32 Mechanism of Thrombosis in Antiphospholipid Syndrome: Binding to Platelets Joan-Carles Reverter and Dolors Tàssies Introduction Antiphospholipid antibodies (aPL) are related to thrombosis in the antiphospho- lipid syndrome (APS) [1, 2] and numerous pathophysiological mechanisms have been suggested involving cellular effects, plasma coagulation regulatory proteins, and fibrinolysis [3, 4]: aPL may act as blocking agents directly inhibiting antigen enzymatic or co-factor function of hemostasis; may bind fluid-phase antigens of hemostasis involved proteins and then decrease plasma antigen levels by clearance of immune complexes; may form immune complexes with their antigens that may be deposited in blood vessels causing inflammation and tissue injury; may cause dysregulation of antigen–phospholipid binding due to cross-linking of membrane bound antigens; and may trigger cell mediated events by cross-linking of antigen bound to cell surfaces or cell surface receptors [3, 4]. Moreover, several characteris- tics of the aPL, such as the concentration, class/subclass, affinity or charge, and several characteristics of the antigens, as the concentration, size, location or charge, may influence which of the theoretical autoantibody actions will occur in vivo [3]. Among the cellular mechanisms supposed to be involved, platelets have been considered as one of the most promising potential target for circulating aPL that may cause antibody mediated thrombosis as a part of the clinical spectrum of the autoimmune disorder of the APS. In the present chapter we will focus on the inter- actions that involve aPL binding to platelet membrane or platelet membrane bound antigens. Platelets as Target for aPL Platelets play a central role in primary hemostasis involving platelet adhesion to the injured blood vessel wall, followed by platelet activation, granule release, shape change, and rearrangement of the outer membrane phospholipids and proteins, transforming them into a highly efficient procoagulant surface [5]. In addition, thrombocytopenia is also a clinical manifestation of APS. For these reasons platelets 403 404 Hughes Syndrome itself have been considered as a potential target for aPL and this fact has been extended to thrombotic mechanisms [3]. Several facts support platelets as the target for aPL. Studies performed in the aggregometer or in flowing conditions and the evaluation of platelet activation markers in vitro and in vivo in patients with the APS are used as demonstration. Activation and spontaneous aggregation of platelets was reported in aggregomet- ric studies to be caused directly by aPL in early reports [6, 7]. Other authors did not find this ability of aPL to initiate platelet activation [8, 9] or report inhibition of aggregation caused by aPL [10]. However, the most realistic interpretation is that in the aggregometric studies aPL may cooperate in platelet activation by making platelets more reactive to the action of weak or low-dose agonists [8, 11–13]. A calcium independent platelet aggregation (thromboagglutination) has also been found in patients with APS [14]. Other studies performed using flowing systems that simulate physiological condi- tions [15, 16] demonstrated, in both systemic lupus erythematosus (SLE) patients and in primary APS patients, increased formation of platelet thrombi when small amounts of patients’ plasmas or purified immunoglobulins with anticardiolipin activity were added to normal blood, but this increase only occurs when plasma or immunoglobulins from patients with thrombotic history was employed. Similar results were obtained in the same system when the experiments were performed β using human monoclonal anticardiolipin (anti– 2-glycoprotein I) antibodies [17], β and the 2-glycoprotein–dependence of this phenomenon has been evidenced [18]. β β Additionally, dimers of 2-glycoprotein I, that mimic effects of 2-glycoprotein β I–anti– 2-glycoprotein I antibody complexes, have been found to increase platelet adhesion and thrombus formation in a flow system but not increased aggregation in an aggregometer [19]. Several studies have been performed to identify platelet activation markers in patients with APS. Some of them investigate the eicosanoid regulation in these patients showing inhibition of prostacyclin synthesis [20–22] and/or increased platelet thromboxane production [13, 22–24]. These results were found in in vitro experiments and in vivo in patients with the APS. However, these results have not β been found by others [25]. More recently, anticardiolipin– 2-glycoprotein I com- plexes have been found to induce platelet overactivity resulting in excessive pro- duction of thromboxane A2, presumably by decreased platelet cyclic AMP activity, causing increased platelet aggregation [26]. Looking