Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Shear rate produced by arterial blood flow over stenotic vessels plays an important role in platelet adhesion, activation, and accretion onto the site of vascular injury (Baumgartner, 1973; Frojmovic, 1998). This process of platelet adhesion plays a key role in thrombus growth and is believed to be mediated by platelet glycoprotein Ib and IIb/IIIa receptors that engage in binding to multiple ligands, such as fibrinogen, von Willebrand factor (vWF), fibronectin, and other plasma and matrix proteins (Lefkovits et al., 1995). In addition, the interaction between collagen and platelets (through collagen receptors) plays a critical role in normal hemostasis and pathological thrombosis (Alberio and Dale, 1999). Under high shear conditions, vWF predominantly mediates platelet adhesion by interacting with both glycoprotein Ib and IIb/IIIa receptors (Ruggeri, 1993; Lankhof et al., 1995; Kroll et al., 1996; Goto et al., 1998). Binding studies revealed that vWF attachment to glycoprotein Ib precedes its binding to glycoprotein IIb/IIIa (Kroll et al., 1996), suggesting that interaction of vWF with glycoprotein Ib (Ib/IX complex) is the initial step leading to platelet activation, followed by its engagement with the glycoprotein IIb/IIIa to induce firm platelet aggregation (Chow et al., 1992; Konstantopoulos et al., 1997). Antibodies against vWF, glycoprotein Ib, or IIb/IIIa have all been shown to block shear-induced platelet aggregation and adhesion (Ruggeri, 1993). The administration of a specific monoclonal antibody against glycoprotein IIb/IIIa (c7E3 or abciximab) to patients in association with coronary angioplasty results in almost complete inhibition of shear-induced aggregation subsequent to vWF-mediated adhesion onto type collagen I (Turner et al., 1995). In this context, the Cone and Plate(let) Analyzer (CPA), but not aggregometry, has been recently shown to monitor prolonged platelet inhibition following gradual recovery of abciximab therapy (Osende et al., 2001).

To further investigate the role of glycoprotein IIb/IIIa receptors on platelet adhesion under shear conditions in vitro, we characterized various experimental conditions using the CPA system (Varon et al., 1997) to monitor platelet activation and adhesion onto an artificial (polystyrene) surface under a high shear rate (Shenkman et al., 2000; Osende et al., 2001). In the present study, we have investigated the pharmacological property of small molecule glycoprotein IIb/IIIa antagonists in various experimental conditions including aspirin, various anticoagulants, temperature, and Ca²⁺ concentration. Because substantial differences have been noted among glycoprotein IIb/IIIa antagonists in respect to their ability to bind to resting glycoprotein IIb/IIIa receptors and their "off-rate" (Mousa et al., 2000), we postulated that small-molecule glycoprotein IIb/IIIa antagonists with higher binding affinity and slower off-rate (class I) than antagonists with low affinity and high off-rate (class II) may provide better efficacy in blocking shear-induced platelet adhesion. Furthermore, the effect of non-Arg-Gly-Asp-based glycoprotein IIb/IIIa antagonists (abciximab and eptifibatide) on platelet adhesion has been compared with Arg-Gly-Asp-based glycoprotein IIb/IIIa antagonists.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials. The free acid form of roxifiban (XP280), orbofiban (YZ202), sibrafiban (YZ211), lotrafiban (DPC-A38628), and DPC802 were synthesized by Bristol-Myers Squibb Co. (Stamford, CT). Distilled water was used as vehicle for each compound in the present study. Abciximab (ReoPro), a chimeric monoclonal antibody specific for glycoprotein IIb/IIIa receptors, was obtained from Centocor, Inc. (Marvin, PA). Eptifibatide (Integrilin), a synthetic peptide antagonist of glycoprotein IIb/IIIa, was obtained from COR Therapeutics, Inc. (manufactured by Key Pharmaceuticals, Kenilworth, NJ). Tirofiban (Aggrastat), a nonpeptide glycoprotein IIb/IIIa antagonist, was from Merck Research Labs (West Point, PA). XT199, a specific small-molecule antagonist of v3 (with an IC₅₀ of 0.05 µM at the v3 receptor as opposed to an IC₅₀ of >10 µM at the glycoprotein IIb/IIIa receptor), was synthesized by Bristol-Myers Squibb Co. Citrate blood Vacutainer tubes (containing 3.8% sodium citrate) were obtained from BD Biosciences (San Jose, CA). Hirudin (Refludan, 10 µM) was purchased from Hoechst Marion Roussel, Inc. (Kansas City, MO). Four-well polystyrene plates were purchased from Nalge Nunc International (Naperville, IL). Teflon cones and the shear stress operating instrument were from Matis, Ltd. (Ramat-Gan, Israel).

