|Home | About | Journals | Submit | Contact Us | Français|
EphB4 receptor (EphB4) and its ligand (EphrinB2) play an important role in the regulation of cell adhesion, growth, and migration. The purpose of this study was to determine the effects of EphB4 blockade by soluble EphB4 (sEphB4) on retinal pigment epithelial (RPE) cell migration and proliferation, induced by platelet-derived growth factor-BB (PDGF), and to establish its relevance to proliferative vitreoretinopathy (PVR)
The expression of EphB4 and EphrinB2 in early-passage human RPE cells and in human PVR membranes was evaluated by confocal microscopy. The effect of sEphB4 (0.1–3 μg/mL) on PDGF (20 ng/mL)-induced RPE migration and proliferation was evaluated using a modified Boyden chamber assay and an MTT assay, respectively. Attachment to basement membrane matrix and fibronectin was assayed by MTT. Phosphorylation of FAK and p42/44 mitogen-activated protein kinase (MAPK) in retinal pigment epithelium was determined by Western blot analysis after exposure to sEphB4. The effect of sEphB4 on the phosphorylation of EphB4/EphrinB2 was demonstrated with the use of immunoprecipitation assays
EphrinB2 and EphB4 were expressed on human RPE cells in vitro and in cells within human PVR membranes. sEphB4 blocked EphB4 and EphrinB2 phosphorylation in RPE cells in vitro. sEphB4 reduced RPE migration in response to PDGF stimulation (P < 0.01). Similarly, sEphB4 inhibited RPE attachment and proliferation in a dose-dependent manner (P < 0.05). PDGF-induced phosphorylation of FAK and MAPK was inhibited by sEphB4
EphB4 and EphrinB2 are expressed in RPE cells and PVR membranes. sEphB4 inhibits PDGF-induced RPE cell attachment, proliferation, and migration. This effect may result from the inhibition of FAK and MAPK phosphorylation.
The retinal pigment epithelium plays an important role in maintaining photoreceptor cell survival and function. Normal retinal pigment epithelial (RPE) cells are quiescent without migration or proliferation.1–3 RPE cells in the stationary monolayer start to migrate in response to pathologic changes in the microenvironment associated with conditions such as proliferative vitreoretinopathy (PVR),4 age-related macular degeneration,5 and diabetic retinopathy.6 RPE cell migration is a complex biological process involving changes in cell attachment, spreading, and cytoskeletal reorganization and is regulated by cell matrix, matrix-dependent enzymes, cytokines, and growth factors.7–9 RPE cell migration is also mediated by cell membrane-associated signaling, especially receptor tyrosine kinases.10,11 Recent reports show that EphB4 and its ligand EphrinB2 mediate cell migration in a variety of cells, such as neurons, retinal vascular endothelial cells, and choroidal endothelial cells.12,13
The Ephs and Ephrins constitute the largest of the receptor tyrosine kinase families, with 14 receptors and eight ligands.14–16 This family is subdivided into EphA and EphB groups and regulates a diverse array of the cellular functions (migration, repulsion, adhesion, and vessel maturation).14–17 The interaction between the Eph receptor and the Ephrin ligand activates forward and reverse signaling through interactions with cytoplasmic signaling proteins. The nature of the downstream signaling pathways and the consequent regulation of cell migration had been demonstrated in various cell culture systems.18–22 Kertesz23 has highlighted the role of EphB4 in angiogenesis through its effects on endothelial cell migration.23 Steinle12 reported that the agonistic EphB4/Fc chimera can stimulate retinal endothelial cell migration by activation of PI3K, Src, and other signaling pathways. In a previous study, we found that sEphB4 inhibits vascular endothelial growth factor (VEGF)-induced choroidal endothelial cell migration.13 We have also demonstrated that platelet-derived growth factor-BB (PDGF) increases RPE cell migration through the activation of p42/44 mitogen-activated protein kinase (MAPK).24,25 Furthermore, PDGF has been found to be expressed in fibrotic PVR membranes and choroidal neovascular membranes, which are associated with RPE migration.26,27 Although previous studies have focused on the roles of EphB4 and EphrinB2 in endothelial cell function and cancer cells, little is known about their role in migration of ocular cells such as RPE cells. Therefore, the aim of the present study was to determine the expression of EphB4 and EphrinB2 in retinal pigment epithelium, to evaluate the effect of sEphB4 on RPE cell migration induced by PDGF, and to investigate the signaling pathways involved.
