Journal of Autoimmunity (2000) 15, 241–247 doi:10.1006/jaut.2000.0420, available online at http://www.idealibrary.com on Review Article Experimental Thrombosis and Antiphospholipid Antibodies: New Insights Silvia S. Pierangeli, Azzudin E. Gharavi and E. Nigel Harris Morehouse School of Medicine, Atlanta, GA, USA Keywords: antiphospholipid antibodies, thrombosis, antiphospholipid syndrome, animal models, lupus anticoagulant Introduction The association of antiphospholipid (aPL) antibodies with recurrent arterial or venous thrombosis is well recognized but the molecular mechanisms are not well understood [1]. Antiphospholipid antibodies are now recognized as one important cause of acquired or secondary hypercoagulable state. Thrombotic events in APS involve either venous or arterial circulations, and they involve various organs. Evidence exists that thrombotic events in patients with aPL segregate into venous or arterial. For example patients that present with a venous event such as deep vein thrombosis tend to have recurrent deep venous thrombosis, whereas patients with stroke tend to have recurrent stroke [2]. Furthermore, APL antibodies are hetero- genous and it is likely that more than one mechanism may be involved in causing thrombosis. Although, both obstetric complications and thrombotic compli- cations are frequently associated with the presence of aPL, they may rise from different and independent mechanisms. This implies perhaps that ‘thrombotic’ mechanisms may be heterogenous or that differences exist in host susceptibilities. The theories of pathogenesis of the thromboembolic phenomena associated with aPL revolve around either abnormalities in platelet function, activation or induc- tion of endothelial cell (EC) dysfunction or aPL interfering with phospholipid–protein complexes that play a critical role in regulation of the coagulation, such as protein C, thrombomodulin or prothrombin [3–8]. A disruption of EC function is an attractive hypoth- esis, because this could explain the non-inflammatory vasculopathy present in APS patients. Furthermore, a large body of evidence indicated that aPL antibodies bind to ECs [9, 10]. Hence, antibody-EC mediated injury and EC activation has been identified as a significant potential factor that may be involved in the pathogenesis of thrombosis by aPL. The conversion of a normal anti-thrombotic state into a prothrombotic state may be the primary pathophysiological event in acquired hypercoagulable states, including the APS [11, 12]. Some of the most provocative data indicate that aPL activate ECs an induce upregulation of EC adhesion molecules [9, 13, 14]. Actually, endothelial adhesion is able per se to induce leukocyte activation and it is widely accepted that activated monocytes for example are able to display a procoagulant activity [11]. Del Papa et al. were the first to demonstrate that aPL or anti- 2 -GPI antibodies upregulate EC adhesion molecules and this effect is directly related to EC binding of the antibodies and IL-1a, and induce an increase of interleukin-6, accompanied by production of Il-1b, which in turn upregulates the adhesion mol- ecule expression [9, 13]. In addition, the same group showed that anti- 2 -GPI antibodies also induced a dose-dependent increase in the endothelial produc- tion of 6-keto prostaglandin F [13]. The findings of Del Papa et al. were confirmed by Simantov and colleagues [14], who showed upregulation of adhe- sion molecules (I-CAM-1, V-CAM-1 and E-selectin) on HUVEC and increased adhesion of monocytes to ECs by aPL antibodies in the presence of  2 -GPI in vitro [14]. Correspondence to: Silvia S. Pierangeli, PhD, Department of Micro- biology and Immunology, Rm 1236, 720 Westview Dr SW, Atlanta, GA 30310-1495, USA. Fax: 404-752-1644. E-mail: pierans@ msm.edu or pierans@aol.com 241 0896–8411/00/060241+07 $35.00/0 © 2000 Academic Press