Matrix metalloproteinase 12dependent cleavage of urokinase receptor in systemic sclerosis microvascular endothelial cells results in impaired angiogenesis.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 50, No. 10, October 2004, pp 3275–3285 DOI 10.1002/art.20562 © 2004, American College of Rheumatology Matrix Metalloproteinase 12–Dependent Cleavage of Urokinase Receptor in Systemic Sclerosis Microvascular Endothelial Cells Results in Impaired Angiogenesis Silvia D’Alessio,1 Gabriella Fibbi,1 Marina Cinelli,1 Serena Guiducci,1 Angela Del Rosso,1 Francesca Margheri,1 Simona Serratı̀,1 Marco Pucci,1 Bashar Kahaleh,2 Pansheng Fan,2 Francesco Annunziato,1 Lorenzo Cosmi,1 Francesco Liotta,1 Marco Matucci-Cerinic,1 and Mario Del Rosso1 shown by flow cytometry, enzyme-linked immunosorbent assay, and Western blotting, a cleavage that is known to impair uPAR functions. These properties of SSc MVECs were associated with poor spontaneous and uPA-dependent invasion, proliferation, and capillary morphogenesis. The uPAR cleavage occurring in SSc MVECs was associated with overexpression of MMP-12. SSc MVEC–conditioned medium impaired uPAdependent proliferation and invasion as well as capillary morphogenesis in normal MVECs in vitro. Both a general hydroxamate inhibitor of MMP activity and anti–MMP-12 antibodies restored this SSc MVEC– induced impaired functioning. Conclusion. Overproduction of MMP-12 by SSc MVECs accounts for the cleavage of uPAR and the impairment of angiogenesis in vitro and may contribute to reduced angiogenesis in SSc patients. Objective. Defective angiogenesis, resulting in tissue ischemia, is particularly prominent in the diffuse form of systemic sclerosis (SSc). The present study was undertaken to identify possible differences between normal and SSc microvascular endothelial cells (MVECs) in the expression of the cell-associated urokinase-type plasminogen activator (uPA)/uPA receptor (uPAR) system, which is critical in the angiogenic process. Methods. MVECs were isolated from the dermis of healthy individuals and from the dermis of patients with diffuse SSc. The uPA/uPAR system was examined at the protein and messenger RNA levels. Angiogenesis was assayed on Matrigel-coated porous filters and plates to evaluate cell proliferation, invasion, and capillary morphogenesis. Cleavage of uPAR and the activity of matrix metalloproteinase 12 (MMP-12) were evaluated by Western blotting. Results. Compared with MVECs from healthy skin, MVECs from SSc patients showed higher expression of uPAR. However, in SSc MVECs, uPAR undergoes truncation between domain 1 and domain 2, as Systemic sclerosis (SSc) affects the skin and the internal organs, resulting in tissue fibrosis and organ dysfunction. The pathogenesis of SSc is characterized by 3 distinct processes: 1) microvascular alterations, including capillary endothelial cell injury, 2) perivascular extravasation of lymphocytes, and 3) accumulation of collagen in the dermis. The microvascular alterations usually precede the appearance of clinically detectable fibrosis. Capillaroscopy has shown that microvascular modifications are characterized by megacapillaries and eventual desertification of the nailfold (1). The mechanism of microvascular injury is still unknown. Some authors report that sera from patients with acrosclerosis potentiate the angiogenic capability of normal human mononuclear cells (2,3). In diffuse SSc, a subset of mononuclear cells displays enhanced angio- Supported by grants from the Scleroderma Foundation (004/ 01) and the Ministero dell’Istruzione, Università e Ricerca (Progetti di Ricerca di Interesse Nazionale). 1 Silvia D’Alessio, PhD, Gabriella Fibbi, PhD, Marina Cinelli, PhD, Serena Guiducci, MD, Angela Del Rosso, MD, Francesca Margheri, PhD, Simona Serratı̀, PhD, Marco Pucci, PhD, Francesco Annunziato, PhD, Lorenzo Cosmi, MD, Francesco Liotta, MD, Marco Matucci-Cerinic, MD, Mario Del Rosso, MD: University of Florence, Florence, Italy; 2Bashar Kahaleh, MD, Pansheng Fan, PhD: Medical College of Ohio, Toledo. Address correspondence and reprint requests to Mario Del Rosso, MD, Professor of General Pathology, Department of Experimental Pathology and Oncology, Viale G. B. Morgagni, 50, Florence 50134, Italy. E-mail: firstname.lastname@example.org. Submitted for publication February 3, 2004; accepted in revised form June 30, 2004. 3275 3276 D’ALESSIO ET AL genic activity (4). In contrast, impaired angiogenic activity has been reported in peripheral blood lymphocytes and monocytes in the diffuse subset of SSc (5,6), although Ribatti et al (7) have shown that angiogenesis is stimulated on the chick embryo chorioallantoic membrane by skin biopsy tissue from patients with diffuse SSc. Circulating endostatin concentrations are significantly increased in patients with SSc, and this may contribute to an overall decrease of angiogenesis in the course of the disease (8). Purportedly, microvascular endothelial cells (MVECs) can perform neoangiogenesis only when provided with the proper enzymatic machinery, enabling them to lyse extracellular matrix (ECM) and to enter the surrounding tissue. The cell-associated plasminogen activator system is known to play a crucial role in the angiogenesis process by modulating the adhesive properties of endothelial cells in their interactions with ECM and in the degradation of ECM (9–13). We have demonstrated that urokinase-type plasminogen activator (uPA) induces neovascular growth in the avascular rabbit cornea and promotes growth, chemotaxis, and matrix invasion of endothelial cells in vitro, by interaction with its receptor, uPAR (11). The uPA required for invasion can be produced by endothelial cells themselves, as well as by other cells of the microenvironment in which angiogenesis develops, such as fibroblasts and macrophages. In this study, we isolated MVECs from the dermis of normal, healthy individuals and from the dermis of patients with diffuse SSc, to study the relationship between angiogenesis and the endothelial cell uPA/ uPAR system. Our results indicate that SSc MVECs are prevented from entering an angiogenic program by matrix metalloproteinase 12 (MMP-12)–dependent cleavage of domain 1 (D1) of uPAR, a structural modification that prevents the functioning of uPAR in endothelial cell invasion and proliferation. PATIENTS AND METHODS Antibodies and reagents. Anti-CD87 mouse monoclonal antibody (mAb), which identifies total uPAR, mouse mAb anti–uPAR domain 1, and mouse mAb anti–uPAR domain 2 were purchased from American Diagnostica (Montreal, Quebec, Canada). Mouse mAb 5B4, which recognizes the kringle domain of the A-chain of uPA and prevents uPA–uPAR interaction (11,14), was a kind gift from Dr. M. L. Nolli (Lepetit Research Center, Varese, Italy). Rabbit mAb against the hinge region of MMP-12 was from Chemicon International (Temecula, CA). The uPA was obtained from Serono (Rome, Italy), and endothelial cell growth supplement (ECGS) was purchased from Calbiochem (Nottingham, UK). Reagents for messenger RNA (mRNA) studies were from Promega Italia (Milan, Italy), and other reagents for cell culture, including fetal calf serum (FCS), were purchased from Sigma (St. Louis, MO). Diff-Quick was from Mertz-Dade (Dade International, Milan, Italy). Matrigel was obtained from Becton Dickinson (Bedford, MA). Cells. Patients with SSc and healthy controls were used as the sources of MVECs. MVECs were isolated from biopsy samples of involved skin of the hands of 6 SSc patients who underwent skin biopsies for diagnostic purposes at the Department of Medicine (Section of Rheumatology) of Florence University, and from 15 healthy patients undergoing surgery for trauma to the hands. The study was approved by the local ethics committee, and written informed consent was obtained from each subject enrolled. The patients were not receiving steroids, cyclophosphamide, D-penicillamine, relaxin, or other disease-modifying drugs. Calcium channel blockers were stopped 10 days before the biopsy. Only proton-pump inhibitors and cisapride were allowed. To obtain the cells, skin biopsy samples were mechanically cleaned of epidermis and adipose tissue in order to obtain a pure specimen of vascularized dermis, and were treated as described elsewhere (15). In some cases, clusters of round-shaped cells were squeezed from microvessels and formed colonies composed of polygonal elements. Such colonies were detached with EDTA, and CD31-positive cells were subjected to immunomagnetic isolation with Dynabeads CD31 (Dynal ASA, Oslo, Norway) (16). Isolated cells were further identified as MVECs by labeling with anti–factor VIII–related antigen and by reprobing with anti-CD31 antibodies. Cells were maintained in complete MCDB medium supplemented with 30% FCS, 20 g/ml ECGS, 10 g/ml hydrocortisone, 15 UI/ml heparin, and antibiotics (100 UI/ml penicillin, 100 g/ml streptomycin, 50 g/ml amphotericin). MVECs from healthy individuals and from SSc patients were used between the third and tenth passages in culture. Enzyme-linked immunosorbent assays (ELISAs) for uPA, uPAR, and uPAR fragments. The IMUBIND total uPAR ELISA kit (American Diagnostica) was used to determine the expression of uPAR in aliquots of cell lysates. The D1 domain of uPAR was measured by a modification of the total uPAR ELISA kit, using murine mAb anti-human uPAR D1 as the capture antibody. Quantification of uPAR D1 was carried out by comparing the values with those of a standard curve, while uPA was quantified by U-PA Actibind ELISA (Technoclone, Wien, Austria). Flow cytometry analysis. Flow cytometry analysis on MVEC suspensions was performed as described previously (17,18). Briefly, 3 ⫻ 105 cells were suspended in phosphate buffered saline (PBS) containing 0.5% bovine serum albumin and 0.02% sodium azide, and after saturation of nonspecific binding sites with total rabbit IgG, cells were incubated for 20 minutes with specific or isotype control mAb. Cells were then washed and incubated with a fluorochrome-conjugated antiisotype antibody for an additional 20 minutes. Finally, cells were washed and analyzed on a BD-LSR cytofluorimeter (Becton Dickinson). Migration assays. The Boyden chamber was used to evaluate spontaneous and stimulated invasion (chemoinvasion) of cells through Matrigel-coated porous filters, as described previously (11,14). For assessment of spontaneous IMPAIRED SSc MVEC ANGIOGENESIS BY MMP-12 CLEAVAGE OF uPAR invasion, 50 l of cell suspension (6.25 ⫻ 103 cells) was placed in the upper compartment of the chamber, and migration was allowed to occur for 6 hours. For assessment of chemoinvasion, test solutions containing increasing concentrations of uPA (5–500 ng/ml) were dissolved in serum-free medium and placed in the lower wells. In the experiments with neutralizing antibodies, the anti-uPA mAb (1.5 g/ml) was added to the lower well with the test substance (uPA), while the anti-uPAR mAb (1.5 g/ml) was incubated with the cell suspension. Irrelevant mouse IgG was used to verify the specificity of the effect. The number of cells moving across the filter was used as the measure of mobilization. Experiments were performed in triplicate. Migration values were expressed as the mean ⫾ SD number of total cells counted per filter, or as a percentage of the basal response value. Proliferation studies. Cell growth was quantified in subconfluent cell monolayers, as described previously (11,14). MVECs were seeded onto 24 multiwell plates (Sarstedt, Verona, Italy) (15 ⫻ 103 cells/well) in complete MCDB medium, and were left to adhere overnight. Cells were then extensively washed in PBS and maintained for 24 hours in 1% FCS–MCDB medium. Medium was removed and cells were incubated with 1% FCS–MCDB medium containing increasing concentrations of uPA for 48 hours. The proliferative effect of 30% FCS was defined as the optimal growth. In the experiments with neutralizing mAb, cells were challenged with uPA in the presence of the mAb (1.5 g/ml). Irrelevant mouse IgG was used as a negative control. Each experimental point was performed in triplicate. At the end of the incubation, cells were fixed by adding 1 ml of ice-cold methanol, stained with Diff-Quick, and counted. Values were expressed as the percentage increase or decrease over the basal response value. Detection of mRNA transcripts by semiquantitative reverse transcription–polymerase chain reaction (RT-PCR), and quantification of PCR products. Levels of mRNA for uPAR and uPA genes in normal, healthy MVECs and SSc MVECs were determined by an internal-based semiquantitative RT-PCR, using procedures and primers previously described (19). For MMP-12, 2 l of reverse-transcribed DNA was amplified, using MMP-12–specific sense (5⬘-CCA-CTGCTT-CTG-GAG-CTC-TT-3⬘) and antisense (5⬘-GCG-TAGTCA-ACA-TCC-TCA-CG-3⬘) primers, or GAPDH-specific sense (5⬘-CCA-CCC-ATG-GCA-AAT-TCC-ATG-GCA-3⬘) and antisense (5⬘-TCT-AGA-CGG-CAG-GTC-AGG-TCCACC-3⬘) primers as a control. Semiquantitative PCR was performed for 35 cycles at 58°C in a thermocycler. The reaction products were analyzed by electrophoresis in 1% agarose gel containing ethidium bromide, followed by photography under ultraviolet illumination using Polaroid positive/ negative instant films, and quantified as reported elsewhere (19). In vitro capillary morphogenesis assay. Matrigel (0.5 ml; 10–12 mg/ml) was pipetted into 13-mm (diameter) tissueculture wells and polymerized for 30 minutes to 1 hour at 37°C (20,21). MVECs were plated (60 ⫻ 103/ml) in complete MCDB medium supplemented with 30% FCS and 20 g/ml ECGS. Capillary morphogenesis was also performed in the presence of 1.5 g/ml CD87 mAb (anti-uPAR) or 5B4 mAb (anti-uPA). Positive controls were obtained upon stimulation with 10 ng/ml vascular endothelial cell growth factor (VEGF). Irrelevant mouse IgG was used as a negative control. In other cases, normal MVECs were incubated with conditioned me- 3277 dium containing SSc MVECs, either alone or supplemented with antibodies to MMP-12 (10 g/ml). Plates were photographed at 6 hours and at 24 hours. Results were quantified at 6 hours by measuring the percentage of the photographic field occupied by endothelial cells, as determined by image analysis. Six to nine photographic fields from 3 plates were scanned for each point. Preparation of SSc cell–conditioned medium. Confluent cultures of normal MVECs and SSc MVECs were washed twice with PBS and incubated overnight in the presence of MCDB medium supplemented with 2% FCS. The culture supernatant was centrifuged at 1,500 revolutions per minute for 10 minutes, and either used immediately or stored at ⫺20°C. Western blotting, treatment with protease inhibitors, and anti–MMP-12. Normal MVECs were grown to 70% confluence and were serum-starved overnight in MCDB medium supplemented with 2% FCS. The cells were then incubated with or without SSc MVEC–conditioned medium for 24 hours, in the presence or in the absence of 10 g/ml of anti–MMP-12 mAb or protease inhibitors (a general hydroxamate inhibitor of MMP activity, Ilomastat, or GM6001, at 25 M [Chemicon International]; a serine-protease inhibitor, aprotinin, at 1 g/ml [Sigma]; or a selective low molecular weight inhibitor of cysteine protease, E-64, at 200 M, kindly provided by Professor Francesco Paoletti from the University of Florence). Cells were then suspended in lysis buffer (10 mM Tris HCl, pH 7.4, containing 150 mM NaCl, 1% Triton X-100, 15% glycerol, 1 mM sodium orthovanadate, 1 mM NaF, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, and 10 g/ml aprotinin). Sixty micrograms of the cell extract protein was electrophoresed in sodium dodecyl sulfate–12% polyacrylamide gels under reducing conditions and then blotted to a polyvinylidene difluoride membrane (Hybond-C Extra; Amersham Biosciences, Piscataway, NJ) for 3 hours at 35V. The membrane was incubated with 5% skim milk in 20 mM Tris buffer, pH 7.4, for 1 hour at room temperature to block nonspecific binding and then probed with a mouse mAb directed against the first domain (D1) as well as the second domain (D2) of uPAR (American Diagnostica, Stamford, CT) overnight at 4°C. The same procedure was used to perform Western blotting of aliquots of MVEC culture medium, to identify MMP-12 with anti–MMP-12 antibodies. After incubation with horseradish peroxidase–conjugated donkey antimouse IgG (1:5,000) for 1 hour (Amersham Biosciences), immune complexes were detected with the enhanced chemiluminescence detection system (Amersham Biosciences). The membranes were exposed to autoradiographic films (Hyperfilm MP; Amersham Biosciences) for 1–30 minutes. Statistical analysis. Results are expressed as the mean ⫾ SD. Multiple comparisons were performed by the Student-Newman-Keuls test, after demonstration of significant differences among medians by nonparametric variance analysis using the Kruskal-Wallis test. P values less than 0.05 were considered significant. RESULTS Hypofunctionality of the uPA/uPAR system in SSc MVECs. MVEC expression. Characterization of the uPA/uPAR system was performed on confluent mono- 3278 D’ALESSIO ET AL Figure 1. Spontaneous invasion, urokinase-type plasminogen activator (uPA)–induced chemoinvasion, and proliferation of normal, healthy microvascular endothelial cells (N-MVEC) and systemic sclerosis endothelial cells (SSc-MVEC). A, To assess spontaneous invasion of Matrigel-coated filters in the Boyden chamber assay, 6.25 ⫻ 103 cells were placed in the upper well and invasion was allowed to occur for 6 hours. Each blocking monoclonal antibody (mAb) as well as irrelevant mouse IgG were used at 1.5 g/ml. B, uPA-dependent chemoinvasion of Matrigel in the Boyden chamber. Results are the total number of cells/filter after 6 hours. C, Effect of blocking mAb and of irrelevant IgG on uPA-dependent chemoinvasion. Results are the percentage variation in relation to the migration of unstimulated cells (control), considered to be 100%. D, Response of N-MVEC and SSc-MVEC to uPA challenge. E, uPA-dependent proliferation with blocking mAb and irrelevant mouse IgG. Results are the percentage variation in relation to the number of cells at the moment of plating (150 ⫻ 103/well), considered to be 100%. Bars show the mean ⫾ SD of 3 experiments performed in triplicate in 3 normal and 3 SSc cell lines. ⴱ ⫽ P ⬍ 0.05 versus control. layers of cells for which CD31 staining yielded positive results in more than 90% of the cell population. No difference in cell density, once at confluence, was observed between normal MVECs and SSc MVECs, which allowed us to perform experiments in conditions of similar cell density. ELISA determination of uPA in the culture medium indicated a level of 31.0 ⫾ 9.1 ng/ml in both cell lines. In contrast, we found that uPAR was lightly overexpressed in SSc MVECs (15.3 ⫾ 3.0 ng/106 cells in normal MVECs compared with 23.1 ⫾ 4.3 ng/106 cells in SSc MVECs; P ⬍ 0.05). Semiquantitative RTPCR analysis of the specific mRNA confirmed the data obtained at the protein level (results not shown). MVEC invasion. Invasion of the surrounding matrix by endothelial cells is a necessary step in the angiogenic program. Figure 1A shows that normal MVECs and SSc MVECs were differentially prone to spontaneous invasion of Matrigel-coated filters. Invasiveness of normal MVECs (650 ⫾ 57 cells/filter) was about twice that of SSc MVECs (330 ⫾ 35 cells/filter). IMPAIRED SSc MVEC ANGIOGENESIS BY MMP-12 CLEAVAGE OF uPAR 3279 Figure 2. Capillary morphogenesis in N-MVEC and SSc-MVEC. N-MVEC and SSc-MVEC were plated on Matrigel (60 ⫻ 103/ml), in complete MCDB medium supplemented with 30% fetal calf serum and 20 g/ml endothelial cell growth supplement. N-MVEC spontaneously formed anastomosing cords of cells, resembling a capillary plexus, which were well organized by 6 hours (A), and showed a complete honeycomb-like network after 24 hours (B). When the branching cords of N-MVEC are completely organized, the total number of cells tend to occupy less surface. In the presence of 250 ng/ml uPA, meshwork formation was increased (C). The process of endothelial cell organization after 6 hours was impaired in the presence of 1.5 g/ml anti-uPA mAb 5B4 (G) and anti–uPA receptor mAb CD87 (H). Positive control with 10 ng/ml vascular endothelial cell growth factor (VEGF) (I) is compared with the effect of irrelevant mouse IgG (J). SSc-MVEC failed to form an organized network in Matrigel after 6 hours (D) and after 24 hours (E). The effect of 250 ng/ml uPA on SSc-MVEC is shown in F. Results are representative of 3 different experiments performed on 3 normal and 3 SSc cell lines. The values shown are the mean ⫾ SD percentage of the photographic field occupied by cells, and are reported only for N-MVEC, because SSC-MVEC organization was scarce and irregular. ⴱ ⫽ P ⬍ 0.05 versus control N-MVEC at 6 hours after plating on Matrigel. (Original magnification ⫻ 100.) See Figure 1 for other definitions. The mAb against uPA (5B4) and the mAb against uPAR (anti-CD87) were able to impair the extent of invasion of normal MVECs and SSc MVECs, whereas irrelevant mouse IgG had minimal effect. Since the 5B4 mAb prevents uPA–uPAR interaction, it is likely that in the spontaneous invasion assay, MVECs exploit the uPAR association with uPA as it is picked up during the autocrine process or from the microenvironment. In turn, the activity of the anti-CD87 mAb, which binds equally to occupied and unoccupied uPAR, may rely on impairment of the adhesive interactions of uPAR with Matrigel-coated ECM macromolecules. Since the angiogenic microenvironment is rich in uPA produced by MVECs themselves and by fibroblasts and macrophages, MVECs were also challenged with increasing concentrations of uPA (5–500 ng/ml) to stimulate invasion of Matrigel-coated filters (Figure 1B). The uPA-dependent chemoinvasion was observed to be 3280 dose-dependent, with a maximal effect at 250 ng/ml, which was much higher for the normal MVECs. The same mAb that efficiently blocked spontaneous invasion also impaired uPA-dependent mobilization of MVECs, whereas irrelevant mouse IgG did not (Figure 1C). The antibodies had no effect on the viability of normal and SSc MVECs, as evaluated by trypan blue dye exclusion. MVEC proliferation. As shown in Figure 1D, uPA dose-dependently induced proliferation of normal MVECs and SSc MVECs, with normal MVECs showing a stronger proliferative response. The same mAb used in the invasion experiments were able to inhibit uPAdependent proliferation (Figure 1E). MVEC capillary morphogenesis. The final angiogenic event requires MVECs to organize in interconnecting tubular structures to form the new capillaries. To study the role of the fibrinolytic system, we undertook studies of capillary morphogenesis on Matrigel matrices. Normal MVECs and SSc MVECs were seeded at low density on plates coated with Matrigel (20,21). In this assay, MVEC produce elongated processes which eventuate in the formation of anastomosing cords of cells resembling a capillary plexus (20). The ability to form a complete network can be sustained by the addition of angiogenic growth factors to culture medium. Figure 2A shows that, after 6 hours of plating on Matrigel, normal MVECs showed abundant networks of branching cords of cells. By 24 hours, normal MVECs formed an interconnected network of anastomosing cells that, by low-power light microscopy, had a honeycomb appearance (Figure 2B). In comparison, by 6 hours of assay, SSc MVECs were unable to produce elongated processes and were unable to organize as branching cords (Figure 2D). Even after 24 hours (Figure 2E), SSc MVECs were less prone to spontaneously organize into a honeycomb morphologic pattern, producing only an incomplete endothelial network. Although Matrigel is rich in uPA, the exogenous addition of uPA was able to augment in vitro morphogenesis by normal MVECs (Figure 2C) but not by SSc MVECs (Figure 2F). Figures 2G and H show that antagonism of the uPA/uPAR system affects capillary morphogenesis. Normal MVECs were unable to form capillary-like structures after 6 hours when cultivated in the presence of the anti-uPA mAb 5B4 (Figure 2G) and anti-uPAR mAb CD87 (Figure 2H), confirming previous data that indicate the critical role of uPA, uPAR, and uPA–uPAR interaction in angiogenesis (10–13). Figures 2I and J show capillary morphogenesis in the presence of 10 ng/ml VEGF and irrelevant mouse IgG, respectively. Addition of VEGF to SSc MVECs did not produce any effect on the organization of the endothelial network D’ALESSIO ET AL Figure 3. Surface expression of CD87 and uPA receptor (uPAR) domain 2 on N-MVEC and SSc-MVEC. A, Dot plots represent the surface expression of CD87 and uPAR domain 2 in N-MVEC and SSc-MVEC. B, Histograms represent the surface expression of domain 2 on N-MVEC before and after treatment with 10 nM uPA for 60 minutes. The gate is established using an isotope-matched mAb. Results in A and B are from 2 distinct experiments. See Figure 1 for other definitions. (results not shown). The different angiogenic response between normal MVECs and SSc MVECs was independent of cell senescence and was observed by performing experiments with cells from the third to the tenth passage in culture. Truncation of uPAR in SSc MVECs. Taken together, our data indicate that the uPA/uPAR system appears to be impaired in SSc MVECs, in spite of the presence of a higher amount of endothelial cell– associated uPAR. We investigated whether such a defect involved alteration of the uPAR structure in SSc MVECs such that SSc MVECs would be rendered less responsive to uPA challenge. RT-PCR showed that primers that encompass the whole uPAR mRNA yielded IMPAIRED SSc MVEC ANGIOGENESIS BY MMP-12 CLEAVAGE OF uPAR amplification products that were indicative of the presence of a complete, nontruncated uPAR in both normal MVECs and SSc MVECs. This suggests no alteration of uPAR at the transcriptional level. These data prompted us to study whether a possible epigenetic uPAR modification could account for the overall decrease of the uPA/uPAR functions in SSc MVECs. Figure 3 shows the flow cytometry results in normal MVECs and SSc MVECs. As shown in Figure 3A, purified normal MVECs and SSc MVECs exhibited comparable CD87 (uPAR) surface expression when stained with an anti-uPAR mAb, whereas by using an mAb specific for the D2 domain of the receptor, the percentage of cells positive for D2 was higher in SSc MVECs in comparison with normal MVECs. Treatment of normal MVECs with 10 nM uPA for 60 minutes, which has been shown to be able to truncate the receptor at the D1 level (22), consistently enhanced the percentage of cells stained with anti-D2 mAb (Figure 3B). These data strongly suggest that anti-D2 mAb recognizes its epitope when D1 is removed. Taken together, these data support the hypothesis that an active uPAR cleavage is likely to occur in SSc MVECs. The ELISA for total uPAR and the D1 domain of uPAR indicated that D1 was present in almost all of the uPAR molecules in normal MVECs, whereas it was detectable in only ⬃40% of the SSc MVEC–associated uPAR, as shown in Table 1. Parallel experiments were performed by Western blotting of aliquots of cell lysates of normal MVECs and SSc MVECs. Figure 4A shows the immunostaining of uPAR with mAb against D1 and mAb against D2. The antibody against uPAR D2 revealed 2 bands, both in normal MVECs and in SSc MVECs, corresponding to the native form (44 kd) and the truncated form (30 kd) of uPAR. Of note, the cleaved form was barely detectable in normal MVECs. The antibody against uPAR D1, which recognizes only the native form of uPAR, extracted from cell lysates showed a more intense staining in normal MVECs than in SSc MVECs, again indicating uPAR truncation in SSc MVECs. Because of the different uPAR affinity of each mAb, this differential intensity of immunostaining be- Table 1. Levels of total uPAR and D1 domain of uPAR in cell lysates of normal MVECs and SSc MVECs* Cells Total uPAR, ng/106 cells D1, ng/106 cells Normal MVECs SSc MVECs 15.3 ⫾ 3.0 23.1 ⫾ 4.3 14.05 ⫾ 3.1 9.33 ⫾ 1.7 * Values are the mean ⫾ SD. uPAR ⫽ urokinase-type plasminogen activator receptor; MVECs ⫽ microvascular endothelial cells; SSc ⫽ systemic sclerosis. 3281 Figure 4. Cleavage of uPAR and effects of matrix metalloproteinase 12 (MMP-12) in N-MVEC and SSc-MVEC. A, Western blots with mAb against uPAR domains 1 and 2 were obtained using proteins (50 g) from cell lysates loaded in sodium dodecyl sulfate–12% polyacrylamide gels under reducing conditions and blotted to a polyvinylidene difluoride membrane. Immune complexes were detected with enhanced chemiluminescence. The third lane in the blots for each mAb shows immunostaining with irrelevant mouse IgG. B, Western blots of proteins of cell lysates (50 g) of N-MVEC incubated with SScMVEC–conditioned medium were examined under control conditions (first lane) or in the presence of the indicated protease inhibitors. Values on the right in A and B indicate the positions of standards with known molecular weights. ␤-actin was used as a control. C, For Western blotting with anti–MMP-12 mAb of 20-l aliquot of culture medium, and reverse transcription–polymerase chain reaction (RTPCR) analysis of MMP-12 and GAPDH (control) in N-MVEC and SSc-MVEC, 5 g of total RNA was reversely transcribed, and reversely transcribed DNA was then amplified with primers. PCR products were analyzed by electrophoresis in 1% agarose gels containing ethidium bromide. Aprot ⫽ aprotinin; Ilom ⫽ ilomastat (see Figures 1 and 3 for other definitions). tween the D1 and D2 mAb must not be considered quantitative. Overexpression of MMP-12 in cleavage of uPAR and impairment of capillary morphogenesis in SSc MVECs. The cleavage of uPAR between D1 and D2 has been shown to depend on the proteolytic activity of uPA, directly or indirectly via the plasminogen/plasmin pathway (22,23) or other proteases, such as chymotrypsin and elastase (24,25). In basic fibroblast growth factor 3282 D’ALESSIO ET AL Figure 5. Reversal of the antiangiogenic effects of SSc-MVEC–conditioned medium (CM SSc) on N-MVEC by inhibition of MMP-12. A, To assess invasion of N-MVEC, unstimulated (control) and uPA-stimulated N-MVEC were subjected to Matrigel migration assay as described in Figure 1. To study the effect of CM SSc on invasion, cells were preincubated 48 hours with CM SSc, and 6.25 ⫻ 103 cells suspended in CM SSc were then placed in the upper well of the migration chamber. Invasion was allowed to occur for 6 hours following the addition in the lower well of CM SSc alone or CM SSc supplemented with the indicated inhibitors. Bars show the mean and SD total number of cells/filter after 6 hours. To assess cell proliferation, 60 ⫻ 103 N-MVEC were seeded in 6 multiwell plates and allowed to grow in MCDB medium supplemented with 30% fetal calf serum (FCS). After 24 hours, cells were washed and left to grow under the various conditions shown (in 2% FCS). Cells were counted 24 hours later. Bars show the mean and SD percentage variations of cell number in relation to the number of plated cells. Results are from 3 experiments performed in triplicate. ⴱ ⫽ P ⬍ 0.05 versus control. B, For capillary morphogenesis assay, N-MVECs were seeded on Matrigel, as described in Figure 2, under the various indicated conditions. Irrelevant mouse IgG was unable to revert the effect of CM SSc (results not shown). Results were obtained after 6 hours and are representative of 3 different experiments. Values shown indicate the mean ⫾ SD percentage of the photographic field occupied by cells (original magnification ⫻ 100). See Figures 1 and 4 for other definitions. (bFGF)/tumor necrosis factor ␣ (TNF␣)– or phorbol myristate acetate (PMA)–stimulated human MVECs, truncation of uPAR D1 and D2 has been shown to be attributable to the hyperactivity of MMP-12 (13). Therefore, we performed a series of experiments to evaluate the class of protease involved in uPAR cleavage in SSc MVECs. Figure 4B shows the results of Western blotting obtained with the mAb against D1. Twenty-four–hour incubation of normal MVECs with conditioned medium containing SSc MVECs strongly reduced D1 expression in normal MVECs (Figure 4B). A similar reduction in the levels of D1 in normal MVECs was observed when SSc-conditioned medium contained a serine-protease inhibitor, aprotinin, or a cysteine-protease inhibitor, E64. In contrast, in the presence of a general hydroxamate inhibitor of MMP activity, Ilomastat, D1/D2 uPAR cleavage did not occur. These results suggested that the presence of a member of the MMP family accounted for the truncation of uPAR D1 in SSc MVECs. Therefore, on the basis of our results and those of previous studies indicating overexpression of MMP-12 in activated human MVECs (13), we incubated normal MVECs for 24 hours with SSc IMPAIRED SSc MVEC ANGIOGENESIS BY MMP-12 CLEAVAGE OF uPAR MVEC–conditioned medium containing a rabbit mAb against the hinge region of MMP-12. As shown in Figure 4B, uPAR D1 cleavage was inhibited by this antibody. RT-PCR analysis showed a significant up-regulation of MMP-12 mRNA in SSc MVECs (Figure 4C, right), and Western blot of aliquots of the culture medium indicated active release of MMP-12 (Figure 4C, left). It is noteworthy that the difference in MMP-12 production between normal MVECs and SSc MVECs is a permanent alteration, which was observed by using cells from the third to the tenth passage in culture. As a consequence of these findings, we performed experiments on uPA-dependent proliferation and chemoinvasion of normal MVECs preincubated overnight with SSc MVEC–conditioned medium, and capillary morphogenesis experiments on normal MVECs maintained in the presence of SSc MVEC– conditioned medium. Under these conditions, chemoinvasion and proliferation of normal MVECs (Figure 5A) were inhibited, and capillary morphogenesis was heavily impaired (Figure 5B), even in the presence of 250 ng/ml uPA. The same experiments, performed in the presence of SSc MVEC–conditioned medium together with Ilomastat or with 10 g/ml anti–MMP-12 antibodies, showed inhibition of the effects obtained with SSc MVEC–conditoned medium alone (Figures 5A and B). DISCUSSION In this study, we observed that the cell-associated uPA/uPAR system is hypofunctional in SSc MVECs. Spontaneous invasion of Matrigel-coated filters as well as the chemoinvasive and mitogenic effects of uPA were low for SSc MVECs, whose fibrinolytic pattern was characterized by higher uPAR expression than that in normal MVECs. When seeded on Matrigel, normal MVECs spontaneously formed capillary-like structures, an activity which was inhibited by uPA and uPAR antagonism and potentiated by exogenous uPA. SSc MVECs showed a lower ability to form capillary-like structures, giving rise to endothelial networks that appeared incomplete, were formed by thinner vessels, and were not sensitive to exogenous uPA. Taken together, these data indicate that a lower performance of the uPA/uPAR system accounts for the decreased ability of SSc MVECs to perform a complete angiogenic program. Given the critical role of uPA and uPAR in angiogenesis (9–13) and the abundance of uPAR in SSc MVECs, their weak response to exogenous uPA in terms of invasion, proliferation, and differentiation was difficult to understand, unless uPAR interactions were pre- 3283 vented by a structural alteration. The uPAR has a 3-domain structure: D1 is the N-terminal domain, D2 connects D1 to D3, and D3 is the C-terminal domain which anchors uPAR to the cell membrane through a glycosyl phosphatidylinositol tail (26). D1 is involved in uPA binding and in the interaction with vitronectin (27), but D2 and D3 are also required for high-affinity interaction (28). Cell-surface uPAR can be cleaved within the D1–D2 linker region by several proteolytic enzymes (22). Such a cleavage leaves, on the cell surface, a 2-domain uPAR, D2-D3, which is a form expressed in various tissues and cell lines (22), including human MVECs (13). In human MVECs, uPAR D1 is cleaved from the full-size receptor by MMP-12, following stimulation either with a combination of bFGF and TNF␣ or with PMA (13). Truncation of uPAR results in a decrease of angiogenesis in fibrin matrices (13). We have shown that in SSc MVECs, uPAR undergoes an MMP-12– dependent cleavage between D1 and D2, which leaves, on the cell surface, the D2-D3 form of the receptor without the ability to bind uPA and to interact with vitronectin and integrins (29). We have observed that MMP-12 is constitutively overexpressed in SSc MVECs, and that uPAR cleavage, impairment of capillary morphogenesis, and reduced uPA-dependent chemoinvasion and proliferation, which were all induced in normal MVECs by incubation with SSc MVEC–conditioned medium, could be reversed by coincubation with anti– MMP-12 antibodies. The uPAR is a multifunctional protein involved in the regulation of cell-associated proteolysis, cell adhesion, chemotaxis, migration, and proliferation (30). These activities of uPAR depend on the presence of the full-size enzyme, which facilitates its ability to bind uPA and vitronectin and to interact with integrins. Removal of uPAR D1 abolishes its interactions with uPA and vitronectin (28,30) and affects the ability of uPAR to associate with various ␣- and ␤-integrin chains (29). Therefore, our observation of an overexpression of MMP-12 in SSc MVECs, which results in uPAR D1 cleavage, may be critical in the reduced angiogenic potential of the endothelial cell in the diffuse form of the disease, by reducing the cell “grip-and-go” properties related to the full-size form of uPAR (31). The results described in this report were obtained with MVECs isolated from biopsy samples of skin, which were independent from any influence of the microenvironment. Nevertheless, in SSc, MVECs undergo hypoxia due to subintimal and adventitial fibrosis and intimal proliferation of arterioles, which leads to a marked 3284 D’ALESSIO ET AL narrowing of the vessel lumen (32). Many observations related to tumor angiogenesis indicate that prolonged hypoxia increases genomic instability (33) and gene amplification (34), and may act as a selective pressure for induction of cell variants by altering gene regulation (35). The endothelium has been considered to be less likely to develop antiangiogenic resistance, because normal endothelial cells are genetically stable. However, although endothelial cells have genetic stability and a unique ability to tolerate hypoxia for prolonged periods, the situation in SSc may be unique, since MVECs are subjected to hypoxia for periods ranging from years to decades, which could determine a unique adaptive response to hypoxic stress. Perivascular inflammation is a prominent feature in SSc (36), and the products of inflammatory cells may induce epigenetic alterations in SSc MVECs. In addition to producing uPA within the inflammatory microenvironment (37), the activated monocyte/macrophages release both TNF␣ and bFGF (38), which have been shown to up-regulate MMP-12 (13) and uPAR (39) in human MVECs. Furthermore, under hypoxia, uPAR undergoes up-regulation (40). Therefore, we suggest the possibility that the alterations we have observed in SSc MVECs (uPAR and MMP-12 up-regulation) may initially be stimulated in an epigenetic manner and then become the product of the hypoxia-induced selection of an endothelial cell population more suitable to survive in the hypoxic microenvironment. Genetic expression profiling, performed in our laboratory on normal MVECs and SSc MVECs, indicates that at least 50 genes are differentially expressed in the 2 cell lines, including uPAR and MMP-12 (D’Alessio S, et al: unpublished observations). Many of these genes are related to the onset of angiogenesis and hypoxia, suggesting that multiple alterations determine the diminished angiogenic capability of SSc MVECs. However, the data reported in the present study indicate that uPAR cleavage is critical to the pathogenesis of defective angiogenesis, and this a typical feature of the terminal forms of diffuse SSc. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. REFERENCES 1. Cutolo M, Grassi W, Matucci Cerinic M. Raynaud’s phenomenon and the role of capillaroscopy. Arthritis Rheum 2003;48:3023–30. 2. Majewski S, Skopinska-Rozewska E, Jablonska S, Polakowski IJ, Pawinska M, Marczak M, et al. Modulatory effect of sera from scleroderma patients on lymphocyte-induced angiogenesis. Arthritis Rheum 1985;28:1133–9. 3. Polakowski IJ, Majewski S, Skopinska-Rozewska E, Zukowska M, Wlorarska B, Jablonska S. Modulatory effect of sera from scleroderma patients on lymphocyte-induced angiogenesis: II. Effector 19. 20. cells for the enhancing effect of acroscleroderma patients’ sera. Arch Dermatol Res 1988;280:395–8. Marczak M, Majewsk S, Skopinska-Rozewska E, Polakowski I, Jablonka S. Enhanced angiogenic capability of monocyte-enriched mononuclear cell suspension from patients with systemic scleroderma. J Invest Dermatol 1986;86:355–8. Kaminski MJ, Majewski S, Jablonska S, Pawinska M. Lowered angiogenic capability of peripheral blood lymphocytes in progressive systemic sclerosis (scleroderma). J Invest Dermatol 1984;82: 239–43. Koch AE, Litvak MA, Burrows JC, Polverini PJ. Decreased monocyte-mediated angiogenesis in scleroderma. Clin Immunol Immunopathol 1992;64:153–60. Ribatti D, Cantatore FP, Vacca A, D’Amore M, Ria R, Roncali L, et al. Systemic sclerosis stimulates angiogenesis in the chick embryo chorioallantoic membrane. Clin Rheumatol 1998;17: 115–20. Hebbar M, Peyrat JP, Hornez L, Hatron PY, Hachulla E, Devulder B. Increased concentrations of the circulating angiogenesis inhibitor endostatin in patients with systemic sclerosis. Arthritis Rheum 2000;43:889–93. Mazar AP, Henkin J, Goldfarb RH. The urokinase plasminogen activator system in cancer: implications for tumor angiogenesis and metastasis. Angiogenesis 1999;3:15–32. Koolwijk P, van Erck MG, de Vree WJ, Vermeer MA, Weich HA, Hanemaaijer R, et al. Cooperative effect of TNF␣, bFGF, and