The institutional review board of the University of Southern California approved our use of cultured human RPE cells and human PVR membranes that had been surgically excised at the time of vitrectomy. All procedures conformed to the Declaration of Helsinki for research involving human subjects.
Human RPE cells were isolated from fetal human eyes of >22 weeks' gestation (Advanced Bioscience Resources, Inc., Alameda, CA), as previously described in detail.28 Cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Fisher Scientific, Pittsburgh, PA) with 2 mM l-glutamine, 100 U/mL penicillin, 100 μg/mL streptomycin (Sigma, St. Louis, MO), and 10% heat-inactivated fetal bovine serum (FBS; Irvine Scientific, Santa Ana, CA). The culture method used regularly yields >95% RPE cells (cytokeratin-positive). Cells used were from passages 2 to 4.
Recombinant soluble extracellular domain of EphB4 was generated by Vasgene Inc. (Los Angeles, CA), as reported previously.13 The coding region, representing amino acids 1 to 537 and including the signal peptide (N-terminal 15 amino acids), was amplified from full-length EphB4 cDNA using TACTAGTCCGCCATGGAGCTCCGGGTGCTGCT as direct and TGCGGCCGCTTAATGGTGATGGTGATGATGCTGCTCCCGCCAGCCCTCGCTCTCAT as reverse primers. The DNA fragment was cloned into the mammalian expression vector pEF6/V5-His-TOPO (Invitrogen, Carlsbad, CA,), followed by digestion with NotI and self-ligation to allow in-frame fusion to V5 and His-tag. Protein was expressed in the 293 human embryonic kidney cell line. The supernatant containing the secreted proteins was harvested 72 to 96 hours later, clarified by centrifugation, and used for purification on agarose (Ni-NTA; Qiagen, Valencia, CA). The purity and quantity of the recombinant proteins was tested by sodium dodecyl sulfate-polyacrylamide gel (SDS-PAGE) electrophoresis with Coomassie blue staining, Western blot analysis, and ultraviolet spectroscopy. Purified proteins, free of endotoxin, were dialyzed against 20 mM Tris-HCl, 0.15 M NaCl, pH 8, and stored at −70°C. The recombinant protein was functionally active, as demonstrated by the ability of sEphB4-immobilized beads to precipitate EphrinB2-alkaline phosphatase fusion protein in an alkaline phosphatase activity assay (data not shown).
The MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay is a colorimetric assay used to measure the activity of enzymes in living cells that reduce MTT to formazan and produce a purple color. Absorbance of the colored solution can be quantified using a spectrophotometer to provide an estimate of the number of attached living cells. RPE cells (2 × 103) were seeded in 96-well plates in DMEM with 10% FBS. After overnight incubation to allow for cell attachment, the medium was replaced with DMEM containing 0.4% FBS, and the cells were treated with different concentrations of sEphB4 (1 and 3 μg/mL) for 48 hours. Four hours before harvest 20 μL MTT (5 mg/mL; Sigma) was added to each well. After 4-hour incubation, the supernatants were decanted, and the formazan precipitates were solubilized by the addition of 150 μL of 100% dimethyl sulfoxide (DMSO; Sigma) and placed on a plate shaker for 10 minutes. Absorbance at 550 nm was determined on a multiwell plate reader (Benchmark Plus; Bio-Rad, Tokyo, Japan).
