Cell density-dependent downregulation of urokinase-type plasminogen activator in normal but not in transformed human epidermal keratinocytes

TGFBI gene transcript is transforming growth factor-beta1-responsive and cell density-dependent in a human corneal epithelial cell line

PURPOSE

To determine the roles of transforming growth factor-beta( 1) (TGF-beta(1)), epidermal growth factor (EGF), and cell density on the regulation of the TGFBI gene transcript in a human corneal epithelial cell line (HCE-T).

METHODS

HCE-T cells (40-50% confluent), which possess the biochemical and morphological phenotype of human corneal epithelial cells, were treated with either TGF-beta(1) (20 ng/ml) or EGF (20 ng/ml) for 0, 24, 48, and 72 h. Total RNA was isolated at each time point from the treated cells and the placebo-treated controls. The TGFBI mRNA transcript level was quantitated using Northern blot analysis.

RESULTS

TGF-beta(1) upregulates TGFBI gene transcript in this human corneal epithelial cell line, reaching a peak of 2.5-fold of upregulation at 48 h after TGF-beta(1) treatment and a 1.5-fold at 72 h. In contrast, EGF treatment showed no effect on the TGFBI mRNA level. In addition, the TGFBI gene transcript appears to be cell density-dependent because the transcript level shows a trend of reduction as the cell density increases from 40-50% to full confluency.

CONCLUSION

The TGFBI gene transcript in this human corneal epithelial cell line is upregulated by growth factor TGF-beta( 1), but is not affected by growth factor EGF. Furthermore, cell density appears to be an important regulatory mechanism in controlling the level of TGFBI gene expression in corneal epithelial cells.

Expression pattern of the plasminogen activator-plasmin system in human cholesteatoma

The plasminogen activator-plasmin system plays a pivotal role in the delicately regulated process of extracellular matrix remodeling. Recent studies have shown that an imbalance of proteolytic enzymes over specific inhibitors in this system may lead to an aggressive, expanding, and infiltrating cellular phenotype. As cholesteatoma resembles a tumor in many ways, we investigated the pattern of expression for members of the plasminogen activator-plasmin system in 12 human cholesteatomas, using immunohistochemistry. As controls, 3 tympanic membranes and 4 ear canal skin specimens were used. In contrast to the tympanic membranes, all cholesteatoma specimens showed a strong expression of plasminogen at the basal epithelial cell layers. In ear canal skin, only the basal surface of the most basal epithelia stained discretely positive. The urokinase-type plasminogen activator (uPA) could be detected in the basal stratum of the cholesteatoma matrix and in the surrounding granulation tissue, while tissue-type plasminogen activator (tPA) was not detectable at all. Plasminogen activator inhibitor-1 (PAI-1) was expressed in both the granulation tissue and the granular cell layer of the matrix, but not in the basal epithelial cells; PAI-2 showed a pericellular expression pattern in the subbasal and granular cell layers. Neither uPA, tPA, nor the PAIs could be detected in tympanic membrane controls; ear canal skin showed the same staining pattern as cholesteatoma only for PAI-2. Our data demonstrate that there is a clear imbalance in favor of proteolytic activity in the basal epithelial layers of the cholesteatoma matrix, which might at least partly account for the aggressive behavior of this tumorlike lesion. Further, the pattern of expression resembles the pattern described for several epithelial malignancies.

Fibrinolytic activity of earthworms extract (G-90) on lysis of fibrin clots originated from the venous blood of patients with malignant tumors

u-PA is secreted by the most malignant tumors. As a response to u-PA synthesis surrounding cells synthetize inhibitors of plasminogen activators for tissue protection. Plasminogen activators were found also in earthworm tissue. From the tissue homogenate of earthworm Eisenia foetida the glycolipoprotein mixture named G-90 was isolated. It contains two serine proteases (P I, P II) with fibrinolytic and anticoagulative activities. The fibrinolytic activity of G-90, P I and P II was tested in an in vitro euglobulinic test applied to fibrin clot from blood plasma of patients suffered from malignant tumors. G-90 and above-mentioned proteases applied in this study showed euglobulinic time proportionally with the concentrations of added substances. The influence of G-90 on the fibrinolysis rate does not depend only on its concentration, but depends too on histological type of tissue (organ) where the malignant tumors are located. Enzyme P I and P II do not show this activity.