl sulfoxide; labeled Control) or with 50 µM 5,5'-dimethyl-BAPTA acetoxymethyl ester as indicated. The arrow indicates the addition of 10 µM ADP to a cuvette maintained at 37 °C with stirring (900 rpm). Platelet samples that were incubated with 10 µM Y-27632 for 5 min before the addition of ADP are labeled Y-27632. Samples labeled Y-27632 (PRP) indicate the addition of 10 µM Y-27632 to the platelet-rich plasma (PRP) followed by incubation at 37 °C for 30 min. The platelets were then washed and exposed to 10 µM ADP in the same manner as the corresponding control.

Effect of HA 1077 and Y-27632 on ADP- stimulated Myosin Light Chain Phosphorylation-- We investigated the contribution of the p160^ROCK to myosin light chain phosphorylation in Ca²⁺-dependent and -independent pathways. The effect of Y-27632 (10 µM) on the extent of ADP-induced myosin light chain phosphorylation in platelets at different time points was measured. This phosphorylation was significantly decreased at all time points in comparison with control platelets (Fig. 8). The difference in the levels of phosphorylated myosin light chain at 2 and 5 s were the most significant (p < 0.01; n = 3). The increase in myosin light chain phosphorylation that occurred in 5,5'-dimethyl-BAPTA-treated platelets in response to 10 µM ADP was completely abolished by 10 µM Y-27632 (Fig. 8). The effects of HA 1077 on ADP-induced myosin light chain phosphorylation were very similar to those of Y-27632. At 2 s following the addition of ADP, the peak level of phosphorylated myosin light chain in control platelets was reduced to similar levels by both 10 µM Y-27632 and 10 µM HA 1077 in 5,5'-dimethyl-BAPTA-treated platelets (Fig. 9). Furthermore, in 5,5'-dimethyl-BAPTA-treated platelets, both Y-27632 and HA 1077 abolished the peak level of myosin light chain phosphorylation observed at 30 s to levels observed in unstimulated platelets.

View larger version (23K): [in this window] [in a new window]   Fig. 8.   The effect of Y-27632 and 5,5'-dimethyl-BAPTA on the time course of myosin light chain phosphorylation induced by ADP. ADP (10 µM) was added at time 0 to samples (0.5 ml) of washed and aspirinated platelets (2 × 109 cells/ml) with stirring at 37 °C. Samples of platelets were either untreated or previously treated with 5,5'-dimethyl-BAPTA as described under "Experimental Procedures." Each data point is the mean ± S.E. percentage value (n = 3) of the amount of myosin light chain in the phosphorylated form. The samples that make up each of these data points are from the platelets of different donors. The reaction was stopped by the addition of 25 µl of 6.6 N HClO4 at indicated times directly into the cuvette. The points represent mean ± S.E. percentage values (n = 3) of total myosin light chain that is phosphorylated. All agonist-induced increases in the level of myosin light chain phosphorylation are significantly greater than unstimulated levels (p < 0.05, Student's unpaired t test) except for the samples treated with both 5,5'-dimethyl-BAPTA and 10 µM Y-27632. For comparison purposes, control responses from Fig. 5 are shown as thin lines.

