Suppression of Laser-Induced Choroidal Neovascularization by a CCR3 Antagonist

Abstract

Purpose.: To evaluate the efficacy of a novel CCR3 antagonist for laser injury–induced choroidal neovascularization (CNV) in mice.

Methods.: We evaluated YM-344031, a novel and selective small-molecule CCR3 antagonist. CNV was induced by laser injury in C57BL/6J mice, and its volume was measured after 7 days by confocal microscopy. Leakage from the CNV was also measured after 7 days by fluorescein angiography. The CCR3 antagonist was administered by gavage at 1 hour before and 1 day after the laser injury, or intravitreous injection immediately after the laser injury. After the laser injury, ELISA, Western blot analysis, and real-time RT-PCR for VEGF-A expression in the RPE/choroid, and immunohistochemistry for CCR3, CCL11, Ki67, and Rac1 was performed.

Results.: Both oral administration and intravitreous injection of YM-344031 significantly suppressed the CNV volume (P < 0.0001 and P < 0.01, respectively). Pathologically significant leakage was significantly less common in YM-344031–injected mice (P < 0.0001). The mean VEGF protein level was significantly increased in vehicle-injected eyes after the laser injury (P < 0.05). Although the YM-344031–injected eyes did not show VEGF-A suppression after the laser injury, VEGF164 mRNA upregulation was significantly suppressed in YM-344031–injected mice (P < 0.05), and intravitreous injection of YM-344031 appeared to suppress CCR3, CCL11 (eotaxin), Ki67, and Rac1 expression after the laser injury.

Conclusions.: The present data suggest that the CCR3 antagonist YM-344031 can suppress CNV, via suppression of the upregulation of VEGF164 mRNA in VEGF isoform after the laser injury. Although our findings may warrant further investigation, YM-344031 may have potential as a new therapy for age-related macular degeneration.

Introduction

Age-related macular degeneration (AMD) is a leading cause of blindness in most industrialized nations. ^1,2 The blindness in AMD usually arises from invasion of the retina by choroidal neovascularization (CNV). Enhanced expression of the proangiogenic cytokine VEGF-A has been validated in patients with CNV, ³ and anti–VEGF-A antibody therapy is the current standard of care for CNV. ^4–6 However, this treatment requires repeated intravitreous injections for an indefinite period, and safety concerns regarding continual blockade of VEGF-A, which is constitutively expressed in the normal adult human retina, are emerging. ^7–10 In addition, other concerns have been raised by a report that anti–VEGF-A antibody therapy may increase the risk of cerebrovascular accidents. ¹¹ Based on these backgrounds, a new therapeutic target for CNV other than VEGF-A has been desired. ¹²  

Takeda et al. ¹³ showed that the eosinophil/mast cell chemokine receptor CCR3 is specifically expressed in CNV endothelial cells in humans with AMD, and that genetic or pharmacological targeting of CCR3 inhibited injury-induced CNV in mice. Interestingly, the CNV suppression by CCR3 blockade was caused by direct inhibition of endothelial cell proliferation and not via VEGF-A blockade. ¹³ However, a recent study showed that CCR3 was not involved in CNV development when a gelatinous protein mixture was used to induce CNV. ¹⁴ Therefore, further studies regarding the potential role of CCR3 in AMD are needed. 

In this study, we sought to further explore the therapeutic potential of CCR3 antagonist treatment for CNV. Instead of intravitreous injection of the CCR3 antagonist SB328437, which was used in the previous studies, ^13,14 we used YM-344031, a novel and selective small-molecule CCR3 antagonist that potently inhibits the ligand binding, ¹⁵ to clarify the antiangiogenic class effect of CCR3 antagonists. 

