Oxygen modulates the release of urokinase and plasminogen activator inhibitor-1 by retinal pigment epithelial cells

The International Journal of Biochemistry & Cell Biology 33 (2001) 237–247 www.elsevier.com/locate/ijbcb

Oxygen modulates the release of urokinase and plasminogen activator inhibitor-1 by retinal pigment epithelial cells Jonathan T. Erichsen a, John Jarvis-Evans b, Asud Khaliq b, Mike Boulton a,* b

a Department of Optometry and Vision Sciences, Cardiff Uni6ersity, Cardiff CF10 3NB, Wales, UK Department of Ophthalmology and School of Biological Sciences, Uni6ersity of Manchester, Manchester, UK

Received 3 July 2000; received in revised form 3 January 2001; accepted 4 January 2001

Abstract The aims of this study were to examine the effect of oxygen, in the presence or absence of exogenous growth factors, on the release of plasminogen activators and plasminogen activator inhibitor-1 by cultured human retinal pigment epithelial cells. Antigen and activity levels of urokinase, tissue plasminogen activator and plasminogen activator inhibitor were measured in conditioned media after cells were exposed to three different oxygen environments: hypoxia, normoxia and hyperoxia. Overall proteolytic balance was determined by zymography. The effects of exogenous basic fibroblast growth factor and transforming growth factor-beta were also examined. it was found that retinal pigment epithelial cells released urokinase, tissue plasminogen activator and plasminogen activator inhibitor in measurable quantities. After 48 h, urokinase levels were highest at normoxia, reaching 7.2ng/106 cells ( 9 2.0 SEM), whereas plasminogen activator inhibitor 1 levels were highest at hyperoxia, reaching 67.5ng/106 cells ( 93.7 SEM). Tissue plasminogen activator levels were minimal ( B0.5ng/106 cells) and unaffected by both oxygen and growth factors. Overall proteolytic activity was also greatest at normoxia. Fibroblast growth factor stimulated urokinase production dose-dependently, but plasminogen activator inhibitor only minimally. Transforming growth factor-beta stimulated plasminogen activator inhibitor production dose-dependently but urokinase only at higher concentrations. These results suggest that both oxygen tension and growth factors may interact to modulate the proteolytic properties of the human retinal pigment epithelium. © 2001 Elsevier Science Ltd. All rights reserved. Keywords: Oxygen; Plasminogen activator; Plasminogen activator inhibitor-1; Retinal pigment epithelial cells

1. Introduction Plasminogen activators (PAs) are serine proteases that convert plasminogen to plasmin. * Corresponding author. Tel.: + 44-29-20875100; fax: +4429-20874859. E-mail address: [email protected] (M. Boulton).

They are ubiquitous enzymes which are secreted by many cell types and play a central role in regulating proteolysis in a wide variety of normal and pathological processes [1,2]. These include tissue remodelling, cell migration, fibrinolysis, thrombolysis, and tumour metastasis and invasiveness [1–3]. Two types of PA have been identified in mammals, tissue-type PA (tPA) and

1357-2725/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S 1 3 5 7 - 2 7 2 5 ( 0 1 ) 0 0 0 0 9 - 7

238

J.T. Erichsen et al. / The International Journal of Biochemistry & Cell Biology 33 (2001) 237–247

urokinase-type PA (uPA) [1]. Both types are inhibited predominantly by plasminogen activator inhibitor-1 (PAI-1), the major PAI in plasma [2,4]. However, uPA is receptor-mediated [5] and is able to catalyse plasminogen activation on the cell surface [6]. The resulting receptor-bound plasmin is consequently protected from the serumderived inhibitors a2-antiplasmin and a2-macroglobulin [7,8]. Such an activation mechanism allows for directional proteolysis and has been implicated in cell migration through the extracellular matrix [3,6,9,10]. The release of PAs and PAI-1 from cells is controlled by such factors as local enzyme concentrations, specific receptors, precursor activation, up or down regulation by various cytokines and the local oxygen tension [11 – 19]. In vitro, human retinal pigment epithelial (hRPE) cells are known to release tPA, uPA and PAI-1 [20 –23] and the proliferation, as well as the growth factor production/responsiveness, of these cells has been previously shown to be dependent on the local oxygen environment [24 – 27]. In proliferative vitreoretinopathy (PVR), hRPE cells proliferate and migrate through full-thickness breaks in the retina to reach the vitreoretinal interface and the vitreous gel where they are involved in epiretinal membrane formation and subsequent tractional retinal detachment [28 – 30]. Epiretinal membranes are characterised by a fibrovascular appearance and by a high extracellular matric content, including collagen, fibronectin, vitronectin, laminin and thrombospondin [31]. In their normal location on Bruch’s membrane, hRPE cells are exposed to relatively high oxygen tensions (i.e. between 70 and 90 mmHg [24,25]). However, in their ectopic location in the vitreous cavity, the local pO2 is much reduced; 30–40 mmHg for attached retina [25] and probably even lower if the neuroretina is separated from the pigment epithelium and choroid owing to rhegmatogenous retinal detachment. Levels in the midvitreous are quoted as low as 12 mmHg [32]. We therefore hypothesise that the oxygen environment regulates the PA system and that this contributes to the abundance of matrix associated with pathologies. We have sought to elucidate the relationship between the

local oxygen environment and plasminogen activator release by hRPE cells in vitro in an effort to better understand the regulation of the PA system in pathological situations. We have also investigated the effect of oxygen on the release of PAI-1 and the concomitant role of two growth factors, basic fibroblast growth factor (FGF2) and transforming growth factor-beta (TGFb), which are reported (a) to regulate plasminogen activator and PAI-1 expression in other cell types [10,13,14] and (b) to be expressed in proliferative diseases such as PVR [33,34]. FGF2 has been localised to epiretinal membranes [35]. TGF-b is upregulated in the vitreous of patients with PVR [36] and has been shown to regulate RPE phenotype in vitro [37].

2. Materials and methods

2.1. Cell culture Human RPE cells were isolated and grown in either multiwell dishes or in 75 cm2 tissue culture flasks as previously described [38]. In brief, human donor eyes of varying ages with no reported ophthalmic disease and permission for research were obtained from the Manchester Eye Bank after the cornea had been removed for transplantation. Each eye was dissected slightly posterior to the limbus, the vitreous and retina removed gently from the RPE layer. The exposed RPE was washed twice with Dulbecco’s phosphate buffered saline (PBS) and treated with 0.25% trypsin solution for 1 h at 37°C. RPE cells were detached by gentle aspiration with a micropipette and the cell suspension transferred to growth medium [Ham’s F10 medium supplemented with 20% fetal calf serum (FCS), 0.4% glucose, antibiotics (100 mg/ml streptomycin, 100 mg/ml kanomycin, 100 units/ml benzyl penicillin) and fungizone (2.5 mg/ml amphotericin B)]. The cells were centrifuged at 70 g and plated out in growth medium. Upon confluence, cells were subcultured at a split ratio of 1:3. All cells were used at confluence between the third and sixth passage.

J.T. Erichsen et al. / The International Journal of Biochemistry & Cell Biology 33 (2001) 237–247

2.2. Control of oxygen en6ironment Human RPE cultures were placed in either a standard incubator (95% air/5% CO2 atmosphere; i.e. 20% O2) or in modular incubators (Flow, Ayrshire, UK) as previously described [26]. Modular incubators were purged with either 95% N2/ 5% CO2 (i.e. 0% O2) or 5% O2 (containing 5% CO2, balance N2). Media pO2 measurements were determined using a Ciba Corning 178 blood gas analyser. Typically, the pO2 values were as follows: 0% O2 = 199 3 mmHg, 5% O2 =68.5 910 mmHg, 20% O2 =136 9 12 mmHg; variations in media pO2 at each atmospheric oxygen concentration are due to inconsistencies inherent in gassing modular incubators [25]. With respect to the normal range of oxygen tensions experienced by hRPE cells, these three mean media pO2s can be considered relatively hypoxic (19 mmHg), normoxic (68.5 mmHg), and hyperoxic (136 mmHg). For ease of presentation we will refer to hypoxia, normoxia and hyperoxia in the text.

2.3. Measurements of plasminogen acti6ators and their inhibitor The following methods were utilised: 1. Total PA and PAI-1 activity levels were measured in culture media using Spectrolyse/fibrin diagnostic kits according to manufacturer’s instructions (Biopool AB, Sweden). 2. Antigen levels of uPA, tPA and PAI-1 were measured in culture media using Tintelize uPA ELISA kits, Imulyse tPA ELISA kits and Tintelize PAI-1 kits according to manufacturer’s instructions (Biopool AB). 3. PA activity in culture supernatant samples was measured by the casein – plasminogen zymography method as described by GranelliPiperno and Reich [21,39] with modifications. In brief, samples were subjected to SDS – polyacrylamide gel electrophoresis (10% polyacrylamide). The gels were washed at room temperature for 30 min with two changes of 2.5% Triton X-100. The gels were washed in water and layered onto substrate indicator gels and incubated for 24 – 48 h. Agarose substrate indicator gels were prepared containing 2%

239

non-fat dried milk as a source of casein and plasminogen (2 mg/ml) and poured onto a petri dish. Lysis zones were seen as clear areas in an opaque background, indicating PA activity. Levels of endogenous uPA, tPA and PAI-1 in plasma were subtracted from the values in conditioned medium.

2.4. Release of uPA, tPA and PAI-1 under standard oxygen conditions Human RPE cells were grown to confluence under standard incubator conditions (37°C, 95% air/5% CO2 atmosphere; i.e. 20% O2) in 2× 24 well multiwell plates in growth medium. At confluence, cells were washed twice in Dulbecco’s phosphate buffered saline and the medium replaced with 1ml per well of either (i) Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 7.5% human platelet-poor plasma [40] or (ii) plasma-free DMEM. Cultures were then placed in a conventional incubator. Culture fluid from three wells was removed from both plates after 6, 12, 24 and 48 h and stored at − 70°C prior to measurement of uPA, tPA and PAI-1 as described above. Cells from each well were trypsinised (0.25% trypsin, 0.02% EDTA) and numbers determined using a haemocytometer. Experiments were repeated on at least two further occasions.

2.5. Effect of FGF2 and TGFi on uPA, tPA and PAI-1 release at 6arying oxygen tensions Confluent hRPE cultures were washed twice in PBS prior to exposure to either FGF2 concentrations of 0, 0.1, 1.0 or 10 ng/ml DMEM 9 plasma or TGFb1 concentrations of 0, 0.2, 2.0 or 20 ng/ml DMEM 9plasma (both growth factors obtained from British Biotechnology). Cultures were subjected to three oxygen environments: hypoxia, normoxia or hyperoxia. Cultures were incubated for 48 h, after which the culture medium was collected and stored at −70°C until assay and cell numbers were determined. These experiments were repeated at least three times for each cell line.

240

J.T. Erichsen et al. / The International Journal of Biochemistry & Cell Biology 33 (2001) 237–247

2.6. Cell 6iability Cell toxicity was assessed by determining cell viability after each experiment utilising the trypan blue exclusion assay [41].

2.7. Statistics Statistical analysis was performed on appropriate data using an unpaired t-test. Where multiple comparisons were made, the t-test was supplemented with a Bonferroni correction factor [42]. Significance levels were confirmed using multiple analysis of variance. Results are presented for individual experiments due to the variation between different donors and passage.

3. Results

enhanced by the presence of plasma. After 6 h, tPA antigen release became detectable at low levels (0.5 –1.2 ng/106 cells) and showed no obvious change with time (90.04 ng/106cells: data not shown). PAI-1 antigen levels (Fig. 1b) reached a maximum at 6 h incubation then decreased by up to 55% thereafter (in plasma-containing medium), depending on the cell line used. This trend was also evident in plasma-free medium, but as in the case of uPA antigen accumulation, PAI-1 antigen levels were reduced at all time points compared to those observed in plasma-containing medium.

3.2. Effect of oxygen on the release of uPA, tPA and PAI-1 Varying medium oxygen tensions modulated the levels of uPA antigen (Fig. 2a,b) and PAI

Confluent human RPE cultures maintained in either plasma-containing or plasma-free medium released detectable amounts of uPA, tPA and PAI-1 into the overlying media. The release of these components was always significantly less in plasma-free medium than that observed in plasma-containing medium. While similar trends were observed in all cell lines, the maximum levels of PAs and PAI-1 released by different cell lines varied. In all experiments, cell numbers remained constant throughout the time course of the study. Trypan blue exclusion determined that there was no loss of cell viability in any experiment.

3.1. Time-dependent release of uPA, tPA and PAI-1 under normal incubator conditions Time course studies under standard incubator conditions demonstrated different profiles for the medium accumulation of uPA, tPA or PAI-1. In both plasma-containing and plasma-free medium, uPA antigen accumulated during the initial 6 h, plateaued until 12 h and thereafter steadily increased until the experiment was terminated at 48 h (Fig. 1a). This profile was observed in all experiments. Maximum levels of uPA antigen (i.e. at 48 h) varied between 0.93 and 14.95 ng/106 cells depending on the cell line used and were always

Fig. 1. (a) Urokinase (uPA) and (b) plasminogen activator inhibitor-1 (PAI-1) antigen release by confluent cultures of human retinal pigment epithelial (hRPE) cells (cultured in DMEM containing 7.5% human platelet-poor plasma and plasma-free DMEM) with time under standard incubator conditions. Each point represents the average of three assays from a single donor (39 years). Bars indicate standard error.

J.T. Erichsen et al. / The International Journal of Biochemistry & Cell Biology 33 (2001) 237–247

241

cells) and did not appear to be affected by oxygen modulation. Relative to the accumulation at normoxic pO2, production of uPA antigen decreased significantly in both the hypoxic and hyperoxic conditions (PB 0.01). The amount of the reduction depended on the cell line used but was not donor age dependent. A similar trend to uPA antigen levels was observed when ‘PA activity’ was measured (Fig. 2b); PA activity was highest at a normoxic media pO2 and decreased as the media pO2 was decreased or increased. For both antigen and activity the decrease was always greatest at hypoxia. By contrast, PAI-1 antigen levels in conditioned medium progressively decreased as the media pO2 was decreased from hyperoxia to hypoxia (Fig. 2c). This trend was similar after 6 and 48 h incubation, but the net accumulations of PAI-1 were always significantly higher at 6 h at all oxygen tensions. In zymography, lysis bands were seen in all samples by 48 h incubation (Fig. 3). Plasma-free samples demonstrated a single lysis band at 50 kDa which corresponded to that of the uPA standard. Samples containing plasma showed the 50 kDa band in all samples together with a second band at approximately 90 kDa. Lysis bands were always greatest at normoxic pO2 and decreased at higher and lower media pO2. This effect was apparent in both plasma-containing and plasma-free medium although lysis was generally less in plasma-free medium.

Fig. 2. (a) uPA antigen levels, (b) plasminogen activator (PA) activity and (c) PAI-1 activity in conditioned media of confluent hRPE cultures derived from separate donors exposed to three oxygen environments (18, 72 and 140 mmHg). Medium was collected 48 h after the addition of fresh medium (DMEM supplemented with 7.5% human platelet-poor plasma). Bars indicate standard error.

antigen (Fig. 2c) released into the culture medium. tPA antigen levels were minimal (B0.5 ng/106

Fig. 3. Casein– agarose zymography following SDS– polyacrylamide gel electrophoresis of conditioned-media from confluent hRPE cells (from a 39 year donor). Lanes 1 – 3, cells conditioned with serum-free medium, and lanes 4 – 6, cells conditioned in medium containing 7.5% human platelet-poor plasma. Cells were exposed to three oxygen environments. Lanes 1 and 4: 20 mmHg; lanes 2 and 5: 61 mmHg; lanes 3 and 6: 135 mmHg; lane 7: sample buffer; lane 8: uPA standard (1U/ml); and lane 9: uPA standard (0.5U/ml).

242

J.T. Erichsen et al. / The International Journal of Biochemistry & Cell Biology 33 (2001) 237–247

Fig. 4. uPA (a,c) and PAI-1 (b,d) antigen release by confluent cultures of hRPE cells (from a 40 year donor) exposed to three oxygen environments (media pO2 = 20, 58 and 135 mmHg) for 48 h. At each oxygen tension, cultures were treated with FGF2 at initial media concentrations of 0.1 (), 1 () or 10 ng/ml ( ) (a,b) or TGFb at initial media concentrations of 0.2 (), 2 () or 20 ng/ml ( ) (c,d). (2) represents control, i.e. plasma-containing DMEM without exogenous growth factor. Each point represents the average of three assays. Bars indicate standard error.

3.3. Effect of FGF2 and TGFi on uPA, tPA and PAI-1 release uPA release by hRPE cells was modulated by both FGF2 and TGFb1 (Fig. 4). The addition of FGF2 caused a dose-dependent increase in uPA antigen in the medium. However, the oxygen-induced trend in uPA production remained evident (Fig. 4a) with maximal release at normoxia. PAI -1 release was stimulated above control by FGF2. PAI-1 release was similar for all FGF2 concentrations tested (Fig. 4b). TGFb1 at 20 ng/ml resulted in a significant increase in uPA release compared to the control and was not observed at lower TGFb1 concentrations (Fig. 4c). Interestingly, TGFb1-stimulated uPA release was greatest at hyperoxia, unlike controls which showed maximal release at

normoxia. By contrast, TGFb1 stimulated a dosedependent release of PAI-1 by confluent cells which appeared to be independent of media pO2 (Fig. 4d); release was greatest at 20 ng/ml TGFb1. At hypoxia, both FGF2 and TGF-b1 caused a small but significant increase in uPA. By contrast, PAI production was more than 1.8 and 100 fold greater than at normoxia for FGF2 and TGF-b1, respectively. tPA production was unaffected by the addition of TGFb1 or FGF2 into the culture medium.

4. Discussion Our results showing that hRPE cells secrete significant amounts of uPA and PAI-1, but little tPA, are in accordance with previous findings

J.T. Erichsen et al. / The International Journal of Biochemistry & Cell Biology 33 (2001) 237–247

[20,21,23]. Furthermore, we have demonstrated that the release of both uPA and PAI by hRPE cells can be modulated by both the local media pO2 and the addition of exogenous growth factors. Furthermore, the extent of such growth factor modulation is determined by changes in the local media pO2. uPA is normally secreted as an inactive, single chain zymogen, which can be converted to the two-chain active enzyme by various proteases including plasmin. Both the active and inactive forms are capable of binding to the uPA-receptor, whereupon activated uPA can modulate extracellular proteolysis either directly or via activation of other latent proteases, e.g. plasminogen and metalloproteinases [1,43,44]. Receptor-associated uPA facilitates focal proteolysis and extracellular matrix turnover [1,6,45]. Therefore, in order to maintain proteolytic balance in tissues, the activities of the enzymes in the uPA cascade must be tightly controlled, e.g. by either the action of specific fast-acting PA inhibitors (predominantly PAI-1) [4] or other environmental agents (in particular growth factors). This was confirmed in our study wherein both FGF2 and TGFb1 increased uPA and PAI production, respectively, in a dosedependent manner, supporting previous studies using a variety of cell types: (i) FGF2 stimulates the expression of uPA and its receptor in cultured vascular endothelial cells [10,15]; (ii) TGFb1 transiently increases mRNA levels for uPA and PAI-1 in bronchial epithelial cells [13]; (iii) TGFb1 stimulates uPA synthesis in human synovial fibroblasts [14]; (iv) TGFb induces uPA receptor expression in cultured RPE [46]; (v) production of PA and PAI by hRPE cells has been reported following exposure to IGF-1 [11]; and (vi) hRPE mRNA levels for PAI-1 are increased when incubated with TGFb [47] while uPA mRNA levels are decreased [48]. During the pathogenesis of PVR, RPE cells also become exposed to lower oxygen environments which, in this study, have been demonstrated to be a significant factor in uPA and PAI-1 regulation, particularly in the presence of growth factors. Oxygen has been shown to be a potent modulator of both transcription and translation of proteins in cultured cells [49,50], including hRPE

243

cells [24]. However, it is unclear whether oxygen exerts its effect (i) directly on some aspect of the PA system or (ii) indirectly through the modulation of either (a) endogenous growth factors levels or (b) growth factor receptor expression [24,26,49 –52]; we have previously reported that oxygen modulates the production of FGF2 and TGF-b by hRPE cells [24] as well as affecting growth factor receptor expression [26]. Whatever the oxygen-inducible mechanism, it results in different profiles of uPA and PAI release. Although the levels of uPA and PAI were both at a minimum in hypoxic conditions, uPA release peaked at a normoxic media pO2 whereas PAI release was highest at a hyperoxic pO2. This was confirmed by zymography which demonstrated that overall proteolytic activity was due to uPA and was greatest at a pO2 of 68 mmHg. The 90 kDa band observed in serum-containing medium is in agreement with that reported by Siren et al. and probably reflects an activated tPA –PAI complex derived from the tPA endogenous to the serum [21]. uPA, tPA and PAI-1 are all likely to make a significant contribution to the normal maintenance of the interphotoreceptor matrix and Bruch’s membrane [20,53,54] as well as extracellular matrix turnover/accumulation in pathological situations such as PVR and age-related macular degeneration (AMD). Matrix turnover is associated with changes in growth factor levels/profiles, thus indicating a potential role for growth factors in the modulation of the PA system. In PVR, for example, hRPE cells form a significant component of epiretinal membranes. High levels of expression of acidic FGF, epidermal growth factor (EGF), insulin-like growth factor-1 (IGF-1) and platelet-derived growth factor (PDGF) have all been immunolocalised to epiretinal membranes [55,56]. Furthermore, intravitreal levels of EGF, IGF-1, TGFb and interleukin-6 have been shown to be increased in PVR as compared to controls [57 –59]. In AMD, a change in proteolytic capacity may contribute to: (a) the accumulation of sub-RPE deposits, (b) the breakdown of Bruch’s membrane, and/or (c) the development of subretinal new vessels. Support for the latter derives from the observation that uPA plays a critical role in angiogenesis in other tissues [60 –62].

244

J.T. Erichsen et al. / The International Journal of Biochemistry & Cell Biology 33 (2001) 237–247

The role of oxygen in modulating proteolytic activity in vivo is an interesting concept which receives support from studies in other tissues [16,17,63,64]. The response is likely to be cell type specific since uPA has been previously reported to both increase and decrease under hypoxic conditions [65,66]. In general, studies in other tissues indicate that PAI is greater at both hypoxia and hyperoxia when compared to normoxia [17,67] The RPE is clearly different since PAI, while being highest under hyperoxic conditions, was lowest under hypoxia. Vitreous PAI levels have been reported to be elevated in diabetic retinopathy, a condition associated with localised retinal ischaemia. However, this may well be related to localised VEGF levels rather than a direct response to hypoxia [68]. Our observations that FGF2 dose-dependently increases uPA production in vitro, but only increases PAI-1 release minimally, suggests that the effect of FGF2 may be to push matrix turnover towards proteolysis. This is especially so in oxygen environments where the pO2 is in the region of normoxia for hRPE cells. By contrast, hypoxia may shift the equilibrium away from proteolysis since both FGF2 and TGF-b1 caused a greater increase in PAI than uPA. This reduced matrix turnover will result in the accumulation of matrix components seen in PVR such as fibronectin, laminin and type IV collagen. This may help explain the fibrotic nature of PVR membranes. It should be noted that the release of growth factors by cultured RPE cells may also affect the uPA/PAI-1 levels. RPE cells are known to synthesise a range of growth factors including IGF-1, PDGF, FGF, VEGF and TGF-b [24,69 – 71], which are known to regulate uPA/PAI levels [10,13,14,72 –74]. Furthermore, the synthesis of these factors may also be directly regulated by oxygen and exogenous growth factors. First, FGF2 and TGF-b can regulate the production of other growth factors in cultured cells [75], and second, oxygen can regulate the production of growth factors by cultured cells [24,76,77].

Acknowledgements This study was supported by the Guide Dogs for

the Blind Association, the Wellcome Trust, British Diabetic Association, Manchester Eye Hospital Endowments, and the North West Regional Health Authority.

References [1] K. Dano, P. Andreasen, J. Grondahl-Hansen, B. Kristensen, L. Nielsen, L. Skriver, Plasminogen activators, tissue degradation and cancer, Adv. Cancer Res. 44 (1985) 146– 239. [2] O. Saksela, Plasminogen activation and regulation of pericellular proteolysis, Biochim. Biophys. Acta. 823 (1985) 35 – 65. [3] M.S. Pepper, J.D. Vassalli, R. Montesano, L. Orci, Urokinase type plasminogen activator is induced in migrating capillary endothelial cells, J. Cell. Biol. 105 (1987) 2535– 2545. [4] E.D. Sprengers, C. Kluft, Plasminogen activator inhibitors, Blood 69 (1987) 381– 387. [5] L. Nielsen, G. Kellerman, N. Behrendt, R. Picone, K. Dano, F. Blasi, A 55 000– 60 000 Mr receptor protein for urokinase-type plasminogen activator, J. Biol. Chem. 263 (1988) 2358– 2363. [6] A. Vaheri, R.W. Stephens, E-M. Salonen, J. Pollanen, H. Tapiovaara, Plasminogen activation at the cell surface– matrix interface, Cell Diff. Dev. 32 (1990) 255– 262. [7] E.F. Plow, L.A. Miles, Plasminogen receptors in the mediation of pericellular proteolysis, Cell Diff. Dev. 32 (1990) 293– 298. [8] D. Collen, H.R. Lijnen, Fibrin-specific fibronolysis, Annals N.Y. Acad. Sci. 667 (1992) 259– 271. [9] J-D. Vassalli, D. Baccino, D. Belin, A cellular binding site for the Mr 55000 form of the human plasminogen activator, urokinase, J. Cell Biol. 100 (1985) 86 – 92. [10] P. Mignatti, R. Mazzieri, D. Rifkin, Expression of the urokinase receptor in vascular endothelial cells is stimulated by basic fibroblast growth factor, J. Cell Biol. 113 (1991) 1193– 1201. [11] M.B. Grant, C. Guay, R. Marsh, Insulin-like growth factor-1 stimulates proliferation, migration, and plasminogen activator release by human retinal pigment epithelial cells, Curr. Eye Res. 9 (1990) 323– 335. [12] M.J. Niedbala, M. Stein-Picarella, Tumor necrosis factor regulation of endothelial cell extracellular proteolysis: the role of urokinase plasminogen activator, Biol. Chem. Hoppe-Seyler 373 (1992) 555– 566. [13] B.I. Gerwin, J. Keski-Oja, M. Seddon, J.F. Lechner, C.C. Harris, TGF-b1 modulation of urokinase and PAI-1 expression in human bronchial epithelial cells, Am. J. Physiol. 259 (1990) L262– L269. [14] J.A. Hamilton, D.S. Piccoli, T. Leizer, D.M. Butler, M. Croatto, A.K. Royston, TGFb stimulates urokinase-type plasmingen activator and DNA synthesis, but not

J.T. Erichsen et al. / The International Journal of Biochemistry & Cell Biology 33 (2001) 237–247

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

prostaglandin E2 production, in human synovial fibroblasts, Proc. Natl. Acad. Sci. USA 88 (1991) 7180–7184. D.B. Rifkin, D. Moscatelli, J. Bizik, N. Quarto, F. Blei, P. Dennis, P. Flaumenhaft, P. Mignatti, Growth factor control of extracellular proteolysis, Cell Diff. Dev. 32 (1990) 313– 318. A. Maity, D. Solomon, Both increased stability and transcription contribute to the induction of the urokinase plasminogen activator receptor (uPAR) message by hypoxia, Exp. Cell Res. 255 (2000) 250–257. T. Uchiyama, M. Kurabayashi, Y. Ohyama, T. Utsugi, N. Akuzawa, M. Sato, S. Tomono, S. Kawazu, R. Nagai, Hypoxia induces transcription of the plasminogen activator inhibitor-1 gene through genistein-sensitive tyrosine kinase pathways in vascular endothelial cells, Arterioscl. Throm. Vas. 20 (2000) 1155–1161. J. Tjarnstrom, L. Holmdahl, P. Arnell, M. Falkenberg, B. Risberg, Treatment with hyperbaric oxygen affects endothelial cell fibrinolysis, Eur. J. Surgery 165 (1999) 834– 838. T.E. Fitzpatrick, C.H. Graham, Stimulation of plasminogen activator inhibitor-1 expression in immortalized human trophoblast cells cultured under low levels of oxygen, Exp. Cell Res. 245 (1998) 155–162. P.A. Campochiaro, J. Mimuro, R. Sugg, D.J. Loskutoff, Retinal pigment epithelial cells produce a latent fibrinolytic inhibitor that is antigenically and biochemically related to type 1 plasminogen activator inhibitor produced by vascular endothelial cells, Exp. Eye Res. 49 (1989) 195– 203. V. Siren, R.W. Stephens, E-M. Salonen, A. Vaheri, P. Summanen, I. Immonen, Retinal pigment epithelial cells secrete urokinase-type plasminogen activator and its inhibitor PAI-1, Ophthalmic Res. 24 (1992) 203–212. B.J. Tripathi, J.K. Park, R.C. Tripathi, Extracellular release of tissue plasminogen activator is increased with the phagocytic activity of the retinal pigment epithelium, Invest. Ophthalmol. Vis. Sci. 30 (1989) 2470–2473. H. Matsunaga, J.T. Handa, C.M. Gelfman, L.M. Hjelmeland, The mRNA phenotype of a human RPE cell line at replicative senescence, Molec. Vis. 5 (1999) 39. A. Khaliq, B. Patel, J. Jarvis-Evans, P. Moriarty, D. McLeod, M. Boulton, The effect of oxygen on the production of bFGF and TGFb by retinal cells in vitro, Exp. Eye Res. 60 (1995) 415–424. M.E. Boulton, A.M. Khaliq, D. McLeod, P.A. Moriarty, Effect of ‘hypoxia’ on the proliferation of microvascular cells in vitro, Exp. Eye Res. 59 (1994) 243–246. A. Khaliq, J. Jarvis-Evans, D. McLeod, M.E. Boulton, Oxygen modulates the response of retinal pigment epithelium to bFGF and EGF via receptor regulation, Invest. Ophthalmol. Vis. Sci. 37 (1996) 436–443. R.W. Nash, B.S. McKay, J.M. Burke, The response of cultured human retinal pigment epithelium to hypoxia: a comparison to other cell types, Invest. Ophthalmol. Vis. Sci. 35 (1994) 2850–2856.

245

[28] B.M. Glaser, M. Lemor, Pathobiology of proliferative vitreoretinopathy, in: S. Ryan, B.M. Glaser (Eds.), Retina, vol 3, Mosby, St Louis, 1994, pp. 2249– 2263. [29] H. Laqua, R. Machemer, Glial cell proliferation in the retinal detachment (massive periretinal proliferation), Am. J. Ophthalmol. 80 (1975) 602– 618. [30] R. Machemer, H. Laqua, Pigment epithelium proliferation in retinal detachment (massive periretinal proliferation), Am. J. Ophthalmol. 80 (1975) 1 – 23. [31] P. Hiscott, I. Grierson, Retinal Detachment, in: A. Garner, G.K. Klintworth (Eds.), Pathobiology of Ocular Disease — A Dynamic Approach, Marcell Decker, New York, 1994, pp. 675– 699. [32] N. Maeda, Y. Tano, T. Ikeda, T. Imai, H. Hamano, R. Manabe, Oxygen tensions in the vitreous cavity of a diabetic patient, Invest. Ophthalmol. Vis. Sci. 31 (1990) S124. [33] S.R. Planck, J. Andresevic, J.C. Chen, D.L. Holmes, W. Rodden, I. Westra, S.C. Wu, X.N. Huan, G. Kay, D.G. Wilson, J.E. Robertson, J.T. Rosenbaum, Expression of growth factor mRNA in rabbit PVR model systems, Curr. Eye Res. 11 (1992) 1031– 1039. [34] P. Weidemann, Growth factors in retinal diseases: proliferative vitreoretinopathy, proliferative diabetic retinopathy, and retinal degeneration, Surv. Ophthalmol. 36 (1992) 373– 384. [35] A. Hueber, P. Wiedemann, P. Esser, K. Heimann, Basic fibroblast growth factor mRNA, bFG peptide and FGF receptor in epiretinal membranes of intreaocular proliferative disorders (PVR and PDR), Int. Ophthalmol. 20 (1996) 345– 350. [36] T.B. Conner, A.B. Roberts, M.B. Sporn, D. Danielpour, L.L. Dart, R.G. Michels, S. de Bustros, C. Enger, H. Kato, M. Lansing, H. Hayasshi, B. Glaser, Correlation of fibrosis and transforming growth factor beta type 2 levels in the eye, J. Clin. Invest. 83 (1989) 1661– 1666. [37] R. Osusky, D. Soriano, J. Ye, S.J. Ryan, Cytokine effect on fibronectin release by retinal pigment epithelial cells, Curr. Eye Res. 13 (1994) (1994) 569– 574. [38] M. Boulton, J. Marshall, J. Mellerio, Retinitis pigmentosa: a preliminary report on tissue culture studies of retinal pigment epithelial cells from eight affected human eyes, Exp. Eye Res. 37 (1983) 307– 313. [39] A. Granelli-Piperno, E. Reich, A study of protease– inhibitor complexes in biological fluids, J. Exp. Med. 146 (1978) 223– 234. [40] H.C. Wong, M.E. Boulton, J. Marshall, P. Clark, Growth of retinal capillary endothelial cells using pericyte conditioned medium, Invest. Ophthalmol. Vis. Sci. 28 (1987) 1767– 1775. [41] R.I. Freshney, Culture of animal cells: a manual of basic techniques, Wiley-Liss Inc., New York, 1987. [42] J.M. Bland, D.G. Altman, Multiple significance test: the Bonferroni method, BMJ 310 (1995) 170. [43] F. Blasi, N. Behrendt, M.Vittoria Cubellis, V. Ellis, L.R. Lund, M. Teresa Masucci, L.B. Moller, D.P. Olson, N.

246

[44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54]

[55]

[56]

J.T. Erichsen et al. / The International Journal of Biochemistry & Cell Biology 33 (2001) 237–247 Pedersen, M. Ploug, E. Ronne, K. Dano, The urokinase receptor and regulation of cell surface plasminogen activation, Cell Diff. Dev. 32 (1990) 247–254. J.P. Alexander, J.M.B. Bradley, J.D. Gabuorel, T.S. Acott, Expression of matrix metalloproteinases and inhibitor by human retinal pigment epithelium, Invest. Ophthalmol. Vis. Sci. 31 (1990) 2520–2528. M.S. Pepper, R. Montesano, Proteolytic balance and capillary morphogenesis, Cell Diff. Dev. 32 (1990) 319– 328. V. Siren, H. Myohanen, A. Vaheri, I. Immonen, Transforming growth factor beta induces urokinase receptor expression in cultured retinal pigment epithelial cells, Ophthalmic Res. 31 (1999) 184–191. S.M. Wileman, N.A. Booth, N. Moore, B. Redmill, J.V. Forrester, R.M. Knott, Regulation of plasminogen activation by TGF-B in cultured human retinal endothelial cells, Br. J. Ophthalmol. 84 (2000) 417–422. S.F. Hackett, P.A. Campochiaro, Modulation of plasminogen activator inhibitor-1 and urokinase in retinal pigment epithelial cells, Invest. Ophthalmol. Vis. Sci. 34 (1993) 2055– 2061. S. Kourembanas, R. Hannon, D. Faller, Oxygen tension regulates the expression of the platelet-derived growth factor-B chain gene in human endothelial cells, J. Clin. Invest. 86 (1990) 670–674. S. Kourembanas, P.A. Marsden, L.P. McQuillan, D.V. Faller, Hypoxia induces endothelin gene expression and secretion in cultured human endothelium, J. Clin. Invest. 88 (1991) 1054– 1057. V. Falanga, S.W. Qian, D. Danielpour, M.H. Katz, A.B. Roberts, M.B. Sporn, Hypoxia upregulates the synthesis of TGFb1 by human dermal fibroblasts, J. Invest. Dermatol. 97 (1991) 634– 637. R. Shreeniwas, S. Ogawa, F. Cozzolino, G. Torcia, N. Braunstein, C. Butura, J. Brett, H.B. Lieberman, M.B. Furie, J.J. Silverstein, D. Stern, Macrovascular and microvascular endothelium during long term hypoxia: alterations in cell growth, monolayer permeability, and cell surface coagulant properties, J. Cell. Physiol. 146 (1991) 8– 17. G.A. Lutty, K. Ikeda, C. Chandler, D.S. McLeod, Immunolocalization of tissue plasminogen activator in the diabetic and nondiabetic retina and choroid, Invest. Ophthalmol. Vis. Sci. 32 (1991) 237–245. R.C. Tripathi, B.J. Tripathi, J.K. Park, Localization of urokinase-type plasminogen activator in human eyes: an immunocytochemical study, Exp. Eye Res. 51 (1990) 545– 552. S.G. Robbins, R.N. Mixon, D.J. Wilson, C.E. Hart, J.E. Robertson, I. Westra, S.R. Planck, J.T. Rosenbaum, Platelet-derived growth factor ligands and receptors immunolocalized in proliferative retinal diseases, Invest. Ophthalmol. Vis. Sci. 35 (1994) 3475–3485. D. Fredj-Reygrobellet, C. Baudouin, F. Negre, J.P. Caruelle, P. Gastaud, P. Lapalus, Acidic FGF and other

[57]

[58]

[59]

[60]

[61]

[62]

[63]

[64]

[65]

[66]

growth factors in preretinal membranes from patients with diabetic retinopathy and proliferative vitreoretinopathy, Ophthalmic Res. 23 (1991) 154– 163. P.A. Campochiaro, J.A. Jerdan, B.M. Glaser, A. Cardin, R.G. Michels, Vitreous aspirates from patients with proliferative vitreoretinopathy stimulate retinal pigment epithelial cell migration, Arch. Ophthalmol. 103 (1985) 1403– 1405. C. Baudouin, D. Fredj-Reygrobellet, F. Brignole, F. Negre, P. Lapalus, P. Gastaud, Growth factors in vitreous and subretinal fluid cells from patients with proliferative vitreoretinopathy, Ophthalmic Res. 25 (1993) 52 – 59. D.J.H. KauffArtermann, J.C. van Meurs, D.A.E. Mertens, E. Peperkamp, C. Master, M.E. Gerritsen, Cytokines in vitreous humor: Interleukin-6 is elevated in proliferative vitreoretinopathy, Invest. Ophthalmol. Vis. Sci. 35 (1994) 900– 906. D. Ribatti, D. Leali, A. Vacca, R. Giuliani, A. Gualandris, L. Roncali, M.L. Nolli, M. Presta, In vivo angiogenic activity of urokinase: role of endogenous fibroblast growth factor-2, J. Cell Sci. 112 (1999) 4213– 4221. S. Heymans, A. Luttun, D. Nuyens, G. Theilmeirer, E. Creemers, L. Moons, G.D. Dyspersin, J.P. Cleutjens, M. Shipley, A. Angellilo, M. Levi, O. Nube, A. Baker, E. Keshet, F. Lupu, J.M. Herbert, J.F. Smits, S.D. Shapiro, M. Baes, M. Borgers, D. Collen, M.J. Daemen, P. Carmeliet, Inhibition of plasminogen activators or matrix metalloproteinases prevents cardiac rupture but impairs therapeutic angiogenesis and causes cardiac failure, Nat. Med. 5 (1999) 1135– 1142. R. Swiercz, E. Skrzypczak-Jankun, M.M. Merrell, S.H. Selman, J. Jankun, Angiostatic activity of synthetic inhibitors of urokinase type plasminogen activator, Oncol. Rep. 6 (1999) 523– 526. T. Kietzmann, U. Roth, K. Jungermann, Induction of the plasminogen activator inhibitor-1 gene expression by mild hypoxia via a hypoxia response element binding the hypoxia-inducible factor-1 in rat hepatocytes, Blood 94 (1999) 4177– 4185. S.F. Yan, N. Mackman, W. Kisiel, D.M. Stern, D.J. Pinsky, Hypoxia/hypoxemia induced activation of the procoagulant pathways and the pathogenesis of ischemiaassociated thrombosis, Arterioscl. Throm. Vas. 19 (1999) 2029– 2035. T. Ito, K. Takenaka, S. Yoshimura, K. Hayashi, S. Noda, N. Sakai, Elevation of mRNA levels of tissue-type plasminogen activator and urokinase-type plasminogen activator in hippocampus and cerebel cortex following middle cerebral artery occlusion in rats, Neurol. Res. 22 (2000) 413– 419. D.J. Pinsky, H. Liao, C.A. Lawson, S.F. Yan, J. Chen, P. Carmeliet, D.J. Loskutoff, D.M. Stern, Co-ordinated induction of plasminogen activator inhibitor-1 (PAI-1) and inhibition of plasminogen activator gene expression by hypoxia promotes pulmonary vascular fibrin dsposition, J. Clin. Invest. 102 (1998) 919– 928.

J.T. Erichsen et al. / The International Journal of Biochemistry & Cell Biology 33 (2001) 237–247 [67] C. Barazzone, D. Belin, P.F. Piguet, J.D. Vassalli, A.P. Sappino, Plasminogen activator inhibitor-1 in acute hyperoxic mouse lung injury, J. Clin. Invest. 98 (1996) 2666– 2673. [68] L.O. Hattenbach, A. Aller, H.O. Gumbel, I. Scharrer, F.H. Koch, Vitreous concentrations of TPA and plasminogen activator inhibitor are associated with VEGF in proliferative diabetic vitreoretinopathy, Retina 19 (1999) 383– 389. [69] P. Moriarty, M. Boulton, A. Dickson, D. McLeod, Production of IGF-I and IGF binding proteins by retinal cells in vitro, Brit. J. Ophthalmol. 78 (1994) 638–642. [70] J.T. Handa, K.M. Reiser, H. Matsunaga, L.M. Hjelmeland, The advanced glycation endproduct pentosidine induces the expression of PDGF-B in human retinal pigment epithelial cells, Exp. Eye Res. (1998) 411–419. [71] Y. Seko, Y. Seko, H. Fujikura, J. Pang, T. Tokoro, H. Shimokawa, Induction of vascular endothelial growth factor after application of mechanical stress to retinal pigment epithelium of the rat in vitro, Invest. Ophthalmol. Vis. Sci. 40 (1999) 3287–3291. [72] B. Cullen, D. Silcock, L.J. Brown, A. Gosiewska, J.C. Geesin, The differential regulation and secretion of proteinases from fetal and neonatal fibroblasts by growth factors, Int. J. Biochem. Cell Biol. 29 (1997) 241–250.

.

247

[73] G. Fibbi, M. Pucci, C. Grappone, G. Pellegrini, R. Salzano, A. Casini, S. Milani, M. Del Rosso, Functions of the fibrinolytic system in human Ito cells and its control by basic fibroblast and platelet-derived growth factor, Hepatology 29 (1999) 868– 878. [74] S.J. Mandriota, M.S. Pepper, Vascular endothelial growth factor-induced in vitro angiogenesis and plasminogen activator expression are dependent on endogenous basic fibroblast growth factor, J. Cell Sci. 110 (1997) 2293– 2302. [75] P. Choudhury, W. Chen, R.C. Hunt, Production of platelet-derived growth factor by interleukin-1 beta and transforming growth factor-beta-stimulated retinal pigment epithelial cells leads to contraction of collagen gels, Invest. Ophthalmol. Vis. Sci. 38 (1997) 824– 833. [76] S.A. Mousa, W. Lorelli, P.A Campochiaro, Role of hypoxia and extracellular matrix-integrin binding in the modulation of angiogenic growth factors secretion by retinal pigmented epithelial cells, J. Cell Biochem. 741 (1999) 135– 143. [77] R.S. Punglia, M. Lu, J. Hsu, M. Kuroki, M.J. Tolentino, K. Keough, A.P. Levy, N.S. Levy, M.A. Goldberg, R.J. D’Amato, A.P. Adamis, Regulation of vascular endothelial growth factor expression by insulin-like growth factor I, Diabetes 46 (1997) 1619– 1626.