Embed Size (px)
Citation preview
1
A2E, A PIGMENT OF THE LIPOFUSCIN OF RETINAL PIGMENT 1 EPITHELIAL CELLS, IS AN ENDOGENOUS LIGAND FOR 2
RETINOIC ACID RECEPTOR 3 4
Aya Iriyama1, Ryoji Fujiki2, Yuji Inoue1, Hidenori Takahashi1, Yasuhiro Tamaki1, Shinichiro 5 Takezawa2, Kenichi Takeyama2, Woo-Dong Jang3, Shigeaki Kato2, and Yasuo Yanagi1 6
From Department of Ophthalmology, University of Tokyo School of Medicine1, The Institute of 7 Molecular and Cellular Biosciences, University of Tokyo2, and Department of Chemistry, College 8
of Science, Yonsei University3 9 Running head: A2E is a ligand for RAR 10
Address correspondence to: Yasuo Yanagi Department of Ophthalmology, University of Tokyo 11 School of Medicine, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan. Tel: +81-3-5800-8660. 12 Fax: +81-3-3817-0798. E-mail: [email protected] 13 14 15 Lipofuscin contains the fluorophores, 16 which represent a biomarker for cellular 17 aging. Although it remains unsubstantiated 18 clinically, experimental results support that 19 the accumulation of lipofuscin is related to 20 an increased risk of choroidal 21 neovascularization (CNV) due to age-related 22 macular degeneration, a leading cause of 23 legal blindness. Here, we report that a major 24 lipofuscin component, A2E, activates the 25 retinoic acid receptor (RAR). In vitro 26 experiments using luciferase reporter assay, 27 competitional binding assay, analysis of 28 target genes and chromatin 29 immunoprecipitation (ChIP) assay strongly 30 suggest that A2E is a bona fide ligand for 31 RAR and induces sustained activation of 32 RAR target genes. A2E induced vascular 33 endothelial growth factor (VEGF) 34 expression in a human retinal pigment 35 epithelial cell line (ARPE-19) and RAR 36 antagonist blocked the upregulation of 37 VEGF. The conditioned medium of 38 A2E-treated ARPE-19 cells induced tube 39 formation in human umbilical vascular 40 endothelial cells (HUVEC), which was 41 blocked by the RAR antagonist and 42 anti-VEGF antibody. These results suggest 43 that A2E accumulation results in the 44 phenotypic alteration of retinal pigment 45 epithelial cells, predisposing the 46
environment to CNV development. This is 47 mediated through the agonistic function of 48 A2E, at least in part. The results of this 49 study provide a novel potential therapeutic 50 target for this incurable condition. 51 52 Retinal age pigments, or lipofuscin granula, 53 contain the fluorophores that accumulate with 54 age and which are thought to represent a 55 biomarker for cellular aging (1). Lipofuscin 56 results from an incomplete degradation of 57 altered material trapped in lysosomes (2) and 58 the accumulation of lipofuscin is related to an 59 increased risk of choroidal neovascularization 60 (CNV) due to age-related macular degeneration 61 (AMD) (1,2). AMD is a leading cause of legal 62 blindness in developed countries (3), and even 63 with the recent advent of several treatment 64 options (4), treatment of AMD remains 65 difficult (5). Thus, a better understanding of the 66 pathogenesis is needed to pursue a novel 67 potential pharmaceutical target. 68 Visual loss in AMD is caused by CNV, i.e., 69 the neovascular vessels extending from the 70 choroid underneath the sensory retina, and the 71 subsequent atrophy of the RPE. The process 72 preceding AMD, early age-related 73 maculopathy (ARM) (4), is pathologically 74 characterized by age-related changes in the 75 RPE, such as the accumulation of a deposit 76 called drusen in the basement membrane of the 77
http://www.jbc.org/cgi/doi/10.1074/jbc.M708989200The latest version is at JBC Papers in Press. Published on March 6, 2008 as Manuscript M708989200
Copyright 2008 by The American Society for Biochemistry and Molecular Biology, Inc.
