Directed plasminogen activation at the surface of normal and malignant cells

Production and activation of matrix metalloprotease-9 (MMP-9) by HL-60 promyelocytic leukemia cells

Human promyelocytic HL-60 cells have been used as a model of acute leukemia to investigate the expression and the regulation of matrix metalloproteases (MMPs), known to contribute to the degradation of extracellular matrix components. As shown by gelatin zymography, HL-60 cells constitutively released significant amounts of proMMP-9 (92 kDa) and moderate amounts of proMMP-2 (72 kDa). Furthermore, casein zymography confirmed the presence of serine proteases in the form of pro-urokinase. Activation of proMMP-9 was dependent on the plasminogen activator/plasmin (PA/plasmin) system and was inhibited by aprotinin. MMP-9 was only detected in cellular extracts or conditioned media incubated with HL-60 cells, indicating that cells are essential to the activation process. Addition of plasminogen increased by 3-fold the basal invasive rate of these cells across a matrigel layer (2.1% versus 0.7% in control cells after 4 h of incubation). Taken together, these results indicate that HL-60 cells exhibit an autocrine activation mechanism of proMMP-9 via the PA/plasmin system and that activation of proMMP-9 increases their invasive potential.

Binding of high molecular weight kininogen to human endothelial cells is mediated via a site within domains 2 and 3 of the urokinase receptor

The urokinase receptor (uPAR) binds urokinase-type plasminogen activator (u-PA) through specific interactions with uPAR domain 1, and vitronectin through interactions with a site within uPAR domains 2 and 3. These interactions promote the expression of cell surface plasminogen activator activity and cellular adhesion to vitronectin, respectively. High molecular weight kininogen (HK) also stimulates the expression of cell surface plasminogen activator activity through its ability to serve as an acquired receptor for prekallikrein, which, after its activation, may directly activate prourokinase. Here, we report that binding of the cleaved form of HK (HKa) to human umbilical vein endothelial cells (HUVEC) is mediated through zinc-dependent interactions with uPAR. These occur through a site within uPAR domains 2 and 3, since the binding of 125I-HKa to HUVEC is inhibited by vitronectin, anti-uPAR domain 2 and 3 antibodies and soluble, recombinant uPAR (suPAR), but not by antibody 7E3, which recognizes the beta chain of the endothelial cell vitronectin receptor (integrin alphavbeta3), or fibrinogen, another alphavbeta3 ligand. We also demonstrate the formation of a zinc-dependent complex between suPAR and HKa. Interactions of HKa with endothelial cell uPAR may underlie its ability to promote kallikrein-dependent cell surface plasmin generation, and also explain, in part, its antiadhesive properties.

Surface-bound plasmin induces selective proteolysis of insulin-like-growth-factor (IGF)-binding protein-4 (IGFBP-4) and promotes autocrine IGF-II bio-availability in human colon-carcinoma cells

Limited proteolysis of insulin-like-growth-factor (IGF)-binding proteins (IGFBPs) represents a key process to modulate IGF bio-availability at the cellular level. In human colon carcinomas, urokinase-type plasminogen activator (u-PA) produced by stroma cells can bind to cancer-cell-associated u-PA receptor (u-PAR), and then catalyze the conversion of plasminogen (Pg) into plasmin (Pm). We therefore investigated the interplay between the IGF and Pm systems in the HT29-D4 human colon-carcinoma-cell model. HT29-D4 cells secreted IGF-II totally complexed to IGFBP-2, IGFBP-4 and IGFBP-6. Approximately 15% of IGFBP-4 was associated with the extracellular matrix. HT29-D4 cells produced neither u-PA- nor IGFBP-specific proteases. However, activation of Pm at the HT29-D4 cell surface obtained by the sequential addition of exogenous u-PA and Pg to mimic the stromal complementation induced selective proteolysis targeted to IGFBP-4 only (>95%). IGFBP-2 and IGFBP-6, though sensitive to proteolysis by soluble Pm, were not altered by cell-bound Pm. IGFBP-4 proteolysis yielded 18- and 14-kDa immunoreactive fragments which were not detectable by Western ligand blotting, indicating that they bound IGF-II with poor affinity. Release of IGF-II from IGF-II-IGFBP complexes after IGFBP-4 proteolysis by cell-bound Pm was indicated by the observation that approximately 20% of the 125I-IGF-II initially associated with endogenous IGFBP in reconstituted complexes was transferred to HT29-D4 cell-surface IGF-I receptors. These results suggest that IGFBP-4 proteolysis by cell-bound Pm can promote autocrine/paracrine IGF-II bio-availability in colon-cancer cells. This may have important consequences on the behavior of cancer cells at the interface between stroma and malignant cells in carcinomas of the colon in vivo.

Plasminogen activation system in human milk

BACKGROUND

Plasmin is the major endogenous protease present in milk. The level of plasmin activity is controlled by the availability of the precursor plasminogen and by the levels of plasminogen activators and inhibitors. Recently, a differential distribution of tissue-type plasminogen activator (t-PA) and urokinase-type plasminogen activator (u-PA) has been demonstrated in bovine milk. To assess whether this distribution pattern is a general feature, the occurrence of components of the plasminogen activation system in different fractions of human milk was investigated.

METHODS

Milk samples were separated into the following fractions; milk fat, skim milk, and milk cells by centrifugation. The different fractions were detected for the presence of plasminogen and plasminogen activators by immunoblotting and zymography. The distribution of t-PA and u-PA was investigated by ligand binding analysis. t-PA-catalyzed plasminogen activation was examined by a coupled chromogenic assay.

RESULTS

A differential distribution of plasminogen, t-PA, and u-PA was found. Casein micelles were found to exhibit t-PA and plasminogen binding activity, whereas the u-PA receptor was identified as the u-PA binding component in the cell fraction. Furthermore, human casein enhanced t-PA-catalyzed plasminogen activation, comparable to the enhancing effect obtained with fibrinogen fragments.

CONCLUSION

The finding of a differential distribution of u-PA and t-PA in milk suggests that the two activators may have different physiological functions, which involve protection against invading microorganisms and maintenance of patency and fluidity in the ducts of mammary gland, respectively.

ELISA for complexes between urokinase-type plasminogen activator and its receptor in lung cancer tissue extracts

A sandwich-type ELISA has been developed for the assessment of complexes between urokinase-type plasminogen activator (uPA) and its receptor (uPAR) in extracts of squamous cell lung carcinomas. The assay is based on a combination of rabbit polyclonal anti-uPA antibodies and a biotinylated mouse anti-uPAR monoclonal antibody (MAb). The detection limit of the assay is approximately 0.5 fmol/ml. A linear dose-response is obtained with up to 40 fmol/ml of uPA:uPAR complexes, while uPA and uPAR separately do not cause any response in the ELISA. A buffer which has been used previously for optimal extraction of uPAR yields the highest amounts of uPA:uPAR complexes. Absorption of tumor extracts with anti-uPA or anti-uPAR MAbs results in a complete disappearance of the ELISA signal, demonstrating the specificity of the ELISA. The recovery of chemically cross-linked uPA:uPAR complexes added to tumor extracts varies between 80% and 105%. The intra- and inter-assay variation coefficients are 5.3% and 9.8%, respectively. Furthermore, a peptide antagonist for uPAR was employed to evaluate de novo uPA:uPAR complex formation during tumor tissue extraction and the immunoassay procedure. Our results strongly indicate that de novo complex formation is a major factor to consider and that complexes analyzed in the presence of this antagonist represent original uPA:uPAR complexes present prior to tumor tissue processing. The present ELISA appears suitable for studying the potential prognostic impact of uPA:uPAR complexes in lung tumor tissue as well as other types of cancer.

Defective cell migration in an ovarian cancer cell line is associated with impaired urokinase-induced tyrosine phosphorylation

The urokinase receptor (u-PAR), a protein anchored to cell membrane by a glycosyl phosphatidylinositol, plays a central role in cancer cell invasion and metastasis by binding urokinase plasminogen activator (u-PA), thereby facilitating plasminogen activation. Plasmin can promote cell migration either directly or by activating metalloproteinases that degrade some of the components of the extra cellular matrix. However, the IGR-OV1-Adria cell line contains the u-PAR but does not migrate even in the presence of exogenous u-PA, although the parental IGR-OV1 cell line migrates normally in the presence of u-PA. We therefore investigated the role of cell signalling for u-PA induced cell locomotion. We show that cell migration induced by u-PA-u-PAR complex is always associated with tyrosine kinase activation for the following reasons: (1) the blockade of the u-PAR by a chimeric molecule (albumin-ATF) inhibits not only the u-PA-induced cell migration, but also the signalling in IGR-OV1 line; (2) the binding of u-PA to u-PAR on non-migrating IGR-OV1-Adria cells was not associated with tyrosine kinase activation; (3) the inhibition of tyrosine kinase also blocked cell migration of IGR-OV1. Therefore tyrosine kinase activation seems to be essential for the u-PA-induced cell locomotion possibly by the formation of a complex u-PAR-u-PA with a protein whose transmembrane domain can ensure cell signalling. Thus, IGR-OV1 and IGR-OV1-Adria cell lines represent a good model for the analysis of the mechanism of u-PA-u-PAR-induced cell locomotion.