Blood Samples. Venous blood was collected from healthy volunteers into either citrate or hirudin Vacutainer tubes and allowed to stay at room temperature for 30 min while gently rocking. Blood samples (500 µl) were distributed into a 2-ml Eppendorf tube and incubated with various concentrations of glycoprotein IIb/IIIa antagonists for 15 min at room temperature. In studies where the effect of aspirin was examined, donors had taken aspirin (325 mg/day) for 7 days. Blood samples from these donors were collected and handled as described above. Platelet counts (Coulter Model T540 instrument; Beckman Coulter, Inc., Fullerton, CA) were at normal ranges for all donors in this study [(2.24 ± 0.14) × 10⁸ platelets/ml, n = 15].

CPA. The procedure of the CPA operation was recently described (Osende et al., 2001). Briefly, after incubating blood samples with fibans or vehicle for 15 min at room temperature, 200 µl was transferred into each of four-well polystyrene plates. Blood samples were spun at 750 rpm × 2.5 shear rate (= 1875 s¹) for 2 min with a rotating Teflon cone, as described in detail previously (Osende et al., 2001). Wells were washed with phosphate-buffered saline for 2 to 3 times and stained with May-Grunwald dye (Sigma Catalog no. MG500; Sigma-Aldrich, St. Loius, MO) for 1 min. After washing, the stained platelets were inspected by an inverted light microscope. Four individual images were captured from each well using a digital camera and quantitated using Image Pro Plus 4.1 software (Media Cybernetics, Silver Spring, MD). Each sample was repeated for at least five times/individuals (n > 5), as indicated in the figure legends. Results were expressed as the percentage of the well surface (on the bottom) covered by platelets. Drug effect was calculated as a percentage of inhibition compared with vehicle treated blood samples.

Platelet Aggregation Assay. Blood was collected from healthy, medication-free volunteers in a sodium citrate Vacutainer. The platelet-rich plasma was collected after centrifugation for 10 min at 150g. The remaining blood was centrifuged for 15 min at 1,500g to collect platelet-poor plasma. Platelet count was determined using a Coulter Counter. For the receptor binding assay, platelet count was adjusted to 100,000 platelets/µl with platelet-poor plasma in binding buffer containing 20 mM HEPES, 137 mM NaCl, 2.7 mM KCl, 1 mM MgCl₂, 5.6 mM glucose, 1 mM CaCl₂, 1 µM hirudin, and 1 mg/ml bovine serum albumin, pH 7.4. Prostaglandin E1 (4 µg/ml) or 0.4% ethanol (vehicle) was added to resting platelets for 10 min to ensure quiescence. Platelet aggregation was determined using PAP-4 platelet aggregometer (Bio/Data, Horsham, PA) in addition to 10 or 20 µM ADP as specified in Table 1 and the legend to Fig. 6. Various concentrations of glycoprotein IIb/IIIa antagonists were incubated with platelet-rich plasma for 10 min before ADP stimulation. Samples (n = 5) were run in duplicate for 4 min at room temperature.

