Identification of a thrombomodulin interaction site on thrombin-activatable fibrinolysis inhibitor that mediates accelerated activation by thrombin

Identification of heparin interaction sites on thrombin-activatable fibrinolysis inhibitor that modulate plasmin-mediated activation, thermal stability, and antifibrinolytic potential

Background

Thrombin-activatable fibrinolysis inhibitor (TAFI) is a plasma zymogen that provides a molecular link between coagulation and fibrinolysis. Studies have shown that the presence of glycosaminoglycans accelerates TAFI activation by plasmin and stabilizes activated TAFI (TAFIa).

Objectives

We aimed to define the elements of TAFI structure that allow these effects.

Methods

Based on crystallographic studies and homology to heparin-binding proteins, we performed mutagenesis of surface-exposed charged residues on TAFI that putatively constitute heparin-binding sites. We determined heparin binding, kinetics of activation by plasmin in the presence or absence of heparin, thermal stability, and antifibrinolytic potential of each variant.

Results

Mutagenesis of Lys211 and Lys212 did not impair heparin binding but affected the ability of TAFI to be activated by plasmin. Mutagenesis of Lys306 and His308 did not impair heparin binding, but mutation of His308 had a severe negative effect on TAFI/TAFIa function. Mutation of Arg320 and Lys324 in combination markedly decreased heparin binding but had no effect on heparin-mediated acceleration of TAFI activation by plasmin while somewhat decreasing TAFIa stabilization by heparin. Mutagenesis of Lys327 and Arg330 decreased (but did not eliminate) heparin binding while decreasing the ability of heparin to accelerate plasmin-mediated TAFI activation, stabilize TAFIa, and increase the antifibrinolytic ability of TAFIa. A quadruple mutant of Arg320, Lys324, Lys327, and Arg330 completely lost heparin-binding ability and stabilization of the enzyme by heparin.

Conclusion

Basic residues in the dynamic flap of TAFIa define a functionally relevant heparin-binding site, but additional heparin-binding sites may be present on TAFI.

Profibrinolytic effects of rivaroxaban are mediated by thrombin-activatable fibrinolysis inhibitor and are attenuated by a naturally occurring stabilizing mutation in enzyme

Rivaroxaban is a direct factor Xa inhibitor, recently implemented as a favorable alternative to warfarin in anticoagulation therapy. Rivaroxaban effectively reduces thrombin generation, which plays a major role in the activation of thrombin activatable fibrinolysis inhibitor (TAFI) to TAFIa. Based on the antifibrinolytic role of TAFIa, we hypothesized that rivaroxaban would consequently induce more rapid clot lysis. In vitro clot lysis assays were used to explore this hypothesis and additionally determine the effects of varying TAFI levels and a stabilizing Thr325Ile polymorphism (rs1926447) in the TAFI protein on the effects of rivaroxaban. Rivaroxaban was shown to decrease thrombin generation, resulting in less TAFI activation, thus enhancing lysis. These effects were also shown to be less substantial in the presence of greater TAFI levels or the more stable Ile325 enzyme. These findings suggest a role for TAFI levels and the Thr325Ile polymorphism in the pharmacodynamics and pharmacogenomics of rivaroxaban.

SHock-INduced Endotheliopathy (SHINE): A mechanistic justification for viscoelastography-guided resuscitation of traumatic and non-traumatic shock

