The plasminogen activator activity of human breast tumors was determined quantitatively in extracts of 33 primary tumors, 11 lymph node metastases, and 11 normal tissues. Activities varied more widely in the neoplastic tissues than in the normal ones. The mean activator content of the tumors (primary and metastasis) was significantly higher (p < 0.01) than that of normal tissues: 4.4-fold on a g wet tissue basis and 2.1-fold when the activities were referred to the protein content of the extracts. Primary and metastatic tumors were not significantly different from one another in their plasminogen activator content. In the six cases where tumors and normal tissue pairs from the same patients could be compared, the mean of the individual tumor:normal ratios was 10.0 (wet weight) or 3.6 (extract protein). In addition to the activator content, the molecular type of the enzymes was also determined, using goat antibody formed against human urinary urokinase as an inhibitor of activator activity. On the average, 80% of the activity of the tumor extracts could be inhibited by the antibody; extracts of normal breast tissue contained only 62% urokinase-like activator. The activators were also characterized by sodium dodecyl sulfate gel electrophoresis using two different methods for the zymographic visualization of activator bands. The fraction of urokinase-like enzyme determined by the fibrinogen copolymerization method agreed well with the results obtained by immunoinhibition. No correlation was found between activator activity or activator composition, on the one hand, and the degree of differentiation of the tumors, the fraction of positive lymph nodes found, or the amount of estrogen-binding protein (or their molecular size) in the cytosol of the tumors, on the other. This apparent lack of correlation may reflect variations in local physiological conditions which necessitate the presence of widely differing amounts of activator in order to ensure a critical amount of the product of the reaction catalyzed by this enzyme, i.e., plasmin.
This research was supported by Grant BC-235 from the American Cancer Society.