Abstract
Human tissue factor pathway inhibitor-2 (TFPI-2) is a 32 kDa extracellular matrix-associated Kunitz-type serine proteinase inhibitor that regulates the plasmin and trypsin- mediated activation of zymogen matrix metalloproteinases and growth factors essential for tumor growth, invasion, and metastasis. We previously demonstrated that HT-1080 human fibrosarcoma cells do not express TFPI-2, but restoration of TFPI-2 expression in these cells by stable transfection with the human TFPI-2 cDNA markedly inhibited their growth and metastasis in-vivo. In this study, 1 μM concentrations of either recombinant human TFPI-2 or its mutated first Kunitz-type domain (R24K KD1), were offered to HT-1080 cells, incubated for 48 h, and the degree of apoptosis assessed by nuclear fragmentation, ethidium bromide/acridine orange (EB/AO) staining, fluorescence-activated cell sorting (FACS) and immunoblotting. Agarose gel electrophoresis of DNA from HT-1080 cells treated with either TFPI-2 or R24K KD1 for 48h indicated DNA fragmentation with a ladder pattern typically associated with apoptosis. EB/AO staining of TFPI-2- and R24K KD1- treated cells revealed that 40–70% of the cells were apoptotic in relation to vehicle-treated cells as judged by fluoresence microscopy. Co-administration of TFPI-2 with polymyxin B produced similar results, suggesting that apoptosis was not endotoxin-dependent. Consistent with our earlier studies showing its enhanced inhibitory activity relative to TFPI-2, R24K KD1 was able to induce apoptosis in 68% of HT-1080 cells after 48h of treatment compared to 39% for the parent TFPI-2 at an equivalent concentration. Moreover, HT-1080 cells treated with a KD1 preparation lacking the reactive site arginine residue (R24Q KD1) produced only an 18% apoptosis rate, thereby linking the observed apoptosis with serine proteinase inhibition. FACS analysis, using propidium iodide and annexin-V labeling, revealed similar apoptotic rates to that seen by EB/AO staining assays. In addition, immunoblotting experiments of vehicle and TFPI-2-treated cells indicated increased caspase-3 activation in TFPI-2-treated cells, thus providing evidence that apoptosis is caspase-mediated. We also observed up-regulation of the proapoptotic Bax protein by immunoblotting following treatment of HT-1080 cells with either TFPI-2 or R24K KD1. Taken together, our results demonstrate that treatment of HT-1080 cells with exogenous TFPI-2 or R24K KD1 activates caspase-mediated, proapoptotic signaling pathways and induces apoptosis. In addition, our data provides suggestive evidence that peritumor application of either TFPI-2 or R24K KD1 may facilitate tumor apoptosis in-vivo.
Paradoxical Pro-invasive Effect of the Serine Proteinase Inhibitor Tissue Factor Pathway Inhibitor-2 on Human Hepatocellular Carcinoma Cells