REDUCTION OF ANNEXIN-V (PLACENTAL ANTICOAGULANT PROTEIN-I) ON PLACENTAL VILLI OF WOMEN WITH ANTIPHOSPHOLIPID ANTIBODIES AND RECURRENT SPONTANEOUS-ABORTION

OBJECTIVE: The mechanism by which antiphospholipid antibodies are associated with pregnancy loss and thromboembolic conditions has yet to be elucidated. Annexin-V, an anticoagulant phospholipid-binding protein, is normally present in syncytiotrophoblasts lining the placental villi, where it may play a role in the maintenance of intervillous blood fluidity. We therefore investigated the distribution of annexin-V in placentas of patients with antiphospholipid antibodies in situ and then used short-term villous cultures to study the direct effect of antiphospholipid antibodies on the immunolocation of annexin-V. STUDY DESIGN: We performed a blinded study by means of computerized morphometric analysis of placental tissues that were stained for annexin-V with affinity-purified polyclonal antibody in an avidin-biotin peroxidase system. The distribution of villous surface annexin-V on cross sections of placentas of patients with antiphospholipid antibodies was compared with that of placentas from patients with uncomplicated pregnancies, elective abortions, and pregnancy losses not associated with antiphospholipid antibodies (n = 8 for each group). We quantitated villous surface annexin-V in cultured placental villi that were incubated with antiphospholipid antibodies immunoglobulin G compared with normal immunoglobulin G and measured annexin-V levels by enzyme-linked immunosorbent assay in conditioned media and in the villi. RESULTS: The mean villous surface annexin-V of the group with antiphospholipid antibodies was 26.2% +/- 17% (SD) versus 93.9% +/- 5.7% in the normal control group (p < 0.0001). Villi from patients undergoing elective abortions and with pregnancy losses that were not attributed to antiphospholipid antibodies also showed higher mean villous surface annexin-V levels (86.9% +/- 10.6% and 83.5% +/- 11.3%, respectively, p < 0.0001). Organ culture of normal placental villi with affinity-purified immunoglobulin G from patients with antiphospholipid antibodies showed a dose-dependent decrease of villous surface annexin-V over a concentration range of 1.5 mu g/ml to 1.5 mg/ml. Annexin-V concentrations in conditioned media were significantly lower in the presence of antiphospholipid antibodies immunoglobulin G compared with normal immunoglobulin G (49.4 +/- 8.9 ng/gm wet weight vs 57.2 + 11.5 ng/gm, respectively, p < 0.05). In contrast, the mean level of annexin-V in placental villi incubated with antiphospholipid antibodies immunoglobulin G was greater than in villi incubated with normal immunoglobulin G, 1328 +/- 130 ng/gm wet weight versus 1183 +/- 165 ng/gm (p < 0.02). CONCLUSIONS: Patients with antiphospholipid antibodies and a history of previous pregnancy losses have a significant reduction in annexin-V immunostaining on placental villous surfaces, and antiphospholipid antibodies immunoglobulin G can directly decrease levels of villous surface annexin-V on cultured placental villi. Assays of annexin-V in the conditioned media and cell pellets of cultured placental villi suggest that the mechanism for antiphospholipid antibodies-mediated reduced annexin-V surface staining is an inhibition of annexin-V transport to the villous surface rather than displacement by antiphospholipid antibodies from the surface. This antiphospholipid antibodies-induced deficiency of placental surface annexin-V may contribute to the placental thrombosis observed in these patients.