The urokinase-type plasminogen activator system in cancer metastasis: a review

The urokinase-type plasminogen activator (u-PA) system consists of the serine proteinases plasmin and u-PA; the serpin inhibitors alpha2-anti-plasmin, PAI-1 and PAI-2; and the u-PA receptor (u-PAR). Two lines of evidence have strongly suggested an important and apparently causal role for the u-PA system in cancer metastasis: results from experimental model systems with animal tumor metastasis and the finding that high levels of u-PA, PAI-1 and u-PAR in many tumor types predict poor patient prognosis. We discuss here recent observations related to the molecular and cellular mechanisms underlying this role of the u-PA system. Many findings suggest that the system does not support tumor metastasis by the unrestricted enzyme activity of u-PA and plasmin. Rather, pericellular molecular and functional interactions between u-PA, u-PAR, PAI-1, extracellular matrix proteins, integrins, endocytosis receptors and growth factors appear to allow temporal and spatial re-organizations of the system during cell migration and a selective degradation of extracellular matrix proteins during invasion. Differential expression of components of the system by cancer and non-cancer cells, regulated by paracrine mechanisms, appear to determine the involvement of the system in cancer cell-directed tissue remodeling. A detailed knowledge of these processes is necessary for utilization of the therapeutic potential of interfering with the action of the system in cancers.

Inhibition of in vivo tumorigenicity and invasiveness of a human glioblastoma cell line transfected with antisense uPAR vectors

Our previous studies showed that glioblastomas express increased urokinase-type plasminogen activator receptors (uPARs) in comparison to low-grade gliomas (Yamamoto et al., Cancer Res., 54, 5016-5020, 1994). To explore whether downregulation of uPAR inhibits tumor formation and invasiveness, a human glioblastoma cell line was transfected with a cDNA construct corresponding to 300 bp of the human uPAR's 5' end in an antisense orientation, resulting in a reduced number of uPA receptors. Co-culture studies with tumor spheroids and fetal rat brain aggregates showed that antisense SNB19-AS1 cells expressing reduced uPAR failed to invade fetal rat brain aggregates. Intracerebral injection of SNB19-AS1 stable transfectants failed to form tumors and were negative for uPAR expression in nude mice. Thus uPAR appears in this model to be essential for tumorigenicity and invasion of glioblastomas in vivo.

Plasmin Is a Potent and Specific Chemoattractant for Human Peripheral Monocytes Acting Via a Cyclic Guanosine Monophosphate–Dependent Pathway

We have previously reported that the serine protease plasmin generated during contact activation of human plasma triggers biosynthesis of leukotrienes (LTs) in human peripheral monocytes (PMs), but not in polymorphonuclear neutrophils (PMNs). We now show that purified plasmin acts as a potent chemoattractant on human monocytes, but not on PMNs. Human plasmin or plasminogen activated with urokinase, but not active site-blocked plasmin or plasminogen, elicited monocyte migration across polycarbonate membranes. Similarly, stimulation of monocytes with plasmin, but not with active site-blocked plasmin or plasminogen, induced actin polymerization. As assessed by checkerboard analysis, the plasmin-mediated monocyte locomotion was a true chemotaxis. The plasmin-induced chemotactic response was inhibited by the lysine analog trans-4-(aminomethyl)cyclohexane-1-carboxylic acid (t-AMCA), which prevents binding of plasmin/ogen to the appropriate membrane binding sites. In addition, active site-blocked plasmin inhibited monocyte migration triggered by active plasmin. Further, plasmin-induced monocyte chemotaxis was inhibited by pertussis toxin (PTX) and 1-O-hexadecyl-2-O-methyl-rac-glycerol (HMG) and chelerythrine, two structurally unrelated inhibitors of protein kinase C (PKC). Plasmin, but not active site-blocked plasmin or plasminogen, triggered formation of cyclic guanosine monophosphate (cGMP) in monocytes. LY83583, an inhibitor of soluble guanylyl cyclase, inhibited both plasmin-induced cGMP formation and the chemotactic response. The latter effect could be antagonized by 8-bromo-cGMP. In addition, KT5823 and (Rp)-8-(p-chlorophenylthio)guanosine-3′,5′-cyclic monophosphorothioate [(Rp)-8-pCPT-cGMPs], two structurally unrelated inhibitors of cGMP-dependent protein kinase, inhibited plasmin-mediated monocyte chemotaxis. Thus, beyond being a stimulus for lipid mediator release, plasmin is a potent and specific chemoattractant for human monocytes acting via a cGMP-dependent mechanism. Therefore, plasmin represents a proinflammatory activator for human monocytes.

Expression of stromelysin-3 in the human placenta and placental bed

Human placentation is mediated by fetal trophoblastic cells which penetrate into the decidualized uterine endometrium. Trophoblast invasion requires the precisely regulated secretion of specific proteinases able to degrade the endometrial basement membranes and extracellular matrix. To document further the involvement of these proteinases during human placentation, we evaluated in vivo the expression of stromelysin-3, a member of the metalloproteinase family, during the first and third trimesters of pregnancy, by means of immunohistochemistry, in situ hybridization and Northern blot analysis. Human extravillous trophoblasts invading the maternal decidua produced stromelysin-3 during both, the first and third trimesters of pregnancy, but to a lesser extent during the latter. In floating villi, stromelysin-3 expression was restricted to the syncytiotrophoblasts that line intervillous vascular spaces. In conclusion, stromelysin-3 is expressed by differentiated, non-proliferative villous and extravillous trophoblastic cells in early and late placental beds and villi, and its pattern of expression evolves during pregnancy. Our observations suggest that stromelysin-3 could play a role in human placentation.

Plasminogen activator inhibitor-1 represses integrin- and vitronectin-mediated cell migration independently of its function as an inhibitor of plasminogen activation

Cell migration involves the integrins, their extracellular matrix ligands, and pericellular proteolytic enzyme systems. We have studied the role of plasminogen activator inhibitor-1 (PAI-1) in cell migration, using human amnion WISH cells and human epidermoid carcinoma HEp-2 cells in an assay measuring migration from microcarrier beads and a modified Boyden-chamber assay. Active, but not latent or reactive center-cleaved, PAI-1 inhibited migration. A PAI-1 mutant without ability to inhibit plasminogen activation was as active as wild-type PAI-1 as a migration inhibitor, showing that inhibition of plasminogen activation was not involved. PAI-1 specifically interfered with intergrin- and vitronectin-mediated migration: Migration onto vitronectin-coated but not onto fibronectin-coated surfaces was inhibited by PAI-1, a cyclic RGD peptide inhibited migration, and both cell lines expressed vitronectin-binding alpha v-integrins. In addition, active PAI-1, but not latent or reactive center-cleaved PAI-1, inhibited vitronectin binding to integrins in an in vitro binding assay, without affecting binding of fibronectin. Monoclonal antibodies against the urokinase receptor, another vitronectin binding protein, did not affect cell migration in the beads assay, while some inhibitory effect was observed in the Boyden-chamber assay. We conclude that PAI-1, independently of its role as a proteinase inhibitor, inhibits cell migration by competing for vitronectin binding to integrins, while the interference of PAI-1 with binding of vitronectin to the urokinase receptor may play a secondary role. These data define a novel function for the serpin PAI-1, enabling it to regulate cell migration over vitronectin-rich extracellular matrix in the body.

Evidence of a non-conventional role for the urokinase tripartite complex (uPAR/uPA/PAI-1) in myogenic cell fusion

Urokinase can form a tripartite complex binding urokinase receptor (uPAR) and plasminogen activator inhibitor type-1 (PAI-1), a component of the extracellular matrix (ECM). The components of the tripartite complex are modulated throughout the in vitro myogenic differentiation process. A series of experiments aimed at elucidating the role of the urokinase tripartite complex in the fusion of human myogenic cells were performed in vitro. Myogenic cell fusion was associated with increased cell-associated urokinase-type plasminogen activator (uPA) activity, cell-associated uPAR, and uPAR occupancy. Incubation of cultures with either uPA anticatalytic antibodies, or the amino-terminal fragment of uPA (ATF), which inhibits competitively uPA binding to its receptor, or anti-PAI-1 antibodies, which inhibit uPA binding to PAI-1, resulted in a 30 to 47% decrease in fusion. Incubation of cultures with the plasmin inhibitor aprotinin did not affect fusion. Decreased fusion rates induced by interfering with uPAR/uPA/PAI-1 interactions were not associated with significant changes in mRNA levels of both the myogenic regulatory factor myogenin and its inhibitor of DNA binding, Id. Incubation of cultures with purified uPA resulted in a decrease in fusion, likely due to a competitive inhibition of PAI-1 binding of endogenous uPA. We conclude that muscle cell fusion largely depends on interactions between the members of the urokinase complex (uPAR/uPA/PAI-1), but does not require proteolytic activation of plasmin. Since the intrinsic muscle cell differentiation program appears poorly affected by the state of integrity of the urokinase complex, and since cell migration is a prerequisite for muscle cell fusion in vitro, it is likely that the urokinase system is instrumental in fusion through its connection with the cell migration process. Our results suggest that the urokinase tripartite complex may be involved in cell migration in a non conventional way, playing the role of an adhesion system bridging cell membrane to ECM.

