Human Platelet Ca2+ Mobilization, Glycoprotein IIb/IIIa Activation, and Experimental Coronary Thrombosis In Vivo in Dogs Are All Inhibited by the Inotropic Agent Amrinone
Background Inotropic drugs are often used to treat acute, severe heart failure resulting from acute myocardial infarction and other unstable coronary artery syndromes. However, catecholamine inotropic agents may potentiate coronary thrombosis via a platelet α2-adrenergic mechanism, thus exacerbating the original problem. The present studies were designed to determine whether the nonadrenergic inotropic and vasodilator drug amrinone, which elevates platelet cAMP levels, would both inhibit human platelet Ca2+ mobilization and adhesion molecule expression ex vivo and protect against experimental coronary thrombosis in vivo in dogs.
Methods and Results Human platelets in suspension were preincubated with amrinone 2.5 to 15 μg/mL; stimulated with the agonists thrombin 0.1 U/mL, ADP 10−6 mol/L, or arginine vasopressin 10−7 mol/L; and studied for Ca2+ mobilization, glycoprotein IIb/IIIa activation, and P-selectin expression by fluorescent flow cytometry methods. Experimental coronary thrombosis in vivo was studied in an open-chest dog model with critical coronary artery stenosis and deep vessel wall injury. Results showed that at the cellular level, amrinone inhibited agonist-induced Ca2+ mobilization and had modest inhibitory effects on adhesion molecule expression. In vivo in dogs, intravenous amrinone 2 mg/kg plus infusion at 20 μg·kg−1·min−1 completely abolished coronary thrombosis.
Conclusions The fact that amrinone inhibited human platelet activation at the cellular level and protected against experimental coronary thrombosis in vivo in dogs suggests a potentially advantageous antithrombotic action for this inotropic and vasodilator drug.
Acute myocardial infarction and other unstable coronary artery syndromes can cause acute cardiac failure, sometimes with a severity such that inotropic agents are indicated. For example, the GUSTO-1 investigators report that as many as a quarter of patients admitted to the hospital for acute myocardial infarction receive intravenous inotropic drugs at some time during their hospital course.1 A quandary arises when catecholamine agents are used in this setting, because those with α2-adrenergic activity may potentiate coronary thrombosis via platelet α2-adrenergic receptor mechanisms.2 3 4 5 6 In contrast to the catecholamines, the inotropic and vasodilator agent amrinone7 lacks α2-adrenergic activity. Furthermore, amrinone is a phosphodiesterase III inhibitor.8 9 Phosphodiesterase III inhibitors increase cellular levels of the second messenger 3′,5′-cAMP not only in cardiac myocytes but also in platelets.10 11 12 13 14 15 16 Although cAMP serves to increase contractility in cardiac myocytes, in platelets cAMP opposes cellular activation and evokes platelet quiescence.2 13 Previous studies concerning the platelet effects of amrinone have largely been restricted to platelet aggregation experiments using light transmittance aggregometry methods.14 15 The purpose of the present study was to determine whether amrinone would inhibit both human platelet activation at the cellular level and experimental coronary thrombosis in vivo in dogs.
The first specific aim was to determine whether amrinone would inhibit agonist-induced Ca2+ mobilization in human platelets ex vivo. Within platelets, Ca2+ mobilization serves as an initial signaling step whereby external agonists such as thrombin and ADP activate platelets.2 17 Inhibition of the Ca2+ signal results in inhibition of platelet activation. The second aim, again at the cellular level, ex vivo, was to determine whether amrinone would inhibit agonist-induced adhesion molecule expression. Platelet adhesion molecules, when activated, provide the means whereby platelets become adhesive, a step essential to aggregation and thrombus formation.18 19 20 21 22 23 Two receptors, P-selectin and GP IIb/IIIa, were studied. P-selectin is released from platelet granules to the surface membrane during platelet activation, where it permits platelet binding to white blood cells and to the endothelium.23 GP IIb/IIIa is constitutively expressed on the platelet surface membrane and, when activated, serves as an anchoring site for soluble fibrinogen20 and von Willebrand factor.21 Pharmacological manipulation of this receptor has recognized relevance to the treatment of unstable coronary artery syndromes.18 19
The third specific aim was to test amrinone in vivo in a well-characterized dog model in which critical coronary artery stenosis plus intimal and medial damage generates platelet-mediated periodic thrombosis followed by distal embolizations.3 4 5 24 25 26 27 28 29 30 31 This model has been used to examine the role of α-adrenergic and serotonergic receptors, ADP and platelet-activating factor agonists, prostacyclin, GP IIb/IIIa receptor antagonists, tissue factor, and von Willebrand GP Ib binding in in vivo thrombosis in several species.3 4 5 24 25 26 27 28 29 30 31
Preparation of Platelets for Ca2+ Studies
After approval from the Institutional Human Studies Review Board, 30-mL blood samples were obtained from healthy volunteers free of platelet-active drugs, and the blood was immediately anticoagulated with 1:7 vol/vol acid citrate dextrose. Samples were centrifuged (190g for 15 minutes), platelet-rich plasma was collected, EDTA 5 mmol/L was added, and the sample was centrifuged (900g for 10 minutes) to obtain a platelet pellet. The pellet was washed, repelleted, suspended in room-temperature Tyrode’s buffer, and warmed to 37°C, and the platelets were loaded with the Ca2+-sensitive indicator indo 1, 10 μmol/L, for 30 minutes in the cell-permeant acetoxy-methyl ester form. Platelets were washed to remove excess indo 1 and resuspended in Tyrode’s buffer with 0.4% BSA and 1.4 mmol/L CaCl2.
Estimation of Apparent Platelet [Ca2+]i
Platelet suspension was incubated with and without 2.5 or 10 μg/mL amrinone for 10 minutes at 37°C. Samples (0.5 mL) were placed in the reservoir chamber of a FACStar Plus flow cytometer (Becton Dickinson), and platelets, ≈800 per second, were propelled through an argon ion laser beam tuned to the 351- to 364-nm band. Fluorescent light emitted by each platelet was captured at 490 to 500 nm and 385 to 395 nm, emission wavelengths corresponding approximately to the intensity maximal for Ca2+-free and Ca2+-bound indo 1. Platelets were identified by light scatter. Indo 1 signal ratios were collected from quiescent platelets, a Ca2+-mobilizing agonist was added, and indo 1 signals were monitored in the ensuing ≈3 minutes. Agonists were thrombin 0.1 U/mL, ADP 10−6 mol/L, and AVP 10−7 mol/L (approximately ED50 in this setting). Eight to 10 donors were studied for each agonist. Indo 1 signal ratios were converted to approximate values for apparent [Ca2+]i by standard methods.32
Preparation of Whole Blood for Platelet Adhesion Receptor Studies
After Human Studies Review Board approval, free-flowing venous blood was collected from each of 12 healthy, medication-free volunteers, the first 2 mL was discarded, and a 100-μL blood sample was immediately ultradiluted 1:5000 vol/vol in room-temperature Hanks’ balanced salt solution (with HEPES 25 mmol/L, BSA 2%, and CaCl2 1.4 mmol/L). Ultradilution permits avoidance of anticoagulants, centrifugation, and washing steps and protects against platelet-induced platelet activation and against aggregate formation.
