Soluble EphB4 Regulates Choroidal Endothelial Cell Function and Inhibits Laser-Induced Choroidal Neovascularization

Abstract

purpose. The purpose of this study was to evaluate the effect of a soluble monomeric form of the EphB4 extracellular domain (sEphB4) on choroidal endothelial cell (CEC) migration and tube formation and on experimental laser-induced choroidal neovascularization (CNV).

methods. EphrinB2 and EphB4 expression in CECs was investigated by Western blot analysis and immunohistochemistry. Effects of sEphB4 (0.5–3 μg/mL) on CEC migration were evaluated with a modified Boyden chamber assay. Tube formation was assayed in CEC cultures in collagen gel. CNV was induced in rats by laser photocoagulation. The effects of intravitreal injection of sEphB4 on CNV development were evaluated at day 14 by fluorescein angiography (FA), confocal volumetric analysis of isolectin-B4 labeled flatmounts, and histologic examination of CNV membranes.

results. CEC cells express both EphB4 and EphrinB2, according to Western blot analysis. Immunohistochemical sections of rat eye showed immunoreactivity for both EphB4 and EphrinB2 in the choroidal endothelium. sEphB4 reduced CEC migration in response to vascular endothelial growth factor (P < 0.01). Similarly, sEphB4 inhibited CEC tube formation in a dose-dependent manner. EphB4, and to a lesser extent EphrinB2, were detected on vascular channels within laser-induced CNV membranes. Intravitreal injection of sEphB4 inhibited laser-induced CNV formation. CNV membranes showed a reduction in leakage score (P < 0.05), and membrane volumes were reduced in size (P < 0.05). Histologic analysis revealed that vascularity was reduced in sEphB4-treated membranes.

conclusions. Recombinant soluble monomeric EphB4 exerts an inhibitory effect on choroidal angiogenesis in vitro and in vivo. It should be further evaluated for its potential as a novel therapy for CNV.

Choroidal neovascularization (CNV) is a serious complication of age-related macular degeneration (AMD), ^{1 2} the leading cause of blindness in the elderly. Choroidal angiogenesis may result from a local imbalance between pro- and antiangiogenic growth factors. ^{1 3 4 5} Steps involved in the development of CNV include endothelial cell adhesion, migration, proliferation, and extracellular matrix production. ^{1 3 4} Newly formed vessels generated from the choriocapillaris infiltrate through Bruch’s membrane, leading to the formation of a neovascular submacular membrane. ^{1 6 7}  

Endothelial cell receptor tyrosine kinases (RTKs) have been recognized as important mediators of angiogenesis ^{1 4 5} ; they include vascular endothelial growth factor (VEGF) receptors, angiopoietin (Tie) receptors, and Ephrin receptors (Ephs). The functions of both the VEGF/VEGF receptor and the angiopoietins/Tie-2 receptor families in vascular development and angiogenesis are well studied. ^{1 4 5} The Ephs and Ephrins together comprise the largest of the RTK subfamilies, with 14 receptors and eight ligands. ^{8 9 10} This family is subdivided into EphA and EphB groups, based on sequence homologies and binding to Ephrin ligands. Eph RTKs and their Ephrin ligands regulate a diverse array of cellular functions such as cell migration, repulsion, and adhesion, ^{8 9 10} but notably lack effects on cell proliferation. EphB4 and its ligand EphrinB2 profoundly influence the development of the vascular system. EphrinB2 is specifically expressed in the arterial angioblast and endothelial and perivascular mesenchymal cells, whereas EphB4 is predominantly expressed in the venous lineage. ^{11 12 13} EphB4 expressing cells interact with EphrinB2 expressing cells to guide cellular positioning and pattern formation in the embryo. At the capillary level, EphrinB2 on arterial endothelial cells interacts with EphB4 on venous endothelial cells to establish a hierarchical pattern in the vasculature and mature native vessels. The physiologic function of EphrinB2 and EphB4 in the adult vasculature is unknown; however, it may be in place to play a role in directing or orienting physiologic angiogenesis or pathologic wound healing in response to tissue injury. Targeted disruption of either Ephrin or Eph receptor can disrupt angiogenesis. ¹⁴ The binding of Eph receptor and its ligand results in receptor clustering and activation of receptor signaling pathways. ^{8 9 12} EphrinB2 is the sole ligand for EphB4, whereas EphrinB2 can be activated by several EphB receptor members. ^{8 9 10}  