Migration was measured using a modified Boyden chamber assay, as previously described.13 Briefly, 5 × 104 RPE cells were seeded in the upper part of a Boyden chamber in 24-well plates after treatment with sEphB4 (0.1, 0.5, 1, and 3 μg/mL; Vasgene Therapeutics) for 24 hours, with inserts coated with fibronectin (2 μg/cm2). The lower chamber was filled with 0.4% FBS-DMEM containing 20 ng/mL recombinant PDGF (R&D Systems Inc., Minneapolis, MN).24 After 5 hours of incubation, the inserts were washed three times with PBS, fixed with cold methanol (4°C) for 10 minutes, and counterstained with hematoxylin for 20 minutes The number of migrated cells was counted by phase-contrast microscopy (×320). Four randomly chosen fields were counted per insert. The experiment was repeated three times. An independent migration assay was performed to test the direct effects of sEphB4 on RPE cell migration induced by PDGF; this assay was identical with the one described except that the RPE cells were not pretreated with sEphB4.
Attachment was measured with a modified MTT assay using 96-well plates coated with fibronectin or growth factor-depleted basement membrane matrix (BD Matrigel; BD Biosciences, Franklin Lakes, NJ). After treatment with PDGF (20 ng/mL) and sEphB4 (0.1, 0.5, 1, and 3 μg/mL) for 24 hours, RPE cells were trypsinized and resuspended in DMEM with 0.4% FBS. One hundred microliters of cell suspension (104 cells) were added to each well and allowed to attach for 5 and 15 minutes. The cells were washed gently with PBS twice, and fresh medium (150 μL) was added to each well with MTT (5 mg/mL, 20 μL; Sigma). After 4 hours of incubation, the supernatants were decanted, the formazan precipitates were solubilized by the addition of 150 μL of 100% DMSO (Sigma), and the 96-well plate was placed on a plate shaker for 10 minutes. Absorbance at 550 nm was determined on a multiwell plate reader (Benchmark Plus; Bio-Rad). The number of attached cells was proportional to the absorbance of MTT at 550 nm.
A separate assay was performed to test the direct effects of sEphB4 on RPE cell attachment to fibronectin (see Figs. 3E, E,3F).3F). The same MTT method was used except that RPE cells were not pretreated with sEphB4, and the fibronectin precoated 96-well plates were incubated with sEphB4 (1 μg/mL) for 2 hours, after which the cells were seeded in the wells for the evaluation of the cell attachment at different time points (5 and 15 minutes). A dose-response experiment was also performed (sEphB4; 0.1–3 μg/mL) at the 15-minute time point using denatured (95°C for 30 minutes) sEphB4 (1 μg/mL) as the control.
Three surgically excised epiretinal membranes from patients with PVR were prepared for immunostaining. Tissues were snap-frozen and sectioned at 6 μm using the cryostat. Thawed tissue sections were air dried and rehydrated with PBS (pH 7.4). The slides were fixed with 3.7% paraformaldehyde for 30 minutes, and then rinsed in PBS twice for 10 minutes. After fixation blocked with 5% normal goat serum for 60 minutes, the sections were incubated with primary goat anti-human EphrinB2 or anti-EphB4 (R&D Systems Inc.) at 4°C overnight. Binding of the primary antibody was visualized with rhodamine-conjugated anti-goat secondary antibody (Vector Laboratories, Burlingame, CA) for 30 minutes. For double staining, sections were washed and then incubated with anti-cytokeratin cocktail (AE1/AE3; Abcam, Cambridge, MA) for 1 hour at room temperature. After washing, sections were incubated with fluorescein-conjugated anti-mouse secondary antibody (Vector Laboratories). Sections were mounted in mounting medium (Vectashield; Vector Laboratories), and the sections were examined with a confocal microscope (LSM510; Zeiss, Thornwood, NY) using the accompanying image acquisition and analysis software (LSM version 3.2; Zeiss). Cells were imaged using a 1.3 numerical aperture, 40× objective lens (Plan Neofluar; Zeiss) using identical settings for laser power, pinhole size, and detector sensitivity. FITC staining was captured using a 488-nm argon laser for excitation and a 505- to 530-nm band-pass filter to detect emission. Rhodamine staining was captured using a 543-nm HeNe laser for excitation and a 560-nm long-pass filter to detect emission. The same method of staining was used to evaluate the expression of EphB4 and EphrinB2 in cultured human retinal pigment epithelium grown to 70%–80% confluence in four-well chamber slides (Laboratory-Tek; Electron Microscopy Services, Hatfield, PA). Cell nuclei were labeled with 4′,6-diamidino-2-phenylindole (DAPI) added to the mounting medium. DAPI staining was captured using a 800-nm titanium sapphire laser for excitation and a 390-to 465-nm band-pass filter to detect emission. Negative controls included omitting the primary antibody and the use of IgG in place of the primary antibody (at the same concentration).