View larger version (33K): [in this window] [in a new window]   Fig. 9.   Comparison of the effect of Y-27632 with HA 1077 on peak levels of myosin light chain phosphorylation induced by 10 µM ADP in both normal and 5,5'-dimethyl-BAPTA-treated platelets. Densitometrical analysis of alkaline-urea-PAGE. The values represent the percentage of total myosin light chain in the phosphorylated form. A 0.5-ml volume of washed and aspirinated platelets (2 × 109 cells/ml) was stimulated either with 10 µM ADP for 2 s in vehicle-treated, control platelets or for 30 s in 5,5'-dimethyl-BAPTA-treated platelets before the addition of HClO4. Samples were incubated with 10 µM Y-27632 or HA 1077 for 3 min with stirring at 37 °C before agonist stimulation. Data are expressed as the means ± S.E. from three experiments. The effects of HA 1077 and Y-27632 in 5,5'-dimethyl-BAPTA-treated platelets as compared with each matched control are significant (p < 0.01, Student's unpaired t test).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Shape change is considered to be the first measurable physiological response produced by platelets following exposure to an agonist. We and others have previously shown that ADP or serotonin-induced shape change is solely mediated by G_q-coupled receptors on platelets (6, 9, 10). Since activation of G_q leads to mobilization of calcium from cytosolic stores and activation of PKC, it is expected that these signaling events play an important role in agonist-induced platelet shape change. Offermanns et al. (47) determined that G_q is essential for platelet aggregation by producing G_q-deficient mice. Interestingly, platelets from these mice still undergo agonist-stimulated shape change even in the absence of an increase in the cytosolic Ca²⁺ concentration. Previous observations indicated that substantially lower cytosolic Ca²⁺ concentrations were sufficient for agonist-stimulated shape change and myosin light chain phosphorylation than the Ca²⁺ concentrations required for the Ca²⁺ ionophore, ionomycin, to produce the same responses (48, 49). Ionomycin induces a maximum of 70% phosphorylated myosin light chain at a cytosolic Ca²⁺ concentration of 1 µM. However, in the absence of external Ca²⁺ and in the presence of the fluorescent Ca²⁺ indicator quin2, thrombin (0.5 unit/ml) caused a ~40% increase in myosin light chain phosphorylation, and platelet-activating factor (20 ng) caused a 22% increase in myosin light chain phosphorylation. Both increases in myosin light chain phosphorylation occurred at a cytosolic Ca²⁺ concentration of ~200 nM (49). Hence, we investigated the role of increasing cytosolic Ca²⁺ concentration in platelet shape change and the contribution of Ca²⁺-independent mechanisms to shape change, using the following platelet agonists: ADP, thrombin, and U46619.

We used 5,5'-dimethyl-BAPTA, a high affinity Ca²⁺ chelator, to prevent increases in cytosolic Ca²⁺. 5,5'-Dimethyl-BAPTA (50 µM) abolished agonist-induced cytosolic Ca²⁺ increases (Fig. 1). Under these conditions, we clearly show that shape change still occurs, although an increase in the cytosolic Ca²⁺ concentration is essential for both the rapid initiation and completion of shape change (Figs. 2 and 3).

An agonist-stimulated increase in the concentration of cytosolic Ca²⁺ is required for rapid myosin light chain phosphorylation (Fig. 5). We have confirmed that shape change is preceded by myosin light chain phosphorylation, as has been previously reported (14). Our results indicate that the maximum myosin light chain phosphorylation is occurring ~2 s following the addition of ADP and that the extent of phosphorylation is substantially diminished by 15 s. In the absence of an increase in cytosolic Ca²⁺, the onset of platelet shape change was delayed. Changes in the kinetics of shape change were mirrored by a dramatic decline in the time and peak level of myosin light chain phosphorylation. In control platelets, we observed shape change beginning soon after the peak in myosin light chain phosphorylation, which is in agreement with the findings reported earlier (14). In the absence of an increase in cytosolic Ca²⁺, the level of myosin light chain phosphorylation peaked well after shape change had begun. It appears that an increase of ~50% above the basal level of phosphorylated myosin light chain is sufficient to initiate shape change. In contrast to control platelets, where the level of phosphorylated myosin light chain decreased after 2 s, the phosphorylation of myosin light chain in 5,5'-dimethyl-BAPTA-treated platelets continued to increase throughout shape change. Thus, both calcium-dependent and -independent pathways contribute to agonist-induced platelet shape change.