The role of the EphrinB2/EphB4 system in the development of CNV is unknown. Although retinal endothelial cells have been shown to express EphrinB2, ¹⁵ the Ephrin expression pattern in the choroidal endothelium has not been studied. We hypothesized that a monomeric soluble form of the extracellular portion of the EphB4 receptor, functioning as a dominant negative inhibitor of receptor activation, will block endothelial cell function and choroidal angiogenesis in general. We initiated these studies to determine the involvement of the EphrinB2/EphB4 system in the pathogenesis of CNV, in vitro and in vivo, by blockade of the Ephrin-signaling pathway. 

Materials and Methods

Choroidal Endothelial Cell Culture and Isolation

Magnetic beads bound to the specific endothelial marker Lycopersicon esculentum (Sigma-Aldrich, St.Louis, MO) were used to isolate choroidal endothelial cells (CECs) from bovine eyes, as previously described. ¹⁶ All animal experiments were approved by the Animal Use Committee for the University of Southern California. Positive immunostaining for von Willebrand factor and uptake of di-acetylated low-density lipoprotein (LDL) confirmed endothelial cell specificity. Cells were cultured in endothelial cell growth medium (EGM; BioWhittaker, Walkersville, MD) containing 10% fetal bovine serum (FBS; Invitrogen-Gibco, Gaithersburg, MD), 2 mM glutamine, 100 μg/mL streptomycin, and 100 μg/mL penicillin (Sigma-Aldrich). 

Preparation of Soluble EphB4

Recombinant soluble extracellular domain of EphB4 (sEphB4) was provided by Vasgene Inc. (Los Angeles, CA). 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 TGCGGCCGCTTAATGGTGATGGTGATGATGCTGCTCCCGCCAG-CCCTCGCTCTCAT as reverse primers. The DNA fragment was cloned into the mammalian expression vector pEF6/V5-His-TOPO (Invitrogen), 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 Ni-NTA Agarose (Qiagen, Valencia, CA). The purity and quantity of the recombinant proteins was tested by SDS-PAGE electrophoresis with Coomassie blue, Western blot analysis, and UV spectroscopy. Purified proteins, free of endotoxin, were dialyzed against 20 mM Tris-HCl and 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). 

Migration Assay

The CEC migration assay ¹⁷ was performed using fibronectin-coated, 24-well permeable cell inserts (Boyden chamber, Transwell; Corning Costar, Corning, NY) with 8-μm pore size. Briefly, 100 μL of 5 × 10⁵ cells were seeded onto the upper compartment of a modified Boyden chamber. After a 1-hour attachment and 5-hour incubation, with or without sEphB4, the inserts were fixed with methanol and stained with hematoxylin. The number of migrated cells was counted using phase-contrast microscopy (320×). Four randomly chosen fields were counted per insert. The experiment was repeated three times. 

Tube Formation Assay

Two-dimensional tube formation was measured in a collagen gel. A collagen mixture (Vitrogen 100; Cohesion, Palo Alto, CA) was placed in 24-well plates and incubated at 37°C for 1 hour to form gels. After polymerization of the gels, 1.0 × 10⁵ CECs were seeded on each well and incubated with endothelial basal medium (1.2 mL) containing VEGF (20 ng/mL; R&D Systems Inc., Minneapolis, MN), with or without sEphB4 (0.1 to 3 μg/mL) for 24 hours. Tube formation was documented by photography with a phase-contrast microscope. The experiment was repeated three times. 

Western Blot Analysis

CECs were grown in six-well plates in EGM with 10% FBS until confluence. Cells were lysed and proteins were separated by use of a Tris-HCl 4% to 12% polyacrylamide gradient gel (Ready Gel, Bio-Rad Laboratories, Hercules, CA) at 120 V. The proteins were transferred to a polyvinylidene difluoride (PVDF) blotting membrane (Millipore, Bedford, MA). The membranes were blocked in 5% milk and probed with polyclonal anti-EphrinB2 (Santa Cruz Biotech, Santa Cruz, CA) or mouse monoclonal EphB4 (Clone 265; Vasgene Therapeutics) antibody for 2 hours at room temperature. Membranes were washed and incubated with a horseradish peroxidase (HRP)–conjugated secondary antibody (Vector Laboratories, Burlingame, CA) for 1 hour at room temperature. Images were developed by adding enhanced chemiluminescence (ECL) detection solution (GE Healthcare, Cleveland, OH). ¹⁷  

