Free
Retina  |   February 2015
The Effect of AMA0428, a Novel and Potent ROCK Inhibitor, in a Model of Neovascular Age-Related Macular Degeneration
Author Affiliations & Notes
  • Karolien Hollanders
    KU Leuven-University of Leuven, Department of Neurosciences, Laboratory of Ophthalmology, Leuven, Belgium
  • Tine Van Bergen
    KU Leuven-University of Leuven, Department of Neurosciences, Laboratory of Ophthalmology, Leuven, Belgium
  • Nele Kindt
    Amakem Therapeutics, Diepenbeek, Belgium
  • Karolien Castermans
    Amakem Therapeutics, Diepenbeek, Belgium
  • Dirk Leysen
    Amakem Therapeutics, Diepenbeek, Belgium
  • Evelien Vandewalle
    KU Leuven-University of Leuven, Department of Neurosciences, Laboratory of Ophthalmology, Leuven, Belgium
    KU Leuven-University of Leuven, University Hospitals Leuven, Department of Ophthalmology, Leuven, Belgium
  • Lieve Moons
    KU Leuven-University of Leuven, Department of Neurosciences, Laboratory of Ophthalmology, Leuven, Belgium
    KU Leuven-University of Leuven, Department of Biology, KU Leuven, Leuven, Belgium
  • Ingeborg Stalmans
    KU Leuven-University of Leuven, Department of Neurosciences, Laboratory of Ophthalmology, Leuven, Belgium
    KU Leuven-University of Leuven, University Hospitals Leuven, Department of Ophthalmology, Leuven, Belgium
  • Correspondence: Ingeborg Stalmans, University Hospitals Leuven, Department of Ophthalmology, Kapucijnenvoer 33, B-3000, Leuven, Belgium; Ingeborg.Stalmans@uzleuven.be
Investigative Ophthalmology & Visual Science February 2015, Vol.56, 1335-1348. doi:10.1167/iovs.14-15681
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Karolien Hollanders, Tine Van Bergen, Nele Kindt, Karolien Castermans, Dirk Leysen, Evelien Vandewalle, Lieve Moons, Ingeborg Stalmans; The Effect of AMA0428, a Novel and Potent ROCK Inhibitor, in a Model of Neovascular Age-Related Macular Degeneration. Invest. Ophthalmol. Vis. Sci. 2015;56(2):1335-1348. doi: 10.1167/iovs.14-15681.

      Download citation file:


      © 2016 Association for Research in Vision and Ophthalmology.

      ×
  • Supplements
Abstract

Purpose.: Rho kinase (ROCK) is associated with VEGF-driven angiogenesis, as well as with inflammation and fibrosis. Therefore, the effect of AMA0428, a novel ROCK inhibitor, was studied in these processes, which highly contribute to the pathogenesis of neovascular AMD.

Methods.: The effect of AMA0428 (0.5–5.0 μM) on human umbilical vein endothelial cells (HUVECs), human brain microvascular endothelial cells (HBMECs), and human brain microvascular pericytes (HBVPs) was determined using cell viability (WST-1), apoptosis (caspase 3/7), and migration (scratch and under-agarose) assays. The in vivo response was investigated using a laser-induced choroidal neovascularization (CNV) mouse model, in which intravitreal injections of AMA0428, murine anti–VEGF-R2 mAb (DC101), or placebo was given. Outcome was assessed by analysis of inflammation (CD45), angiogenesis (FITC-dextran), vessel leakage (Texas Red-conjugated Dextran and FITC-labeled lectin) and fibrosis (Sirius Red/Collagen I).

Results.: The AMA0428 dose-dependently reduced proliferation and VEGF-induced migration of HUVEC and HBMEC (P < 0.05). No significant effect was seen on HBVP proliferation; however, migration and pericyte recruitment were enhanced (P < 0.05) by AMA0428 administration. There was no apoptosis induction. The AMA0428 significantly reduced CNV and vessel leakage 2 weeks after laser treatment, comparable to DC101. In addition, AMA0428 inhibited inflammation on day 5 by 42% (P < 0.05) and collagen deposition on day 30 by 43% (P < 0. 05), whereas DC101 had no effect on inflammation nor on fibrosis.

Conclusions.: The results suggest that targeting ROCK with AMA0428 not only reduces neoangiogenesis, but also blocks inflammation and fibrosis (contrary to VEGF suppression). These results point to a potential therapeutic benefit of ROCK inhibition in neovascular AMD.