The complex between urokinase plasminogen activator and its type-1 inhibitor in breast cancer extracts quantitated by ELISA

ELISAs for urokinase plasminogen activator (uPA) and plasminogen activator inhibitor type-1 (PAI-1) have shown that tumor levels of these molecules are prognostic parameters in breast cancer as well as other types of cancer. These ELISAs measure the total amount of the given component, including preforms, active, inactive, and complex-bound forms. However, the amount of the active forms of a component may more closely reflect the ongoing level of proteolytic activity and thereby be particularly related to prognosis. Since the inactive complex between uPA and PAI-1 can only be formed the active forms of the individual components, we have developed a sensitive and specific uPA:PAI-1 complex ELISA consisting of a sandwich format with two monoclonal antibodies (MAbs) against PAI-1 as capture antibodies and three biotinylated MAbs against uPA as detector antibodies. The data were collected as kinetic measurements of bound alkaline phosphatase activity. A standard of uPA:PAI-1 complex could be specifically measured in the assay with a detection limit of 8 pg/ml and a linear relationship between signal and complex concentration up to 4 ng/ml. Neither free uPA nor free PAI-1 were detected by this assay and the addition to the internal standard of free PAI-1 in amounts up to 20 ng/ml or uPA did not reduce the detection of complex by the assay. This ELISA was applied to extracts from 20 individual breast cancers. Each tumor was extracted in two different buffers and the median concentration of uPA:PAI-1 complex in the optimal extraction buffer was 0.8 ng/mg protein, range 0.4-2.8 ng/mg protein. Extraction of the tumor tissue at a low pH prevented de novo formation of complex from free uPA and PAI-1 in the tissue without destabilizing preformed uPA:PAI-1 complex. During incubation of the assay plate at neutral pH further uPA:PAI-1 complex formation from free components in the extracts was blocked by p-nitrophenyl guanidinobenzoate (NPGB). Thus, the present assay selectively quantifies preformed complex in tumor extracts and will enable us, for the first time, to evaluate the potential prognostic value of the uPA:PAI-1 complex in cancer.

Characterization of the binding of different conformational forms of plasminogen activator inhibitor-1 to vitronectin. Implications for the regulation of pericellular proteolysis

Plasminogen activator inhibitor type 1 (PAI-1), the primary physiologic inhibitor of plasminogen activation, is associated with the adhesive glycoprotein vitronectin (Vn) in plasma and the extracellular matrix. In this study we examined the binding of different conformational forms of PAI-1 to both native and urea-purified vitronectin using a solid-phase binding assay. These results demonstrate that active PAI-1 binds to urea-purified Vn with approximately 6-fold higher affinity than to native Vn. In contrast, inactive forms of PAI-1 (latent, elastase-cleaved, synthetic reactive center loop peptide-annealed, or complexed to plasminogen activators) display greatly reduced affinities for both forms of adsorbed Vn, with relative affinities reduced by more than 2 orders of magnitude. Structurally, these inactive conformations all differ from active PAI-1 by insertion of an additional strand into beta-sheet A, suggesting that it is the rearrangement of sheet A that results in reduced Vn affinity. This is supported by the observation that PAI-1 associated with beta-anhydrotrypsin, which does not undergo rearrangement of beta-sheet A, shows no such decrease in affinity, whereas PAI-1 complexed to beta-trypsin, which does undergo sheet A rearrangement, displays reduced affinity for Vn similar to PAI-1.plasminogen activator complexes. Together these data demonstrate that the interaction between PAI-1 and Vn depends on the conformational state of both proteins and suggest that the Vn binding site on PAI-1 is sensitive to structural changes associated with loss of inhibitory activity.

Native and multimeric vitronectin exhibit similar affinity for heparin. Differences in heparin binding properties induced upon denaturation are due to self-association into a multivalent form

For many years, the concept that the heparin-binding sequence is sequestered within vitronectin and exposed upon denaturation of the protein has guided experimental design and interpretation of related structure-function studies on the protein. To evaluate binding of heparin to both native and denatured/renatured vitronectin, methods for monitoring binding in solution have been developed. A fluorescence method based on changes in an extrinsic probe attached to heparin has been used to evaluate heparin binding to native and denatured/renatured vitronectin. This approach indicates that there are not major differences in intrinsic heparin-binding affinities between native and renatured protein and invalidate the currently accepted model for a cryptic heparin-binding sequence in the protein. Denaturation and renaturation of vitronectin under near physiological solution conditions is accompanied invariably by self-association of the protein into a multimeric form (Zhuang, P., Blackburn, M. N., and Peterson, C. B. (1996) J. Biol. Chem. 271, 14323-14332), resulting in exposure of multiple heparin-binding sites on the surface of the oligomer. On the basis of the binding data from solution studies and interaction of the native monomer and the denatured multimeric form of vitronectin with a heparin column, along with evaluation of the ionic strength dependence of heparin binding to these vitronectin forms in solution, an alternative model is favored to account for the altered heparin binding properties of vitronectin associated with denaturation of the protein. This model proposes that multivalent interactions between heparin and multimeric vitronectin are responsible for differences in heparin affinity chromatography and ionic strength dependence compared with the native protein.

Detection of fibronectin expression by human endothelial cells using a enzyme-linked immunosorbent assay (ELISA): enzymatic degradation by activated plasminogen

An enzyme-linked immunosorbent assay (ELISA) has been developed to measure cellular fibronectin (cFN) in association with human umbilical vein endothelial cells (HUVEC) in culture. The expression of a number of functional domains on the cFN molecule was demonstrated using three specific murine monoclonal antibodies. This system was found to be sensitive, detecting as little as 0.156 microg/ml of cFN, and required only 1.3 x 10(5) cells per well confluent cells per experimental condition. This allowed multiple experiments to be performed on one batch of endothelial cells. cFN was detected on both viable and methanol fixed endothelial cells without significant non-specific antibody binding. The utility of this experimental model was studied by exploring the effect of urokinase activated plasminogen, a potent protease, on the expression of cFN and its functional domains.

Induced PAI-1 mRNA expression and targeted protein accumulation are early G1 events in serum-stimulated rat kidney cells

Expression of plasminogen activator inhibitor type-1 (PAI-1), a member of the SERPIN gene family that functions to regulate the plasmin-based pericellular proteolytic cascade, is growth state-regulated in normal rat kidney (NRK) cells (Ryan and Higgins, 1990, J. Cell. Physiol., 155:376-384; Ryan et al., 1996, Biochem. J., 314:1041-1046). Comparative analysis of arrest states induced in NRK cells upon exposure to serum-deficient (0.5% FBS) or serum-free culture conditions served to define the kinetics of PAI-1 gene expression and fate of de novo-synthesized PAI-1 protein. While cells rendered quiescent in serum-free or serum-deficient media were equivalent with regard to the time course of PAI-1 mRNA induction, the level of expressed transcripts (27-fold vs. 12-fold) and accumulated saponin fraction PAI-1 protein (12-fold vs. 6-fold) were consistently greater in cells recruited into exponential growth phase from a serum-free as compared to a serum-deficient arrest condition. Relative PAI-1 mRNA abundance increased within 1-2 hr post-serum addition, was maximal at 4 hr, and declined rapidly thereafter; this time course of expression coupled with placement of entry into DNA synthetic phase at approximately 12 hr after stimulation indicates that PAI-1 induction is an early-to-mid G1 phase event. Induced PAI-1 protein was evident immunocytochemically within 2 hr of serum stimulation as a peripheral "rim" of accumulated protein restricted to the cellular ventral surface at the plane of the substrate. No PAI-1 was detected between individual cells suggesting that this protein may be targeted directly to the undersurface region. By 6 hr post-stimulation, the rim of PAI-1 deposition increased in intensity and broadened to occupy approximately 30 to 50% of the total undersurface area. Double-label immunocytochemistry indicated that accumulated PAI-1 was deposited in close proximity to, but not actually within, vinculin-containing focal contact structures. Potential functionality of induced PAI-1 expression to either the initiation or maintenance of the serum-stimulated phenotype was assessed using antibodies to PAI-1. The IgG fractions of two different antisera which neutralize the ability of PAI-1 to complex with and thereby inhibit the catalytic activity of urokinase plasminogen activator significantly reduced (by 25-35%) the incidence of cells displaying the serum-stimulated phenotype; antibodies that bind PAI-1 but do not block PAI-1 inhibitory activity were without effect. In view of the vagaries of antibody accessibility and in situ neutralizing activity (particularly in a region as structurally complex as the focal contact), these data may actually underestimate the importance of PAI-1 in maintaining the activated phenotype.

Cerebrospinal fluid plasminogen activator inhibitor-1 in patients with neurological disease

AIM

To study cerebrospinal fluid (CSF) concentrations of plasminogen activator inhibitor type-1 (PAI-1) in patients with neurological disease.

METHODS

CSF PAI-1 concentrations were measured in 51 patients with neurological disease and 20 reference subjects using an ELISA. The patient group comprised three patients with viral meningitis, 20 with encephalitis, nine with acute lymphoblastic (n = 7) and myeloid (n = 2) leukaemia (with central nervous system involvement), and 19 with multiple sclerosis.