Laser-Induced CNV

A modified rat laser- model of CNV was used to evaluate the effects of sEphB4 on choroidal angiogenesis in vivo. ¹⁸ Thirty-five adult male pigmented rats (Brown-Norway; Charles River Laboratories, Wilmington, MA) were used in the study, and all procedures were conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and approved by the University of Southern California (USC) Animal Use Committee. The rats were anesthetized for all procedures with an intramuscular injection of 0.2 mL of a 50:50 mixture of ketamine hydrochloride (20 mg/mL) and xylazine hydrochloride (100 mg/mL; both from Phoenix Pharmaceutical Inc., St. Joseph, MO). The pupils were dilated with 5% phenylephrine hydrochloride and 0.8% tropicamide, and four photocoagulation lesions were delivered with a diode green laser (150 mW, 0.05 s, 75 μm) between the retinal vessels in a peripapillary distribution in each fundus. Production of a subretinal bubble at the time of laser treatment confirmed the rupture of Bruch’s membrane. sEphB4 (10 μL, containing 0.1, 0.3, 1, or 3 μg soluble protein), or vehicle (0.9% saline) was injected with a 30-gauge needle (Hamilton, Reno, NV) introduced into the vitreous 200 μm posterior to the limbus, 3 and 7 days after laser photocoagulation. Fluorescein angiograms and histologic examinations were performed at day 14 after laser surgery. 

Fluorescein Angiography

Fluorescein angiography was performed on day 14, as described. ¹⁹ The intensity of staining in late-phase FA (180–200 seconds after fluorescein injection) was scored in a blinded manner (0, no staining; 1, slight leakage; 2, moderate leakage; and 3, heavy leakage), using standardized photographs for each stage. 

Immunohistochemical Staining and Immunofluorescence

Cryostat sections of rat eyes were obtained from control animals or animals 14 days after laser surgery. Thawed tissue sections were air dried, rehydrated with phosphate-buffered saline (pH 7.4), and blocked with 5% normal goat serum for 15 minutes. Sections were incubated sequentially with rabbit anti-EphrinB2 or goat anti-EphB4 (1:100; Santa Cruz Biotechnology, Santa Cruz, CA) antibody for 60 minutes, biotinylated secondary anti-rabbit or anti-goat antibody (1:200; Vector Laboratories, Burlingame, CA), and streptavidin peroxidase (Vector Laboratories), with three PBS washes in between. Immunoreactivity was visualized with the peroxidase substrate amino ethyl carbazole (AEC kit; Zymed Laboratory Inc., South San Francisco, CA). Slides were rinsed with tap water, counterstained with hematoxylin, and mounted with glycerin-gelatin medium. For immunofluorescence studies, frozen sections were incubated with fluorescein-tagged monoclonal anti-EphB4 (Clone 138; Vasgene Therapeutics) and anti-EphrinB2 antibody (Clone 2B5; Vasgene Therapeutics), both at 2 μg/mL final concentration, for 60 minutes. After three washes with PBS, slides were counterstained with propidium iodide and mounted in mounting medium for fluorescence (Vector Laboratories). 

Histologic Analysis of CNV Lesions

On day 14, after the rats were killed with an overdose of pentobarbital sodium, the eyes were enucleated and snap frozen in optimal cutting temperature solution (Ames/Miles, Elkhart, IN). Each lesion was cryostat sectioned at 6-μm intervals and sections placed on glass slides coated with poly-l-lysine (Sigma-Aldrich). The sections were fixed in acetone for 5 minutes and stored at −70°C. Hematoxylin and eosin (H&E)–stained sections were evaluated for CNV formation, and immunohistochemical staining for CD31 (Sigma-Aldrich) was used as a marker to identify endothelial cells. 