For evaluation of sEphB4 effects on FAK expression, RPE cells were plated on four-well chamber slides, treated with sEphB4 (0.1, 0.5, 1, 3 μg/mL) for 24 hours with or without the addition of PDGF (20 ng/mL), fixed with 3.7% paraformaldehyde, and stained with rabbit anti-FAK (total FAK) antibody (C-932; 1:200; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) or rabbit anti-FAK [PY379] (1:100; Invitrogen, Carlsbad, CA) for 1 hour at room temperature, followed by fluorescein-labeled secondary goat anti-rabbit antibody (Chemicon, Inc., Temecula, CA) for 30 minutes. After washing and mounting, the slides were examined with a confocal microscope (LSM-510; Zeiss).
Confluent cells grown in six-well plates were starved for 24 hours in serum-free media and treated with sEphB4 for 24 hours at various concentrations (0.1–3 μg/mL) with or without PDGF (20 ng/mL). To assay phosphorylated FAK and p42/44 MAPK, the cells were stimulated with PDGF for 10 minutes after cells had been pre-exposed to sEphB4 for 24 hours. Cells were harvested and lysed, supernatants were collected, and proteins were resolved on Tris-HCl 10% polyacrylamide gels (Ready Gel; Bio-Rad, Hercules, CA) at 120 V. The proteins were transferred to polyvinylidene blotting membrane (Millipore, Bedford, MA), and the membranes were probed with polyclonal antibody specific for a phosphorylated form of FAK (Biosource, Camarillo, CA) or polyclonal rabbit antibody specific for the dually phosphorylated forms of p42/44 MAPK (at amino acids Thr202 and Tyr204; 1:1000 dilution; New England Biolabs, Ipswich, MA). Membranes were washed and then incubated with a horseradish peroxidase-conjugated secondary antibody (Vector Laboratories) for 30 minutes at room temperature. Images were developed by adding enhanced chemiluminescence detection solution (Amersham Pharmacia Biotech, Cleveland, OH). After stripping, the membranes were reprobed with anti-total p42/44 MAPK (New England Biolabs) or anti-total FAK antibodies (Santa Cruz Biotechnology, Inc.), and the detection procedure was followed as described.
RPE cells grown to 80% confluence were serum-starved overnight. EphrinB2/Fc, EphB4/Fc (both from R&D Systems) or Fc fragment (Jackson Laboratories, Bar Harbor, ME) alone was clustered by incubation with anti-human Fc antibody (1:1000 dilution; Jackson Laboratories) for 1 hour at 4°C and were added to the culture medium at a concentration of 2 μg/mL, with or without sEphB4 (3 μg/mL) for 20 minutes. To study the role of PDGF, serum-starved cells were stimulated with PDGF (10 ng/mL) for 20 minutes. After stimulation, media were aspirated and cells were immediately harvested with protein extraction buffer containing 20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% (vol/vol) Triton X-100, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, and 1 mM sodium vanadate, pH 7.8. Protein extracts were clarified by centrifugation at 18,000g for 20 minutes at 4°C. Cleared cell lysates were incubated with 2 μg/mL EphB4 monoclonal antibody (clone 47; Vasgene Therapeutics, Inc.) or EphrinB2 monoclonal antibody (clone 2B5; Vasgene Therapeutics, Inc.) for 2 hours. Antigen-antibody complexes were immunoprecipitated by shaking with 10 μL protein G-Sepharose beads (Santa Cruz Biotechnology, Inc.) for 1 hour at 4°C. Immunoprecipitates were immunoblotted with anti-phosphotyrosine antibody (clone 4G10; Upstate Biotechnology, Lake Placid, NY) to detect phosphorylation status. To monitor immunoprecipitation efficiency, a duplicate membrane was probed with EphB4-specific (clone 265; Vasgene Therapeutics, Inc.) or EphrinB2-specific (clone P20; Santa Cruz Biotechnology) antibody.
