Membrane microvesicles: a circulating source for fibrinolysis, new antithrombotic messengers

Thrombus lysis is the consequence of a restricted number of reactions localized on the surface of fibrin and cell membranes. A functional defect or an insufficient fibrinolytic response may lead to thrombosis with severe or fatal clinical consequences. Despite this clinical need, and real progress in the knowledge of the different components of this system (plasminogen activators, inhibitors and receptors), including the structure-function relationship unveiled by the crystal structure of plasminogen, the functional evaluation of fibrinolysis still remains a challenge in hemostasis. The absolute requirement of a template for molecular assembly of plasminogen and its activators (tPA or uPA) restricts the formation of plasmin onto the surface of fibrin and cells. In contrast, during fibrinolysis, plasmin and tPA released from the clot are immediately neutralized by their respective inhibitors, α2-antiplasmin and plasminogen activator inhibitor 1, PAI-1. It is, therefore, almost impossible to detect fibrinolytic activity in plasma with current methods or to measure the degree of fibrinolysis directly on clots that are unavailable in a clinical setting.

Interestingly, it was recently discovered that circulating membrane microvesicles, which behave as messengers of cell activation, might be indicators of the fibrinolytic response to an inflammatory or prothrombotic process.^1,2 These fibrinolytic microvesicles transport at their membrane the plasminogen activators expressed by the cell of origin: tPA from endothelial cells and uPA from leukocytes.³ These microvesicles generate plasmin in situ upon binding of plasminogen to carboxy-terminal lysine residues of membrane receptors. Thus, co-assembly of plasminogen and its activator onto the same surface is required to trigger the fibrinolytic or proteolytic process classically described on cell membranes and fibrin.

We recently discovered that moving surfaces such as microvesicles might also participate in a new mechanism of plasmin formation requiring a cross-talk between two different surfaces.² In this fibrinolytic cross-talk, one of the surfaces bearing uPA/uPAR (leukocyte or its microvesicles) can recognize and activate plasminogen carried by the other surface (platelets, fibrin or extracellular matrix). Plasminogen thus bound adopts an open conformation that is readily transformed into plasmin. Recently published studies^4–6 are in agreement with these hitherto unknown fibrinolytic pathways and potentially novel bio-markers in clinical practice.

This novel mechanism of plasmin formation at the surface of platelets, extracellular matrix or fibrin by microvesicles bearing uPA raises the question of its involvement in different pathophysiological situations.

Platelets and derived microvesicles do not bear plasminogen activators. However, they can immobilize plasminogen on their surface via carboxy-terminal lysine residue-dependent interactions.⁷ Platelets may thus contribute to increase the concentration of plasminogen within the clot. Microvesicles bearing uPA could then cross-talk with platelet-bound plasminogen thus allowing in situ plasmin formation and re-canalization of an occluded vessel. Similarly, activation of plasminogen bound to fibrin by leukocytes bearing uPA plays a role in endogenous fibrinolysis.⁶

Apart from its fibrinolytic function, plasmin formation by the uPA/uPAR system is involved in tissue remodeling and plays a critical role in cell migration and angiogenesis. The ability of microvesicles to generate plasmin influences and modulates the repair process of endothelial progenitor cells. Small amounts of microvesicles bearing the uPA/uPAR system promote cell migration and angiogenesis whereas at high concentrations excess plasmin leads to matrix degradation, decreased cell adhesiveness and finally apoptosis.¹

Platelet microvesicles may promote metastasis and angiogenesis,⁸ and high amounts of uPA/uPAR were associated with matrix degradation and loss of cell adhesion in advanced metastatic cancers.⁹ It is interesting to note that the described fibrinolytic/proteolytic cross-talk mechanism is only possible in the presence of uPA, which has been found on microvesicles emitted by cancer cells.

The structure and function of the plasminogen activation system and its role in the maintenance of hemostasis and thrombosis prevention is now well established. However, detection of a dysfunction of this system remains a major challenge in clinical practice. Actually, plasminogen activators circulate at extremely low concentrations as inactive complexes with PAI-1 whereas active plasminogen activators are exclusively located on the cell membrane or fibrin. Since most measurements performed in plasma or its euglobulin fraction do not take into account the contribution of surface-bound plasminogen activators, it is impossible to quantify a lack of tPA or uPA activity that may be the cause of a fibrinolytic insufficiency. The recent discovery of fibrinolytic microvesicles and their role in fibrinolytic cross-talk has opened up new perspectives. We propose that the fibrinolytic activity conveyed by microvesicles could be the real source of fibrinolysis in circulating blood. These microvesicles would act on and within the clot, thus explaining the lack of systemic fibrinolysis. We suggest that the fibrinolytic activity of endothelial and leukocyte microvesicles compensates the activity of procoagulant microvesicles at a local level. Spontaneous re-canalization in 15%–20% ST-segment elevation myocardial infarction¹⁰ could, therefore, be explained by fibrinolytic microvesicles counterbalancing procoagulant microvesicles. On the contrary, a failure of timely fibrinolytic response by microvesicles would result in persistence of occluding thrombi. Accordingly, the functional balance between these two types of microvesicles would result in a physiological hemostatic response, while the lack of fibrinolytic microvesicles may allow thrombus formation. Within this context, it will be possible to use microvesicles as vectors of fibrinolysis and pericellular proteolysis. The existence of a hemorrhagic syndrome (Quebec platelet disorder) caused by profibrinolytic platelets with an abnormal expression of uPA is consistent with this hypothesis.¹¹ Transgenic mice that express uPA in their platelets are resistant to arterial thrombosis and transfusion of these platelets to control mice prevents occlusive arterial thrombi formation.¹² Similarly, addition of fibrinolytic microvesicles to plasma or euglobulins decreases the microplate clot lysis time in a concentration-dependent manner ((E. Anglés-Cano, unpublished results, 2012). Although the venous occlusion test induces the release of fibrinolytic microvesicles, the majority of fibrinolytic assays, including the euglobulin clot lysis time, exclude microvesicle-bound plasminogen activators ((L. Plawinski, unpublished results, 2012)). Thus, development of a new test for the detection of fibrinolytic microvesicles directly in plasma will throw a new light on both the pathophysiology of fibrinolysis and the management of thrombosis in clinical practice.¹³