Releasing the Brakes on the Fibrinolytic System in Pulmonary Emboli

Abstract

P ulmonary embolism (PE) affects millions of patients each year and is a leading cause of hospital mortal-ity. 1–3 Physiological or endogenous fibrinolysis usu-ally fails to dissolve pulmonary emboli that may cause acute obstructive complications such as hypotension and shock. The persistence of thromboemboli also may lead to serious complications like chronic thromboem-bolic pulmonary hypertension and right ventricular dys-function. 4 Anticoagulation is the standard therapy for PE, but anticoagulation only prevents new thrombus forma-tion on preexisting thromboemboli and does not cause thrombus dissolution. 5 In patients with massive PE, phar-macological, r-tPA (recombinant tissue-type plasminogen activator) therapy is given to rapidly dissolve thrombo-emboli to enhance hemodynamic function, relieve hypo-tension and reduce mortality. 6,7 However, treatment with r-tPA and other plasminogen activators causes serious bleeding that restricts its use to patients with massive PE who have a high risk of mortality. 6,7 Why endogenous fibrinolysis fails to dissolve acute pulmonary emboli is not well understood. Studies using genetically modified mice and inhibitors have shown that tissue-type (t) 8 and urinary-type (u) plasminogen activator (PA), 9 which convert plasminogen to the active enzyme plasmin, contribute to the dissolution of experimental pulmonary emboli. In a similar fashion, their inhibitor, plasminogen activator inhibitor-1 (PAI-1) suppresses fi-brinolysis. 10 The plasmin inhibitor α2-antiplasmin 11,12 af-fects the dissolution of experimental pulmonary emboli and epidemiological studies in humans identify higher α2-antiplasmin levels as a risk factor for venous throm-boembolism. 13 Whether these components of the fibri-nolytic system are expressed at the site of acute pul-monary emboli is unknown. Similarly, little information exists about the relative contribution of the endogenous plasminogen activation and plasmin inhibition, and their interactions, to the rate and extent of fibrinolysis in vivo. Finally, it is uncertain whether promoting fibrinolysis by selectively altering plasminogen activation or plasmin inhibition has the same downstream effects on coagula-tion and bleeding. We examined the expression of the plasminogen acti-vation and plasmin inhibition system at the site of experi-mental pulmonary emboli. We compared the effects of plasminogen activation with α2-antiplasmin inactivation on experimental fibrinolysis, fibrinogen levels, and bleed-ing. Experimental α2-antiplasmin inactivation enhances endogenous fibrinolysis to levels comparable to levels achieved with higher-dose r-tPA, but unlike r-tPA, does not cause fibrinogen destruction or enhance surgical bleeding. These data suggest that plasmin inhibition by α2-antiplasmin is a key rate-limiting step in the acute dis-solution of experimental pulmonary emboli. METHODS Proteins and Reagents Reagents were purchased from the following sources: human α2-antiplasmin (Athens Research and Technology); human plasmin (Calbiochem); bovine thrombin (Sigma); citrated fro-zen human plasma (Lampire Biological Laboratories); chime-ric α2-antiplasmin–inactivating antibody (TS23, Translational Sciences); 125 I-fibrinogen (Perkin-Elmer); fluorescein isothio-cyanate-fibrinogen (Molecular Innovations); r-tPA (Alteplase, Genetech Inc.); and all the other reagents, if not specified (Sigma). Pulmonary Embolism Animal studies were approved by the Institutional Animal Care and Use Committee. The dissolution of experimental pulmo-nary emboli, fibrinogen degradation, and bleeding were exam-ined in a humanized model, in which adult male and female α2-antiplasmin –/– mice (KOMP, UC Davis) on a C57Bl/6 back-ground were supplemented with physiological amounts of human α2-antiplasmin. To prepare plasma clots, pooled fresh frozen human plasma (5 µL) was mixed with trace amounts of 125 I-fibrinogen (≈5000 cpm), 0.25 U/mL (NIH units) bovine thrombin, and 100 mmol/L CaCl 2 in a total clot volume of 12.5 µL. After overnight incubation at 37°C, the clots were compressed, washed thoroughly with saline, and cut into 20 pieces before embolization. Mice were anesthetized by ket-amine, xylazine, and atropine and kept on continuous anes-thesia during the surgery. An anterior midline incision was made in the neck, and the left jugular vein was exposed. A small incision was made in the jugular vein and a nonoc-clusive PE-10 catheter was inserted. Human α2-antiplasmin (4.7 mg/kg) was promptly infused intravenously to achieve physiological levels. Then, clots in saline were embolized into the lungs through the PE-10 catheter. After 30 minutes,