Pretreatment with Proteasome Inhibitors Protects against Oxidative Injuries via PPARα-Dependent and -Independent Pathways in ARPE-19 Cells

Abstract

Purposes.: Oxidative processes may play important roles in age-related macular degeneration. Previous studies have suggested that enhancing proteasome activity by pretreatment with low doses of proteasome inhibitors reduces injury from oxidative damage in neuronal cultures. The objective of the current study was to determine whether proteasome inhibitors could ameliorate the toxicity from oxidative stresses in ARPE-19 cells and to dissect the pathways that may mediate these protective effects.

Methods.: The toxicity of oxidative stressors menadione (VK3) and 4-hydroxynonenal (4-HNE) and the protective effects of proteasome inhibitors, including MG-132 and clasto-lactacystin-β-lactone (LA), were studied in ARPE-19 cells. Binding and activation of the peroxisome proliferator-activated receptors (PPARs) family of transcription factors were studied using electrophoretic mobility shift assay (EMSA) and a peroxisome proliferator-activated response element (PPRE)–driven dual-luciferase reporter gene.

Results.: An 18-hour pretreatment with 30 to 300 nM MG-132 or 300 to 1000 nM LA reduced the toxicity of menadione or 4-HNE in ARPE-19 cells. The protective effects of MG-132 pretreatment were partially reversed by the PPARα antagonist GW6471 but not by the PPARγ antagonist GW9662; in contrast, neither agent reduced the protective effects of LA. MG-132 but not LA induced increased expression of a PPRE-driven luciferase reporter gene in a dose-dependent manner. Nuclear proteins isolated from ARPE-19 cells treated by MG-132 had increased binding to PPRE sequences as measured by EMSA.

Conclusions.: Our data suggest that pretreatment with proteasome inhibitors reduces oxidative injury in ARPE-19 cells and that the underlying mechanisms are different for different proteasome inhibitors, with PPARα-dependent effects for MG-132 and PPAR-independent effects for LA.

Introduction

Age-related macular degeneration (AMD) is a primary cause of visual impairment and blindness among older adults in developed countries. ^1,2 There is no effective prophylaxis for AMD and no effective treatment for most cases of AMD. The pathogenesis is poorly understood and likely multifactorial, involving a complex interaction of metabolic, functional, genetic, and environmental factors. The retinal pigment epithelial (RPE) cell is considered a primary target; dysfunction, dystrophy, and death of RPE cells play a vital role in the pathogenesis of AMD. ³ Oxidative injuries and chronic inflammation play important roles in degeneration and death of RPE cells in several retinal degenerative diseases including AMD. ^4,5  

The RPE subserves multiple functions in the retina, including absorption of blue light, protection from photooxidation via endogenous enzymatic and nonenzymatic strategies, maintenance of the blood–retina barrier, and isomerization of all-trans-retinol to 11-cis-retinal. ⁶ Cumulative oxidative stresses due to exposure to sunlight, the near-arterial level of oxygen, and phagocytosis of shed outer segment of photoreceptors ^3,7,8 eventually lead to degeneration of RPE cells and thus disruption of the blood–retina barrier. Antiinflammatory and antioxidant reagents are thus attractive therapeutic targets for RPE cells. Although antioxidant vitamins (e.g., vitamins E and C) and minerals (e.g., zinc) reduce oxidative injuries in cell culture models, ^9,10 there is no clinical evidence that dietary supplements of antioxidant vitamins or minerals would prevent or delay the onset of AMD. ¹¹  

Peroxisome proliferator-activated receptors (PPARs) are nuclear receptor proteins that function as transcriptional factors regulating gene expression. PPARs form heterodimers with the retinoid X receptor (RXR), which influences the transcription of numerous target genes by recognizing specific sequences termed peroxisome proliferator response elements (PPREs) in the promoter regions of the target genes. There are three members of the PPAR family: PPARα, PPARδ, and PPARγ. ¹² Recent observations suggest that the PPARs, especially PPARα and PPARγ, are involved in the regulation of the immune and inflammatory responses. ^13,14  

The ubiquitin-proteasome system (UPS) is a cytoplasmic multiprotein complex that degrades proteins that have been modified by addition of ubiquitin; addition of ubiquitin requires activation of a series of enzymes (E1, E2, and E3). ¹⁵ A loss or inhibition of proteasome function has been associated with aging, as well as a number of age-related diseases such as Alzheimer's Disease, Parkinson's Disease, and other neurodegenerative disorders. ^16,17 Recent studies suggest that retinal proteasome activity decreases with aging. ¹⁸ Proteasome inhibitors have been tested clinically as treatments for a variety of neoplasms; bortezomib (Velcade; Millennium Pharmaceuticals, Cambridge, MA) is now used in the treatment of cancers, especially in multiple myeloma. ^19,20 In addition to their application in cancer treatment, proteasome inhibitors have been found to enhance proteasome activity and counteract oxidative injuries in a variety of cell types when used in lower doses than those that inhibit proteasome activity sufficiently to cause cell death. ^21,22 Thus, we tested the possibility that pretreatment with proteasome inhibitors might reduce oxidative damage in RPE cells and set out to identify the pathways that might mediate these protective effects. 

