Increases Expression of Receptor by Activation of the Thrombin Receptor

Zeren Yang,* Robert L. Cohen,* Ge Ming Lui,~f David A. Laxvrence* and Marc A. Shuman*

Purpose. To investigate the effect of thrombin on the urokinase receptor (u-PAR) in retinal pigment epithelial (RPE) cells. Methods. The authors analyzed u-PAR mRNA by Northern blot hybridization. Retinal pigment epithelial cell surface u-PAR was assayed by measuring the amount of functional urokinase plasminogen activator (u-PA) bound to cells at saturation. Retinal pigment epithelial cells were derived from fetal retinal tissue and established in primary cell culture. Results. Thrombin increased u-PAR mRNA 4-fold in RPE cells examined by Northern blot hybridization, whereas the amount of thrombin receptor mRNA was unchanged. Thrombin stimulated u-PA binding to RPE cells 2.5- to 5-fold in a time- and dose-dependent manner. , a thrombin antagonist, completely blocked the effects of thrombin on u-PAR expres- sion in RPE cells. Phosphatidylinositol treatment of RPE cells resulted in the abolition of thrombin-induced u-PA binding. Recombinant soluble u-PAR competitively inhibited two-chain u-PA binding to the surface of thrombin-treated RPE cells. A thrombin receptor agonist peptide (SFLLRNPNDKYEPF) also induced a 2.5-fold increase in binding of u-PA to the surface of RPE cells. Conclusion. Thrombin increases u-PAR expression by RPE cells by a mechanism involving activation of the seven transmembrane thrombin receptor. Invest Ophthalmol Vis Sci. 1995; 36:2254-2261.

In addition to initiating , urokinase plas- mature, glycosylated form ranges from 40,000 to 55,000 minogen activator (u-PA) has been implicated in a daltons in different cell lines.16"20 u-PAR binds active two- variety of biologic processes, including , chain u-PA and its , single-chain u-PA (pro-u- tumor invasion, tissue remodeling, cellular differentia- PA).21 Phorbol ester, endothelial cell growth factor, fi- tion and proliferation, ovulation, and embryonic im- broblast growth factor, a-tumor necrosis factor, and inter- plantation.1"3 Some of the actions of u-PA appear to feron-7 increase u-PAR expression without a change in be mediated by its interaction with a specific, high- receptor affinity in U937 and endothelial cells and mono affinity membrane urokinase plasminogen activator cytes.4"21 Reuning et al22 recently reported that thrombin receptor (u-PAR).4"6 and other mitogens increased l25I-labeled u-PA binding u-PAR is expressed by many types of cells, including to vascular smooth muscle cells of the bovine aorta; how- U937 cells, endothelial cells, fibroblasts, leukocytes, tumor ever, the mechanism underlying this effect was not deter- cell lines, gastric mucosa, and ovarian tumor cells.4"l5 The mined. The serine thrombin, a , also has the potential to regulate fibrinolysis by limited From the * Cancer Research Institute and the ^Department of Ophthalmology, University of California at San Francisco. and inactivation of single-chain urokinase. In Supported in part by a grant from the Joseph Drown Foundation. ZY was a addition, thrombin stimulates synthesis of types 1 and 2 Biomedical Research Trainee in the Cheng Scholar Program at the University of California at San Francisco. plasminogen activator inhibitors (PAI-1 and PAI-2) by Submitted for publication Decemljer 28, 1994; revised April 3, 1995; accepted May endothelial cells.I ":V23"25 19, 1995. Proprietary interest category: N. A significant manifestation of disordered angiogen- Reprint requests: Marc A. Shuman, Cancer Research Institute, University of esis occurs in neovascularization of the retina. Increasing California at San Francisco, 3rd Avenue and Parnassus Street, San Francisco, CA 94143-0128. evidence suggests that retinal pigment epithelial (RPE)

Investigative Ophthalmology & Visual Science, October 1995, Vol. 36, No. 11 2254 Copyright © Association lor Research in Vision and Ophthalmology

Downloaded from iovs.arvojournals.org on 09/30/2021 Thrombin Increases 2255

cells play a pivotal role in the inhibition of intraocular were from Stratagene (#400701; La Jolla, CA). Im- neovascularization, a key event in the pathogenesis of a muno microliter plate (#01 1-010-3350) was from Dy- number of ophthalmic disorders, including age-related natech Laboratories (Chantilly, VA). Fast-track mRNA macular degeneration and proliferative diabetic retinopa- isolation kit, vector pCR 1000, and vector pCEP were thy.2(>"27 Regulation of plasminogen activation by RPE from Invitrogen (San Diego, CA). (#K1593-02). 5,5'- cells, therefore, may play an important role in these pro- dithiobis (2-nitrobenzoic acid; DTNB) was from Calbi- 520.27 cesses. ochem (LaJolla, CA). Z-lysine thiobenzyl ester (BLT) We investigated the regulation of the expression of was from Peninsula Laboratories (Belmont, CA). En- u-PAR in RPE cells by thrombin to elucidate how throm- dothelial cell growth supplement (#40006) was from bin may alter various biologic processes by modulating Collaborative Biomedical Products (Bedford, MA). cell surface PA activity.1"' We have demonstrated that u- YM-30 membrane, Centricon 10, and Centricon 30 PAR expression by RPE cells is increased by thrombin micro-concentrator were from Amicon (Lexington, through activation of a recently described seven trans- MA). membrane thrombin receptor. Recombinant Soluble u-PAR A recombinant form of u-PAR30 was produced as fol- METHODS lows: A u-PAR cDNA encoding the full-length was cloned by the polymerase chain reaction using Materials U937 cDNA as a template and primers based on the Two-chain u-PA was a generous gift from Jack Henkin, published u-PAR sequence.1'20 In a second round of Abbott Laboratories (Abbott Park, IL). a-thrombin polymerase chain reaction, a stop codon was intro- was a generous gift from John Fenton, Department of duced to truncate the protein after arg281. The resul- Health, New York State University (Albany, NY). The tant 957-bp fragment was cloned into pCR 1000, subse- synthetic agonist of the thrombin receptor (SFLLRN- quently excised with Kpnl and Notl, and recloned into PNDKYEPF), mimicking the new amino terminus cre- the corresponding sites of the expression vector pCEP ated by cleavage of the receptor, the control peptide to yield pCEP/Trunc. A stable cell line expressing (FSLLRNPNDKYEPF), and a thrombin receptor truncated u-PAR was obtained by transfection of 293 cDNA probe (708 bp) were kindly provided by S. R. cells with pCEP/Trunc using the calcium phosphate Coughlin, University of California at San Fran- method.1' Serum-free conditioned media were pre- cisco.282'1 The human u-PA cDNA probe used in these pared, concentrated 10-fold on a YM-30 membrane, studies has been described previously.15 Polyclonal and collected by affinity chromatography on u-PA- anti-u-PAR antiserum was generously provided by Dr. Sepharose. Bound u-PAR was eluted with acid- R. F. Todd III (Ann Arbor, MI).H<) The following were as described,15 and the neutralized eluate was concen- purchased from commercial sources: rabbit polyclonal trated using a Centricon 30 micro-concentrator before against u-PA (#398), plasminogen (#412), use. Sodium dodecyl sulfate-polyacrylamide gel elec- and (#411) were from American Diagnostica trophoresis followed by silver staining showed a single (Greenwich, CT).ir>32P-dCTP and Resolution were from broad band centered at 55 kDa that cross-reacted with Amersham (Arlington Heights, IL. Acrylamide, glyc- a polyclonal anti-u-PAR antiserum. erol, and nylon niters were from Bio-Rad (Richmond, CA). 3-(N-Morpholino) propane sulfonic acid, basic Cell Culture fibroblast growth factor (bFGF), collagenase, bovine Human fetal RPE cells derived from two different fetal serum albumin (BSA), N-2-hydroxythylpiperazine-N'- retinas were harvested and maintained as described 2-ethane sulfonic acid (Hepes), Tris, ethylenediamine previously.31 Tenets of the Declaration of Helsinki tetraacetic acid (EDTA) formaldehyde, Triton X-100, were followed. Informed consent for human research and Tween-20 were from Sigma (St. Louis, MO). Ko- was obtained, and institutional human experimenta- dak XAR autoradiography film was from Eastman Ko- tion committee approval was granted. Briefly, 8- to 10- dak (Rochester, NY). Fibronectin, guanidinium isothi- passage RPE cells were plated in 75 mm2 flasks coated ocynate, cesium chloride, and a random DNA labeling with fibronectin for 10 days, and basic fibroblast system were from BRL (Gaithersburg, MD). Phospha- growth factor (1 ng/ml) was added as a growth supple- tidylinositol phospholipase C (PI-PLC, #1143069) ment to cells every other day as previously reported.31 was from Boehringer Mannheim Biochemicals (India- For mRNA extraction, cells were plated on 24 X 24 napolis, IN). 293 cells (#CRL1573) and glyceralde- cm2 plates. For u-PAR functional assays, the final con- hyde-3-phosphate (GAPDH, #57090) centration was approximately 2 to 2.5 X 107 cells/75 probe from PBR 322 were from American Type Cul- mm2 flask (protein concentration: 5.5 to 6 mg/ml). ture Collection (Rockville, MD). Imidazole was from J. After cells became confluent, they were treated with T. Baker Chemical Company. Nuctran push columns 0 to 10 U/ml of thrombin at 37°C for 24 hours for

Downloaded from iovs.arvojournals.org on 09/30/2021 2256 Investigative Ophthalmology 8c Visual Science, October 1995, Vol. 36, No. 11

determining the effects of thrombin concentration, 5 BSA, then incubated with elution buffer at room tem- U/ml of thrombin for 0 to 48 hours at 37°C for the perature for 3 minutes with mild shaking. The eluate time course study, 5 U/ml of thrombin and/or 10 U/ was collected and 1/10 volume of 1 M Tris-HCl, pH ml of hirudin at 37°C for 24 hours for hirudin blocking 7.6, was added to the solution. Samples were concen- study,2' or 50 to 200 /xM of the thrombin receptor trated 20-fold with a Centricon 10 apparatus and ei- peptide or the same amount of the control peptide at ther assayed for u-PA activity or stored at — 70°C. In 37°C for 24 hours for demonstrating the effect of the competition experiments, 200 ng/ml of two-chain u- thrombin receptor peptide on u-PAR synthesis. PA and 400 ng/ml of recombinant soluble u-PAR were incubated at 4°C for 1 hour before incubation with PI-PLC Treatment of Retinal Pigment RPE cells. Epithelial Cells Retinal pigment epithelial cells were incubated with u-PA Activity Assay 5 U/ml of thrombin for 24 hours. Then cells were Briefly, the samples and standards (0.625 to 24 ng/ washed with phophated-buffered saline (PBS) three ml of two-chain u-PA) were incubated in microtiter times and incubated with 0.2 U/ml of PI-PLC in Tris- plates coated with a polyclonal antibody against u-PA maleate-NaCl buffer (20 mM Tris-maleate, 150 mM at 37°C for 120 minutes. After washing, plasminogen NaCl, pH 7.2) at 37°C for an additional hour. The substrate solution containing 100 //g/ml plasmino- cells were washed three times with PBS before incuba- gen, 0.1% Triton X-100, 0.1% gelatin, 10 mM EDTA, tion of the cells with u-PA to obtain the u-PA binding 50 mM glycine-NaOH, pH 7.8, and 10 ng/ml plasmin assay. • were incubated with the bound u-PA at 370°C for 30 minutes. Then, plasmin substrate solution containing Northern Blot Hybridization 200 mM KC1, 200 mM Tris-HCl, pH7.5, 0.01% Triton After RPE cells were stimulated with and without 5 X-100, 200 fiM BLT, and 220 (iM DTNB were added. U/ml of thrombin at 37°C for 6 hours, total RNA was The samples were incubated at room temperature and extracted from confluent RPE cells using 4 M sodium then analyzed at 405 nm in a microtiter plate spectro- isothiocynate and centrifuged on a 5.7 M cesium chlo- photometer. ride cushion at 150,000g- for 16 hours. The samples were electrophoresed in 1.2% agarose-2.2 M formal- RESULTS dehyde gels and blotted on nylon filters. The filters were respectively hybridized with u-PAR, thrombin re- Initially, we examined the effect of thrombin on u- ceptor, and GAPDH cDNA probes for the correspond- PAR mRNA in RPE cells. u-PAR mRNA was increased ing transcripts. Probe preparation was as described up to 4-fold in thrombin-treated RPE cells, as indi- previously.15 Briefly, probes were labeled by the Ran- cated by densitometry of Northern blots. The throm- 32 dom DNA labeling system in the presence of P-dCTP bin receptor mRNA was expressed in RPE cells but and purified with Nuctran push columns. Hybridiza- was unchanged after thrombin treatment of RPE cells tion was completed at 42°C for 16 hours, and the filters (Fig. 1). were washed in 0.1 X SSC, 0.1% sodium dodecyl sul- We analyzed the dose-dependent manner of fate, at room temperature briefly, then for 30 minutes thrombin stimulation on the functional u-PAR. As at 42°C, before exposure to X-ray film. The density of shown in Figure 2, at a concentration as low as 0.1 U/ autoradiographic signals was quantitated witfi a laser ml, thrombin increased u-PA binding to RPE cells 2.5- scanning densitometer equipped with an integrator. fold, whereas binding was increased 4-fold by 5 U/ml GAPDH band values were used to correct for loading of thrombin. of different amounts of mRNA. The time-dependent effect of thrombin stimula- tion on u-PA binding to RPE cells is shown in Figure Assay of Cell Surface u-PAR 3. Thrombin stimulated a 2-fold increase in u-PA bind- u-PA binding to cells was used to assess cell surface u- ing to RPE cells after 2 hours of incubation. After 48 PAR51': Confluent cells were washed three times with hours of incubation with thrombin, u-PA binding to ice-cold PBS and acid treated with elution buffer (50 the RPE cell surface was 5-fold more than the control. mM glycine-HCl, 0.1 M NaCl, pH 3.0) at room tem- We next examined the effects of hirudin, a potent perature for 3 minutes with mild shaking. After neu- thrombin antagonist,28'34 on thrombin-induced u-PAR tralization with 0.1 M Tris-HCl, pH 7.6, the cells were synthesis. Compared with the control and treatment washed three times with cold PBS containing 0.1% with thrombin only, hirudin completely blocked the BSA; 200 ng/ml of two-chain u-PA was then added in effect of thrombin on u-PA binding to the RPE cell ice-cold M199 medium containing 0.1% BSA, and the surface, as indicated in Figure 4. cells were incubated at 4°C for 1 hour. Cells were Further evidence that two-chain u-PA specifically washed three times with ice-cold PBS containing 0.1% binds to u-PAR was obtained using PI-PLC. Pre-

Downloaded from iovs.arvojournals.org on 09/30/2021 Thrombin Increases Urokinase Receptor 2257 1 2 i UPAR Thrombin I" receptor

GAPDH 10 30 30 4 0 SO FIGURE l. u-PAR mRNA expression by retinal pigment epi- FIGURE 3. Time course of u-PAR expression on the surface thelial cells stimulated by thrombin. Total RNA was ex- of retinal pigment epithelial cells in response to thrombin tracted from retinal pigment epithelial cells after incubation stimulation. Retinal pigment epithelial cells were incubated with 5 U/ml of thrombin at 37°C for 6 hours. Twenty-five with 5 U/ml of thrombin at 37°C for varying periods. Func- micrograms of total RNA was loaded in each lane. The filter tional u-PAR was measured with u-PA binding assay. Data was hybridized with glyceraldehyde-3-phosphate dehydroge- shown are die mean ± SD of u-PA binding activity from nase (GAPDH), thrombin receptor cDNA, and u-PAR cDNA three separate experiments. probes as described in Methods. Lane 1 = control; lane 2 = thrombin treatment. GAPDH band is a control to show that the same amount of RNA was loaded in each lane. binding to the cell surface was increased 2.5-fold in RPE cells by the thrombin receptor peptide, whereas the control peptide, at a concentration equivalent to viously, it was shown that u-PAR is attached to the cell that of the agonist, did not induce a significant change membrane by a phosphatidylinositol linkage and that the receptor can be solubilized with PI-PLC.3^35 in u-PA binding to the RPE cell surface (Fig. 7). Compared with the control and treatment with throm- bin only, PI-PLC treatment abolished most of the DISCUSSION effect of the thrombin on the u-PA binding to the Thrombin-induced synthesis of u-PAR mRNA by any surface of RPE cells (Fig. 5). cell type has not been reported. In this study, we dem- Additional evidence that two-chain u-PA specifi- onstrated diat thrombin increases u-PAR mRNA in cally binds to cell surface u-PAR was provided with RPE cells. We also found that u-PAR mRNA contained recombinant soluble u-PAR. Recombinant soluble u- in human umbilical vein endothelial cells was upregu- PAR competitively inhibited binding of two-chain u- lated by thrombin (data not shown). Thrombin also PA to the surface of the cells (Fig. 6). To determine whether thrombin induces u-PAR synthesis by activating the thrombin receptor, we incu- bated RPE cells with the thrombin receptor peptide previously shown to autoactivate the receptor. u-PA

3

Control Thr Hlr Hlr+Thr 0 0-1 1 5 10 Thrombin U/ml FIGURE 4. Effect of hirudin on stimulation of u-PAR expres- FIGURE 2. Dose-dependent effect of thrombin on expression sion by thrombin. Retinal pigment epithelial cells were incu- of functional u-PAR by retinal pigment epithelial cells bated with 5 U/ml of thrombin with or without 10 U/ml of treated with 0 to 10 U/ml of thrombin at 37°C for 24 hours. hirudin at 37°C for 24 hours. Functional u-PAR was mea- Functional u-PAR was measured with a u-PA binding assay sured by the u-PA binding assay. Data shown are the mean ± as described in Methods. Data shown are the mean ± SD SD of u-PA binding activity from three separate experiments. of u-PA binding from three separate experiments. Thr = thrombin; Hir = hirudin.

