Abstract
Abstract 535
In vitro studies have suggested that reactive oxygen species such as superoxide can produce several potentially prothrombotic effects, including enhanced platelet activation, increased tissue factor (TF) expression, and an oxidative modification in thrombomodulin that impairs its capacity to enhance the generation of activated protein C (APC) by thrombin. It is not known, however, if elevated levels of superoxide accelerate susceptibility to experimental thrombosis in vivo. Using a murine model that is genetically deficient in superoxide dismutase-1 (SOD1, an antioxidant enzyme that dismutates superoxide to hydrogen peroxide), we tested the hypothesis that lack of superoxide dismutase enhances susceptibility to thrombosis. Additionally, we investigated the mechanisms of superoxide-enhanced thrombosis. First, we examined the susceptibility to carotid artery thrombosis in a photochemical injury model. We found that Sod1−/− mice formed stable occlusions significantly faster than Sod1+/+ or Sod1+/− mice (P<0.05). Further, using an inferior vena cava (IVC) stasis method we observed that Sod1−/− mice developed significantly larger thrombi 48 hours after IVC ligation (P<0.05 compared with Sod1+/+ or Sod1+/− mice). These findings suggest that deficiency of SOD1 leads to increased susceptibility to both arterial and venous thrombosis in mice. To address the mechanism of accelerated thrombosis, we first examined activation of platelets in response to multiple agonists using flow cytometry. After activation by thrombin (0.5 U/ml) and convulxin (200 ng/ml), no differences in surface expression of P-selectin or binding of fibrinogen to activated platelets were observed between Sod1−/−, Sod1+/+, or Sod1+/− mice, suggesting that increased susceptibility to thrombosis in Sod1−/− mice is not platelet mediated. Next, we measured expression of TF mRNA in lung by real time qPCR. TF mRNA levels in Sod1−/− mice were similar to those in Sod1+/+ mice, suggesting that deficiency of SOD1 does not influence TF expression in mice. Finally, we measured the activation of protein C in vivo in response to infusion of thrombin (40 U/Kg). Generation of activated protein C was significantly lower in Sod1−/− mice compared with Sod1+/+ mice (P<0.05). No differences in mRNA levels for thrombomodulin or endothelial protein C receptor were detected in Sod1−/− mice compared with Sod1+/+ mice (P=0.4 and 0.6 respectively), suggesting that altered generation of activated protein C in Sod1−/− mice may be related to a direct oxidative effect on thrombomodulin rather than to decreased expression of thrombomodulin or EPCR. We conclude that lack of SOD1 in mice accelerates thrombosis and impairs the protein C anticoagulant response to thrombin.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.