RESULTS

Raised PAI-1 concentrations were observed in patients with leukaemia, encephalitis and multiple sclerosis. There was no difference in the mean concentrations of PAI-1 in patients with meningitis when compared with the reference subjects. The highest mean (SEM) PAI-1 concentration was found in patients with leukaemia (1.28 (0.36) ng/ml), and the next highest in those with encephalitis (1.19 (0.20) ng/ml). these values were much higher than those in patients with viral meningitis. In a previous report, raised CSF tissue-type plasminogen activator (tPA) activities were detected in patients with multiple sclerosis, leukaemia and encephalitis, with mean activities in decreasing order. PAI-1 concentrations in the same patients were the reverse of their corresponding tPA activities, being higher in those with leukaemia and encephalitis, than in patients with multiple sclerosis. There was no association between CSF PAI-1 concentrations and age in either patients or controls. Similarly, there was no association between CSF PAI-1 concentrations and urokinase-type plasminogen activator (uPA).

CONCLUSIONS

Raised CSF PAI-1 concentrations may be used as a non-specific marker of neurological disease. Moreover, PAI-1 may play an important role in regulating the functions tPA, and probably uPA, in CSF.

P52PAI-1 gene expression in butyrate-induced flat revertants of v-ras-transformed rat kidney cells: mechanism of induction and involvement in the morphological response

Sodium n-butyrate-induced flat reversion in v-K-ras oncogene-transformed rat kidney (KNRK) cells is associated with transcriptional activation of the p52PAI-1 gene (which encodes the type-1 inhibitor of plasminogen activator). Butyrate-initiated expression of p52PAI-1 mRNA and protein correlated with induced cell spreading and preceded development of cell-to-substrate focal adhesions. Such undersurface matrix contact structures, which are absent from parental KNRK cells, require proximal PAI-1 deposition for their stabilization. Stimulated p52PAI-1 expression by flat revertants (approximating 25-fold that of control cells) and the accompanying cytoarchitectural reorganization appeared to be programmed responses to butyrate as both events required de novo RNA and protein synthesis, metabolic characteristics consistent with a secondary pathway of gene regulation. To assess the relevance of p52PAI-1 induction to the process of flat reversion more critically, a molecular genetic approach was designed to maintain high-level constitutive p52PAI-1 synthesis in the absence of butyrate. KNRK cells transfected with a Rc/CMVPAI plasmid construct, in which expression of a p52PAI-1 cDNA insert was driven by enhancer-promoter sequences from the immediate-early gene of human cytomegalovirus, formed colonies comprised of flat-revertant-like cells with a greater frequency than did cells transfected with the Rc/CMV vector alone (24.8% and 1.7% respectively). Comparative analysis of randomly selected Rc/ CMVPAI clones indicated that a 10-fold increase in immunoreactive p52PAI-1 protein, relative to Rc/CMV isolates, correlated with generation of the flat phenotype. These data suggest that induced p52PAI-1 expression probably functions in the development of morphological revertants in the KNRK cell system.

The human ENO1 gene product (recombinant human alpha-enolase) displays characteristics required for a plasminogen binding protein

Plasminogen binds with low affinity in a lysine-dependent manner to many cell types. Previously, a 54 kDa plasminogen receptor found on the surface of U-937 cells was identified as an alpha-enolase-like molecule. The aims of this study were to determine whether recombinant alpha-enolase (r-alpha-enolase), encoded by ENO1, was a plasminogen binding protein and to generate polyclonal antibodies against this antigen. Plasminogen specifically bound r-alpha-enolase with a Kd 1.9 microM and approached saturation at 10 microM. Lysine-dependent plasminogen binding to r-alpha-enolase was demonstrated by a greater than 80% inhibition of binding by the lysine analogues epsilon-amino caproic acid and tranexamic acid, whilst only 14% inhibition occurred with the arginine analogue benzamidine. Removal of the C-terminal lysine residue of r-alpha-enolase with carboxy-peptidase B significantly reduced its plasminogen binding capacity, suggesting that binding required C-terminal lysine residue of r-alpha-enolase. Binding to r-alpha-enolase enhanced the activation rate of plasminogen by urokinase but prevented alpha 2-antiplasmin from binding plasminogen. Taken together, these data suggest that the gene product of human ENO1 encodes an authentic plasminogen binding protein.

Expression and induction by IL-6 of the normal and variant genes for human plasminogen

We examined the promoter activity of the gene for human plasminogen (PLG) employing its 1.1 kb fragment of the 5'-flanking region inserted in front of a reporter gene. Deletion analysis revealed that a region surrounding the transcription start site was essential for the PLG expression. Since the PLG gene has three sequences for the interleukin-6 (IL-6) responsive element, we examined the effect of IL-6 on the PLG expression. IL-6 stimulation of PLG resulted in a 2.5-fold increase in its transcription. This is also true for the PLG gene of a case with dysplasminogenemia. Although the patient's gene had six mutations in the 5'-flanking region, its promoter activity was 1.8-fold that of normal PLG.

Induced PAI-1 mRNA expression and targeted protein accumulation are early G1 events in serum-stimulated rat kidney cells

Expression of plasminogen activator inhibitor type-1 (PAI-1), a member of the SERPIN gene family that functions to regulate the plasmin-based pericellular proteolytic cascade, is growth state-regulated in normal rat kidney (NRK) cells (Ryan and Higgins, 1990, J. Cell. Physiol., 155:376-384; Ryan et al., 1996, Biochem. J., 314:1041-1046). Comparative analysis of arrest states induced in NRK cells upon exposure to serum-deficient (0.5% FBS) or serum-free culture conditions served to define the kinetics of PAI-1 gene expression and fate of de novo-synthesized PAI-1 protein. While cells rendered quiescent in serum-free or serum-deficient media were equivalent with regard to the time course of PAI-1 mRNA induction, the level of expressed transcripts (27-fold vs. 12-fold) and accumulated saponin fraction PAI-1 protein (12-fold vs. 6-fold) were consistently greater in cells recruited into exponential growth phase from a serum-free as compared to a serum-deficient arrest condition. Relative PAI-1 mRNA abundance increased within 1-2 hr post-serum addition, was maximal at 4 hr, and declined rapidly thereafter; this time course of expression coupled with placement of entry into DNA synthetic phase at approximately 12 hr after stimulation indicates that PAI-1 induction is an early-to-mid G1 phase event. Induced PAI-1 protein was evident immunocytochemically within 2 hr of serum stimulation as a peripheral "rim" of accumulated protein restricted to the cellular ventral surface at the plane of the substrate. No PAI-1 was detected between individual cells suggesting that this protein may be targeted directly to the undersurface region. By 6 hr post-stimulation, the rim of PAI-1 deposition increased in intensity and broadened to occupy approximately 30 to 50% of the total undersurface area. Double-label immunocytochemistry indicated that accumulated PAI-1 was deposited in close proximity to, but not actually within, vinculin-containing focal contact structures. Potential functionality of induced PAI-1 expression to either the initiation or maintenance of the serum-stimulated phenotype was assessed using antibodies to PAI-1. The IgG fractions of two different antisera which neutralize the ability of PAI-1 to complex with and thereby inhibit the catalytic activity of urokinase plasminogen activator significantly reduced (by 25-35%) the incidence of cells displaying the serum-stimulated phenotype; antibodies that bind PAI-1 but do not block PAI-1 inhibitory activity were without effect. In view of the vagaries of antibody accessibility and in situ neutralizing activity (particularly in a region as structurally complex as the focal contact), these data may actually underestimate the importance of PAI-1 in maintaining the activated phenotype.

Plasminogen activator system in pemphigus vulgaris

Pemphigus vulgaris (PV) is caused by autoantibodies against desmosomes and is characterized by intra-epidermal blisters. The pathology of PV has been linked with plasminogen activation in lesional epidermis. The plasminogen activator system (PA system) consists of urokinase-type plasminogen activator (uPA), tissue-type PA (tPA), as well as the two types of plasminogen activator inhibitors (PAI-1 and PAI-2). In keratinocytes, uPA binds to a specific cell surface receptor for uPA (uPA-R = CD87) in an autocrine manner. Cell-bound uPA is regulated by PAIs. The central PA system component plasminogen, which is present in plasma and interstitial fluids, is bound to the keratinocyte surface via plasmin(ogen) binding sites, where it can be activated by uPA-R-bound uPA. Cell surface-associated plasmin then mediates pericellular proteolysis. As the topographical organization of the distinct PA system components in lesional epidermis of PV remained elusive, we have performed the present immunohistological analysis of lesional and non-lesional epidermis of PV. In keratinocytes directly involved in the epidermal split formation, plasmin(ogen) was stained in nine of 10 cases, uPA-R and uPA in four of 10 cases and PAI-2 in seven of 10 cases. Together, acantholytic plasmin(ogen)+ keratinocytes appeared in three different phenotypes: uPA-R+/uPA+ and PAI-2+, uPA-R-/uPA- and PAI-2+, as well as uPA-R-/uPA- and PAI-2-. Our findings demonstrate that, in acantholytic keratinocytes of PV, PAs and PAIs appear as differentially regulated components of the PA system.