Fluorescent Immunostaining of Platelet Adhesion Receptors
Samples (200 μL) of ultradilute whole blood suspension were preincubated for 10 minutes at room temperature with 0, 5, or 15 μg/mL amrinone and activated with thrombin 0.1 U/mL or ADP 10−6 mol/L. Immunostaining for activated GP IIb/IIIa was done with saturating concentrations of PAC-1, a FITC-conjugated mouse anti-human IgM monoclonal antibody directed against the fibrinogen binding site on activated GP IIb/IIIa.33 34 PAC-1 does not bind to inactivated GP IIb/IIIa on quiescent platelets. After 20 minutes, samples were fixed for 2 hours with 1 mL 1% paraformaldehyde plus 0.1% azide in 0.9% saline, washed, and resuspended in a solution of the same composition at 4°C. Samples were stored in the dark and analyzed within 24 hours. Immunostaining for P-selectin was done in a similar manner with saturating concentrations of a PE-conjugated mouse anti-human IgG monoclonal antibody with immunostaining after 1% paraformaldehyde fixation. In all studies, two color methods were used in which platelets labeled with PAC-1 were colabeled with a PE-conjugated monoclonal antibody against CD41, this second marker serving solely as a platelet-identifying label.33 34 Platelets labeled for P-selectin were colabeled with a FITC-conjugated monoclonal antibody against pan-platelet marker CD61.
Platelet GP IIb/IIIa (Activated) and P-Selectin Assay by Flow Cytometry
A flow cytometer (FACScan, Becton Dickinson) was used to measure immunostain fluorescence with excitation via argon laser light at 488 nm, and emissions were collected at 530 nm for FITC and at 585 nm for PE. Threshold and electronic gatings were set by use of both light scatter and fluorescence from the pan-platelet label. Positivity threshold was set by use of fluorescence intensity from the pan-platelet label along with appropriate isotype controls. Data (10 000 events per sample) were displayed in dot plot format, dots representing simultaneous signals from both the label of interest and the pan-platelet marker. Results are expressed as percent of events designated positive for the marker of interest.33 34
Preparation of Dogs for Experimental Coronary Thrombosis Studies
The model has been described previously.35 After approval by the Institutional Animal Care Committee, eight mongrel dogs (25 to 35 kg) were anesthetized with sodium pentobarbital (35 mg/kg IV), intubated, and mechanically ventilated with oxygen/air. A catheter was placed in a femoral artery for blood pressure monitoring and a thermodilution catheter placed in the pulmonary artery via a femoral vein for measuring cardiac output and for drug infusions. After a left thoracotomy, the heart was exposed, a 2- to 3-cm segment of proximal left circumflex coronary artery dissected free, and an EMF probe (Spectra Med) placed around the artery. Distal to the flow probe, the artery was damaged by squeezing with cushioned forceps, producing intimal and medial damage as previously described.31 Reduction in vessel diameter by 70% was produced by a constricting plastic cylinder placed around the outside of the artery.31 35 (Critical stenosis was confirmed by documenting absence of hyperemic response to temporary 20-second snare occlusion of the vessel.) The lesion generates platelet-rich thrombus, with a resultant decrease in circumflex blood flow. At the nadir of the decrease in blood flow, the thrombus either spontaneously breaks loose and embolizes distally or can be made to break loose by gentle shaking of the plastic constrictor. The cycle then repeats. CFR frequency and magnitude serve as a semiquantitative measure of thrombus formation.
Protocol for Coronary Thrombosis Studies
The experiment was done during four study periods. (1) Spontaneous CFRs were assessed during an initial 30-minute baseline control period. (2) Thrombosis was augmented by infusion of epinephrine, a prothrombotic agonist, at 0.2 μg·kg−1·min−1 IV for 20 minutes, a dose known to restore CFRs in dogs previously inhibited with aspirin. The preparation was allowed to rest for 10 minutes to permit return to baseline. (3) Amrinone was infused during 10 minutes as repeated 10-mg IV doses to a total of 2 mg/kg and continued at 20 μg·kg−1·min−1 IV for 30 minutes. (This dose schedule was based on results from a previous limited experiment in two dogs. No attempt was made to determine an ED50 for amrinone but rather to determine an effective amrinone dose regimen.) (4) In four of the eight dogs, amrinone infusion was continued. Epinephrine was reintroduced at 0.2 μg·kg−1·min−1 IV for 20 minutes. In the other four of the eight dogs, intravenous amrinone was discontinued and recurrence of CFRs was recorded for 30 minutes.