VEGF on the formation of tubular structures of human microvascular endothelial cells in a fibrin matrix: role of urokinase activity. J Cell Biol 1996;132:1177–88. Fibbi G, Caldini R, Chevanne M, Pucci M, Schiavone N, Morbidelli L, et al. Urokinase-dependent angiogenesis in vitro and diacylglycerol production are blocked by antisense oligonucleotides against the urokinase receptor. Lab Invest 1998;78:1109–19. Kroon ME, Koolwijk P, van Goor H, Weidle UH, Collen A, van der Pluijm G, et al. Role and localization of urokinase receptor in the formation of new microvascular structures in fibrin matrices. Am J Pathol 1999;154:1731–42. Koolwijk P, Sidenius N, Peters E, Sier CF, Hanemaaijer R, Blasi F, et al. Proteolysis of the urokinase-type plasminogen activator receptor by metalloproteinase-12: implication for angiogenesis in fibrin matrices. Blood 2001;97:3123–31. Fibbi G, Pucci M, Grappone C, Pellegrini G, Salzano R, Casini A, et al. Function of the fibrinolytic system in human Ito cells and its control by basic fibroblast and platelet-derived growth factor. Hepatology 1999;29:868–78. Scott PA, Bicknell R. The isolation and culture of microvascular endothelium. J Cell Sci 1993;105:269–73. Manconi F, Markham R, Fraser IS. Culturing endothelial cells of microvascular origin. Method Cell Sci 2000;22:89–99. Annunziato F, Romagnani P, Cosmi L, Beltrame C, Steiner BH, Lazzeri E, et al. MDC and ELC attract human thymocytes in different stages of development and are produced by distinct subsets of medullary epithelial cells: possible implications for negative selection. J Immunol 2000;165:238–46. Romagnani P, Annunziato F, Lazzeri E, Cosmi L, Beltrame C, Lasagni L, et al. Interferon-inducible protein 10, monokine induced by interferon ␥, and interferon-inducible T-cell chemoattractant are produced by thymic epithelial cells and attract T-cell receptor (TCR) ␣␤ CD8⫹ single-positive T cells, TCR␥␦ T cells, and natural killer-type cells in human thymus. Blood 2001;97: 601–7. Fibbi G, Barletta E, Dini G, Del Rosso A, Pucci M, Cerletti M, et al. Cell invasion is affected by differential expression of the urokinase plasminogen activator/urokinase plasminogen activator receptor system in muscle satellite cells from normal and dystrophic patients. Lab Invest 2001;81:27–39. Kubota Y, Kleinman HK, Martin GR, Lawley TJ. Role of laminin IMPAIRED SSc MVEC ANGIOGENESIS BY MMP-12 CLEAVAGE OF uPAR 21. 22. 23. 24. 25. 26. 27. 28. 29. and basement membrane in the morphological differentiation of human endothelial cells into capillary-like structures. J Cell Biol 1988;107:1589–98. Grant DS, Tashiro K, Segui-Real B, Yamada Y, Martin GR, Kleinman HK. Two different laminin domains mediate the differentiation of human endothelial cells into capillary-like structures in vitro. Cell 1989;58:933–43. Hoyer-Hansen G, Ploug M, Behrendt N, Ronne E, Dano K. Cell-surface acceleration of urokinase-catalyzed receptor cleavage. Eur J Biochem 1997;243:21–6. Hoyer-Hansen G, Ronne E, Solberg H, Behrendt N, Ploug M, Lund LR, et al. Urokinase plasminogen activator cleaves its cell surface receptor releasing the ligand-binding domain. J Biol Chem 1992;267:18224–9. Behrendt N, Ploug M, Patthy L, Houen G, Blasi F, Dano K. The ligand-binding domain of the cell surface receptor for urokinasetype plasminogen activator. J Biol Chem 1991;266:7842–7. Ploug M, Ellis V. Structure-function relationships in the receptor for urokinase-type plasminogen activator: comparison to other members of the Ly-6 family and snake venom ␣-neurotoxins. FEBS Lett 1994;349:163–8. Ploug M, Ronne E, Behrendt N, Jensen AL, Blasi F, Dano K. Cellular receptor for urokinase plasminogen activator: carboxylterminal processing and membrane anchoring by glycosyl-phosphatidylinositol. J Biol Chem 1991;266:1926–33. Wei Y, Waltz DA, Rao N, Drummond RJ, Rosenberg S, Chapman HA. Identification of the urokinase receptor as cell adhesion receptor for vitronectin. J Biol Chem 1994;269:32380–8. Hoyer-Hansen G, Behrendt N, Ploug M, Dano K, Preissner KT. The intact urokinase receptor is required for efficient vitronectin binding: receptor cleavage prevents ligand interaction. FEBS Lett 1997;420:79–85. Montuori N, Carriero MV, Salzano S, Rossi G, Ragno P. The 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 3285 cleavage of the urokinase receptor regulates its multiple functions. J Biol Chem 2002;277:46932–9. Sidenius N, Blasi F. The urokinase plasminogen activator system in cancer: recent advances and implication for prognosis and therapy. Cancer Metastasis Rev 2003;22:205–22. Del Rosso M, Fibbi G, Pucci M, D’Alessio S, Del Rosso A, Magnelli L, et al. Multiple pathways of cell invasion are regulated by multiple families of serine proteases. Clin Exp Metastasis 2002;19:193–207. LeRoy EC. Systemic sclerosis: a vascular perspective. Rheum Dis Clin North Am 1996;22:675–94. Reynolds TY, Rockwell S, Glazer PM. Genetic instability induced by the tumor microenvironment. Cancer Res 1996;56:5754–7. Hill RP. Tumor progression: potential role of unstable genomic changes. Cancer Metastasis Rev 1990;9:137–47. Subarsky P, Hill RP. The hypoxic tumour microenvironment and metastatic progression. Clin Exp Metastasis 2003;20:237–50. Ishikawa O, Ishikawa H. Macrophage infiltration in the skin of patients with systemic sclerosis. J Rheumatol 1992;19:1202–6. Stacey KJ, Fowles LF, Colman MS, Ostrowski MC, Hume DA. Regulation of urokinase-type plasminogen activator gene transcription by macrophage colony-stimulating factor. Mol Cell Biol 1995;15:3430–41. Sunderkotter C, Steinbrink K, Goebeler M, Bhardwaj R, Sorg C. Macrophages and angiogenesis. J Leukoc Biol 1994;55:410–22. Dekkers PE, ten Hove T, te Velde AA, van Deventer SJ, van Der Poll T. Upregulation of monocyte urokinase plasminogen activator receptor during human endotoxemia. Infect Immun 2000;68: 2156–60. Graham CH, Fitzpatrick TE, McCrae KR. Hypoxia stimulates urokinase receptor expression through a heme protein-dependent pathway. Blood 1998;9:3300–7.