Matrix metalloproteinase-2 (72 kD type IV collagenase) expression occurs in the early stage of human melanocytic tumour progression and may have prognostic value

Matrix metalloproteinase-2 (MMP-2), a member of the matrix metalloproteinase family, participates in degradation of the pericellular and extracellular matrix during neoplastic growth and metastasis. Experimental data have substantiated its role in melanoma invasion, but there is no information at present concerning its expression in histological specimens from human melanocytic tumors. This study describes the occurrence and immunolocalization of MMP-2 in human melanocytic lesions, defining distinct steps in melanoma progression. Paraffin-embedded sections from 118 melanocytic lesions were immunostained using a specific antibody to 72 kD type IV collagenase. The material included 34 common naevocellular naevi, 14 dysplastic naevi, 21 in situ melanomas, 20 primary malignant melanomas, and 29 melanoma metastases. Intracytoplasmic MMP-2 immunoreactive protein was found in the 'naevocytic nests' of common naevi, in junctional naevus cells, and in melanoma cells. The surrounding normal skin stained negatively, except for occasional macrophages, sweat glands, and hair follicles. The number of MMP-2-positive cells increased with decreasing architectural organization and increasing atypia in the melanocytic lesions. The MMP-2 positivity in the primary and subcutaneous melanoma lesions correlated with later haematogenous metastasis. The data suggest that MMP-2 expression is an early event in melanocytic tumour progression, but is nevertheless prognostic for haematogenous metastasis in melanoma.

Conformational changes of the reactive-centre loop and beta-strand 5A accompany temperature-dependent inhibitor-substrate transition of plasminogen-activator inhibitor 1

We have studied conformational changes of type-1 plasminogen-activator inhibitor (PAI-1) during a temperature-dependent inhibitor-substrate transition by measuring susceptibility of the molecule to non-target proteinases. When incubated at 0 degree C instead of the normally used 37 degrees C, a tenfold decrease in the specific inhibitory activity of active PAI-1 was observed. Accordingly, PAI-1 was recovered in a reactive-centre-cleaved form from incubations with urokinase-type plasminogen activator (uPA) and tissue-type plasminogen activator (tPA) at 0 degree C, but not at 37 degrees C. It thus behaved as a substrate for the target proteinases at the lower temperature. Active PAI-1 was exposed to a variety of non-target proteinases, including elastase, papain, thermolysin, trypsin, and V8 proteinase. It was found that specific peptide bonds in the reactive centre loop (RCL) and strand 5 in beta-sheet A (s5A) had a temperature-dependent proteolytic susceptibility, while the P17-P16 (E332-S333) bond, forming the hinge between s5A and the RCL, showed indistinguishable susceptibility to proteolysis by V8 proteinase at 0 degree and 37 degrees C. In latent and reactive-centre-cleaved PAI-1, all the bonds were resistant to proteolysis at the higher as well as the lower temperature. An anti-PAI-1 monoclonal antibody maintained the inhibitory activity of PAI-1 and prevented reactive centre cleavage at 0 degree C, and thus prevented substrate behaviour. Concomitantly, it caused specific changes in proteolytic susceptibility of s5A and the RCL, but it did not affect cleavage of the P17-P16 bond by V8 proteinase. Our observations suggest that temperature-dependent conformational changes of beta-sheet A and the RCL determine whether the serpin act as an inhibitor or a substrate. Furthermore they suggest that the RCL of PAI-1 is fully extracted from beta-sheet A in the inhibitory as well as in the substrate form, favoring a so-called induced conformational state model to explain why inhibitory activity requires partial insertion of the RCL into beta-sheet A.

The proteolytic potential of normal human melanocytes: comparison with other skin cells and melanoma cell lines

To understand the contribution of epidermal melanocytes in the proteolytic potential of human skin, we have studied melanocytes grown in a low-serum medium deprived of phorbol esters, cholera toxin, and other non-physiological supplements. We focused on the plasminogen activation system and certain matrix metalloproteinases (gelatinases). Supposing that the proteolytic activity of cells can influence binding to collagen matrix and its reorganization, we have analyzed these parameters as well. We found that human melanocytes secreted tissue-type plasminogen activator and utilised it to generate cell-bound plasmin. No urokinase-type plasminogen activator was detected in the cultures but its receptor was found in cell extracts. Both the 72 kDa and 92 kDa gelatinases were secreted by the cells and in equal amounts. In addition, melanocytes secreted the wide-spectrum proteinase inhibitor alpha-2-macroglobulin. Melanocytes cast into collagen matrices retained a rounded morphology, did not extend processes, and were unable to contract collagen lattices. As a control, these parameters were investigated in parallel in cultures of human keratinocytes, dermal fibroblasts, and two melanoma cell lines. The obtained characteristics suggest that normal human melanocytes are proteolytically active cells. This function may pertain to skin physiology and pathophysiology.

Degradation of human plasma and extracellular matrix fibronectin by tissue type plasminogen activator and urokinase

Fibronectins and plasminogen activators, both tissue and urokinase types, are involved in the physiopathological degradation of the extracellular matrix. Previous reports indicate that fibronectin can be degraded by urokinase without plasminogen. Also, we have shown that tissue-type plasminogen activator can cleave fibronectin, without plasminogen, generating fragments of 30 and 220-250 kDa detectable by immunoblotting analysis. A comparison with urokinase-induced degradation indicates that the cleavage sites are the same for both plasminogen activators. One is close to the carboxyl-terminal, disrupting the fibronectin dimeric structure, and one is near the amino-terminal, generating a 30 kDa fragment. In solution, the activity of tissue-type plasminogen activator was prevalent on the amino-terminal site, while urokinase activity was prevalent on the carboxyl-terminal site. On fibronectin immobilized onto a gelatin coated surface, only the 30 kDa fragment was released when treated with both plasminogen activators. Plasminogen activators also were active on fibronectin assembled into the extracellular matrix of cultured fibroblasts. Urokinase caused the complete disappearance of extracellular matrix fibronectin, together with the release of the 30 and the 220-250 kDa fibronectin fragments, but left a flat morphology, while tissue-type plasminogen activator induced the release of the 30 kDa fragment associated with changes in cellular morphology. The plasminogen-independent fibronectin degradation exerted by tissue-type plasminogen activator and urokinase is 100 times lower than that exerted by plasmin. This may provide a mechanism for localized and limited degradation of fibronectin preventing the generalized proteolysis associated with plasminogen activation.

Bone resorption and response to calcium-regulating hormones in the absence of tissue or urokinase plasminogen activator or of their type 1 inhibitor

Plasminogen activators (PA) are implicated in cell migration and tissue remodeling, two components of the bone resorption processes. Using mice with inactivated tissue PA (tPA), urokinase PA (uPA), or type 1 PA inhibitor (PAI-1) genes, we evaluated whether these processes, or their stimulation by parathyroid hormone (PTH) or 1,25-dihydroxyvitamin (1,25[OH]2D3) are dependent on these genes. Two culture models were used, one involving 19-day fetal calvariae, to evaluate the direct resorptive activity of osteoclasis, and the other involving 45Ca-labeled 17-day fetal metatarsals, in which this activity depends on preliminary (pre)osteoclast migration. PTH similarly increased (about 10-fold) PA activity in calvariae from wild-type tPA+/+ and uPA+/+ or deficient uPA-/- and PAI-/- mice; it affected only tPA, not uPA. In tPA-/- bones, the low PA levels, due to uPA, were not influenced by PTH. Calcitonin did not affect PA responses to PTH. No differences were observed between tPA+/+, tPA-/-, uPA+/+, and uPA-/- calvariae for any parameter related to bone resorption (development of lacunae, release of calcium and lysosomal enzymes, accumulation of collagenase, loss of hydroxyproline), indicating similar responses to PTH or calcitonin. The progressive 45Ca release was largely similar in cultures of tPA+/+, tPA-/-, uPA+/+, uPA-/-, PAI+/+, or PAI-/- metatarsals and it was similarly enhanced by PTH or 1,25(OH)2D3. However, uPA-/- metatarsals released 45Ca at a slower rate at the beginning of the cultures, suggesting an impaired recruitment of the (pre)osteoclasts, which migrate at that time from the periosteum into the calcified cartilage. Thus, it appears that the direct resorptive activity of the osteoclasts does not necessitate the presence of either tPA or uPA, but uPA is likely to facilitate the migration of the (pre)osteoclasts toward the mineralized surfaces. Although considerably enhanced by PTH, tPA does not mediate the actions of PTH (nor of 1,25[OH]2D3) evaluated in these models.

Tetranectin and plasmin/plasminogen are similarly distributed at the invasive front of cutaneous melanoma lesions

The induction of expression of the components of the proteolytic plasminogen activation system in cutaneous melanocytic tumour progression has previously been reported. Plasminogen activators, their inhibitors, and the receptor for urokinase were present only in advanced primary melanomas and melanoma metastases. The present study reports on the presence of tetranectin and plasmin/ plasminogen, two proteins connected with plasminogen activation, in cutaneous melanocytic lesions. The distribution of tetranectin and plasminogen was studied by immunohistochemistry in 105 freshly frozen melanocytic lesions of common naevocellular naevi (n = 24), atypical naevi (n = 14), early (n = 12) and advanced (n = 20) primary melanomas, and melanoma metastases (n = 35). Both tetranectin and plasminogen were detected in a variety of tissue components. In all stages of melanocytic tumour progression, tetranectin was found in endothelium, perivascular dendritic cells, and leukocytes. Plasminogen was present in endothelium and in the basal layer of the normal skin. Tetranectin and plasminogen staining of fibroblastic cells at the invasive front and of extracellular matrix was, however, restricted to malignant lesions. Co-localization of tetranectin and plasminogen was found in 50 per cent of the early primary melanomas and in more than 75 per cent of the advanced melanomas and melanoma metastases. These results suggest a coordinated role for tetranectin and plasminogen at the invasive front of melanomas. Tetranectin-bound plasminogen may facilitate the migration of tumour cells.

