Diagnostic and prognostic usefulness of neutrophil extracellular traps in disseminated intravascular coagulation: Histone-induced pro-coagulative phenotype of endothelial cells

Abstract

Introduction: Neutrophils can be induced to release DNA combined with histones. The resulting neutrophil extracellular trap (NET) provides a scaffold for growing hemostatic plug. NET might be actively formed in clinical conditions that are characterized by formation of vascular thrombi. Endothelium usually provides anticoagulant surface, but the thrombotic event may change the anticoagulant endothelium to procoagulant phenotype. Resting endothelial cells express anticoagulant thrombomodulin (TM), but not procoagulant tissue factor (TF). Stress condition, however, can induce TF. This study hypothesized that active NET formation may occur in overt- disseminated intravascular coagulation (DIC) and that the circulating levels of NET may have clinical diagnostic and prognostic implication. In addition, the elevated histones may alter endothelial TF and TM expression levels. Methods: The plasma levels of DNA-histone complexes and double-stranded DNA (dsDNA), considered to be in vivo markers of NET, were measured in 199 patients suspected of having DIC and 20 healthy controls. Surface antigens, soluble forms, and mRNA levels of TF and TM were measured by flow cytometry, ELISA, and real-time RT-PCR, respectively. TF and TM activity were measured using procoagulant activity, thrombin generation, or chromogenic assays. Involvement of NF-κB pathway and toll-like receptors (TLR) were assessed. Result: The circulating levels of DNA-histone complexes and dsDNA were significantly elevated in overt-DIC. The increased levels of these two markers correlated with the severity of coagulopathy including DIC score and D-dimer. Multivariable Cox regression analysis, adjusted for the conventional DIC markers, revealed that elevated DNA-histone complexes and dsDNA are poor independent prognostic markers. Histones dose-dependently induced surface antigens, activity and mRNA levels of endothelial TF. Histone-treated endothelial cells significantly shortened the lag time and enhanced the endogenous thrombin potential of normal plasma, which was normalized by a TF neutralizing antibody. Histones induced phosphatidylserine and protein-disulfide isomerase expression in endothelial cells. Histones also reduced the surface antigen, activity, and mRNA levels of endothelial TM. TLR9 inhibitor significantly blocked the TF up-regulation and TM down-regulation, whereas TLR2, and TLR4 inhibitors partly blocked the TF up-regulation. Polysialic acid and heparin reversed the histone-induced TF up-regulation and TM down-regulation. Activated protein C did not affect the TF up-regulation, but interrupted TM down-regulation. Conclusion: The circulating levels of NET release reflect the coagulation activation and adverse clinical outcomes in patients with DIC, thereby providing potential clinical relevance for mortality prediction in DIC. Histones induced the endothelial procoagulant phenotype through TF up-regulation and TM down-regulation. The effects of histones were partly mediated by TLR2, TLR4, and TLR9. Strategies to inhibit the harmful effects of histones in endothelial cells may be required in order to prevent a thrombotic environment.