All data are reported as mean±SEM. Statistical analysis was done with repeated-measures ANOVA, with multivariate analysis, and by paired Student’s t test. Values of P<.05 were considered significant.
Amrinone was obtained from Sanofi Winthrop Pharmaceuticals. Indo 1 was from Molecular Probes. PAC-1 was obtained with the help of Dr S. J. Shattil, from the University of Pennsylvania Cell Center. Other monoclonal antibodies were from commercial sources. Anti-CD41 clone P2 was from Immunotech. Anti–P-selectin clone AC 1.2 and anti-CD61 clone RUU-PL7FI2 were from Becton Dickinson. Other agonists and chemicals were from Sigma Chemical Co.
Platelet Ca2+ Mobilization
The aim was to determine whether amrinone would inhibit agonist-induced Ca2+ mobilization in human platelets. Baseline [Ca2+]i was stable in unstimulated, quiescent platelets. Preincubation with amrinone 2.5 μg/mL and with amrinone 10 μg/mL had no effect on baseline [Ca2+]i in quiescent platelets. The agonists thrombin 0.1 U/mL, ADP 10−6 mol/L, and AVP 10−7 mol/L each evoked a vigorous and typical increase in platelet [Ca2+]i. Amrinone at the two concentrations tested attenuated Ca2+ mobilization evoked by the agonists. Results from an individual experiment are shown in Fig 1⇓, in which indo 1 fluorescence emission ratios are plotted against time. Results are summarized in Table 1⇓, in which peak indo 1 fluorescence emission ratios have been converted to approximate apparent [Ca2+]i by standard transformation methods.32
Platelet Adhesion Molecule Expression
Platelet GP IIb/IIIa activation. The aim was to determine whether amrinone would inhibit agonist-induced expression of activated GP IIb/IIIa recognized by monoclonal antibody PAC-1. Results show that ≈1% of unstimulated, quiescent platelets expressed activated GP IIb/IIIa. Amrinone 5 μg/mL and amrinone 15 μg/mL had no effect on activated GP IIb/IIIa expressed by unstimulated platelets. Both thrombin 0.1 U/mL and ADP 10−6 mol/L evoked modest increases in PAC-1 binding, ie, expression of activated GP IIb/IIIa. Amrinone at both concentrations tested attenuated expression of activated GP IIb/IIIa. An example of an individual experiment is shown in Fig 2⇓. Dots in this figure represent dual fluorescence, both from FITC-conjugated PAC-1, which labels conformationally active GP IIb/IIIa, and from a PE-conjugated monoclonal antibody against CD41, which served as a pan-platelet marker. Data are summarized in Table 2⇓.
Platelet P-selectin expression. The purpose was to determine whether amrinone would inhibit agonist-induced platelet P-selectin expression. Approximately 1% of quiescent, unstimulated platelets expressed P-selectin. Preincubation with amrinone 5 μg/mL or 10 μg/mL had no effect on P-selectin expression by quiescent platelets. Thrombin 0.1 U/mL and ADP 10−6 mol/L each evoked typical increases in percentage of platelets expressing P-selectin, with the response to thrombin being vigorous. Amrinone had a modest inhibitory effect on agonist-induced P-selectin expression. Results are shown in Fig 3⇓ and Table 3⇓.