Irrespective of the reason for hypoperfusion, hypocoagulable and/or hyperfibrinolytic hemostatic aberrancies afflict up to one-quarter of critically ill patients in shock. Intensivists and traumatologists have embraced the concept of SHock-INduced Endotheliopathy (SHINE) as a foundational derangement in progressive shock wherein sympatho-adrenal activation may cause systemic endothelial injury. The pro-thrombotic endothelium lends to micro-thrombosis, enacting a cycle of worsening perfusion and increasing catecholamines, endothelial injury, de-endothelialization, and multiple organ failure. The hypocoagulable/hyperfibrinolytic hemostatic phenotype is thought to be driven by endothelial release of anti-thrombogenic mediators to the bloodstream and perivascular sympathetic nerve release of tissue plasminogen activator directly into the microvasculature. In the shock state, this hemostatic phenotype may be a counterbalancing, yet maladaptive, attempt to restore blood flow against a systemically pro-thrombotic endothelium and increased blood viscosity. We therefore review endothelial physiology with emphasis on glycocalyx function, unique biomarkers, and coagulofibrinolytic mediators, setting the stage for understanding the pathophysiology and hemostatic phenotypes of SHINE in various etiologies of shock. We propose that the hyperfibrinolytic phenotype is exemplified in progressive shock whether related to trauma-induced coagulopathy, sepsis-induced coagulopathy, or post-cardiac arrest syndrome-associated coagulopathy. Regardless of the initial insult, SHINE appears to be a catecholamine-driven entity which early in the disease course may manifest as hyper- or hypocoagulopathic and hyper- or hypofibrinolytic hemostatic imbalance. Moreover, these hemostatic derangements may rapidly evolve along the thrombohemorrhagic spectrum depending on the etiology, timing, and methods of resuscitation. Given the intricate hemochemical makeup and changes during these shock states, macroscopic whole blood tests of coagulative kinetics and clot strength serve as clinically useful and simple means for hemostasis phenotyping. We suggest that viscoelastic hemostatic assays such as thromboelastography (TEG) and rotational thromboelastometry (ROTEM) are currently the most applicable clinical tools for assaying global hemostatic function-including fibrinolysis-to enable dynamic resuscitation with blood products and hemostatic adjuncts for those patients with thrombotic and/or hemorrhagic complications in shock states.

Blood coagulation dissected

Hemostasis is the physiological control of bleeding and is initiated by subendothelial exposure. Platelets form the primary vascular seal in three stages (localization, stimulation and aggregation), which are triggered by specific interactions between platelet surface receptors and constituents of the subendothelial matrix. As a secondary hemostatic plug, fibrin clot formation is initiated and feedback-amplified to advance the seal and stabilize platelet aggregates comprising the primary plug. Once blood leakage has been halted, the fibrinolytic pathway is initiated to dissolve the clot and restore normal blood flow. Constitutive and induced anticoagulant and antifibrinolytic pathways create a physiological balance between too much and too little clot production. Hemostatic imbalance is a major burden to global healthcare, resulting in thrombosis or hemorrhage.

Lys 42/43/44 and Arg 12 of thrombin-activable fibrinolysis inhibitor comprise a thrombomodulin exosite essential for its antifibrinolytic potential

The thrombin-thrombomodulin (TM) complex activates thrombin-activable fibrinolysis inhibitor (TAFI) more efficiently than thrombin alone. The exosite on TAFI required for its TM-dependent activation by thrombin has not been identified. Based on previous work by us and others, we generated TAFI variants with one or more of residues Lys 42, Lys 43, Lys 44 and Arg 12 within the activation peptide mutated to alanine. Mutation of one, two, or three Lys residues or the Arg residue alone decreased the catalytic efficiency of TAFI activation by thrombin-TM by 2.4-, 3.2-, 4.7-, and 15.0-fold, respectively, and increased the TAFI concentrations required for half-maximal prolongation of clot lysis times (K_1/2) by 3-, 4,- 15-, and 24-fold, respectively. Mutation of all four residues decreased the catalytic efficiency of TAFI activation by 45.0-fold, increased the K_1/2 by 130-fold, and abolished antifibrinolytic activity in a clot lysis assay at physiologic levels of TAFI. Similar trends in the antifibrinolytic activity of the TAFI variants were observed when plasma clots were formed using HUVECs as the source of TM. When thrombin was used as the activator, mutation of all four residues reduced the rate of activation by 1.1-fold compared with wild-type TAFI, suggesting that these mutations only impacted activation kinetics in the presence of TM. Surface plasmon resonance data suggest that mutation of the four residues abrogates TM binding with or without thrombin. Therefore, Lys 42, Lys 43, Lys 44 and Arg 12 are critical for the interaction of TAFI with the thrombin-TM complex, which modulates its antifibrinolytic potential.

Fibrinolysis inhibitors in plaque stability: a morphological association of PAI-1 and TAFI in advanced carotid plaque

Essentials Fibrinolysis inhibitors are localized in advanced atheroma by immunohistology of endarterectomies. Neovascular endothelium/neocapillaries show thrombin-activatable fibrinolysis inhibitor (TAFI). Macrophage areas show free plasminogen activator inhibitor (PAI-1), notably in the vulnerable part. Free PAI-1 and TAFI stabilize active plaque area by inhibition of fibrinolysis and inflammation.

