Summary
Urokinase with an activity of 45,200 CTA nnits/mg was used in 0.4 mg amounts with Freund’s adjuvant to immunize four individual rabbits by two injections spaced 4 weeks apart. All antisera were titrated for inhibition of urokinase activation of plasminogen, for double diffusion in Agarose, and for immunoelectrophoresis. Against constant portions of 125I-labelled IgG from each antiserum, 131I-labelled urokinase was titrated by the quantitative precipitin technique. The results from these 4 methods of titration were compared and out of them the following conclusions were drawn :
1. There were 3 main immunogens in the urokinase preparations, one inactive as an activator, one with questionable activity, and one definitely and strongly active as a urokinase. The first two comprised less than 5% of the total antigen mixture; the active third component was more than 20% of the total. About 75% of the total protein was non-precipitable with antibody and was inactive as an activator. One inactive immunogen, I, was a fast β-protein in immunoelectrophoresis at pH 8.2, ionic strength 0.05: the questionably active immunogen, II, was a γ-protein; the active third component, III, was a slow β-protein whose specific activity was estimated to be 230,000 CTA units/mg protein.
2. Antibodies against I and II were in considerably higher titer than the ones to III and their equivalence points were reached at much higher concentrations of added urokinase preparation. A single precipitin line instead of 3 could be obtained by the use of very dilute antisera, but the line was due to the inactive I, not the slow active β-protein, III, nor the questionably active γ-protein, II.
3. All 4 antisera were 100% inhibitory of plasminogen activation activity at the equivalence point of III, but were not inhibitory at the equivalence points of I and II.
4. Urokinase when added to normal human serum could be visualized as easily in double diffusion as when tested in buffer alone.
