Search

There have been major advances in our understanding of thrombosis and antithrombotic drugs. This review focuses on the molecular aspects of thrombus formation and antithrombotic therapy. Molecules involved in arterial thrombosis are derived from inflammatory cells in the atherosclerotic plaque and blood platelets. These molecules work in concert to promote plaque instability and thrombogenicity. Thrombus formation on the ruptured plaque is mediated by platelet and coagulation activation. By contrast, molecules involved in venous thrombosis are derived from the activated coagulation cascade. Platelets appear to play a secondary role. The antithrombotic drugs are classified according to their targeted constituents: antiplatelet agents and anticoagulants; the latter are further divided into non-specific anticoagulants, such as vitamin K antagonists and heparin, and direct thrombin inhibitors, including hirudin and argatroban. Currently available antiplatelet agents target glycoprotein IIbIIIa (abciximab, tirofiban, eptifibatide), cyclooxygenase-1 (aspirin) or adenosine diphosphate receptor, P2Y12 (clopidogrel).

Genetic evidence from a large Vermont kindred indicates that an unknown gene promotes thrombosis when inherited in conjunction with type I protein C deficiency. Cyclooxygenase-1 [prostaglandin H synthase 1 gene (PTGS1)] was tested as a plausible candidate for the unknown gene because of its role in primary hemostasis. The complete sequence of PTGS1 (25 638 nucleotides) was determined from a 37 kb human genomic cosmid clone to characterize intronic regions and subsequently to allow the search for mutations by direct sequencing of genomic DNA. Northern blot analysis confirms usage of a newly described distal poly-adenylation signal. Short tandem repeat (STR) sequences found in intron 2 and the 3' flanking region were developed as new genetic markers for PTGS1. The position of PTGS1 was refined on the CHLC chromosome 9 linkage map using the new markers scored in four Centre d'Etude du Polymorphisme Humain families and multipoint linkage analysis. Direct sequencing of DNA from members of the Vermont kindred led to the discovery of two new single nucleotide polymorphisms (SNPs) that give rise to non-conservative amino acid changes in the signal peptide (Arg(8) to Trp and Pro(17) to Leu) of cyclooxygenase-1. Linkage analysis of the SNP and STR markers indicated that PTGS1 is not the interacting gene associated with an increased incidence of thrombosis in the Vermont kindred.

Thrombomodulin (TM) binds thrombin, changes thrombin conformation and allows thrombin to activate protein C and thrombin-activatable fibrinolysis inhibitor (TAFI). Activated protein C and TAFI inhibit coagulation and fibrinolysis, respectively. TM is, thus, a linker of endogenous control of coagulation and fibrinolysis. Plasma soluble TM, cleaved products of cellular TM, also have anticoagulant and antifibrinolytic properties. TM plays an important role in thromboresistance. Prospective studies show that a high plasma TM level is associated with a low risk of developing coronary heart disease. The plasma TM level may reflect the level of endothelial TM expression. TM expression levels are influenced by multiple gene polymorphisms. Several of the polymorphisms are probably associated with coronary heart disease.

Nitric oxide (NO) and prostacyclin (PGI2) are potent molecules produced by endothelial cells that act synergistically to maintain normal vascular functions. Recent studies indicate that key enzymes that catalyze the synthesis of these two molecules are induced by chemical and physical factors. Because NO and PGI2 and their synthetic enzymes have short half-lives, transcriptional induction by injurious agents, such as lysophosphatidylcholine (lysoPC), represents an important defense mechanism. LysoPC is capable of inducing constitutive endothelial NO synthase (eNOS) and cyclooxygenase-2 (COX-2), an immediate early gene, with distinct kinetics and possibly distinct transcriptional mechanisms. The importance of injury-coupled eNOS and COX overexpression in vasoprotection is supported by powerful effects of virus-mediated transfer of COX and eNOS genes on defense against thrombosis and intimal hyperplasia in angioplasty-injured carotid arteries in animal models.

This article summarizes the genetic markers of human venous and arterial thrombotic disorders. For venous thromboembolism, a factor V mutation (Arg 506-->Gln) has the highest risk, followed by protein C, S and antithrombin III gene defects. By contrast, these genetic defects are not associated significantly with arterial atherothrombotic disorders. Instead, a glycoprotein IIIa polymorphism (Pro33 versus Leu 33) has been reported to be associated with myocardial infarction. Fibrinogen Bbeta chain, factor VII, and plasminogen activator inhibitor-1 gene polymorphisms have been reported to influence the plasma levels of these factors and may indirectly be risk factors for arterial thrombotic disorders. Further studies will uncover additional genetic markers for thrombosis.