Glycoprotein IIb/IIIa Antagonist Binding Affinity to Resting and Activated Human Platelets. This assay was used to determine the saturable binding of a compound to platelets using platelet-rich plasma prepared from citrated blood. Platelet count was adjusted to 33,000 platelets/µl. Platelets were incubated with the binding buffer (described above) containing 0.5 nM of radiolabeled [³H]fibans or ¹²⁵I-abciximab in a total volume of 150 µl for 30 min at room temperature. For platelet activation, 10 µM ADP was added to all samples, followed by a 30-min incubation at room temperature. Bound ligand was separated from free ligand by filtration using a Packard Plate Harvester (Filtermate 196) GF/B filter plates (Hewlett Packard, Palo Alto, CA) presoaked in 0.2% polyethylenimine. Plates were washed three times with 250 µl of cold phosphate-buffered saline and allowed to dry before being read on the counter (Packard TopCount NXT plate reader). All the experiments were performed in duplicate on platelets obtained from five different donors.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Representative saturation binding curve of roxifiban and orbofiban on activated platelets. Platelet-rich plasma was prepared using Refludan anticoagulant blood, stimulated with a mixture of 10 µM ADP, 10 µM epinephrine, and 10 µM TRAP and used for saturation binding analysis (n = 4), as described in detail under Materials and Methods. 3H-Labeled roxifiban and orbofiban were mixed with unlabeled compounds to determine their saturation bindings to the receptors. Nonspecific binding was defined using 2.5 µM unlabeled roxifiban and 100 µM unlabeled orbofiban, respectively. Hill slope = 1.95 and 1.07 for roxifiban and orbofiban, respectively. , Kd value of 1.4 nM for roxifiban; , Kd value of 94 nM for orbofiban.

Dissociation Rates. For dissociation studies, 2.5 nM of radiolabeled [³H]fibans or ¹²⁵I-abciximab was incubated with platelet-rich plasma (33,000 platelets/µl) in a final volume of 150 µl for 60 min at room temperature to ensure equilibrium binding. Unlabeled compound was added at 500-fold excess from 1 to 30 min. Nonspecific binding was assessed in the presence of 1,000-fold excess concentration of unlabeled compound and was typically less than 5% of the total binding. Bound ligand was separated from free ligand by filtration and measured as described above.

Statistical Analysis. Data are presented as mean ± standard errors. IC₅₀ values and Hill slopes for concentration-response curves were generated according to best fit of all data points using GraphPad Prism software (GraphPad, San Diego, CA). Statistical comparisons were made by analysis of variance followed by Fisher's protected t test, and values were considered to be significant when p is less than 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Pharmacological Data of Various Oral Glycoprotein IIb/IIIa Antagonists. Table 1 shows the IC₅₀ of various glycoprotein IIb/IIIa antagonists using platelet-rich plasma stimulated with 10 µM ADP (except for the use of 100 µM ADP for abciximab), as well as their binding kinetics to resting and activated human platelets. DPC802 and roxifiban demonstrated a similar high affinity to both resting and activated human platelets, with a K_d value of 0.2 to 0.5 and 1.0 to 1.4 nM for DPC802 and roxifiban, respectively. Likewise, abciximab showed a high affinity to the resting and activated platelets (with a K_d value of 9.1 and 9.2 nM, respectively) and slow dissociation rate. In contrast, sibrafiban, lotrafiban, and orbofiban revealed higher affinity to activated human platelets (with a K_d value of 27, 62, and 271 nM, respectively) than to resting platelets (with a K_d value of 46, 422, and 987 nM, respectively). Orbofiban, lotrafiban, and sibrafiban also demonstrated faster dissociation rates (half-time of 2-18 s) from both resting and activated human platelets compared with those of roxifiban.