Age-related macular degeneration (AMD) was first recognized approximately 130 years ago and has been extensively investigated over the past decades in an attempt to unravel the underlying disease mechanisms. However, the pathogenesis remains incompletely understood. Age-related macular degeneration is the most frequent cause of irreversible vision loss in the developed world1,2 and is becoming an important socio-medical problem, due to the increased life expectancy in the western world. Two different subgroups of AMD are classically distinguished: an atrophic lesion or dry form and an exudative, wet form. Dry AMD is the most common, and less sight-threatening type of macular degeneration.3 The crucial difference between the two subgroups is the development of choroidal neovascularization (CNV) in wet AMD,4 which causes rapid progression to significant sight loss.5 
In addition to neovascularization, wet AMD is characterized by inflammation and fibrosis,6 processes that generally contribute to wound healing. Anti-VEGF administration is currently the most commonly used treatment option for active CNV.7,8 However, there is only limited evidence that anti-VEGF therapy might reduce fibrosis and inflammation.9,10 A major issue related to anti-VEGF therapy is that VEGF is a key player in vascular and neuronal outgrowth and maintenance,11 and its inhibition can result in blood vessel regression and neurodegenerative processes.12,13 The use of anti-VEGF may therefore give rise to local side effects, but also potentially to important systemic side effects, including cardiovascular effects, cerebrovascular accidents, and myocardial infarction.14 Moreover, conventional therapy that focuses primarily on inhibiting angiogenesis may not be optimal because of the inflammatory and fibrotic component in AMD. Therefore, new therapeutic strategies that address the different pathological processes are required to provide additional options to better manage AMD. 
As our understanding of the pathophysiologic mechanism of AMD improves, the opportunity to refine treatment approaches increases. An attractive novel therapeutic option in the treatment of wet AMD is the inhibition of Rho kinase. To date, there are two Rho kinase isoforms that have been described, ROCK1 (ROKβ) and ROCK2 (ROKα).15 They regulate numerous cellular processes, such as cell proliferation, survival, permeability,16 endothelial cell (EC) polarity, and directed cell migration.17 The Rho/ROCK pathway also mediates vascular smooth muscle cell and pericyte recruitment,18 which is essential for blood vessel maturation and stability. Indeed, pericyte recruitment lags behind endothelial sprouting, leading to a window of pericyte absence that allows for vascular plasticity resulting in survival, growth, and remodeling, dependent on the presence of different growth factors such as VEGF.19 This process is responsible for the “leaky nature” of the proliferating vessels and, together with a breakdown of the blood-retinal barrier (BRB), causes edema formation.20 Importantly, Ren et al.21 demonstrated that cell spreading requires transient downregulation of RhoA activity. Defects in cell spreading and directional motility, present in the pathology of AMD, may therefore be rescued by ROCK inhibitors.20 Moreover, a potential advantage of ROCK inhibition, compared with anti-VEGF therapy, would be its additional effect on inflammation and fibrosis.2226 Although AMD is not a predominantly inflammation-driven disease, unlike uveitis and scleritis, inflammation has been found to play a critical role in the disease pathogenesis and progression.2729 Local inflammation accompanied by complement activation, cell lysis, and immune responsiveness are important facets of AMD pathogenesis and progression to wet AMD.30,31 Both innate immunity and autoimmune components are known to be involved in AMD development.27 As such, age-related drusen, a by-product of these chronic inflammatory events, accumulate between the RPE and Bruch's membrane. This will lead to recruitment of tissue-destructive macrophages, microglia, and inflammatory cell-derived cytokines and chemokines.32 Moreover, several diseases, such as wet AMD and diabetic retinopathy, which are characterized by macular edema, retinal and vitreous hemorrhage, and fibrovascular scarring, have a final common pathophysiological denominator, namely the retinal response to injury, with chronic wound healing leading to fibrosis. 
In this study, AMA0428, a potent and novel ROCK inhibitor derived from recent work of Amakem Therapeutics was used (Ref. 33 and Boland S, oral communication of unpublished results, 2013). We focused on the three major processes involved in AMD: inflammation, angiogenesis, and fibrosis. Inhibition of ROCK could be an alternative/additive therapy for anti-VEGF in the treatment of wet AMD, without the systemic side effects of anti-VEGF. 
Materials and Methods
Cells and Culture Conditions
Human brain microvascular ECs (HBMECs; Sciencell Research Laboratories, Carlsbad, CA, USA) and human umbilical vein ECs (HUVECs; Lonza, Walkersville, MD, USA) were cultured in 0.1% gelatin- (Invitrogen, Carlsbad, CA, USA) coated T75 flasks (Sciencell Research Laboratories). Both cell types were maintained in complete endothelial growth medium (EGM)-2, supplemented with 5% fetal bovine serum (FBS; Thermo Fisher Scientific, Rochester, NY, USA), 0.1% human recombinant epidermal growth factor, 0.4% recombinant human FGF, 0.1% VEGF, 0.1% recombinant lung insulin-like growth factor, 0.04% hydrocortisone, 0.1% ascorbic acid, 0.1% heparin, and 0.1% gentamicin sulphate amphotericin-B (EBM-2 bullet kit; Lonza). 
Human brain vascular pericytes (HBVPs; Sciencell Research Laboratories) were cultured in poly-L-lysine- (Sciencell Research Laboratories) coated T75 flasks. The HBVPs were maintained in complete pericyte medium, supplemented with 2% FBS, 10% pericyte growth supplement, 100 U/mL penicillin, and 100 μg/mL streptomycin (Sciencell Research Laboratories). 
All cell lines were routinely maintained at 37°C in a humidified atmosphere of 5% CO2 and medium was replaced every 2 to 3 days. Cells between third and sixth passage were used in all experiments. All in vitro assays were repeated at least three times. 
Apoptosis Assay
Human brain microvascular ECs and HBVPs were trypsinized and were seeded in coated (0.1% gelatin for ECs; poly-L-lysine for pericytes) 96-well plates (Costar; Corning, Inc., Corning, NY, USA) at an initial density of 10 × 103 cells per well in 100 μL complete medium. Plates were incubated at 37°C in a humidified atmosphere of 5% CO2. Twenty-four hours later, the medium was replaced with fresh medium, with or without AMA0428 (0.5, 1, 2.5, 5, 10, 25 μM; Amakem Therapeutics, Diepenbeek, Belgium) or the equivalent of vehicle/dimethyl sulfoxide (DMSO; Sigma-Aldrich Corp., St. Louis, MO, USA). Two days later, Caspase-GloR 3/7 Reagent (Promega, Leiden, The Netherlands) was added to the cell suspension (1:1 ratio) and incubated at room temperature for 1 hour. Afterward, the supernatant was transferred into a white-bottom 96-well plate (Thermo Scientific, Roskilde, Denmark), in which the luminescence of each sample was measured in a plate-reading luminometer (Microplate Luminometer LB96V, EG&G Berthold, VIC, Australia). 
Proliferation Assay
Subconfluent HBMECs, HUVECs, and HBVPs were trypsinized and were seeded at 37°C in coated 96-well plates at an initial density of 3 × 103 cells per well in 100 μL complete medium. The following day, cells were incubated with AMA0428 or DMSO at different concentrations (0.5, 1, 2.5, 5 μM) in their respective complete media. For HBMECs and HUVECs, administration of an anti-VEGF antibody (bevacizumab [Avastin]; Genentech/Roche Therapeutic, Welwyn Garden City, UK), at a concentration of 2.5 mg/mL and 5.0 mg/mL, was used as a positive control in full medium. In serum-free medium, EC were challenged with recombinant human VEGF165 (rhVEGF165; 100 ng/mL or 5.22nM; R&D Systems, Minneapolis, MN, USA) to increase proliferation and treated with AMA0428 (0.5, 1, 2.5, 5 μM) or bevacizumab (2.5 mg/mL or 16.8 μM, 5 mg/mL or 33.6 μM) as positive control. 
For HBVPs, recombinant human platelet-derived growth factor (rhPDGF-BB; R&D Systems) was used as a positive control and rhVEGF165 (R&D Systems) was added as a negative control. In another series of experiments, HBVPs were serum starved (medium supplemented with 0.1% FBS) overnight, 6 hours after cell seeding. The medium of the cells was changed to fresh serum-free medium containing AMA0428 or DMSO at different concentrations (0.5, 1, 2.5, 5 μM). 
Forty-eight hours after growth factor or compound administration, cell proliferation was assessed in all experiments using the WST-1 Cell Proliferation Assay System (Roche, Mannheim, Germany). Complete or serum-free medium was used as control. 
Scratch Assay
The effect on cell migration was evaluated using an in vitro scratch assay, performed as previously described.34 Human brain microvascular endothelial cells and HBVPs were grown to confluence in 48-well tissue culture dishes and were serum starved overnight. The next day, the monolayer was scratched with a plastic pipette tip to remove the confluent cells by two perpendicular linear scratches to create linear wounds. The HBMECs were incubated at 37°C in starvation medium with or without AMA0428 (0.5–5 μM), DMSO, and/or 20 ng/mL rhVEGF165. Recombinant human PDGF-BB (20 and 40 ng/mL) and/or AMA0428 (0.5–5 μM), was added to HBVP. After 2 hours, the movement of cells into the wound area was measured using a crossline micrometer on a Primo Vert 1 microscope (both from Zeiss, Oberkochen, Germany). The shortest distance between the edges of migrated cells (including protrusions) from both sides was measured to determine the effect of AMA0428 after wounding of a monolayer of cells. These measurements were at four predefined points at start and at 2, 4, 6, 8, and 24 hours after wounding.35 
Under-Agarose Migration Assay
To investigate the effect of ROCK inhibition on pericyte recruitment, an under-agarose migration assay36 was performed as described. Briefly, 0.48 g ultrapure agarose was suspended in 10 mL double-distilled water (ddH2O). The agarose solution was briefly heated in a microwave until it boiled. Next, 2% of the agarose solution was added to 1% FBS EGM-2. This solution was added into tissue culture dishes, and was allowed to solidify at room temperature. Afterward, two 5-mm wells were punched in the gels at 1 mm distance from each other. The gels were equilibrated at 37°C, 5% CO2 for 1 hour. Next, any condensation that might have formed in the wells was removed and twenty thousand HUVEC were plated in one well in 20 μL EGM-2 medium with 1% FBS. After 3 hours the HUVEC were treated with AMA0428 in different concentrations (1, 2.5, 5 μM) or the equivalent of DMSO. Thirty minutes later, HBVPs (2 × 104 cells in the same medium) were added to the other 5-mm well. The distance of HBVP migration toward the HUVEC was measured with a crossline micrometer in a Primo Vert 1 microscope (both from Zeiss) every day for 4 successive days. 
Permeability Assay
Endothelial cell permeability was determined by measuring the passage of FITC-labeled dextran through a tight monolayer of cells. For this, HUVECs were seeded at a concentration of 24,000 cells/transwell on 0.2% gelatin-coated transwell membranes (Corning 0.4-μm polyester membrane, CLS3470; Sigma-Aldrich Corp.). After 2 days, a tight monolayer of cells was formed and the transwell inserts were transferred to a new 24-well plate. The culture medium was supplemented with vehicle, VEGF165 (2.08 μM = 40 ng/mL, V7259; Sigma-Aldrich Corp.) with/without AMA0428 (0.01, 0.1, 1 μM). Next, 70 kDa FITC-dextran (1 mg/mL, FD70S; Sigma-Aldrich Corp.) was added on top of the cells and medium samples were collected from the lower compartment every 15 minutes for 1 hour. Fluorescence intensity was evaluated using a Fluostar Omega V1.30 microplate reader (BMG Labtech, Ortenberg, Germany). Data of three independent experiments (±SEM) are presented as mean relative values compared to VEGF treatment. 
Animals
Male C57BL/6J mice (8–10 weeks old) were obtained from Charles River Laboratories, L'Arbresle Cedex, France, and were used in all experiments. Housing and all experimental animal procedures were approved by the Institutional Animal Care and Research Advisory Committee of the KU Leuven. All animal procedures were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and the 2010/63/EU Directive for animal experiments, using the guidelines of our own institution. 
Compounds
In this study, the ROCK inhibitor AMA0428 was tested in comparison to vehicle (both provided by Amakem NV). For vehicle, H2O-PEG was used (PEG 400; Fagron NV, Waregem, Belgium). First, a safety assessment was made by intravitreally injecting 1 μL of the highest dose of the compound (1000 ng) and compared with noninjected/naïve mice, sham (puncture, no fluid injection) and vehicle-injected mice (10 mice/group; only one eye per mouse was used). The potential side effects were examined at day 14 post injection on hematoxylin and eosin (H&E)–stained retinal sections by measuring the total retinal thickness, the thickness of the outer and inner nuclear/plexiform layer, and photoreceptor layer. Also, retinal ganglion cell density was quantified two times on the same sections by counting ganglion cell nuclei at 500 μm on four sections on each side of the optic nerve head. 
Murine Model of CNV
Mice were anesthetized by an intraperitoneal (IP) injection of 60 mg/kg sodium pentobarbital (Nembutal, CEVA; Sante Animale, Brussels, Belgium) and their pupils dilated with tropicamide (Tropicol; Thea, Clermont-Ferrand, France). As described in literature,3739 three laser burns were placed with a 532-nm green laser (Ophthalas 532 EyeLite Laser Phothocoagulator; Alcon Surgical, Puurs, Belgium) at the 9, 12, and 3 o'clock positions around the optic disk using a slit lamp delivery system with a hand-held cover slide as a contact lens and Genteal Gel (Novartis, Vilvoorde, Belgium). Each spot had a size of 50 μm, laser duration of 100 ms, and a power of 400 mW, based on previous experiments in our laboratory.7 Bubble production as a sign for the rupture of the Bruch's membrane was necessary to include the spot. In all experiments, one eye per mouse was lasered, except for the expression studies (ELISA), where two eyes per mouse received laser treatment to obtain sufficient tissue. 
Expression of ROCK, VEGF, and Placental Growth Factor (PlGF).
A slight variation of the CNV model, using 30 separate laser spots on Bruch's membrane, was performed to investigate the expression of ROCK, VEGF, and PlGF in the injured choroid and retina. Mice were divided into one naïve group (nonlasered, nontreated) and three lasered groups (five mice per group; two eyes were lasered): a first nontreated group, a second AMA0428 (1000 ng) group, and a third vehicle group. Mice were treated at day 0 and day 2 via intravitreal (IVT) injections. Three days after laser treatment, the eyes were enucleated and the retina–choroid–sclera complexes were microdissected and snap-frozen. The levels of ROCK, VEGF, and PlGF were determined by ELISA. To obtain sufficient protein concentrations, each sample contained pooled tissue of two posterior segment complexes of one mouse. Cell lysates from these segments were obtained using the Mammalian Cell Lysis kit (Sigma-Aldrich Corp.) and expression levels were determined with the Quantikine ELISA kit for mouse VEGF and PlGF-2 (both from R&D Systems) and the Rho-associated kinase (ROCK) Activity ELISA Assay (Merck Millipore, Overijse, Belgium), with detection limit of 7.8 pg/mL, 1.49 pg/mL, and 100 nU, respectively. 
Treatment Regimen.
Immediately after laser treatment, C57BL/6J mice were divided into different groups and their eye was consequently injected IVT by using an analytic science syringe (SGE Analytic Science, Crownhill, Milton Keynes, UK) and glass capillaries with a diameter of 50 to 70 μm at the end, controlled by the UMP3I Microsyringe Injector and Micro-4 Controller (all from World Precision Instruments, Inc., Hertfordshire, UK). Before injection, the eye was anesthetized by topical instillation of 0.4% oxybuprocaine (Unicaine; Théa Pharma, Schaffhausen, Switzerland). One group was treated with the ROCK inhibitor, AMA0428 (1 μL; 100–1000 ng); the second group was used as control and received an injection of vehicle (H2O-PEG; 1 μL). In the positive control groups, the eye was treated IVT respectively with the well-described rat anti-mouse VEGF-R2 antibody, DC10140,41 (1 μL, 6200 ng) or an isotype-matched control antibody (1C8, 1 μL, 4800 ng; both provided by ThromboGenics NV, Heverlee, Belgium). 
In the first experiment, repeated IVT injections of the compound/control antibodies were given at similar concentrations on days 0, 4, 10, and 20 after laser treatment. These time points were based on previously performed IVT injections by Van de Veire et al.7 Mice were killed on day 5 (to evaluate inflammation), day 14 (to evaluate angiogenesis), or day 30 (to evaluate fibrosis) after laser treatment. In a second experiment, a single IVT injection with AMA0428 or vehicle was given on day 0 or day 3 and mice were killed on either day 5 or day 14 after laser treatment. In a last experiment, a combination of the most effective dose of AMA0428 and a suboptimal dose of DC101 (anti–VEGF-R2) was tested. For this, a dose-response curve was first established for both compounds. Based on these results, a dose of 1550 ng DC101 (25% of full concentration) was chosen and a dose of 1000 ng AMA0428 (full dose). All the experimental groups consisted of 10 mice per group (unless stated otherwise). 
(Immuno)histochemistry
Mice were anesthetized with an IP injection of Nembutal and killed by cervical dislocation. Afterward, both eyes were enucleated and fixed in 1% paraformaldehyde (PFA) overnight. 
To analyze the effect of AMA0428 on inflammatory cell infiltration in the CNV lesions, the retina was removed from the dissected posterior segments. These posterior eye cups, which included RPE, the choroid, and the sclera, were stored in PBS. To stain all leukocytes, a rat anti-mouse CD45 antibody (1:100; Pharmingen, Erembodegem, Belgium) was used overnight, diluted in Tris-buffered saline (TBS)-Triton 0.3%. The following day, the tissues were incubated for 2 hours with rabbit anti-rat biotin (RaRat-B)-labeled antibody (1:300; DakoCytomation A/S, Copenhagen, Denmark), diluted in TBS-Triton 0.3%. Antibody binding was visualized by fluorescent staining using streptavidin-Alexa-568 (1:200; Molecular Probes, Life Technologies, Eugene, OR, USA) in TBS-Triton 0.3% for 2 hours. The flat mounts of the posterior eye cups were mounted with Prolong Gold with 4′,6-diamidino-2-phenylindole (DAPI; Molecular Probes). 
To investigate the effect of AMA0428 on angiogenesis, retrobulbar perfusion with 200 μL FITC-conjugated dextran (50 mg/mL, Mr 2 × 106 Da; Sigma-Aldrich Corp.),42 was performed and after 2 minutes mice were killed. Vessel leakage was assessed, as previously described,7 by retrobulbar injection using a total volume of 200 μL per mouse, of which 100 μL Texas Red-conjugated Dextran 70 kDa (5 mg/mL; Molecular Probes) together with 100 μL FITC-labeled lectin (1 mg/mL, Lycopersicon esculentum; Vector Laboratories, Inc., Burlingame, CA, USA). Five minutes later, the red dye was washed out by intracardial perfusion with NaCl 0.9% for 5 minutes, to remove the tracer form the actual blood vessels. All eyes were enucleated and fixed overnight in 1% PFA. The next day, all RPE–choroid–sclera complexes were dissected and flatmounted on a slide containing a drop of Vectashield mounting medium (Vector Laboratories) to prevent the fluorescent dyes from bleaching. 
Deposition of collagen was analyzed by Sirius Red staining and collagen I immunostaining on paraffin-embedded eyes. Serial sections were cut at 7-μm thickness in five series on five glass slides. First, H&E staining was performed to localize the laser spots. For the Sirius Red and collagen I staining, the consecutive slides were used. Slides were deparaffinized, washed, and placed in Sirius Red solution (Direct Red 80 and 1.3% picric acid solution; both Sigma-Aldrich Corp.) for 60 minutes. Sections were then placed in 0.01N HCl (Prolabo, Amsterdam, The Netherlands) for 2 minutes, dehydrated and mounted with DPX mounting medium (Prosan, Gent, Belgium). Collagen I staining was also performed on deparaffinized sections. After antigen retrieval (Dako Antigen Retrieval Solution; Dako, Heverlee, Belgium), endogenous peroxidase activity (0.3% hydrogen peroxide in methanol for 20 minutes) and aspecific binding was blocked using pre-immune goat serum (DakoCytomation A/S). Next, sections were incubated with rabbit anti-collagen I (1:500 in tris/NaCl blocking reagent; Abcam, Cambridge, UK) overnight. As secondary antibody, RaRat-B (DakoCytomation A/S) was used. The bound antibodies were visualized using signal amplification with the Perkin Elmer kit (Renaissance TSA Indirect, Waltham, MA, USA) with cyanine 3 as fluorophore. Afterward, the slides were mounted with Prolong Gold containing DAPI (Molecular Probes). 
Microscopic Analysis
Using a microscope (Leica Microsystems, Wetzlar, Germany) equipped with a digital camera (Axiocam MrC5; Zeiss), images were analyzed at a magnification of ×20 and a resolution of 2584 × 1936 pixels. Commercial software (KS300; Zeiss) was used for morphometric analyses, as previously described.4345 The area taken by immunopositive leukocytes and blood vessels was quantified by calculating respectively the CD45-positive and the FITC-dextran-positive CNV area in the samples. Deposition of collagen was determined by first defining the middle of each spot on H&E staining and secondly by measuring the Sirius Red-positive or collagen I-positive area on five serial sections in the middle of each laser spot. Sirius Red was analyzed under polarized light to distinguish mature from immature collagen fibers. For each staining, the average area of typically three laser spots per eye was used of 10 mice (n = 10), generating 10 data points (only one eye per animal was used) per group, based on literature.46 However, lesions were excluded (1) if there were signs of eye infection during the experiment, excluding the entire eye; (2) if there was hemorrhage in the eye, excluding all three lesions of one eye; (3) if the lesion had fused with another lesion, excluding those two lesions; or (4) if the lesion was an outlier, meaning that the CNV area exceeded 100,000 μm2 and was more than five times larger than the next largest lesion in the eye; or when the CNV area was “too small,” that is, it was less than one-fifth the area of the next smallest lesion in the eye. These exclusion criteria were based on the study performed by Poor et al.46 
To analyze for leaky vessels, confocal imaging was performed, with a Laser Scanning Microscope (LSM; Olympus, Melville, NY, USA). Laser scanning microscope–scanning software FluoView (FV10-ASW; Olympus) was used for digital acquisition and processing of the images. Vessel leakage was quantified by calculating the leaky area (Texas Red-conjugated Dextran; red area) as a proportion to the blood vessel area (FITC-lectin; green area). An average of 11 frames per final image/spot were used. 
Statistical Analysis
All statistical data were generated in GraphPad Prism 5.0 (GraphPad Software, Inc., San Diego, CA, USA). A one-way ANOVA with vehicle as comparator was performed on all in vivo immunomorphometric data with a Dunnet's post hoc analysis test, as previously reported.46 For the expression studies, a one-way ANOVA was also performed with the lasered, nontreated group as comparator. In vitro cell apoptosis and proliferation was analyzed by a one-way ANOVA. The migration assays were evaluated using mixed model analysis for repeated measures (two-way ANOVA). Data are represented as mean ± SEM, unless otherwise stated, and considered to be statistically significant at P ≤ 0.05. All experiments were analyzed blindly and randomized. 
Results
AMA0428 Inhibits Proliferation of ECs, but Not of Pericytes
To investigate the direct effect of AMA0428 on cell proliferation, HUVECs, HBMECs, and HBVPs were cultured. First, possible toxic effects of the ROCK inhibitor were checked using a caspase 3/7 apoptosis assay. Hereto, the cells were incubated in complete medium with AMA0428 or vehicle/DMSO at different concentrations. There was no significant increase in apoptosis after 48 hours when treated with AMA0428 from 0.5 μM up to 25 μM for EC and up to 10 μM for HBVPs, as compared with the cells treated with the equivalent of DMSO or control medium (data not shown). Thus, AMA0428 did not induce caspase 3/7 activity, suggesting that there are no obvious toxic effects at the used doses. 
To determine the effect of AMA0428 on proliferation of HBMECs and HUVECs, cells were grown in complete medium or in serum-free medium to which rhVEGF165 was added. In both conditions, AMA0428 (0.5, 1, 2.5, and 5 μM) induced a significant decrease in cell proliferation of HBMECs, as compared with vehicle/DMSO solutions in, respectively, complete/full medium or serum-free medium enriched with rhVEGF165 (P < 0.05 for all concentrations; Figs. 1A, 1B). Also, the growth of HUVECs was inhibited in both conditions after AMA0428 (0.5, 1, 2.5, and 5 μM) administration, as compared with the respective control solutions (P < 0.05 for all concentrations; Figs. 1C, 1D). The VEGF inhibitor, bevacizumab (Avastin; Roche), which was used as a positive control in a concentration of 2.5 mg/mL and 5 mg/mL,43,47 reduced EC growth comparable to AMA0428 (Figs. 1A–D). 
Figure 1
 
The AMA0428 reduces cell growth of HUVEC and HBMEC, without affecting HBVP proliferation in vitro. (A, B) Administration of AMA0428 (0.5, 1, 2.5, 5 μM) significantly inhibited cell proliferation of HBMECs (P < 0.05), similar to anti-VEGF (bevacizumab; 2.5–5.0 mg/mL) that was used as control in full medium by 13.0% ± 3.3%, 13.0% ± 2.9%, 19.0% ± 4.7%, and 39.0% ± 11%, respectively (A), and in serum-free medium after rhVEGF165 (100 ng/mL or 5.22 μM) challenge by 23.0% ± 1.0%, 26.0% ± 3.9%, 25.0% ± 5.8%, and 31.0% ± 4.8% (B). (C, D) AMA0428 significantly reduced proliferation of HUVECs at all concentrations (0.5, 1, 2.5, 5 μM; P < 0.05), which was comparable to bevacizumab in full medium by 29.0% ± 9.4%, 30.0% ± 9.1%, 34.0% ± 11.7%, and 44.0% ± 9.9%, respectively (C), and in serum-free medium after rhVEGF165 (100 ng/mL or 5.22 μM) challenge by 27.0% ± 2.1%, 28.0% ± 2.5%, 27.0% ± 2.4%, and 34.0% ± 3.1% (D). (E) Administration of AMA0428 did not induce any specific inhibitory effects on basal cell proliferation of HBVP in full medium (P = nonsignificant [NS]). (F) In starvation medium, AMA0428 did not stimulate proliferation of HBVP (P = NS). Recombinant human platelet-derived growth factor was used as a positive control, which gave a significant increase in proliferation (P < 0.05). Data represent mean ± SEM of three independent experiments (n = 3). Each experiment was performed in triplicate.
Figure 1
 
The AMA0428 reduces cell growth of HUVEC and HBMEC, without affecting HBVP proliferation in vitro. (A, B) Administration of AMA0428 (0.5, 1, 2.5, 5 μM) significantly inhibited cell proliferation of HBMECs (P < 0.05), similar to anti-VEGF (bevacizumab; 2.5–5.0 mg/mL) that was used as control in full medium by 13.0% ± 3.3%, 13.0% ± 2.9%, 19.0% ± 4.7%, and 39.0% ± 11%, respectively (A), and in serum-free medium after rhVEGF165 (100 ng/mL or 5.22 μM) challenge by 23.0% ± 1.0%, 26.0% ± 3.9%, 25.0% ± 5.8%, and 31.0% ± 4.8% (B). (C, D) AMA0428 significantly reduced proliferation of HUVECs at all concentrations (0.5, 1, 2.5, 5 μM; P < 0.05), which was comparable to bevacizumab in full medium by 29.0% ± 9.4%, 30.0% ± 9.1%, 34.0% ± 11.7%, and 44.0% ± 9.9%, respectively (C), and in serum-free medium after rhVEGF165 (100 ng/mL or 5.22 μM) challenge by 27.0% ± 2.1%, 28.0% ± 2.5%, 27.0% ± 2.4%, and 34.0% ± 3.1% (D). (E) Administration of AMA0428 did not induce any specific inhibitory effects on basal cell proliferation of HBVP in full medium (P = nonsignificant [NS]). (F) In starvation medium, AMA0428 did not stimulate proliferation of HBVP (P = NS). Recombinant human platelet-derived growth factor was used as a positive control, which gave a significant increase in proliferation (P < 0.05). Data represent mean ± SEM of three independent experiments (n = 3). Each experiment was performed in triplicate.
Next, the effect of the ROCK inhibitor, AMA0428, on pericyte proliferation was investigated using HBVPs. Analysis of the WST-1 proliferation assay showed no significant difference in number of HBVPs after treatment with AMA0428 (Figs. 1E, 1F) as compared with DMSO-treated cells (data not shown). As previously described,48 administration of rhVEGF165 resulted in a significantly decreased HBVP cell growth. On the contrary, rhPDGF significantly increased pericyte proliferation in starvation medium (Figs. 1E, 1F). 
Thus, AMA0428 reduces the proliferation of ECs, without affecting pericyte growth. 
AMA0428 Inhibits VEGF-Induced Migration of HBMECs, but Stimulates HBVP Migration
The effect of AMA0428 on cell migration was evaluated using an in vitro scratch assay.34 All cells were incubated in serum-free medium overnight. Immediately following scratch injury, HBMEC and HBVP were treated with AMA0428 vehicle (DMSO) and/or with rhVEGF165 (20 ng/mL) or PDGF (20 ng/mL). Sole administration of AMA0428 did not inhibit nor stimulate migration of HBMECs. On the other hand, VEGF administration led to near complete (90.0% ± 3.5%) wound closure in HBMECs at 8 hours after scratching. Furthermore, addition of AMA0428 to VEGF-stimulated HBMECs significantly blocked migration compared with vehicle (overall P < 0.05; Fig. 2A). In HBVPs, VEGF administration inhibited migration, whereas PDGF or AMA0428 treatment led to near complete wound closure after 8 hours, respectively to 85.0% ± 4.2% and to 85.0% ± 4.5%, as compared with DMSO (overall P < 0.05; Fig. 2C). The level of wound closure in the DMSO solution did not significantly differ from serum-free medium after 8 hours in both assays. These data were confirmed using the under-agarose migration assay.36 Analysis showed that, in comparison to vehicle (DMSO), treatment with AMA0428 significantly stimulated migration of HBVPs toward HUVECs (overall P < 0.05 for all concentrations; Fig. 3). To determine whether this effect was not solely induced by exposure of HBVPs to AMA0428, as HBVP migration was increased after AMA0428 administration in the scratch assay, the under-agarose migration assay was performed with and without HUVECs. These experiments revealed an additional stimulatory effect on migration by AMA0428 in the presence of HUVECs (data not shown). 
Figure 2
 
AMA0428 inhibits VEGF-induced HBMEC migration, but stimulates HBVP migration and recruitment in vitro. (A, C) Quantification of wound repair in a scratch assay. (A) Human brain microvascular endothelial cells were treated with DMSO (control), AMA0428 (0.5–5.0 μM), and/or RhVEGF165, (20 ng/mL). AMA0428 significantly inhibited VEGF-induced HBMEC migration, compared to VEGF-stimulated cells (overall P < 0.05). (B) Representative images of HBMEC in the scratch assay (2, 4, 6, 8 hours) after treatment with RhVEGF165 and AMA0428 0.5μM. (C) The HBVPs were treated with AMA0428 (0.5–5.0 μM), DMSO, and/or RhPDGF-BB (20–40 ng/mL). Migration of HBVPs was significantly stimulated by AMA0428 in both concentrations compared to DMSO/starvation medium (overall P < 0.05). Data are mean ± SEM of three independent experiments (n = 3). Each experiment was performed in triplicate. (C) Representative images of HBVP in the scratch assay (2, 4, 6, 8 hours) after treatment with AMA0428 0.5 μM.
Figure 2
 
AMA0428 inhibits VEGF-induced HBMEC migration, but stimulates HBVP migration and recruitment in vitro. (A, C) Quantification of wound repair in a scratch assay. (A) Human brain microvascular endothelial cells were treated with DMSO (control), AMA0428 (0.5–5.0 μM), and/or RhVEGF165, (20 ng/mL). AMA0428 significantly inhibited VEGF-induced HBMEC migration, compared to VEGF-stimulated cells (overall P < 0.05). (B) Representative images of HBMEC in the scratch assay (2, 4, 6, 8 hours) after treatment with RhVEGF165 and AMA0428 0.5μM. (C) The HBVPs were treated with AMA0428 (0.5–5.0 μM), DMSO, and/or RhPDGF-BB (20–40 ng/mL). Migration of HBVPs was significantly stimulated by AMA0428 in both concentrations compared to DMSO/starvation medium (overall P < 0.05). Data are mean ± SEM of three independent experiments (n = 3). Each experiment was performed in triplicate. (C) Representative images of HBVP in the scratch assay (2, 4, 6, 8 hours) after treatment with AMA0428 0.5 μM.
Figure 3
 
Migration of HBVPs toward HUVECs was assessed in an under-agarose migration coculture assay. (A) The HUVECs were treated with AMA0428 (1, 2.5, 5 μM) or control (DMSO). The AMA0428 significantly stimulated HBVP recruitment by HUVECs compared to DMSO/starvation medium (overall P < 0.05). Data are represented as mean ± SEM of three independent experiments (n = 3). Each experiment was performed in triplicate. (B) Representative images of the under-agarose migration assay 24, 48, and 72 hours after treatment with 1 μM of AMA0428.
Figure 3
 
Migration of HBVPs toward HUVECs was assessed in an under-agarose migration coculture assay. (A) The HUVECs were treated with AMA0428 (1, 2.5, 5 μM) or control (DMSO). The AMA0428 significantly stimulated HBVP recruitment by HUVECs compared to DMSO/starvation medium (overall P < 0.05). Data are represented as mean ± SEM of three independent experiments (n = 3). Each experiment was performed in triplicate. (B) Representative images of the under-agarose migration assay 24, 48, and 72 hours after treatment with 1 μM of AMA0428.
Thus, administration of AMA0428 was able to inhibit migration of VEGF-stimulated ECs and to stimulate pericyte recruitment. Overall, these data indicate that AMA0428 has the potential to reduce new blood vessel formation, on the one hand by inhibiting EC proliferation and migration and, on the other hand, by stimulating pericyte recruitment, thereby promoting maturation of blood vessels. 
AMA0428 Decreases EC Permeability In Vitro
Endothelial cell permeability was determined by measuring the passage of FITC-labeled dextran through a tight monolayer of HUVECs. Cells were treated with vehicle or AMA0428 with or without VEGF165 (2.08 μM = 40 ng/mL). Cell permeability was determined by evaluating fluorescence intensity every 15 minutes for 1 hour. A significant decrease in permeability of EC monolayers was observed after treatment with AMA0428 at 15 and 30 minutes after VEGF administration (0.1 μM; P < 0.05; Fig. 4). 
Figure 4
 
The AMA0428 significantly decreased permeability in a monolayer of VEGF-stimulated HUVECs (P < 0.05 versus VEGF165-treated cells). The permeability in HUVEC monolayers was analyzed by evaluating the passage of fluorescence-conjugated FITC-dextran from upper to lower compartments. Numbers represent mean ± SEM of three independent experiments (n = 3).
Figure 4
 
The AMA0428 significantly decreased permeability in a monolayer of VEGF-stimulated HUVECs (P < 0.05 versus VEGF165-treated cells). The permeability in HUVEC monolayers was analyzed by evaluating the passage of fluorescence-conjugated FITC-dextran from upper to lower compartments. Numbers represent mean ± SEM of three independent experiments (n = 3).
Upregulation of ROCK, VEGF, and PlGF in the CNV Mouse Model Is Inhibited by AMA0428
To investigate whether ROCK signaling contributes to CNV in vivo, we first analyzed levels of ROCK, VEGF, and PlGF in choroidal tissue of mice that received 30 laser spots. This procedure is known to induce the upregulation of growth factors such as VEGF49 and PlGF.7,50 Enzyme-linked immunosorbent assay for ROCK, VEGF, and PlGF showed a 2-fold increase in ROCK activity, a 1.5-fold increase in VEGF levels, and 2.3-fold increase in PlGF levels after laser treatment compared with naïve nonlasered mice, suggesting that these molecules are involved in the pathogenesis of CNV. The upregulation of these factors could be blocked significantly by the IVT administration of AMA0428 (P < 0.05; Fig. 5). Of note, the results of the vehicle group were comparable to the lasered, nontreated group (data not shown). 
Figure 5
 
Upregulation of ROCK, murine VEGF, and PlGF in choroid and retina 3 days after laser induction in C57Bl/6 mice measured by ELISA (P < 0.05). The results showed a 2-fold increase of ROCK activity, a 1.5-fold increase of VEGF concentration, and a 2.3-fold increase of PlGF levels after laser treatment as compared to baseline (gray horizontal line). Intravitreal administration of AMA0428 (1000 ng) decreased the upregulation of these factors, as compared to nontreated conditions by, respectively, 31%, 28%, and 43% (P < 0.05). Data are mean ± SEM of three independent experiments (n = 5 mice/group).
Figure 5
 
Upregulation of ROCK, murine VEGF, and PlGF in choroid and retina 3 days after laser induction in C57Bl/6 mice measured by ELISA (P < 0.05). The results showed a 2-fold increase of ROCK activity, a 1.5-fold increase of VEGF concentration, and a 2.3-fold increase of PlGF levels after laser treatment as compared to baseline (gray horizontal line). Intravitreal administration of AMA0428 (1000 ng) decreased the upregulation of these factors, as compared to nontreated conditions by, respectively, 31%, 28%, and 43% (P < 0.05). Data are mean ± SEM of three independent experiments (n = 5 mice/group).
AMA0428 Inhibited Inflammation, Angiogenesis, Vascular Leakage, and Fibrosis
The possible in vivo efficacy of IVT administered AMA0428 was determined in the laser-induced murine model for CNV. In a first approach, the potential side effect of AMA0428 (1000 ng) on retinal degeneration was investigated. Hereto, repeated injections did not reveal any gross changes in the thickness of the neuronal cell layers and in number of retinal ganglion cells in the AMA0428-treated eyes on H&E staining (Fig. 6). These data suggest no significant observable toxicity of AMA0428 to the neurosensory retina. 
Figure 6
 
The AMA0428 (1000 ng) seems safe by IVT injection in naïve mice. (A) Compared to vehicle, sham injection, or naïve mice, IVT administration of AMA0428 did not alter the total retinal thickness, the thickness of the outer and inner nuclear and plexiform layer, or photoreceptor layer as investigated on H&E staining. (B) The AMA0428 did not reduce the retinal ganglion cell number compared to vehicle, sham, or naïve mice. Data are represented as mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points.
Figure 6
 
The AMA0428 (1000 ng) seems safe by IVT injection in naïve mice. (A) Compared to vehicle, sham injection, or naïve mice, IVT administration of AMA0428 did not alter the total retinal thickness, the thickness of the outer and inner nuclear and plexiform layer, or photoreceptor layer as investigated on H&E staining. (B) The AMA0428 did not reduce the retinal ganglion cell number compared to vehicle, sham, or naïve mice. Data are represented as mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points.
Next, to compare the effect of this new ROCK inhibitor with a reference treatment, additional groups of mice were injected IVT with the murine anti–VEGF-R2 antibody, DC101, or with an irrelevant mouse antibody (1C8). To unravel which disease processes AMA0428 could affect, mice were killed at defined time points: at the peak of inflammation (day 5), angiogenesis (day 14), and fibrosis (day 30).7 
Inflammation is known to be involved in the earliest changes of CNV.6 A CD45 immunostaining was performed to study the effect of ROCK inhibition on inflammation. We compared the effect of repeated IVT injections of AMA0428 (1000 ng) versus DC101 on leukocyte infiltration in the laser-injured eyes. After 5 days, repeated administration of AMA0428 reduced the CD45-positive area in the lesions by 31.0% ± 11.8%, as compared to vehicle-treated eyes (n = 10 eyes; data not shown), whereas DC101 did not affect the inflammatory process. To investigate whether single administration would be sufficient, one IVT injection was given on day 0 or on day 3. Either AMA0428 1000 ng or 100 ng reduced inflammation, respectively, by 42.0% ± 13.1% and 38.0% ± 6.7% compared with vehicle-treated eyes (n = 10 eyes; P < 0.05; Fig. 7A) when supplied at the day of injury. Inflammation was also decreased after a single injection on day 3 with AMA0428 1000 ng or 100 ng, respectively, by 44.0% ± 14.4% and 38.0% ± 16.6% (n = 10 eyes; P < 0.05; Fig. 7B). These results showed that a single IVT injection on day 0 with AMA0428 was the most effective. Of note: A single injection on day 0 or day 3 with DC101 did not result in a significant reduction of leukocytes (n = 10 eyes; P > 0.05; Figs. 7A, 7B). 
Figure 7
 
Intravitreal administration of AMA0428 reduces inflammation in the CNV mouse model. (A) A significant inhibition of inflammation was seen with a single IVT injection on day 0 with AMA0428 (100 ng: P < 0.01; 1000 ng: P < 0.001). (B) A single IVT injection on day 3 with AMA0428 gave a significant reduction of leukocytes at day 5 (100 ng: P < 0.05; 1000 ng: P < 0.01). Early and late administration of DC101 had no effect on inflammation in the CNV model (P > 0.05). Data represent mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points. (C) Representative pictures of CD45 whole-mount staining (injection day 0) of eyes treated with vehicle or AMA0428 (1000 ng). Scale bars: 50 μm.
Figure 7
 
Intravitreal administration of AMA0428 reduces inflammation in the CNV mouse model. (A) A significant inhibition of inflammation was seen with a single IVT injection on day 0 with AMA0428 (100 ng: P < 0.01; 1000 ng: P < 0.001). (B) A single IVT injection on day 3 with AMA0428 gave a significant reduction of leukocytes at day 5 (100 ng: P < 0.05; 1000 ng: P < 0.01). Early and late administration of DC101 had no effect on inflammation in the CNV model (P > 0.05). Data represent mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points. (C) Representative pictures of CD45 whole-mount staining (injection day 0) of eyes treated with vehicle or AMA0428 (1000 ng). Scale bars: 50 μm.
To confirm the antiangiogenic capacities of AMA0428 in vivo, neovascularization was investigated at 14 days after laser treatment via FITC-dextran perfusion. At a dose of 1000 ng and 100 ng, repeated administration of AMA0428 reduced the total vessel area, respectively, by 79.0% ± 21.2% and 57.0% ± 21.0% (n = 10 eyes, P < 0.001, Figs. 8A, 8B), 14 days after laser treatment. Similarly, repeated injection of DC101 inhibited vessel area by 75.0% ± 17.3% (n = 10 eyes; P < 0.001), with no significant differences between the two compounds (P > 0.05, Fig. 8A). A single IVT injection on day 0, immediately after laser treatment, with AMA0428 1000 ng or 100 ng was less effective, because it only inhibited angiogenesis by, respectively, 55.0% ± 22.6% and by 34.0% ± 9.9% (n = 10 eyes; P < 0.05; data not shown). A single injection on day 3 gave a reduction of the active angiogenic process of 39.0% ± 15.9% for AMA0428 1000 ng (n = 10 eyes; P < 0.05), which was less potent compared with the injection on day 0 (data not shown). 
Figure 8
 
Repeated IVT injections of AMA0428 blocks angiogenesis and vessel leakage in the CNV mouse model. (A) Repeated (days 0, 4, 10) IVT administration of AMA0428 gave a dose-dependent reduction of angiogenesis at day 14 (1000 ng and 100 ng: P < 0.001), similar to DC101. (B) Representative pictures of FITC-dextran flatmounts; eyes were treated with either vehicle or AMA0428 (1000 ng). (C) Repeated administration of AMA0428 (1000 ng) was as effective as DC101 in reducing vessel leakage at day 14 after laser compared to their respective controls (P < 0.05). Data are mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points. (D) Representative pictures of flatmounts of eyes perfused with Texas Red–conjugated dextran 60 kDa (red) together with 0.05 mg FITC-labeled lectin (green); eyes were treated with either vehicle or AMA0428 (1000 ng).
Figure 8
 
Repeated IVT injections of AMA0428 blocks angiogenesis and vessel leakage in the CNV mouse model. (A) Repeated (days 0, 4, 10) IVT administration of AMA0428 gave a dose-dependent reduction of angiogenesis at day 14 (1000 ng and 100 ng: P < 0.001), similar to DC101. (B) Representative pictures of FITC-dextran flatmounts; eyes were treated with either vehicle or AMA0428 (1000 ng). (C) Repeated administration of AMA0428 (1000 ng) was as effective as DC101 in reducing vessel leakage at day 14 after laser compared to their respective controls (P < 0.05). Data are mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points. (D) Representative pictures of flatmounts of eyes perfused with Texas Red–conjugated dextran 60 kDa (red) together with 0.05 mg FITC-labeled lectin (green); eyes were treated with either vehicle or AMA0428 (1000 ng).
The effect of AMA0428 on the integrity of the BRB was also investigated at day 14 using a vessel leakage assay with Texas Red-conjugated Dextran and FITC-labeled lectin perfusion. The AMA0428 (1000 ng) blocked choroidal vessel leakage by 29.0% ± 13.9% (n = 10 eyes, P < 0.05; Figs. 8C, 8D) when administered IVT on day 0, day 4, and day 10. The effect was similar to that induced by repeated injections of DC101, namely 22.0% ± 12.3% (n = 10 eyes, P < 0.05; Fig. 8C). 
Choroidal neovascularization is also characterized by fibrosis. Therefore, Sirius Red staining and collagen I staining were used to check the effect of Rho kinase inhibition on the process of collagen deposition in vivo. Repeated (days 0, 4, 10, 20) IVT injections of AMA0428 1000 ng and 100 ng reduced collagen deposition in the lesions at day 30 by 34.0% ± 10.2% and 25.0% ± 5.9%, respectively (n = 10 eyes; P < 0.05, Sirius Red staining; data not shown), whereas DC101 did not reduce fibrosis 30 days after laser treatment (n = 10 eyes; P > 0.05). In addition, a collagen I staining was used to confirm the fibrotic reaction, as this type of collagen is most widely expressed in the CNV model.51 Analysis gave similar results compared with the Sirius Red staining, namely a reduction of 43.0% ± 9.9% and 39.0% ± 9.9% when treated, respectively, with 1000 ng and 100 ng AMA0428 (n = 10 eyes; P < 0.05; Figs. 9A, 9B). 
Figure 9
 
The AMA0428 inhibits fibrosis 30 days after laser treatment in the CNV mouse model. (A) Administration of repeated (days 0, 4, 10, 20) IVT injections with AMA0428 decreased fibrosis significantly compared to vehicle (P < 0.05), whereas DC101 did not affect fibrosis. Data are represented as mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points. (B) Representative pictures of Collagen I staining of eyes treated with vehicle or AMA0428 (1000 ng). Borders of the CNV lesion are marked by a dashed line, the immunopositive area inside the dashed line represents the new fibrotic area, which was used for analysis. Scale bars: 50 μm.
Figure 9
 
The AMA0428 inhibits fibrosis 30 days after laser treatment in the CNV mouse model. (A) Administration of repeated (days 0, 4, 10, 20) IVT injections with AMA0428 decreased fibrosis significantly compared to vehicle (P < 0.05), whereas DC101 did not affect fibrosis. Data are represented as mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points. (B) Representative pictures of Collagen I staining of eyes treated with vehicle or AMA0428 (1000 ng). Borders of the CNV lesion are marked by a dashed line, the immunopositive area inside the dashed line represents the new fibrotic area, which was used for analysis. Scale bars: 50 μm.
Overall, these data suggest that AMA0428 exhibits anti-inflammatory, antiangiogenic, and antifibrotic effects in the laser-induced CNV mouse model. 
Partial Substitution Effect of Anti–VEGF-R2 by AMA0428
Because complications of anti-VEGF treatment are mostly dose-dependent,13,5254 the possibility to partially substitute DC101 by AMA0428 was explored. Therefore, first a dose-response experiment was set up for both AMA0428 and DC101 (Figs. 10A, 10B). Based on these results, a combination of the most effective dose of AMA0428 and a suboptimal dose of anti–VEGF-R2 (25%) was given. This low dose of anti–VEGF-R2 was chosen to leave a larger window for the substitution effect. The combination therapy was given IVT on day 0, day 4, and day 10, and mice were killed on day 14. Neovascularization was reduced by 83.0% ± 11.4% with the combination therapy (n = 10 eyes; P < 0.001), that is, comparable to the highest dose of anti–VEGF-R2 therapy alone (75.0% ± 17.3% reduction; n = 10 eyes; P < 0.001; Fig. 10C). 
Figure 10
 
Intravitreal administration of AMA0428 allows reducing the dose of anti–VEGF-R2 (DC101) by 4-fold. (A) The AMA0428 gave a dose-dependent reduction in CNV (P < 0.001) (B) Dose-dependent inhibition of CNV by DC101 (P < 0.001). (C) Partial replacement of anti–VEGF-R2 by AMA0428 yielded a similar anti-CNV effect as the maximal dose of anti–VEGF-R2 therapy alone. Data represent mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points.
Figure 10
 
Intravitreal administration of AMA0428 allows reducing the dose of anti–VEGF-R2 (DC101) by 4-fold. (A) The AMA0428 gave a dose-dependent reduction in CNV (P < 0.001) (B) Dose-dependent inhibition of CNV by DC101 (P < 0.001). (C) Partial replacement of anti–VEGF-R2 by AMA0428 yielded a similar anti-CNV effect as the maximal dose of anti–VEGF-R2 therapy alone. Data represent mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points.
Thus, by adding AMA0428, the dose of DC101 can be reduced by 4-fold, while maintaining the efficacy of the full dose of anti-VEGF therapy. This opens the possibility to reduce the adverse effects sporadically linked to vascular and neuronal regression caused by VEGF inhibition.12,13 
Discussion
The effect of a novel ROCK inhibitor (AMA0428) on inflammation, angiogenesis, vascular leakage, and fibrosis, processes known to be involved in neovascular AMD, was investigated. Our results demonstrate that AMA0428 was effective in reducing these different processes in vitro and in the in vivo laser-induced CNV mouse model. Moreover, we showed that this ROCK inhibitor had a more profound effect compared with VEGF-R2 inhibition (DC101). Our previous findings have consistently demonstrated that DC101 is as effective as bevacizumab, the reference treatment in clinical practice, in mouse models of CNV, oxygen-induced retinopathy, and glaucoma filtration surgery, thus confirming its suitability for translational ophthalmological research.55 
The present study indeed indicates that administration of AMA0428 might reduce the formation of new blood vessels by inhibiting EC proliferation and VEGF-mediated migration. These results are comparable to published data of Y-27632, another reference ROCK inhibitor.56,57 Because it is known that VEGF-induced changes in the endothelial F-actin cytoskeleton require activation of Rho kinase,58 disruption of RhoA/ROCK signaling might inhibit VEGF-mediated alterations to cytoskeletal morphology.18,59,60 Aepfelbacher et al.61 described some evidence for a role of ROCK in EC migration, based on experiments performed in the presence of an endothelial growth supplement. 
Moreover, AMA0428 might also improve vessel maturation, by stimulating pericyte recruitment, as shown by the in vitro under-agarose migration assay. Previous studies showed that exposure of human vascular pericytes to different ROCK inhibitors led to a migratory phenotype in various assays.18,56 A previous study also demonstrated that activated polymorphonuclear neutrophils induced pericyte relaxation in the cremaster muscles of mice by inhibiting the RhoA-ROCK pathway.18 Overall, it is known that ROCK inhibitors decrease myosin light chain phosphorylation in pericytes and induce loss of stress fibers and focal adhesions,18,62,63 thereby facilitating migration and thus recruitment of pericytes. 
To obtain our results in vivo, we used the standard laser-induced mouse model for CNV, as this model is widely used to investigate wet AMD.3739 As anti-VEGF therapy is an important treatment modality in this pathology, we directly compared the in vivo effect of AMA0428 to the well-established VEGF-R2 antibody, DC101. We demonstrated for the first time that ROCK inhibition is as effective as DC101 in reducing neovascularization, with an additional inhibitory effect on inflammation and fibrosis in the CNV mouse model. Besides a clear decrease in ROCK activity, administration of AMA0428 also resulted in a reduced VEGF and PlGF upregulation in CNV lesions. Although a direct effect of ROCK inhibition on VEGF or PlGF levels cannot be excluded, it seems unlikely, as VEGF and PlGF are upstream of ROCK; however, the interconnection between the pathways is still unclear and needs to be unraveled further. Therefore, we rather hypothesize that the reduced expression of both growth factors can be explained by an indirect effect of the working mechanism of the ROCK inhibitor. Indeed, the immediate action of AMA0428 on neovascularization and inflammation can lead to a diminished production of additional growth factors. 
It is described that anti-VEGF therapy reduces the inflammatory response in the in vivo CNV model, through a VEGF-R1 effect.64 However, when both VEGF-R1 and 2 are blocked, an accumulation of inflammatory cells is seen at the CNV surface.50 In the clinical setting, anti-VEGF ligands affecting all receptors may thus induce inflammation in patients with neovascular AMD, in part due to an upregulation of different proinflammatory growth factors, such as PlGF.7,65 It is indeed known that PlGF is increased in the CNV mouse model,7,50 and after treatment with bevacizumab and different receptor tyrosin kinase inhibitors in colorectal, renal, and glioblastoma patients.6668 In the present study, we showed that anti–VEGF-R2 therapy did not affect the inflammatory process in the CNV model, whereas AMA0428 was able to reduce inflammation after laser. The PlGF pathway is known to interact with the ROCK pathway, which had modulating effects on cell motility.69,70 Therefore, the anti-inflammatory effect of AMA0428 can partially be explained by the inhibition of PlGF upregulation in the CNV model. The Rho/ROCK pathway is also a key mediator in several other inflammatory processes involved in retinopathies, such as TNF-α signaling,71 anchoring of leukocytes72,73 and transcription of inflammatory cytokines (e.g., IL-6, IL-8) through nuclear factor–κβ activation.74,75 Therefore, ROCK inhibition might be a new way to address the underlying inflammatory pathophysiology of several retinopathies. 
Fluid accumulation in the posterior retina, known as macular edema, is also involved in wet AMD and is the major cause of severe visual loss in neovascular AMD patients. This could result from a breakdown of the existing BRB at the level of the Bruch's membrane or from the “leaky nature” of the proliferating choroidal vessels. Anti-VEGF treatment can exert an effect on macular exudation, but might have only a limited and short-acting impact. Importantly, we showed that AMA0428 reduced vessel leakage. Moreover, IOP elevation has been described after IVT anti-VEGF administration, resulting in neuronal retinal damage.76 An additional likely benefit is that ROCK inhibitors, such as AMA0076, Y-27632, Y-39983, and others are known to have IOP-lowering capacities.77,78 We indeed confirmed IOP reduction with AMA0428 in mice and rabbits after topical installation (data not shown). 
Although several approaches have been extensively investigated to inhibit promoters of angiogenesis in AMD,79 efforts to minimize fibrosis have met with limited success. Local destruction of neuronal cells in the retina, due to fibrosis, leads to reduction in function and, thus, in vision. We showed that IVT administration of AMA0428 was effective in reducing fibrosis, whereas anti–VEGF-R2 had no clear effect on the fibrotic response in the CNV model. From the literature it is known that the Rho/ROCK pathway is a downstream signaling mechanism of major ocular degenerative and fibrotic disease drivers, such as TGF-β,80,81 suggesting that it represents a new possible therapeutic target for fibrotic-related diseases. Inhibiting the initial inflammatory process and the subsequent angiogenesis might also indirectly decrease fibrosis, the final stage of CNV. Cao et al.82 already demonstrated that VEGF Trap, an inhibitor of VEGF and PlGF, was able to prevent fibrosis associated with CNV,82 due to its inhibition of leukocyte infiltration and angiogenesis. Acting on the inflammatory process to block collagen deposition is of utmost importance, as proinflammatory factors contribute to the development of experimental CNV and are present in human CNV fibrotic membranes.65 
In a subpopulation of patients,83 the currently used VEGF inhibitors are known to cause adverse effects linked to vascular and neuronal regression in many organs, including the eye.13,5254,83 These complications are related to the dose of anti-VEGF treatment and, thus, lowering this dose would be beneficial for patient outcome. Therefore, we explored whether we could substitute part of the anti–VEGF-R2 dose with AMA0428, while maintaining the efficacy of the full-dose anti–VEGF-R2. Importantly, our results indicated that the combination therapy of AMA0428 and anti–VEGF-R2 was effective in reducing neovascularization; however, the overall inhibitory effect of AMA0428 and the combination therapy was similar. Therefore, the combination of a lower dose of AMA0428 that does not give maximal inhibition, and various doses of DC101 can be considered in future studies to determine an additive effect between two compounds. Nevertheless, our data clearly reveal that ROCK inhibition might be a new approach as alternative or additive to anti-VEGF in the treatment of neovascular AMD patients. 
The improvement of effective therapy for AMD will become even more significant as the population ages and as the prevalence of AMD and vision loss becomes more and more widespread. Inhibition of ROCK might improve current treatment options and thus help to fight the expected exponential rise in number of legally blind patients. The AMA0428 is an inhibitor of Rho kinases, which distinguishes itself from other agents by its efficacy profile and pleiotropic working mechanism. 
Conclusions
In summary, targeting ROCK with AMA0428, a novel potent Rho kinase inhibitor, efficiently treated CNV in the laser-induced mouse model. Compared with anti-VEGF therapy, AMA0428 not only reduced neoangiogenesis but also blocked inflammation and fibrosis in the CNV lesions. These findings make ROCK inhibitors promising new candidates with potential additional therapeutic benefit in the treatment of neovascular AMD. 
Acknowledgments
The authors thank Sofie Beckers, Ann Verbeek, and Martine Leijssen for their technical support. The AMA0428 was kindly provided by Amakem Therapeutics; DC101 and 1C8 were kindly provided by ThromboGenics NV. 
Disclosure: K. Hollanders, None; T. Van Bergen, None; N. Kindt, Amakem Therapeutics (E); K. Castermans, Amakem Therapeutics (E); D. Leysen, P; E. Vandewalle, None; L. Moons, None; I. Stalmans, Amakem Therapeutics (F, C) 
References
Bressler NM. Age-related macular degeneration is the leading cause of blindness. JAMA. 2004; 291: 1900–1901.
Jager RD, Mieler WF, Miller JW. Age-related macular degeneration. N Engl J Med. 2008; 358: 2606–2617.
Dixon JA, Oliver SC, Olson JL, Mandava N. VEGF Trap-Eye for the treatment of neovascular age-related macular degeneration. Expert Opin Investig Drugs. 2009; 18: 1573–1580.
Bressler NM. Early detection and treatment of neovascular age-related macular degeneration. J Am Board Fam Pract. 2002; 15: 142–152.
Lassota N. Clinical and histological aspects of CNV formation: studies in an animal model. Acta Ophthalmol. 2008; 86: 1–24.
Schlingemann RO. Role of growth factors and the wound healing response in age-related macular degeneration. Graefes Arch Clin Exp Ophthalmol. 2004; 242: 91–101.
Van de Veire S, Stalmans I, Heindryckx F et al. Further pharmacological and genetic evidence for the efficacy of PlGF inhibition in cancer and eye disease. Cell. 2009; 141: 178–190.
de Oliveira Dias JR, Rodrigues EB, Maia M, Magalhães OJr Penha FM, Farah ME. Cytokines in neovascular age-related macular degeneration: fundamentals of targeted combination therapy. Br J Ophthalmol. 2011; 95: 1631–1637.
Rosenfeld PJ, Brown DM, Heier JS et al. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med. 2006; 355: 1419–1431.
Brown DM, Kaiser PK, Michels M, et al. Ranibizumab versus verteporfin for neovascular age-related macular degeneration. N Engl J Med. 2006; 355: 1432–1444.
Kim I, Ryan AM, Rohan R et al. Constitutive expression of VEGF, VEGFR-1, and VEGFR-2 in normal eyes. Invest Ophthalmol Vis Sci. 1999; 40: 2115–2121.
Tobin KA. Macugen treatment for wet age-related macular degeneration. Insight. 2006; 31: 11–14.
Fraunfelder FW. Pegaptanib for wet macular degeneration. Drugs Today (Barc). 2005; 41: 703–709.
Semeraro F, Morescalchi F, Parmeggiani F, Arcidiacono B, Costagliola C. Systemic adverse drug reactions secondary to anti-VEGF intravitreal injection in patients with neovascular age-related macular degeneration. Curr Vasc Pharmacol. 2011; 9: 629–646.
Hahmann C, Schroeter T. Rho-kinase inhibitors as therapeutics: from pan inhibition to isoform selectivity. Cell Mol Life Sci. 2010; 67: 171–177.
Jaffe AB, Hall A. Rho GTPases: biochemistry and biology. Annu Rev Cell Dev Biol. 2005; 21: 247–269.
Danen EH, van Rheenen J, Franken W et al. Integrins control motile strategy through a Rho-cofilin pathway. J Cell Biol. 2005; 169: 515–526.
Wang S, Cao C, Chen Z, et al. Pericytes regulate vascular basement membrane remodeling and govern neutrophil extravasation during inflammation. PLoS One. 2012; 7: e45499.
Bondjers C, Kalen M, Hellstrom M et al. Transcription profiling of platelet-derived growth factor-B-deficient mouse embryos identifies RGS5 as a novel marker for pericytes and vascular smooth muscle cells. Am J Pathol. 2003; 162: 721–729.
Montanez E, Wickstrom SA, Altstatter J, Chu H, Fassler R. Alpha-parvin controls vascular mural cell recruitment to vessel wall by regulating RhoA/ROCK signalling. EMBO J. 2009; 28: 3132–3144.
Ren XD, Kiosses WB, Schwartz MA. Regulation of the small GTP-binding protein Rho by cell adhesion and the cytoskeleton. EMBO J. 1999; 18: 578–585.
Honjo M, Tanihara H, Kameda T, Kawaji T, Yoshimura N, Araie M. Potential role of Rho-associated protein kinase inhibitor Y-27632 in glaucoma filtration surgery. Invest Ophthalmol Vis Sci. 2007; 48: 5549–5557.
Liu M, Gu M, Wu Y et al. Therapeutic effect of Y-27632 on chronic allograft nephropathy in rats. J Surg Res. 2009; 157: e117–e127.
Arita R, Hata Y, Ishibashi T. ROCK as a therapeutic target of diabetic retinopathy. J Ophthalmol. 2010; 2010: 175163.
Doe C, Bentley R, Behm DJ et al. Novel Rho kinase inhibitors with anti-inflammatory and vasodilatory activities. J Pharmacol Exp Ther. 2007; 320: 89–98.
Mong PY, Wang Q. Activation of Rho kinase isoforms in lung endothelial cells during inflammation. J Immunol. 2009; 182: 2385–2394.
Coleman HR, Chan CC, Ferris FLIII Chew EY. Age-related macular degeneration. Lancet. 2008; 372: 1835–1845.
Grossniklaus HE, Green WR. Choroidal neovascularization. Am J Ophthalmol. 2004; 137: 496–503.
Wang Y, Wang VM, Chan CC. The role of anti-inflammatory agents in age-related macular degeneration (AMD) treatment. Eye. 2011; 25: 127–139.
Anderson DH, Radeke MJ, Gallo NB et al. The pivotal role of the complement system in aging and age-related macular degeneration: hypothesis re-visited. Prog Retin Eye Res. 2010; 29: 95–112.
Hageman GS, Luthert PJ Victor Chong NH, Johnson LV, Anderson DH, Mullins RF. An integrated hypothesis that considers drusen as biomarkers of immune-mediated processes at the RPE-Bruch's membrane interface in aging and age-related macular degeneration. Prog Retin Eye Res. 2001; 20: 705–732.
Patel M, Chan CC. Immunopathological aspects of age-related macular degeneration. Semin Immunopathol. 2008; 30: 97–110.
Boland S, Defert O, Alen J et al. 3-[2-(Aminomethyl)-5-[(pyridin-4-yl)carbamoyl]phenyl] benzoates as soft ROCK inhibitors. Bioorg Med Chem Lett. 2013; 23: 6442–6446.
Bryan BA, Dennstedt E, Mitchell DC, et al. RhoA/ROCK signaling is essential for multiple aspects of VEGF-mediated angiogenesis. FASEB J. 2010; 24: 3186–3195.
Castermans K, Tabruyn SP, Zeng R et al. Angiostatic activity of the antitumor cytokine interleukin-21. Blood. 2008; 112: 4940–4947.
Zheng Y, Bando H, Ikuno Y, et al. Involvement of rho-kinase pathway in contractile activity of rabbit RPE cells in vivo and in vitro. Invest Ophthalmol Vis Sci. 2004; 45: 668–674.
Ryan SJ. The development of an experimental model of subretinal neovascularization in disciform macular degeneration. Trans Am Ophthalmol Soc. 1979; 77: 707–745.
Dobi ET, Puliafito CA, Destro M. A new model of experimental choroidal neovascularization in the rat. Arch Ophthalmol. 1989; 107: 264–269.
Tobe T, Ortega S, Luna JD et al. Targeted disruption of the FGF2 gene does not prevent choroidal neovascularization in a murine model. Am J Pathol. 1998; 153: 1641–1646.
Bocci G, Man S, Green SK, et al. Increased plasma vascular endothelial growth factor (VEGF) as a surrogate marker for optimal therapeutic dosing of VEGF receptor-2 monoclonal antibodies. Cancer Res. 2004; 64: 6616–6625.
Bocci G, Danesi R, Marangoni G et al. Antiangiogenic versus cytotoxic therapeutic approaches to human pancreas cancer: an experimental study with a vascular endothelial growth factor receptor-2 tyrosine kinase inhibitor and gemcitabine. Eur J Pharmacol. 2004; 498: 9–18.
Lambert V, Munaut C, Jost M, et al. Matrix metalloproteinase-9 contributes to choroidal neovascularization. Am J Pathol. 2002; 161: 1247–1253.
Li Z, Van Bergen T, Van de Veire S et al. Inhibition of vascular endothelial growth factor reduces scar formation after glaucoma filtration surgery. Invest Ophthalmol Vis Sci. 2009; 50: 5217–5225.
Van Bergen T, Vandewalle E, Van de Veire S, et al. The role of different VEGF isoforms in scar formation after glaucoma filtration surgery. Exp Eye Res. 2011; 93: 689–699.
Van Bergen T, Marshall D, Van de Veire S et al. The role of LOX and LOXL2 in scar formation after glaucoma surgery. Invest Ophthalmol Vis Sci. 2013; 54: 5788–5796.
Poor SH, Qiu Y, Fassbender ES, et al. Reliability of the mouse model of choroidal neovascularization induced by laser photocoagulation. Invest Ophthalmol Vis Sci. 2014; 55: 6525–6534.
Van Bergen T, Jonckx B, Hollanders K et al. Inhibition of placental growth factor improves surgical outcome of glaucoma surgery. J Cell Mol Med. 2013; 17: 1632–1643.
Greenberg JI, Shields DJ, Barillas SG, et al. A role for VEGF as a negative regulator of pericyte function and vessel maturation. Nature. 2008; 456: 809–813.
Krause TA, Alex AF, Engel DR, Kurts C, Eter N. VEGF-production by CCR2-dependent macrophages contributes to laser-induced choroidal neovascularization. PLoS One. 2014; 9: e94313.
Huang H, Shen J, Vinores SA. Blockade of VEGFR1 and 2 suppresses pathological angiogenesis and vascular leakage in the eye. PLoS One. 2011; 6: e21411.
Caballero S, Yang R, Grant MB, Chaqour B. Selective blockade of cytoskeletal actin remodeling reduces experimental choroidal neovascularization. Invest Ophthalmol Vis Sci. 2011; 52: 2490–2496.
Iriyama A, Chen YN, Tamaki Y, Yanagi Y. Effect of anti-VEGF antibody on retinal ganglion cells in rats. Br J Ophthalmol. 2007; 91: 1230–1233.
Klein KS, Walsh MK, Hassan TS et al. Endophthalmitis after anti-VEGF injections. Ophthalmology. 2009; 116: e1221.
Nishijima K, Ng YS, Zhong L, et al. Vascular endothelial growth factor-A is a survival factor for retinal neurons and a critical neuroprotectant during the adaptive response to ischemic injury. Am J Pathol. 2007; 171: 53–67.
Hollanders K, Van Bergen T, Van de Velde S et al. Bevacizumab revisited: its use in different mouse models of ocular pathologies. Curr Eye Res. 2014; 1–11.
Bryan BA, Dennstedt E, Mitchell DC, et al. RhoA/ROCK signaling is essential for multiple aspects of VEGF-mediated angiogenesis. Faseb J. 2010; 24: 3186–3195.
van Nieuw Amerongen GP, Koolwijk P, Versteilen A, van Hinsbergh VW. Involvement of RhoA/Rho kinase signaling in VEGF-induced endothelial cell migration and angiogenesis in vitro. Arterioscler Thromb Vasc Biol. 2003; 23: 211–217.
van der Meel R, Symons MH, Kudernatsch R et al. The VEGF/Rho GTPase signalling pathway: a promising target for anti-angiogenic/anti-invasion therapy. Drug Discov Today. 2011; 16: 219–228.
Hoang MV, Whelan MC, Senger DR. Rho activity critically and selectively regulates endothelial cell organization during angiogenesis. Proc Natl Acad Sci U S A. 2004; 101: 1874–1879.
Zeng H, Zhao D, Mukhopadhyay D. KDR stimulates endothelial cell migration through heterotrimeric G protein Gq/11-mediated activation of a small GTPase RhoA. J Biol Chem. 2002; 277: 46791–46798.
Aepfelbacher M, Essler M, Huber E, Sugai M, Weber PC. Bacterial toxins block endothelial wound repair. Evidence that Rho GTPases control cytoskeletal rearrangements in migrating endothelial cells. Arterioscler Thromb Vasc Biol. 1997; 17: 1623–1629.
Kimura K, Ito M, Amano M et al. Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase). Science. 1996; 273: 245–248.
Feng J, Ito M, Ichikawa K, et al. Inhibitory phosphorylation site for Rho-associated kinase on smooth muscle myosin phosphatase. J Biol Chem. 1999; 274: 37385–37390.
Huang H, Parlier R, Shen JK, Lutty GA, Vinores SA. VEGF receptor blockade markedly reduces retinal microglia/macrophage infiltration into laser-induced CNV. PLoS One. 2013; 8: e71808.
Rakic JM, Lambert V, Devy L et al. Placental growth factor, a member of the VEGF family, contributes to the development of choroidal neovascularization. Invest Ophthalmol Vis Sci. 2003; 44: 3186–3193.
Batchelor TT, Sorensen AG, di Tomaso E, et al. AZD2171, a pan-VEGF receptor tyrosine kinase inhibitor, normalizes tumor vasculature and alleviates edema in glioblastoma patients. Cancer Cell. 2007; 11: 83–95.
Motzer RJ, Michaelson MD, Redman BG et al. Activity of SU11248, a multitargeted inhibitor of vascular endothelial growth factor receptor and platelet-derived growth factor receptor, in patients with metastatic renal cell carcinoma. J Clin Oncol. 2006; 24: 16–24.
Willett CG, Boucher Y, Duda DG, et al. Surrogate markers for antiangiogenic therapy and dose-limiting toxicities for bevacizumab with radiation and chemotherapy: continued experience of a phase I trial in rectal cancer patients. J Clin Oncol. 2005; 23: 8136–8139.
Fazal MA, Saksena J, Haddad FS. Temporary coracoclavicular screw fixation for displaced distal clavicle fractures. J Orthop Surg (Hong Kong). 2007; 15: 9–11.
Zhang B, Yan L, Fazal I et al. Slow light on Gbit/s differential-phase-shift-keying signals. Opt Express. 2007; 15: 1878–1883.
Arita R, Nakao S, Kita T, et al. A key role for ROCK in TNF-alpha-mediated diabetic microvascular damage. Invest Ophthalmol Vis Sci. 2013; 54: 2373–2383.
Lee H, Lin CI, Liao JJ et al. Lysophospholipids increase ICAM-1 expression in HUVEC through a Gi- and NF-kappaB-dependent mechanism. Am J Physiol Cell Physiol. 2004; 287: C1657–C1666.
Wojciak-Stothard B, Williams L, Ridley AJ. Monocyte adhesion and spreading on human endothelial cells is dependent on Rho-regulated receptor clustering. J Cell Biol. 1999; 145: 1293–1307.
Fazal F, Bijli KM, Minhajuddin M, Rein T, Finkelstein JN, Rahman A. Essential role of cofilin-1 in regulating thrombin-induced RelA/p65 nuclear translocation and intercellular adhesion molecule 1 (ICAM-1) expression in endothelial cells. J Biol Chem. 2009; 284: 21047–21056.
Anwar KN, Fazal F, Malik AB, Rahman A. RhoA/Rho-associated kinase pathway selectively regulates thrombin-induced intercellular adhesion molecule-1 expression in endothelial cells via activation of I kappa B kinase beta and phosphorylation of RelA/p65. J Immunol. 2004; 173: 6965–6972.
Adelman RA, Zheng Q, Mayer HR. Persistent ocular hypertension following intravitreal bevacizumab and ranibizumab injections. J Ocul Pharmacol Ther. 2010; 26: 105–110.
Van de Velde S, Van Bergen T, Sijnave D et al. AMA0076, a novel, locally acting Rho kinase inhibitor, potently lowers intraocular pressure in New Zealand white rabbits with minimal hyperemia. Invest Ophthalmol Vis Sci. 2014; 55: 1006–1016.
Wang SK, Chang RT. An emerging treatment option for glaucoma: Rho kinase inhibitors. Clin Ophthalmol. 2014; 8: 883–890.
Michels S, Schmidt-Erfurth U, Rosenfeld PJ. Promising new treatments for neovascular age-related macular degeneration. Expert Opin Investig Drugs. 2006; 15: 779–793.
Tsapara A, Luthert P, Greenwood J, Hill CS, Matter K, Balda MS. The RhoA activator GEF-H1/Lfc is a transforming growth factor-beta target gene and effector that regulates alpha-smooth muscle actin expression and cell migration. Mol Biol Cell. 2010; 21: 860–870.
Nishikimi T, Matsuoka H. Molecular mechanisms and therapeutic strategies of chronic renal injury: renoprotective effect of rho-kinase inhibitor in hypertensive glomerulosclerosis. J Pharmacol Sci. 2006; 100: 22–28.
Cao J, Zhao L, Li Y et al. A subretinal matrigel rat choroidal neovascularization (CNV) model and inhibition of CNV and associated inflammation and fibrosis by VEGF trap. Invest Ophthalmol Vis Sci. 2010; 51: 6009–6017.
Falavarjani KG, Nguyen QD. Adverse events and complications associated with intravitreal injection of anti-VEGF agents: a review of literature. Eye. 2013; 27: 787–794.
Figure 1
 
The AMA0428 reduces cell growth of HUVEC and HBMEC, without affecting HBVP proliferation in vitro. (A, B) Administration of AMA0428 (0.5, 1, 2.5, 5 μM) significantly inhibited cell proliferation of HBMECs (P < 0.05), similar to anti-VEGF (bevacizumab; 2.5–5.0 mg/mL) that was used as control in full medium by 13.0% ± 3.3%, 13.0% ± 2.9%, 19.0% ± 4.7%, and 39.0% ± 11%, respectively (A), and in serum-free medium after rhVEGF165 (100 ng/mL or 5.22 μM) challenge by 23.0% ± 1.0%, 26.0% ± 3.9%, 25.0% ± 5.8%, and 31.0% ± 4.8% (B). (C, D) AMA0428 significantly reduced proliferation of HUVECs at all concentrations (0.5, 1, 2.5, 5 μM; P < 0.05), which was comparable to bevacizumab in full medium by 29.0% ± 9.4%, 30.0% ± 9.1%, 34.0% ± 11.7%, and 44.0% ± 9.9%, respectively (C), and in serum-free medium after rhVEGF165 (100 ng/mL or 5.22 μM) challenge by 27.0% ± 2.1%, 28.0% ± 2.5%, 27.0% ± 2.4%, and 34.0% ± 3.1% (D). (E) Administration of AMA0428 did not induce any specific inhibitory effects on basal cell proliferation of HBVP in full medium (P = nonsignificant [NS]). (F) In starvation medium, AMA0428 did not stimulate proliferation of HBVP (P = NS). Recombinant human platelet-derived growth factor was used as a positive control, which gave a significant increase in proliferation (P < 0.05). Data represent mean ± SEM of three independent experiments (n = 3). Each experiment was performed in triplicate.
Figure 1
 
The AMA0428 reduces cell growth of HUVEC and HBMEC, without affecting HBVP proliferation in vitro. (A, B) Administration of AMA0428 (0.5, 1, 2.5, 5 μM) significantly inhibited cell proliferation of HBMECs (P < 0.05), similar to anti-VEGF (bevacizumab; 2.5–5.0 mg/mL) that was used as control in full medium by 13.0% ± 3.3%, 13.0% ± 2.9%, 19.0% ± 4.7%, and 39.0% ± 11%, respectively (A), and in serum-free medium after rhVEGF165 (100 ng/mL or 5.22 μM) challenge by 23.0% ± 1.0%, 26.0% ± 3.9%, 25.0% ± 5.8%, and 31.0% ± 4.8% (B). (C, D) AMA0428 significantly reduced proliferation of HUVECs at all concentrations (0.5, 1, 2.5, 5 μM; P < 0.05), which was comparable to bevacizumab in full medium by 29.0% ± 9.4%, 30.0% ± 9.1%, 34.0% ± 11.7%, and 44.0% ± 9.9%, respectively (C), and in serum-free medium after rhVEGF165 (100 ng/mL or 5.22 μM) challenge by 27.0% ± 2.1%, 28.0% ± 2.5%, 27.0% ± 2.4%, and 34.0% ± 3.1% (D). (E) Administration of AMA0428 did not induce any specific inhibitory effects on basal cell proliferation of HBVP in full medium (P = nonsignificant [NS]). (F) In starvation medium, AMA0428 did not stimulate proliferation of HBVP (P = NS). Recombinant human platelet-derived growth factor was used as a positive control, which gave a significant increase in proliferation (P < 0.05). Data represent mean ± SEM of three independent experiments (n = 3). Each experiment was performed in triplicate.
Figure 2
 
AMA0428 inhibits VEGF-induced HBMEC migration, but stimulates HBVP migration and recruitment in vitro. (A, C) Quantification of wound repair in a scratch assay. (A) Human brain microvascular endothelial cells were treated with DMSO (control), AMA0428 (0.5–5.0 μM), and/or RhVEGF165, (20 ng/mL). AMA0428 significantly inhibited VEGF-induced HBMEC migration, compared to VEGF-stimulated cells (overall P < 0.05). (B) Representative images of HBMEC in the scratch assay (2, 4, 6, 8 hours) after treatment with RhVEGF165 and AMA0428 0.5μM. (C) The HBVPs were treated with AMA0428 (0.5–5.0 μM), DMSO, and/or RhPDGF-BB (20–40 ng/mL). Migration of HBVPs was significantly stimulated by AMA0428 in both concentrations compared to DMSO/starvation medium (overall P < 0.05). Data are mean ± SEM of three independent experiments (n = 3). Each experiment was performed in triplicate. (C) Representative images of HBVP in the scratch assay (2, 4, 6, 8 hours) after treatment with AMA0428 0.5 μM.
Figure 2
 
AMA0428 inhibits VEGF-induced HBMEC migration, but stimulates HBVP migration and recruitment in vitro. (A, C) Quantification of wound repair in a scratch assay. (A) Human brain microvascular endothelial cells were treated with DMSO (control), AMA0428 (0.5–5.0 μM), and/or RhVEGF165, (20 ng/mL). AMA0428 significantly inhibited VEGF-induced HBMEC migration, compared to VEGF-stimulated cells (overall P < 0.05). (B) Representative images of HBMEC in the scratch assay (2, 4, 6, 8 hours) after treatment with RhVEGF165 and AMA0428 0.5μM. (C) The HBVPs were treated with AMA0428 (0.5–5.0 μM), DMSO, and/or RhPDGF-BB (20–40 ng/mL). Migration of HBVPs was significantly stimulated by AMA0428 in both concentrations compared to DMSO/starvation medium (overall P < 0.05). Data are mean ± SEM of three independent experiments (n = 3). Each experiment was performed in triplicate. (C) Representative images of HBVP in the scratch assay (2, 4, 6, 8 hours) after treatment with AMA0428 0.5 μM.
Figure 3
 
Migration of HBVPs toward HUVECs was assessed in an under-agarose migration coculture assay. (A) The HUVECs were treated with AMA0428 (1, 2.5, 5 μM) or control (DMSO). The AMA0428 significantly stimulated HBVP recruitment by HUVECs compared to DMSO/starvation medium (overall P < 0.05). Data are represented as mean ± SEM of three independent experiments (n = 3). Each experiment was performed in triplicate. (B) Representative images of the under-agarose migration assay 24, 48, and 72 hours after treatment with 1 μM of AMA0428.
Figure 3
 
Migration of HBVPs toward HUVECs was assessed in an under-agarose migration coculture assay. (A) The HUVECs were treated with AMA0428 (1, 2.5, 5 μM) or control (DMSO). The AMA0428 significantly stimulated HBVP recruitment by HUVECs compared to DMSO/starvation medium (overall P < 0.05). Data are represented as mean ± SEM of three independent experiments (n = 3). Each experiment was performed in triplicate. (B) Representative images of the under-agarose migration assay 24, 48, and 72 hours after treatment with 1 μM of AMA0428.
Figure 4
 
The AMA0428 significantly decreased permeability in a monolayer of VEGF-stimulated HUVECs (P < 0.05 versus VEGF165-treated cells). The permeability in HUVEC monolayers was analyzed by evaluating the passage of fluorescence-conjugated FITC-dextran from upper to lower compartments. Numbers represent mean ± SEM of three independent experiments (n = 3).
Figure 4
 
The AMA0428 significantly decreased permeability in a monolayer of VEGF-stimulated HUVECs (P < 0.05 versus VEGF165-treated cells). The permeability in HUVEC monolayers was analyzed by evaluating the passage of fluorescence-conjugated FITC-dextran from upper to lower compartments. Numbers represent mean ± SEM of three independent experiments (n = 3).
Figure 5
 
Upregulation of ROCK, murine VEGF, and PlGF in choroid and retina 3 days after laser induction in C57Bl/6 mice measured by ELISA (P < 0.05). The results showed a 2-fold increase of ROCK activity, a 1.5-fold increase of VEGF concentration, and a 2.3-fold increase of PlGF levels after laser treatment as compared to baseline (gray horizontal line). Intravitreal administration of AMA0428 (1000 ng) decreased the upregulation of these factors, as compared to nontreated conditions by, respectively, 31%, 28%, and 43% (P < 0.05). Data are mean ± SEM of three independent experiments (n = 5 mice/group).
Figure 5
 
Upregulation of ROCK, murine VEGF, and PlGF in choroid and retina 3 days after laser induction in C57Bl/6 mice measured by ELISA (P < 0.05). The results showed a 2-fold increase of ROCK activity, a 1.5-fold increase of VEGF concentration, and a 2.3-fold increase of PlGF levels after laser treatment as compared to baseline (gray horizontal line). Intravitreal administration of AMA0428 (1000 ng) decreased the upregulation of these factors, as compared to nontreated conditions by, respectively, 31%, 28%, and 43% (P < 0.05). Data are mean ± SEM of three independent experiments (n = 5 mice/group).
Figure 6
 
The AMA0428 (1000 ng) seems safe by IVT injection in naïve mice. (A) Compared to vehicle, sham injection, or naïve mice, IVT administration of AMA0428 did not alter the total retinal thickness, the thickness of the outer and inner nuclear and plexiform layer, or photoreceptor layer as investigated on H&E staining. (B) The AMA0428 did not reduce the retinal ganglion cell number compared to vehicle, sham, or naïve mice. Data are represented as mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points.
Figure 6
 
The AMA0428 (1000 ng) seems safe by IVT injection in naïve mice. (A) Compared to vehicle, sham injection, or naïve mice, IVT administration of AMA0428 did not alter the total retinal thickness, the thickness of the outer and inner nuclear and plexiform layer, or photoreceptor layer as investigated on H&E staining. (B) The AMA0428 did not reduce the retinal ganglion cell number compared to vehicle, sham, or naïve mice. Data are represented as mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points.
Figure 7
 
Intravitreal administration of AMA0428 reduces inflammation in the CNV mouse model. (A) A significant inhibition of inflammation was seen with a single IVT injection on day 0 with AMA0428 (100 ng: P < 0.01; 1000 ng: P < 0.001). (B) A single IVT injection on day 3 with AMA0428 gave a significant reduction of leukocytes at day 5 (100 ng: P < 0.05; 1000 ng: P < 0.01). Early and late administration of DC101 had no effect on inflammation in the CNV model (P > 0.05). Data represent mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points. (C) Representative pictures of CD45 whole-mount staining (injection day 0) of eyes treated with vehicle or AMA0428 (1000 ng). Scale bars: 50 μm.
Figure 7
 
Intravitreal administration of AMA0428 reduces inflammation in the CNV mouse model. (A) A significant inhibition of inflammation was seen with a single IVT injection on day 0 with AMA0428 (100 ng: P < 0.01; 1000 ng: P < 0.001). (B) A single IVT injection on day 3 with AMA0428 gave a significant reduction of leukocytes at day 5 (100 ng: P < 0.05; 1000 ng: P < 0.01). Early and late administration of DC101 had no effect on inflammation in the CNV model (P > 0.05). Data represent mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points. (C) Representative pictures of CD45 whole-mount staining (injection day 0) of eyes treated with vehicle or AMA0428 (1000 ng). Scale bars: 50 μm.
Figure 8
 
Repeated IVT injections of AMA0428 blocks angiogenesis and vessel leakage in the CNV mouse model. (A) Repeated (days 0, 4, 10) IVT administration of AMA0428 gave a dose-dependent reduction of angiogenesis at day 14 (1000 ng and 100 ng: P < 0.001), similar to DC101. (B) Representative pictures of FITC-dextran flatmounts; eyes were treated with either vehicle or AMA0428 (1000 ng). (C) Repeated administration of AMA0428 (1000 ng) was as effective as DC101 in reducing vessel leakage at day 14 after laser compared to their respective controls (P < 0.05). Data are mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points. (D) Representative pictures of flatmounts of eyes perfused with Texas Red–conjugated dextran 60 kDa (red) together with 0.05 mg FITC-labeled lectin (green); eyes were treated with either vehicle or AMA0428 (1000 ng).
Figure 8
 
Repeated IVT injections of AMA0428 blocks angiogenesis and vessel leakage in the CNV mouse model. (A) Repeated (days 0, 4, 10) IVT administration of AMA0428 gave a dose-dependent reduction of angiogenesis at day 14 (1000 ng and 100 ng: P < 0.001), similar to DC101. (B) Representative pictures of FITC-dextran flatmounts; eyes were treated with either vehicle or AMA0428 (1000 ng). (C) Repeated administration of AMA0428 (1000 ng) was as effective as DC101 in reducing vessel leakage at day 14 after laser compared to their respective controls (P < 0.05). Data are mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points. (D) Representative pictures of flatmounts of eyes perfused with Texas Red–conjugated dextran 60 kDa (red) together with 0.05 mg FITC-labeled lectin (green); eyes were treated with either vehicle or AMA0428 (1000 ng).
Figure 9
 
The AMA0428 inhibits fibrosis 30 days after laser treatment in the CNV mouse model. (A) Administration of repeated (days 0, 4, 10, 20) IVT injections with AMA0428 decreased fibrosis significantly compared to vehicle (P < 0.05), whereas DC101 did not affect fibrosis. Data are represented as mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points. (B) Representative pictures of Collagen I staining of eyes treated with vehicle or AMA0428 (1000 ng). Borders of the CNV lesion are marked by a dashed line, the immunopositive area inside the dashed line represents the new fibrotic area, which was used for analysis. Scale bars: 50 μm.
Figure 9
 
The AMA0428 inhibits fibrosis 30 days after laser treatment in the CNV mouse model. (A) Administration of repeated (days 0, 4, 10, 20) IVT injections with AMA0428 decreased fibrosis significantly compared to vehicle (P < 0.05), whereas DC101 did not affect fibrosis. Data are represented as mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points. (B) Representative pictures of Collagen I staining of eyes treated with vehicle or AMA0428 (1000 ng). Borders of the CNV lesion are marked by a dashed line, the immunopositive area inside the dashed line represents the new fibrotic area, which was used for analysis. Scale bars: 50 μm.
Figure 10
 
Intravitreal administration of AMA0428 allows reducing the dose of anti–VEGF-R2 (DC101) by 4-fold. (A) The AMA0428 gave a dose-dependent reduction in CNV (P < 0.001) (B) Dose-dependent inhibition of CNV by DC101 (P < 0.001). (C) Partial replacement of anti–VEGF-R2 by AMA0428 yielded a similar anti-CNV effect as the maximal dose of anti–VEGF-R2 therapy alone. Data represent mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points.
Figure 10
 
Intravitreal administration of AMA0428 allows reducing the dose of anti–VEGF-R2 (DC101) by 4-fold. (A) The AMA0428 gave a dose-dependent reduction in CNV (P < 0.001) (B) Dose-dependent inhibition of CNV by DC101 (P < 0.001). (C) Partial replacement of anti–VEGF-R2 by AMA0428 yielded a similar anti-CNV effect as the maximal dose of anti–VEGF-R2 therapy alone. Data represent mean ± SEM. Numbers in the bars represent the used CNV lesions, which were averaged for 10 data points.
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×