Quantitative Assessment of CNV Lesion Size

Rat eyecups were fixed in 4% paraformaldehyde and then permeabilized in 1% Triton X-100 for 2 hours. After removal of the anterior segment and neural retina, fluorescein-labeled isolectin B4 (specific endothelial cell marker, 1:100; Vector Laboratories) was added to the eyecup and incubated at 4°C overnight. Samples were coverslipped (Vectashield medium; Vector Laboratories), and the lectin labeling was examined with the 20× objective of a laser scanning confocal microscope (LSM; model 510; Carl Zeiss Meditec, Dublin, CA). Fluorescence volume measurements were accomplished by creating image stacks of optical slices within lesions. The image stacks were generated in the z-plane with the confocal microscope set to excite at 488 nm and detect at 505 to 530 nm. Images were further processed using the LSM software, by closely circumscribing and digitally extracting the fluorescent lesion areas throughout the entire image stack. The extracted lesion was processed through the LSM topography software to generate a digital topographic image representation of the lesion and an implicit image volume. Units for implicit volume were designated as VolSurfArea (in square micrometers). Results reported represent mean total surface area (VolSurfArea) ± SEM. 

Statistical Analysis

All in vitro experiments were performed at least three times. Data are presented as the mean ± SEM. For Western blot analysis, band intensity after electrophoresis was evaluated with ImageJ software (version 1.32j; available by ftp at zippy.nimh.nih.gov/ or at http://rsb.info.nih.gov/nih-image; developed by Wayne Rasband, National Institutes of Health, Bethesda, MD) and normalized to β-actin. The area of fluorescent signal (for immunofluorescent stains) and tube length and branch points (for tube formation assay) were evaluated by two masked observers who used morphometric analyses from the ImageJ software. Statistical evaluations were performed with Student’s t-test. For evaluation of multiple groups, ANOVA was performed; one-way, repeated-measures analysis of variance was used to demonstrate statistically significant differences in the mean values among the treatment groups, whereas a pair-wise multiple-comparison procedure was used to demonstrate statistically significant differences among each of the groups. The accepted level of significance for all tests was P < 0.05. 

Results

Expression of EphrinB2 and EphB4 in CECs and Rat Retina

Western blot analysis of CEC cultures showed that these early-passage cells expressed both EphrinB2 and EphB4; however, the expression of EphB4 was more than eight times more than that of EphrinB2 (Fig. 1) . Immunohistochemical staining of normal rat retina revealed that the choroidal arteries were positive for EphrinB2, whereas the endothelial cells of the choriocapillaris were positive for both EphrinB2 and EphB4 (Fig. 2)The retinal pigment epithelium (RPE) and photoreceptor cells were negative for both EphrinB2 and EphB4. 

Inhibition of CEC Migration

Early passage CECs migrate effectively in serum-free conditions supplemented with VEGF. VEGF at a dose of 20 ng/mL increased the number of migrating cells by 50% in 5 hours. The addition of sEphB4 significantly reduced the migration of CECs in response to VEGF over the 5-hour assay time by greater than 50%, to a level significantly lower than those cells migrating in the absence of VEGF (Fig. 3) . ANOVA revealed that each of the groups (Control versus VEGF, Control versus VEGF+sEphB4, and VEGF versus VEGF+sEphB4) showed statistically significant differences (P < 0.001). 

Inhibition of CEC Tube Formation

Early passage CECs grown in collagen matrix gel alone showed minimal tube formation (Fig. 4A) . When 20 ng/mL VEGF was added to the gel, prominent tube formation occurred (2.7-fold greater vessel length per high-power field [HPF]) with clearly defined branching (3.4-fold increase in number of branch points per HPF; Fig 4B ). The addition of sEphB4 to the gel at dose levels ranging from 0.5 to 3 μg/mL resulted in a dose-dependent inhibition of tube formation. At the highest concentration of sEphB4 (3 μg/mL), the inhibition of tube formation was nearly complete, with no difference in number of branch points or length of vessels, compared with nonstimulated CECs. 

Expression of EphrinB2 and EphB4 in Rat CNV

Immunohistochemical staining revealed that the cells lining the neovascular channels in the CNV lesions were positive for EphB4; in contrast, these cells showed only weak, focal positivity for EphrinB2 (Fig. 5) . Control sections without primary antibody showed no background staining. These data indicate that laser-induced CNV membranes have an increase in EphB4-positive cells. Lack of apparent induction of EphrinB2 may be due in part to the lower signal intensity of the EphrinB2 antibody. 

