Statistics from Altmetric.com
Antiphospholipid syndrome (APS) is characterised by arterial and/or venous thrombosis and recurrent fetal loss in association with the presence of antiphospholipid antibodies (aPL). In addition to β2-glycoprotein I, prothrombin (PT) is an important autoantigen recognised by aPL. PT is a coagulation proenzyme abundantly present in blood (70–100 μg/ml), and binds to negatively charged phospholipids such as phosphatidylserine. PT comprises two major domains, fragment-1 (F-1) and prethrombin-1 (Pre-1). We previously reported that antiprothrombin antibody (aPT) is a mixture of antibodies against both F-1 and Pre-1, and that there are significant clinical differences between anti-F-1 and anti-Pre-1.1,2
aPL, including aPT, are not mere serological markers of the disease, but are important players in the pathogenesis of APS. Therefore, antigen-specific immunosuppressive treatments, if developed, will help to prevent thrombosis in APS. We aimed to locate the T cell epitopes of PT on either F-1 or Pre-1, as an initial approach to understanding the mechanism of aPT formation, and to search for possible PT-specific immunosuppressive treatments.
Samples from 15 patients with systemic lupus erythematosus (SLE) were positive for aPT. Fourteen patients were taking prednisolone at a mean of 10.8 mg/day (range 5.0–22.5). For T cell proliferation assays, samples from nine healthy volunteers served as controls.
Prothrombin was cleaved as described3 with some modifications. Briefly, 1 ml (1.46 mg/ml) of purified human PT (Haematologic Technologies, Essex Junction, VT) was incubated with 3 U bovine thrombin (Itoham Foods, Osaka, Japan) for three hours at 37°C. Cleaved products were run on sodium dodecyl sulphate-polyacrylamide gel electrophoresis, cut out, and purified using a model 422 Electro-Eluter (Bio-Rad, Hercules, CA).
Anti-PT, anti-Pre-1, and anti-F-1 enzyme linked immunosorbent assays (ELISAs) were performed as described previously.3
For T cell proliferation assays, peripheral blood mononuclear cells (2.0×105 cells/well) were cultured in RPMI 1640 containing 10% fetal bovine serum and 1% l-glutamine-penicillin-streptomycin (Sigma), in the presence of PT, F-1 , Pre-1 (1 μmol/l) or controls, for three days under standard conditions. Antigen induced T cell proliferation was assayed by a cell proliferation ELISA, BrdU (Roche Diagnostics, Mannheim, Germany). The stimulation index (SI) was calculated as the mean optical density (OD) of wells containing the antigen/mean OD of the wells without antigen. An SI ⩾2 was considered positive.
Samples from only two patients showed positive responses to PT. Because a large amount of PT is present in sera, T cells may be rendered tolerant in normal conditions, or some modifications may be necessary for PT to be incorporated by antigen presenting cells. On the other hand, three samples showed positive responses to F-1, and four to Pre-1, indicating the presence of at least one T cell epitope in those domains (fig 1). Among the six patients with positive T cell responses, the putative localisation of T and B cell epitopes was on the same domain in four patients, but was on different domains in two patients, suggesting that the location of T cell epitopes and the production of pathogenic aPT are not necessarily related (table 1).
HLA-DRB1*1501, which is suggested to be related to SLE,4 was present in 5/15 (33%) patients. No significant skewing of the HLA-DRB1 alleles, or relationships with T cell epitopes were seen.
Samples from three controls showed positive responses to PT, F-1, or Pre-1. Autoantigen-specific T cells do exist in the sera of healthy subjects,5 and the presence of autoantigen reactive T cells by itself is not sufficient to cause autoimmune diseases, other factors are necessary.
Our study shows that PT-specific T cells present in aPT positive patients do not recognise a single common T cell epitope, and that T cell epitopes are different among individual patients.