Concentration-Dependent Effect of Roxifiban and Orbofiban on Shear-Induced Platelet adhesion in Citrated Blood at Room Temperature. Because previous studies were carried out using citrated blood at room temperature (Varon et al., 1997; Shenkman et al., 2000; Osende et al., 2001), the same condition was applied for our initial study to validate our CPA system and its capacity to study the efficacy of small-molecule glycoprotein IIb/IIIa inhibitors. Quantitative data (n = 6) confirmed the dose-dependent inhibition of roxifiban, along with orbofiban, on shear-induced platelet adhesion (Fig. 2). IC₅₀ values were 38 and 396 nM for roxifiban and orbofiban, respectively.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Concentration-dependent effects of roxifiban and orbofiban on shear-induced platelet adhesion in citrated blood. Various concentrations of roxifiban (0, 10, 20, 30, 40, 50, 100, and 250 nM) and orbofiban (0, 25, 50, 250, 500, 1000, and 2000 nM) were incubated with a citrated whole-blood sample for 15 min and then subjected to the CPA test at room temperature. Each sample was quantitated using four images, and the data (the percentage of surface coverage) illustrated are the mean values ± standard errors of six independent studies (n = 6). Percentage of inhibition was calculated based on each dose-dependent study relevant to the vehicle treated samples. Hill slope = 7.17 and 2.09 for roxifiban and orbofiban, respectively. , IC50 value of 38 nM for roxifiban; , IC50 value of 396 nM for orbofiban.

Effects of Aspirin, Anticoagulant, and Temperature on Shear-Induced Platelet Adhesion. Aspirin is a widely used antiplatelet drug that blocks the cyclooxygenase pathway. Parallel experiments (n = 5) performed on citrated blood samples from donors free of aspirin or after 7 days on aspirin produced similar platelet adhesion profiles in response to either roxifiban or orbofiban (Fig. 3A).

View larger version (32K): [in this window] [in a new window]   Fig. 3.   Effects of roxifiban and orbofiban on shear-induced platelet adhesion in various experimental conditions. A, effect of blood samples from donors with or without aspirin. Blood samples were taking from the paired healthy donors before and after 7 days of aspirin. Blood samples (n = 5) were drawn in sodium citrate Vacutainer and applied to the CPA test at room temperature in the presence of 0, 40, and 100 nM roxifiban or 0, 500, and 1000 nM orbofiban, as described in the legend to Fig. 2. B, effect of roxifiban or orbofiban on citrate or hirudin anticoagulated whole blood on shear-induced platelet adhesion at room temperature (n = 11). C, effect of roxifiban or orbofiban on shear-induced platelet adhesion in citrated blood at room temperature and 37 °C (n =5). D, concentration-dependent study of roxifiban or orbofiban on shear-induced platelet adhesion in hirudin anticoagulant blood at room temperature and 37 °C (n = 6).

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Comparative analysis of small-molecule glycoprotein IIb/IIIa antagonists on shear-induced platelet adhesion in hirudin anticoagulant whole blood at room temperature. Various concentrations of roxifiban (0, 10, 30, 40, 50, 75, and 100 nM; n = 11), sibrafiban (0, 25, 50, 100, 200, 500, 1,000, and 2,000 nM; n = 6), orbofiban (0, 100, 250, 500, 1,000, 2,000, and 10,000 nM; n = 11), lotrafiban (0, 20, 200, 5,000, 1,000, and 2,000 nM; n = 5), and DPC802 (0, 10, 20, 30, 40, 50, 100, and 250 nM; n = 4) were incubated with hirudin anticoagulant blood and used for the CPA study, as described in the legend to Fig. 2. The IC50 value of each glycoprotein IIb/IIIa antagonist was determined based on the present study using the CPA system and [D]p is the plasma levels of drug applied for clinical studies. The [D]p of DPC802 is predicted based on its pharmacokinetics profile. Hill slope = 4.59, 1.55, 1.03, 1.74, and 14.18 for roxifiban (; n = 11; IC50 = 35; [D]p = 25 nM), sibrafiban (; n = 6; IC50 = 91; [D]p = 75 nM), orbofiban (; n = 11; IC50 = 606; [D]p = 150 nM), lotrafiban (; n = 5; IC50 = 438; [D]p = 42 nM), and DPC802 (; n = 4; IC50 = 34; [D]p = 10 nM), respectively.