The chemical structure of YM-344031 is N-{(3R)-1-[(6-fluoro-2-naphthyl) methyl] pyrrolidin-3-yl}-2-{1-[(3-methyl-1-oxidopyridin-2-yl) carbonyl] piperidin-4-ylidene} acetamide, and it has previously been shown to inhibit mouse and cynomolgus monkey CCR3 function. ¹⁵ We examined the in vivo effects of this CCR3 antagonist in a mouse model of CNV induced by laser injury, ^16–20 which is the most widely used animal model of this disease. We examined whether the CCR3 antagonist could induce a shift in VEGF isoform, and also explored a downstream component of the CCR3 and chemokine (C-C motif) ligand 11 (CCL11) pathway, Rac1, to elucidate the mechanism of CNV suppression by this CCR3 antagonist. Rac1 has been reported to be necessary for choroidal endothelial cell transmigration of the RPE. ^21,22  

Materials and Methods

Animals

Male wild-type C57BL/6J mice (Japan SLC, Shizuoka, Japan) between 6 and 8 weeks of age were used to minimize variability. For all procedures, the animals were anesthetized with Avertin (2.5% 2,2,2-tribromoethyl and tertiary amyl alcohol; Sigma, Tokyo, Japan; 10 μL/g, 400 mg/kg intraperitoneal injection) and the pupils were dilated with topical 1% tropicamide (Santen, Osaka, Japan). The study protocol was approved by the Nagoya City University Animal Care and Use Committee. All animal experiments were performed in accordance with the guidelines of the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Mouse Model of CNV

Laser photocoagulation (532 nm, 200 mW, 100 ms, 100 μm; Elite; Lumenis, Salt Lake City, UT) was bilaterally performed (volume studies: four to six spots per eye; protein analyses: 20 spots per eye) in each animal on day 0 by one individual masked to the drug group assignment as described previously. ^16,17 The laser spots were created in a standard fashion around the optic nerve using a slit lamp delivery system and a coverslip as a contact lens. The morphologic end point of the laser injury was the appearance of a cavitation bubble, which is thought to be correlated with disruption of Bruch's membrane. 

CCR3 Antagonist Treatment

The CCR3 antagonist YM-344031 (provided by Astellas Pharma, Inc., Tokyo, Japan) was dissolved in dimethylsulfoxide (DMSO) (Sigma). YM-344031 (50 mg/kg/day) or the same volume of vehicle (DMSO) was given to one group of mice by gavage at 1 hour before and 1 day after the laser treatment using a 1-mL syringe and a 20-gauge 3.5-cm gavage needle (Thermo Fisher Scientific K.K., Kanagawa, Japan). Another group of mice was injected with 0.1 to 10.0 μg/μL of YM-344031 or the same volume of vehicle (DMSO) into the vitreous humor using a 33-gauge needle (Ito Corporation, Tokyo, Japan) immediately after the laser injury. 

Fluorescein Angiography

Fluorescein angiography was performed by an operator masked to the treatment group assignments of the animals using a fundus camera (TRC-50AX; Topcon, Tokyo, Japan) at 1 week after the laser photocoagulation. Photographs were captured with a 20-diopter lens in contact with the fundus camera lens after intraperitoneal injection of 0.1 mL of 1% fluorescein sodium (Alcon, Tokyo, Japan). 

The lesions were graded as described previously. ¹⁷ Briefly, the lesions were graded on an ordinal scale based on the spatial and temporal evolution of fluorescein leakage as follows: 0 (nonleaky) = no leakage, faint hyperfluorescence, or mottled fluorescence without leakage; 1 (questionable leakage) = hyperfluorescent lesion with no progressive increase in size or intensity; 2 (leaky) = hyperfluorescence increasing in intensity but not in size, no definite leakage; 3 (pathologically significant leakage) = hyperfluorescence increasing in intensity and size, definite leakage. 

CNV Volume

At 1 week after the laser injury, the eyes were enucleated and fixed with 4% paraformaldehyde. The eyecups obtained by removing the anterior segments were incubated with 0.5% fluorescein-isothiocyanate (FITC)-isolectin B4 (Vector Laboratories, Burlingame, CA). CNV was visualized using a blue argon laser wavelength (488 nm) and a scanning laser confocal microscope (LSM 5 Pascal; Carl Zeiss Meditec GmbH, Oberkochen, Germany). 

Horizontal optical sections were obtained at 1-μm intervals from the surface of the RPE/choroid/sclera complex. The deepest focal plane in which the surrounding choroidal vascular network connected to the lesion could be identified was judged to be the floor of the lesion. Any vessel in the laser-treated area and superficial to this reference plane was judged as CNV. The images of each section were digitally stored. The area of CNV-related fluorescence was measured using ImageJ software (developed by Wayne Rasband, National Institutes of Health, Bethesda, MD; available at http://rsb.info.nih.gov/ij/index.html). The summation of the whole fluorescent area in each horizontal section was used as an index for the volume of CNV, as described previously. ^16,17,19 The average volume obtained from all four to six laser spots per eye was generated (n = number of eyes). Imaging was performed by an operator masked to the treatment group assignments. 

ELISA for VEGF

Previous reports showed the mean VEGF level in the RPE/choroid peaked at 3 days after laser photocoagulation. ^17,19,20,23 The RPE/choroid complex was carefully isolated from the eyes at 3 days after photocoagulation and sonicated in lysis buffer (20 mM imidazole HCl, 10 mM KCl, 1 mM MgCl₂, 10 mM EGTA, 1% Triton X-100, 10 mM NaF, 1 mM Na molybdate, 1 mM EDTA, protease inhibitor cocktail [Sigma-Aldrich, St. Louis, MO]) on ice for 15 minutes. The lysate was centrifuged at 15,000 rpm for 15 minutes at 4°C, and the VEGF protein levels in the supernatants were determined using a Quantikine M ELISA kit (threshold of detection, 3 pg/mL; R&D Systems, Minneapolis, MN) that recognizes all splice variants at wavelengths of 450 to 570 nm (SpectraMax340; Molecular Devices, Sunnyvale, CA). All data were normalized by the total protein levels measured by the Bio-Rad Protein assay (Bio-Rad, Hercules, CA). 

VEGF Western Blot Analysis

To confirm the specificity of the ELISA data, and clarify the VEGF isoforms, we performed Western blot analyses 3 days after the laser photocoagulation. The eyes were enucleated and the RPE/choroid complex was sonicated in the above-described lysis buffer on ice for 15 minutes. 

Aliquots (60 μg of total protein) of each sample from the RPE/choroid were separated in sodium dodecyl sulfate-10% to 20% polyacrylamide gradient gels (e-PAGEL; Atto Corporation, Tokyo, Japan) and electrophoretically transferred onto polyvinylidene fluoride membranes (Immobilon Transfer Membrane; Millipore, Bedford, MA). The resulting membranes were blocked with 10% skim milk for 1 hour at room temperature. Two different isoforms of VEGF-A were detected by incubating the membranes with an anti-VEGF_120/164 rat antibody (1:500; R&D Systems) at 4°C overnight. Subsequently, the membranes were incubated with an alkaline phosphatase–conjugated antirat IgG secondary antibody (AP 136 A, 1:500; Chemicon International, Temecula, CA) for 2 hours at room temperature. The signals were visualized by chromogenic Western blot assays using 5-bromo-4-chloro-3-indolyl-phosphate and 4-nitroblue tetrazolium chloride (Roche Diagnostics GmbH, Mannheim, Germany). Quantification of band intensity (optical density) was carried out on scanned Western blot images using ImageJ from blots of independent experiments. Each experiment was repeated five times. 

RNA Isolation and Real-Time RT-PCR

To confirm the specificity of the Western blot data, and clarify the VEGF isoforms, we performed real-time RT-PCR at 0, 12, and 24 hours after the laser photocoagulation. The eyes were enucleated and the RPE/choroid complex was sonicated in the above-described lysis buffer on ice for 15 seconds. Total RNA was isolated from aliquots of each sample from the RPE/choroid and was further purified with a PureLink RNA Mini Kit (Life Technologies, Carlsbad, CA). After RNase-free DNase treatment by using a TURBO DNA-free Kit (Applied Biosystems, Foster City, CA), total RNA was reverse transcribed in a 20-μL reaction system using Super Script VILO Master Mix (Life Technologies) under conditions described by the supplier. The mRNA level of each VEGF isoform was quantified by real-time RT-PCR on the Applied Biosystems 7500 Fast Real-Time PCR System. PCR was performed using Fast Advanced Master Mix (Life Technologies) containing 900 nM forward primer, 900 nM reverse primer, and 250 nM Taqman probe at 20 μL per tube according to the manufacturer's instructions. The pairs of primers and the TaqMan probes for the target mRNAs were designed according to a previous report (Table). ²⁴ PCR cycles consisted of an incubation step at 50°C for 2 minutes, and polymerase activation step at 95°C for 20 seconds, followed by 40 cycles of denature steps at 95°C for 3 seconds and annealing and extending steps at 60°C for 30 seconds. PCR amplification of the housekeeping gene, murine glyceraldehyde-3-phosphate dehydrogenase (GAPDH), using TaqMan Gene Expression Assay (Life Technologies), was used for each sample as control of sample loading and to allow normalization between samples. Each sample was analyzed three times and each PCR experiment included two nontemplate control and a negative reverse transcription control wells. 

Table

 

Primer Sequence

Primer Name	Sequence
VEGF common forward	GCC AGC ACA TAG AGA GAA TGA GC
VEGF 120 reverse	CGG CTT GTC ACA TTTT TCT GG
VEGF 164 reverse	CAA GGC TCA CAG TGA TTT TCT GG
VEGF common probe	ACA GCA GAT GTG AAT GCA GAC CAA AGA AAG

Immunohistochemistry

To evaluate the proliferation of endothelial cells, double-staining for Ki67 and isolectin B4 was performed after the laser injury. Ki67 detects cells in G1 phase, S phase, G2 phase, or mitosis, but not in G0 phase. ^25,26 The expression of CCR3, CCL11 (eotaxin), and Rac1 was also evaluated immunohistochemically. 

Mouse eyes were enucleated at 3 days after the laser injury, snap-frozen in optimal cutting temperature compound (Tissue Tek; Miles Laboratories, Elkhart, IN), and cryosectioned. Frozen sections (7 μm) were fixed in Histochoice MB (Amresco, Euclid, OH), and blocked with Dako Cytomation Protein Block (Dako Cytomation, Carpinteria, CA) for 1 hour at room temperature. The sections were then incubated overnight at 4°C with the following primary antibodies: rabbit anti-Ki67 (1:200; Abcam, Inc., Cambridge, MA); rabbit antimouse Rac1 (1:50; Santa Cruz Biotechnology, Santa Cruz, CA); rabbit antimouse CCR3 (1:50; Abcam, Inc.); goat antimouse CCL11 (eotaxin1) (1:50; R&D Systems); and 0.5% FITC-isolectin B4 (1:100; Vector Laboratories). Subsequently, the sections were incubated with Cy 3-conjugated donkey antirabbit or antigoat IgG secondary antibodies (1:200; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) for 3 hours at room temperature. Nuclei were counterstained in 4,6-diamidino-2-phenylindole (DAPI)-containing mounting medium (Vector Laboratories). The sections were examined using a microscope (AX70; Olympus, Tokyo, Japan). Primary antibody omission or substitution with an irrelevant antibody of the same species and staining with chromogen alone were used as negative controls. All immunohistochemistry was performed for at least three specimens from each time point and each treatment. 

Statistical Analysis

All results were expressed as mean ± SEM (n = number of eyes). Values were analyzed statistically using the Mann-Whitney U test. Differences were considered to be statistically significant at P < 0.05. 

Results

Oral Administration of the CCR3 Antagonist Suppresses CNV

Oral administration of YM-344031 suppressed the laser-induced CNV in wild-type mice. The volume of laser-induced CNV was significantly reduced in the mice treated by oral administration of YM-344031 compared with control mice (148,891.6 ± 27,311.9 vs. 382,762.4 ± 15,047.7 μm³, P < 0.0001) (Figs. 1A, 1B). 

Figure 1

 

The CCR3 antagonist suppresses the CNV volume. (A, B) Oral administration of the CCR3 antagonist suppresses the CNV volume. (A) Representative examples of CNV in mice treated with DMSO (control) and the CCR3 antagonist YM-344031 (50 mg/kg/day). Bars: 100 μm. (B) The CNV volume is significantly suppressed by the CCR3 antagonist at 50 mg/kg/day compared with the vehicle control (DMSO). Data are means ± SEM (n = 8). *P < 0.0001. (C, D) Intravitreous injection of the CCR3 antagonist suppresses the CNV volume. (C) Representative examples of CNV in mice treated with DMSO (a) or the CCR3 antagonist at 0.1 μg (b), 1.0 μg (c), or 10.0 μg (d). Bars: 100 μm. (D) The CNV volume is significantly suppressed by the intravitreous injections of the CCR3 antagonist in a dose-dependent manner. Data are means ± SEM (n = 8). *P < 0.01 versus control (DMSO).

Intravitreous Injection of the CCR3 Antagonist Suppresses CNV

Intravitreous injection of YM-344031 also suppressed the laser-induced CNV in wild-type mice. The volume of laser-induced CNV was significantly reduced in the mice injected with the CCR3 antagonist at 0.1, 1.0, or 10.0 μg in a dose-dependent manner (P < 0.01) (Figs. 1C, 1D). The CCR3 antagonist at 10 μg caused significant suppression of the CNV volume compared with the vehicle (193,775.6 ± 19,265.9 vs. 339,812.2 ± 43,581.9 μm³, P < 0.01); however, there was no significant difference in the CNV sizes between the mice injected with the CCR3 antagonist at 0.1 and 10.0 μg. 

Intravitreous Injection of the CCR3 Antagonist Suppresses Vascular Leakage from CNV

At 1 week after the laser photocoagulation, pathologically significant leakage (grade 3 lesions) developed in most of the DMSO-injected mice, but in significantly fewer YM-344031–injected mice (P < 0.001) (Fig. 2). However, there was no significant difference in the vascular leakage between the mice injected with the CCR3 antagonist at 0.1 and 10.0 μg. 

Figure 2

 

Grade 3 lesions are significantly less common in CCR3 antagonist–injected eyes than in DMSO-injected eyes. (A) Representative late-phase (6–8 minutes) fluorescein angiograms of eyes treated with DMSO (a) or the CCR3 antagonist at 0.1 μg (b), 1.0 μg (c), or 10.0 μg (d). (B) Percentages of grade 3 lesions. Data are means ± SEM (n = 8–14). *P < 0.001.

The CCR3 Antagonist Does Not Suppress Total VEGF Expression but Suppresses VEGF164 Expression

The VEGF levels in the RPE/choroid at 3 days after the laser injury were significantly increased in the DMSO-injected eyes (7.32 ± 1.37 pg VEGF/mg total protein, P < 0.05) and YM-344031–injected eyes (7.06 ± 0.76 pg VEGF/mg total protein, P < 0.05) compared with the level on day 0. There was no significant difference in the VEGF expression levels between the DMSO-injected eyes and YM-344031–injected eyes (Fig. 3A). 

Figure 3

 

VEGF protein levels in the RPE/choroid. (A) The VEGF levels in the RPE/choroid are significantly increased at 3 days after the laser injury in control DMSO– and CCR3 antagonist–injected eyes. Data are means ± SEM (n = 5). *P < 0.05 versus day 0. Intravitreous injection of the CCR3 antagonist at 10.0 μg does not suppress VEGF expression. NS, not significant. (B, C) Western blot analyses showing induction of two isoforms of VEGF protein. The VEGF level in the RPE/choroid is upregulated at 1 day after the laser photocoagulation, and reaches a peak at 3 days after the laser injury. In DMSO-injected eyes, the VEGF164 isoform shows a very strong response, whereas VEGF120 shows only a marginal response. In contrast, in YM-344031–treated eyes, the VEGF120 isoform shows a very strong response, whereas VEGF164 shows only a marginal response at 3 days after laser photocoagulation. The membranes were scanned using a flatbed scanner and the band intensities analyzed using ImageJ software. The expression levels are calculated as percentages of the VEGF expression in the day 0 control. (D, E) Real-time RT-PCR analysis showing CCR3 antagonist treatment suppressed VEGF164 upregulation and induced VEGF120 expression. Vegfa120 mRNA expression in RPE/choroid of CCR3 antagonist–treated eyes was significantly upregulated and Vegfa164 mRNA expression was significantly suppressed compared with DMSO-injected eyes at 24 hours after the laser injury. Gene expression measured by real time RT-PCR was normalized to GAPDH mRNA levels and to baseline levels (n = 6). *P < 0.05 compared to control (day 0). (F) Real-time RT-PCR analysis showing CCR3 antagonist resulted in a reversal of the relative ratio of VEGF120 to VEGF164 mRNA, with VEGF120 upregulation at 24 hours after the laser injury (*P = 0.0003, compared to DMSO-treated eyes).

Western blot analysis showed the VEGF level in the RPE/choroid at 3 days after the laser photocoagulation. Semiquantitative analysis demonstrated VEGF 164 levels were significantly lower in CCR3 antagonist-injected eyes compared with DMSO-injected eyes (P < 0.05) (Figs. 3B, 3C). 

To confirm the result of Western blot analysis, we performed real-time RT-PCR for VEGF isoforms. The real time RT-PCR analysis showed that VEGF120 mRNA expression in RPE/choroid of CCR3 antagonist–treated eyes was significantly upregulated (P < 0.05) and VEGF164 mRNA expression was significantly suppressed (P < 0.05) compared with DMSO-injected eyes at 24 hours after the laser injury (Figs. 3D, 3E). CCR3 antagonist resulted in a reversal of the relative ratio of VEGF120 to VEGF164 mRNA, with VEGF120 upregulation at 24 hours after the laser injury (P = 0.0003) (Fig. 3F). 

The CCR3 Antagonist, CCR3, CCL11, Ki67, and Rac1 Expression after Laser Injury

To examine the expression and localization of CCR3, CCL11, Ki67, and Rac1 in CCR3 antagonist–treated eyes, sections were stained with antibodies together with isolectin B4, as a marker for vascular endothelial cells. The immunohistochemical analyses showed CCR3 staining in isolectin B4–positive endothelial cells after the laser injury, but less expression of CCR3 in YM-344031–treated eyes (Fig. 4A). CCL11, one of the ligands for CCR3, was located in the RPE cell layer in vehicle-treated eyes, but showed reduced localization in YM-344031–treated eyes (Fig. 4B). 

Figure 4

 

The CCR3 antagonist suppresses CCR3 and CCL11 expression after laser injury. Bars: 50 μm. (A) Immunohistochemical analyses of CCR3. (a) CCR3 staining is detected in isolectin B4–positive endothelial cells. Green fluorescence from isolectin B4 (left) and red fluorescence from the anti-CCR3 antibody (middle) identifies the isolectin B4–positive endothelial cells as expressing CCR3 when the images are superimposed (right). (b) CCR3 antagonist-injected eyes show lower expression of isolectin B4–positive and CCR3-positive cells after the laser injury. (B) Immunohistochemical analyses of CCL11. (a) CCL11 is detected in the RPE cell layer. Red fluorescence from the anti-CCL11 antibody is merged with the Nomarski image (left) and with DAPI staining (right). (b) CCR3 antagonist-injected eyes express less CCL11 in the RPE cell layer.

To compare the proliferation status of endothelial cells after the laser injury between vehicle– and YM-344031–treated eyes, double-immunostaining for Ki67 and isolectin B4 was performed. The staining revealed relatively less proliferating isolectin-positive cells within the laser-injured area of the CCR3 antagonist-treated mice (Fig. 5). Ki67 is usually expressed in the nucleus, and cytoplasmic expression of Ki67 is thought to indicate the active mitosis phase. 

Figure 5

 

The CCR3 antagonist suppresses Ki67-positive cells after laser injury. (A) DMSO-treated eyes. (B) CCR3 antagonist-treated eyes. Bars: 50 μm. (A) Red fluorescence from the anti-Ki67 antibody (middle) identifies isolectin B4–positive endothelial cells (green) within the laser-injured area. (B) CCR3 antagonist–treated eyes show lower expression of both Ki67– and isolectin B4–positive endothelial cells.

Rac1 is a downstream component of the eotaxin–CCR3 pathway, and activates the migration of choroidal endothelial cells. Rac1 was expressed on isolectin B4–positive endothelial cells after the laser injury, but less expression of Rac1 was noted in YM-344031–treated eyes (Fig. 6). 

Figure 6

 

The CCR3 antagonist suppresses Rac1 expression after laser injury. (a) DMSO-treated eyes. (b) CCR3 antagonist–treated eyes. Bar: 50 μm. Rac1 (red) is expressed on isolectin B4–positive endothelial cells (green) after the laser injury (a), but lower expression of Rac1 is noted in CCR3 antagonist-treated eyes (b).

Discussion

Previously, Takeda et al. ¹³ showed that CNV suppression by CCR3 blockade was caused by direct inhibition of endothelial cell proliferation and not via VEGF-A blockade. We demonstrated that another CCR3 antagonist, YM-344031, also suppressed CNV, supporting the previous report. ¹³ The CCR3 antagonist did not suppress VEGF upregulation after laser injury. We found that the CCR3 antagonist might induce a shift in VEGF isoforms, suppress VEGF164 upregulation, and induce VEGF120 expression. 

In contrast, Li et al. ¹⁴ previously reported that CCR3 targeting was ineffective for CNV inhibition when a gelatinous protein mixture (Matrigel) was injected into the subretinal space to induce CNV in an animal model. Because we examined the antiangiogenic effects of our CCR3 antagonist in a mouse model of CNV induced by laser injury, the most obvious difference between our study and their study is the CNV models used. Wang et al. ²⁷ demonstrated that CCR3 is expressed under certain conditions of aging or oxidative stress, which are known factors associated with advanced AMD. We used a laser-induced CNV model, and laser injury to mouse retinas is known to cause increased generation of reactive oxidative species (ROS) through activated nicotinamide adenine dinucleotide phosphate (NADPH) oxidase. ²⁸ It is not clear whether injection of Matrigel into the subretinal space caused ROS upregulation in the CNV assay. ¹⁴  

The CCR3–eotaxin axis is a newly identified pathway that contributes to ocular pathological angiogenesis, not only in CNV ^13,29 but also in corneal angiogenesis. ³⁰ The CCR3–eotaxin pathway has been reported to directly induce angiogenic responses in human, mouse, rat, and chick endothelial cells. ³¹ In vitro, eotaxin induced human endothelial cell migration in a dose-responsive manner that was correlated with CCR3 expression at the RNA and protein levels. ³¹ In a corneal neovascularization model, Liclican et al. ³⁰ showed that inflammatory neovascularization with prostaglandin E2 was independent of the VEGF circuit, and was instead associated with significant induction of the CCR3–eotaxin pathway. In agreement with their report, our experiments showed that CNV suppression by the CCR3 antagonist was independent of the total VEGF expression, and that the CCR3 antagonist suppressed eotaxin expression after the laser injury. 

Clinically, vascular leakage from CNV causes exudative changes that can directly affect visual symptoms. Vascular leakage from CNV has been thought to enhance immature vessel formation and VEGF expression. In our study, the CCR3 antagonist did not suppress VEGF expression, but did suppress vascular leakage from CNV. The CCR3–eotaxin pathway was reported to increase endothelial permeability in human coronary artery endothelial cells. ³² Rac1 is known to regulate a sequential process that includes enhancement of endothelial permeability, and promotion of endothelial proliferation, migration, and tubulogenesis. ²² Wang et al. ²⁷ showed that a CCR3 inhibitor prevented Rac1 activation in choroidal endothelial cells in vitro. Further investigation is still required, but since our immunohistochemical study showed that the CCR3 antagonist suppressed Rac1 expression after the laser injury, the Rac1 pathway may be one of the pathways involved in the suppression of vascular leakage by the CCR3 antagonist. 

Although there were no significant differences in the expression of VEGF evaluated by ELISA between vehicle- and CCR3 antagonist–injected eyes, we found that the CCR3 antagonist appeared to suppress VEGF164 upregulation and induce VEGF120 expression after the laser injury by Western blot analyses and real time RT-PCR. VEGF164 is a more proinflammatory isoform than VEGF120, and is preferentially induced in pathological neovascularization. ^33,34 In contrast, VEGF120 is more potent at neuroprotection. ⁸ The inhibition of VEGF164 by the CCR3 antagonist may be one of the explanations for why the CCR3 antagonist was able to suppress CNV. 

Although anti-VEGF therapy is the standard care for AMD, ^4–6 anti-VEGF therapy is also associated with concerns about retinal toxicity and systemic complications. ⁹ To combat AMD, which is a leading cause of visual impairment in elderly individuals, clarification of an alternative pathway besides the VEGF pathway is important for the development of new therapies. Although further investigations are needed, a new therapeutic strategy targeting CCR3 for CNV may have a large advantage because it would not suppress VEGF121, which is constitutively expressed in the normal adult human retina and is an important neuroprotectant. 

Acknowledgments

The authors thank Mayumi Kato for technical assistance. 

Footnotes

 Supported by a Grant-in-Aid for Scientific Research (C) from the Japan Society for the Promotion of Science and a Grant-in-Aid for Scientific Research from the Ministry of Health, Labour and Welfare of Japan.

Footnotes

 Disclosure: T. Mizutani, None; M. Ashikari, None; M. Tokoro, None; M. Nozaki, None; Y. Ogura, None