Hemostatic components play major roles in the pathogenesis of human atherothrombotic disease. There has been interest in determining whether tests for hemostatic components predict this disorder. Recent population-based and clinically-oriented prospective studies have begun to provide useful information. Northwick Park Heart (NPH) study made the initial observation that the plasma fibrinogen level is an independent risk factor for non-fatal and fatal coronary heart disease (CHD). This has been confirmed by other population-based studies. By contrast, factor VIIc levels which were reported by NPH to be an independent risk factor for CHD, especially fatal myocardial infarction (MI) has not been confirmed by other studies. Factor VIIIc, von Willebrand factor (vWF), tPA and PAI-1 are reported to be associated with CHD. Prospective angina pectoris studies have identified fibrinogen, C-reactive protein (CRP) and von Willebrand factor as independent risk factors for acute MI. These findings raise a possible mechanism of inflammation in CHD. A number of studies imply an association of platelet activation with CHD. A prospective study has shown that a persistent positive spontaneous platelet aggregation (SPA) test is an independent risk factor for recurrent MI. Hence, prospective studies indicate a potential value for certain hemostatic tests to predict atherothrombotic disorder. However, clinical utility of these tests remains to be established. Controlled clinical trials may be required to provide a more definite information regarding the use of these tests for selecting high risk patients for preventive strategies.

Prostacyclin (PGI2) and nitric oxide (NO) are potent vascular mediators, playing key roles in protecting arterial wall from injury-induced lesions. The key enzyme that catalyzes PGI2 biosynthesis is cyclooxygenase (COX). COX-1 undergoes auto-inactivation, which severely limits PGI2 synthesis. Overexpression of COX-1 in cultured endothelial cells by COX-1 gene transfer was accompanied by a higher capacity for and sustained synthesis of PGI2. Adenovirus-mediated COX-1 gene transfer to angioplasty damaged carotid arteries in pigs augmented PGI2 synthesis and prevents thrombus formation. Transfer of endothelial NO synthase (eNOS) into angioplasty injured, carotid arteries was reported to suppress intimal hyperplasia in rats. Transfer of PGI2 and NO synthetic enzymes restores the vasoprotective properties and represents an exciting new strategy for treating arterial thrombotic disorders.

Human arterial thrombotic disorders are triggered by many agents, with participation of platelets and monocytes, blood coagulation factors and vascular cells. Platelet hyperaggregability appears to be an important risk factor for these disorders. Vascular endothelium possesses several properties to defend against vascular insults and thrombotic atherosclerotic lesions. Two molecules, prostacyclin (PGI2) and nitric oxide (NO), are of particular importance. The rate-limiting step of PGI2 synthesis is cyclooxygenase (COX). Constitutive and upregulated constitutive COX (COX-1) expression and inducible COX (COX-2) expression are important in PGI2 production required for the physiologic and pathologic defense of blood vessels and blood fluidity. NO synthesis is catalyzed by endothelial nitric oxide synthase (eNOS), which can be stimulated by lipid mediators. Virus or non-virus mediated transfer of COX-1 and eNOS are accompanied by augmented PGI2 and NO synthesis, respectively. In animal angioplasty models, it has been shown that transfer of these two genes has a dramatic antithrombotic and anti-intimal hyperplastic effect. Transfers of these two enzymes may have potential therapeutic uses.

Vascular endothelium is strategically located at the interface between tissue and blood. It is pivotal for protecting against vascular injury and maintaining blood fluidity. Normal endothelium releases prostacyclin and nitric oxide, potent inhibitors of platelet and monocyte activation and vasodilators. Their syntheses are governed by isoforms of enzymes. Normal endothelial surface expresses ecto-adenosine diphosphatase, which degrades adenosine diphosphate and inhibits platelet aggregation; thrombomodulin, which serves as a binding site for thrombin to activate protein C; and heparin-like molecules, which serve as a cofactor for antithrombin III. Normal endothelium secretes tissue plasminogen activator, which activates the fibrinolysis system. Endothelium produces and secretes von Willebrand factor, which mediates platelet adhesion and shear-stress-induced aggregation. Injury to endothelium is accompanied by loss of protective molecules and expression of adhesive molecules, procoagulant activities, and mitogenic factors, leading to development of thrombosis, smooth muscle cell migration, and proliferation and atherosclerosis.