Concentration-Dependent Effect of Small Molecule Glycoprotein IIb/IIIa Antagonists on Shear-Induced Platelet Adhesion in Hirudin Anticoagulant Blood At Room Temperature. Figure 4 illustrates the concentration-dependent effect of roxifiban, sibrafiban, orbofiban, lotrafiban, and DPC802 on shear-induced platelet adhesion in hirudin anticoagulant blood: IC₅₀ = 35, 91, 606, 438, and 34 nM, respectively (n = 4-11).

Concentration-Dependent Effect of Glycoprotein IIb/IIIa Antagonists Used as Intravenous Injection on Shear-Induced Platelet Adhesion in Hirudin Anticoagulant Blood. As shown in Fig. 5, the intravenous glycoprotein IIb/IIIa antagonists abciximab, tirofiban, and eptifibatide revealed concentration-dependent effect on shear-induced platelet adhesion in hirudin anticoagulant blood. IC₅₀ values were 43, 430, and 5781 nM for abciximab, tirofiban, and eptifibatide, respectively (n = 6). Due to the limited highest drug concentration available for eptifibatide, the data on maximal inhibition could not be obtained for this drug. In addition, because abciximab is known to have high affinity to both glycoprotein IIb/IIIa and v3 receptors (Coller, 2001), we applied a specific v3 receptor antagonist, XT-199 (with an IC₅₀ value of 0.05 µM for the v3 receptor versus an IC₅₀ >10 µM for glycoprotein IIb/IIIa) and evaluated its effect on shear-induced platelet adhesion using the CPA system. Very limited inhibition (15%, n = 4) of platelet adhesion was seen at 1000 nM XT-199.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Comparative analysis of intravenous glycoprotein IIb/IIIa antagonists on shear-induced platelet adhesion in hirudin anticoagulant whole blood at room temperature. Various concentrations of abciximab (0, 10, 30, 40, 50, 75, and 100 nM), tirofiban (0, 20, 40, 80, 200, 500, 2000, 4000, and 9010 nM), and eptifibatide (0, 800, 4000, 6000, and 9010 nM) were incubated with hirudin anticoagulant blood and used for the CPA study (n = 6 of each sample), as described in the legend to Fig. 4. Hill slope = 3.68, 2.16, and 2.6 for abciximab (; IC50 = 43; [D]p = 70 nM), tirofiban (; IC50 = 430; [D]p = 75 nM), and eptifibatide (; IC50 = 5781; [D]p = 900 nM), respectively.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Comparative analysis of glycoprotein IIb/IIIa antagonists on platelet aggregation in platelet-rich plasma stimulated with ADP. Platelet-rich plasma was prepared and used to measure platelet aggregation using an optical aggregometer upon the stimulation of 20 µM ADP. Various concentrations of roxifiban (; IC50 = 27 nM) and orbofiban (; IC50 = 105 nM) (A) or abciximab (; IC50 = 74 nM), tirofiban (; IC50 = 55 nM), and eptifibatide (; IC50 = 238 nM) (B) were added 10 min before the addition of ADP (n = 6 of each sample). Hill slope = 4.49, 2.19, 1.5, 3.46, and 1.96 for roxifiban, orbofiban, abciximab, tirofiban, and eptifibatide, respectively.