SUMMARY

Background Fibrinolysis plays an important role in destabilization of atherosclerotic plaques and is tightly regulated by specific inhibitors. Objective The fibrinolysis inhibitors plasminogen activator inhibitor type-1 (PAI-1) and thrombin-activatable fibrinolysis inhibitor (TAFI) were quantified and described in the morphological context of advanced carotid plaques American Heart Association VI-VIII to elucidate their role in plaque stability. Methods Immunohistochemistry in serial sections along the longitudinal axis of endarterectomies from patients with symptomatic carotid stenosis (n = 19) were studied using an antibody specific for free PAI-1 (I205), an antibody with high affinity for TAFI/TAFIa (CP17) and established antibodies for smooth muscle cells (α-actin), endothelial cells (von Willebrand factor [VWF]), macrophages (CD68) and platelets (CD42). Results PAI-1 and TAFI show a specific distribution in these advanced plaques with a maximum corresponding to the internal carotid artery (ICA). Free PAI-1 was mainly detected in macrophages and in intravascular thrombi, and TAFI in endothelial cells (ECs) but also macrophages. The one-way ANOVA analysis with Bonferroni's correction showed a significant increase of macrophages and ECs, TAFI and PAI-1 in areas with high neovascularization in endarterectomy sections corresponding to ICA. High Spearman factors for TAFI, PAI-1 and VWF indicate neovascularization as the main source of plasma proteins, transported by platelets into the atheroma (PAI-1) or expressed by ECs (TAFI). CD68 was highly associated with VWF, PAI-1 and especially TAFI, underlining the role of macrophages in fibrinolytic activity and inflammation. Conclusion The abundance of free PAI-1 and TAFI in the plaque may inhibit plasmin generation and thereby counteract plaque destabilization by fibrinolysis, cell migration and inflammation.

Elucidation of the molecular mechanisms of two nanobodies that inhibit thrombin-activatable fibrinolysis inhibitor activation and activated thrombin-activatable fibrinolysis inhibitor activity

Essentials Thrombin-activatable fibrinolysis inhibitor (TAFI) is a risk factor for cardiovascular disorders. TAFI inhibitory nanobodies represent a promising step in developing profibrinolytic therapeutics. We have solved three crystal structures of TAFI in complex with inhibitory nanobodies. Nanobodies inhibit TAFI through distinct mechanisms and represent novel profibrinolytic leads.

SUMMARY

Background Thrombin-activatable fibrinolysis inhibitor (TAFI) is converted to activated TAFI (TAFIa) by thrombin, plasmin, or the thrombin-thrombomodulin complex (T/TM). TAFIa is antifibrinolytic, and high levels of TAFIa are associated with an increased risk for cardiovascular disorders. TAFI-inhibitory nanobodies represent a promising approach for developing profibrinolytic therapeutics. Objective To elucidate the molecular mechanisms of inhibition of TAFI activation and TAFIa activity by nanobodies with the use of X-ray crystallography and biochemical characterization. Methods and results We selected two nanobodies for cocrystallization with TAFI. VHH-a204 interferes with all TAFI activation modes, whereas VHH-i83 interferes with T/TM-mediated activation and also inhibits TAFIa activity. The 3.05-Å-resolution crystal structure of TAFI-VHH-a204 reveals that the VHH-a204 epitope is localized to the catalytic moiety (CM) in close proximity to the TAFI activation site at Arg92, indicating that VHH-a204 inhibits TAFI activation by steric hindrance. The 2.85-Å-resolution crystal structure of TAFI-VHH-i83 reveals that the VHH-i83 epitope is located close to the presumptive thrombomodulin-binding site in the activation peptide (AP). The structure and supporting biochemical assays suggest that VHH-i83 inhibits TAFIa by bridging the AP to the CM following TAFI activation. In addition, the 3.00-Å-resolution crystal structure of the triple TAFI-VHH-a204-VHH-i83 complex demonstrates that the two nanobodies can simultaneously bind to TAFI. Conclusions This study provides detailed insights into the molecular mechanisms of TAFI inhibition, and reveals a novel mode of TAFIa inhibition. VHH-a204 and VHH-i83 merit further evaluation as potential profibrinolytic therapeutics.