for more direct markers of platelet activation, other authors have found an increase of the CD62p (P-selectin), an integral protein found in the α-granules, on platelet surface [27, 28] and/or the soluble CD62 [29, 30], loss from the activated platelet membranes to the plasma, in patients with APS, but not all the authors found the same results [31, 32]. Moreover, platelet CD63 expression, a lysosomal granule protein exposed after platelet activation, has been found increased in primary APS patients [30], but not in all the studies [28]. PAC-1, an antibody that detects the conformational change of the platelet membrane glycoprotein IIb–IIIa complex that occurs when the platelet is activated and probably a more sensitive index of platelet activation than degranulation proteins, binding has been also reported significantly increased in patients with primary APS [30]. In these patients, a significantly increased number of circulating platelet microparticles have been found [33], supporting an increased platelet activation in vivo, but these results have not been confirmed by others [30]. Annexin V binding to platelets, which indi- Mechanism of Thrombosis in Antiphospholipid Syndrome: Binding to Platelets 405 cates phosphatidyl serine exposure on the outer leaflet of the membranes during platelet activation, measured by flow cytometry has been reported as increased in a series of patients with SLE with thrombosis [34], most of them with aPL, but has been found normal in primary APS [30]. In several studies, increased platelet–leukocyte complexes, generated after platelet activation or as an inflamma- tory feature of autoimmune disease, have been found in APS patients [30, 35]. Finally, β-thromboglobulin, another platelet activation marker, has been evaluated in vitro or in vivo in few studies with discrepant results [34, 36, 37]. The interaction of aPL with platelets can occur in at least three different ways. First, immunoglobulins may bind through the Fab terminus with specific platelet antigens (or with other antigens deposed on platelets) in a classical antigen–anti- body reaction; second, immune complexes may bind to platelets via Fcγ receptor (FcγR); and third, aPL, as other immunoglobulins, may bind to platelets in a non- specific manner by mechanisms not well characterized but probably related to platelet membrane injuries [38]. The last mechanism, non-specific binding, does not seem to have a pathophysiological role in APS-related thrombosis. Binding of aPL to Phospholipids on Platelet Membrane Normal platelet membranes have a clear phospholipid asymmetry. The outer leaflet of the phospholipid bilayer is rich in choline-phospholipids, whereas amino-phos- pholipids are located in the inner leaflet [39]. Then, in resting platelets phosphatidyl serine is predominantly located in the cytoplasmic leaflet of the platelet mem- branes, but platelets undergoing activation lose their physiological phospholipid asymmetry and increase the exposure of anionic phospholipids, mainly phos- phatidyl serine, on the external cell membrane [40]. During platelet activation a fast transbilayer movement of phospholipids (flip-flop) is produced [39]. In addition, during platelet activation exposure of anionic phospholipids is accompanied by shedding of procoagulant microvesicles [41]. The extent of the platelet membrane phospholipid expression depends on the type of platelet activator, being calcium- ionophore the most potent, followed in order of potency by the complement complex C5b-9, collagen plus thrombin, collagen, and thrombin [39]. ADP and epi- nephrine are very weak activators considering the induction of the phospholipid flip-flop. Some studies have demonstrated that aPL may bind to platelet surface, and this binding is higher on activated or damaged platelets than in resting ones [16, 37, 42]. In addition, several agonists, like collagen, may be more able to activate platelets in a way that makes them reactive to aPL than others, like ADP [16]. These results agree with the differential expression of phospholipids during platelet activation. APL with lupus anticoagulant (LA) activity also bind to platelets and to platelet- derived microvesicles at least in part by membrane-bound prothrombin [43]. However, it has been reported that LA antibodies can bind to activated platelets in absence of any plasma component, but it can not be excluded that the required co- factor may be released from the platelet granules [43]. Membranes from activated platelets are an important source of negatively charged phospholipids to provide a catalytic surface for interacting coagulation factors [43]. The ability of platelets to support tenase and prothrombinase activity 406 Hughes Syndrome (and also protein C activity) correlates with the extent to the