Effects of Glycoprotein IIb/IIIa Inhibitors on Platelet Aggregation Using Optical Aggregometry. Figure 6 illustrates the effects of small molecule glycoprotein IIb/IIIa antagonists (roxifiban and orbofiban) and intravenous antagonists (abciximab, tirofiban, and eptifibatide) on platelet aggregation measured by an optical aggregometer using platelet-rich plasma stimulated with 20 µM ADP. IC₅₀ = 27, 105, 74, 55, and 238 nM for roxifiban, orbofiban, abciximab, tirofiban, and eptifibatide, respectively (n = 6).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The ability of high shear to activate platelets and induce their adhesion and aggregation has been well studied (Ruggeri, 1993; Kroll et al., 1996; Frojmovic, 1998). The importance of shear in platelet adhesion and aggregation has been viewed in the perspective of arterial stenosis caused by atherosclerotic plaques in the coronary, carotid, and peripheral arteries. The interaction of vWF with the glycoprotein Ib/IX/V complex is the initial step that is followed by the binding of vWF to glycoprotein IIb/IIIa to induce platelet activation and aggregation (Kroll et al., 1996). Shear-induced platelet aggregation is often associated with platelet thrombus formation (Kroll et al., 1996). The exposed subendothelium in pathological conditions can serve as the substrate for platelet adhesion and aggregation under elevated shear stress. Such shear-induced platelet-surface interaction is known to be mediated between the functional glycoprotein IIb/IIIa receptors and collagen, vWF, fibrin(ogen), thrombospondin, laminin, or fibronectin (Kroll et al., 1996). The role of glycoprotein IIb/IIIa in shear-induced platelet adhesion and aggregation has also been demonstrated by in vivo administration of specific antagonists such as abciximab and eptifibatide (Turner et al., 1995; Kamat et al., 1997; Osende et al., 2001). Taking advantage of a recently developed CPA system (Varon et al., 1997; Shenkman et al., 2000), our present study further explored the effects of various oral and intravenous glycoprotein IIb/IIIa antagonists on shear-induced platelet activation and adhesion in vitro.

Although the present CPA system used an artificial surface for platelet adhesion under shear condition, its biological relevance has been previously demonstrated by comparing the polystyrene surface with extracellular matrix surface, showing that blocking of vWF or glycoprotein IIb/IIIa resulted in substantial decrease in platelet adhesion in both surfaces (Shenkman et al., 2000). Therefore, differences in potency for various glycoprotein IIb/IIIa antagonists as demonstrated in the present work is likely to resemble physiological conditions. It is of interest to note that among these glycoprotein IIb/IIIa antagonists tested in the CPA system, only abciximab demonstrated a lower concentration of IC₅₀ (43 nM) than the plasma levels attained in clinical application (70 nM). The high efficacy of abciximab on shear-induced platelet adhesion is independent on the high-affinity nature of abciximab to v3 receptors since the specific v3 antagonist (XT-199) failed to show such effect at relevant concentration at 1000 nM.

The substantial differences in potency of these small molecule antagonists may reflect their differences in binding affinity to resting and activated glycoprotein IIb/IIIa receptors and their off-rates. As noted, a much steeper binding curve was observed for roxifiban than orbofiban (1.8-fold increase in Hill slope; Fig. 1), which is also the case for the concentration-response curve for platelet aggregation (2-fold increase in Hill slope; Fig. 6), and shear-induced platelet adhesion (3.4- and 4.5-fold increase in Hill slope in citrated blood and hirudin anti-coagulant blood, respectively; Figs. 2 and 4). Because platelet activation and adhesion occurred rapidly (within 2 min and in fact less than 30 s for most platelets using the CPA system; our unpublished observation), the high binding affinity (or quick "on-rate") for fibans to the resting glycoprotein IIb/IIIa receptors and slow off-rate may be critical in the CPA system. Of note, roxifiban and DPC802 have very low K_d values to both resting and activated glycoprotein IIb/IIIa receptors (ranging at 0.2-1.4 nM, with a ratio of 0.7-2.5) and slow off-rate, and thus, these two antagonists demonstrated their high potency in blocking shear-induced platelet activation and adhesion. In contrast, orbofiban and lotrafiban require a high concentration to block shear-induced platelet adhesion, which was also observed in the respective 4- and 10-fold higher IC₅₀ value than plasma levels applied for clinical studies. Poor binding affinity was also observed for orbofiban and lotrafiban to the resting glycoprotein IIb/IIIa receptors compared with the activated receptors, with a ratio of 3.6:1 and 6.8:1 for orbofiban and lotrafiban, respectively. Sibrafiban was at a moderate range of potency in blocking shear-induced platelet adhesion, as was its binding affinity to the resting platelet and dissociation rate. Although no parallel study has been conducted for tirofiban and eptifibatide in the present study, the high receptor binding affinity and slow off-rate for abciximab are also associated with its high efficacy in blocking shear-induced platelet adhesion.

