- Email: [email protected]
41 The intact urokinase receptor is required for high affinity vitronectin binding: Receptor cleavage prevents ligand interaction
12
SESSION IV: Cell Adhesion, Migration, Signalling Posters
39 C O - O R D I N A T E OVEREXPRESSION OF U R O K I N A S E RECEPTOR AND t~v[]5 V I T R O N E C T I N R E C E P T O R IN BREAST CANCER. M.V. Carriero,S. Del Vecchiot, P. Franco, M. Capozzoli, L. Corvino, M.t.Potena, A. Zannetti, M.P. Stoppelli~, M. Salvatorej.lstituto NazionaleTumori, IUniversita'degli Studi "FedericoIt"; 21stitutoInternazionaledi Geneticae Bioflsica, Naples,Italy. Perturbation of adhesive interactions at cell-substratum and cell-cell contact sites is a critical event in the multistep process of cancer invasion. We have previously shown that the urokinase receptor can simultaneously bind to the amino-terminal fragment (ATF) of urokinase and to the matrix-like form of vitronectin in cultured cell lines and in breast carcinomas. This interaction occurs at physiological concentrations of the reagents and results in the formation of temary high Mr complexes (Carriero et at. Clin Cancer Res 1997, in press). To test the involvement of vitronectin receptor to a such complex array of interactions on the ceil surface, we simultaneously determined the expression of uPAR, vitronectin and vitronectin receptor ~v~3 and owl]5 in HT1080 fibmsarcoma cells, MCF-7 and MDA-MB 231 human breast carcinoma ceil lines, bearing high levels of uPAR and all capable of forming high Mr ternary complexes. Immunostaining with a panel of monoclonal antibodies showed a co-ordinate overexpression of uPAR and ctv[~5 in the cell lines tested, whereas only a faint reaction could be observed with anti-ow~3 monoclonal antibody. Then we compared the histological localization of uPAR, vitronectin and vitronectin receptors cw[~3 and ctv135 in frozen sections of 35 breast carcinomas and 8 benign breast lesions using the same panel of monoclonal antibodies. We found that uPAR-overexpressing breast cancer cells showed a strong chromogenic reaction with anti-~v135 monoclanal antibody whereas a positive immunostaining with anti-owl33 monoclonal antibody was observed only at the level of endothelial cells.On the contrary, ductai cells in benign breast lesions showed undetectable levels of uPAR and a markedly staining with both ~v[~3 and t~v~5 antibodies. Our study showed a co-ordinate overexpression of uPAR and c(v[]5 vitronectin receptor in bo[h cultured cell lines and breast carcinomas supporting the hypothesis of its cooperation to uPAR-dependent regulation of cell adhesion and iuvasiveness.
40 T H E UROKINASE-TYPE PLASMINOGEN ACTIVATOR PROMOTES PROLIFERATION OF HUMAN OVARIAN CARCINOMA CELLS Kerstin Fischer, Verena Lutz, Manfred Schmitt, Henner Graeff, *Thomas Luther, Olaf Wilhelm, Viktor Magdolea, and Ute Reuning Frauenklinik der TU Miinchen, Klinikum rechts der Isar, Munich, Germany, *Iastitut fiir Pathologie der TU Dresden, Dresden, Germany The mitogenic effect of the urokinase-type plasminogen activator (uPA) was investigated on the human ovarian carcinoma cell line OV-MZ-6 which synthesizes and secretes high concentrations of uPA and expresses considerable amounts of its receptor uPAR. Prior to cell stimulation experiments, we significantly reduced endogenous production of uPA by stable UPA "antisense" transfection of OV-MZ-6 cells. High-Molecular-Weight (HMW)-uPA independent of its catalytic activity induced an up to 90 % increase in OV-MZ-6 cell number after 96 h with maximal efficiency at about 1 nM. The amino-terminal fragment (ATF) of UPA, which lacks the catalytic domain but encompasses the uPAR binding region of uPA, exhibited a comparable mitogenic activity upon stimulation of OV-MZ-6 ceils compared to HMW-uPA. The same observation was made using the synthetic uPA-derived peptide cyclo'9'31uPAlg.3t spanning the uPAR binding region. In contrast, catalytically active Low-Molecular-Weight(LMW)-uPA, devoid of the uPAR binding region, did not provoke growth stimulation. A moneclonal antibody directed against the uPA binding region of uPAR (IIIFI0) significantly suppressed uPA-dependent enhancement of cell proliferation as did recombinant soluble uPAR as a "scavenger" for uPA. OVMZ-6 cells pretreated with phosphoinositol (PI)-specific pbospholipase C, known to release glycosylphosphatidyl-iansitol (GPI)-anchored proteins, such as uPAR, from cell surfaces, did not longer respond to uPA stimulation. The uPARdeficient lymphoblast-like cell line Raji served as control. Raji ceils could not be stimulated by uPA, but gained responsiveness upon stable transfection with fulllength uPAR cDNA. The present study clearly demonstrates that binding of uPA to uPAR causes a significant growth promotion of OV-MZ-6 cells, an effect independent of the proteolytic activity of uPA.
41
42
THE INTACT UROKINASE RECEPTOR IS REQUIRED FOR HIGH AFFINITY VITRONECT1N BINDING: RECEPTOR CLEAVAGE PREVENTS LIGAND INTERACTION. ~ a a i l ] ~ , Niels Behrendtt, Michael Ploug', Keld Dane I and Klaus T. Preissner2. ~Finsen Laboratory, Rigshospitalet, Strandboulevarden 49, DK-2100 Copenhagen O., Denmark. 2Max-Planek-Institut, Haemostasis Research Unit, Kerekhoff-Klinik, D-61231 Bad Nauhnim, Germany
BEYOND PROTEOLYSIS: EVIDENCE FOR A NONPROTEOLYTIC ROLE FOR uPAR 1N CELLULAR MIGRATION 1N VIVO
The urokinase receptor (uPAR) is a high affinity receptor for both urokinase plasminogen activator (uPA) and the adhesion protein vitronectin and is a central component in peficelhilar proteolysis and cellular adhesion. Two forms of cell surface bound uPAR are present on many cell types; the full length uPAR and a cleaved form, UPAR(2+3). On enitured U937 cells UPAR(2+3) is formed by uPA catalyzed cleavage of uPAR. In ligand-blotting experimants using uPAR purified from U937 cell lysatus we found, that vitronectin and uPA bind uPAR but not uPAR(2+3). In order to study various aspects of vitronectin binding to intact and cleaved forms of UPAR we employed BIA technology and recombinant, soluble UPAR (suPAR) preparations. Both plasma and multimeric forms of vilronectin bound to intact, antibody-immobilized suPAR and uPA augmented this binding. Plasminogan activator inhibitor 1 (PAl- 1) inhibited suPAR binding to vitroncetin in a dose-dependent manner with maximal inhibition at equimolar concentrations of vitronectin and PAl-1. Monoclonal antibodies against domain 1 of uPAR blocked suPAR binding to vitroneetin and vitronectin did not interact with suPAR(2+3). Both suPAR(2+3) and the isolated domain 1 failed to compete with the intact suPAR in binding to vitronectin, indicating that the inteet receptor is required for high affinity vitroneetin binding. Thus, cleavage of uPAR prevents cell surface potentiation of plasminogen activation as well as uPAR-dependent vitronectin binding with consequences for cellular adhesion.
David A. Waltz Ross M. Fujita, Lisa Natkin, and Harold A. Chapman, Children's Hospital and Brigham & Women's Hospital, Boston, MA, USA Introduction: PericeUular proteolysis mediated by urokinase (uPA) and its receptor (uPAR) have been implicated in cellular migration. Reduced tumor cell metastasis in response to nonproteolytic receptor-binding amino-terminal fragments (ATF) of uPA has thus been ascribed to displacement of active uPA. We have shown that uPAR also functions as a cellular receptor for extraoellular matrix vitronectin, and that occupancy by ATF induces cellular adhesion to and retards migration acmss vitronectin in vitro. We sought to determine 1) the effect of uPAR occupancy on cellular migration in vivo, and 2) the relevance of uPAR-mediated proteolysis to migration utilizing mice lacking functional uPA. Methods: Recombinant murine KC (1.5 p.g), a potent neutrophil chemoattractant, was delivered to the ltmgs of uPA +/+ or uPA -/- C57b16 mice via nasal insufflation 18h after intraperitoneal (IP) injection of 200 I.tg human IgG/murine ATF fusion protein or human IgG (control). Whole lung lavage was performed 4h later and total and differential white blood cell counts obtained. Results: Expressed as mean (SD) of 4-6 animals per treatment, mouse: +/+ -/. +/+ -/+/+ -H-/-/KC: + + + + + + IP: IgG ATF IgG ATF lavage wbc x 10-6: 0.7 0.6 6.4 4.6 8.5 3.7* 6.0 2.5* (0.2) (0.3) (2.4) (2.9) (2.3) (0.8) (1.7) (0.8) In the absence of KC, lavage wbc were 99% maerophages; KC treatment resulted in 85-97% neutrophils. * = p < 0.01 vs. IgG control. Conclusions: The reduction in wbc influx induced by ATF in mice lacking functional uPA suggests that the effect of ATF is not due to disruption of uPARmediated proteolysis. Possible explanations for these observations include altered cellular signalling or adhesiveness. These data provide in vivo evidence of a function of uPAR distinct from the regulation of pericelhilar proteolysis.