All experiments were performed at least three times. Data were analyzed using the Student's t-test, and P < 0.05 was accepted as significant.
We previously reported that the normal, resting retinal pigment epithelium in an intact RPE monolayer does not show immunopositivity for either EphB4 or EphrinB2 by standard immunohistochemistry.13 We therefore evaluated the expression on EphB4 and EphrinB2 in human RPE cells activated by subconfluent cell culture or retinal pigment epithelium present within active, surgically excised PVR membranes. Three PVR membranes were evaluated, and all showed prominent immunoreactivity for both EphrinB2 and EphB4 (Figs. 1A, A,1B).1B). Double labeling with anti-cytokeratin antibodies, which specifically label RPE cells in the human retina,26 showed numerous cytokeratin-positive transdifferentiated retinal pigment epithelium within the PVR membrane. Focal double staining for EphB4 and EphrinB2 with cytokeratin was found in each membrane, indicating that a subset of the EphrinB2 or EphB4 labeling was on cells derived from retinal pigment epithelium (Figs. 1A, A,1B,1B, and insets). Cultured human retinal pigment epithelium showed strong immunolabeling with EphB4 and weaker immunolabeling for EphrinB2 (Fig. 1C). Background labeling was negligible (Fig. 1C, right).
Cultured human RPE cells express both EphB4 and EphrinB2, as demonstrated by immunofluorescence (Fig. 1C) and Western blot (Fig. 2A). Stimulation of RPE cells with EphrinB2/Fc induced EphB4 phosphorylation, and EphrinB2 phosphorylation was activated after exposure to EphB4/Fc (Fig. 2A). Addition of sEphB4 completely blocked EphB4 phosphorylation by EphrinB2/Fc and EphrinB2 phosphorylation by EphB4/Fc (Fig. 2B). PDGF induced a lower level of EphrinB2 phosphorylation, but sEphB4 did not affect PDGF-induced EphrinB2 phosphorylation (Fig. 2B). PDGF had no effect on the phosphorylation status of EphB4 (Fig. 2C). These results suggest that PDGF induces EphrinB2 phosphorylation, which is not mediated by its binding to the EphB4 binding domain of EphrinB2.
RPE cell attachment was measured on fibronectin (Figs. 3A, A,3B)3B) and basement membrane matrix (Figs. 3C, C,3D).3D). After 24-hour exposure to sEphB4, RPE cell attachment in control and low-dose sEphB4 (0.1 μg/mL)-treated cells was similar. However, a dose-dependent decrease in attachment was seen with the application of higher concentrations of sEphB4 (0.5 μg/mL and greater) in both fibronectin- and basement membrane matrix-coated plates (P < 0.01). Maximal inhibition (40%) was observed at 1 μg/mL sEphB4 in 5- and 15-minute attachment assays.
In additional experiments, fibronectin precoated wells were incubated with an overlay of sEphB4 (Figs. 3E, E,3F).3F). RPE cell attachment was significantly reduced in wells by overlaid sEphB4 at both 5 and 15 minutes of cell attachment; denatured sEphB4 did not significantly interfere with cell attachment (Fig. 3E; P < 0.05). The dose-response of sEphB4 (0.1–3 μg/mL) on the direct inhibition of RPE cells attachment was also tested (Fig. 3F); the results showed that RPE attachment was progressively reduced with increasing concentrations of sEphB4 beginning at a dose of 0.5 μg/mL (P < 0.05). Denatured sEphB4 did not significantly interfere with RPE cell attachment except at the highest dose (3 μg/mL denatured sEphB4 vs. untreated control; P < 0.05), at which there was a partial inhibition compared with the stronger effect of nondenatured sEphB4 (Fig. 3F).