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
2
RPE, i.e. Bruch’s membrane, and fluorescent 1 lipofuscin granules in the RPE cells (6). It is 2 generally considered that the age-related 3 accumulation of these potentially toxic deposits 4 affects normal RPE functions (4). Although it 5 remains unsubstantiated whether the 6 accumulation of lipofuscin is related to the 7 development of exudative AMD, the fact that 8 most abundant accretion is in the RPE cells 9 under the central retina suggests that there may 10 be a causal relationship between lipofuscin 11 accumulation and exudative AMD (2). 12 Laboratory studies also support that there may 13 be a causal relationship between lipofuscin 14 accumulation and the progression of exudative 15 AMD (7,8). There is a strong impetus to 16 understand how the accumulation of these 17 potentially toxic lipofuscin fluorophores 18 contribute to AMD; however, little is known 19 about the underlying molecular mechanism of 20 CNV development. 21 RPE lipofuscin is a byproduct of the 22 phagocytosis of lipid-rich photoreceptor outer 23 segments and consists of a complex mixture of 24 pigments. A major fluorophore is A2E 25 (N-retinyledin-N-retinylethanolamin) (9), 26 which arises from a Schiff-base reaction from 27 ethanolamine and vitamin-A-aldehyde (10). 28 In vitro, A2E affects normal RPE functions by 29 causing membrane permeabilization inhibiting 30 lysosomal function (11), inhibiting cytochrome 31 c oxigenase (12), acting as a detergent 32 inhibiting the ATP-driven proton pump (13) 33 and partly mediating light damage by acting as 34 a photosensitizer (14), targeting DNA (15). 35 This study demonstrates that A2E is an 36 endogenous ligand for retinoic acid receptor 37 (RAR). The data suggest that A2E 38 accumulation results in the pro-angiogenic 39 conversion of retinal pigment epithelial cell 40 phenotype predisposing the environment to 41 CNV development via RAR activation, at least 42 in part. 43 44
Experimental Procedures 45 46 A2E Synthesis- A2E was prepared from 47
all-trans-retinal and ethanolamine according to 48 the method in literature (10). Typically, a 49 mixture of all-trans-retinal (50 mg, 176 µmol) 50 and ethanolamine (4.6 mg, 78 µmol) in ethanol 51 (1.5 mL) was stirred in the presence of acetic 52 acid (4.7 µL, 78 µmol) at room temperature 53 under dark conditions for two days. The 54 reaction mixture was evaporated and then 55 purified by silica gel column chromatography. 56 After elution with MeOH:CH2Cl2 (5:95) to 57 remove less-polar byproducts, further elution 58 with MeOH:CH2Cl2 (5:95) including 0.1% 59 trifluoroacetic acid gave A2E (21 mg, 38%). 60 The product was confirmed by NMR (JEOL 61 GSX 270), UV-VIS (JASCO-V550), 62 MALDI-TOF-MS (Bruker model Reflux III), 63 and HPLC (TSK gel Silica 60) analysis. All 64 the spectral data was consistent with previous 65 reports (9,10). Also, the chromatogram of A2E 66 shows a single sharp peak without any other 67 retinoid contamination as shown in Figure 2C. 68 Cell Culture- Human retinal pigment epithelial 69 (ARPE19) cells and human embryonic kidney 70 293T cells were maintained in Dulbecco’s 71 modified Eagle’s medium (DMEM). Cells 72 were cultured at 37°C in a humidified 73 atmosphere with 10% CO2. 74 For uptake into cultured cells, A2E was 75 delivered at 1, 10 and 100nM concentrations 76 into the culture media, and confluent cultures 77 of the cells were incubated with A2E, all-trans 78 retinoic acid (atRA) or ethanol (control) for the 79 indicated period. Preliminary experiments have 80 demonstrated that the cells six weeks and 24h 81 after confluence showed similar responses to 82 A2E exposure, the cells 24h after confluence 83 were used for analysis. ELISA was used for 84 determining VEGF in ARPE-19 cells. 85 For the ELISA experiment, ARPE-19 cells 86 were plated in a 24-well plate in phenol 87 red-free DMEM containing 10% FBS, 88 pretreated with dextram-coated charcoal, and 89 incubated overnight. The spent medium was 90 collected for the measurement of VEGF. The 91 ELISA for VEGF was performed using the 92 human VEGF immunoassay system (R&D 93 Systems, Minneapolis, MN). 94
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
3