The plasminogen activation system in tumour invasion and metastasis

The involvement of proteases in the metastatic spread of tumour cells and in tumour related processes, such as angiogenesis and ulceration, has been known for many decades. This chapter reviews the involvement of one proteolytic system--the plasminogen activation system--in tumour progression. In recent years, many biochemical properties of the various components of the plasminogen activation system have become known. These properties and the functional relationship between the components are discussed in the first section. Since interfering with proteolysis by tumour cells and by newly formed endothelial cells can be an objective for future therapy, experimental tumour models have been used to study the effects of inhibitors of plasminogen activation. The second section deals with this issue. Finally, the presence of the various components of the plasminogen activation system in human tumours is reviewed. Following the availability of specific ELISAs, antibodies and molecular probes, the content and the cellular distribution of the components of the plasminogen activation system have recently been mapped in various human tumours.

Cerebrospinal fluid activity of tissue plasminogen activator in patients with neurological diseases

AIM

To study cerebrospinal fluid (CSF) activity of tissue plasminogen activator (tPA) in patients with neurological diseases.

METHODS

CSF tPA and urokinase (uPA) activities were studied using an immunocapture assay and zymography in 44 patients with neurological disease and 20 reference subjects. The patient group comprised three patients with meningitis, 21 with encephalitis, nine with acute lymphoblastic (n = 7) and myeloid (n = 2) leukaemia, seven with multiple sclerosis, three with facial paresis, and one with polyradiculitis.

RESULTS

Raised tPA activities were observed in patients with multiple sclerosis, leukaemia and encephalitis. In contrast, there were no differences in the mean activities of tPA in patients with meningitis or other diseases compared with the reference subjects. The highest tPA activities were found in patients with multiple sclerosis. The mean activity in patients with leukaemia was higher than in those with meningitis and polyradiculitis, but not encephalitis and facial paresis. Although the CSF tPA activity correlated positively with age in reference subjects, no correlation was observed in patients. Samples were qualitatively screened for both tPA and uPA activity by zymography and positive samples were quantitated. Some of the samples had quantifiable levels of uPA activity: three of seven multiple sclerosis samples, 10 of 21 samples from patients with encephalitis and five of nine leukaemic samples. The highest activities were recorded in patients with leukaemia. uPA was not detected in the CSF of the patients with meningitis, facial paresis or polyradiculitis.

CONCLUSIONS

Plasminogen activator activity can be measured reliably in CSF and the assessment of tPA activity may be useful for studying the pathogenesis of neurological diseases.

Protein kinase C mediates up-regulation of urokinase and its receptor in the migrating keratinocytes of wounded cultures, but urokinase is not required for movement across a substratum in vitro

Both in cell culture and in vivo, keratinocytes that are migrating in response to a wound express enhanced levels of both urokinase-type plasminogen activator (uPA) and the uPA cell surface receptor (uPA-R). To explore the mechanism of this up-regulation, keratinocyte cultures were treated proir to wounding with a variety of metabolic and growth factor inhibitors in order to evaluate their effect on uPA and uPA-R expression. Actinomycin D and cycloheximide inhibited the up-regulation of both uPA and uPA-R, as determined by immunohistochemistry, indicating that RNA and protein syntheses are required for their induction in migrating keratinocytes. Neither removal of protein growth factors from the medium nor addition of inhibitory antibodies to a number of growth factors depressed uPA or uPA-R induction; these findings suggest that a variety of exogenous or endogenous growth factors [i.e., basic fibroblast growth factor (bFGF), epidermal growth factor (EGF), transforming growth factor-alpha (TGF-alpha), amphiregulin, and tumor necrosis factor-alpha (TNF-alpha) do not have a critical role in the induction of uPA or uPA-R. In contrast, when protein kinase C (PKC) was either down-regulated with bryostatin 5 or inhibited with Ro31-8220 or staurosporine, the expression of both uPA and uPA-R was greatly decreased in migrating keratinocytes. Furthermore, pharmacologic activation of PKC enhanced uPA levels in non-wounded cultures. These data suggest that the enhanced expression of uPA and uPA-R in migrating keratinocytes is mediated by selective activation of PKC in these cells, perhaps secondary to alterations in the cytoskeleton induced by wounding. To test the requirement for uPA during keratinocyte migration in vitro, the extent of migration was quantified in the presence and absence of a variety of inhibitors in the wounded culture model. Migration was not altered by actinomycin D, cycloheximide, any of the above growth factor inhibitors, anti-uPA antibodies, a variety of inhibitors of uPA or plasmin enzymatic activity, or exogenous uPA. The independence of keratinocyte migration in vitro from uPA was further suggested by experiments which combined the phagokinetic assay of migration and the zymographic assay for pericellular uPA activity; no relationship was observed between pericellular uPA activity and the motility of individual cells.

The role of cell adhesion molecules and proteases in tumor invasion and metastasis

Over the past 10 years it has become apparent that invasion and metastasis are extremely complex processes; neoplastic cells must escape from the primary tumor, degrade the extracellular matrix, migrate to distant sites, arrest in the capillaries, and migrate through the basement membrane and underlying connective tissue to the metastatic site. Therefore, tumor cells must exhibit considerable flexibility in their adhesive interactions, and this is reflected in a complex and dynamic expression pattern of cell adhesion molecules, proteases, protease inhibitors, motility factors, and growth factors. Despite the recent explosion of information regarding adhesion-related molecules, questions as to their possible roles in normal tissue architecture and as to how alterations in their expression or structure may be responsible for the progression from a single malignant cell to a lethal metastatic disease need further investigation. Moreover, efforts should be made to use the obtained knowledge to contribute to improvements in the clinical management of cancer. In this review the different classes of cell adhesion molecules and proteases are summarized, with special emphasis being placed on molecules that have been shown to correlate with invasion and metastasis. Furthermore, the role of E-cadherin in cell adhesion and invasive processes is discussed in more detail, since E-cadherin may be considered promising as a candidate among cell-adhesion-regulating molecules to be used as a biomarker for malignancy. We also elaborate on the role of the catenins, which associate with and are important for the functioning of E-cadherin.

Inverse correlation between mRNA expression of plasminogen activator inhibitor-2 and lymph node metastasis in human breast cancer

We examined mRNA expressions of urokinase-type plasminogen activator (u-PA), its specific receptor (u-PR), and plasminogen activator inhibitors (PAI-1 and PAI-2) in 50 human breast cancers by the reverse transcriptase-polymerase chain reaction method. The expressions of the genes are discussed in relation to the clinicopathological findings. In the overall population in breast cancers, a low level of PAI-2 expression was significantly associated with lymph node involvement (P < 0.0001). The u-PA, u-PR, and PAM expressions tended to be at high levels in such metastatic cancers. Also, positive expression of u-PA, u-PR, and PAI-1 was significantly correlated with negative expression of PAI-2. These results indicate that PAI-2 may play a critical role in the regulation of extracellular matrix degradation during tumor cell invasion and metastasis, and the expression of PAI-2 may be useful as a marker to evaluate the prognosis of breast cancers.

Combined overexpression of urokinase, urokinase receptor, and plasminogen activator inhibitor-1 is associated with breast cancer progression: an immunohistochemical comparison of normal, benign, and malignant breast tissues

BACKGROUND

A strong positive correlation exists between the breast cancer tissue content of either urokinase-plasminogen activator (uPA) or plasminogen activator, inhibitor type I (PAI-1), quantified in the tissue extracts by immunoassays, and the survival of patients with breast cancer. Furthermore, several studies assign to the urokinase-type plasminogen activator receptor (uPAR) a pivotal role in triggering the proteolytic activity of the urokinase pathway involved in tumor stroma degradation, tumor spread and metastasis. However, the pattern of distribution of uPAR in normal and cancerous human tissue and the pattern of coexpression of activators and inhibitors that occurs in breast cancer tissues is not completely known.

METHODS

The immunohistochemical localization of uPAR, uPA, tPA) and PAI-1 was evaluated by using the avidin-biotin immunoperoxidase technique and affinity-purified monoclonal antibodies from American Diagnostica Inc. Studies were performed in formalin fixed, paraffin-embedded tissue prepared from 23 surgically excised non-neoplastic breast tissues and 18 ductal breast carcinomas.

RESULTS

While the expression of uPAR protein represents a constant feature of invasive ductal breast cancer, it was also observed in most of the breast tissue samples, including the normal breast tissues. The staining for uPAR was mainly localized on normal or tumoral epithelial cells, even if the co-expression of uPAR in stromal cells was frequently observed in adjacent slides. A semiquantitative analysis of immunohistochemical results showed that uPAR and PAI-1 were overexpressed in invasive breast cancer in comparison with normal and benign breast tissues. In addition, uPA was higher in both invasive breast carcinomas and benign breast lesions with respect to normal breast tissues.

CONCLUSIONS

We showed that overexpression of uPAR, uPA, and its main inhibitor, PAI-1, is a constant feature of invasive ductal breast carcinomas. However, the expression of the above fibrinolytic reactants is not specific for breast cancer since positive staining for these molecules was frequently observed in benign breast lesions as well as in normal breast tissues. The combined increased expression of uPA and its cellular receptor, uPAR on the surface of tumor epithelial cells may account for the activation of the proteolytic system which occurs in breast cancer.

