Free
Retina  |   April 2014
Angiopoietin-1 Suppresses Choroidal Neovascularization and Vascular Leakage
Author Affiliations & Notes
  • Junyeop Lee
    National Research Laboratory of Vascular Biology and Stem Cells, Graduate School of Medical Science and Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea
  • Dae-Young Park
    National Research Laboratory of Vascular Biology and Stem Cells, Graduate School of Medical Science and Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea
  • Do Young Park
    National Research Laboratory of Vascular Biology and Stem Cells, Graduate School of Medical Science and Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea
  • Intae Park
    National Research Laboratory of Vascular Biology and Stem Cells, Graduate School of Medical Science and Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea
  • Woohyok Chang
    Department of Ophthalmology, College of Medicine, Yeungnam University, Daegu, Korea
  • Yoshikazu Nakaoka
    Department of Cardiovascular Medicine, Osaka University Graduate School of Medicine, Suita, Osaka, Japan
  • Issei Komuro
    Department of Cardiovascular Medicine, Tokyo University Graduate School of Medicine, Tokyo, Japan
  • Gou Young Koh
    National Research Laboratory of Vascular Biology and Stem Cells, Graduate School of Medical Science and Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea
  • Correspondence: Gou Young Koh, Graduate School of Medical Science and Engineering, KAIST 373-1, Guseong-dong, Daejeon, 305-701, Republic of Korea; gykoh@kaist.ac.kr
  • Ook-Joon Yoo, Graduate School of Medical Science and Engineering, KAIST 373-1, Guseong-dong, Daejeon, 305-701, Republic of Korea; ojyoo@kaist.ac.kr
  • Footnotes
     Current affiliation: *Department of Ophthalmology, University of Ulsan, College of Medicine, Asan Medical Center, Seoul, Korea.
  • Footnotes
     Graduate School of Biomedical Science and Engineering, KAIST 373-1, Guseong-dong, Daejeon, 305-701, Republic of Korea; .
Investigative Ophthalmology & Visual Science April 2014, Vol.55, 2191-2199. doi:10.1167/iovs.14-13897
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Junyeop Lee, Dae-Young Park, Do Young Park, Intae Park, Woohyok Chang, Yoshikazu Nakaoka, Issei Komuro, Ook-Joon Yoo, Gou Young Koh; Angiopoietin-1 Suppresses Choroidal Neovascularization and Vascular Leakage. Invest. Ophthalmol. Vis. Sci. 2014;55(4):2191-2199. doi: 10.1167/iovs.14-13897.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: To investigate the role of angiopoietin-1 (Ang1) in choroidal neovascularization (CNV) and vascular leakage.

Methods.: We generated laser-induced CNV in mice and measured the size of CNV and vascular leakage after intravitreal administration of Ang1. The expressions and distributions of endothelial junctional proteins were analyzed using immunohistochemistry and Western blot. Moreover, we compared the sizes of CNV and vascular leakage in Ang1-overexpressing, Ang1-deficient, and their littermate control mice. In addition, following the transplantation of GFP+ bone marrow cells into these Ang1-genetically modified mice, we evaluated the recruitment of VEGF-A producing macrophages from the bone marrow after CNV induction.

Results.: Intravitreal administration of Ang1 was as effective as VEGF-Trap in inhibiting CNV formation. Furthermore, Ang1 suppressed vascular leakage by increasing endothelial junctional proteins, which was more effective than VEGF-Trap. Genetic deletion of Ang1 exacerbated, while overexpression of Ang1 suppressed CNV formation and vascular leakage. We attribute these Ang1-induced, anti-angiogenic, and anti-leakage effects to its inhibitory actions against the recruitment and infiltration of VEGF-A–producing macrophages from bone marrow into the inflammatory lesions.

Conclusions.: Ang1 supplementation can be established as a therapeutic strategy to suppress the CNV formation and vascular leakage by inhibiting the recruitment of angiogenic macrophages and tightening the endothelial junctions.

Introduction
Choroidal neovascularization (CNV), the pathologic growth of new blood vessels from preexisting choroidal vessels, is a vision-threatening complication occurring in the late stage of AMD. Progressive accumulation of lipofuscin, a photoreceptor byproduct in the dysfunctional RPE, provokes a localized inflammation, breaking the balance between pro-angiogenic and anti-angiogenic growth factors and eventually leading to CNV formation. 13 In this neovascular AMD, vascular leakage from immature choroidal neovessels is the main underlying cause of severe manifestations such as RPE detachment, vitreous or retinal hemorrhage, and fibrovascular disciform scarring, consequently giving rise to severe visual loss and blindness. 13  
Because VEGF-A plays a cardinal role in the pathogenesis and progression of CNV, several drugs targeting VEGF-A have been developed and widely used for the treatment of neovascular AMD. 4,5 Anti-VEGF-A antibodies (the antigen-binding fragment ranibizumab, and the full form bevacizumab) and VEGF-Trap fusion protein (aflibercept) are currently being used for patients with neovascular AMD, and each has been demonstrated to have substantial advantages. 4,5 Accumulating evidence indicates an excellent safety profile for periodic, intravitreal administration of anti–VEGF-A drugs, with no overt ocular or systemic adverse effects attributable to VEGF-A inhibition. 6 However, a practical difficulty remains in that the current therapies must be performed repeatedly and continuously. 5 In addition, some sets of patients with AMD showed refractoriness or tachyphylaxis against anti–VEGF-A antibodies. 79 Intriguingly, one possible mechanism for the tachyphylaxis was suggested to be that the infiltrated macrophages within CNV membrane may upregulate VEGF-A production as a compensatory response to intravitreal VEGF-A blockade. 9 This hypothesis was supported by another previous report that demonstrated that the surgically excised human CNV membrane of eyes that previously received intravitreal anti–VEGF-A antibody displayed an increased density and proliferative activity of inflammatory macrophages compared with those of non-treated eyes. 10 Thus, it would be desirable to develop an alternative or combinative strategy to overcome the unexpected potential adverse effects and limitations of VEGF-A blockades on the treatment of CNV. 
Angiopoietins are the second family of vascular growth factor ligands known to have diverse but selective effects on vascular endothelial cell (EC). 11 Among them, angiopoietin-1 (Ang1) is essential for vessel assembly and maturation in embryos, and is necessary for vessel quiescence and stabilization in adults. 11,12 Supplemental Ang1 promotes blood vessel remodeling and formation in several organs; these blood vessels are nonleaky, noninflammatory, functional, and stable, and thus can relieve ischemic side effects. 1317 Furthermore, Ang1 reduces inflammation-induced vascular leakage and inflammatory cell infiltration by tightening cell junctions and reducing adhesion molecules in the targeted ECs. 18,19 Previous studies using CNV animal models have demonstrated that transgenic Ang1 overexpression in retinas or intravitreal Ang1 administrations reduced the CNV size 2022 ; however, the underlying mechanisms of how Ang1 inhibits CNV and the functional relevance of the Ang1-treated CNV is yet to be clearly defined. 
In the present study, we investigated whether Ang1 inhibits CNV formation. To do this, we performed an intravitreal administration of COMP-Ang1, a soluble and potent variant of Ang1, 23 to the laser-induced CNV mouse model and compared the effect of Ang1 against VEGF-Trap. In addition, we examined whether Ang1 is effective for the prevention of vascular leakage from CNV using fluorescein angiography (FA) and indocyanine green angiography (ICGA). We demonstrated that Ang1 indeed suppresses vascular leakage from CNV by increasing the expressions of inter-EC junctional proteins. Furthermore, we evaluated the underlying mechanisms by which Ang1 inhibits CNV formation and vascular leakage using genetically modified mice with Ang1 gain- and loss-of-function. Our results show that Ang1 suppresses CNV formation by inhibiting the recruitment and infiltration of VEGF-A-producing macrophages from the bone marrow (BM). 
Methods
Mice and CNV Mice Models
Animal care and experimental procedures were performed with the approval of the Animal Care Committee of KAIST. Mice were handled in accordance and with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Specific pathogen-free C57BL/6J mice (Jackson #000664); Tie2-GFP mice (Jackson #003658); and ROSA26-Cre ERT2 mice (Jackson #008463) were purchased from The Jackson Laboratory (Bar Harbor, ME, USA). GFP+ transgenic mice (C57BL/6J genetic background) were a gift from Masaru Okabe (Osaka University, Japan). To knockdown or overexpress Ang1 globally in a tamoxifen-dependent manner, Ang1 flox/flox (Ang1fl/fl) 16 or COMP-Ang1-Tg mice 24 were intercrossed with ROSA26-Cre ERT2 mice. 25 Tamoxifen (Sigma-Aldrich, St. Louis, MO, USA) was dissolved in corn oil (Sigma-Aldrich), and 2 mg of the resulting tamoxifen solution was injected into the peritoneal cavity. Laser-induced CNV in mice was generated by a previously described method 26 with some modifications. Male C57BL/6J mice were anesthetized with an intramuscular injection of 40 mg/kg ketamine and 12 mg/kg xylazine. Topical 0.5% proparacaine was applied and the pupils were dilated with 2.5% phenylephrine (Alcon, Fort Worth, TX, USA). Laser photocoagulator (Carl Zeiss AG, Oberkochen, Germany) with a slit-lamp delivery system was used with a coverslip as a contact lens to visualize the retina. Sufficient laser energy (250 mW, 100 ms, 50 μm) was delivered to rupture Bruch's membrane in three locations for each eye (the 2, 6, and 10 o'clock positions of the posterior pole). The CNV formation was confirmed by the production of a bubble at the time of laser photocoagulation, which indicates the rupture of Bruch's membrane, an important factor in obtaining experimental CNV. 26  
Generation and Intravitreal Administration of Recombinant Proteins
To generate recombinant proteins, FLAG-tagged COMP-Ang1, VEGF-Trap, and dimeric-Fc, stable CHO cell lines that secrete these recombinant proteins were used as previously described. 23,27 Recombinant proteins in supernatant were purified by column chromatography with anti-Flag M2 antibody-agarose affinity gel (Sigma-Aldrich) or Protein A-agarose gel (Oncogene; Millipore Corp., Billerica, MA, USA) and Flag peptide (Sigma-Aldrich) or acid. To deliver the recombinant proteins into the mouse choroid, indicated amount of reagents were injected into the vitreous cavity using a micro-injector (Nanoliter 2000; World Precision Instruments, Sarasota, FL, USA) fitted with glass capillary pipettes under anesthesia. As a control, Fc or BSA was injected in the same manner. The mice were killed and examined 2 weeks after laser injury unless indicated otherwise. 
Simultaneous Confocal Scanning Laser FA and ICGA
Two weeks after the laser photocoagulation, in vivo fundus FA and ICGA of the CNV lesion was performed using commercial retinal angiography system (Heidelberg retinal angiography system [HRA]; Heidelberg Retinal Engineering, Heidelberg, Germany). Under systemic anesthesia and pupil dilation, a mixture of 10 mg of fluorescein sodium (Alcon) and 0.15 mg of ICG (Dongin-dang, Shihung-shi Gyeonggi, South Korea) was administered intraperitoneally. Late-phase FA and ICGA images were captured simultaneously using HRA with 55° angle lens. Leaky areas from CNV were calculated as total measured hyperfluorescent areas in FA images divided by the total measured CNV area in ICGA images. 
Isolation of Bone Marrow Cells and Bone Marrow Transplantation (BMT)
Bone marrow cells (2 × 106) were harvested from femurs and tibias of GFP + transgenic mice by flushing with ice-cold Dulbecco's PBS (DPBS; Sigma-Aldrich). Eight-week-old recipient mice (ROSA26-Cre ERT2/Ang1 flox/flox and ROSA26-Cre ERT2/COMP-Ang1-Tg and their littermates) were sublethally irradiated at a dose of 4.5 Gy with a gamma irradiator (Gammacell 3000; MDS Nordion, Inc., Ottawa, Ontario, Canada). Bone marrow cells were then injected intravenously into the recipient mice 16 hours after irradiation. The choroid/sclera tissues of BMT mice were examined at the indicated time. 
Histologic and Morphometric Analyses
Immunohistochemistry (IHC) of a whole-mounted choroid/sclera complex or trachea was performed as previously described. 27,28 The choroid/sclera complex or trachea were incubated with hamster anti-CD31 monoclonal antibody (2H8; Millipore Corp.) or with one or more of the following antibodies: rabbit anti–zonnula occludens (ZO)-1 polyclonal antibody (Invitrogen, Carlsbad, CA, USA); rat anti–vascular endothelial (VE)-cadherin monoclonal antibody (11D4.1, BD Pharmingen; BD Biosciences, Franklin Lakes, NJ, USA); rat anti-F4/80 monoclonal antibody (BM8; eBioscience, San Diego, CA, USA); goat anti-VEGF-A polyclonal antibody (R&D Systems, Minneapolis, MN, USA). After several washes, the samples were incubated for 4 hours at room temperature with FITC- or Cy3- conjugated anti-hamster IgG antibody (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA) or with the following antibodies: Cy3- or Cy5-conjugated anti-rat antibody (Jackson ImmunoResearch Laboratories, Inc.); FITC- or Cy3-conjugated anti-rabbit antibody (Jackson ImmunoResearch Laboratories, Inc.). For the control experiments, the primary antibody was omitted or substituted with pre-immune serum. Flatmount stained choroid/sclera complex were visualized and digital images were obtained using a confocal microscope (Zeiss LSM 510; Carl Zeiss Microscopy, Jena, Germany) equipped with argon and helium-neon lasers (Carl Zeiss Microscopy). Morphometric analyses on the choroid/sclera complex were conducted with Java-based imaging software (ImageJ, in the public domain at http://rsb.info.nih.gov/ij; National Institutes of Health [NIH], Bethesda, MD, USA) or by commercial image browser (LSM Image Browser; Carl Zeiss Microscopy). CD31+ CNV volumes were analyzed using the image analysis toolbox of commercial computing software (MATLAB 8.1; MathWorks, Natick, MA, USA), which are presented as μm3. The number of BM-derived macrophages was calculated by GFP and F4/80 copositive cells per 0.05 mm2 field of each site of laser injury. All parameters were measured with 6 to 10 mice per group. 
Western Blot
For Western blot analyses, the tissue containing retina, choroid, and sclera, or only choroidal lysates were separated by SDS-PAGE and transferred to polyvinylidene fluoride (PVDF) membranes. After blocking with 5% skim milk, the membrane was incubated overnight at 4°C with the following primary antibodies: rabbit anti-ZO-1 (Invitrogen); goat anti-VE-cadherin (VE-Cad; Santa Cruz Biotechnology, Inc., Dallas, TX, USA); goat anti-VEGF-A (R&D Systems); rabbit anti-β-actin (Santa Cruz Biotechnology, Inc.); rabbit anti-Ang1 (Santa Cruz Biotechnology, Inc.); and mouse anti-Flag (Sigma-Aldrich). Horseradish peroxidase (HRP)–conjugated secondary antibodies were used to capture primary antibody and the signals were developed with enhanced chemiluminescence HRP substrate (Millipore Corp.) and imaged with a luminescent image analyzer (LAS-1000 mini; Fujifilm, Tokyo, Japan). Quantitative densitometric measurements of signal intensities were performed using Java-based imaging software (NIH). 
Semiquantitative RT-PCR (sqRT-PCR) and Quantitative Real-Time PCR (rtPCR)
Total RNA was extracted from eyes using a commercial reagent (TRIzol Reagent; Invitrogen) according to the manufacturer's instructions. Two micrograms of the RNA was reverse-transcribed into cDNA using reverse transcriptase (SuperScript II Reverse Transcriptase; Invitrogen). sqRT-PCR was performed using PCR (Ex Taq; TaKaRa Bio, Inc., Otsu, Shiga, Japan) with the indicated primers (Supplementary Table S1). Quantitative rtPCR was performed with the indicated primers using real-time PCR detection system (Bio-Rad TM CFX96; Bio-Rad Laboratories, Inc., Hercules, CA, USA). The rtPCR data were analyzed with amplification software (Bio-Rad CFX Manager; Bio-Rad Laboratories, Inc.). 
Statistical Analysis
Values are presented as mean ± SD. Significant differences between means were determined by unpaired Student's t-test. Statistical significance was set at P < 0.05. 
Results
Ang1 Suppresses CNV Formation and Vascular Leakage in the Mouse CNV Model
To investigate the potential therapeutic effects of Ang1 on neovascular AMD, we compared the effects of COMP-Ang1 (hereafter “Ang1”) with VEGF-Trap (Aflibercept) in a laser-induced CNV mouse model that mimics human neovascular AMD. 26 We performed laser photocoagulation on mouse retinas to disrupt the retinal pigment epithelium and Bruch's membrane. As demonstrated by FA and ICGA, the CNV mouse model exhibited leaky new vessels around the site of laser photocoagulation (Fig. 1A). Consistent with a previous report, 29 intravitreal injection of VEGF-Trap (5 μg) effectively suppressed CNV formation (55.5% reduction) and vascular leakage (32.2% reduction; Figs. 1A–C). While intravitreal supplementation of Ang1 (5 μg) was as effective as VEGF-Trap in CNV inhibition (58.6% reduction), Ang1 suppressed vascular leakage (49.1% reduction) to a superior degree compared with VEGF-Trap (32.2% reduction; Figs. 1A–C). Thus, supplemental Ang1 exerts antileakage and anti-angiogenic roles in the mouse CNV model. 
Figure 1
 
Ang1 inhibits CNV formation and vascular leakage. Laser-induced CNV mice were given 5 μg of Fc (control, C), VEGF-Trap (VT), or COMP-Ang1 (A1, hereafter “Ang1”) intravitreally at 24 hours after the laser injury. Two weeks later, FA and ICGA were performed and choroid/sclera complexes were harvested. (A) Late-phase (6 min) FA for detecting vascular leakages surrounding the site of laser injury, and ICGA and CD31 staining for measuring the extent of CNV. Intravitreal injection of Ang1 suppressed vascular leakage to a superior degree compared with VEGF-Trap. Scale bars: 100 μm. (B) Comparisons of CNV volume were calculated by the total measured CD31+ CNV volume. Each group, n = 9. *P < 0.05 versus control. (C) Comparisons of leaky area were calculated by the total measured hyperfluorescent areas in FA divided by the total measured CNV area in ICGA. Each group, n = 9. *P < 0.05 versus C. #P < 0.05 versus VT.
Figure 1
 
Ang1 inhibits CNV formation and vascular leakage. Laser-induced CNV mice were given 5 μg of Fc (control, C), VEGF-Trap (VT), or COMP-Ang1 (A1, hereafter “Ang1”) intravitreally at 24 hours after the laser injury. Two weeks later, FA and ICGA were performed and choroid/sclera complexes were harvested. (A) Late-phase (6 min) FA for detecting vascular leakages surrounding the site of laser injury, and ICGA and CD31 staining for measuring the extent of CNV. Intravitreal injection of Ang1 suppressed vascular leakage to a superior degree compared with VEGF-Trap. Scale bars: 100 μm. (B) Comparisons of CNV volume were calculated by the total measured CD31+ CNV volume. Each group, n = 9. *P < 0.05 versus control. (C) Comparisons of leaky area were calculated by the total measured hyperfluorescent areas in FA divided by the total measured CNV area in ICGA. Each group, n = 9. *P < 0.05 versus C. #P < 0.05 versus VT.
Ang1 Prevents Vascular Leakage by Increasing Inter-EC Junctional Proteins in the Mouse CNV Model
Compared with the VEGF-Trap-treated CNV, Ang1-treated CNV displayed less vascular leakage, although the sizes of CNV were not significantly different in both groups (Figs. 1A–C). These results led us to investigate the inter-endothelial cell (EC) junctions of CNV since Ang1-Tie2 signaling is directly involved in limiting EC permeability by regulating inter-EC junctional proteins, such as vascular endothelial (VE)-cadherin and ZO-1. 11,30 High-magnification images of CNV in control mice displayed faint expressions of ZO-1 and VE-cadherin along the EC junctions, indicating the disruption of adherens and tight junctions due to the laser-induced choroidal inflammation (Fig. 2A). On the other hand, the CNV of mice treated with VEGF-Trap exhibited an increased expression and reassembly of ZO-1 and VE-cadherin in the cell-to-cell junctions, which was consistent with previous reports demonstrating the actions of VEGF-A on endothelial permeability through junctional proteins disassembly 3134 (Fig. 2A). Interestingly, Ang1-treated mice displayed markedly increased expression and distributions of both junctional proteins in the ECs of CNV, compared with that of control and VEGF-Trap–treated mice (Fig. 2A). These results matched the results of Western blot analysis of ZO-1 and VE-cadherin proteins of the choroidal lysate (Figs. 2B–D). The present data demonstrate that Ang1 increases the protein expressions and distributions of ZO-1 and VE-cadherin in the EC of the CNV. 
Figure 2
 
Ang1 increases inter-endothelial cell junctional proteins in CNV. Laser-induced CNV mice were given 5 μg of Fc (C), VT, or A1 intravitreally at 24 hours after laser injury. Two weeks later, choroid/sclera complexes were harvested for immunostaining and choroidal tissues was lysated for Western blot. (A) ZO-1 and VE-Cad expressions in CD31+ blood vessels of the CNV. Scale bars: 10 μm. (B) Western blot of ZO-1 and VE-cadherin in choroidal lysate. (C, D) Densitometric analysis of relative levels of ZO-1 and VE-cadherin, where the value of C is regarded as 1. Each group, n = 8. *P < 0.05 versus C. #P < 0.05 versus VT.
Figure 2
 
Ang1 increases inter-endothelial cell junctional proteins in CNV. Laser-induced CNV mice were given 5 μg of Fc (C), VT, or A1 intravitreally at 24 hours after laser injury. Two weeks later, choroid/sclera complexes were harvested for immunostaining and choroidal tissues was lysated for Western blot. (A) ZO-1 and VE-Cad expressions in CD31+ blood vessels of the CNV. Scale bars: 10 μm. (B) Western blot of ZO-1 and VE-cadherin in choroidal lysate. (C, D) Densitometric analysis of relative levels of ZO-1 and VE-cadherin, where the value of C is regarded as 1. Each group, n = 8. *P < 0.05 versus C. #P < 0.05 versus VT.
Genetic Deletion of Ang1 Exacerbates, While Overexpression of Ang1 Inhibits CNV and Vascular Leakage in the Mouse CNV Model
To investigate the role of endogenous or supplemental Ang1 in CNV formation and vascular leakage, genetically modified Ang1-deficient and -overexpressed mice were utilized. In order to globally deplete Ang1 expression, a transgenic mouse (ROSA26-Cre ERT2/Ang1 flox/flox) 16 designed to be under the control of the ROSA26 promoter and tamoxifen was used. We confirmed a significant deletion of Ang1 in the tissue containing retina, choroid, and sclera after tamoxifen injection by Western blots and semiquantitative RT-PCR (Supplementary Figs. S1A, S1B). When compared with littermate controls (ROSA26-Cre ERT2/Ang1 +/+), Ang1-depleted mice (Ang1DE) displayed increased CNV volume (27.9%) and vascular leakage (40.5%; Figs. 3A–C). To globally overexpress Ang1, a transgenic mouse (ROSA26-Cre ERT2/COMP-Ang1-Tg) 24 designed to be under the control of the ROSA26 promoter and tamoxifen was used. We confirmed a significant overexpression of Ang1 by Western blots and verifying the enlarged tracheal blood vessels (Supplementary Figs. S1C, S1D) as previously described. 35 When compared with controls (tamoxifen-treated transgenic mice that possessed only either the ROSA26-Cre ERT2 or COMP-Ang1 allele), Ang1-overexpressing mice (Ang1OE) displayed decreased CNV volume (46.2%) and vascular leakage (35.7%; Figs. 3A–C). Thus, endogenous and supplemental Ang1 indeed plays anti-leakage and anti-angiogenic roles in the mouse CNV model. 
Figure 3
 
Genetic deletion of Ang1 exacerbates, while overexpression of Ang1 inhibits CNV formation and vascular leakage. Eight-week-old ROSA26-Cre ERT2/COMP-Ang1-Tg mice, ROSA26-Cre ERT2/Ang1 flox/flox mice, and their littermate controls, designated as Ang1OE (AOE), Ang1DE (ADE), and Ang1+/+ (A+), respectively, were given tamoxifen 3 times every other day, and laser photocoagulation was performed 1 week after the last tamoxifen treatment. Two weeks after laser photocoagulation, FA and ICGA were performed and choroid/sclera complexes were harvested. (A) Late-phase (6 min) FA for detecting vascular leakages surrounding the laser spot, and ICGA and CD31 staining for measuring extent of CNV. Scale bars: 100 μm. (B) Comparisons of CNV volume were calculated by the total measured CD31+ CNV volume. Each group, n = 8. *P < 0.05 versus A+. (C) Comparisons of leaky area were calculated by the total measured hyperfluorescent areas in FA divided by the total measured CNV area in ICGA. Each group, n = 8. *P < 0.05 versus A+.
Figure 3
 
Genetic deletion of Ang1 exacerbates, while overexpression of Ang1 inhibits CNV formation and vascular leakage. Eight-week-old ROSA26-Cre ERT2/COMP-Ang1-Tg mice, ROSA26-Cre ERT2/Ang1 flox/flox mice, and their littermate controls, designated as Ang1OE (AOE), Ang1DE (ADE), and Ang1+/+ (A+), respectively, were given tamoxifen 3 times every other day, and laser photocoagulation was performed 1 week after the last tamoxifen treatment. Two weeks after laser photocoagulation, FA and ICGA were performed and choroid/sclera complexes were harvested. (A) Late-phase (6 min) FA for detecting vascular leakages surrounding the laser spot, and ICGA and CD31 staining for measuring extent of CNV. Scale bars: 100 μm. (B) Comparisons of CNV volume were calculated by the total measured CD31+ CNV volume. Each group, n = 8. *P < 0.05 versus A+. (C) Comparisons of leaky area were calculated by the total measured hyperfluorescent areas in FA divided by the total measured CNV area in ICGA. Each group, n = 8. *P < 0.05 versus A+.
Ang1 Reduces the Recruitment and Infiltration of VEGF-A–Producing Macrophages From BM in the Mouse CNV Model
The aforementioned results encouraged us to explore the underlying mechanism behind the Ang1-induced anti-leakage and anti-angiogenic effect in CNV model. We examined the recruitment and infiltration of macrophages derived from the BM, since BM-derived macrophages are known as a critical cellular component initiating CNV formation and vascular leakage by increasing VEGF-A level. 36,37 To elucidate the role of Ang1 on the recruitment of BM-derived macrophages, we generated laser-induced CNV models in Ang1 genetically modified mice that previously received BMT from GFP + transgenic mice. Because macrophages invade the site of laser injury within 1 day, with a peak response for 3 days followed by a rapid disappearance from day 5 to 7, 36 we analyzed the distribution of BM-derived macrophages at the site of laser injury 3 days after laser photocoagulation to minimize secondary macrophage infiltrations derived from the subsequent vascular leakage after CNV inductions. When compared with controls (Ang1 +/+ BMT), Ang1-depleted BMT mice (Ang1DE BMT) displayed 103.1% increased recruitment of BM-derived GFP+ F4/80+ macrophages (Figs. 4A, 4B). In contrast, Ang1-overexpressing BMT mice (Ang1OE BMT) displayed 56.1% decreased GFP+ F4/80+ macrophages at the site of laser injury compared with controls (Figs. 4A, 4B). Intriguingly, the macrophages of Ang1DE BMT mice expressed more VEGF-A compared with control mice (Ang1+/+ BMT; Fig. 4C). In contrast, the macrophages of Ang1OE BMT mice expressed less VEGF-A compared with control mice (Fig. 4C). In agreement with these observations, Western blot analysis of choroidal lysate revealed that total VEGF-A protein level of Ang1DE mice was higher than control Ang1+/+ mice, while the VEGF-A level of Ang1OE mice was lower than control mice (Fig. 4D). Thus, Ang1 suppresses CNV formation and vascular leakage by inhibiting the recruitment and infiltration of VEGF-A expressing macrophages from the BM in the mouse CNV model. 
Figure 4
 
Ang1 inhibits the recruitment and infiltration of VEGF-A–producing macrophages from bone marrow. Eight-week-old ROSA26-Cre ERT2/COMP-Ang1-Tg mice, ROSA26-Cre ERT2/Ang1 flox/flox mice, and their littermate controls, designated as AOE, ADE, and A+, respectively, were sublethally irradiated and received BMT from GFP + transgenic mice. Two months later, they were given tamoxifen three times every other day, and laser photocoagulation was performed 1 week after the last tamoxifen treatment. Three days after laser photocoagulation, choroid/sclera complexes were harvested. (A) Bone marrow–derived GFP+ cells and F4/80+ macrophages are clustered at the site of laser injury. Scale bars: 50 μm. (B) Comparisons of the number of GFP+ F4/80+ cells. Each group, n = 6. *P < 0.05 versus A+. (C) More abundant distribution of VEGF-A+ F4/80+ cells (arrowheads) around the CNV in Ang1DE compared with the control group (Ang1+/+). (D) Western blot of VEGF-A from choroidal lysate from each group.
Figure 4
 
Ang1 inhibits the recruitment and infiltration of VEGF-A–producing macrophages from bone marrow. Eight-week-old ROSA26-Cre ERT2/COMP-Ang1-Tg mice, ROSA26-Cre ERT2/Ang1 flox/flox mice, and their littermate controls, designated as AOE, ADE, and A+, respectively, were sublethally irradiated and received BMT from GFP + transgenic mice. Two months later, they were given tamoxifen three times every other day, and laser photocoagulation was performed 1 week after the last tamoxifen treatment. Three days after laser photocoagulation, choroid/sclera complexes were harvested. (A) Bone marrow–derived GFP+ cells and F4/80+ macrophages are clustered at the site of laser injury. Scale bars: 50 μm. (B) Comparisons of the number of GFP+ F4/80+ cells. Each group, n = 6. *P < 0.05 versus A+. (C) More abundant distribution of VEGF-A+ F4/80+ cells (arrowheads) around the CNV in Ang1DE compared with the control group (Ang1+/+). (D) Western blot of VEGF-A from choroidal lysate from each group.
Ang1 Reduces Inflammatory Cytokines and Adhesion Molecules in the Mouse CNV Model
To investigate how Ang1 can inhibit the macrophage recruitment from the BM to the site of laser injury, we examined the expression of Tie2, the corresponding receptor for Ang1, using Tie2-GFP transgenic mice. Tie2 was strongly expressed on the infiltrated macrophages around the site of laser injury as well as on the ECs of CNV (Fig. 5A). Combining with the previous report 38 demonstrating that Ang1/Tie2 system directly inhibits the activation and migration of Tie2-expressing macrophages under inflammatory conditions, our findings suggest that Ang1 may inhibit the migration and infiltration of Tie2-expressing macrophages into CNV lesions. Furthermore, quantitative rtPCR analyses of the tissue containing retina, choroid, and sclera from genetically modified Ang1-deficient and -overexpressed mice revealed that Ang1OE mice present decreased expressions of inflammatory cytokines SDF-1, MCP-1, and IL-6, and adhesion molecules ICAM-1 and VCAM-1, compared with controls (Ang1+/+; Fig. 5B). In contrast, Ang1DE mice displayed increased expressions of these inflammatory adhesion molecules and cytokines (Fig. 5B). These findings suggest that endogenous and supplemental Ang1 plays anti-inflammatory actions by reducing inflammatory adhesion molecules and cytokines in the mouse CNV model. 
Figure 5
 
Ang1 reduces inflammatory cytokines and adhesion molecules in Tie2-expressing CNV. (A) Tie2 expression in CNV lesions displayed using Tie2-GFP transgenic mice 2 weeks after laser photocoagulation. CD31+ endothelial cells (arrowheads) and F4/80+ macrophages (arrows) express Tie2. Scale bars: 50 μm. (B) Three days after laser photocoagulation, the tissue containing retina, choroid, and sclera of AOE, ADE, and A+ were harvested. Total RNAs were extracted from the choroid/sclera complexes, and mRNA expressions of each gene were analyzed by rtPCR. GAPDH, internal control. The mRNA level of each gene of A+ was arbitrarily regarded as 1. Each group, n = 5, *P < 0.05 versus A+.
Figure 5
 
Ang1 reduces inflammatory cytokines and adhesion molecules in Tie2-expressing CNV. (A) Tie2 expression in CNV lesions displayed using Tie2-GFP transgenic mice 2 weeks after laser photocoagulation. CD31+ endothelial cells (arrowheads) and F4/80+ macrophages (arrows) express Tie2. Scale bars: 50 μm. (B) Three days after laser photocoagulation, the tissue containing retina, choroid, and sclera of AOE, ADE, and A+ were harvested. Total RNAs were extracted from the choroid/sclera complexes, and mRNA expressions of each gene were analyzed by rtPCR. GAPDH, internal control. The mRNA level of each gene of A+ was arbitrarily regarded as 1. Each group, n = 5, *P < 0.05 versus A+.
Discussion
Here, we demonstrated that supplemental Ang1 inhibits CNV formation and vascular leakage in the experimental laser-induced CNV model using genetically modified mice as well as intravitreal injections of recombinant proteins. During the initial inflammatory period after laser-induced RPE and Bruch's membrane destruction, Ang1 reduces the recruitment and infiltration of macrophages from the BM to the injury site. Because these BM-derived macrophages are known to be the major sources of angiogenic growth factor VEGF-A for the CNV formation, 36,37 Ang1-induced inhibition of macrophage infiltrations leads to the prevention of CNV formation and the subsequent vascular leakage. Once CNV forms, Ang1 directly binds to endothelial Tie2 and it stabilizes the EC in CNV and decreases vascular leakage by increasing the inter-EC junctional proteins, which prevents the secondary inflammation from propagating (Fig. 6). These findings indicate that Ang1 can efficiently inhibit CNV formation as well as preventing related complications by breaking the vicious cycle of inflammatory CNV (Fig. 6). This implies that Ang1 supplementation can be established as a therapeutic strategy for neovascular AMD. 
Figure 6
 
Scheme depicting how Ang1 inhibits CNV and vascular leakage. Ang1 breaks the vicious cycle of inflammatory CNV. Bone marrow (BM)-derived macrophages (MØ) are the major sources of several pro-inflammatory factors including VEGF-A during the initiation of CNV. Ang1 suppresses inflammatory angiogenesis by inhibiting the recruitment and infiltration of MØ from the BM to the inflammatory lesion. Furthermore, Ang1 binds to Tie2-expressing EC in the already-established CNV and tightens EC junctions, which decrease vascular leakage and prevents the propagation of secondary inflammation.
Figure 6
 
Scheme depicting how Ang1 inhibits CNV and vascular leakage. Ang1 breaks the vicious cycle of inflammatory CNV. Bone marrow (BM)-derived macrophages (MØ) are the major sources of several pro-inflammatory factors including VEGF-A during the initiation of CNV. Ang1 suppresses inflammatory angiogenesis by inhibiting the recruitment and infiltration of MØ from the BM to the inflammatory lesion. Furthermore, Ang1 binds to Tie2-expressing EC in the already-established CNV and tightens EC junctions, which decrease vascular leakage and prevents the propagation of secondary inflammation.
Our study indicates that intravitreal administration of Ang1 upregulates the expression of VE-cadherin and ZO-1, the key components of endothelial cell-to-cell junctions maintaining vascular integrity. Although both Ang1 and VEGF-A are essential for vessel assembly and normal EC functions, they exert opposing effects on vascular integrity. 11 Previous studies revealed that VEGF-A causes endothelial permeability by disrupting intercellular junctions through the phosphorylation, endocytosis, and degradation of VE-cadherin and ZO-1. 33,34 In contrast, Ang1 prevents the VEGF-A–mediated junctional disruption by interfering with VEGF receptor signaling pathways, thereby rescuing the endothelial barrier function. 39 Furthermore, one recent report suggests that Ang1 does not merely inhibit the VEGF-mediated junctional disassembly, but also it directly stabilizes VE-cadherin and ZO-1 by regulating the RhoA-specific guanine nucleotide exchange factor Syx. 40 Thus, these previous studies support our findings that demonstrate that Ang1 is more effective than VEGF-trap in the maintenance and restoration of the endothelial integrity, and thus suppress the vascular leakage from CNV. 
While anti–VEGF-A drugs are widely used as a primary therapy for neovascular AMD, several controversies exist in the use of the anti–VEGF-A drugs. 1 One of the issues is that some sets of patients with AMD show refractoriness or tachyphylaxis against anti–VEGF-A antibodies. 79 One possible mechanism for tachyphylaxis was suggested to be that infiltrated macrophages within CNV membrane may upregulate VEGF-A production as a compensatory response to intravitreal VEGF-A blockade. 9 In addition, the majority (>90%) of macrophages infiltrated within the CNV were recruited from the BM. 37 Moreover, one previous study reported that depletion of these macrophages using liposomal clodronate reduced the size and leakage of the CNV, which was accompanied by decreased VEGF-A level. 36 All these previous reports indicate that BM-derived macrophage is a critical cellular component in CNV formation and vascular leakage as well as tachyphylaxis against anti–VEGF-A antibodies. Based on our observations that Ang1 inhibits the recruitment and infiltration of VEGF-A-producing macrophages from the BM, Ang1 would be beneficial to overcome the potential adverse effects of anti–VEGF-A drugs. Therefore, further study is necessary to demonstrate the applicability of Ang1 as an alternative or combinative strategy in these tachyphylactic cases. 
Although previous reports 2022 have already described Ang1 as a potential therapeutic strategy in inhibiting CNV and related vascular leakage, no work has been done so far to unravel the functional relevance and the underlying mechanisms of how Ang1 inhibits CNV. Nambu et al. 21 reported that photoreceptor-specific Ang1 overexpression at the seventh day after laser injury exhibited a preventative effect for the additional growth of CNV, while it did not regress the already-established CNV. It has been reported that macrophages invade the site of laser injury within 24 hours, with a peak response for 3 days, followed by rapid disappearance at the laser injury site from day 5 to 7. 36 Therefore, Ang1-induced CNV stabilization in the experiment by Nambu et al. is thought to be derived mainly from the Ang1-induced anti-leakage effect on EC rather than from the Ang1-induced anti-inflammatory effect on macrophages, because the starting point of Ang1 overexpression in their experiments were at the seventh day after laser injury, after the highly inflammatory period. 21 Nevertheless, it is notable that Ang1 suppressed the further progression of CNV efficiently during the following 7 days. 21 These findings suggest that supplemental Ang1 could be beneficial for the treatment of atypical CNV that is derived from low- or non-inflammatory disease entities as well, such as pathologic myopia, choroidal ruptures, and angioid streaks. 
The present study using genetically Ang1-depleted mice revealed that endogenous Ang1 has a substantial role in inhibiting CNV. However, the precise information about the source of Ang1 during CNV formation is not yet clearly demonstrated. Although a previous study 41 reported that Ang1 was detected in the stromal cells of human CNV membrane, we used the ROSA26-Cre ERT2 driver line to globally deplete the Ang1 allele because Ang1 is known to be constitutively expressed in several types of non-EC as well, such as perivascular smooth muscle cells or mesenchymal cells. 11 Moreover, Hangai et al. 42 reported that Ang1 was also expressed in cultured RPE cells. Intriguingly, they demonstrated that RPE-derived Ang1 was upregulated upon VEGF-A stimulation. In the present study, we found that Ang1 downregulates VEGF-A by inhibiting the recruitment of inflammatory macrophages into the CNV area. Collectively, if VEGF-A is upregulated at the RPE/choroid complex in the inflammatory condition such as neovascular AMD, stimulated RPE-derived Ang1 would downregulate VEGF-A as a negative feedback mechanism, which contributes to the prevention of further pathologic angiogenesis and vascular leakage. Consequently, endogenous Ang1 is a physiologic negative regulator inhibiting the propagation of inflammation and pathologic angiogenesis. 
In summary, the present study provides new observations demonstrating that endogenous and supplemental Ang1 can suppress CNV formation and related vascular leakage by inhibiting the recruitment and infiltration of VEGF-A-producing macrophages from the BM and by increasing the inter-EC junctional protein. This implies that Ang1 supplementation can be established as a therapeutic strategy for neovascular AMD. 
Supplementary Materials
Acknowledgments
Supported by Grant R2011-0019268 (GYK) from the Ministry of Science, ICT and Future Planning, Korea. 
Disclosure: J. Lee, None; D.-Y. Park, None; D.Y. Park, None; I. Park, None; W. Chang, None; Y. Nakaoka, None; I. Komuro, None; O.-J. Yoo, None; G.Y. Koh, None 
References
Lim LS Mitchell P Seddon JM Holz FG Wong TY. Age-related macular degeneration. Lancet . 2012; 379: 1728–1738. [CrossRef] [PubMed]
Holz FG Pauleikhoff D Klein R Bird AC. Pathogenesis of lesions in late age-related macular disease. Am J Ophthalmol . 2004; 137: 504–510. [CrossRef] [PubMed]
Jager RD Mieler WF Miller JW. Age-related macular degeneration. N Engl J Med . 2008; 358: 2606–2617. [CrossRef] [PubMed]
Heier JS Brown DM Chong V Intravitreal Aflibercept (VEGF Trap-Eye) in Wet Age-related Macular Degeneration. Ophthalmology . 2012; 119: 2537–2548. [CrossRef] [PubMed]
Martin DF Maguire MG Fine SL Ranibizumab and bevacizumab for treatment of neovascular age-related macular degeneration: two-year results. Ophthalmology . 2012; 119: 1388–1398. [CrossRef] [PubMed]
Tolentino M. Systemic and ocular safety of intravitreal anti-VEGF therapies for ocular neovascular disease. Surv Ophthalmol . 2011; 56: 95–113. [CrossRef] [PubMed]
Lux A Llacer H Heussen FM Joussen AM. Non-responders to bevacizumab (Avastin) therapy of choroidal neovascular lesions. Br J Ophthalmol . 2007; 91: 1318–1322. [CrossRef] [PubMed]
Schaal S Kaplan HJ Tezel TH. Is there tachyphylaxis to intravitreal anti-vascular endothelial growth factor pharmacotherapy in age-related macular degeneration? Ophthalmology . 2008; 115: 2199–2205. [CrossRef] [PubMed]
Forooghian F Cukras C Meyerle CB Chew EY Wong WT. Tachyphylaxis after intravitreal bevacizumab for exudative age-related macular degeneration. Retina . 2009; 29: 723–731. [CrossRef] [PubMed]
Tatar O Yoeruek E Szurman P Effect of bevacizumab on inflammation and proliferation in human choroidal neovascularization. Arch Ophthalmol . 2008; 126: 782–790. [CrossRef] [PubMed]
Augustin HG Koh GY Thurston G Alitalo K. Control of vascular morphogenesis and homeostasis through the angiopoietin-Tie system. Nat Rev Mol Cell Biol . 2009; 10: 165–177. [CrossRef] [PubMed]
Yancopoulos GD Davis S Gale NW Rudge JS Wiegand SJ Holash J. Vascular-specific growth factors and blood vessel formation. Nature . 2000; 407: 242–248. [CrossRef] [PubMed]
Thurston G Rudge JS Ioffe E Angiopoietin-1 protects the adult vasculature against plasma leakage. Nat Med . 2000; 6: 460–463. [CrossRef] [PubMed]
Cho CH Kim KE Byun J Long-term and sustained COMP-Ang1 induces long-lasting vascular enlargement and enhanced blood flow. Circ Res . 2005; 97: 86–94. [CrossRef] [PubMed]
Uemura A Ogawa M Hirashima M Recombinant angiopoietin-1 restores higher-order architecture of growing blood vessels in mice in the absence of mural cells. J Clin Invest . 2002; 110: 1619–1628. [CrossRef] [PubMed]
Lee J Kim KE Choi DK Angiopoietin-1 guides directional angiogenesis through integrin alphavbeta5 signaling for recovery of ischemic retinopathy. Sci Transl Med . 2013; 5: 203ra127.
Koh GY. Orchestral actions of angiopoietin-1 in vascular regeneration. Trends Mol Med . 2013; 19: 31–39. [CrossRef] [PubMed]
Gamble JR Drew J Trezise L Angiopoietin-1 is an antipermeability and anti-inflammatory agent in vitro and targets cell junctions. Circ Res . 2000; 87: 603–607. [CrossRef] [PubMed]
Kim I Moon SO Park SK Chae SW Koh GY. Angiopoietin-1 reduces VEGF-stimulated leukocyte adhesion to endothelial cells by reducing ICAM-1, VCAM-1, and E-selectin expression. Circ Res . 2001; 89: 477–479. [CrossRef] [PubMed]
Nambu H Nambu R Oshima Y Angiopoietin 1 inhibits ocular neovascularization and breakdown of the blood-retinal barrier. Gene Ther . 2004; 11: 865–873. [CrossRef] [PubMed]
Nambu H Umeda N Kachi S Angiopoietin 1 prevents retinal detachment in an aggressive model of proliferative retinopathy, but has no effect on established neovascularization. J Cell Physiol . 2005; 204: 227–235. [CrossRef] [PubMed]
Wang Y Bi H Teng D Potential protective effect of angiopoietin-1 on the leakage of rat choroidal neovascularization. Saudi Med J . 2013; 34: 584–590. [PubMed]
Cho CH Kammerer RA Lee HJ Designed angiopoietin-1 variant, COMP-Ang1, protects against radiation-induced endothelial cell apoptosis. Proc Natl Acad Sci U S A . 2004; 101: 5553–5558. [CrossRef] [PubMed]
Hato T Kimura Y Morisada T Angiopoietins contribute to lung development by regulating pulmonary vascular network formation. Biochem Biophys Res Commun . 2009; 381: 218–223. [CrossRef] [PubMed]
Soriano P. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet . 1999; 21: 70–71. [CrossRef] [PubMed]
Tobe T Ortega S Luna JD Targeted disruption of the FGF2 gene does not prevent choroidal neovascularization in a murine model. Am J Pathol . 1998; 153: 1641–1646. [CrossRef] [PubMed]
Koh YJ Kim HZ Hwang SI Double antiangiogenic protein, DAAP, targeting VEGF-A and angiopoietins in tumor angiogenesis, metastasis, and vascular leakage. Cancer Cell . 2010; 18: 171–184. [CrossRef] [PubMed]
Campos M Amaral J Becerra SP Fariss RN. A novel imaging technique for experimental choroidal neovascularization. Invest Ophthalmol Vis Sci . 2006; 47: 5163–5170. [CrossRef] [PubMed]
Saishin Y Takahashi K Lima e Silva R VEGF-TRAP(R1R2) suppresses choroidal neovascularization and VEGF-induced breakdown of the blood-retinal barrier. J Cell Physiol . 2003; 195: 241–248. [CrossRef] [PubMed]
Yu H Wang P An P Xue Y. Recombinant human angiopoietin-1 ameliorates the expressions of ZO-1, occludin, VE-cadherin, and PKCalpha signaling after focal cerebral ischemia/reperfusion in rats. J Mol Neurosci . 2012; 46: 236–247. [CrossRef] [PubMed]
Gavard J Gutkind JS. VEGF controls endothelial-cell permeability by promoting the beta-arrestin-dependent endocytosis of VE-cadherin. Nat Cell Biol . 2006; 8: 1223–1234. [CrossRef] [PubMed]
Wang W Dentler WL Borchardt RT. VEGF increases BMEC monolayer permeability by affecting occludin expression and tight junction assembly. Am J Physiol Heart Circ Physiol . 2001; 280: H434–H440. [PubMed]
Murakami T Felinski EA Antonetti DA. Occludin phosphorylation and ubiquitination regulate tight junction trafficking and vascular endothelial growth factor-induced permeability. J Biol Chem . 2009; 284: 21036–21046. [CrossRef] [PubMed]
Giannotta M Trani M Dejana E. VE-cadherin and endothelial adherens junctions: active guardians of vascular integrity. Dev Cell . 2013; 26: 441–454. [CrossRef] [PubMed]
Kim KE Cho CH Kim HZ Baluk P McDonald DM Koh GY. In vivo actions of angiopoietins on quiescent and remodeling blood and lymphatic vessels in mouse airways and skin. Arterioscler Thromb Vasc Biol . 2007; 27: 564–570. [CrossRef] [PubMed]
Sakurai E Anand A Ambati BK van Rooijen N Ambati J. Macrophage depletion inhibits experimental choroidal neovascularization. Invest Ophthalmol Vis Sci . 2003; 44: 3578–3585. [CrossRef] [PubMed]
Espinosa-Heidmann DG Caicedo A Hernandez EP Csaky KG Cousins SW. Bone marrow-derived progenitor cells contribute to experimental choroidal neovascularization. Invest Ophthalmol Vis Sci . 2003; 44: 4914–4919. [CrossRef] [PubMed]
Gu H Cui M Bai Y Angiopoietin-1/Tie2 signaling pathway inhibits lipopolysaccharide-induced activation of RAW264.7 macrophage cells. Biochem Biophys Res Commun . 2010; 392: 178–182. [CrossRef] [PubMed]
Gavard J Patel V Gutkind JS. Angiopoietin-1 prevents VEGF-induced endothelial permeability by sequestering Src through mDia. Dev Cell . 2008; 14: 25–36. [CrossRef] [PubMed]
Ngok SP Geyer R Liu M VEGF and Angiopoietin-1 exert opposing effects on cell junctions by regulating the Rho GEF Syx. J Cell Biol . 2012; 199: 1103–1115. [CrossRef] [PubMed]
Otani A Takagi H Oh H Koyama S Matsumura M Honda Y. Expressions of angiopoietins and Tie2 in human choroidal neovascular membranes. Invest Ophthalmol Vis Sci . 1999; 40: 1912–1920. [PubMed]
Hangai M Murata T Miyawaki N Angiopoietin-1 upregulation by vascular endothelial growth factor in human retinal pigment epithelial cells. Invest Ophthalmol Vis Sci . 2001; 42: 1617–1625. [PubMed]
Figure 1
 
Ang1 inhibits CNV formation and vascular leakage. Laser-induced CNV mice were given 5 μg of Fc (control, C), VEGF-Trap (VT), or COMP-Ang1 (A1, hereafter “Ang1”) intravitreally at 24 hours after the laser injury. Two weeks later, FA and ICGA were performed and choroid/sclera complexes were harvested. (A) Late-phase (6 min) FA for detecting vascular leakages surrounding the site of laser injury, and ICGA and CD31 staining for measuring the extent of CNV. Intravitreal injection of Ang1 suppressed vascular leakage to a superior degree compared with VEGF-Trap. Scale bars: 100 μm. (B) Comparisons of CNV volume were calculated by the total measured CD31+ CNV volume. Each group, n = 9. *P < 0.05 versus control. (C) Comparisons of leaky area were calculated by the total measured hyperfluorescent areas in FA divided by the total measured CNV area in ICGA. Each group, n = 9. *P < 0.05 versus C. #P < 0.05 versus VT.
Figure 1
 
Ang1 inhibits CNV formation and vascular leakage. Laser-induced CNV mice were given 5 μg of Fc (control, C), VEGF-Trap (VT), or COMP-Ang1 (A1, hereafter “Ang1”) intravitreally at 24 hours after the laser injury. Two weeks later, FA and ICGA were performed and choroid/sclera complexes were harvested. (A) Late-phase (6 min) FA for detecting vascular leakages surrounding the site of laser injury, and ICGA and CD31 staining for measuring the extent of CNV. Intravitreal injection of Ang1 suppressed vascular leakage to a superior degree compared with VEGF-Trap. Scale bars: 100 μm. (B) Comparisons of CNV volume were calculated by the total measured CD31+ CNV volume. Each group, n = 9. *P < 0.05 versus control. (C) Comparisons of leaky area were calculated by the total measured hyperfluorescent areas in FA divided by the total measured CNV area in ICGA. Each group, n = 9. *P < 0.05 versus C. #P < 0.05 versus VT.
Figure 2
 
Ang1 increases inter-endothelial cell junctional proteins in CNV. Laser-induced CNV mice were given 5 μg of Fc (C), VT, or A1 intravitreally at 24 hours after laser injury. Two weeks later, choroid/sclera complexes were harvested for immunostaining and choroidal tissues was lysated for Western blot. (A) ZO-1 and VE-Cad expressions in CD31+ blood vessels of the CNV. Scale bars: 10 μm. (B) Western blot of ZO-1 and VE-cadherin in choroidal lysate. (C, D) Densitometric analysis of relative levels of ZO-1 and VE-cadherin, where the value of C is regarded as 1. Each group, n = 8. *P < 0.05 versus C. #P < 0.05 versus VT.
Figure 2
 
Ang1 increases inter-endothelial cell junctional proteins in CNV. Laser-induced CNV mice were given 5 μg of Fc (C), VT, or A1 intravitreally at 24 hours after laser injury. Two weeks later, choroid/sclera complexes were harvested for immunostaining and choroidal tissues was lysated for Western blot. (A) ZO-1 and VE-Cad expressions in CD31+ blood vessels of the CNV. Scale bars: 10 μm. (B) Western blot of ZO-1 and VE-cadherin in choroidal lysate. (C, D) Densitometric analysis of relative levels of ZO-1 and VE-cadherin, where the value of C is regarded as 1. Each group, n = 8. *P < 0.05 versus C. #P < 0.05 versus VT.
Figure 3
 
Genetic deletion of Ang1 exacerbates, while overexpression of Ang1 inhibits CNV formation and vascular leakage. Eight-week-old ROSA26-Cre ERT2/COMP-Ang1-Tg mice, ROSA26-Cre ERT2/Ang1 flox/flox mice, and their littermate controls, designated as Ang1OE (AOE), Ang1DE (ADE), and Ang1+/+ (A+), respectively, were given tamoxifen 3 times every other day, and laser photocoagulation was performed 1 week after the last tamoxifen treatment. Two weeks after laser photocoagulation, FA and ICGA were performed and choroid/sclera complexes were harvested. (A) Late-phase (6 min) FA for detecting vascular leakages surrounding the laser spot, and ICGA and CD31 staining for measuring extent of CNV. Scale bars: 100 μm. (B) Comparisons of CNV volume were calculated by the total measured CD31+ CNV volume. Each group, n = 8. *P < 0.05 versus A+. (C) Comparisons of leaky area were calculated by the total measured hyperfluorescent areas in FA divided by the total measured CNV area in ICGA. Each group, n = 8. *P < 0.05 versus A+.
Figure 3
 
Genetic deletion of Ang1 exacerbates, while overexpression of Ang1 inhibits CNV formation and vascular leakage. Eight-week-old ROSA26-Cre ERT2/COMP-Ang1-Tg mice, ROSA26-Cre ERT2/Ang1 flox/flox mice, and their littermate controls, designated as Ang1OE (AOE), Ang1DE (ADE), and Ang1+/+ (A+), respectively, were given tamoxifen 3 times every other day, and laser photocoagulation was performed 1 week after the last tamoxifen treatment. Two weeks after laser photocoagulation, FA and ICGA were performed and choroid/sclera complexes were harvested. (A) Late-phase (6 min) FA for detecting vascular leakages surrounding the laser spot, and ICGA and CD31 staining for measuring extent of CNV. Scale bars: 100 μm. (B) Comparisons of CNV volume were calculated by the total measured CD31+ CNV volume. Each group, n = 8. *P < 0.05 versus A+. (C) Comparisons of leaky area were calculated by the total measured hyperfluorescent areas in FA divided by the total measured CNV area in ICGA. Each group, n = 8. *P < 0.05 versus A+.
Figure 4
 
Ang1 inhibits the recruitment and infiltration of VEGF-A–producing macrophages from bone marrow. Eight-week-old ROSA26-Cre ERT2/COMP-Ang1-Tg mice, ROSA26-Cre ERT2/Ang1 flox/flox mice, and their littermate controls, designated as AOE, ADE, and A+, respectively, were sublethally irradiated and received BMT from GFP + transgenic mice. Two months later, they were given tamoxifen three times every other day, and laser photocoagulation was performed 1 week after the last tamoxifen treatment. Three days after laser photocoagulation, choroid/sclera complexes were harvested. (A) Bone marrow–derived GFP+ cells and F4/80+ macrophages are clustered at the site of laser injury. Scale bars: 50 μm. (B) Comparisons of the number of GFP+ F4/80+ cells. Each group, n = 6. *P < 0.05 versus A+. (C) More abundant distribution of VEGF-A+ F4/80+ cells (arrowheads) around the CNV in Ang1DE compared with the control group (Ang1+/+). (D) Western blot of VEGF-A from choroidal lysate from each group.
Figure 4
 
Ang1 inhibits the recruitment and infiltration of VEGF-A–producing macrophages from bone marrow. Eight-week-old ROSA26-Cre ERT2/COMP-Ang1-Tg mice, ROSA26-Cre ERT2/Ang1 flox/flox mice, and their littermate controls, designated as AOE, ADE, and A+, respectively, were sublethally irradiated and received BMT from GFP + transgenic mice. Two months later, they were given tamoxifen three times every other day, and laser photocoagulation was performed 1 week after the last tamoxifen treatment. Three days after laser photocoagulation, choroid/sclera complexes were harvested. (A) Bone marrow–derived GFP+ cells and F4/80+ macrophages are clustered at the site of laser injury. Scale bars: 50 μm. (B) Comparisons of the number of GFP+ F4/80+ cells. Each group, n = 6. *P < 0.05 versus A+. (C) More abundant distribution of VEGF-A+ F4/80+ cells (arrowheads) around the CNV in Ang1DE compared with the control group (Ang1+/+). (D) Western blot of VEGF-A from choroidal lysate from each group.
Figure 5
 
Ang1 reduces inflammatory cytokines and adhesion molecules in Tie2-expressing CNV. (A) Tie2 expression in CNV lesions displayed using Tie2-GFP transgenic mice 2 weeks after laser photocoagulation. CD31+ endothelial cells (arrowheads) and F4/80+ macrophages (arrows) express Tie2. Scale bars: 50 μm. (B) Three days after laser photocoagulation, the tissue containing retina, choroid, and sclera of AOE, ADE, and A+ were harvested. Total RNAs were extracted from the choroid/sclera complexes, and mRNA expressions of each gene were analyzed by rtPCR. GAPDH, internal control. The mRNA level of each gene of A+ was arbitrarily regarded as 1. Each group, n = 5, *P < 0.05 versus A+.
Figure 5
 
Ang1 reduces inflammatory cytokines and adhesion molecules in Tie2-expressing CNV. (A) Tie2 expression in CNV lesions displayed using Tie2-GFP transgenic mice 2 weeks after laser photocoagulation. CD31+ endothelial cells (arrowheads) and F4/80+ macrophages (arrows) express Tie2. Scale bars: 50 μm. (B) Three days after laser photocoagulation, the tissue containing retina, choroid, and sclera of AOE, ADE, and A+ were harvested. Total RNAs were extracted from the choroid/sclera complexes, and mRNA expressions of each gene were analyzed by rtPCR. GAPDH, internal control. The mRNA level of each gene of A+ was arbitrarily regarded as 1. Each group, n = 5, *P < 0.05 versus A+.
Figure 6
 
Scheme depicting how Ang1 inhibits CNV and vascular leakage. Ang1 breaks the vicious cycle of inflammatory CNV. Bone marrow (BM)-derived macrophages (MØ) are the major sources of several pro-inflammatory factors including VEGF-A during the initiation of CNV. Ang1 suppresses inflammatory angiogenesis by inhibiting the recruitment and infiltration of MØ from the BM to the inflammatory lesion. Furthermore, Ang1 binds to Tie2-expressing EC in the already-established CNV and tightens EC junctions, which decrease vascular leakage and prevents the propagation of secondary inflammation.
Figure 6
 
Scheme depicting how Ang1 inhibits CNV and vascular leakage. Ang1 breaks the vicious cycle of inflammatory CNV. Bone marrow (BM)-derived macrophages (MØ) are the major sources of several pro-inflammatory factors including VEGF-A during the initiation of CNV. Ang1 suppresses inflammatory angiogenesis by inhibiting the recruitment and infiltration of MØ from the BM to the inflammatory lesion. Furthermore, Ang1 binds to Tie2-expressing EC in the already-established CNV and tightens EC junctions, which decrease vascular leakage and prevents the propagation of secondary inflammation.
×
×

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.

×