Downloaded from iovs.arvojournals.org on 09/30/2021 2258 Investigative Ophthalmology & Visual Science, October 1995, Vol. 36, No. 11

f 3 I

Control Thrombin A50 A200 CSO C200 FIGURE 7. u-PAR expression in retinal pigment epithelial Control Thr PI-PLC PUThr cells stimulated by the thrombin receptor agonist peptide. FIGURE 5. Effect of phosphatidylinositol phospholipase C Retinal pigment epithelial cells were incubated with either (PI-PLC) on binding of u-PA to retinal pigment epithelial 5 U/ml of thrombin, 50 to 200 //M of the agonist peptide cells after stimulation by thrombin. Retinal pigment epithe- (SFLLRNPNDKYEPF), or control peptides (FSLLRN- lial cells were incubated with 0.2 U/ml of PI-PLC at 37°C PNDKYEPF) at 37°C for 24 hours, respectively. Functional for 1 hour after treatment with 5 U/ml of thrombin at 37°C u-PAR was measured with the u-PA binding assay. Data for 24 hours. Functional u-PAR was measured with the u- shown are the mean ± SD of u-PA binding from three sepa- PA binding assay. Data shown are the mean ± SD of u- rate experiments. Thr = thrombin; A50 = 50 fj,M of the PA binding activity for three separate experiments. Thr = agonist peptide; C50 = 50 /JM of control peptide. thrombin treatment; PI = PI-PLC treatment. resulted in an abolition of u-PA binding is consistent stimulates expression of functional u-PAR on the sur- with evidence shown by others that PI-PLC cleaves face of RPE cells in a time- and concentration-depen- the PI linkage site of u-PAR to the plasma membrane, 32 33 35 dent manner, as indicated by increased u-PA binding. releasing soluble receptor from the cell surface. ' ' That hirudin, a specific thrombin antagonist,28'34 com- Furthermore, that two-chain u-PA binding to RPE cells pletely blocked the effect of thrombin on u-PA bind- was competitively inhibited by recombinant soluble u- ing to RPE cells indicates that an unblocked thrombin PAR has provided additional evidence for the specific- 30 anion binding exosite is necessary for the stimulation ity of the binding of u-PA to the surface of the cells. of RPE cells.25 That PI-PLC treatment of RPE cells It is unclear whether RPE cells contain u-PA or tissue plasminogen activator (t-PA).27'36"44 Although we were unable to detect u-PA in thrombin-treated RPE cells (data not shown), our results raise the possi- bility that paracrine interactions with other cell types occur in vivo, resulting in a complete plasminogen activator system. Recent results in other organ systems suggest that the generation and regulation of extracel- lular proteolysis during invasive processes is a complex cooperative process involving components provided by different cell types. For example, in colon cancer, u-PA is expressed by fibroblast-like stromal cells in the vicinity of invasive foci, whereas u-PAR is found on tumor-infiltrating macrophages and some tumor cells.45"50 PAI-1 is expressed by endothelial cells in the 45 51 Control Thr UPAR UPAFUThr tumor stroma. " It is likely that u-PA secreted by one type of cell can bind to u-PAR on another type FIGURE 6. Effect of recombinant soluble u-PAR on u-PA bind- in a paracrine fashion.15'40"50 Thus, it is possible that u- ing to retinal pigment epidielial cells stimulated by throm- PA synthesized and secreted by another type of ocular bin. Retinal pigment epithelial cells treated with 5 U/ml of tissue could be bound to the u-PAR expressed on RPE thrombin were incubated with two-chain u-PA and recombi- nant soluble u-PAR as described in Methods. Functional u- cells. Plasminogen activation within the retina may be PAR on retinal pigment epithelial cells was measured with the result of regulation by several different cell types. the u-PA binding assay. Data shown are the mean ± SD of Several different thrombin receptors have been u-PA binding activity from three separate experiments. Thr described.28'52"54 Binding of thrombin to any of these = thrombin; u-PAR = recombinant soluble u-PAR. could conceivably serve as a mechanism for induction

Downloaded from iovs.arvojournals.org on 09/30/2021 Thrombin Increases Urokinase Receptor 2259

of u-PAR synthesis. Our observations that the seven bleeding, although this process has the potential for transmembrane thrombin receptor is expressed by negative regulation.2540"44 RPE cells and that stimulation of u-PAR expression by thrombin is mimicked by the thrombin receptor Key Words agonist peptide28'29i55r><1 suggest that thrombin stimu- retinal pigment epithelium, thrombin, thrombin receptor, lates u-PAR synthesis by activation of this receptor. urokinase plasminogen activator, urokinase plasminogen ac- A balance between u-PA and PAI-1 is likely to be tivator receptor important in vascular morphogenesis and several Acknowledgments other processes in which there is degradation of extra- cellular matrix. '~3'49 In addition, it has been postu- The authors thank Margaret Lee for her assistance with lated that u-PAR, as well as other involved in densitometry. the activation or inhibition of plasminogen activation, References are involved in neovascularization under physiologic and pathologic situations.53'1"44'57"59 Thrombin, acting 1. Blasi F, VassalliJ-D, Dan0 K. Urokinase-type plasmin- through the seven transmembrane receptor, may play ogen activator: Prozyme, receptor, and inhibitors. / an important role in physiologic and pathologic neo- CellBiol. 1987; 104:801-804. vascularization in the eye by increasing expression of 2. Mayer M. Biochemical and biological aspects of the u-PAR and PAI-1.5'26'27'37'40-44 plasminogen activator system. Clin Biochem. 1990; 23: 197-211. One of the important functions of RPE cells is 3. Loskutoff DJ, Sawdey M, Mimuro J. Type 1 plasmino- continuous phagocytosis of the outer segment of rods gen activator inhibitor. Prog . shed by the photoreceptor tips during the course of 1989;9:87-115. renewal of the photoreceptor membrane/10 Interpho- 4. Masucci MT, Blasi F. The receptor for urokinase: A toreceptor matrix (IPRM) consists of glycoproteins, potential target for anti-invasive and anti-metastatic including fibronectin and proteoglycans.b0 During re- therapy. Thrombosis Res. 1990(suppl XII):49-60. newal of the outer segments, the apical processes of 5. Mignatti P, Mazzieri R, Rifkin DB. Expression of the the RPE cells move through the IPRM as the microvilli urokinase receptor in vascular endothelial cells is stim- engulf packets of outer segments. Some digestion of ulated by basic fibroblast growth factor. J Cell Biol. the shed photoreceptor discs also occurs before inges- 30 3957 60 6. Hollas W, Blasi F, Boyd D. Role of the urokinase recep- tion by the RPE cells. - - It is possible that, in tor in facilitating invasion by cul- vivo, u-PA is secreted by RPE or different types of tured colon cancer. Cancer Res. 1991;51:3690-3695. retinal cells, resulting in plasmin formation and the 7. Stoppelli MP, Corti A, Soffientini G, Cassani F, Blasi degradation of various components of the IPRM. This F, Assoian RK. Differentiation-enhanced binding of process would be facilitated by u-PA binding to the u- the amino-terminal fragment of human plasminogen PAR on RPE cells.3"375758 Evidence for a role of cell activator to a specific receptor on U937 monocytes. surface proteolysis of matrix by plasmin has been ob- Proc Natl Acad Sci USA. 1985; 82:4939-4943. tained in cancer. Receptor-bound u-PA mediates plas- 8. Vassalli JD, Baccino D, Belin D. A cellular min generation on some types of tumor cells, which, for the Mr 55,000 form of the human plasminogen in turn, contributes significandy to their potential to activator, urokinase. / Cell Biol. 1985; 100:86-92. degrade extracellular matrix.6'15'45"48 9. Schlechte JW, Boyd D. Insensitivity of laminin degra- dation directed by receptor bound urokinase to PAI- Degradation of the extracellular matrix permits 36 39 5760 1 in cultured colon cancer. Cancer Comm. 1990;2:261- the movement and migration of cells. - ' Thus, 270. u-PA bound to the surface of RPE cells also may be 10. Boyd D. Expression of the effects of epidermal growth important in the development of the retina by facilitat- factor on the production of urokinase and the expres- ing cellular migration during the organization of reti- sion of the plasminogen activator receptor in a human nal tissue. Thrombin and the seven transmembrane colon cancer cell line. Cancer Res. 1989; 49:2427-2432. receptor may be involved in these processes by increas- 11. Barnathan ES, Kuo A, Kariko K, et al. Characterization ing u-PAR on RPE cells. of human endothelial cell urokinase-type plasmino- Finally, thrombin, the penultimate coagulation gen activator-receptor protein and messenger RNA. factor, may play a complex role in regulating fibrino- . 1990; 76:1795-1806. lysis in the retina. By increasing the synthesis of u-PAR, 12. Kristensen P, Eriksen J, Blasi F, Dan0 K. Two alterna- tively spliced mouse urokinase receptor mRNAs with thrombin facilitates u-PA-mediated plasmin formation different histological localization in the gastrointesti- and rapid clot dissolution. Thrombin, on the other nal tract. / Cell Biol. 1990; 115:1763-1771. hand, inactivates u-PA and stimulates PAI-1 synthesis 13. Casslen B, Gustavsson B, Astedt B. in the retina, resulting in the inhibition of t-PA and receptor for urokinase plasminogen activator are in- u-PA and further plasmin generation. Thus, thrombin creased in malignant ovarian tumor. Eur J Cancer. formation in the retina has the potential to increase 1991;27:1445-1448.

Downloaded from iovs.arvojournals.org on 09/30/2021 2260 Investigative Ophthalmology 8c Visual Science, October 1995, Vol. 36, No. 11

14. Haddock RC, Spell ML, Baker CD, et al. Urokinase ered ligand agonist peptides. / Biol Chem. 1992; 267: binding and receptor identification in cultured endo- 13146-13149. thelial cells. J Biol Chem. 1991;266:21466-21473. 30. Mizukami IF, Garni-Wagner BA, DeAngelo LM, et al. 15. Cohen RL, Xi XP, Crowley CW, Luces BK, Levinson Immunologic detection of the cellular receptor for AD, Shuman MA. Effects of urokinase receptor occu- urokinase plasminogen activator. Clin Immunol Immu- pancy on plasminogen generation and proteolysis of nopathol. 1994; 71:96-104. basement membrane by human tumor cells. Blood. 31. Soung MY, Lui GM. Propagation of fetal human RPE 1991; 78:479-487. cells: Preservation of original culture morphology 16. Behrendt N, Ronne E, Ploug M, et al. The human after serial passage. / Cell Physiol. 1990; 143:196-203. receptor for urokinase plasminogen activator: NH2- 32. Mizukami IF, Vinjauri SD, Perini F, Liu DY, Todd RF terminal sequence and glycosylation variants. / Biol HI. Purification, biochemical composition, and bio- Chem. 1990; 265:6453-6460. synthesis of the Mo3 activation expressed on 17. Estreicher A, Wohlwend A, Belin D, Schleuning WD, the plasma membrane of human mononuclear phago- Vassalli JD. Characterization of the cellular binding cytes. / Immunol. 1991; 147:1331-1337. site for the urokinase-type plasminogen activator. / 33. Behrendt N, Ploug M, Houen G, Blasi F, Dan0 K. Biol Chem. 1989;264:1180-1189. The human receptor for the urokinase plasminogen 18. Hajjar K, Hamel NM. Identification and characteriza- activator. JBiol Chem. 1991;266:7842-7848. tion of human endothelial cell membrane binding 34. Liu LW, Vu T-KH, Esmon CT, Charo I, Coughlin sites for tissue plasminogen activator and urokinase. SR. The region of the thrombin receptor resembling J Biol Chem. 1990;265:2908-2916. hirudin binds to thrombin and alters enzyme specific- 19. Nielsen LG, Kellerman GM, Behrendt N, Picone R, ity. JBiol Chem. 1991;266:16977-16980. Dan0 K, Blasi F. A 55,000-60,000 Mr receptor protein 35. Ploug M, Ronne E, Behrendt N, Jensen AL, Blasi F, for urokinase-type plasminogen activator: Identifica- Dan K. Cellular receptor for urokinase plasminogen tion in human tumor cell lines and partial purifica- activator. / Biol Chem. 1991; 266:1926-1932. tion. JBiol Chem. 1988;263:2358-2363. 36.( Immonen I, Stephens RW, Salonen E-M, Lattikainen 20. Roldan AL, Cubellis MV, Masucci MT, et al. Cloning L, Sim P-S, Vaheri A. Tissue-type plasminogen activa- and expression of the receptor for human urokinase tor in subretinal fluid. Curr Eye Res. 1989; 8:249-255. plasminogen activator, a central molecular in cell sur- 37. Tripathi BJ, Genon JD, Tripathi RC. Distribution of face, plasmin dependent proteolysis. EMBO J. 1990; tissue plasminogen activator in human and monkey 9:467-474. eye. / Ophthalmol. 1987; 94:1434-1438. 21. Cubellis MV, Nolli ML, Cassani G, Blasi F. Binding of 38. Mirshahi M, Mirshahi S, Soria C, et al. Production of single-chain prourokinase to the urokinase receptor type plasminogen activator and their inhibi- of human U937 cells. J Biol Chem. 1986;261:15819- tor in cornea. Biochem Biophys Res Comm. 1989; 160: 15822. 1021-1025. 22. Reuning U, Bang NU. Regulation of the urokinase- 39. Moisseiev J, Jerdan JA, Dyer K, Maglione A, Glaser type plasminogen activator receptor on vascular BM. Retinal cell plasminogen activator. Invest Ophthal- smooth muscle cells is under the control of thrombin mol Vis Sri. 1990; 31:1070-1078. and other mitogens. Atherosclerosis Thromb. 1992; 12: 40. Siren V, Stephens RW, Salonen EM, Vaheri A, Summa- 1161-1168. nen P, Immonen I. Retinal pigment epithelial cells 23. Knudsen BS, Harpel PC, Nachman RL. Plasminogen secrete urokinase-type plasminogen activator and its activator inhibitor is associated with the extracellular inhibitor PAI-1. Ophthalmic Res. 1992; 24:203-210. matrix of cultured bovine smooth muscle cells. J Clin 41. Hackett SF, Campochiaro PA. Modulation of plasmin- Invest. 1987;80:1082-1089. ogen activator inhibitor-1 and urokinase in retinal pig- 24. Gelehrter TD, Sznycer-Laszuk R. Thrombin induc- mented epithelial cells. Invest Ophthalmol Vis Sri. tion of plasminogen activator-inhibitor in cultured hu- 1993; 34:2055-2062. man endothelial cells. / Clin Invest. 1986; 77:165-169. 42. Schmidt D, Schumacher M, Wakhloo AK Microca- 25. Shuman MA. Thrombin-cellular interactions. Ann theter urokinase infusion in central retinal artery oc- NYAcadSci. 1986; 485:349-368. clusion. Am J Ophthalmol. 1992; 113:429-434. 26. MoisseievJ, Jerdon J, Dyer K, Maglione A, Glaser BM. 43. Zheng YL, Tang DH, Peng HM. Genetic predisposi- Retinal pigment epithelial cells can influence endo- tion to retinoblastoma (Rb). Eye Sri. 1993;9:149-152. thelial cell plasminogen activators. Invest Ophthalmol 44. Mach R, Kessler P, Susicky P, Kubickova M. Thrombol- Vis Sri. 1990;31:1070-1078. ysis of arterial retinal occlusion using urokinase. 27. Tripathi BJ, ParkJK, Tripathi RC. Extracellular release Ceskoslovenska OJtalmologie. 1992;48:42-47. In Czecho- of tissue plasminogen activator is increased with the slovakian. phagocytic activity of the retinal pigment epithelium. 45. Brucker A, Fildman AE, Kirchheimer JC, Binder BR, Invest Ophthalmol Vis Sri. 1989;30:2470-2473. Remold HG. Endogenous receptor-bound urokinase 28. Vu T-KH, Hung DT, Virginia I, Wheaton VI, Cough- mediates tissue invasion of the human lung carcinoma lin SR. Molecular cloning of a functional thrombin cell lines. Cancer Res. 1992; 52:3043-3047. receptor reveals a novel proteolytic mechanism of re- 46. Zini JM, Murray SC, Graham CH, et al. Characteriza- ceptor activation. Cell. 1991;64:1057-1068. tion of urokinase receptor by placenta trophoblasts. 29. Scarborough RM, Naughton M, Teng W, et al. Teth- Blood. 1992;79:29l7-2929.

Downloaded from iovs.arvojournals.org on 09/30/2021 Thrombin Increases Urokinase Receptor 2261

47. Hoosein NM, Boyd D, Hollas WJ, Mazar A, Henkin J, receptor of the surface. J Biol Chem. Chung LW. Involvement of urokinase and its receptor 1978; 253:3435-3443. in the invasiveness of human prostatic carcinoma cell 54. Esmon CT, Owen WG. Identification of an endothelial lines. Cancer Comm. 1991;3:2550-2564. cell for thrombin-catalyzed activation of pro- 48. Quax PH, Pedersen N, Masucci MT, et al. Comple- tein C. Proc Natl Acad Sci USA. 1981;78:2249-2252. mentation between urokinase-producing and recep- 55. Vu T-K H, Virginia I, Wheaton VI, Hung DT, Charo I, tor-producing cells in extracellular matrix degrada- Coughlin SR. Domains specifying thrombin-receptor tion. CellRegul 1991;2:793-803. interaction. Nature. 1991; 353:674-677. 49. Liotta LA, Steeg PS, Stetler-Stevenson WG. Cancer 56. Vassallo RRJr, Keiber-Emmons T, Cichowski K, Brass and angiogenesis: An imbalance of positive L. Structure-function relationships in the activation and negative regulation. Cell. 1991;64:327-336. of platelet thrombin receptors by receptor-derived 50. Olson D, Pollanen J, Hoyer-Hansen G, et al. Internal- peptides. / Biol Chem. 1992; 267:6081-6085. ization of the urokinase-plasminogen activator inhibi- 57. Glaser BM, Campochiaro PA, DavisJL, Sato M. Retinal tor type 1 complex is mediated by the urokinase recep- pigment epithelial cells release an inhibitor of neovas- cularization. Arch Ophthalmol. 1985; 103:1870-1875. tor. JBiol Chem. 1991;267:9129-9133. 58. Adler AJ, Klucznik KM. Proteins and glycoproteins of 51. Hollas WJ, Boyd D. Regulation of the urokinase recep- the bovine interphotoreceptor matrix: Composition tor by its plasminogen activator. Thromb Haemost. and fraction. Exp Eye Res. 1981;34:423-430. 1991;66:678-683. 59. Bok D. Retinal photoreceptor-pigment epithelium in- 52. Gronke RS, Bergman BL, Baker JB. Thrombin interac- teractions. Invest Ophthalmol Vis Sci. 1985;26:1659- tion with : Influence of a platelet protease 1665. nexin. JBiol Chem. 1987;262:3030-3036. 60. Pino RM. Immunocytochemical localization of fibro- 53. Okamura T, Hasitz M, Jamieson GA. Platelet glycocali- nection to the retinal pigment epithelium of the rat. cin: Interaction with thrombin and role as thrombin Invest Ophthalmol Vis Sci. 1986; 27:840-848.

Downloaded from iovs.arvojournals.org on 09/30/2021