RPE cell migration to the lower compartment of a Boyden chamber was significantly increased in response to PDGF stimulation (Fig. 4A). PDGF-induced RPE migration was inhibited when cells were pretreated with sEphB4 (0.5, 1, 3 μg/mL) in a dose-dependent manner. Maximal inhibition (50%) was seen at 1 μg sEphB4 (Fig. 4A; P < 0.01). When sEphB4 was added to RPE cells in the upper chamber in the absence of pretreatment (Fig. 4B), cell migration was even more prominently inhibited over the range of 0.1 μg/mL sEphB4 (P < 0.05) to 3 μg/mL sEphB4 (P < 0.01).
RPE cell proliferation was significantly inhibited by the addition of sEphB4 (1 and 3 μg/mL) to the media for 48 hours at both dosages (Fig. 5; P < 0.05).
In the present study, we evaluated the activation of FAK by PDGF and the effect of sEphB4 on this effect (Fig. 6C). FAK expression was seen in a distinct, punctate staining pattern at the cell periphery by immunofluorescent staining (Figs. 6A, A,6B).6B). Although PDGF did not significantly alter total FAK expression, FAK phosphorylation was increased in the confocal imaging and Western blots analysis (Figs. 6A, A,6C)6C) by PDGF. Remarkably, even at the dose of 0.5 μg/mL, sEphB4 inhibited PDGF-induced FAK phosphorylation, as demonstrated by immunofluorescence and Western blot analysis (Figs. 6A, A,6C).6C). The inhibition of PDGF-induced FAK phosphorylation by sEphB4 was dose dependent (Figs. 6A, A,6C).6C). sEphB4 at a dose of 3 μg/mL almost completely inhibited PDGF-induced FAK phosphorylation, as shown in Western blot analysis (Fig. 6C). Inhibition was statistically significant when evaluated by densitometric analysis (Fig. 6C; P < 0.01). Total FAK expression levels, as measured by immunofluorescence and Western blot, were reduced only at doses of 1 and 3 μg/mL sEphB4 (Figs. 6B).
PDGF is known to activate p42/44 MAPK in RPE cells in vitro.24 We found that stimulation of RPE cells with PDGF resulted in prominent phosphorylation of MAPK (Fig. 7), whereas RPE cells without PDGF treatment showed weak MAPK phosphorylation. After 24 hours of sEphB4 pretreatment (0.5 to 3 μg/mL), the phosphorylation of MAPK was significantly reduced after PDGF stimulation (Fig. 7); however, sEphB4 pretreatment did not alter the total MAPK expression (Fig. 7). The quantitative effect of sEphB4 on MAPK activation was shown by densitometry analysis of three independent experiments; at sEphB4 levels of 0.5 μg/mL and greater, the phosphorylation of MAPK was significantly inhibited (Fig. 7) (P < 0.05).
EphB4 and EphrinB2 are expressed mainly on endothelial cells and neurons.14–16 EphB4 selectively binds EphrinB2, leading to bidirectional signal activation and regulation of cell proliferation, cell attachment, cytoskeletal organization, and migration. These functions play important roles in embryonic development, segmentation vessel maturation, neuronal pathfinding, and angiogenesis.14–16 Eph/ephrin interaction also regulates colonic epithelial cell differentiation.29 In pathologic conditions associated with vascular disease and neoplasia, EphB4 and EphrinB2 expression was found to be high, particularly in inflammatory30 and tumor cells.31 In the present study, we demonstrated expression of EphB4 and EphrinB2 in cultured human RPE cells and cells derived from RPE cells in surgically excised PVR membranes. Interestingly, we had previously shown that RPE cells within the normal RPE monolayer do not demonstrate EphB4 or EphrinB2 by standard immunohistochemistry.13 We reevaluated this lack of expression in several normal adult and fetal human retinas by immunohistochemistry and again did not find expression of EphB4 or EphrinB2 (data not shown). Because the RPE cells in the nonpathologic RPE monolayer do not proliferate or migrate, it is logical that their expression of a receptor-mediating migration would be downregulated; however, it will be of interest in future studies to determine the factors that upregulate the expression of EphB4 and EphrinB2 in retinal pigment epithelium. To the best of our knowledge, this is the first documentation of EphB4 and EphrinB2 expression in ocular epithelial cells.
