Correlation of in vivo and in vitro inhibition of thrombin by plasma inhibitors.

Rabbit antithrombin III and thrombin were purified to homogeneity to determine the in vivo relationship of these proteins in an autologous system. These proteins, radiolabeled with Na[125I], were injected into rabbits to determine the circulatory half-life. The mean half-life values were 125I-antithrombin III, 54.75 +/- 3.10 hr; 125I-thrombin, 7.25 +/- 1.49 hr; 125I-thrombin-antithrombin III, 7.25 +/- 1.09 hr; 125I[thrombin-antithrombin III], 11.13 +/- 0.88 hr; and Tos-Lys-CH2Cl-125I-thrombin, 27.75 +/- 3.18 hr. All the mean half-life values were statistically different from that of thrombin alone except for the two forms of thrombin-antithrombin III complex. Following injection of the radiolabeled proteins, plasma samples were obtained and gel-filtered to analyze the molecular weight distribution of the radiolabel. An identical elution position on gel filtration of 125I-antithrombin III with native antithrombin III was observed. The 125I-thrombin distributed into two peaks of radioactivity, with a molecular weight of 100,000 (79%) and a molecular weight greater than 200,000 (21%). The 100,000 dalton peak is consistent with a thrombin--antithrombin III complex, and the greater than 200,000 dalton peak is consistent with a thrombin-alpha 2-macroglobulin complex as confirmed by in vitro immunochemical studies. Thrombin inactivated with Tos-Lys-CH2Cl also showed two peaks of radioactivity on gel filtration, one peak which was excluded from the column and the other peak with an elution volume that was consistent with the position of native thrombin.