Search

HE INTACT endothelial cell (EC) lining of the vascuT lar wall is not reactive to the circulating blood However, ECs acquire thrombogenic properties when they are stimulated by cytokines or thrombin, as shown both in vivo and in vitro. Thus, there is expression of plasminogen activator inhibit~r,~ expression of tissue fa~tor,~ and downregulation of thrombomod~lin,~ all of which can facilitate thrombogenesis. We have made a number of observations that suggest that changes in EC lipoxygenase metabolism after stimulation also influence the thrombogenic properties of ECs. First, we found that unstimulated ECs metabolize linoleic acid into 13-hydroxyoctadecadienoic acid (1 3HODE) via the lipoxygenase pathway, the intracellular levels of which correlate inversely with EC thrombogenecity.6 Second, we and others have reported that, when ECs are stimulated by thrombin or by any of a number of cytokines, 13-HODE synthesis is inhibited and EC reactivity to platelets and cancer cells increases, both in vitro and in Finally, recent studies indicate that the increased EC reactivity is associated with an enhanced expression of the vitronectin receptor (VnR) on ihe apical surface of the ECs.I6,l7 Therefore, we hypothesized that these two phenomena are related and constitute an important mechanism by which EC reactivity for platelets, cancer cells, and other circulating blood cells is regulated. One possible mechanism by which 13-HODE may influence VnR expression is that 13-HODE may interact with the VnR, altering its conformational presentation and, therefore, its ability to recognize its specific ligands. Consistent with that possibility, Conforti et all8 found that the ability of the VnR to recognize von Willebrand factor (vWF), fibronectin, and vitronectin was dependent, in part, on the plasma membrane fatty acid(s) with which it associated. Other investigators have also reported that various fatty acids derived from activated platelets and polymorphonuclear leukocytes enhance the abilities of other argineglycine-aspartic acid (RGD)-recognizing integrins to bind their ligands in purified sy~tems.’~,~~ Thus, a number of investigators have provided evidence that lipid moieties may be important in modulating the ligand-binding properties of integrins. The studies cited above raise the possibility that

Polymorphonuclear leukocytes (PMN) adhere to the vascular endothelial lining in vivo and to the surfaces of cultured endothelial cells in vitro, but the mechanisms of these cellular interactions remain unclear. Arachidonic acid metabolites, both cyclooxygenase- and lipoxygenase-derived, have been shown to influence PMN locomotion, secretion, and adhesion to artificial surfaces. To determine whether such mediators also are involved in regulating PMN-endothelial cell interactions, we have examined the effects of prostacyclin and various inhibitors of arachidonic acid metabolism on the adherence of radiolabeled PMN to cultured bovine aortic endothelial cells. Confluent endothelial monolayers were incubated with washed suspensions of radiolabeled human PMN (which contained less than 1% platelet contamination) at 37 degrees C for 30 min, then subjected to a standardized wash procedure and the number of adherent leukocytes determined radiometrically. Under basal conditions, i.e., in the absence of exogenous activating stimuli, 4,163 +/- 545 PMN adhered per square millimeter of endothelial surface (mean +/- SEM, n = 12). This basal adhesion (which corresponds to approximately 4–5 leukocytes per endothelial cell) was unaffected when the leukocytes and endothelial monolayers were pretreated with cyclooxygenase inhibitors (100 microM aspirin or 1–5 microM indomethacin) or PGI2 (10(-9)-10(6) M). Thus, basal PMN-endothelial adhesion in this in vitro model system does not appear to be dependent on endogenous cyclooxygenase derivatives of arachidonate or to be sensitive to inhibition by exogenous prostacyclin. In contrast, leukocyte adhesion was significantly reduced by pretreatment with 5,8,11,14- or 4,7,10,13-eicosatetraynoic acid, 0.5- 5 mM sodium salicylate, or 10–1,000 microM indomethacin, antiinflammatory agents that can interfere with the metabolism of arachidonic acid via non-cyclooxygenase-dependent mechanisms. These observations may be relevant to the interactions of circulating PMN with vascular endothelium under both physiologic and pathophysiologic conditions in vivo.

The effect of experimental hematomas on serum levels of fibrinogen- related antigen (FRA) was studied in rabbits. Homologous blood used for production of hematomas was mixed with 131I-fibrinogen, and the measurement of FRA was made using a radioactive technique (representing FRA from hematoma) and immunologically (representing total FRA). Total body radiation of animals bearing hematomas containing 131I-fibrinogen decayed slowly (mean t 1/2 = 189 hr) over the first 2–3 days and then rapidly (mean t 1/2 = 43 hr) during the subsequent days. Serum FRA levels measured immunologically rose transiently from 4 to 11 mug/ml during the slow phase of hematoma resolution, and returned to baseline levels during the rapid phase of resolution. The appearance of a small amount of protein-bound radioactivity in the blood of animals with hematomas paralleled the time course of the changes in immunologically determined FRA, but, even at its peak, circulating levels only averaged 0.36% of the administered tracer. From these observations, it seemed unlikely that the dissolution of fibrin which may form as the result of bleeding into tissues would significatnly contribute to levels of circulating FRA.

The effect of streptokinase on hemostasis was studied by direct observation of hemostatic plugs in the rabbit mesentery. Infusion of streptokinase did not produce dissolution of hemostatic plugs which were already formed. It did cause a delay in the formation of a platelet mass capable of arresting bleeding from injured vessels. In contrast to systemic infusion, topical application of streptokinase did produce disruption of hemostatic plugs which were already formed. Comparison of the electronmicroscopic appearances of plugs from the control and streptokinase infused animals revealed a lack of fibrin around the plugs in the streptokinase treated animals but no significant difference in platelet morphology. In vitro tests with ADP, thrombin and collagen on platelet-rich plasma from the streptokinase infused animals did not show any difference in platelet aggregation between the pre-and postinfusion samples, even though streptokinase infusion produced a marked increase in plasma fibrinolytic activity. It is suggested that streptokinase infusions, which produce only a slight fall in the plasma fibrinogen level, impair hemostatic plug formation by digestion of fibrin as it is formed around the aggregated platelets.