Detergent and antigen fragility affect the ELISA for measurement of anti-prothrombin autoantibodies

Some investigators have reported that anti-prothrombin autoantibodies (aPT) in lupus anticoagulant positive sera were detectable by ELISA. Discrepancies in aPT ELISA were observed by some investigators. To clarify this situation, we tested the binding of aPT positive sera to purified prothrombin under various conditions.

We performed aPT ELISA under different conditions. The variables we tested were: ELISA plate (untreated or gamma irradiated polystyrene plates), buffer (phosphate buffered saline or Tris buffered saline), detergent (presence or absence of Tween-20), and antigen condition (intact or fragmented prothrombin).

Anti-PT from patients with lupus or antiphospholipid syndrome were similarly bound to prothrombin with both buffers. Addition of Tween-20 to the buffer increased reactivity in the irradiated plate assay, but decreased reactivity in the untreated plate assay. Reactivities in 90% of lupus sera were decreased by the use of fragmented prothrombin. In contrast, the reactivity of serum from a healthy subject was remarkably increased by antigen fragmentation.

The discrepant ELISA results in measurement of aPT in the various reports may have been due to the use of detergent in the buffer and condition of the prothrombin used as antigen. In our experiments the best ELISA condition for measurement of aPT was achieved using buffer with Tween-20 detergent, with prothrombin directly coated onto irradiated polystyrene plates.