Luciferase Assay- For the transfection 1 experiment, 2 × 104 Hec-293T cells were plated 2 in a 24-well plate in phenol red-free DMEM 3 containing 10% FBS, pretreated with 4 dextram-coated charcoal, and incubated 5 overnight. The MH100-tkLuc reporter plasmid 6 (100ng) was cotransfected with the internal 7 control plasmid pRL-CMV (Promega, Madison, 8 WI) and the expression plasmids, pCMX-Gal, 9 p-CMX-Gal-RARα, p-CMX-Gal–RXRα, 10 p-CMX-Gal-PNR, p-CMX-Gal-PPARγ and 11 p-CMX-Gal-FXR using the SuperFect 12 transfection reagent (QIAGEN, Valencia, CA) 13 following the manufacturer’s protocol. 14 Luciferase reporter vector for VEGF promoter 15 was constructed as previously described (16). 16 The amount of each transfected DNA sample 17 was adjusted with carrier DNA to make a total 18 of 200ng. After transfection, ligands were 19 added to the medium and the cells were 20 cultured for 24h. The cells were lysed in 21 100µl/well passive lysis buffer, and the 22 luciferase assay was performed in accordance 23 with the protocols of the Dual-Luciferase 24 Reporter Assay System, using a Lumat LB 25 9507 luminometer (Berthold Technologies, 26 Bad Wildbad, Germany). 27 Binding Assay - For expression of the 28 ligand-binding domain (DEF) of RARα in 29 Escherichia coli, the expression vector pGEX 30 was used. GST fusion protein 31 (GST-RARα-DEF) was expressed in E. coli 32 M-15 after isopropyl β-D 33 thiogalactopyranoside induction and purified 34 on gluthathione-Sepharose beads as has been 35 previously described (17). The expression of 36 the protein of the predicted size was monitored 37 by SDS-PAGE. The GST-RARα protein (2-3p 38 mol) was incubated in 0.2 ml of binding buffer 39 (50 mM Tris, pH 7.6, 150mM NaCl, 0.2% 40 (v/v) NP-40, 0.1 % (v/v) Triton X-100) 41 containing 270pM [3H] atRA, and various 42 concentrations of atRA or A2E. After 3h of 43 incubation at room temperature, the tubes were 44 centrifuged at 12,000 × g for 10min and the 45 supernatants were subjected to liquid 46 scintillation counting. To determine the Kd, 47
nonlinear regression analysis of the 48 competition curves was carried out using Prism 49 4 (GraphPad). 50 Reverse transcriptase polymerase chain 51 reaction (RT-PCR)- RNA for RT-PCR was 52 isolated using an SV Total RNA Isolation Kit 53 (Promega, Madison, WI) in accordance with 54 the manufacturer’s instructions. 55 cDNA was prepared using Superscript III for 56 RT-PCR (Invitrogen). Each PCR was carried 57 out in a 20µl volume using Platinum SYBR 58 Green qPCR SuperMix UDG (Invitorogen) for 59 15min at 95°C denature, followed by 55 cycles 60 at 95°C for 30s and 60°C for 1 min in Roche 61 LightCycler. Values for each gene were 62 normalized to expression levels of GAPDH. 63 The sequences of the primers used for RT-PCR 64 were as follows; human-GAPDH, left, 65 5’-gagtcaacggatttggtcgt-3’, right, 66 5’-ttgattttggagggatctcg-3’. human-VEGF, left 67 5’-atgtgcatggtgatgttgga-3’, right, 68 5’-gcttgctgctgtacctccac-3’. human-RARβ2, left 69 5’-ggtttcactggcttgaccat-3’, right, 70 5’-ggcaaaggtgaacacaaggt-3’. human-NORPEG, 71 left, 5’-caagagccccataaacctca-3’, right, 72 5’-aaggcttccacagc-3’. human-SCD, left, 73 5’-tcctat gggaagctggtcac-3’, right, 74 5’-ggatagacagggcagagcag-3’. 75 Chromatin immunoprecipitation (ChIP) assay- 76 ChIP analysis was performed using the ChIP 77 assay kit (Upstates), according to the 78 manufacturer’s instructions. ARPE-19 cells 79 were cultured in the presence of A2E or atRA. 80 Soluble chromatin prepared from 1× 106 cells 81 was immunoprecipitated with an antibody 82 against αAcH4 (Upstates). A negative control 83 experiment was performed using anti-IgG, 84 which gave no positive signals. 85 High performance liquid chromatography 86 (HPLC) - Samples of A2E were analyzed using 87 normal-phase HPLC on a silica column (TSK 88 gel silica-60 Toso) using the mobile phase: 89 dichloromethane/methanol/triethylamine 90 (200:10:0.25 vol/vol) at 1.0ml/min. The eluted 91 peaks were analyzed with a UV detector 92 (430nm). Samples of atRA were analyzed 93 using normal-phase HPLC on a silica column 94
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
4
(TSK gel silica-60 Toso) using the mobile 1 phase: dichloromethane/TFA (2000:1 vol/vol) 2 at 1.0ml/min. The eluted peaks were analyzed 3 with a UV detector (340nm). 4 In vitro tube formation assay 5 In vitro anti-angiogenesis activity was 6 evaluated with in vitro tube formation assay. 7 HUVECs starved of serum for 4h were 8 seeded at a cell density of 40,000 cells per 9 well in twenty-four-well culture plates 10 (Nalge Nunc International), pre-coated 11 with 0.4 mL low-growth-factor synthetic 12 matrix (Matrigel; BD Bioscience, San Jose, 13 CA) and cultured with DMEM medium or 14 the conditioned medium (DMEM) 15 collected from ARPE19 cells exposed to 16 A2E for 24h, either in the presence or 17 absence of Ro415253 or neutralizing 18 antibody against VEGF. Tube formation 19 was determined 16h after cells were plated 20 on Matrigel, by counting the number of 21 connected cells in five randomly selected 22 fields at x200 magnification, and dividing 23 that number by the total number of cells in 24 the same field. Micrographs were taken 25 under a phase contrast light microscope 26 (Olympus, Tokyo, Japan). 27 28
RESULTS 29 30 A2E is a ligand for RAR. Given the 31 lipid-soluble nature of A2E, the effects of A2E 32 on the nuclear receptor function were 33 investigated by means of a luciferase assay. 34 Subsequently, it was found that A2E 35 transactivates retinoic acid receptor (RAR) in 36 HEC-293T cells using GAL-4 DBD fused 37 RARα ligand binding domain (LBD) with a 38 luciferase reporter gene plasmid containing 39 GAL4-DBD binding site in a dose-dependent 40 manner (Figure 1a and b). Similar results were 41 obtained using full-length RARα with a 42 reporter plasmid containing consensus RARE 43 sequence (DR1) (data not shown). A2E 44 transactivated only RAR, and not other nuclear 45
receptors including the retinoid X receptor 46 (RXR) α, peroxisome proliferator activator 47 receptor (PPAR) γ or orphan nuclear receptors 48 including the farnesoid X receptor (FXR) or 49 the photoreceptor cell-specific receptor (PNR) 50 (Figure 1a). Based on the fact that the nuclear 51 receptor recruits coactivators such SRC-1 52 (steroid receptor coactivator-1), TIF2 53 (transcriptional intermediary factor 2), AIB-1 54 (amplified in breast cancer-1) upon ligand 55 binding, the present study examined 56 A2E–induced interactions of RARα with 57 coactivators in the mammalian two-hybrid 58 system, which showed A2E-induced bindings 59 of p160 coactivators (SRC-1, TIF2 and AIB-1) 60 to RARα (data not shown). Next, to verify 61 whether A2E acts as a cognate ligand for 62 RARα, it was investigated as to whether A2E 63 could compete for the specific binding of [3H] 64 all-trans retinoic acid (atRA) to the RARα 65 LBD-GST fusion protein (GST-RARα) in vitro. 66 It was demonstrated that A2E bound to 67 GST-RARα with a half-maximal inhibitory 68 concentration (IC 50) of 8.74nM, whereas that 69 of atRA was 0.49nM (Figure 1c). The control 70 vehicle alone did not interfere with the binding 71 of [3H] atRA to the GST-RARα, confirming 72 the specific binding of A2E. 73 Furthermore, RT-PCR analysis demonstrated 74 that A2E treatment transactivated established 75 RAR target genes, such as NORPEG (Novel 76 Retinal Pigment Epithelial Cell Gene) (18), 77 SCD (Stearoyl-CoA Desaturase) (19) 78 RARβ2 (20) in a human retinal pigment 79 epithelial cell line, ARPE-19 cells (Figure 2a). 80 Specifically, the real time RT-PCR 81 measurements indicated that while 82 vehicle-treated cells showed no change in the 83 expression of these genes, greater than 3 fold 84 increases were observed in levels of NORPEG, 85 SCD and RARβ2 mRNA following atRA 86 treatment for 24h and 48h, as expected. 87 Likewise, 2.5 to 3 fold increases in levels of 88 NORPEG, SCD and RARβ2 mRNA were 89 observed at 24h and 48h after A2E treatment. 90 Together, these experiments demonstrate that 91 A2E is a bona fide RAR ligand. 92
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
5
A2E induces sustained activation of RAR target 1 genes. Noteworthy was that in the ARPE-19 2 cells at 72h and 96h after the atRA treatment, 3 the expression levels of RAR target genes had 4 returned back to normal levels, whereas they 5 remained upregulated or steadily increased 6 after A2E treatment. RAR recruits histone 7 acetyltransferase upon ligand binding, and 8 acetylate histones including histone H4, to 9 regulate gene transcription (21). To confirm the 10 acetylation of histone H4 of the promoter 11 region of these genes, chromatin 12 immunoprecipitation (ChIP) assay was 13 performed. The results revealed that both 14 A2E and atRA treatment enhanced the 15 acetylation of histone H4 associated with both 16 SCD and RARβ2 promoter at 24h after 17 treatment. The acetylation was sustained even 18 96h after A2E treatment, while it declined at 19 96h after atRA-treatment (Figure 2b). Using 20 RARα antibodies, the promoter was occupied 21 with RARα, regardless of the presence or 22 absence of the ligands as had been reported 23 previously (20) (data not shown). Furthermore, 24 HPLC analysis demonstrated that atRA was 25 turned over or expelled from the ARPE-19 26 cells after 72h, which is in contrast to A2E, 27 which remained for at least 72h after treatment 28 (Figure 2c). Together with the results of 29 RT-PCR analysis and ChIP assays these results 30 suggest that residual A2E activated the 31 expression of these genes. After the A2E 32 treatment, atRA was not detected at any point 33 tested in the ARPE-19 cells, suggesting that 34 A2E is unlikely to be metabolically converted 35 to produce atRA. 36 A2E induces VEGF through the activation of 37 RAR. Vascular endothelial growth factor 38 (VEGF) production is also upregulated by 39 atRA in some cell lines (16), raising the 40 possibility that VEGF is upregulated by A2E in 41 RPE cells. Thus, following treatment with A2E, 42 the changes in expression of the VEGF protein 43 were investigated. The cells expressed higher 44 levels of the VEGF protein after treatment with 45 A2E as well as atRA (Figure 3a). The real time 46 PCR measurements indicated a 3-fold increase 47
in VEGF mRNA as a result of treatment by 48 A2E (Figure 3a). Luciferase assays 49 demonstrated that as well as atRA, A2E 50 increased the luciferase levels approximately 51 3-5 fold in Hec 293 cells transfected with the 52 luciferase plasmid containing VEGF promoter, 53 suggesting that A2E regulated these 54 expressions at the transcriptional level (Figure 55 3c). Upregulated expression of VEGF mRNA 56 by A2E treatment was diminished by siRNA 57 against RARα, but not by scramble RNA, and 58 by the addition of Ro415253, a potent RARα 59 antagonist. But it was not diminished by the 60 treatment of specific inhibitors for 61 mitogen-activated protein kinase (MEK), 62 U0126, or for phosphoinositide 3-kinase 63 (PI3K), wortmannin, or in the presence of the 64 antioxidants, N-acetyl-L-cysteine (NAC) or 65 pyrolidinedithiocarbamate (PDTC) (Figure 3b). 66 Moreover, VEGF luciferase reporter activation 67 induced by A2E was diminished by siRNA 68 against RARα, but not by scramble siRNA 69 (Figure 3c), suggesting that A2E stimulates 70 VEGF expression by RARα-mediated 71 transactivation. Furthermore, luciferase assay 72 demonstrated that the effects of both atRA and 73 A2E on the VEGF promoter were abrogated 74 using the plasmid containing mutations of four 75 Sp-1 binding sites, previously shown to be 76 indispensable for RAR-mediated VEGF 77 transactivation (16), suggesting that the binding 78 sites were necessary for VEGF induction by 79 A2E, as well as atRA (Figure 3c). 80 In order to examine whether A2E stimulates 81 VEGF expression and promote angiogenesis, it 82 was examined whether human umbilical vein 83 endothelial cells (HUVECs) form capillary-like 84 structures in response to secreted factors from 85 the ARPE19 cells exposed to A2E. The results 86 demonstrated that HUVECs cultured with the 87 conditioned medium of ARPE-19 after A2E 88 treatment induced capillary-like tube formation, 89 suggesting that ARPE19 cells secrete factor(s) 90 to promote angiogenesis after exposure to A2E. 91 To examine whether activation of RARα and 92 VEGF expression is involved, the in vitro tube 93 formation was performed by coincubation with 94
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
6
either by Ro415253 or human VEGF antibody, 1 which demonstrated that tube-formation was 2 inhibited when Ro415253 or human VEGF 3 antibody was present (Figure 4), suggesting 4 that inactivation of either RARα or VEGF 5 leads to decreased angiogenesis activity. In 6 conclusion, these data support that A2E 7 stimulates VEGF expression and promotes 8 angiogenesis presumably by activating RARα 9 in cells. 10 11
DISCUSSION 12 13 It was demonstrated that A2E is an 14 endogenous ligand for RARα. A2E binds 15 RARα in vitro, and transactivates RARα. 16 Furthermore, A2E treatment transactivated 17 established RAR target genes, such as 18 NORPEG, SCD and RARβ2 in ARPE-19 cells, 19 and ChIP assay revealed that A2E treatment 20 induces acetylation of histone H4 associated 21 with SCD and RARβ2 promoters. These results 22 support that A2E acts as a bona fide ligand for 23 RAR. Previous investigations have 24 demonstrated several examples where 25 biological metabolites act as ligands for nuclear 26 receptors (22). For example, bile acid, which is 27 the biological metabolite of cholesterol is well 28 known to act as the ligand for FXR and to 29 negatively regulate its own biosynthesis 30 through FXR (22). Similarly, cholesterol 31 metabolites produced from squalene by a shunt 32 in the classical cholesterol biosynthesis 33 pathway acts as natural LXR ligand. 34 Pregnane X receptor (PXR) is activated by a 35 wide variety of compounds including natural 36 and synthetic steroids (22). The results of this 37 study, in conjunction with those of these 38 previous investigations, support the importance 39 of biological metabolites on the nuclear 40 receptor functions. 41 Accumulation of A2E in RPE cells induces 42 the sustained activation of RAR and stimulates 43 the expression of a potent pro-angiogenic 44 factor, VEGF. Several groups have previously 45 reported the effects of atRA on angiogenesis, 46 although the results obtained are controversial 47
(23-25). Some groups have reported an 48 anti-angiogenic effect (23,24). For example, 49 Majewski et al. reported that atRA functions to 50 inhibit angiogenesis using in vivo tumor 51 cell-induced angiogenesis assay (23). Similarly, 52 Pepper et al. reported the anti-angiogenic 53 effects of atRA using an in vitro 54 three-dimensional collagen gel assay (24). On 55 the other hand, pro-angiogeic effects have been 56 reported by in vitro tube formation assay using 57 HUVECs and normal human dermal fibroblast 58 (25). It has also been demonstrated that atRA 59 and RARα ligand AM80 upregulates VEGF 60 production in fibroblast cells (26). The 61 conflicting reports may be attributed, at least in 62 part, to that fact that nuclear receptor ligands 63 have different actions depending on the cellular 64 context. The difference in the response to the 65 same ligand may be due to the quantity of 66 cofactor or other regulatory factors in cells (27). 67 The data in the present study support that A2E, 68 as well as atRA, stimulate VEGF expression by 69 activating RAR in RPE cells. The molecular 70 mechanism that leads to different angiogenic 71 response in different cells with the same ligand 72 is an important issue for future research. 73 The ARM is pathologically characterized by 74 age-related changes in the RPE, such as the 75 accumulation of lipofuscine in the Bruch’s 76 membrane (4). Although it is not substantiated 77 whether accumulation of lipofuscin is related to 78 the development of exudative AMD clinically, 79 some research results suggest that there may be 80 a causal relationship between lipofuscin 81 accumulation and exudative AMD (2,7,8). The 82 data in the present study support that the 83 accumulation of A2E induces the enhancement 84 of VEGF expression, suggesting a role of A2E 85 in the progression of exudative AMD. 86 Anti-VEGF therapy is currently showing 87 favorable results in the treatment of CNV (28), 88 suggesting VEGF plays a pivotal role in the 89 progression of CNV. However, the 90 overexpression of VEGF in RPE cells is not 91 sufficient to induce CNV. Recent studies have 92 demonstrated that the loss of the integrity of 93 the Bruch’s membrane, as well as the 94
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
7
overexpression of VEGF in RPE cells, is 1 required to induce CNV (29,30). Schwesinger 2 et. al reported that in transgenic mice that 3 overexpress VEGF in the RPE an intact 4 Bruch’s membrane prevents CNV from 5 penetrating into the subretinal space (29). 6 Yoshida et. al also reported that the expression 7 of VEGF is increased in neprilysin gene 8 disrupted mice but that no CNV was observed 9 (30). Clinically, it is well known that the 10 accumulation of A2E is seen in Stargardt 11 disease, but the occurrence of CNV is rare (31). 12 These findings suggest that the loss of the 13 integrity of Bruch’s membrane, in combination 14 with the accumulation of A2E, is needed for 15 the occurrence of CNV. 16 Photodynamic therapy with verteporfin 17 (PDT) has been shown to reduce the overall 18 risk of moderate visual loss and lesion growth, 19 and has been the standard therapy for patients 20 with exudative AMD (5). However, 21 histopathological studies have shown that the 22 effect of PDT is limited, and clinical studies 23 have demonstrated that in most cases, repeated 24 retreatment is needed every three months. 25 Recently, it has been demonstrated that 26 intravitreous injection of anti-VEGF antibody 27 is effective for patients with neovascular AMD, 28 improving visual acuity and reducing retinal 29 edema, but the inhibition of VEGF signaling is 30 considered to elevate the risk of adverse 31 systemic effects. For example, Lee et al. 32 demonstrated that genetic deletion of VEGF 33 specifically in the endothelial lineage lead to 34 progressive endothelial degeneration and 35 sudden death in mutant mice (32). Another 36 group demonstrated that the intravitreous 37 injection of anti-VEGF antibody fraction, 38 ranibizumab, may tend to increase the risk of 39 myocardial infarction and cerebral infarction in 40 a dose-dependant manner (33,34). Thus, to 41 pursue a novel therapeutic target is mandatory. 42 Much effort has been directed to 43 understanding the mechanism of lipofuscin 44 toxicity on RPE cells. The current investigation 45 suggests that the endogenous accumulation of 46 A2E both induces sustained RAR activation in 47
RPE cells and activates the expression of 48 pro-angiogenic factors. The induction and 49 activation contribute to the phenotypic 50 alteration of RPE to the chronic pro-angiogenic 51 state observed in AMD and predisposes eyes to 52 CNV formation. The pro-angiogenic effect of 53 A2E is diminished by RAR antagonist and 54 siRNA against RARα, suggesting that RAR 55 may be a new potential pharmaceutical target 56 for the treatment of AMD. As a variety of 57 important biological functions are regulated 58 through RAR, further investigations will be 59 needed for the therapeutic application of these 60 findings. Also, understanding how A2E is 61 metabolized would be interesting to pursue 62 novel potential pharmacological targets. 63 Additionally, further identification of 64 lipofuscin components and their actions on 65 nuclear receptors might contribute to a better 66 understanding of the pathogenesis of AMD. 67
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
8
REFERENCES 1 1. Eldred, G. E., Miller, G. V., Stark, W. S., and Feeney-Burns, L. (1982) Science 2
216(4547), 757-759 3 2. Sparrow, J. R., and Boulton, M. (2005) Exp Eye Res 80(5), 595-606 4 3. Ferris, F. L., Fine, S. L., and Hyman, L. (1984) Arch Ophthalmol 102(11), 5
1640-1642. 6 4. Fine, S. L., Berger, J. W., Maguire, M. G., and Ho, A. C. (2000) N Engl J Med 7
342(7), 483-492 8 5. Wormald, R., Evans, J., and Smeeth, L. (2000) Cochrane Database Syst Rev (2), 9
CD002030 10 6. Sarks, S. H., and Sarks, P. S. (2001) Age-Related Maculopathy: Nonneovascular 11
Age-Related Macular degeneration and the Evolusion of Geographic Atrphy. In: 12 Ryan, S. J. (ed). RETINA, Mosby, Inc., St. Louis 13
7. Zhou, J., Cai, B., Jang, Y. P., Pachydaki, S., Schmidt, A. M., and Sparrow, J. R. 14 (2005) Exp Eye Res 80(4), 567-580 15
8. Zhou, J., Jang, Y. P., Kim, S. R., and Sparrow, J. R. (2006) Proc Natl Acad Sci U S 16 A 103(44), 16182-16187 17
9. Eldred, G. E., and Lasky, M. R. (1993) Nature 361(6414), 724-726 18 10. Parish, C. A., Hashimoto, M., Nakanishi, K., Dillon, J., and Sparrow, J. (1998) Proc 19
Natl Acad Sci U S A 95(25), 14609-14613 20 11. Finnemann, S. C., Leung, L. W., and Rodriguez-Boulan, E. (2002) Proc Natl Acad 21
Sci U S A 99(6), 3842-3847 22 12. Suter, M., Reme, C., Grimm, C., Wenzel, A., Jaattela, M., Esser, P., Kociok, N., 23
Leist, M., and Richter, C. (2000) J Biol Chem 275(50), 39625-39630 24 13. Bergmann, M., Schutt, F., Holz, F. G., and Kopitz, J. (2004) Faseb J 18(3), 562-564 25 14. Sparrow, J. R., Nakanishi, K., and Parish, C. A. (2000) Invest Ophthalmol Vis Sci 26
41(7), 1981-1989 27 15. Sparrow, J. R., Vollmer-Snarr, H. R., Zhou, J., Jang, Y. P., Jockusch, S., Itagaki, Y., 28
and Nakanishi, K. (2003) J Biol Chem 278(20), 18207-18213 29 16. Akiyama, H., Tanaka, T., Maeno, T., Kanai, H., Kimura, Y., Kishi, S., and 30
Kurabayashi, M. (2002) Invest Ophthalmol Vis Sci 43(5), 1367-1374 31 17. Yanagi, Y., Suzawa, M., Kawabata, M., Miyazono, K., Yanagisawa, J., and Kato, S. 32
(1999) J Biol Chem 274(19), 12971-12974 33 18. Kutty, R. K., Kutty, G., Samuel, W., Duncan, T., Bridges, C. C., El-Sherbeeny, A., 34
Nagineni, C. N., Smith, S. B., and Wiggert, B. (2001) J Biol Chem 276(4), 35 2831-2840 36
19. Samuel, W., Kutty, R. K., Nagineni, S., Gordon, J. S., Prouty, S. M., Chandraratna, 37 R. A., and Wiggert, B. (2001) J Biol Chem 276(31), 28744-28750 38
20. Perissi, V., Aggarwal, A., Glass, C. K., Rose, D. W., and Rosenfeld, M. G. (2004) 39 Cell 116(4), 511-526 40
21. Korzus, E., Torchia, J., Rose, D. W., Xu, L., Kurokawa, R., McInerney, E. M., 41 Mullen, T. M., Glass, C. K., and Rosenfeld, M. G. (1998) Science 279(5351), 42 703-707 43
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
9
22. Francis, G. A., Fayard, E., Picard, F., and Auwerx, J. (2003) Annu Rev Physiol 65, 1 261-311 2
23. Majewski, S., Szmurlo, A., Marczak, M., Jablonska, S., and Bollag, W. (1993) 3 Cancer Lett 75(1), 35-39 4
24. Pepper, M. S., Vassalli, J. D., Wilks, J. W., Schweigerer, L., Orci, L., and 5 Montesano, R. (1994) J Cell Biochem 55(4), 419-434 6
25. Gaetano, C., Catalano, A., Illi, B., Felici, A., Minucci, S., Palumbo, R., Facchiano, 7 F., Mangoni, A., Mancarella, S., Muhlhauser, J., and Capogrossi, M. C. (2001) Circ 8 Res 88(4), E38-47 9
26. Saito, A., Sugawara, A., Uruno, A., Kudo, M., Kagechika, H., Sato, Y., Owada, Y., 10 Kondo, H., Sato, M., Kurabayashi, M., Imaizumi, M., Tsuchiya, S., and Ito, S. 11 (2007) Endocrinology 148(3), 1412-1423 12
27. Kodera, Y., Takeyama, K., Murayama, A., Suzawa, M., Masuhiro, Y., and Kato, S. 13 (2000) J Biol Chem 275(43), 33201-33204 14
28. Moshfeghi, A. A., Rosenfeld, P. J., Puliafito, C. A., Michels, S., Marcus, E. N., 15 Lenchus, J. D., and Venkatraman, A. S. (2006) Ophthalmology 113(11), 2002 16 e2001-2012 17
29. Schwesinger, C., Yee, C., Rohan, R. M., Joussen, A. M., Fernandez, A., Meyer, T. 18 N., Poulaki, V., Ma, J. J., Redmond, T. M., Liu, S., Adamis, A. P., and D'Amato, R. 19 J. (2001) Am J Pathol 158(3), 1161-1172 20
30. Yoshida, T., Ohno-Matsui, K., Ichinose, S., Sato, T., Iwata, N., Saido, T. C., 21 Hisatomi, T., Mochizuki, M., and Morita, I. (2005) J Clin Invest 115(10), 22 2793-2800 23
31. Cideciyan, A. V., Aleman, T. S., Swider, M., Schwartz, S. B., Steinberg, J. D., 24 Brucker, A. J., Maguire, A. M., Bennett, J., Stone, E. M., and Jacobson, S. G. 25 (2004) Hum Mol Genet 13(5), 525-534 26
32. Lee, S., Chen, T. T., Barber, C. L., Jordan, M. C., Murdock, J., Desai, S., Ferrara, 27 N., Nagy, A., Roos, K. P., and Iruela-Arispe, M. L. (2007) Cell 130(4), 691-703 28
33. Gillies, M. C., and Wong, T. Y. (2007) N Engl J Med 356(7), 748-749; author reply 29 749-750 30
34. Dafer, R. M., Schneck, M., Friberg, T. R., and Jay, W. M. (2007) Semin 31 Ophthalmol 22(3), 201-204 32
33 34
FOOTNOTES 35 This work was supported in part by a grant-in-aid from the Ministry of Education, Science, Sports 36 and Culture of Japan. 37 38 The abbreviations used are: atRA; all-trans retinoic acid, AMD; age-related macular degeneration, 39 ARM; age-related maculopathy, ChIP; chromatin immunoprecipitation, CNV; choroidal 40 neovascularization, RAR; retinoic acid receptor, VEGF; vascular endothelial growth factor, 41 HUVEC; human umbilical vascular endothelial cells, NORPEG; Novel Retinal Pigment Epithelial 42 Cell Gene, SCD; Stearoyl-CoA Desaturase 43 44
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
10
FIGURE LEGENDS 1 2
Fig.1. A2E acts as a ligand for RAR. (a) A2E only transactivates RARα, not other nuclear 3 receptors such as RXRα, PPARγ, PNR and FXR. Luciferase assays were performed in HEC-293T 4 cells transfected with GAL4DBD containing MH100-tkLuc reporter plasmid (400ng), GAL-fused 5 expression vectors (200ng), and ligands (atRA ; 10–8M, Troglitazone; 10-6M, GW4046; 10-4M; A2E; 6 10-8M). (b) A2E transactivates RARα in a dose-dependent manner, similar to atRA. The 7 concentrations of atRA and A2E were 10-9M. 10-8M and 10-7M. Luciferase assays were performed 8 in HEC-293 cells with MH100-tk-Luc reporter plasmid and a GAL-fused RARα vector. (a and b) 9 Data are plotted as means; error bars represent the standard errors of the mean (n = 3). (c) 10 GST-RARα protein (2-3 p mol) was incubated with 270pM [3H] atRA in the presence of various 11 concentrations of A2E and atRA. All experiments were performed in triplicate, and each gave 12 similar results. Representative data is shown (a – c). 13 14 Fig.2. A2E induces sustained activation of RAR target genes. (a) Expression of established RAR 15 target genes such as NORPEG, SCD and RARβ2 in ARPE-19 cells using real-time PCR after cells 16 were treated with atRA, A2E or ethanol (control) at concentration of 10-8M for 24h, 48h, 72h, or 17 96h. The real time PCR measurements indicated increases in levels of NORPEG, SCD, and 18 RARβ2 mRNA following both atRA and A2E treatment for 24h and 48h. At 72h and 96h after 19 the atRA treatment, expression levels of RAR target genes had returned to normal levels, whereas 20 they remained upregulated after A2E treatment. Data are plotted as means; error bars represent the 21 standard errors of the mean (n = 3). (b) ChIP assay demonstrated increased acetylation of histone 22 H4 (AcH4) at promoters of SCD and RARβ2 genes at 24h after A2E (10-8M ) treatment, as well as 23 atRA (10-8M ) treatment. When chromatin prepared from ARPE-19 cells at 96h after A2E or atRA 24 treatment was subjected to analysis, AcH4 on the promoter region was diminished in ARPE-19 25 cells treated with atRA, whereas the acetylation was sustained in cells treated with A2E. (c and d) 26 HPLC analysis showing atRA (c) and A2E (d) in the extract of ARPE-19 cells at 24h, 48h and 72h 27 after A2E and atRA treatment (10-8M ). 28 29 Fig.3. A2E conveys pro-angiogenetic function on RPE cells. (a) ARPE-19 cells expressed higher 30 levels of the VEGF protein after 24h of treatment either with atRA or A2E than non-treated control 31 (left panel). VEGF mRNA production was increased either by atRA or A2E (10-8M) treatment in 32 ARPE-19 cells (right panel). (b) ARPE-19 cells were pretreated with one of U0126 (10-5M), 33 wortmannin (10-7M), NAC (10-2M), PDTC (10-4M) or Ro415253 (10-5M) for 30min, and then 34 treated for 24h with A2E (10-8M). RNA was extracted and analyzed by real-time RT-PCR. 35 VEGF mRNA levels were normalized by GAPDH. (c) VEGF and luciferase reporter activation 36 induced by A2E was diminished by siRNA against RARα, but not by scramble siRNA, and effects 37 of both atRA (10-8M ) and A2E (10-8M ) on the VEGF promoter were abrogated using plasmid 38 containing mutations of four Sp-1 binding sites (VEGF-mutant). 39 40 Fig.4. In vitro tube formation demonstrates A2E treatment induces angiogenesis through the 41 activation of RAR and the upregulation of VEGF. In vitro tube formation assay demonstrated that 42 HUVECs cultured with the conditioned medium of ARPE-19 after A2E (10-8M ) treatment induced 43 capillary-like tube formation, which was inhibited by coincubation with either by Ro415253 or 44 human VEGF antibody. Left panel shows the results from quantitative analysis. Data are plotted as 45 means; error bars represent the standard errors of the mean (n = 3). Right panels show the 46 representative micrographs. Original magnification, x200. All photographs are representative of 47
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
11
five standardized fields from three separate experiments. All experiments were performed in 1 triplicate. 2
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
13
Figure2 1
2 3
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
14
Figure3 1
2 3
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
Takezawa, Kenichi Takeyama, Woo-Dong Jang, Shigeaki Kato and Yasuo YanagiAya Iriyama, Ryoji Fujiki, Yuji Inoue, Hidenori Takahashi, Yasuhiro Tamaki, Shinichiro
ligand for retinoic acid receptorA2E, a pigment of the lipofuscin of retinal pigment epithelial cells, is an endogenous
published online March 6, 2008J. Biol. Chem.
10.1074/jbc.M708989200Access the most updated version of this article at doi:
Alerts:
When a correction for this article is posted•
When this article is cited•
to choose from all of JBC's e-mail alertsClick here
by guest on April 16, 2018
http://ww
w.jbc.org/
Dow
nloaded from
