Abstract

When human, canine, or bovine factor VIII preparations are chromatographed on 4% agarose at ionic strength 0.2, the factor VIII activity elutes as a single peak in the void volume with slight tailing. Incubation of such preparations with dilute (0.01 U/ml) highly purified thrombin results in some activation of factor VIII. Chromatography of such incubation mixtures, under the same conditions as before, results in elution of two peaks of factor VIII activity one in the void volume and one much later with marked tailing. The void volume peak has most of the protein and some factor VIII activity. These void volume fractions also contain all the von Willebrand factor activity of thrombin-treated bovine preparations. Longer treatment with thrombin, or treatment with stronger thrombin, appears to shift much more of the procoagulant activity to the later eluting peak. Also, when the peak of factor VIII activity, found in the void volume after thrombin treatment, was again incubated with dilute thrombin, an increase in factor VIII activity occurred. Chromatography of this incubation mixture demonstrated only a small amount of activity in the void volume, while the bulk of the activity was present in the second peak. On the other hand, thrombin treatment of factor VIII activity from peak 2 caused a rapid decline of activity instead of a further increase. It is proposed that the residual factor VIII activity found in the void volume represents unreacted factor VIII, while the late eluting peak represents thrombin-activated material that is of smaller apparent size. The late eluting peak differs from the small active factor VIII fragment obtained by Ca2+ dissociation, as the latter can be activated by thrombin. A similar set of experiments was performed using ultracentifugation of bovine factor VIII preparations on sucrose density gradients. Results of these experiments agreed completely with those obtained with get chromatography. Preparations made from human hemophilic plasma, by the procudure employed in the purification of human factor VIII, were also incubated with thrombin and chromatographed. von Willebrand factor was again found only in the void volume fractions, but there was no factor VIII activity in any fractions eluted. In other control experiments, activated and unactivated factor VIII fractions did not clot fibrinogen and contained no assayable factor IX or X. The thrombin-modified factor VIII of small size was inactivated by both a naturally occurring human inhibitor to factor VIII and the gamma globulin fraction of a rabbit antisera produced against the calcium-dissociated small active factor VIII fragment.

Authors

H A Cooper, F F Reisner, M Hall, R H Wagner

×

Other pages: