Complement activation in anti-phospholipid syndrome: A clue for an inflammatory process?

Abstract

Anti-phospholipid syndrome (APS) is defined by recurrent arterial/venous thrombosis and/or fetal losses in the persistent presence of anti-phospholipid antibodies (aPL). In in vivo experimental models aPL thrombogenic activity is associated with a pro-inflammatory endothelial phenotype (increased adhesion molecule [ADM] expression and leukocyte adhesion) in addition to a pro-coagulant one (tissue factor [TF] expression). This is in line with the in vitro aPL ability to trigger intracellular signalling and to up-regulate ADM, TF and pro-inflammatory cytokine/chemokine expression at the mRNA and protein level in endothelial cells. Comparable effects were also reported in monocytes in vitro. In addition, complement activation is required by aPL to display their thrombogenic activity in in vivo models. Interestingly, complement activation blocking as well as Tumor Necrosis Factor alpha neutralization protect animals from aPL-induced fetal losses. Altogether these findings speak in favour for a role of inflammation in APS in spite of the absence of a clear inflammatory signature in the patients. We could not find any complement (C3c and C4d) deposition in the placentas from 2 late abortions (20 weeks of gestation) in APS women. Further studies are necessary to investigate whether complement activation and inflammatory processes found in animal models are taking place in APS patients.

Introduction

Anti-phospholipid syndrome (APS) is a clinical disorder, characterised by recurrent thrombosis, fetal losses and circulating anti-phospholipid antibodies (aPL), reacting to PL-binding plasma proteins (namely β2glycoprotein I [β2GPI] and prothrombin) [1].

It is widely accepted that β2GPI-dependent aPL play a pathogenic role. In fact, several in vivo studies showed a direct effect of these autoantibodies in inducing fetal loss [2], [3], [4], [5], endothelial activation [6] and in enhancing the pro-coagulant effect of a trigger factor [6], [7], [8].

The thrombophilic state in APS has been associated with a vasculopathy rather than with a true inflammatory vasculitic process. Actually, even the recent classification criteria state that inflammatory signs should be absent in the thrombotic manifestations of APS [9]. Such a statement is apparently in clash with the evidence that aPL, and in particular anti-β2GPI antibodies, are able to induce an inflammatory phenotype in both monocytes and endothelial cells (EC) in in vitro experimental models [10]. Actually these cell types may up-regulate the synthesis and secretion of pro-inflammatory cytokines, chemokines and adhesion molecules (ADM) when incubated in the presence of β2GPI-dependent aPL [11], [12], [13], [14], [15]. Animals deficient for ADMs or treated with anti-VCAM-1 blocking antibodies were shown to be protected from the thrombogenic effect of human aPL IgG and VCAM-1 up-regulation was shown in the endothelium of carotid arteries in animals injected with human aPL IgG fractions (rev. in ref. [40]). As a whole these findings do suggest that an endothelial pro-inflammatory phenotype may be present also in in vivo APS animal models.

The obstetrical counterpart of APS is represented by the occurrence of pregnancy losses and pregnancy complications such as early pre-eclampsia. Both histopathological studies and experimental evidence of a direct effect of aPL on placental tissues strongly suggest that thrombotic phenomena alone cannot explain the obstetrical manifestations of the syndrome [16], [17]. A local inflammatory process has been suggested to play a role in inducing a defective placentation as a cause of the recurrent fetal losses associated to aPL. Actually local Tumor Necrosis Factor alpha (TNFα) production, neutrophil recruitment and complement activation/deposition have been described in experimental animal models of aPL-mediated fetal loss [18], [19], [20], [21]. In line with these experimental findings, histological evaluation of terminal placental tissues of APS women reported an increased infiltration of monocyte-macrophage lineage cells [22].