Experimental Coronary Thrombosis in Dogs
This four-part study was designed to determine whether amrinone would attenuate experimental coronary artery thrombus formation in vivo in dogs. (1) Seven of eight dogs exhibited spontaneous CFRs during an initial 30-minute baseline control period. (2) Prothrombotic epinephrine was introduced at 0.2 μg·kg−1·min−1 IV, a dose known to restore CFRs previously inhibited by aspirin.5 During epinephrine infusion, all eight dogs exhibited CFRs with frequency increased from baseline. Epinephrine was discontinued, and once again seven of eight dogs demonstrated spontaneous CFRs. (3) Amrinone was administered as repeated 10-mg IV bolus doses during 10 minutes to a total dose of 2 mg/kg, resulting in complete abolition of CFRs in all of the seven animals exhibiting spontaneous CFRs. The amrinone effect was rapid, with abolition of CFRs occurring within 1.1±0.4 minutes of the first 10-mg IV bolus of amrinone. Amrinone was continued at 20 μg·kg−1·min−1 IV for 30 minutes, with continued absence of CFRs. (Recordings from a single dog are shown in Fig 4A⇓.) (4) In a subgroup of four of the eight dogs, amrinone infusion was continued, but prothrombotic tendency was increased by reintroducing epinephrine 0.2 μg·kg−1·min−1 IV. Despite epinephrine, CFRs remained absent in all four dogs receiving simultaneous amrinone infusion (Fig 4B⇓). In the other four dogs, in which amrinone infusion was stopped, spontaneous CFRs recurred in three of four animals. Results are summarized in Table 4⇓. Results from an individual experiment are shown in Fig 4A⇓ and 4B⇓.
Phosphodiesterase III inhibitors, including amrinone, elevate platelet cAMP levels8 9 10 11 12 and inhibit platelet function, at least when assessed by light transmittance aggregometry methods.10 11 12 13 14 15 16 Some experimental phosphodiesterase III compounds also attenuate platelet Ca2+ signaling.10 12 The present results extend these previous findings by demonstrating that the clinically useful drug amrinone attenuates agonist-induced platelet Ca2+ mobilization, inhibits activation and expression of adhesion molecules of importance to coronary thrombosis, and abolishes experimental coronary thrombosis in dogs in vivo.
Amrinone inhibited Ca2+ mobilization evoked by ADP, thrombin, and AVP. Both ADP and thrombin have immediate importance to coronary thrombosis in humans, and thrombin in particular, as a final product of tissue factor activity, initiates, amplifies, and sustains coronary thrombosis. AVP evokes well-characterized Ca2+ signaling responses in many types of animal cells, including platelets.3 All three agonists mobilize Ca2+ via a receptor–G protein–phospholipase C–inositol trisphosphate signaling axis.2 17 Some steps, in particular phospholipase C activity and the Ca2+-releasing actions of inositol trisphosphate at platelet Ca2+ stores, are sensitive to inhibition by increased levels of platelet cAMP,17 one of the known actions of amrinone.
With some caveats, the present results show that amrinone inhibited agonist-induced adhesion molecule expression in human platelets. Adhesion receptors provide a means whereby activated platelets bind to one another, to white blood cells, and to structures within the damaged vessel wall.18 19 20 21 22 23 Amrinone at most only modestly inhibited P-selectin expression. The other receptor studied, GP IIb/IIIa, has considerable importance to coronary thrombosis.18 19 20 When platelets are activated, GP IIb/IIIa takes on a new three-dimensional conformational state and in doing so reveals binding sites for fibrinogen.20 21 22 Platelet-rich thrombus is built on a core of activated platelets linked together via fibrinogen bridges. Present results show that amrinone inhibited GP IIb/IIIa activation, although not with the same inhibitory vigor as seen in current Ca2+ mobilization experiments. The mechanism of amrinone action is unclear, although inhibition of Ca2+ mobilization by amrinone, a result shown in the present experiments, would inhibit GP IIb/IIIa activation to some degree.
The magnitude of adhesion molecule expression evoked by thrombin and ADP in the present experiments was not particularly robust. Ultradilution methods may have been responsible, because at 1:5000 vol/vol dilution, spatial separation between cells would interrupt platelet-to-platelet paracrine signaling, with loss of platelet self-amplifying activation processes via messengers such as thromboxane A2. In addition, 2% albumin used in the Hanks’ buffer, a relatively high concentration by cell culture standards but one that ensures good platelet integrity ex vivo, may decrease responses reported by the immunostains.
