Tumor-associated proteolysis and the urokinase-receptor (u-PAR)
For the accomplishment of almost each step of the metastatic cascade, tumor cells essentially drive the degradation of extracellular matrix components, e.g., collagen type IV, laminin, or vitronectin. To stimulate suchlike activities, tumor cells, and also surrounding stromal cells which interact with them, employ several “tumor-associated proteases”, which are secreted by tumor cells or surrounding stromal cells and can be focused, e.g. at the tumor cell surface, by expression of specific protease receptors. According to the catalytically active site of these proteases they are classified into serine, aspartic, cysteine, threonine, and metalloproteinases. The identity of these proteolytic enzymes is not different from physiologic enzymes which are involved in several tissue remodelling processes such as wound healing, fibrinolysis, inflammation, embryogenesis and angiogenesis. However, the quantity rather than the quality of their expression or activity contributes to the invasive phenotype of malignant cells.
One of the best characterized “tumor-associated proteases” is the urokinase-type plasminogen activator (u-PA), a serine protease of 55 kDa. It converts plasminogen to the active enzyme, plasmin, thus promoting extracellular matrix degradation including fibrin, fibronectin, proteoglycans, laminin and collagen IV (important for a tumor cell’s invasion and intravasation through basement membranes). Its proteolytic activation can occur either in the extracellular space or bound to the specific urokinase-receptor (u-PAR), however, receptor-bound u-PA activates plasminogen much more efficiently, this being reflected by a 40-fold decrease in Km for its substrate. The 55-60 kDa heavily glycosylated, disulfide-linked cell surface receptor (u-PAR) consists of 3 similar repeats approximately 90 residues each, the last of which is anchored to the cell membrane via glycosyl-phosphatidylinositol, this enabling a high intramembrane mobility.
Efficient proteolytic system
Receptor-bound u-PA is inactivated by PAI-1 (-2), the u-PAR/u-PA/PAI complex is internalized into the cell together with α-2-macroglobulin receptor and its ligand, the free u-PAR is recycled to the cell surface, and binding and activation of a second u-PA-molecule can occur. Moreover, the u-PAR is co-localized with integrins, acts as a co-receptor for vitronectin, can be found at cellular focal contacts, lamellipodia or in caveolae, and is able to induce the phosphorylation of focal adhesion kinase (FAK), cytoskeletal proteins and Src-family members. A switch between considerably different signalling cascades brought about by u-PAR-associated molecular interactions can promote such opposing phenomena such as proliferation and dormancy for tumor cells. Thus, the u-PAR/u-PAR system is considered to be one of the most important, dynamic and ubiquitous tumor-associated protease systems, this being reflected by the fact that the u-PAR is overexpressed in the majority of solid carcinoma entities (e.g., breast, gastrointestinal, prostate lung cancers and others), this being associated with a poor clinical prognosis in many instances (for further reading please consider our reviews).
u-PAR system as a novel prognostic factor
Since 1993, our group has been actively contributing to hypothesis generation on how the u-PAR works molecularly, how it is being regulated in malignant cells, and how this could contribute to a better understanding, diagnosis, therapy and classification of tumor diseases. Towards this end, we established the complete u-PAR system as a novel prognostic factor in gastric cancer and demonstrated that the u-PAR is one of the critical molecular characteristics of the metastatically relevant phenotype in minimal residual tumor cells in solid cancer. Moreover, we identified the major transcriptional regulators of u-PAR overexpression in cancer cells and, in a translational approach, demonstrated that the transactivation of u-PAR gene expression via different promoter elements of the u-PAR gene occurs at different levels of tumor specificity, a notion that we currently explore in terms of diagnostic and therapeutic opportunities in a patented approach (Patent-No eu PCT/EPO3/01671. Tissue Specific Exression). Furthermore, we observed that, with specific transcriptional regulators of the u-PAR and further parameters, we were able to modify existing clinical staging concepts toward the identification of completely novel high-risk groups of cancer patients (Maurer). Finally, in recent years we extended translational research projects on u-PAR and identified the EGFR-targeted antibody Cetuximab as an inhibitor of metastasis via blocking u-PAR gene expression, in parallel suggesting the u-PAR as a novel biomarker for Cetuximab sensitivity. Also, we recently implicated the anti-malaria agent Artesunate as a novel metastasis inhibitor in cancer, acting via the u-PAR ligand urokinase (u-PA) and specific matrix-metalloproteinases (MMPs). Furthermore, we established metastasis suppression as a function of the novel tumor suppressor Pdcd4 and showed that this is largely achieved by inhibiting u-PAR gene expression. This was supported by the clinical observation that the loss of Pdcd4, in parallel to an increase of u-PAR gene expression, is an independent prognostic marker in colorectal cancer, Pdcd4 being essentially downregulated in cancer cells by microRNA-21.