Complex regulation of plasminogen activator inhibitor type-1 (PAI-1) gene expression by serum and substrate adhesion

Expression of plasminogen activator inhibitor type-1 (PAI-1), a member of the serine protease inhibitor (SERPIN) superfamily that functions to negatively regulate the plasmin-based pericellular proteolytic cascade, was induced early after exposure of growth-arrested normal rat kidney (NRK) cells to serum-containing medium. Increased PAI-1 transcription was rapid (evident within 10 min of serum addition) and involved immediate-early response kinetics. [3H]Thymidine autoradiography was used to map the time frame of PAI-1 expression during a synchronous growth cycle. PAI-1 transcript accumulation peaked in mid-G1 phase (approx. 4-6 h post-stimulation) and declined prior to, or concomitant with, the onset of DNA synthetic phase. Serum increased PAI-1 expression in NRK cells in agarose suspension, as well as monolayer, culture; induction in suspended cells (similar to monolayer culture conditions) also occurred in the presence of cyclohexamide or puromycin. The serum-inductive pathway leading to PAI-1 gene activation is thus functional regardless of adhesive conditions or capacity for de novo protein synthesis. The amplitude of induction and maintenance of expression in later stages of G1, however, were subject to adhesive influences. PAI-1 transcript accumulation at 4 and 8 h post-stimulation in newly adherent cells, moreover, was blocked by puromycin, indicating that both immediate-early and secondary mechanisms regulate PAI-1 mRNA levels during progression of NRK cells through an 'activated' G1 growth phase.

Limited plasmin proteolysis of vitronectin. Characterization of the adhesion protein as morpho-regulatory and angiostatin-binding factor

The adhesion protein vitronectin is associated with extracellular matrices and serves as cofactor for plasminogen-activator inhibitor-1. Limited proteolysis by plasmin converts vitronectin into defined fragments which are detectable at sites of inflammation and angiogenesis. The loss and gain of binding functions of vitronectin fragments for macromolecular ligands was characterized in the present study. The initially generated 61--63-kDa vitronectin-(1--348)-fragment serves as typical binding component for plasminogen and binding function was lost upon carboxypeptidase B treatment indicating the importance of a C-terminal lysine. Complementary binding sites reside in isolated plasminogen kringles 1--3 (designated angiostatin) as deduced from direct binding and ligand blotting experiments. A synthetic vitronectin-(331--348)-peptide from the C-terminus of the 61--63-kDa fragment could mimic plasminogen and angiostatin binding. Also, the immobilized peptide bound tissue plasminogen-activator and mediated plasmin formation, comparable to fibrinogen-derived peptides. The 61--63-kDa vitronectin fragment was indistinguishable in its adhesive properties to intact vitronectin and bound active but not latent plasminogen-activator inhibitor-1. Late plasminolysis of vitronectin resulted in the processing of the N-terminal region of the protein with the generation of 42 kDa/35-kDa fragments that had Gly89 as new N-terminus and that were ineffective in promoting cell adhesion. Thus, at sites of cell-matrix interactions which become proteolytically modified by plasmin during inflammatory and angiogenic processes, vitronectin serves as plasminogen/angiostatin-binding factor. Due to this differential change in functions particularly at sites of deposition in the vascular system or at wound sites vitronectin is considered to be an important morpho-regulatory factor.

Serum is a potent stimulator of keratinocyte tissue plasminogen activator expression

The plasminogen activator (PA) proteolytic cascade comprises two enzymes known as urokinase PA (uPA) and tissue PA (tPA), both of which activate plasminogen to plasmin. In normal human epidermis uPA is the predominant PA. In lesional epidermis from patients with a variety of cutaneous diseases, including psoriasis, pemphigus foliaceous, pemphigus, vulgaris, bullous pemphigoid, and benign chronic pemphigus, however, tPA is selectively elevated and becomes the predominant PA activity. The enhanced tPA is likely to be a reaction to the disease challenge rather than an initiating event in these clinically and etiologically diverse lesions. In the present study, cultured human keratinocytes, propagated under serum-free conditions, have been shown to respond to the addition of bovine or human serum with an increase in tPA activity and antigen. Furthermore, tPA is found predominantly in the suprabasal keratinocytes both in lesional epidermis and in stratified cultures that have been incubated for approximately 8 d in the presence of serum. These results suggest a possible mechanism by which epidermal tPA may be increased in diverse cutaneous lesions: The plasma infiltrated into lesional epidermis may stimulate the suprabasal keratinocytes in vivo to express tPA.

Differential expression of tissue inhibitor of metalloproteinases-2 by cutaneous squamous and basal cell carcinomas

Tumor cell invasion and metastasis are considered to represent a multistep process leading to the degradation of the extracellular matrix by proteolytic enzymes. The functional activity of matrix metalloproteinases (MMPs) is controlled by tissue inhibitor of metalloproteinases-2 (TIMP-2), which has been shown to inhibit tumor cell invasion and metastasis in vitro and in vivo. To assess the role of TIMP-2 in skin-derived epithelial tumors, we have analyzed the expression of TIMP-2 mRNA in primary tissue samples from human cutaneous basal (BCC) and squamous cell carcinoma (SCC) for a correlation with their different invasive and metastatic potential. Comparative quantitative analysis of TIMP-2 mRNA levels by Northern blot hybridization demonstrated significantly higher TIMP-2 tissue levels in BCC than in SCC, indicating an inverse correlation between TIMP-2 expression and the metastatic capacity of these tumors in vivo. By in situ hybridization, tumor stromal cells were identified as the principal source of TIMP-2 mRNA in both BCC and SCC. A comparable distribution has been reported previously for several matrix metalloproteinases in cutaneous BCC and SCC, indicating co-localization of metalloproteinases with their respective inhibitor. These results may suggest that TIMP-2 substantially contributes to the biological behavior of epithelium-derived skin tumors by significantly inhibiting tumor cell metastasis.

Tumor progression and angiogenesis: cathepsin B & Co

Experimental and clinical evidence reveals that the growth of solid tumors is dependent on angiogenesis. Proteolytic enzymes such as plasminogen activators and matrix metalloproteinases have been implicated in this neovascularization. The role of lysosomal proteases in this process has yet to be explored. Increased expression of the lysosomal cysteine protease cathepsin B has been observed in many etiologically different tumors, including human brain, prostate, breast, and gastrointestinal cancers. Immunohistochemical and in situ histochemical studies have demonstrated expression of cathepsin B in neovessels induced during malignant progression of human glioblastoma and prostate carcinomas. In these two tumor types, neovessels stain strongly for cathepsin B compared with the normal microvasculature. As an initial point to elucidate whether cathepsin B is an important component of the angiogenic response in tumours, we analyzed expression of cathepsin B in endothelial cells during neovessel formation. We present evidence for strong immunostaining of cathepsin B in rat brain microvascular endothelial cells as they form capillary tubes in vitro. This finding is discussed within the general framework of the role of proteolytic enzymes in tumor invasion and angiogenesis.

Transforming growth factor-beta system and its regulation by members of the steroid-thyroid hormone superfamily

TGF-beta s and their receptors are expressed ubiquitously, and they act as key regulators of many aspects of cell growth, differentiation, and function. Steroid action on target tissues is often associated with increase in TGF-beta isoforms. Regulation of TGF-beta expression and activation is crucial for normal development and growth control. The loss of responsiveness of different tumor cells to the antiproliferative effects of TGF-beta is a common feature in carcinogenesis. Multiple changes are required for the cells to gain complete resistance to TGF-beta growth inhibition (Fynan and Reiss, 1993; Kimchi et al., 1988; Samuel et al., 1992). Although many tumor cells are not growth inhibited by TGF-beta, they respond to TGF-beta treatment by changes in the expression of matrix components and enhanced proteolytic activity (KeskiOja et al., 1988). Agents that induce TGF-beta production in target tissues can have a chemopreventive or chemotherapeutic value for the management of epithelial malignancies. Conversely, data supporting a positive role for TGF-beta in established tumor progression are beginning to emerge (Arteaga et al., 1993a,b; Barrett-Lee et al., 1990; Arrick et al., 1992 ; E. A. Thompson et al., 1991). In later stages of tumor development, cell proliferation is often not inhibited by TGF-beta, and tumor cells secrete large amounts of this growth factor (Fynan and Reiss, 1993). In vivo TGF-beta secreted by tumor or stromal cells can influence host responses such as a natural killer cell function and thus indirctly support tumor cell viability (Arteaga et al., 1993b). TGF-beta may also affect tumor growth indirectly by stromal effects and promotion of angiogenesis. TGF-beta may also be involved in the progression of breast tumors from the steroid-sensitive to steroid-insensitive state (King et al., 1989). Understanding of the net effect of TGF-beta in different stages of tumor development is critical for the evaluation of its therapeutic value in cancer treatment.