Limitations of Studies Done Ex Vivo in Human Platelets
The Ca2+ mobilization and adhesion receptor results are encouraging. However, it is well recognized that the study of platelets ex vivo is fraught with problems. For example, sample isolation, centrifugation, platelet pellet preparation, and washing procedures can all perturb platelet responses. Attempts were made to lessen these influences in the present adhesion receptor studies by use of ultradilution methods. Nevertheless, the present results must be interpreted in the light of limitations inevitably imposed by ex vivo conditions. In addition, amrinone concentrations in the 2.5- to 15-μg/mL range were studied. However, indications for amrinone use in terms of plasma levels have not been well established in clinical practice.
Coronary Thrombus Formation In Vivo in Dogs
The sine qua non test for putative platelet-inhibiting drugs is demonstration of platelet inhibition in vivo. Amrinone rapidly abolished coronary thrombosis in vivo in dogs and protected against thrombosis even when epinephrine was infused.4 5 Of the inhibitory agents previously studied in the CFR model, amrinone is one of the few that both completely block CFRs and protect against CFR recurrence when epinephrine is infused. Aspirin, for example, in this model fails to protect against CFRs induced by epinephrine.5 Protection comparable to that currently observed with amrinone is also provided by nitric oxide donors, by the GP IIb/IIIa antagonist antibody Rheopro, and by the thienopyridines clopidrogrel and ticlopidine. This topic has recently been reviewed.36 The mechanism of protection by amrinone against experimental coronary thrombosis is not clear but probably includes platelet inhibition, especially because CFRs in this dog model are platelet-dependent. However, it would be wrong to ascribe antithrombotic amrinone effects observed in vivo in dogs entirely to the antiplatelet activity of amrinone that was currently observed ex vivo in human platelets.
Limitations of the In Vivo Dog Model
This model, in various forms, has served in a number of milestone studies concerning the roles of endogenous and other agonists in coronary thrombosis, including demonstration of the contributory roles of thromboxane A2,24 serotonin,3 α2-adrenergic agonism,3 ADP,25 and platelet-activating factor.26 The model lacks the thrombogenic components present in ruptured atherosclerotic plaque in humans. However, features of the model do mimic unstable coronary disease in humans, notably critical stenosis, high shear stress, turbulent blood flow, endothelial damage, and deep vessel wall injury. Furthermore, a number of drugs investigated for their antiplatelet effects in this dog model have gone on to successful clinical trials.
Results of the present studies demonstrate that in human platelets studied ex vivo, amrinone inhibited Ca2+ signaling and modestly inhibited P-selectin expression and GP IIb/IIIa activation. In dogs, intravenous amrinone abolished experimental coronary thrombosis, even when CFRs were potentiated by epinephrine. Results raise the possibility that the antithrombotic actions of amrinone may be useful in treatment of acute refractory heart failure associated with unstable coronary artery syndromes. However, how closely amrinone actions observed under present experimental conditions can be extrapolated to the human condition needs to be determined.
Selected Abbreviations and Acronyms
|CFR||=||cyclical flow reduction|
|GP IIb/IIIa||=||platelet glycoprotein IIb/IIIa|
Rita Nelson, Terry Halsey, and Richard Koenig are thanked for technical help, and Janet Beckman is acknowledged for secretarial assistance. The Mayo Foundation is thanked for a grant to pursue this study.
- Received October 14, 1996.
- Revision received March 6, 1997.
- Accepted March 11, 1997.
- Copyright © 1997 by American Heart Association
Bush LR, Campbell WB, Kern K, Tilton GD, Apprill P, Ashton J, Schmitz J, Buja LM, Willerson JT. The effects of α2-adrenergic and serotonergic receptor antagonists on cyclic blood flow alterations in stenosed canine coronary arteries. Circ Res. 1984;55:642-652.
Shattil SJ, Haimovich B, Cunningham M, Lipfert L, Parsons JT, Ginsberg MH, Brugge JS. Tyrosine phosphorylation of pp125FAK in platelet requires coordinated signaling through integrin and agonist receptors. J Biol Chem. 1994;269:14738-14745.