We examined the possible contribution of calcium-independent protein kinase activity by using protein kinase-selective inhibitors. Since ADP and serotonin induce shape change through G_q-coupled receptors, leading to activation of PKC, we used selective PKC inhibitors, GF 109203X (36) and Ro 31-8220 (50), to examine the possible role of PKC in calcium-independent shape change. These agents did not have any effect on platelet shape change, either alone or in combination with 5,5'-dimethyl-BAPTA. In agreement with our finding, it has been reported that PKC did not have any effect on agonist-induced Ca²⁺ sensitization of smooth muscle from guinea pig vas deferens (51). Since these compounds are known to inhibit the classical PKC isoforms , , and  as well as the novel isoforms  and  (36-38), these isoforms probably do not contribute to either Ca²⁺-dependent or Ca²⁺-independent mechanisms of platelet shape change. However, the role of other PKC isoforms in these processes cannot be ruled out.

Myosin light chain phosphorylation plays a central role in agonist-stimulated smooth muscle contraction (52, 53). A receptor-mediated increase in cytosolic Ca²⁺ binds calmodulin and activates the Ca²⁺/calmodulin-dependent myosin light chain kinase. Myosin light chain kinase primarily phosphorylates myosin light chain at Ser-19, which induces the interaction of actin and myosin, resulting in increased actin-stimulated myosin ATPase activity and smooth muscle contraction (53, 54). Analogous to the situation in platelets, it has been reported that levels of smooth muscle cell contraction are not always proportional to cytosolic Ca²⁺ concentration (55). In smooth muscle cells, the Rho family of Ras-like small GTPases has been identified as a mediator in the enhancement of smooth muscle cell sensitivity to Ca²⁺-induced contraction. RhoA activates RhoA-binding kinase, which phosphorylates the myosin-binding subunit (MBS) of myosin phosphatase and inhibits its activity (56, 57). RhoA-binding kinase has also been shown to directly phosphorylate myosin light chain and activate myosin in vitro as well as inducing smooth muscle contraction in the absence of Ca²⁺ (40).

Human platelets contain myosin phosphatase consisting of a 38-kDa catalytic subunit of protein phosphatase type 1 , a 130-kDa MBS, and a 20-kDa subunit (46, 58). High levels of RhoA protein are also found in platelets, and both RhoA and p160^ROCK co-immunoprecipitate with anti-MBS antibodies (46). Hence, initially we used staurosporine to test for the possibility that p160^ROCK plays a role in shape change (46, 58). Staurosporine had been shown to prevent the phosphorylation of the MBS of platelet myosin phosphatase by recombinant p160^ROCK (46). In our study, staurosporine (0.3 µM) partially inhibited ADP, thrombin, or U46619-induced shape change in control platelets and completely blocked shape change in 5,5'-dimethyl-BAPTA-treated platelets.

Since staurosporine is a potent inhibitor of many tyrosine and serine/threonine protein kinases including myosin light chain kinase, we decided to use the more selective p160^ROCK inhibitors, Y-27632 and HA 1077. Both of these compounds had a similar inhibitory effect on the extent of shape change (~30%) in control platelets (Fig. 6). In the presence of 5,5'-dimethyl-BAPTA, both Y-27632 and HA 1077 completely abolished agonist-induced platelet shape change. The IC₅₀ values of Y-27632 and HA 1077 for inhibiting shape change in 5,5'-dimethyl-BAPTA-treated platelets (Fig. 6) are similar to those for inhibiting purified human platelet p160^ROCK (27). Furthermore, both Y-27632 and HA 1077 reduced myosin light chain phosphorylation during shape change (Fig. 9). The abrogation of both shape change (Fig. 6) and myosin light chain phosphorylation in 5,5'-dimethyl-BAPTA-treated platelets (Figs. 8 and 9) by both Y-27632 and HA 1077 provides strong evidence that RhoA-activated p160^ROCK is mediating the Ca²⁺-independent shape change.