The involvement of EphB4 and EphrinB2 in the regulation of cell migration has been demonstrated in vascular endothelial cells12 and tumor cells.22 Furthermore, the extracellular domain of EphB4 stimulates endothelial cell migration and proliferation.32 More important, EphB4 is associated with tumor cell survival.33 In contrast, inhibition of EphB4 signaling by EphB4 siRNA and antisense oligodeoxynucleotide reduced tumor cell invasion.34 On the other hand, overexpression of EphrinB2 enhances actin fiber polymerization and migration in melanoma cells.35 EphrinB2 deletion causes a defect in smooth muscle cell spreading and migration.36 These results suggest a possible close relationship between EphB4 function and RPE migration.
RPE migration is prominent in early stages of eye development and in retinal wound healing, inflammation, and angiogenesis. 4–6,26,27 RPE migration is strongly stimulated by PDGF in vitro,24 and PDGF is highly expressed in lesions of many retinal diseases involving RPE migration, such as PVR, choroidal neovascularization complicating age-related macular degeneration, and proliferative diabetic retinopathy.4–6 Thus, inhibition of PDGF-induced RPE migration holds the promise of therapeutic regulation of RPE function. Our research group and Zamora et al.13,37 have shown that blocking of EphB4 and EphrinB2 interaction by competitively acting soluble EphB4 inhibits retinal and choroidal neovascularization in vitro and in vivo. Based on these findings and the known role of bidirectional EphB4/EphrinB2 signaling during cell migration, it seemed likely that soluble EphB4 could interfere with RPE migration. Indeed, our results confirmed the hypothesis by showing that RPE migration induced by PDGF was significantly inhibited by sEphB4. This inhibitory effect on attachment and migration occurs either when the RPE cells are pretreated with sEphB4 for 24 hours or when they are directly exposed to sEphB4 during the course of the assay, without pretreatment. This indicates that sEphB4 can act quickly to prevent EphB4/EphrinB2 function and that effects are unlikely to be mediated by any downstream secondary changes in the retinal pigment epithelium that could be altered by treatment with sEphB4.
Semela et al.38 recently showed that EphrinB2 is a downstream mediator of PDGF signaling; more important, siRNA-induced knockdown of EphrinB2 attenuated hepatic stellate cell-driven angiogenesis in response to PDGF. Along with our data demonstrating that PDGF induces EphrinB2 phosphorylation in retinal pigment epithelium, it seems likely that the effect of sEphB4 on the inhibition of PDGF-induced RPE migration is partially related to inhibition of EphrinB2 action.38 Our immunoprecipitation assays showed that sEphB4 completely blocked both EphB4 and EphrinB2 phosphorylation in RPE cells, suggesting bidirectional or reciprocal signaling interactions between Ephrin and Eph are able to inhibit PDGF-induced migration.
Cell movement involves changes in cell adhesion molecules; accordingly, we found that sEphB4 decreased RPE cell attachment to fibronectin or basement membrane matrix. This confirms the findings of Foo36 and Gill23 that interfering with EphB4 and EphrinB2 signaling reduces focal adhesion formation and cell motility. Therefore, current results suggest that sEphB4 compromised PDGF-induced RPE migration, in part through the inhibition of cell attachment. The inhibition of RPE cell attachment may also be attributed to sEphB4 binding to fibronectin and the formation of a complex of sEphB4 and fibronectin, preventing cell attachment to fibronectin. The hypothesis is further confirmed by overlay of sEphB4 on fibronectin precoated wells, indicating that fibronectin or its receptor can bind sEphB4 and reduce cell binding. Another possibility is that sEphB4 may interact with integrin expression or function,39,40 so that RPE cell binding to fibronectin is reduced.
