Histones Bind Prothrombin to Generate Alternative Prothrombinase Complexes That Can Disseminate Intravascular Coagulation

Background: Increased thrombin generation in vivo is pivotal to the development of disseminated intravascular coagulation (DIC). Typically, thrombin is generated when the prothrombinase complex, composed of activated factor X (FXa), activated co-factor V (FVa) and phospholipids, cleaves prothrombin in the presence of calcium. In critical illness, extensive cell death releases histones into the circulation, which can increase thrombin generation. However, the underlying pathophysiological mechanisms remains to be fully elucidated.

Methods: In vitro: Isolation of histone-binding proteins with mass spectrometry analysis. Surface plasmon resonance binding studies, prothrombin cleavage and thrombin generation assays. In vivo: histone infusion mouse models (C57BL/6 mice) with or without prothrombin fragment F1+F2 infusion. Clinical: a prospective cohort of 129 adult intensive care unit patients (ICU) with sepsis and analysed for DIC.

Results: Histone-conjugated Sepharose beads were used to pull down proteins from human plasma. Histone-binding proteins were subjected to 2D gel electrophoresis and sequenced by liquid chromatography-mass spectrometry. Prothrombin was the only coagulation factor identified. Histones directly bind to prothrombin (H3 [Kd = 6.8 x 10⁷ M] and H4 [Kd = 7.0 x 10⁷ M]), specifically prothrombin fragments F1+F2, to facilitate FXa-induced prothrombin cleavage and thrombin generation (H4 [12.25 ± 1.25 fold] and H3 [8.82 ± 0.67 fold]). FXa levels are the limiting factor of histone-enhanced thrombin generation since this process was inhibited in FX-deficient plasma unless exogenous FXa was added. Specifically, using either heparin or anti-histone antibodies to block histones, histone-prothrombin interactions, prothrombin cleavage and subsequent thrombin generation were significantly reduced. Unlike FVa which requires a phospholipid surface to form functional prothrombinase complexes, histones can substitute for FVa in the absence of phospholipids. The addition of histones to FV-deficient plasma restored thrombin generation, suggesting that histones can bypass FVa to induce thrombin generation.

In vivo, infusion of histones into mice caused significant decreases in platelet counts and fibrinogen levels with elevations in thrombin-antithrombin complexes, D-dimer and prothrombin time in a dose-dependent manner. Pathological examination indicated intravascular thrombi with various organs, particularly in within lung tissues. These histone-induced DIC changes were significantly abrogated when prothrombin fragments F1+F2 were infused prior to histones to act as a decoy for binding of histones to circulating prothrombin. Analysis of DIC scores in ICU patients (n=129) with sepsis showed circulating histone levels to strongly correlate with DIC scores (r=0.446, p<0.0001).

Conclusions: Histones can replace FVa in prothrombinase and not require phospholipid surfaces. This alternative histone-assembled prothrombinase can explain how thrombin could be generated and amplified away from cell surfaces to cause systemic dissemination of its effects and potentiate DIC. This study also identifies circulating histones as a potential target for therapeutic intervention in reducing DIC development and subsequent multi-organ failure in ICU patients.