An outline of the intracellular signaling events leading to platelet shape change and their regulation is shown in Fig. 10. One target of p160^ROCK has been shown to be myosin phosphatase, which is inactivated by phosphorylation (24, 46, 56-58). While myosin phosphatase dephosphorylates the myosin light chain and counteracts Ca²⁺/calmodulin-dependent myosin light chain kinase, inactivation of myosin phosphatase would lead to an increase in myosin light chain phosphorylation. It is also possible that p160^ROCK directly phosphorylates myosin (39, 40). The presence of two systems regulating myosin phosphorylation and shape change may at first appear unnecessarily redundant, but there are countless examples in biology of redundancy. In addition, dual activation of myosin phosphorylation by kinase activation and phosphatase inhibition may allow for a more rapid and robust response to an external signal. In human endothelial cells, thrombin has been shown to activate MLC phosphatase through a Rho/Rho-operated kinase pathway as part of a signaling network that controls myosin phosphorylation and endothelial cell contractility (42).

View larger version (19K): [in this window] [in a new window]   Fig. 10.   Model depicting the intracellular events mediating platelet shape change. The solid arrows indicate a stimulatory effect, and dashed arrows indicate an inhibitory effect. Double bars indicate the inhibitory action of Y-27632 and HA 1077.

We previously have shown that the G_i-coupled receptor activation, by ADP or epinephrine, does not induce shape change (6, 7, 9). Offermanns et al. (47) have shown that shape change occurs in G_q-deficient platelets. G₁₁ is not expressed in human platelets (59). Thus, the calcium-independent shape change observed in our study and in the G_q-deficient mice (47) is mediated by G proteins other than G_i, G_q, or G₁₁. It is possible that G₁₂ or G₁₃ mediate Ca²⁺-independent shape change, since it has already been shown that Rho is regulated by G₁₂ and G₁₃ (60).

Since the submission of this paper, Klages et al. (61) have demonstrated a Ca²⁺-independent mechanism for thromboxane A₂-mediated shape change in G_q-deficient mouse platelets. Agonist stimulation of G_q-deficient mouse platelets occurs without mobilization of Ca²⁺ into the cytoplasm (47). In complete agreement with our results, it was observed that 10 µM Y-27632 totally inhibited shape change in G_q-deficient mouse platelets (61). Klages et al. (61) have proposed that Ca²⁺-independent shape change is mediated by either G₁₂ or G₁₃ (Fig. 10). Since ADP failed to elicit a shape change response in G_q-deficient mouse platelets (47), Offermanns and co-workers (47, 61) did not establish whether the calcium-independent pathway is a general mechanism of platelet shape change or specific for U46619 alone. We have demonstrated with three platelet agonists that both calcium-sensitive and -insensitive pathways independently contribute to platelet shape change and myosin light chain phosphorylation, suggesting that this is the general mechanism of platelet shape change.

In conclusion, we have demonstrated that agonist-induced platelet shape change occurs through both calcium-dependent and -independent mechanisms. RhoA/p160^ROCK appears to play an important role in the calcium-independent pathway leading to shape change.

	ACKNOWLEDGEMENTS

We thank Drs. J. Bryan Smith and Barrie Ashby (Department of Pharmacology) for critically reviewing the manuscript.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grant HL60683 and the Temple University M.D./Ph.D. program (to B. Z. S. P.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Recipient of an Established Investigator award in Thrombosis from the American Heart Association and Genentech. To whom correspondence should be addressed: Dept. of Physiology, Temple University School of Medicine, 3420 N. Broad St., Philadelphia, PA 19140. Tel.: 215-707-4615; Fax: 215-707-4003; E-mail: kunapuli@nimbus.temple.edu.

	ABBREVIATIONS

The abbreviations used are: ROK, RhoA-binding kinase; ROCK, Rho-associated coiled-coil-forming kinase; P2T_AC, platelet ADP receptor coupled to inhibition of adenylate cyclase; MBS, myosin-binding subunit; 5,5'-dimethyl-BAPTA, 5,5'-dimethyl-bis-(o-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid; PKC, protein kinase C.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES