human urokinase receptor concentration in malignant and ... · siveness and proliferation,...

[CANCER RESEARCH 53. 3198-3206, July I, 1993]

Human Urokinase Receptor Concentration in Malignant and Benign Breast Tumorsby in Vitro Quantitative Autoradiography: Comparison with Urokinase Levels1

Silvana Del Vecchio,2 M. Patrizia Stoppelli, Maria V. Carriero, Rosa Fonti, Omelia Massa, PeÃYong Li,Gerardo Botti, MarÃaCerra, Giuseppe D'Aiuto, Giuseppe Esposito, and Marco Salvatore

Medicina Nucleare. UniversitÃ degli Studi "Federico II," FacoltÃ de Medicina, Via S. Pansini 5, 80301 Naples [S. D. V., M. S./: Istituto Nazionale Tumori, Naples Â¡M. V. C.,

K. F., P. K L. G. B.. M. C, G. D., G. E., M. S.]: and Istituto Intemazionale di Genetica e Biofisica, Naples, Italy Â¡M.P. S.. O. U.]

ABSTRACT

We measured the tissue concentration of human urokinase receptor(uPAR) in 22 breast carcinomas and 9 benign breast lesions using in vitroquantitative autoradiography. Tissue sections were incubated with increasing concentrations of 125I-pro-urokinase in the presence or absence of

unlabeled competitor. Breast carcinomas were found to contain 5 timesmore uPAR than benign breast lesions with respect to their protein content[523 Â±72 versus 108 Â±20 (SE) fmol/mg (P < 0.001)]. Simultaneous

quantitation of urokinase (uPA) by immunoenzymatic assay on tissueextracts from the same specimens showed that breast carcinomas alsocontain 19 times more uPA than benign tumors (611 Â±134 versus 32 Â±8fmol/mg) l /' < 0.01). The reliability of quantitative autoradiography meas

urements was confirmed by uPAR cross-linking assay on membrane frac

tion from either U937 histiocytic lymphoma cells or breast carcinomas andimmunoperoxidase staining with an anti-uPAR antibody on tumor sections. Also, immunoperoxidase staining with an anti-uPA monoclonal an

tibody showed that uPA is indeed localized on the plasma membrane ofepithelial tumor cells in confined areas of breast carcinomas whereas cellsfrom benign breast lesions were devoid of uPA under the same experimental conditions. In conclusion, our findings support the hypothesis thatuPAR plays a central role in the acquisition of an invasive phenotype andfavor its potential use as a prognostic factor in patients with breast carcinoma.

INTRODUCTION

uPA3 is one of the two plasminogen activators which convert plas-

minogen to plasmili, a trypsin-like enzyme of broad specificity. An

overexpression of uPA has been reported for many malignant tumorsincluding lung, breast, and colon cancer (1-3). Because uPA can

degrade directly or indirectly all components of the extracellular matrix, it has been hypothesized that this enzyme plays an important rolein tumor invasion and metastasis. Previous studies have shown thatantibodies specific for urokinase can inhibit metastatic disseminationof tumor cells in animal models (4, 5). Furthermore, in vitro invasionassays on cultured human tumor cells demonstrated that uPA activityis essential for the invasive phenotype of these cells (6, 7).

Urokinase is secreted as an inactive single chain protein (pro-urokinase or pro-uPA) of 411 amino acids (8). A proteolytic cleavage

generates the active, two chain form of uPA (9) which also reacts withplasminogen activator inhibitors (10). The expression of urokinase incultured tumor cells is modulated by effectors that are also able topromote cell proliferation, including hormones (11), and growth factors, such as epidermal growth factor (12), fibroblast growth factor(13), transforming growth factor ÃŸ(14), or serum (15). Furthermoreurokinase itself was found to elicit a direct mitogenic effect in certainhuman tumor cell lines (16) and to activate latent growth factorsdirectly (17) or through a plasmin-mediated mechanism (14).

Received 1/21/93; accepted 4/27/93.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1This work was supported by Associazione Italiana Ricerca Cancro and grants from

Consiglio Nazionale delle Ricerche. P.P. A.C.R.O. and P.P. B.T.B.S.2 To whom requests for reprints should be addressed.' The abbreviation used are: uPA, urokinase plasminogen activator; uPAR, urokinase

plasminogen activator receptor; ATF, amino-terminal fragment; PBS. phosphate-buffered

saline; MoAb, monoclonal antibody.

A specific cell surface receptor for uPA (uPAR) has been describedon human peripheral blood monocytes and the monocyte-like U937cells (18, 19). uPA and pro-uPA bind with the same affinity to uPAR(Kd range, 0.1-2 niu) (20). The receptor binding site has been shownto be located in the MT 17,000 ATF in the so-called "growth factor-like" domain of uPA ( 19, 20). High affinity binding sites for uPA have

been described on the surface of several cultured tumor cells (21) andin membranes from human breast carcinoma (22). An autocrine saturation of uPA receptors occurs in cultured tumor cells overexpressinguPA (23) which may serve to focus the proteolytic activity of uPA atthe cell membrane thus enhancing uPA-mediated invasion. Ossowski

et al. (24) showed that the mere production of high levels of uPA inthe absence of surface receptors was inadequate in endowing a tumorcell with maximal invasive potential. Also, extracellular matrix degradation and invasion by cultured tumor cells have been shown to bedependent on the number and fractional occupancy of surface uPAreceptors (25, 26). In agreement with these findings several authorsshowed an enhancement of plasminogen activation catalyzed by receptor-bound uPA as compared to uPA in solution (27, 28).

To test whether uPAR overexpression is part of the invasive machinery of tumor cells, we determined the intratumoral distributionand concentration of urokinase receptors in histological sections ofhuman breast carcinomas and benign breast lesions using in vitroquantitative autoradiography. This technique was chosen because itpreserves tissue anatomy and provides reliable quantitative data. Inaddition, both the distribution and local concentration of urokinasereceptor can be related to the structural components of the tumor,including viable tumor cells, stroma, vascular structures, leading edgeof the tumor. Moreover, to determine the uPAR/uPA molar ratio withrespect to some biological characteristics of the tumor, such as inva-siveness and proliferation, urokinase and pro-urokinase concentration

was assessed on tumor extracts obtained from the same specimensanalyzed by quantitative autoradiography. The tissue concentrationsof uPA and uPAR were then compared with the previously establishedprognostic factors of breast carcinomas, such as tumor size, lymphnode involvement, histology, hormone receptor status, and other factors, such as the rate of proliferation and neovascularization.

MATERIALS AND METHODS

Cell Culture. U937 histiocytic lymphoma cells were grown in RPMI 1640supplemented with 10% heat-inactivated fetal calf serum at 37Â°C,5% CO2.HT1080 fibrosarcoma cells were grown in Dulbecco's modified Eagle's me

dium-10% fetal bovine serum.12SI-Labeling Procedure and Quality Control. Human recombinant pro-

urokinase and the recombinant amino-terminal fragment of human urokinasewere gifts of Dr. F. Blasi, Milan, Italy. Pro-uPA was radioiodinated with 125I

(Sorin Biomedica, Saluggia, Italy) using the lodo-Gen method (29). One nmolof protein was reacted with 800 Â¿Â¿Ciof Na'25I and 12 fig of lodo-Gen (Pierce,

Rockford, IL). After 10 min the reaction was stopped by the addition of l Â¿Â¿molof /V-acetyltyrosine (Sigma Chemical Co, St. Louis, MO). The radiolabeledprotein was purified from unbound iodide by Sephadex G-25 (Pharmacia,Uppsala, Sweden) chromatography. The mean percentage of 125Iincorporation

was 75%. The protein concentration of the radiolabeled product was determined by an enzyme immunoassay (American Diagnostica, Inc., Greenwhich,CT) in which recombinant pro-uPA is used as standard. The specific activity of

3198

Research. on January 2, 2020. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

HUMAN IKOKINASH RECEPTOR CONCI.MKAIION

the product was 14.07 Â±3.18 (SE) n.Ci//ig. T"6 chemical integrity of theradiolabeled pro-uPA was assessed by electrophoresis on a 12.5% sodiumdodecyl sulfate-polyacrylamide gel (30). The functional integrity of the radi

olabeled protein was assessed by a competitive binding assay on HT1080fibrosarcoma cells as described previously (23). More than 90% of displacement was obtained in all cases using 0.04 JAMcold recombinant pro-uPA and

0.4 JIMconcentrations of two chain uPA from human kidney cells (Sigma). Thesame labeling conditions were applied to ATF when used.

Cross-Linking of the uPA Receptor. Extensively ground samples of tumor

tissue and U937 cell pellet were homogenized in a hypotonie solution [20 mmTris-HCI (pH 7.5)-1 ITIMEDTA-1 nun ethyleneglycol bis(ÃŸ-aminoethyl ether)-N,N,N',N'-telraacetic acid-1 HIMphenylmethylsulfonyl fluoride-5 JIM leupep-

tin-0.025 TlU/ml aprotinin-25 imi benzamidine]. The lysates were cleared by

low speed centrifugation and then subjected to centrifugation at 100,000 x g.The resulting pellet (membrane fraction) was resuspended in 50 HIM4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid. pH 7.4. and stored at -80Â°C.Membrane fractions from cells or tissue were incubated with 125I-ATF

(200,000 cpm) with or without unlabeled competitor and cross-linked with 1

m.Mdisuccinimidyl suberate (Pierce) under conditions described by Nielsen el

al. (31 ). The membranes were then washed twice with an acidic buffer (23) andrecovered by centrifugation at 12,000 rpm.

Patients and Tissue Samples. Thirty-one patients undergoing surgery fora breast lump were studied. Twenty-two had a histologically confirmed breast

carcinoma and 9 had benign breast lesions (6 fibroadenomas. 3 fibrocysticdisease). These tumors were classified according to the WHO nomenclature(32) and were staged using the tumor-nodes-metastasis system. Nuclear grad

ing was used to define the degree of tumor differentiation (Gl, well differen

tiated: G2. intermediate; G3, anaplasia) according to the classification of Blackand Speer (33). The mean age of patients with breast carcinomas was 53.4 Â±10.6 years whereas patients with benign lesions had a mean age of 34 Â±9.8years. No patient had received previous systemic therapy or preoperative local

radiotherapy.Tumor biopsy specimens were immediately frozen in liquid nitrogen after

surgical removal and stored at -80Â°C until studied. From these specimens,

consecutive 8-fim frozen sections were cut in a cryomicrotome correspondingto the largest cross-sectional area of the tumor including the leading edge of thelesion. The sections were washed in an acidic buffer (50 mvi glycine-100 m.M

NaCI, pH 3) to remove endogenously bound urokinase. They were then fixedin a solution of 0.25% glutaraldehyde (Sigma) in PBS, pH 7.4. for 20 min andthen washed in cold PBS for another 20 min. For histological comparison,some sections were stained with hematoxylin and eosin. Several 5-fj.m adjacent

sections were cut for immunoperoxidase staining.Tumor specimens were also processed for hormone receptors binding assay

and uPA determination.Saturation Study. Cells were harvested from suspension cultures by cen

trifugation and washed twice in PBS. The cell pellet was included in 20%gelatin (Sigma) and then frozen in liquid nitrogen to form a cell stick (34).Eight-/im frozen sections were then cut in a cryomicrotome and processed as

the tissue sections.The tissue sections used to determine the extent of nonspecific binding were

preincubated in a humidity chamber (30 min 22Â°C)with 75 fil of a 20 JÃŒM

solution of unlabeled urokinase (Sigma) and then washed in cold PBS toremove excess unlabeled ligand. All sections were incubated (30 min 22Â°C)

with PBS containing 2% bovine serum albumin and 10% chicken serum,washed 5 times with cold PBS, and finally incubated in the presence of 0.1 to4 HMI2'I labeled pro-uPA for 2 h at 22Â°C.This incubation was performed in

the presence and absence of 0.2 /AMunlabeled urokinase (Sigma) and wasfollowed by several washes in cold PBS (30 min) and dehydration in 70 and100% ethanol.

At the end of the saturation assay the chemical integrity of the radiolabeledpro-uPA in solution was checked again by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. No conversion of single chain pro-uPA to two chain

uPA was observed.Quantitative Autoradiography. Quantitative autoradiographic techniques

and preparation of autoradiographic standards are described elsewhere (34,35). Briefly, serial dilution of I2lsl-labeled human serum albumin were prepared

and added to 20% gelatin (Sigma). To determine the amount of radioactivityassociated to the standards, aliquots of the gelatin suspension were weighed

and counted in a gamma counter. The remainder of each suspension was frozenin liquid nitrogen. Eight-jxm sections were cut from each frozen gelatin stick,

mounted on glass coverslip, and used as autoradiographic standards. The tissuesections, along with 125Istandards, were placed in a film cassette with Kodak

SB5 X-ray film. After 2 days of exposure, the film was photographically

processed and the resulting images were digitized using an image analysissystem (Sistemi Avanzati, Milan, Italy) including a high resolution CCD camera (High Technology Holland) and the Micro Computer Imaging Device

(Imaging Research, Inc., Ontario, Canada). The absorbance measurementsfrom autoradiograms were used to reconstruct images in a 512 x 512-pixelmatrix. The absorbance readings over the I2Ã•Istandards were plotted against

the respective fiCi/g in each standards. A polynomial fitting of these dataprovided a standard curve. By comparing the autoradiographic images to theimages of adjacent sections stained with hematoxylin and eosin, tumor areas

were defined and regions of interest were drawn over these areas. The meanabsorbance values of the thus selected regions were obtained and the (Â¿Ci/gof

tumor were determined from the standard curve. Using the specific activity ofeach '"I-pro-uPA preparation, the fÂ¿Ci/gvalues were convened to pmol of

ligand bound/g of tissue.For each tumor the nonspecific binding curve was subtracted from the total

binding curve so that the specific binding curve was generated. In all cases, the

specific binding curve reached a plateau corresponding to the maximal amountof l25I-pro-uPA that can specifically bind to tumor or ÃŸmaâ€ž.The dissociation

constant (K,Â¡)of the pro-uPA/uPAR interaction was calculated as the concentration of l25I-pro-uPA that reduces the Sma, by 50%.

Tissue Extraction and uPA Determination. In order to quantitute urokinase and pro-urokinase contained in each tumor biopsy specimen, 200-300 mg

(wet weight) of tumor were cut in small pieces and pulverized in a Mikro-

Dismembrator (Braun. Melsungen, Germany) set to 1 min at maximal power.The resulting powder was suspended in 1:10 (w/v) extraction buffer (36)containing 0.075 M potassium acetate, 0.3 M NaCI, 0.1 M L-arginine, IO imiEDTA. and 0.25% Triton X-100 (pH 4.2) and homogenized for 5 min in ice

cold Potter tubes. Aliquots of crude homogenate were taken for determinationof total protein content. Ultracentrifugation was then performed at 100,000 xg for 45 min at 4Â°C.Supernatants were collected, divided in aliquots, and keptat -80Â°C.

Determination of total protein content was performed on 20-/nl aliquots of

crude homogenate by adding 200 /j.1of a solution of 0.2 NNaOH. After heatingto 85Â°Cfor IO min and neutralizing with HCI, samples were assayed for

protein content by the method of Bradford (37).A commercially available enzyme-linked immunosorbent assay kit (Amer

ican Diagnostica) was used for determination of tissue-associated uPA levels.

Supernatants were diluted from 1:10 to 1:50 with a dilution buffer and assayed.

The results were expressed as pmol/g of wet tissue and tmol/mg of protein.Estrogen and progesterone receptors were measured using the dextran

coated technique according to the European Organization for Research andTreatment of Cancer recommended methods (38).

Immunoperoxidase Staining. Immunoperoxidase staining was performedusing MoAb 3936 (American Diagnostica), an IgG2a that binds to soluble andmembrane-associated uPAR (39). Tumor sections were processed with a slightmodification of previously described standard procedure (40) using 3-amino-9-ethylcarbazole as chromogen. They were given a preliminary wash for 3 min

with an acidic buffer to remove endogenously bound uPA and then incubatedovernight at 4Â°Cwith MoAb 3936 (20 /ig/ml). To demonstrate the specificity

of uPAR staining. MoAb 3936 was preadsorbed with recombinant nativesoluble form of uPAR from concentrated conditioned medium of LB6 recombinant native soluble uPAR-producing clone generously provided by Dr. F.

Blasi and described previously (41 ). Under these conditions a strong reductionof the staining was obtained.

The intratumoral localization of uPA was assessed by MoAb 394 (AmericanDiagnostica) which is specific for the B chain of human uPA and recognizes allknown forms of human uPA including receptor bound uPA. Tumor sectionswere incubated overnight at 4Â°Cwith MoAb 394 at 20 ng/ml. The intensity of

the chromogenic reaction was graded from 0 to 4 plus.The rate of proliferation of tumor cells was evaluated using the monoclonal

antibody KÃŒ67(Dako, Copenhagen, Denmark) directed against a nuclear antigen present in cells which are in G,, S, G;, or M phases of the cell cycle.Quiescent G0 cells are not recognized by this antibody (42). The sections wereexamined by light microscopy at x 400. Each section was divided in 4 to 10regions; a minimum of 100 tumor cells were counted in every region excluding

degenerated and necrotic areas. The results were expressed as the mean percentage of positively stained tumor cells in a section.

3199



HUMAN UROKINASE RECEPrOR CONCENTRATION

30

234nanomoles/liter B

Fig. I. Representative binding curves obtained from the saturation assay performed onfrozen sections of U937 cells (A) and a breast carcinoma (ÃŸ)from patient 19. Abscissa,125I-Pro-uPA concentration; ordinÃ¤res,bound ' 25l-pro-uPA. Total binding of l25I-pro-uPA(â€¢).nonspecific binding of l25I-pro-uPA (O). specific binding of 12M-pro-uPA (*).

Nonspecific binding was determined using excess unlabeled uPA. The specific bindingcurve was obtained by subtracting the nonspecific binding curve from the total bindingcurve. Note that the binding curves obtained from breast carcinoma were comparable tothose obtained from U937 cells.

The number and distribution of intratumoral blood microvessels were assessed by staining endothelial cells for factor VIII using the monoclonalantibody F8/86 (Dako). Each section was divided in 4 to 10 regions. Eachregion was scanned by light microscopy at low power in order to identify theareas with the highest microvessel density. Individual microvessel in each areawere counted on a x 400 field. The results were expressed as the mean number

of microvessels per field in a section.Statistics. Student's i test and simple regression analysis were used as

appropriate (43).

RESULTS

Saturation Study on Cultured Tumor Cells. In order to optimizethe conditions for the saturation assay followed by quantitative auto-

radiography on tissue sections, we used U937 histiocytic lymphomacells bearing a known number of uPA receptors. This cell line has beenreported to express 10,000 to 20,000 receptors/cell with a K,, ranging

between 0.18 and 1 nM(28, 44). Frozen sections from U937 cell stickwere incubated with increasing concentrations of l25I-pro-uPA in

presence or absence of unlabeled urokinase. The cell sections werethen washed and autoradiographed to quantitate the radioactivity associated to each sample. Fig. \A shows representative binding curvesobtained from the saturation study on cultured tumor cells. The ÃŸmaxvalues ranged between 2.12 and 6.58 pmol/g using different cellinclusions and different 125I-pro-uPA preparations. Since the number

of tumor cells per g of suspension was known, the number of receptorbinding sites per cell could be calculated and was 9968 Â±1954. Theseresults were consistent with the previously reported value of the sameparameters determined with different binding techniques (28, 44).Similar results were obtained when radiolabeled recombinant ATFwas used in the saturation assay.

Biochemical Identification of uPAR. To test whether highly malignant breast carcinoma expresses receptors for uPA, we preparedmembrane fractions from U937 cells and tumor which we incubatedwith radiolabeled amino-terminal fragment of human urokinase.Cross-linking products of bound '-AI-ATF to receptor protein with the

bifunctional amino group reactive cross-linking reagent disuccinim-

idyl suberate were analyzed onto a 12.5% gel electrophoresis underreducing conditions (Fig. 2). As for U937 membranes, an approximately M, 70,000 band from the cross-linked membranes of breast

carcinoma resulted which was completely competed for by excessunlabeled rec-pro-uPA, two chain uPA. and ATF (Fig. 2). Differences

in glycosylation may account for the slightly increased electrophoreticmobility of the ATF-uPAR complex from the tumor with respect to theU937 cells. These findings indicate that the amino-terminal fragment

of urokinase binds to a membrane protein exhibiting binding properties and electrophoretic mobility similar to those of uPA receptor.

Quantitation of Tumor-associated uPAR and uPA. Table 1 sum

marizes the clinical data, pathological findings, and receptor status ofpatients with breast cancer. All patients underwent radical surgeryexcept patient 6 who could not undergo axillary node dissectionbecause of impaired left ventricular function.

Frozen sections from tumor biopsy specimens were processed in thesaturation study and bound radioactivity was measured by quantitativeautoradiography. Fig. \B shows representative binding curves obtained from the assay performed on a breast carcinoma (patient 19).These binding curves were comparable to those obtained from thestudy performed on frozen sections of U937 cells. Endogenous uPAlevels were also determined by enzyme-linked immunosorbent assay

on tumor extracts from the same specimens.Tables 2 and 3 report the tissue concentration of uPAR determined

by quantitative autoradiography and uPA levels measured by immu-

noenzymatic assay in breast carcinomas and benign lesions, respectively. The results were normalized for both tumor weight (pmol/g)and total protein content (fmol/mg). In breast carcinomas the 5max

Fig. 2. Cross-linking of radiolabeled ATF louPAR in extracts of U937 cells and breast carcinoma. Membrane fractions from U937 cells (40fig; Lanes 5 and 6) and breast carcinoma ( 120 fig;Lanes 1-4) were incubated in 50 m.w 4-(2-hydrox-yethyl)-l-pÂ¡perazineethanesulfonic acid. pH 7.5,with 12*I-ATF in the absence (Lanes I and 5) andin the presence of 0.1 fiM unlabeled pro-uPA (Lanes2 and 6) or two chain uPA (Lane 3) or ATF (Lane4 ). After the incubation, the extracts were treatedwith disuccinimidyl suberate, dissolved with loading buffer and analyzed by a 12.5% polyacrylamidegel under reducing conditions followed by autoradiography. Arrow, M, 68,000 complex betweenATF and uPAR from tumor tissue; right ordinate,position of the molecular weight marker proteins.K, thousands.

46 K

3200



HUMAN UROKINASE RECEPTOR CONCENTRATION

Table 1 Clinical dala, pathological findings, and hormone receptor status of patients with breast cancer

Patient12345678910AgePathological

(yrs) Histologystage59

LobularT3N,Mâ€ž59DuelaT2N,Mâ€ž60

LobularT,N,Mâ€ž66DuctalT2N,Mâ€ž61LobularT,N,M069LobularT2NXM(147MucinousT2NoM048LobularTiNoM039DuctalT|N(|M062LobularT2Nâ€žMâ€žII

52 DuctalT|Nâ€žMâ€ž12131415161718192021TT36DuctalT,N(,M(>51

Duct59Duci40Duct56Duela51DuelT,Nâ€žMâ€žT,N,Mâ€žT,N,Mâ€žT2N0Mâ€žT2N,Mâ€ž38

DuctalT,N,,M045LobularT,Nâ€žMâ€ž71DuclalT2N(IM041DuctalT2N,M065Duclal TÃ•N, M,,Nuclear

grade2312121221121233232232Size(cm)>53.82.3422.22.5>51.32.41.721.543.42.33.41.23.2332fmol/mg"ER*10941192628793112178829828581337ND*N

D237540212PgR27225143175142SO9635774303297213NDND85176452

" fmol/mg of protein in cytosol.* ER, estrogen receptor; PgR. progesterone receptor; TNM, tumor-nodes-metastasis classification; ND, not determined.

ranged between 89 and 1452 fmol/mg with a mean of 523 Â±72fmol/mg. In benign breast lesions the Smaxranged between 35 and 234fmol/mg with a mean of 108 Â±20 fmol/mg (Fig. 3). There was asignificant statistical difference between the two groups (P < 0.001).On the contrary the average Kd values calculated for the two groupswere similar ( 1.9 Â±0.9 nM for breast carcinomas, 1.2 Â±0.9 nM forbenign lesions).

Also, breast carcinomas contained 19 times more uPA than benignlesions (611 Â±134 fmol/mg versus 32 Â±8 fmol/mg) (P < 0.01) asassessed by immunoenzymatic assay on tissue extracts (Fig. 3). Simple regression analysis showed that the tissue concentration of uPARis significantly correlated to the tissue concentration of uPA (r = 0.62,

P < 0.002) in breast carcinomas. The mean molar ratio between uPARand uPA was 0.85 for breast carcinomas. This ratio was inverted inbenign lesions which contained 3 times more uPAR than uPA.

There was no significant difference in uPAR tissue concentrationand uPA levels between patients with lymph node involvement andpatients without lymph node metastasis. Neither uPAR nor uPA correlated with patient age, tumor size, nuclear grade, or estrogen orprogesterone receptor levels.

Analysis of regional distribution of uPAR showed 3 different au-

toradiographic patterns depending on the type of tissue analyzed. Inbreast carcinomas the distribution of uPAR was quite heterogeneous.Regions of high concentration were interspersed with areas of lowconcentration. A comparison of autoradiographic and histolÃ³gica! images showed that in most of the cases, uPA receptors were mainlyconcentrated in areas of high cellularity. These areas were often located at the periphery of the tumor (Fig. 4A ), although in a number ofinstances they were also found scattered throughout the tumor section.In fibroadenomas the uPAR concentration was diffuse and homogeneous (Fig. 4B) and again corresponding to cellular elements in thesections. However, the density of uPARs was significantly lower thanin malignant lesions. In fibrocystic breast disease the uPAR distribution was focal corresponding to the walls of cystic structures(Fig. 4C).

A confirmatory experiment was performed by staining sections ofbreast carcinomas with an anti-uPAR MoAb. Although the intensity of

the chromogenic reaction varied in different tumors, all breast carcinomas showed a positive and heterogeneous pattern of staining (Fig.5A). The most intense positivities were found on the membrane ofepithelial tumor cells protruding into the tumor stroma with an alignedpattern of growth or at the periphery of large clusters of tumor cells.

Table 2 Tissue concentration of uPAR determined h\ quantitative autoradio^raphyand uPA levels determined by immunoenzymatic assay in breast carcinomas

uPAR uPA

Patient12345678910111213141516171819202122pmol/g"13.7810.549.7410.406.0015.428.707.086.0212.368.187.046.844.4414.3613.525.944.8613.646.188.445.20fmol/mg''8755104826272911094925265285566350239257540405145226189961229517291pmol/g"11.701.356.024.307.5413.521.572.468.126.192.2010.8121.8611.4425.2018.6013.5211.8136.3012.076.024.95fmol/mg*743652982603659591679238428394367822139371019975942172558447369277

" pmol/g of wet tissue.* fmol/mg of protein in crude homogenate.

Weak and diffuse staining was also observed in the tumor stromaalong with scattered positive fibroblast-like cells. All benign breastlesions were negative for immunoperoxidase staining with anti-uPAR

MoAb except for occasional and rare focal positivity of ductal epithelial cells (Fig. 5B).

Intratumoral Distribution of uPA. Since uPA receptors may beeither free or occupied with endogenous produced uPA, we assessedwhether tumor-associated uPA is localized on the plasma membrane of

breast cancer cells. Immunoperoxidase staining was performed withMoAb 394 which reacts with either free or bound uPA. Most of theepithelial tumor cells were stained with MoAb 394, although theintensity of staining was not always homogeneous. Diffuse stainingwas found in the cytoplasm of tumor cells and a more pronouncedreactivity of plasma cell membranes was observed in confined areas ofthe tumor section (Fig. 6A). A preferential location of uPA at theleading edge of the tumor was not found. Tumor stroma showed onlya faint chromogenic reaction and at higher magnification scatteredpositive fibroblast-like cells were also observed. Again, all benign

3201




breast lesions were unreactive to MoAb 394 except for rare and dottedpositivities in ductal epithelial cells (Fig. dB).

Rate of Proliferation and Neovascularization. Since uPAR isassociated to the highly cellularized areas of the tumor we determinedthe rate of proliferation of tumor cells using the monoclonal antibodyKi67 directed against a nuclear antigen expressed by cells in proliferation. Table 4 reports the percentage of tumor cells that are in aproliferative state as assessed by staining with MoAb Ki67 and thenumber of microvessels per field assessed by staining endothelial cellsfor factor VIII. Simple regression analysis showed a significant positive correlation between tumor concentration of uPA determined byenzyme-linked immunosorbent assay (pmol/g) and rate of proliferation of malignant tumor cells (r = 0.55, P < 0.01). However, the

percentage of tumor cells in a proliferative state was not correlated totissue concentration of uPAR determined by quantitative autoradiog-

raphy. Also, neither uPAR nor uPA appeared to be correlated to thenumber of microvessels per field.

DISCUSSION

The present study reports both the tissue concentration and thedistribution of uPAR in 22 breast carcinomas and 9 benign breastlesions by in vitro quantitative autoradiography. The concentration of

Table 3 Tissue concentration of uPAR determined by quantitative autoradiographyand uPA levels determined bv immunoenzymatic assay in benign breast lesions

uPAR uPA

Patient123456789pmol/g"3.804.741.202.202.584.142.862.842.40fmol/mg''172136357390986723472pmol/g"0.890.820.560.262.471.970.550.410.82fmol/mg''40241688647133424

" pmol/g of wet tissue.h fmol/mg of protein in crude homogenate.

3000

2000

CD

E

0

E

1000

M M

uPAR uPA

Table 4 Rale of proliferation assessed by immunostaining wiÃ¬hKÃŒ67monoclonalantibody and number of microvessels per field determined bv immunostaining of

endothelial cells for factor VIII in sections of breast carcinomas

Patient12345678910111213141516171819202122"

ND. not deterRate

of proliferation(% of KÃŒ67positivecells)50318401013ND"1640ID15314135666540ND33121029nined.Neovasculahzation(no. of microvessels/field)13IX10282618ND271511ND17222022IK33ND13132019

tumor-associated uPA was also determined by enzyme-linked immu

nosorbent assay on tumor extracts. We found that the expression ofuPAR and uPA is coordinately enhanced in breast carcinomas ascompared to benign breast lesions. Malignant tumors contained 5times more uPAR and 19 times more uPA than benign tumors.

Interestingly, an equimolar ratio was found between the mean content of uPAR and the mean level of uPA in those tumors. If theconcentration of uPAR and uPA is expressed in terms of pmol/g of wettissue, i.e., nmol/liter, we can apply the well known binding equation:

fi B

Assuming that the ligand and its specific receptors can freely interact,we can predict 70% of the receptor binding sites be occupied in breastcarcinomas whereas the theoretical value for occupancy in benignbreast lesions is only 21%.4 On this basis, breast carcinomas contain10 times more receptor-bound uPA than benign breast lesions. Although immunostaining with anti-uPA MoAb 394 could not providequantitative data on the fractional occupancy of uPAR, it confirmedthat in breast carcinomas uPA is heavily concentrated on the membranes of tumor cells whereas cells from benign lesions were repro-ducibly devoid of uPA.

As mentioned before, uPA catalyzed plasminogen activation on thecell surface is much more efficient than that occurring in solution (27,28). The 10-fold increase in the concentration of receptor-bound uPA,

which we estimated in breast carcinomas, may then result in a potentself-amplification of the proteolytic system on the surface of malig

nant tumor cells. Functionally active plasmin generated on the plasmamembrane of tumor cells may allow them to cross basement membranes, capillary walls, and interstitial connective tissues and to spreadto other organs or tissues. The increased expression of uPA in malignant breast tumors has already been reported (45). Our data show thatuPA and uPAR are coordinately overexpressed in breast carcinomassupporting the hypothesis that receptor-bound uPA is required for the

4 Substituting F = T - B yields:

B B

T B Kd KjFig. 3. Tissue concentration of uPAR determined by quantitative autoradiography and

uPA levels measured by immunoenzymatic assay in 22 malignant (M) and 9 benign (B) where T is the total concentration of uPA measured by enzyme Â¡mmunoassay, and fim.xtumors of the breast. Bars, mean values. Statistical difference: uPAR concentration in and KtÂ¡are the receptor concentration and dissociation constant determined by quantitativemalignant versus benign tumors, P< 0.001; uPA concentration in malignant versus benign autoradiography. Solving the equation for B gives the molar concentration of uPA spe-tumors, P < 0.01. cifically bound to its receptor. ÃŸ/ÃŸmaxwill give occupancy of the receptor binding sites.

3202




Fig. 4. Left, digitized auloradiograms of a section of breast carcinoma (A), fibroadenoma (B),and fibrocystic breast disease (C) incubated with 4nM l25I-pro-uPA; Righi, digitized autoradiogramsof an adjacent section of the same tumors preincu-

bated with 20 /AMunlabeled uPA and then incubatedwith 4 nM 123I-pro-uPA plus an excess (0.2 /IM) of

unlabeled uPA. The values of bound radioactivityare shown in the color scale on the righi. Note thedifferent autoradiographic patterns depending onthe type of tissue analyzed.

metastatic cascade. In agreement with our findings, uPA activity hasbeen reported to be significantly associated with an increased relapserate in patients with breast carcinoma (45,46). The lack of correlationbetween uPA/uPAR expression and lymph node involvement shown inthe above mentioned studies as well as in ours suggests that tumor-

associated uPA may be mainly involved in the metastatic processoccurring directly through blood vessels rather than in lymphaticdissemination. There is indeed experimental evidence implicating thatuPA is an essential factor in tumor cell intravasation (47).

uPAR was heterogeneously distributed within breast carcinomaswithout a clearcut preferential location at the periphery of the tumor.However, when uPAR localization was studied at cellular level by

immunoperoxidase staining, the most intense staining was found inmalignant cells located in areas of the tumor that, from histolÃ³gica!criteria such as pattern of growth or stroma infiltration, appeared to beinvasive.

In vitro quantitative autoradiography has been extensively used tomeasure the binding parameters of hormones and neurotransmitters(48) to receptor sites and to quantitate tumor-associated antigens in

histolÃ³gica! sections of solid tumors (34). To the best of our knowledge, ours is the first study in which quantitative autoradiography hasbeen used to measure receptors involved in the mechanism of tumorinvasion and metastasis. The reliability of autoradiographic measurements was confirmed testing tumor cells bearing a known number of

3203




'

Fig. 5. Immunoperoxidase staining of uPARwith MoAb 3936 on frozen sections of a breastcarcinoma (A ) and a benign breast lesion (B). Notethe intense staining on the surface of malignanttumor cells (A) and the lack of reactivity in benignductal epithelial cells (B). Counterstaining with he-maloxylin; A. X 400; B, X 250.

*..**â€¢â€¢

B

Â«>

- -

receptor binding sites, by cross-linking experiments, and by immu-

noperoxidase staining with several MoAbs. Previous studies on membranes from breast carcinomas reported an uPAR concentration of 250fmol/mg of protein (22) whereas we found a mean uPAR concentration of 523 fmol/mg of protein. This difference either could be due toa different extraction procedure or reflect the presence of a naturalsoluble form of uPAR in different compartments of the tumor.

A significant positive correlation was found between tissue concentration of uPA, determined by immunoenzymatic assay on tumorextracts, and rate of proliferation of malignant tumor cells as assessedby immunostaining with MoAb KÃ•67.This result was not totallyunexpected since several growth factors, such as epidermal growthfactor, fibroblast growth factor, platelet-derived growth factor, andcolony-stimulating factor 1, besides their ability to elicit DNA syn

thesis, were shown to stimulate the synthesis of urokinase mRNA infibroblasts, endothelial cells, and cultured human tumor cells (49).Recently, it has been reported that uPA can directly activate latent

growth factors (17) thus contributing to an enhanced proliferativeactivity. In any case, the lack of correlation between uPAR concentration and rate of proliferation of malignant tumor cells in our studysuggests that, if a causal relationship exists between uPA activity andproliferation, uPAR is not involved in transducing a mitogenic signalwithin the cell. Furthermore, the correlation between uPA levels andrate of proliferation indicates that uPA expression in vivo is likelymodulated by effectors similar to those which have been shown toregulate uPA production in cultured human tumor cell lines. It is.ofinterest that the tumor with the highest level of uPA in our study alsoshowed overexpression of the oncogene c-erB-2 which encodes for a

protein highly homologous to epidermal growth factor receptor (datanot shown).

Immunoperoxidase staining revealed the presence of uPAR and uPAnot only in malignant tumor cells but also in fibroblast-like cells

scattered in tumor stroma. It is known that normal fibroblasts bearuPAR and if properly stimulated can secrete urokinase (21). Local

3204




Â«:Â«

Fig. 6. Immunoperoxidase staining of uPA withMoAb 394 on frozen sections of an invasive ductalbreast carcinoma with squamous metaplasia (A) anda benign breast lesion (B). Note the intense reactivityof plasma cell membranes and the diffuse cytoplas-mic staining of malignant tumor ceils (A ). No staining is observed in benign ductal epithelial cells (B).Counterstaining with hematoxylin; A, X 400; B, X250.

* >Ã *i.

B

secretion of proteolytic enzymes by fibroblasta might be regulated byfactors produced by tumor cells themselves. For instance, Davies et al.(50) have described a polypeptide secreted by transformed cells thatinduces the production of human plasminogen activator in culturedfibroblasts. Furthermore, many authors reported that fibroblasts isolated from human solid tumors or cocultured with human tumor cellshave an increased capacity to synthesize and secrete hydrolases (51,52). On this basis, the fibroblast-like cells, positively stained for uPAR

and uPA in our study, can be considered as helper cells contributing tothe disruption of matrix structures. Also, they may participate to thevery complex cross-talk which occurs between epithelial tumor cells

and stromal cells during tumor development.In conclusion our study showed that both uPAR and uPA are over-

expressed in breast carcinomas as compared to benign breast lesions.On the basis of simultaneous quantitative measurements receptor-

bound uPA was estimated to be 10 times more concentrated in malignant tumors than in benign lesions. This may result in an enhancedkinetics of plasminogen activation at the surface of malignant tumor

cells thus leading to a solid-phase proteolytic cascade the end point of

which is the pericellular degradation of extracellular matrix. Furtherstudies are needed to define the prognostic significance of uPARconcentration and the correlation with the probability of disease-free

survival in patients with breast cancer. The technique described hereappeared to be particularly suitable for such studies providing quantitative data on uPAR concentration in tumors. Finally the approachthat we used in this study, namely the simultaneous determination ofthe receptor concentration and the levels of its specific ligand, couldbe applied to other systems involved in the mechanism of tumorinvasion and metastasis, thus contributing to clarify the complexnetwork of interactions leading to tumor spread.

ACKNOWLEDGMENTS

We wish to thank Dr. F. Blasi for the generous gift of some monoclonalantibodies and cell lines. We are also indebted to Dr. J. C. Reynolds for many

3205




helpful discussions. The authors acknowledge the expert technical assistanceof Luigi Spiezia and Alfredo De Rosa.

REFERENCES

1. Markus, G., Takita, H., Camiolo. S., Corasanti, J., Evers, J., and Hobika, G. Contentand characterization of plasminogen activators in human lung tumors and normal lungtissue. Cancer Res.. 40: 841-848, 1980.

2. Sappine, A., Busso, N., Belin, D., and Vassalli, J. Increase of urokinase-type plas

minogen activator gene expression in human lung and breast carcinoma. Cancer Res.,47: 4043^Â»046, 1987.

3. Markus, G., Camiolo, S., Konga, S., Madeja, J., and Mittelman, A. Plasminogenactivator secretion of human tumors in short-term organ culture, including a comparison of primary and metastatic colon tumors. Cancer Res., 43: 5517-5525, 1983.

4. Ossowski, L., and Reich, E. Antibodies to plasminogen activator inhibit human tumormetastasis. Cell, 35: 611-619, 1983.

5. Hearing, V. J., Law, L. W., Coni, A., Appella, E., and Blasi, F. Modulation ofmetastatic potential by cell surface urokinase of murine melanoma cells. Cancer Res.,48: 1270-1278. 1988.

6. Mignatti, P., Robbins, E., and Riikin. D. B. Tumor invasion through the humanamniotic membrane: requirement for a proleinase cascade. Cell, 47: 487â€”498,1986.

7. Cajot, J. F., Schleuning, W. D., Medcalf, R. L., Bamat, J.. Testuz, J., Lieberman, L.,and Sordat, B. Mouse L cells expressing human pro-urokinase-type plasminogen

activator, effects on extracellular matrix degradation and invasion. J. Cell Biol., 109:915-925, 1989.

8. Blasi, F., Vassalli, J. D., and DaÃ±o,K. Urokinase-type plasminogen activator: pro-enzyme, receptor, and inhibitors. J. Cell Biol., 104: 801-804, 1987.

9. DaÃ±o,K., Andreasen. P. A., Grondahl-Hansen, J., Kristensen, P., Nielsen, L. S.,Skriver, L. Plasminogen activators, tissue degradation and cancer. Adv. Cancer Res.,44: 139-266, 1985.

10. Andreasen, P. A., Nielsen, L. S., Kristensen, P., Grondahl-Hansen, J., Skriver, L., andDaÃ±o,K. Plasminogen activator inhibitor from human fibrosarcoma cells binds uroki-nase-lype plasminogen activator, but not its proenzyme. J. Biol. Chem., 261: 7644-7651. 1986.

11. Mira-y-Lopez, R., Reich, E., and Ossowski, L. Modulation of plasminogen activatorin rodent mammary tumors by hormones and other effectors. Cancer Res.. 43: 5467-5477. 1983.

12. Niedbala. M. J.. and Sartorelli, A. Regulation by epidermal growth factor of humansquamous cell carcinoma plasminogen activator-mediated proteolysis of extracellularmatrix. Cancer Res., 49: 3302-3309, 1989.

13. Saksela. O., Moscatelli, D.. and Rifkin. B. The opposing effects of basic fibroblastgrowth factor and transforming growth factor ÃŸon the regulation of plasminogenactivator activity in capillary endothelial cells. J. Cell Biol., 105: 957-963, 1987.

14. Keski-Oja, J., Blasi, F., Leof, E. B., and Moses, H. L. Regulation of the synthesis and

activity of urokinase plasminogen activator in A549 human lung carcinoma cells bytransforming growth factor ÃŸ.l. Cell Biol., 106: 451^*59, 1988.

15. Grimaldi, G.. Di Fiore, P., Kajtaniak, L. E., Falco, J., and Blasi, F. Modulation ofurokinase plasminogen activator gene expression during the transition from quiescentto proliferative state in normal mouse cells. EMBO J.. 5: 855-861, 1986.

16. Kirchheimer, J. C. Woita. J.. Christ. J.. and Binder, B. R. Functional inhibition ofendogenously produced urokinase decreases cell proliferation in a human melanomacell line. Proc. Nati. Acad. Sci. USA, 86: 5424-5428, 1989.

17. N.ililnu. L., Tamagnone, L.. Vigna, E., Sachs, M., Hartmann, G., Birchmeier, W.,Daikuhara. Y.. Tsubouchi. H.. Blasi. F., and Comoglio. P. Extracellular proteolyticcleavage by urokinase is required for activation of hepatocyte growth factor/scatterfactor. EMBO J.. //:4825^4833. 1993.

18. Vassalli. J. D.. Baccino. D., and Belin. D. A cellular binding site for the 55.000 formof the human plasminogen activator, urokinase. J. Cell Biol., 100: 86-92, 1985.

19. Stoppelli, M. P., Corti, A., Soffientini, A., Cassani, G., Blasi, F., and Assoian, R. K.Differentiation-enhanced binding of the amino-terminal fragment of human urokinase

plasminogen activator to a specific receptor on U937 monocytes. Proc. Nati. Acad.Sci. USA. Â«2:4939-4943, 1985.

20. Appella, E., Robinson, E. A., Ullrich, S. J., Stoppelli. M. P.. Corti, A.. Cassani, G., andBlasi, F. The receptor-binding sequence of urokinase. A biological function for thegrowth-factor module of proteases. J. Biol. Chem., 262: 4437-4440, 1987.

21. Blasi. F. Surface receptors for urokinase plasminogen activator. Fibrinolysis, 2: 73-

84. 1988.22. Needham. G. K.. Sherbet. G. V.. Farndon, J. R.. and Harris, A. L. Binding of urokinase

to specific receptor sites on human breast cancer membranes. Br. J. Cancer, 55:13-16, 1987.

23. Stoppelli, M. P.. Tacchetti, C., Cubellis, M. V., Corti, A., Hearing, V. J., Cassani. G..Appella, E., and Blasi F. Autocrine saturation of pro-urokinase receptors on humanA431 cells. Cell, 45: 675-684, 1986.

24. Ossowski, L. Clunie, G., Masucci, M. T., and Blasi, F. In vivo paracrine interactionbetween urokinase and its receptor: effect on tumor cell invasion. J. Cell Biol., Â¡IS:1107-1112, 1991.

25. Schlechte. W., Murano. G., and Boyd, D. Examination of the role of the urokinasereceptor in human colon cancer mediated laminin degradation. Cancer Res., 49:6064-6069, 1989.

26. Hollas. W.. Blasi. F., and Boyd. D. Role of the urokinase receptor in facilitatingextracellular matrix invasion by cultured colon cancer. Cancer Res., 5/: 3690-3695,

1991.27. Plow, E. F., Freaney, D. E., Plescia, J.. and Miles, L. A. The plasminogen system and

cell surfaces: evidence for plasminogen and urokinase receptors on the same cell type.J. Cell Biol., 103: 2411-2420, 1986.

28. Ellis. V, Scully, M. F.. and Kakkar. V. V. Plasminogen activation initiated bysingle-chain urokinase-type plasminogen activator. J. Biol. Chem., 264: 2185-2188,

1989.29. Fraker, P. J. and Speck, J. C. Protein and cell membrane iodinations with a sparingly

soluble chloramide l.3.4,6-tetrachloro-3 a.6 a-diphenylglycouril. Biochem. Biophys.Res. Commun.. 80: 849-857, 1978.

30. Laemmli, U. K. Cleavage of structural proteins during the assembly of the head ofbacteriophage T4. Nature (Lond.), 227: 680-685, 1970.

31. Nielsen. L. S., Kellerman, G. M., Behrendt. N., Picone. R.. DaÃ±o.K.. and Blasi, F. A55,000-60,000 M, receptor protein for urokinase-type plasminogen activator. J. Biol.Chem.. 263: 2358-2363. 1988.

32. Hartmann, W. H., Ozzello, L., Sobun, L. H., and Stalsberg, H. HistolÃ³gica! Typing ofBreast Tumors, Ed. 2. Geneva: WHO, 1981.

33. Black, M. M., and Speer. F. D. Nuclear structure in cancer tissues. Surg. Gynecol.Obstet.. 705: 97-102, 1957.

34. Del Vecchio, S., Reynolds, J. C.. Blasberg, R. G., Neumann, R. D.. Carrasquillo, J. A.,HellstrÃ¶m, I., and Larson, S. M. Measurements of local M, 97,000 and 250.000protein antigen concentration in sections of human melanoma tumors using in vitroquantitative autoradiography. Cancer Res., 48: 5475-5481, 1988.

35. Blasberg, R. G., Groothuis, D., and Molnar, P. Application of quantitative autorad-

iographic measurements in experimental brain tumor models. Semin. Neurol., I:203-221, 1981.

36. Camiolo, S. M., Siula, M. R.. and Madeja, J. M. Improved medium for extraction ofplasminogen activator from tissue. Prep. Biochem. 12: 297-305. 1982.

37. Bradford, M. A rapid and sensitive method for the quantification of microgramquantities of protein utilizing the principle of protein-dye binding. Anal. Biochem.,72: 248-254. 1976.

38. EORTC Breast Cancer Cooperative Group. Standards for the assessment of hormonereceptors in human breast cancer. Eur. J. Cancer, 16: 1513-1515, 1980.

39. Chucholowski. N.. Schmitt. M.. Rettenberger. P.. Schuren, E.. Moniwa, N.. Goretzki.L., Wilhelm. O., Weidle, U., Janicke, F., and Graeff. H. Flow cytofluorometricanalysis of the urokinase receptor (uPAR) on tumor cells by fluorescent uPA-ligandor monoclonal antibody 3936. Fibrinolysis, 6: 95-102, 1992.

40. Hsu, S. M.. Raine, L., and Fanger. H. Use of avidin-biotin-peroxidase complex (ABC)

in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody(PAP) procedures. J. Histochem., Cytochem., 29: 577-580, 1981.

41. Masucci, M. T, Pedersen, N., and Blasi, F. A soluble, ligand binding mutant of thehuman urokinase plasminogen activator receptor J. Biol. Chem., 266: 8655-8658,

1991.42. Gerdes, J., Lemke, H., Baisch, H., Wacker, H. H., Schwab, U.. and Stein, H. Cell cycle

analysis of a cell proliferation associated human nuclear antigen defined by themonoclonal antibody KÃ•67.J. Immunol., 133: 1710-1715. 1983.

43. Colton. T. Statistic in Medicine. Boston: Little. Brown, and Co.. 1974.44. Picone. R.. Kajtaniak. E. L., Nielsen, L. S., Behrendt, N., Mastronicola, M. R.,

Cubellis, M. V, Stoppelli, M. P., Pedersen, S., DaÃ±o,K.. and Blasi. F. Regulation ofurokinase receptors in monocytelike U937 cells by phorbol ester myristate acetate. J.Cell Biol.. 108: 693-702, 1989.

45. Janicke, F., Schmitt, M., Hafter, R., Hollrieder, A., Babic, R., Ulm, K., Gossner, W.,and Graeff, H. Urokinase-type plasminogen activator (uPA) antigen is a predictor ofearly relapse in breast cancer. Fibrinolysis, 4: 69-78, 1990.

46. Foekens, J. A.. Schmitt, M., van Putten, W. L. J., Peters, H. A., Bontenbal, M.,Janicke, F., and Klijn, J. G. M. Prognostic value of urokinase-type plasminogenactivator in 671 primary breast cancer patients. Cancer Res., 52: 6101-6105, 1992.

47. Ossowski, L. Plasminogen activator dependent pathways in the dissemination oftumor cells in the chick embryo. Cell, 52: 321-326, 1988.

48. Young, A. B., Frey, K. A., and Agranoff, B. W. Receptor assay: in vitro and in vivo.In: M. Phelps, J. Mazziotta, and H. Scheiben (eds.). Positron Emission Tomographyand Autoradiography: Principles and Applications for the Brain and Heart, pp. 73-

111. New York: Raven Press, 1986.49. Mullins, D. E., and Riikin, D. B. Induction of proteases and protease inhibitors by

growth factors. In: M. B. Spom and A. B. Roberts (eds.), Peptide Growth Factors andtheir Receptors, pp. 481-507. New York: Springer-Verlag, 1990.

50. Davies, R. L., Rifkin, D. B., Tepper, R., Miller. A., and Kucherlapati. R. A polypeptidesecreted by transformed cells that modulates human plasminogen activator production. Science (Washington DC), 22/: 171-173, 1983.

51. Bauer, E. A., Gordon, J. M., Reddick, M. E., and Eisen, A. Z. Quantitation andimmunochemical localization of human skin collagenase in basal cell carcinoma. J.Invest. Dermatol, 69: 363-367, 1977.

52. Dabbous, M. K., Haney, L., Carter, L. M., Paul, A. K., and Reger, J. Heterogeneity offibroblast response in host-tumor cell-cell interactions in metastatic tumors. J. Cell.Biochem., 35: 333-344. 1987.

3206



1993;53:3198-3206. Cancer Res Silvana Del Vecchio, M. Patrizia Stoppelli, Maria V. Carriero, et al. Autoradiography: Comparison with Urokinase Levels

Quantitativein VitroBenign Breast Tumors by Human Urokinase Receptor Concentration in Malignant and

Updated version

http://cancerres.aacrjournals.org/content/53/13/3198

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/53/13/3198To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



human urokinase receptor concentration in malignant and ... · siveness and proliferation,...

Documents