Design of Novel and Selective Inhibitors of Urokinase-type Plasminogen Activator with Improved Pharmacokinetic Properties for Use as Antimetastatic Agents

Targeting of urokinase plasminogen activator receptor in human pancreatic carcinoma cells inhibits c-Met- and insulin-like growth factor-I receptor-mediated migration and invasion and orthotopic tumor growth in mice

Pancreatic carcinomas express high levels of urokinase-type plasminogen activator (uPA) and its receptor (uPAR), both of which mediate cell migration and invasion. We investigated the hypotheses that (a) insulin-like growth factor-I (IGF-I)- and hepatocyte growth factor (HGF)-mediated migration and invasion of human pancreatic carcinoma cells require uPA and uPAR function and (b) inhibition of uPAR inhibits tumor growth, retroperitoneal invasion, and hepatic metastasis of human pancreatic carcinomas in mice. Using transwell assays, we investigated the effect of IGF-I and HGF on L3.6pl migration and invasion. We measured the induction of uPA and uPAR following treatment of cells with IGF-I and HGF using immunoprecipitation and Western blot analysis. The importance of uPA and uPAR on L3.6pl cell migration and invasion was studied by inhibiting their activities with amiloride and antibodies before cytokine treatment. In an orthotopic mouse model of human pancreatic carcinoma, we evaluated the effect of anti-uPAR monoclonal antibodies with and without gemcitabine on primary tumor growth, retroperitoneal invasion, and hepatic metastasis. IGF-I and HGF mediated cell migration and invasion in L3.6pl cells. In addition, IGF-I and HGF induced uPA and uPAR expression in L3.6pl cells. In vitro, blockade of uPA and uPAR activity inhibited IGF-I- and HGF-mediated cell migration and invasion. Treatment of mice with anti-uPAR monoclonal antibody significantly decreased pancreatic tumor growth and hepatic metastasis and completely inhibited retroperitoneal invasion. Our study shows the importance of the uPA/uPAR system in pancreatic carcinoma cell migration and invasion. These findings suggest that uPAR is a potential target for therapy in patients with pancreatic cancer.

A Urokinase-type Plasminogen Activator-inhibiting Cyclic Peptide with an Unusual P2 Residue and an Extended Protease Binding Surface Demonstrates New Modalities for Enzyme Inhibition

To find new principles for inhibiting serine proteases, we screened phage-displayed random peptide repertoires with urokinase-type plasminogen activator (uPA) as the target. The most frequent of the isolated phage clones contained the disulfide bridge-constrained sequence CSWRGLENHRMC, which we designated upain-1. When expressed recombinantly with a protein fusion partner, upain-1 inhibited the enzymatic activity of uPA competitively with a temperature and pH-dependent K(i), which at 25 degrees C and pH 7.4 was approximately 500 nm. At the same conditions, the equilibrium dissociation constant K(D), monitored by displacement of p-aminobenzamidine from the specificity pocket of uPA, was approximately 400 nm. By an inhibitory screen against other serine proteases, including trypsin, upain-1 was found to be highly selective for uPA. The cyclical structure of upain-1 was indispensable for uPA binding. Alanine-scanning mutagenesis identified Arg(4) of upain-1 as the P(1) residue and indicated an extended binding interaction including the specificity pocket and the 37-, 60-, and 97-loops of uPA and the P(1), P(2), P(3)', P(4)', and the P(5)' residues of upain-1. Substitution with alanine of the P(2) residue, Trp(3), converted upain-1 into a distinct, although poor, uPA substrate. Upain-1 represents a new type of uPA inhibitor that achieves selectivity by targeting uPA-specific surface loops. Most likely, the inhibitory activity depends on its cyclical structure and the unusual P(2) residue preventing the scissile bond from assuming a tetrahedral geometry and thus from undergoing hydrolysis. Peptide-derived inhibitors such as upain-1 may provide novel mechanistic information about enzyme-inhibitor interactions and alternative methodologies for designing effective protease inhibitors.

Reduction of Experimental Human Fibrosarcoma Lung Metastasis in Mice by Adenovirus-Mediated Cystatin C Overexpression in the Host

Tumor cell invasion and metastasis are associated with degradation of components of the extracellular matrix by different proteinases. Among those, papain-like cysteine proteases, such as cathepsin B, seem to play an important role, as they are associated with poor clinical outcome in different cancers. In this study, we tested whether cystatin C, a natural extracellular inhibitor of papain-like cysteine proteases, can inhibit metastasis when overexpressed at the tumor-host interface. Local overexpression of cystatin C in liver and lungs of CD1 nu/nu mice was achieved by gene transfer with a novel adenoviral construct, which also led to the presence of 60 ng/mL of cystatin C in the serum. Three days after gene transfer, these mice were challenged by i.v. inoculation of lacZ-tagged human fibrosarcoma cells (HT1080lacZ-K15), leading to the formation of experimental lung and liver metastases. In this model, formation of experimental metastatic foci correlated with expression of cathepsin B in lungs, whereas there was no correlation with metastasis to the liver. In mice overexpressing cystatin C, the number of lung metastases was significantly reduced by 92%, as compared with mice receiving control adenovirus. The efficacy of extravasation of HT1080lacZ-K15 cells into the liver was not affected, indicating the independence of this process from the activity of cysteine-cathepsins. The present report is the first evidence of successful reduction of metastasis by inhibition of cysteine-cathepsins by cystatin C overexpression in the host microenvironment. Furthermore, organ-specific protease expression during tumor-host cell interactions could affect the success of antiproteolytic intervention against metastasis.

RNA interference-directed knockdown of urokinase plasminogen activator and urokinase plasminogen activator receptor inhibits prostate cancer cell invasion, survival, and tumorigenicity in vivo

The invasive ability of tumor cells plays a key role in prostate cancer metastasis and is a major cause of treatment failure. Urokinase plasminogen activator-(uPA) and its receptor (uPAR)-mediated signaling have been implicated in tumor cell invasion, survival, and metastasis in a variety of cancers. This study was undertaken to investigate the biological roles of uPA and uPAR in prostate cancer cell invasion and survival, and the potential of uPA and uPAR as targets for prostate cancer therapy. uPA and uPAR expression correlates with the metastatic potential of prostate cancer cells. Thus, therapies designed to inhibit uPA and uPAR expression would be beneficial. LNCaP, DU145, and PC3 are prostate cancer cell lines with low, moderate, and high metastatic potential, respectively, as demonstrated by their capacity to invade the extracellular matrix. In this study we utilized small hairpin RNAs (shRNAs), also referred to as small interfering RNAs, to target human uPA and uPAR. These small interfering RNA constructs significantly inhibited uPA and uPAR expression at both the mRNA and protein levels in the highly metastatic prostate cancer cell line PC3. Our data demonstrated that uPA-uPAR knockdown in PC3 cells resulted in a dramatic reduction of tumor cell invasion as indicated by a Matrigel invasion assay. Furthermore, simultaneous silencing of the genes for uPA and uPAR using a single plasmid construct expressing shRNAs for both uPA and uPAR significantly reduced cell viability and ultimately resulted in the induction of apoptotic cell death. RNA interference for uPA and uPAR also abrogated uPA-uPAR signaling to downstream target molecules such as ERK1/2 and Stat 3. In addition, our results demonstrated that intratumoral injection with the plasmid construct expressing shRNAs for uPA and uPAR almost completely inhibited established tumor growth and survival in an orthotopic mouse prostate cancer model. These findings uncovered evidence of a complex signaling network operating downstream of uPA-uPAR that actively advances tumor cell invasion, proliferation, and survival of prostate cancer cells. Thus, RNA interference-directed targeting of uPA and uPAR is a convenient and novel tool for studying the biological role of the uPA-uPAR system and raises the potential of its application for prostate cancer therapy.

Geldanamycins exquisitely inhibit HGF/SF-mediated tumor cell invasion

Induction of the urokinase-type plasminogen activator (uPA) by hepatocyte growth factor/scatter factor (HGF/SF) plays an important role in tumor cell invasion and metastasis that is mediated through the Met receptor tyrosine kinase. Geldanamycins (GA) are antitumor drugs that bind and inhibit HSP90 chaperone activity at nanomolar concentrations (nM-GAi) by preventing proper folding and functioning of certain oncoproteins. Previously, we have shown that a subset of GA derivatives exhibit exquisite potency, inhibiting HGF/SF-induced uPA-plasmin activation at femtomolar concentrations (fM-GAi) in canine MDCK cells. Here, we report that (1) inhibition of HGF/SF-induced uPA activity by fM-GAi is not uncommon, in that several human tumor glioblastoma cell lines (DBTRG, U373 and SNB19), as well as SK-LMS-1 human leiomyosarcoma cells are also sensitive to fM-GAi; (2) fM-GAi drugs only display inhibitory activity against HGF/SF-induced uPA activity (rather than basal activity), and only when the observed magnitude of uPA activity induction by HGF/SF is at least 1.5 times basal uPA activity; and (3) not only do fM-GAi derivatives strongly inhibit uPA activity but they also block MDCK cell scattering and in vitro invasion of human glioblastoma cells at similarly low drug concentrations. These effects of fM-GAi drugs on the Met-activated signaling pathway occur at concentrations well below those required to measurably affect Met expression or cell proliferation. We also examined the effect of Radicicol (RA), a drug with higher affinity than GA for HSP90. RA displays uPA activity inhibition at nanomolar levels, but not at lower concentrations, indicating that HSP90 is not likely the fM-GAi molecular target. Thus, we show that certain GA drugs (fM-GAi) in an HGF/SF-dependent manner block uPA-plasmin activation in tumor cells at femtomolar levels. This inhibition can also be observed in scattering and in vitro invasion assays. Our findings also provide strong circumstantial evidence for a novel non-HSP90 molecular target that is involved in HGF/SF-mediated tumor cell invasion.