Carbon monoxide derived from degradation of heme by heme oxygenase or carbon monoxide releasing molecules (CORMs) has been demonstrated to decrease thrombosis in vivo and to weakly inhibit platelet aggregation. We tested the hypothesis that carbon monoxide released from tricarbonyldichlororuthenium (II) dimer (CORM-2) would diminish the velocity of formation and strength of plasma thrombi as determined by thrombelastography. Normal plasma was exposed to 0 or 100 micromol/l CORM-2 or inactivated CORM-2 (iCORM-2), with coagulation initiated with tissue factor or celite (n = 8 per condition). Additional experiments utilized factor XIII (FXIII) deficient plasma activated with celite. Coagulation kinetics was monitored with thrombelastography for 15 min. CORM-2, and to a lesser extent, iCORM-2, significantly (P < 0.05) increased the velocity of formation (122 and 56%, respectively) and strength (66 and 57%, respectively) of plasma clots initiated with either tissue factor or celite compared with thrombi not exposed to CORM-2 or iCORM. In FXIII deficient plasma CORM-2 significantly increased the velocity of clot formation (264%) and strength (240%). Carbon monoxide and iCORM-2 derived from CORM-2 markedly enhance the velocity of clot growth and strength. These findings serve as the rationale for further investigations to determine if CORMs could be utilized as hemostatic agents.

Two parallel lines of investigation elucidating novel mechanisms by which iron (scanning electron microscopy-based) and carbon monoxide (viscoelastic-based) enhance coagulation and diminish fibrinolysis have emerged over the past few years. However, a multimodal approach to ascertain the effects of iron and carbon monoxide remained to be performed. Such investigation could be important, as iron and carbon monoxide are two of the products of heme catabolism via heme oxygenase-1, an enzyme upregulated in a variety of disease states associated with thrombophilia. Human plasma was exposed to ferric chloride, carbon monoxide derived from carbon monoxide-releasing molecule-2, or their combination. Viscoelastic studies demonstrated ferric chloride and carbon monoxide mediated enhancement of velocity of growth, and final clot strength, with the combination of the two molecules noted to have all the prothrombotic kinetic effects of either separately. Parallel ultrastructural studies demonstrated separate types of fibrin polymer cross-linking and matting in plasma exposed to ferric chloride and carbon monoxide, with the combination sharing features of each molecule. In conclusion, we present the first evidence that iron and carbon monoxide interact with key coagulation and fibrinolytic processes, resulting in thrombi that begin to form more quickly, grow faster, become stronger, and are more resistant to lysis.

Carbon monoxide, derived from carbon monoxide-releasing molecules, has been recently demonstrated to enhance the velocity of formation and strength of plasma thrombi. We tested the hypothesis that carbon monoxide-releasing molecule-2 would modulate fibrinolysis of plasma thrombi. Normal plasma was exposed to 0, 25, 50, 100 or 200 μmol/l carbon monoxide-releasing molecule-2, with coagulation activated with tissue factor and fibrinolysis initiated with tissue-type plasminogen activator. Additional experiments utilized factor XIII, plasminogen activator inhibitor-1, thrombin activatable fibrinolysis inhibitor or α₂-antiplasmin-deficient plasmas. Thrombus growth/disintegration kinetics was monitored with thrombelastography. Carbon monoxide-releasing molecule-2, in a concentration-dependent fashion, increased the velocity of thrombus formation and strength, and markedly attenuated fibrinolysis in normal plasma. In factor XIII-deficient plasma, carbon monoxide-releasing molecule-2 mediated effects on thrombus growth/disintegration kinetics were similar to that seen with normal plasma; however, carbon monoxide-releasing molecule-2 had a less marked effect on thrombus growth/disintegration in both plasminogen activator inhibitor-1 and thrombin activatable fibrinolysis inhibitor-deficient plasma, with even less carbon monoxide-releasing molecule-2-mediated effects noted in α₂-antiplasmin-deficient plasma. Carbon monoxide-releasing molecule-2 attenuated fibrinolysis by enhancing the velocity of clot growth and strength while augmenting the effects of plasminogen activator inhibitor-1, thrombin activatable fibrinolysis inhibitor and α₂-antiplasmin. These findings serve as the rationale for further investigations to determine if carbon monoxide-releasing molecules could be utilized as hemostatic agents.

The objective of this study was to determine how carbon monoxide directly modifies fibrinogen utilizing liquid chromatography-mass spectrometry (LC-MS/MS) by examining fibrinogen exposed to carbon monoxide releasing molecule-2 [tricarbonyldichlororuthenium (II) dimer; CORM-2]. Purified fibrinogen was exposed to 0, 25, 50 or 100 μmol/l CORM-2 for 5 min at 37°C and then stored at −80°C before analyses with LC-MS/MS. In a second series of experiments, normal plasma was exposed to 0 or 100 μmol/l CORM-2 in the absence or presence of the nitric oxide donor sodium nitroprusside and hydroquinone (an organic reductant) to compete with carbon monoxide binding to a putative heme group found on fibrinogen. Coagulation was activated with tissue factor (n = 8 per condition). Thrombus growth was monitored with thrombelastography for 15 min. LC-MS/MS did not detect any direct modifications of amino acids in fibrinogen, but detection of small regions of both the alpha and gamma chains was lost following exposure to CORM-2 and endoproteinase digestion with trypsin and Glu-C. An ion with the same m/z and expected retention time as heme was found in the purified fibrinogen. Exposure of plasma to nitric oxide/hydroquinone significantly decreased CORM-2-mediated enhancement of coagulation without affecting the coagulation kinetics of plasma not exposed to CORM-2. Carbon monoxide derived from CORM-2 likely modifies fibrinogen via modulation of a fibrinogen-associated heme group(s). Whereas the precise molecular location of heme attachment and three-dimensional conformational change secondary to carbon monoxide exposure remain to be determined, fibrinogen appears to be a carbon monoxide sensing molecule.

Cigarette smoking is associated with plasmatic hypercoagulability, and carbon monoxide has been demonstrated to enhance coagulation by binding to a fibrinogen-bound heme. Our objective was to design and test a redox-based method to detect carboxyhemefibrinogen. Normal, pooled, citrated plasma was exposed to 0–100 μmol/l carbon monoxide releasing molecule-2 (tricarbonyldichlororuthenium (II) dimer; CORM-2) before or after exposure to the organic reductant phenylhydroxylamine (PHA, 0–30 mmol/l), a compound that rapidly converts Fe⁺² to Fe⁺³ in heme. Addition of calcium and tissue factor activation in disposable thrombelastographic cups was performed, followed by data collection at 37°C for 15 min. Elastic modulus (G, dynes/cm²) was the primary endpoint. CORM-2 significantly increased G values by 67.8% compared to unexposed plasma; pretreatment with 10 mmol/l PHA significantly decreased G values in CORM-2-exposed plasma by 77.1%, whereas 30 mmol/l PHA was required to significantly decrease G values by 64.0% in plasma following CORM-2 pre-exposure. G values were not significantly different between unexposed plasma and plasma exposed to CORM-2 followed by 30 mmol/l PHA addition. Conversion of fibrinogen-bound to the metheme state alone decreased G by 34.3–38.9% following exposure to 10–30 mmol/l PHA. Conversion of fibrinogen-bound heme Fe⁺² to Fe⁺³ with PHA abrogated carbon monoxide-mediated increases in clot strength. Clinical trials are planned to investigate smoking individuals to mechanistically link carboxyhemefibrinogen formation with in-vitro hypercoagulability.