Oral contraceptives (OC) and hormone replacement therapy (HRT) were suspected to increase the risk of thrombosis. Their effects on hemostatic factors (platelets, coagulation proteins, and vascular factors) were unclear. Earlier studies focused on the influence of OCs on platelet function and coagulation activation. Several studies, including ours, show that OCs increased platelet and coagulation activation. These earlier studies are generally performed in small numbers of human subjects. In order to gain insight into the relevance of these earlier observations in the general population, we determined the relationship of hemostatic changes with HRT in a population-based study, called ARIC (Atherosclerosis Risk In Communities). The ARIC study is sponsored by the National Heart, Lung and Blood Institute of the U.S. NIH. It recruited 15,800 middle-aged men and women from four U.S. communities. These participants were examined, including blood tests for hemostatic and lipid factors, every three years. Plasma fibrinogen, factor VII, factor VIII, von Willebrand factor, antithrombin, and protein C as well as cholesterol, apoprotein, lipoprotein (a), glucose, and insulin were measured in all participants. After adjusting for age, race, body mass index, cigarette smoking, alcohol drinking, diabetes, hypertension, level of education, and sports index, the results show that the HRT users had a significantly lower fibrinogen, higher factor VIII, lower antithrombin, and higher protein C levels than non-users. These results provided, for the first time, new insight into the alteration in several key hemostatic risk factors in HRT in the population. They are valuable for understanding the involvement of hemostatic factors in HRT-induced thrombotic complications.

We have also performed a series of prospective analyses of the association of a large number of hemostatic factors with coronary heart disease (CHD) risk. We have identified several coagulation and fibrinolytic factors as independent risk factors that predict the development of CHD. Among them, the association of soluble thrombomodulin with CHD is novel and of considerable importance. Thrombomodulin (TM) is expressed on the endothelial surface. It anchors to membrane via a single transmembrane domain with a large extracellular region that contains lectin-like and epidermal growth factor domains which harbor binding sites for thrombin and protein C. Thrombin binds to TM and becomes catalytically active in converting protein C to activated protein C (APC). APC is a major player in defense against thrombosis and inflammation. The extracellular region of TM is cleaved into multiple fragments which circulate as soluble TM. Soluble TM (sTM) may possess similar biological activities as the endothelial surface TM. We measured sTM and analyzed its association with CHD in a prospective case-cohort study. Our results showed that participants with a high plasma sTM have a lower CHD risk than those with a low sTM. These results suggest that plasma sTM in healthy subjects may be a surrogate marker of endothelial surface TM. Alternatively, it may have a direct protective effect on CHD. Plasma soluble intercellular adhesion molecule-1 (sICAM-1) reflects endothelial cell inflammation, and a high sICAM is associated with an increased CHD risk. To determine whether a high sTM retains protective property in the face of inflammation, we performed a combinatorial analysis in the prospective ARIC cohort study. Our data showed that ARIC participants with high sTM and high sICAM have a CHD risk similar to those with high sTM and low sICAM, suggesting that a high sTM negates the risk of sICAM. By contrast, a low sTM and a high sICAM greatly increased the risk of developing CHD when compared with a high sTM and a low sICAM. These data underscored the importance of combinatorial analysis of hemostatic risk factors. They shed light on the interaction of hemostatic and pro-inflammatory factors in CHD development.

Vascular endothelial cells synthesize several compounds that target platelet activation. Two of these compounds, prostacyclin (PGI₂) and nitric oxide (NO), act synergistically on blocking platelet aggregation. They also control monocyte activation and its transmigration into the arterial wall. Biosynthesis of PGI₂ and NO requires endothelial cell activation by physiological and pathological agonists, which activate the key synthetic enzymes. The extent of PGI₂ and NO production depends on transcriptional upregulation of cyclooxygenase (COX) and NO synthase (NOS), respectively. Our research work has provided new information regarding the control of the expression of these two classes of enzymes. We have shown that aspirin and its metabolite, sodium salicylate, suppress COX-2 and NOS-2 (also known as inducible NOS or iNOS) expressions by blocking C/EBPβ binding to C/EBP enhancer elements on the promoter regions of these two genes. C/EBPβ is inert until it is phosphorylated, when its binding to DNA is vastly increased. We postulate that salicylate targets a specific kinase that phosphorylates C/EBPβ. The work is in progress. Once a specific kinase is identified, it will be a useful target for drug discovery.

PGI₂ production is decreased in injured arteries. We have used gene transfer to restore PGI₂ productions. We have recently developed a bicistronic COX-1/PGI synthase vector, which selectively augmented PGI₂ synthesis after transfection into endothelial cells. Our results have shown that this approach is effective in preventing arterial thrombosis in a porcine arterial injury model and reducing cerebral infarct volume in a rat stroke model. These studies have several implications: (1) they clarify the physiological roles of COX-1 and PGIS in vaso- and tissue protection, (2) they provide new information on the regulation of PGI₂ synthesis, and (3) they offer a valuable strategy for treating ischemia-reperfusion induced tissue damage.

My future research is directed at: (1) search for novel risk factors of CHD and ischemic stroke as well as the influence of HRT on the novel factors, (2) identifying molecules that control COX-2, iNOS, and other pro-inflammatory gene expressions, and (3) therapeutic implications of gene replacement for prevention and treatment of arterial thrombotic diseases and ischemia-reperfusion tissue injury.

Kenneth K. Wu, M.D., Ph.D.
University of Texas-Houston Medical School
and
Brown Foundation Institute of Molecular Medicine for the Prevention of Human Diseases
Houston, TX, USA