Inhibition of complement activation: a novel mechanism for the protective effects of heparin in antiphospholipid antibody-induced pregnancy loss

The antiphospholipid syndrome (APS), defined by thrombosis and recurrent pregnancy loss in the presence of antiphospholipid (aPL) antibodies, is generally treated with anticoagulation. Because complement activation is essential and causative in aPL antibody-induced fetal injury, we hypothesized that heparin protects mice from fetal loss in APS by preventing complement activation on trophoblasts and that anticoagulation, per se, is not sufficient to prevent miscarriage. We used a murine model of APS in which pregnant mice are injected with human IgG-containing aPL antibodies (aPL-IgG) or IgG from healthy individuals (NH-IgG). Passive transfer of aPL-IgG caused a 43.3 ± 3.5% frequency of fetal resorption (P < 0.001 versus NH-IgG) and treatment with either unfractionated heparin (UFH) (10 U or 20 U subcutaneously twice daily) or low molecular weight heparin (LMWH) (enoxaparin) reduced the frequency of fetal resorption to that of mice treated with NH-IgG: UFH10 + aPL, 9.6 ± 3.2; UFH20 + aPL 10.4 ± 2.1; LMWH + aPL, 11.8 ± 2.8; NH-IgG, 10.5 ± 2.5% (P < 0.01 versus aPL). Even in the absence of detectable anticoagulation, as in mice treated with 10 U UFH, heparins prevented aPL-induced pregnancy loss and inhibited systemic complement activation evidenced by generation of C3adesArg (aPL, 1212 ± 101* ng/ml; UFH10U + aPL, 178 ± 25; LMWH + aPL, 180 ± 15; NH-IgG, 215 ± 381, * P < 0.05 versus NH-IgG). Heparin limited C3 deposition in deciduas of mice treated with aPL. Neither fondaparinux nor hirudin inhibited the generation of complement split products or prevented pregnancy loss, despite anticoagulation levels comparable with heparin, indicating that anticoagulation is insufficient therapy for APS-associated miscarriage.

We assessed the effects of anticoagulants on complement activation in vitro using BeWo cells, trophoblast-like cells with externalized phosphatidylserine recognized by aPL. APL-opsonized BeWo cells cultured with mouse serum activated complement, evidenced by increased surface-bound C3 fragments (detected as C3 staining by FACS) and generation of soluble C3adesArg in supernatants. In the presence of UFH or LMWH, however, complement activation was completely inhibited (C3adesArg [ng/ml]: aPL, 680 ± 30; UFH + aPL, 120 ± 20*; LMWH + aPL, 105 ± 18*; % C3-positive BeWo cells by FACS: aPL, 31 ± 11; aPL + UFH, 9 ± 5*; LMWH + aPL, 5 ± 3; * P < 0.01 versus aPL). In contrast, neither fondaparinux nor hirudin inhibited complement activation in vitro.

In conclusion, heparins inhibit complement activation in vitro and in vivo. Heparins may prevent obstetrical complications in APS by blocking activation of complement induced by aPL targeted to deciduas – not by preventing placental thrombosis. These findings explain how subanticoagulant doses of heparin can limit antibody-mediated tissue injury.