Several experimental conditions that have not been studied in the past have been evaluated for their effect on the CPA system, including aspirin, anticoagulant, temperature, and Ca²⁺ concentration. The increase in [Ca²⁺]_i has been associated with the platelet aggregation in response to shear force and vWF (Chow et al., 1992). EDTA chelation of extracellular Ca²⁺ completely inhibited vWF-mediated increases in [Ca²⁺]_i and aggregation responses to shear stress. Of the two anticoagulants used in the present study, sodium citrate is known to maintain only a very low free-calcium condition (i.e., 50-100 µM), whereas hirudin preserves physiological Ca²⁺ levels (Moake et al., 1998). Although Ca²⁺ concentration has been shown to figure prominently in the potency of small glycoprotein IIb/IIIa antagonists in platelet aggregation in platelet-rich plasma (Marciniak et al., 2001), our data indicate that Ca²⁺ ions have little if any bearing on shear-induced platelet activation and adhesion. Another report, however, showed that blood in hirudin anticoagulant reduced (up to 50%) platelet adhesion under shear stress (Alkhamis et al., 1993). This difference may reflect the use of different systems, i.e., the use of 5680-s¹ shear rate in the previous study versus 1875-s¹ shear rate in our current article, as well as the use of different platelet adhesion plate surfaces, i.e., tetrafluoroethylene-propylene copolymer versus polystyrene in our study.

Aspirin is a standard antiplatelet drug used in secondary and primary prevention of atherosclerotic cardiovascular disease morbidity and mortality. Aspirin inhibits platelet function by acetylation of cyclooxygenase-1 at serine 529 and thereby prevents access of substrate to the active site of the enzyme (DeWitt et al., 1990). In the present study, aspirin neither had an effect on the extent of shear-induced platelet adhesion in whole blood nor on roxifiban or orbofiban effects. These data are also in agreement with a previous article in which aspirin had little effect on the inhibition of aggregation in response to shear stress (Moake et al., 1998). Similarly, no significant difference was found for platelet adhesion at 37°C and room temperature using the CPA system.

The effect of glycoprotein IIb/IIIa inhibitors on platelet aggregation was also evaluated using a traditional aggregometer. Similar IC₅₀ values were defined for roxifiban and abciximab using aggregometry and CPA analysis (less than 2-fold), whereas marked differences were noted for orbofiban, tirofiban, and eptifibatide (IC₅₀ values were 5.8-, 7.8-, and 24-fold higher in CPA assay than aggregometry analysis, respectively). Although the clinical relevance of each assay is a subject of ongoing debate, one previous article suggested that the CPA system might more accurately evaluate the role of glycoprotein IIb/IIIa receptor function under shear stress than classical aggregation-based platelet tests (Osende et al., 2001).

In conclusion, our present study demonstrated a dose-dependent inhibition by various oral and intravenous glycoprotein IIb/IIIa inhibitors on shear-induced platelet activation and adhesion, supporting the clinical role of this integrin in platelet adhesion in this condition. Marked differences, however, were observed in the potency of these antagonists in blocking shear-induced platelet adhesion. The substantial differences in potency of these glycoprotein IIb/IIIa inhibitors may reflect their differences in binding affinity to resting and activated glycoprotein IIb/IIIa receptors and their off-rates. Of interest is our observation that of the three glycoprotein IIb/IIIa antagonists used in clinical condition, only abciximab demonstrated inhibition of shear-induced platelet adhesion at a clinically attainable plasma concentration. Taking together, the data generated under shear condition may provide additional measure for the potential efficacy of glycoprotein IIb/IIIa inhibitors in clinic.