Once regarded as an inert lining, the vascular endothelium is now known to synthesize a variety of molecules that can protect against or promote disease. These important agents include prostacyclin, thrombomodulin, plasminogen activators, and ecto-enzymes. Our growing understanding of their protective and pathologic roles has already provided new therapeutic strategies.

The present study evaluated the direct involvement of thromboxane A2/prostaglandin H2 (TXA2/PGH2) in the process of thrombus formation at a site of vascular damage. De-endothelialization of the rabbit aorta was performed by a balloon catheter technique. Platelet deposition to the injured vessel was measured using 111Indium-labeled autologous platelets. Studies using the radiolabeled TXA2/PGH2 antagonist, 13-azaprostanoic acid (13-APA), indicated that 13-APA has an in vivo half-life of approximately 35 min and is excreted by the kidney in the metabolized form. Addition of 13-APA to rabbit plasma samples in vitro produced a dose-dependent inhibition of arachidonic acid-induced platelet aggregation. When comparable plasma levels of 13-APA were achieved by infusion of 13-APA (300 micrograms/kg/min for 90 min), a similar dose-dependency of inhibition of ex vivo aggregation was observed. Furthermore, at a plasma concentration of 40 microM, 13-APA was found to inhibit platelet deposition to the de-endothelialized rabbit aorta by 45%. Because 13-APA does not interfere with arachidonic acid metabolism, the ability of 13-APA to suppress thrombus formation is presumably due to direct antagonism of TXA2/PGH2 at the platelet receptor level. These findings, therefore, provide evidence that TXA2 and/or PGH2 have a major role in platelet deposition at a site of vascular damage.

The arachidonic acid metabolites produced by human peripheral blood monocytes were studied to determine which metabolites could have a role in thrombogenesis. Monocytes were found to be free of platelets by scanning electron microscopy and by measurement of 12-HETE. Human peripheral blood monocytes produce thromboxane as their major metabolite. Thromboxane levels reached a plateau at 12-16 hours of culture. Monocytes produced relatively little prostaglandin E2 or F2. In contrast to our control platelet preparation, neither A23187 (1-10 microM) nor exogenous arachidonic acid (0-40 microM) caused an increase in monocyte thromboxane B2. On the other hand, lipopolysaccharide (20 micrograms per ml), collagen (2.5 mg per 10(7) cells), and thrombin (5-10 units per ml) caused a two- to fivefold increase in monocyte thromboxane B2 in most donors but had no effect on prostaglandin F1 alpha levels. Blockage of thromboxane synthase by 1-benzylimidazole abolished thromboxane B2 production but did not increase prostaglandin F1 alpha. Finally, aspirin-treated platelets from a volunteer donor, which were refractory to 30 microM arachidonate, could be aggregated by isolated blood monocytes. Our data indicate that monocytes are capable of producing thromboxane in large amounts. The regulation of this increase, however, appears to be quite different from platelets. We postulate that monocytes may have a role in hemostasis by virtue of their ability to adhere at sites of vascular injury and release thromboxane, which may enhance platelet aggregation and thrombus formation.

The relation between platelet hyperaggregability and thrombosis was assessed in 28 patients with thrombocythemia due to myeloproliferative diseases and 11 with reactive thrombocytosis. None of the patients with reactive thrombocytosis had thrombotic or hemorrhagic complications, but thrombosis was noted in seven patients and bleeding in two patients with thrombocythemia. Nineteen were asymptomatic. In patients with thrombosis, bleeding time, platelet glass retention, and clot retraction were normal, but evidence of platelet hyperaggregability was present in all but one. Serial studies on six patients revealed a close association between platelet hyperaggregability and ischemic attacks. Neither patient with bleeding complications had evidence of platelet hyperaggregability, although poor platelet function was found in one. Platelet function in asymptomatic patients can be classified as hyperactive, hypoactive, or normal.

Endothelial cells play a pivotal role in hemostasis and thrombosis. They produce a myriad of factors either associated with the membrane or released into the blood stream and the subendothelial matrix which are involved in various steps of hemostasis. The endothelial cell function is modulated by a diversified group of biologically active molecules, notably thrombin, vasoactive amines and cytokines. Mechanism and selectivity of the effects of these molecules differ and the difference may have important physiological implications. Most of the information is gathered through experiments performed in cultured endothelial cells. Availability of the cultured cells has greatly facilitated the understanding of endothelial cell biology. In vivo models, however, are still needed to understand how the endothelial cell function is modulated. Furthermore, as the cultured endothelial cells exhibit nor only species differences but also vascular origin difference in behavior and function, these factors should be carefully considered when designing experiments involving the use of cultured endothelial cells.