SESSION IV: Cell Adhesion, Migration, Signalling Posters
39 C O - O R D I N A T E OVEREXPRESSION OF U R O K I N A S E RECEPTOR AND t~v[]5 V I T R O N E C T I N R E C E P T O R IN BREAST CANCER. M.V. Carriero,S. Del Vecchiot, P. Franco, M. Capozzoli, L. Corvino, M.t.Potena, A. Zannetti, M.P. Stoppelli~, M. Salvatorej.lstituto NazionaleTumori, IUniversita'degli Studi "FedericoIt"; 21stitutoInternazionaledi Geneticae Bioflsica, Naples,Italy. Perturbation of adhesive interactions at cell-substratum and cell-cell contact sites is a critical event in the multistep process of cancer invasion. We have previously shown that the urokinase receptor can simultaneously bind to the amino-terminal fragment (ATF) of urokinase and to the matrix-like form of vitronectin in cultured cell lines and in breast carcinomas. This interaction occurs at physiological concentrations of the reagents and results in the formation of temary high Mr complexes (Carriero et at. Clin Cancer Res 1997, in press). To test the involvement of vitronectin receptor to a such complex array of interactions on the ceil surface, we simultaneously determined the expression of uPAR, vitronectin and vitronectin receptor ~v~3 and owl]5 in HT1080 fibmsarcoma cells, MCF-7 and MDA-MB 231 human breast carcinoma ceil lines, bearing high levels of uPAR and all capable of forming high Mr ternary complexes. Immunostaining with a panel of monoclonal antibodies showed a co-ordinate overexpression of uPAR and ctv[~5 in the cell lines tested, whereas only a faint reaction could be observed with anti-ow~3 monoclonal antibody. Then we compared the histological localization of uPAR, vitronectin and vitronectin receptors cw[~3 and ctv135 in frozen sections of 35 breast carcinomas and 8 benign breast lesions using the same panel of monoclonal antibodies. We found that uPAR-overexpressing breast cancer cells showed a strong chromogenic reaction with anti-~v135 monoclanal antibody whereas a positive immunostaining with anti-owl33 monoclonal antibody was observed only at the level of endothelial cells.On the contrary, ductai cells in benign breast lesions showed undetectable levels of uPAR and a markedly staining with both ~v[~3 and t~v~5 antibodies. Our study showed a co-ordinate overexpression of uPAR and c(v[]5 vitronectin receptor in bo[h cultured cell lines and breast carcinomas supporting the hypothesis of its cooperation to uPAR-dependent regulation of cell adhesion and iuvasiveness.
40 T H E UROKINASE-TYPE PLASMINOGEN ACTIVATOR PROMOTES PROLIFERATION OF HUMAN OVARIAN CARCINOMA CELLS Kerstin Fischer, Verena Lutz, Manfred Schmitt, Henner Graeff, *Thomas Luther, Olaf Wilhelm, Viktor Magdolea, and Ute Reuning Frauenklinik der TU Miinchen, Klinikum rechts der Isar, Munich, Germany, *Iastitut fiir Pathologie der TU Dresden, Dresden, Germany The mitogenic effect of the urokinase-type plasminogen activator (uPA) was investigated on the human ovarian carcinoma cell line OV-MZ-6 which synthesizes and secretes high concentrations of uPA and expresses considerable amounts of its receptor uPAR. Prior to cell stimulation experiments, we significantly reduced endogenous production of uPA by stable UPA "antisense" transfection of OV-MZ-6 cells. High-Molecular-Weight (HMW)-uPA independent of its catalytic activity induced an up to 90 % increase in OV-MZ-6 cell number after 96 h with maximal efficiency at about 1 nM. The amino-terminal fragment (ATF) of UPA, which lacks the catalytic domain but encompasses the uPAR binding region of uPA, exhibited a comparable mitogenic activity upon stimulation of OV-MZ-6 ceils compared to HMW-uPA. The same observation was made using the synthetic uPA-derived peptide cyclo'9'31uPAlg.3t spanning the uPAR binding region. In contrast, catalytically active Low-Molecular-Weight(LMW)-uPA, devoid of the uPAR binding region, did not provoke growth stimulation. A moneclonal antibody directed against the uPA binding region of uPAR (IIIFI0) significantly suppressed uPA-dependent enhancement of cell proliferation as did recombinant soluble uPAR as a "scavenger" for uPA. OVMZ-6 cells pretreated with phosphoinositol (PI)-specific pbospholipase C, known to release glycosylphosphatidyl-iansitol (GPI)-anchored proteins, such as uPAR, from cell surfaces, did not longer respond to uPA stimulation. The uPARdeficient lymphoblast-like cell line Raji served as control. Raji ceils could not be stimulated by uPA, but gained responsiveness upon stable transfection with fulllength uPAR cDNA. The present study clearly demonstrates that binding of uPA to uPAR causes a significant growth promotion of OV-MZ-6 cells, an effect independent of the proteolytic activity of uPA.
41
42
THE INTACT UROKINASE RECEPTOR IS REQUIRED FOR HIGH AFFINITY VITRONECT1N BINDING: RECEPTOR CLEAVAGE PREVENTS LIGAND INTERACTION. ~ a a i l ] ~ , Niels Behrendtt, Michael Ploug', Keld Dane I and Klaus T. Preissner2. ~Finsen Laboratory, Rigshospitalet, Strandboulevarden 49, DK-2100 Copenhagen O., Denmark. 2Max-Planek-Institut, Haemostasis Research Unit, Kerekhoff-Klinik, D-61231 Bad Nauhnim, Germany
BEYOND PROTEOLYSIS: EVIDENCE FOR A NONPROTEOLYTIC ROLE FOR uPAR 1N CELLULAR MIGRATION 1N VIVO
The urokinase receptor (uPAR) is a high affinity receptor for both urokinase plasminogen activator (uPA) and the adhesion protein vitronectin and is a central component in peficelhilar proteolysis and cellular adhesion. Two forms of cell surface bound uPAR are present on many cell types; the full length uPAR and a cleaved form, UPAR(2+3). On enitured U937 cells UPAR(2+3) is formed by uPA catalyzed cleavage of uPAR. In ligand-blotting experimants using uPAR purified from U937 cell lysatus we found, that vitronectin and uPA bind uPAR but not uPAR(2+3). In order to study various aspects of vitronectin binding to intact and cleaved forms of UPAR we employed BIA technology and recombinant, soluble UPAR (suPAR) preparations. Both plasma and multimeric forms of vilronectin bound to intact, antibody-immobilized suPAR and uPA augmented this binding. Plasminogan activator inhibitor 1 (PAl- 1) inhibited suPAR binding to vitroncetin in a dose-dependent manner with maximal inhibition at equimolar concentrations of vitronectin and PAl-1. Monoclonal antibodies against domain 1 of uPAR blocked suPAR binding to vitroneetin and vitronectin did not interact with suPAR(2+3). Both suPAR(2+3) and the isolated domain 1 failed to compete with the intact suPAR in binding to vitronectin, indicating that the inteet receptor is required for high affinity vitroneetin binding. Thus, cleavage of uPAR prevents cell surface potentiation of plasminogen activation as well as uPAR-dependent vitronectin binding with consequences for cellular adhesion.
David A. Waltz Ross M. Fujita, Lisa Natkin, and Harold A. Chapman, Children's Hospital and Brigham & Women's Hospital, Boston, MA, USA Introduction: PericeUular proteolysis mediated by urokinase (uPA) and its receptor (uPAR) have been implicated in cellular migration. Reduced tumor cell metastasis in response to nonproteolytic receptor-binding amino-terminal fragments (ATF) of uPA has thus been ascribed to displacement of active uPA. We have shown that uPAR also functions as a cellular receptor for extraoellular matrix vitronectin, and that occupancy by ATF induces cellular adhesion to and retards migration acmss vitronectin in vitro. We sought to determine 1) the effect of uPAR occupancy on cellular migration in vivo, and 2) the relevance of uPAR-mediated proteolysis to migration utilizing mice lacking functional uPA. Methods: Recombinant murine KC (1.5 p.g), a potent neutrophil chemoattractant, was delivered to the ltmgs of uPA +/+ or uPA -/- C57b16 mice via nasal insufflation 18h after intraperitoneal (IP) injection of 200 I.tg human IgG/murine ATF fusion protein or human IgG (control). Whole lung lavage was performed 4h later and total and differential white blood cell counts obtained. Results: Expressed as mean (SD) of 4-6 animals per treatment, mouse: +/+ -/. +/+ -/+/+ -H-/-/KC: + + + + + + IP: IgG ATF IgG ATF lavage wbc x 10-6: 0.7 0.6 6.4 4.6 8.5 3.7* 6.0 2.5* (0.2) (0.3) (2.4) (2.9) (2.3) (0.8) (1.7) (0.8) In the absence of KC, lavage wbc were 99% maerophages; KC treatment resulted in 85-97% neutrophils. * = p < 0.01 vs. IgG control. Conclusions: The reduction in wbc influx induced by ATF in mice lacking functional uPA suggests that the effect of ATF is not due to disruption of uPARmediated proteolysis. Possible explanations for these observations include altered cellular signalling or adhesiveness. These data provide in vivo evidence of a function of uPAR distinct from the regulation of pericelhilar proteolysis.