Before cell migration, the formation of focal adhesions is critical. The present study has shown that sEphB4 affects both the expression of total FAK and, more impressively, phosphorylated FAK in RPE cells as revealed by confocal immunofluorescence and Western blot analysis. The strong inhibition of FAK expression and phosphorylation is also consistent with the reduction of RPE migration after sEphB4 treatment. It has been shown in other systems that Eph/Ephrin signaling results in alterations in the activation of FAK and the redistribution of paxillin, thus altering cytoskeletal dynamics,41 and that both Ephrin and FAK expression are growth factor regulated.42 Our results further highlight the complexity of Eph/Ephrin-driven cellular interactions controlling RPE migration. Clearly, multiple cellular signaling molecules, including FAK, are regulated by bidirectional EphB4/EphrinB2 signaling during RPE migration.
In many cell types, EphB4/EphrinB2 binding leads to mitogenic responses involving the p42/44 MAPK cascade.12,21,43 Consistent with these reports, we were able to show that sEphB4 inhibited PDGF-induced proliferation in retinal pigment epithelium. This is important in relevance to potential therapeutic potential because RPE proliferation plays an important role in the pathogenesis of PVR.28 We have shown that along with proliferation, PDGF-induced migration is associated with p42/p44 MAPK activity in retinal pigment epithelium and that PDGF-induced migration in RPE cells was significantly inhibited by the MAPK-specific inhibitor PD98059.24 We speculated that MAPK activation might be impaired by pretreatment with sEphB4. In the present study, we demonstrated that sEphB4 was able to inhibit MAPK phosphorylation after PDGF stimulation. Our results are consistent with the previous finding that EphB4 and EphrinB2 signaling is involved in cell activation of MAPK, which is essential in the regulation of cell migration.12,22,33 The effect of sEphB4 on other signals related to RPE cell migration (protein kinase C, phosphoinositide 3-kinase) and other growth factors related to PVR44 remains to be determined.
In this study we demonstrate that EphB4 and EphrinB2 are expressed in cultured retinal pigment epithelium and in cells derived from retinal pigment epithelium in human PVR membranes and that sEphB4 can inhibit EphB4 and EphrinB2 signaling in retinal pigment epithelium. We then show that many of the pathologic events related to PVR, such as RPE proliferation, attachment, and migration, are inhibited by sEphB4. At the molecular level, RPE cells treated with sEphB4 show reduced FAK and p42/44 MAPK phosphorylation, pathways known to be associated with proliferation and migration in response to PDGF (Fig. 8). The experiment provides a rationale for the development of pharmacologic inhibitors of EphB4/EphriB2 signaling to modulate the regulation of RPE attachment, migration, and proliferation as therapeutic tools for disorders such as PVR.
The authors thank Tom Ogden for review of the manuscript, Susan Clarke for editorial assistance, and Ernesto Barron and Chris Spee for technical assistance.
Supported in part by National Institutes of Health Grants EY01545 (SJR, DRH) and CA79218 (PSG) and Core Grant EY03040; the Arnold and Mabel Beckman Foundation; an unrestricted grant to the Department of Ophthalmology from Research to Prevent Blindness Inc.; and Vasgene Therapeutics.
Disclosure: S. He, Vasgene Therapeutics, Inc. (F); S.R. Kumar, Vasgene Therapeutics, Inc. (F); P. Zhou, Vasgene Therapeutics, Inc. (F); V. Krasnoperov, Vasgene Therapeutics, Inc. (F, E); S.J. Ryan, Vasgene Therapeutics, Inc. (F); P.S. Gill, Vasgene Therapeutics, Inc. (F, I), P; D.R. Hinton, Vasgene Therapeutics, Inc. (F)