Plasminogen activation in bullous pemphigoid immunohistology reveals urokinase type plasminogen activator, its receptor and plasminogen activator inhibitor type-2 in lesional epidermis

Keratinocytes synthesize urokinase-type plasminogen activator (uPA) and a specific cell surface receptor for uPA (uPA-R, CD 87). Plasminogen is present in plasma and interstitial fluids from where it is bound to cell surfaces via plasmin(ogen) binding sites. uPA binds to the uPA-R in an autocrine manner and activates cell-bound plasminogen: a mechanism, which provides plasmin for pericellular proteolysis. Cell-bound uPA is regulated by plasminogen activator inhibitor type-1 (PAI-1) or type-2 (PAI-2). Bullous pemphigoid is an autoimmune inflammatory skin disease characterized by subepidermal blisters. Although circumstantial evidence suggested plasminogen activation in lesional epidermis of bullous pemphigoid, immunohistological data on the type of plasminogen activators, on the uPA-receptor or the type of plasminogen activator inhibitors in the lesions of bullous pemphigoid are lacking so far. To obtain this information we have performed the present immunohistological study. The presence of uPA and its receptor as well as PAI-2 was disclosed in epidermal keratinocytes in the roof of the subepidermal blisters. Moreover, keratinocytes at the bottom of the blister, which most likely represent keratinocytes during reepithelialization were stained. Co-localization was found for uPA and its receptor, uPA and plasmin(ogen) as well as for uPA and PAI-2. In non-lesional epidermis of bullous pemphigoid only PAI-2 was found. We propose that the expression of uPA and uPA-R, as well as the upregulation of PAI-2 in keratinocytes of lesional epidermis is part of the repair and reepithelialization process following lesion formation, i.e. epidermo-dermal dyshesion, in bullous pemphigoid.

Differential growth state-dependent regulation of plasminogen activator inhibitor type-1 expression in senescent IMR-90 human diploid fibroblasts

The type-1 inhibitor of plasminogen activator (PAI-1) regulates pericellular proteolytic activity functioning, thereby to control matrix integrity, cell growth, and morphology. Subconfluent late-passage IMR-90 human fibroblasts and normal rat kidney (NRK) cells, both at the stage of replicative senescence accumulated 15- to 30-fold more undersurface PAI-1 protein compared to early-passage, actively-proliferating, cultures. Senescence-associated elevations in PAI-1 expression by IMR-90 cells reflected corresponding 11-fold increases in the 3.0- and 2.2-kb PAI-1 mRNA species. The 2.2-kb transcript exhibited a greater age-dependent increase (7.2-fold) compared to the 3.0-kb mRNA (3.7-fold). Since PAI-1 expression is coupled to growth activation in serum-deprived cultures (Ryan and Higgins, 1993, J. Cell. Physiol., 155:376-384), it was important to determine if PAI-1 gene regulation was altered as a function of cellular aging. In contrast to early-passage cultures, senescent IMR-90 fibroblasts did not down-regulate either PAI-1 protein expression or steady-state levels of PAI-1 mRNA transcripts upon serum-deprivation. Late-passage human fibroblasts at their proliferative end-stage, thus, appear to regulate PAI-1 mRNA levels through different mechanisms than do young, actively-proliferating, cells. PAI-1 overexpression during in vitro cellular aging, therefore, may contribute to the acquisition of specific senescence-associated phenotypic traits (e.g., enlarged cell morphology; increased adhesivity) by altering the pericellular proteolytic balance influencing, in turn, the formation or stability of cell-to-substrate attachment complexes.

Interaction of single-chain urokinase and plasminogen activator inhibitor type 1

Urokinase (u-PA) is synthesized and secreted as a single-chain polypeptide (single-chain u-PA, scu-PA), which has such little enzymatic activity in solution that it has been considered essentially enzymatically inert. We found that plasminogen activator inhibitor type 1 (PAI-1), the major PAI in plasma, demonstrated concentration-dependent inhibition of this solution-phase scu-PA enzymatic activity. 125I-scu-PA formed complexes with PAI-1 in a concentration- and time-dependent manner, as detected by SDS-polyacrylamide gel electrophoresis under reducing conditions. Among a given population of scu-PA molecules, all measurable enzymatic activity was inhibited by a 10-fold molar excess of PAI-1. However, at this stoichiometry, only a minority of 125I-scu-PA molecules formed SDS-stable complexes with PAI-1 (i.e. complexes that formed a covalent bond upon denaturation), even though the uncomplexed PAI-1 molecules remained competent to inhibit u-PA enzymatic activity. Neither the extent nor the time course of complex formation was altered by using PAI-1 that had been pre-incubated with native human vitronectin, compared with native PAI-1 alone. 125I-scu-PA.PAI-1 complexes that would form a covalent bond if denatured were reversible and existed in equilibrium with either non-complexed or loosely complexed reactants. These data suggest that scu-PA has more enzyme-like properties than previously appreciated and raises the possibility that it resembles single-chain tissue type-plasminogen activator in lacking a complete zymogen conformation.

Transcriptional and post-transcriptional regulation of the receptor for urokinase-type plasminogen activator by cytokines and tumour promoters in the human lung carcinoma cell line A549

The receptor for urokinase-type plasminogen activator (uPAR) is an integral membrane protein that specifically binds urokinase-type plasminogen activator (uPA) and plays a crucial role in cell surface plasmin generation. We have previously found that transforming growth factor-beta, type 1 (TGF-beta 1), increases uPAR gene transcription in the human lung carcinoma cell line A549 and now report that also epidermal growth factor (EGF) and the tumour promoter phorbol 12-myristate 13-acetate (PMA) cause increased uPAR transcription and that PMA and TGF-beta 1 in addition increase the stability of uPAR mRNA, while EGF has no effect on this parameter. All three compounds also increase the uPAR protein level, as measured by cell-binding experiments with radiolabelled ligand. The increase in uPAR protein level was however considerably lower with all three compounds than the increase in mRNA level, suggesting that they also exert a translational or post-translational control. Accompanying the increase in the number of uPAR molecules there was a proportional decrease in their ligand-binding affinity, the mechanism of which is unknown. Platelet-derived growth factor, basic fibroblast growth factor and cyclic AMP analogues did not induce any change in the uPAR mRNA level in A549 cells. Previous studies have shown that expression of uPA and its type-1 inhibitor is regulated by a variety of cytokines in a cell-specific manner. The present study indicates that cytokines in addition influence cell surface plasminogen activation by regulating uPAR expression.

Urokinase plasminogen activator expression by primary and HPV 16-transformed keratinocytes

The molecular events underlying progression of human papillomavirus (HPV) 16-associated intraepithelial neoplasia to invasive cancer have not been studied in detail. Penetration of the basement membrane is an early, but poorly understood step in this process and probably involves the action of one or more metallo- and serine proteinases. Urokinase-type plasminogen activator (uPA) is a serine proteinase that has been implicated in the pathogenesis of several epithelial tumors, but its role in HPV-associated tumors is not known. To examine uPA expression by HPV 16-transformed keratinocytes in vitro, primary foreskin keratinocyte cultures were transfected by HPV 16 DNA. The primary parental cells and the HPV 16-transformed keratinocytes were studied using substrate gel zymography, Western blot analysis and an in vitro assay measuring penetration of a Matrigel artificial basement membrane. Both uPA and its inhibitor, plasminogen activator inhibitor-1 (PAI-1), were overexpressed in the HPV 16-transformed cells relative to the parental cell line. The transformed cells, but not the parental cells, were able to degrade and penetrate the Matrigel membrane and penetration was blocked by both PAI-1 and by antibodies to uPA. Our data suggest that HPV 16-induced transformation of keratinocytes is associated with upregulation of uPA expression. In conjunction with other proteinases, uPA plays an important role in the ability of HPV 16-transformed keratinocytes to penetrate artificial basement membranes.

Regulation of urokinase-type plasminogen activator gene transcription by macrophage colony-stimulating factor

The mouse urokinase-type plasminogen activator (uPA) gene was used as a model macrophage colony-stimulating factor 1 (CSF-1)-inducible gene to investigate CSF-1 signalling pathways. Nuclear run-on analysis showed that induction of uPA mRNA by CSF-1 and phorbol myristate acetate (PMA) was at the transcriptional level in bone marrow-derived macrophages. CSF-1 and PMA synergized strongly in the induction of uPA mRNA, showing that at least some components of CSF-1 action are mediated independently of protein kinase C. Promoter targets of CSF-1 signalling were investigated with NIH 3T3 cells expressing the human CSF-1 receptor (c-fms). uPA mRNA was induced in these cells by treatment with CSF-1, and a PEA3/AP-1 element at -2.4 kb in the uPA promoter was involved in this response. Ets transcription factors can act through PEA3 sequences, and the involvement of Ets factors in the induction of uPA was confirmed by use of a dominant negative Ets-2 factor. Expression of the DNA binding domain of Ets-2 fused to the lacZ gene product prevented CSF-1-mediated induction of uPA mRNA in NIH 3T3 cells expressing the CSF-1 receptor. Examination of ets-2 mRNA expression in macrophages showed that it was also induced synergistically by CSF-1 and PMA. In the macrophage cell line RAW264, the uPA PEA3/AP-1 element mediated a response to both PMA and cotransfected Ets-2. uPA promoter constructs were induced 60- to 130-fold by Ets-2 expression, and the recombinant Ets-2 DNA binding domain was able to bind to the uPA PEA3/AP-1 element. This work is consistent with a proposed pathway for CSF-1 signalling involving sequential activation of fms, ras, and Ets factors.