Inhibition of Experimental CNV

Laser-induced photocoagulation reproducibly resulted in the development of CNV 14 days after laser, as demonstrated by fluorescein angiography (Fig. 6) , CNV volume (Fig. 7) , and histology (Fig. 8) . Whereas CNV membranes in buffer-injected control rats showed slight-to-moderate fluorescein leakage score, rats treated with intravitreous injections of sEphB4 showed a dose-dependent decrease in fluorescein leakage (Fig. 6) . At the highest dose of sEphB4 tested (3 μg), fluorescein leakage scores were in the none-to-slight range and were >50% reduced, compared with controls (P < 0.05). CNV volume analysis was performed in FITC-labeled isolectin B4-stained retinal flatmounts. As shown in Figure 7 , the volume of laser-induced CNV membranes was reduced significantly in sEphB4 (3 μg)-treated eyes (VolSurfArea, 620 μm²) compared with control lesions (VolSurfArea, 1484 μm²; P < 0.05). H&E-stained tissue sections of the CNV membranes were also evaluated and showed that both sEphB4-treated eyes and buffer-injected controls contained true CNV membranes centered on a disrupted Bruch’s membrane with vascular channels, fibrous stroma, and macrophage infiltration. The lesions in the sEphB4-injected eyes, however, appeared to have a greater than 55% reduction in endothelial cell area, as identified by CD31 staining (Fig. 9) . Whereas CNV lesions in control eyes showed large vascular channels perfused by erythrocytes, the vascular channels in the sEphB4-treated CNV lesions were often smaller and slitlike, with partially reduced erythrocyte content (Fig. 8) . In addition, the lesions in sEphB4-treated lesions contained fibrous stroma that was decreased in proportion to the extent of neovascularization. Macrophage infiltration was observed in both control and sEphB4-treated lesions. Control nonlasered eyes, evaluated 14 days after injection of sEphB4 at a concentration of 10 μg, did not show any morphologic abnormalities in multiple 1-μm plastic sections of retina (data not shown). In particular, there was no evidence of proliferation of fibroblasts or infiltration of macrophages. 

Discussion

The Eph RTK family represents the largest class of RTKs, and they play a critical role in embryonic patterning and neuronal targeting. ^{1 2 3 4} The role of this family in angiogenesis has recently begun to emerge, in studies demonstrating a role in vascular development and in the interplay between VEGF and Eph ligands. ^{20 21} VEGF induces the expression of EphrinB2, thus increasing the activation potential of EphB4. ²² EphrinB2 induces endothelial cell migration and capillary assembly in vitro, ^{1 2 3 20} and angiogenesis in a corneal pocket assay in vivo. ¹⁴ Indeed, expression of Ephrin and its receptor Eph was observed in breast tumors ²³ and associated vasculature and in adult vascular sites. ²⁴  

In the current experiments, we found that both EphrinB2 and EphB4 are expressed in choroidal vessels of adult rats, suggesting that Eph RTK and its ligand are involved in vascular development in adult vascular remodeling in addition to the known role in embryonic angiogenesis. EphrinB2 was found both in large choroidal vessels and in the choriocapillaris, whereas EphB4 was most prominent in the choriocapillaris. Consistent with this finding, we showed that expression of EphB4 was greater than EphrinB2 in cultured bovine CECs, suggesting that EphB4 forward signaling may be dominant in health. Alternatively, in vitro culture conditions may favor higher EphB4 expression. 

Targeted disruption of EphrinB2/EphB4, results in embryonic lethality due to defects in primary capillary network remodeling and subsequent patterning defects in the embryonic vasculature, ^{9 10 12 22} whereas other studies suggest that Eph signaling is also critical in tumor angiogenesis, ^{25 26} and that inhibition of tumor growth can be achieved with a similar monomeric preparation of sEphB4. ²⁶ In the present experiments, we demonstrate that sEphB4, acting in a dominant negative manner, ²⁶ inhibits choroidal angiogenesis in vitro and in vivo. 

This is the first demonstration of a soluble monomeric form of EphB4 that inhibits endothelial tube formation, even in the presence of VEGF. Our data reinforce the findings described recently by Zamora et al., ²⁷ in which intravitreal injection of EphrinB2/Fc or EphB4/Fc chimeric protein was shown to inhibit oxygen-induced proliferative retinopathy. Apart from using a laser-induced neovascularization model, our study differs from that of Zamora et al. in that we used a monomeric form of sEphB4, which lacks the Fc fragment and thus dimerization capability. The chimeric EphB4/Fc and EphrinB2/Fc proteins are dimers capable of phosphorylating the complementary membrane protein. In contrast, the monomeric form of sEphB4 used in this report inhibits EphrinB2-induced phosphorylation of EphB4 and vice versa (data not shown) by competing for binding EphrinB2. In addition, we provided in vitro evidence that sEphB4 inhibits CEC migration and tube formation, providing a mechanistic basis for the inhibition of CNV membrane observed in vivo. 

The pathogenesis of retinal neovascularization and CNV includes both common and distinct mechanisms. ²⁸ Neovascular lesions from both sites are associated with expression of a similar pattern of angiogenic growth factors including VEGF, basic fibroblast growth factor, transforming growth factor-β, and connective tissue growth factor. ^{17 28 29} In this article, we provide evidence suggesting that the retinal and choroidal vasculature, as well as the neovascular lesions derived from these sites, differentially express EphrinB2 and EphB4. Previous studies have shown that cultured microvascular retinal endothelial cells express EphrinB2 strongly over EphB4. In support of this contention we observed immunoreactivity for EphrinB2 on retinal vessels in tissue sections from rat retina (results not shown). In contrast, in our study cultured CECs showed much stronger expression of EphB4 than of EphrinB2, and the choriocapillaris expressed both EphB4 and EphrinB2 in tissue sections of rat retina. A recent pathology study showed that human retinal fibrovascular lesions were positive for EphrinB2 and not EphB4. ³⁰ Consistent with our findings in the normal choroid, laser-induced CNV membranes were most positive for EphB4, with only weak expression for EphrinB2. Based on these and previous studies showing that EphB4 is associated with veins and EphrinB2 is associated with arteries, ²⁴ we suggest that the choriocapillaris and CNV membranes express antigens associated with both veins and arteries, with an induction of EphB4 expression after laser injury. 

The mechanism by which sEphB4 mediates the inhibition of CNV could involve blocking of EphrinB2 activation, or reverse inhibition of EphB4 signaling and downstream signaling pathways, including PI3K. ¹⁵ This may result in (1) direct inhibition of cell proliferation as shown by reduced number of CD31-positive cells, (2) inhibition of CEC migration, (3) inhibition of tube formation and subsequent new vessel assembly, ^{31 32} and (4) reduction of CNV leakage by effects on vessel maturation or patterning of perivascular cells. The possibility that sEphB4-mediated inhibition of angiogenesis may involve interference with the function of VEGF must also be considered. ^{20 33} Previous studies have shown that VEGF can induce expression of EphrinB2 which in turn activates EphB4. ^{8 9 11} VEGF provides a relatively long-range signal, whereas EphrinB2 provides a contact signal, ⁹ both of which may cooperate to regulate angiogenesis in the laser model. Neovascular membrane from CNV is associated with the induction of several angiogenic growth factors, including VEGF, basic fibroblast growth factor, transforming growth factor-β, and connective tissue growth factors. Inhibition of CNV membrane with sEphB4 indicates that this polypeptide can block the effects of several different growth factors. sEphB4 may also modulate other functions such as interaction of EphB4 receptor with matrix proteins ^{22 34} or other cell-membrane–associated proteins, with eventual antiangiogenic effect. 

In summary, we provide the first evidence that sEphB4 inhibits experimental CNV. In addition, in vitro studies show that treatment of CECs with sEphB4 inhibits migration and tube formation induced by VEGF. Further studies may determine the feasibility of using sEphB4 in the treatment of human CNV. 

 Supported by National Eye Institute Grant EY01545 (SJR, DRH), National Cancer Institute Grant CA79218 (PSG), and core Grant EY03040; the Arnold and Mabel Beckman Foundation; a grant to the Department of Ophthalmology from Research to Prevent Blindness Inc., and Vasgene Therapeutics.

 Submitted for publication April 23, 2005; revised August 5, 2005; accepted October 6, 2005.

 Disclosure: S. He, None; Y. Ding, None; J. Zhou, None; V. Krasnoperov, Vasgene Therapeutics, Inc. (E); S. Zozulya, Vasgene Therapeutics, Inc. (E); S. R. Kumar, None; S. J . Ryan, None; P. S. Gill, Vasgene Therapeutics, Inc. (F, I); D. R. Hinton, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

EphrinB2 and EphB4 expression in bovine CECs. EphB4 expression was assessed in primary cultures of CECs. The positive control was tumor cell line SCC-15 and the negative control was transformed fibroblast cell line T1. An equal amount of total protein lysate was analyzed by Western blot, with EphB4- and EphrinB2-specific antibodies. The blot was stripped and reprobed, with β-actin included to control for loading. The difference of the expression of EphB4 and EphrinB2 in the Western blot was quantified by computer (*P < 0.0).

Figure 2.

 

Immunohistochemical localization of EphrinB2 and EphB4 in choroidal vessels of normal rat eye. Immunoreactivity was demonstrated using aminoethyl carbazole as the chromogen (red) with a hematoxylin (blue) nuclear counterstain. EphrinB2 and EphB4 were both expressed in the choriocapillaris (vertical arrows, top, middle); however, EphrinB2 staining was also prominent in the choroidal arteries (arrow, top inset). There was no EphrinB2 or EphB4 staining in the retinal pigment epithelium or neural retina. Control sections stained in parallel with nonimmune IgG showed no background staining (bottom). Bar, 50 μm.

Figure 3.

 

sEphB4 inhibited VEGF-induced migration of CECs. Bovine CECs were stimulated with VEGF (20 ng/mL), and migration measured in a modified Boyden chamber assay in the presence or absence of sEphB4 (3 μg/mL). ANOVA revealed that each of the groups (+ Control versus VEGF, ♦ Control versus VEGF+sEphB4, and ∗ VEGF versus VEGF+sEphB4) showed statistically significant differences (P < 0.001).

Figure 4.

 

sEphB4 inhibited tube formation. Phase-contrast photomicrographs showing CEC tube formation induced by VEGF (20 ng/mL) was inhibited by treatment with sEphB4. (A) Control without VEGF; (B–F) all received VEGF; and (C–F) received increasing doses of sEphB4 (0.1, 0.5, 1, and 3 μg/mL in C, D, E, and F, respectively). Tube formation decreased with increasing doses of sEphB4. Quantitation of tube formation revealed that sEphB4 treatment resulted in dose-dependent inhibition of total branch points per high-power microscopic field (G), and total length of vessels (H). Bar, 50 μm.

Figure 5.

 

Confocal immunofluorescent analysis of EphrinB2 and EphB4 expression in laser-induced CNV. (A–C) are from one CNV lesion shown in phase contrast in (A), and (D–F) are from a second lesion shown in phase contrast in (D). (B) Focal EphrinB2 expression (green) is shown in cells associated with vascular channels within the CNV membrane (2× in inset); (E) prominent Eph B4 expression in association with vascular channels within the CNV membrane (2× in inset). (C, F) Sections stained in parallel without primary antibody. Sections were counterstained with propidium iodide (red) to identify nuclei. Bar, 25 μm.

Figure 6.

 

Effect of sEphB4 on fluorescein leakage in rat CNV. FA was performed 14 days after laser photocoagulation and intravitreal injection of sEphB4. sEphB4 injection resulted in a dose-dependent inhibition of fluorescein leakage (P < 0.05).

Figure 7.

 

The effects of sEphB4 (3 μg) injection on CNV volume on day 14, as measured by confocal analysis of FITC-labeled isolectin-B4-stained flatmounts. (A) Representative CNV membrane from a control animal; (B) membrane from an animal treated with sEphB4. (C) The reduction in implicit CNV volume (VolSurfArea in square micrometers) was significant in treated rats compared with the control (*P < 0.05). In each group, 14 rats and 65 lesions (FA and CNV volume) were included.

Figure 8.

 

Effect of sEphB4 (3 μg) injections on CNV membrane size and vessel maturation. (A) Light microscopic image of a CNV membrane from a control rat at day 14. (B) CNV membrane from a sEphB4-injected rat at day 14. There was a decrease in the area of CNV membrane (white arrows) in treated rats compared with control animals. Blood vessels in control lesions frequently contained red blood cells (black arrows), whereas intravascular red blood cells in treated lesions appeared to be reduced in number. H&E stain. Bar, 100 μ

Figure 9.

 

Confocal immunofluorescent images of CD31-positive endothelial cells in rat CNV membranes in control (A), and sEphB4-injected (B, 3 μg) animals. CD31 staining was identified by immunofluorescence. (C) There was more extensive CD31 positivity in CNV membranes from control eye sections than in the sEphB4-injected rats (*P < 0.05). Bar, 50 μm.

The authors thank Tom Odgen for review of the manuscript and Ernesto Barron and Chris Spee for technical assistance.