Monoclonal Antibodies Towards the Fast Inhibitor of Tissue Plasminogen Activator from Human Plasma and Serum

Purification of high and low molecular weight plasminogen activator inhibitor 1 from fibrosarcoma cell-line HT 1080 conditioned medium

Functionally active (high-Mr) and inactive (low-Mr) plasminogen activator inhibitor 1 (PAI) have been purified from fibrosarcoma cell-line HT 1080 conditioned medium, containing 1% fetal calf serum. The two forms were first purified by affinity chromatography on heparin-Sepharose and then separated from each other by gel filtration on Sephadex G-150. The final purification was achieved by affinity chromatography on insolubilized monoclonal antibodies towards human PAI. Alternatively, the low-Mr form was purified by chromatography on carboxymethyl-cellulose. Low-Mr PAI purified in this way, could be almost fully reactivated by treatment with guanidinium chloride. Sodium dodecyl sulphate-polyacrylamide gel electrophoresis followed by immunoblotting revealed that the low-Mr form contained nothing but PAI at an Mr of about 50,000. In addition to PAI, the high-Mr form contained a component, which was not antigenically related to PAI. This compound had a molecular weight of about 75,000 and its NH2-terminal amino acid sequence corresponded to that of human vitronectin. We conclude that the high-Mr form of PAI constitutes a complex between 50,000 Mr PAI and vitronectin from fetal calf serum.

Plasminogen activator inhibitor 1 (PAI) is bound to vitronectin in plasma

Functionally active plasminogen activator inhibitor 1 (PAI) is bound to a discrete binding protein in plasma [(1988) Thromb. Haemost. 59, 392-395]. The binding protein has now been partially purified using conventional chromatographic techniques. After addition of active PAI its complex with the binding protein was purified by chromatography on insolubilized monoclonal antibodies towards PAI. Dodecylsulphate (polyacrylamide gel electrophoresis revealed two main compounds with molecular masses of 50 and 75 kDa respectively. NH2-terminal amino acid sequence analysis and immunoblotting analysis suggested that the two compounds were PAI (50 kDa) and vitronectin (75 kDa). We conclude that the PAI-binding protein is identical to vitronectin.

Angiogenic properties of Kaposi's sarcoma-derived cells after long-term culture in vitro

Cells derived from lung biopsies and pleural effusions from AIDS patients with Kaposi's sarcoma (KS) of the lungs were established in long-term culture with the aid of conditioned medium from HTLV-II-transformed T cells (HTLV-II CM). These AIDS-KS cells were similar to the so-called spindle cells in KS lesions and had some of their features. They produced factors that supported their own growth (autocrine) and the growth of other cells (paracrine), including umbilical vein endothelium and fibroblasts. That the AIDS-KS cells also expressed potent angiogenic activity was demonstrated by the chorioallantoic membrane assay and by subcutaneous inoculation of AIDS-KS cells into nude mice, which resulted in the development of angiogenic lesions composed of mouse cells and showing histological features similar to those of human KS lesions. These data suggest that AIDS-associated KS and possibly other types of KS may be initiated by signals that induce the growth of particular cells (spindle cells of lymphatic or vascular origin) and the expression of autocrine and paracrine activities.

Comparison of plasminogen activator inhibitor activity and antigen in plasma samples

A double antibody radioimmunoassay for determination of plasminogen activator inhibitor (PA-inhibitor) in plasma samples has been developed. The reliability of the method as assessed by determining the specificity, the accuracy, the detectability and the variability seems sufficient for use in clinical practice. The correlation between PA-inhibitor antigen as measured with this method and PA-inhibitor activity as measured with a spectrophotometric assay in 111 patients with thrombotic diseases was very good (r = 0.89). As calculated from the regression line or from the mean activity and mean antigen values a specific activity of about 800,000-900,000 arb U/mg was obtained for PA-inhibitor in these samples. In 15 healthy individuals a similar figure was obtained. The results suggest that PA-inhibitor in most plasma samples is fully active or close to fully active. PA-inhibitor activity and PA-inhibitor antigen have also been measured after venous occlusion. The data suggest that small amounts of PA-inhibitor is released on venous occlusion, but at the same time an inactivation takes place, most likely due to the formation of enzymatically inactive complex with simultaneously released t-PA.