Benotti JR, Grossman W, Braunwald E, Carabello BA. Effects of amrinone on myocardial energy metabolism and hemodynamics in patients with severe congestive cardiac failure due to coronary heart disease. Circulation. 1980;62:28-34.
Coller BS. Blockade of platelet GPIIb/IIIa receptors as an antithrombotic strategy. Circulation. 1995;92:2373-2380.
Willerson JT. Inhibitors of platelet glycoprotein IIb/IIIa receptors: will they be useful when given chronically? Circulation. 1996;94:866-868.
Kroll MH, Hellums JD, McIntire LV, Schafer AI, Moake JL. Platelets and shear stress. Blood. 1996;88:1525-1541.
Smyth SS, Joneckis CC, Parise LV. Regulation of vascular integrins. Blood. 1993;81:2827-2843.
Ashton JH, Schmitz JM, Campbell WB, Ogletree ML, Raheja S, Taylor AL, Fitzgerald C, Buja LM, Willerson JT. Inhibition of cyclic flow variations in stenosed canine coronary arteries by thromboxane A2/prostaglandin H2 receptor antagonists. Circ Res. 1986;59:568-578.
Yao S-K, Ober JC, McNatt J, Benedict CR, Rosolowsky M, Anderson HV, Cui K, Maffrand J-P, Campbell WB, Buja LM, Willerson JT. ADP plays an important role in mediating platelet aggregation and cyclic flow variations in vivo in stenosed and endothelium-injured canine coronary arteries. Circ Res. 1992;70:39-48.
Golino P, Ambrosio G, Ragni M, Pascucci I, Triggiani M, Oriente A, McNatt J, Buja LM, Condorelli M, Chiariello M, Willerson JT. Short-term and long-term role of platelet activating factor as a mediator of in vivo platelet aggregation. Circulation. 1993;88:1205-1214.
Mousa SA, Bozarth JM, Forsythe MS, Jackson SM, Leamy A, Diemer MM, Kapil RP, Knabb RM, Mayo MC, Pierce SK, De Grado WF, Thoolen MJ, Reilly TM. Antiplatelet and antithrombotic efficacy of DMP 728, a novel platelet GPIIb/IIIa receptor antagonist. Circulation. 1994;89:3-12.
Coller BS, Smith SR, Scudder LE, Jordan R, Folts JD. Abolition of in vivo platelet thrombus formation in primates with monoclonal antibodies to the platelet GPIIb/IIIa receptor: correlation with bleeding time, platelet aggregation and blockade of GPIIb/IIIa receptors. Circulation. 1989;80:1766-1774.
Demrow H, Batley C, Folts JD. SDZ GPI 562, an oral peptidomimetic inhibitor of the platelet GpIIbIIIa abolishes epinephrine and shear induced cyclic flow reduction in stenosed monkey carotid arteries. J Am Coll Cardiol. 1995;(special edition):512. Abstract.
McGhie AI, McNatt J, Ezov N, Cui K, Mower LK, Hagay Y, Buja LM, Garfinkel LI, Gorecki M, Willerson JT. Abolition of cyclic flow variations in stenosed, endothelium-injured coronary arteries in nonhuman primates with a peptide fragment (VCL) derived from human plasma von Willebrand factor–glycoprotein Ib binding domain. Circulation. 1994;90:2976-2981.
Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J Biol Chem. 1985;260:3440-3450.
Michelson AD, Shattil SJ. The use of flow cytometry to study platelet activation. In: Watson SP, Authi KS, ed. Platelets: A Practical Approach. Oxford, UK: Oxford University Press; 1996:111-129.
Michelson AD. Flow cytometry: a clinical test of platelet function. Blood. 1996;87:4925-4936.
Folts JD. An in vivo model of experimental arterial stenosis, intimal damage, and periodic thrombosis. Circulation. 1991;83:3-14.
Folts JD. Drugs for the prevention of coronary thrombosis: from an animal model to clinical trials. Cardiovasc Drugs Ther. 1995;9:31-43.