Materials and Methods

Materials

The following substances, materials, and reagents (and suppliers) were used in this study: carbobenzoxy-l-leucyl-l-leucyl-l-leucinal, menadione, GW6471, and GW9662 (Sigma, St. Louis, MO); clasto-lactacystin-β-lactone, 4-hydroxynonenal, and protease inhibitor cocktail (Calbiochem, San Diego, CA); cell proliferation assay (MTS) and dual-luciferase reporter assay system (CellTiter 96 AQueous One Solution; Promega, Beijing, China); nylon membrane (Hybond-N+, 30 cm × 3 m; GE Healthcare Life Sciences, Shanghai, China); transfection reagents (Lipofectamine 2000; Invitrogen Life Technologies, Carlsbad, CA); a chemiluminescent EMSA kit (Lightshift; Thermo Scientific Branch, Shanghai, China); mouse monoclonal antibodies, anti-PPARα and anti PPARγ (Perseus Proteomics Inc., Tokyo, Japan); Clear-blue x-ray films (CL-XPosure films; Thermo Scientific Branch); acrylamide–bis-acrylamide solution (37.5:1; Bio-Rad, Shanghai, China); and ARPE-19 cells (American Type Culture Collection [ATCC], Manassas, VA). 

Methods

Cell Cultures.

ARPE-19 cells were cultured in Dulbecco's modified Eagle's medium (DMEM)/F-12 supplemented to 10% with fetal bovine serum (FBS). The cultures were maintained in a humidified 5% CO₂ incubator at 37°C. ARPE-19 cells were seeded in 96-well culture plates at an initial density of 18,000 cells/well and grown for 24 hours before treatments. 

MTS Assay.

After ARPE-19 cells were cultured for 24 hours, culture media were changed to serum-free DMEM/F-12 and cells were used for different treatments. At the end of treatment, cell survival was assayed using the aqueous nonradioactive cell proliferation assay kit (CellTiter96; Promega). Briefly, a 20-μL aliquot of 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium inner salt (MTS) was added to each well. After incubation at 37°C for 2 to 4 hours, the formazan product was detected (Spectra M5; Molecular Devices, Sunnyvale, CA) to measure absorbance at 490 nM. The cellular survival in control groups was set as 100%, and the readings from treatments were normalized to control groups. The plates were also examined by phase-contrast microscopy to visually confirm the results of the MTS assays. 

Analysis of Proteasome Activity In Vitro.

ARPE-19 cells cultured in 6-well plates were scraped into lysis buffer (10 mM Tris-HCl, pH 7.5, 1 mM EDTA, 2 mM ATP, 20% glycerol, 4 mM DTT), sonicated, and then centrifuged at 13,000g at 4°C for 10 minutes. The supernatants (2.5 μg of protein) were incubated with proteasome-activity assay buffer (50 mM Tris-HCl, pH 8, 0.5 mM EDTA, 40 μm N-succinyl-Leu-Leu-Val-Tyr-7-amido-4-methylcoumain [Suc-LLVY-AMC]) for 1 hour at 37°C. The reactions were stopped by adding 0.25 mL of cold water and placing the reaction mixtures on ice for at least 10 minutes. The intensity of fluorescence of each solution was measured by fluorescence spectrophotometry (Spectra M5; Molecular Devices) at 380 nm excitatory (Ex) and 440 nm emission (Em) wavelengths as described previously. ^23,24 The ratios between raw fluorescence and protein contents were used to represent proteasome activity as described before. ¹⁶  

Transient Transfections and Luciferase Assays.

ARPE-19 cells were grown in 96-well plates in DMEM/F12 supplemented to 10% FBS until they had reached 70% confluency. The cells were then transfected using a commercial reagent (Lipofectamine 2000) in serum-free DMEM. Each well received 0.4 μg of the firefly luciferase plasmid, in which three consecutive peroxisome proliferator-activated response elements (PPREs) were cloned downstream of TK promoter and upstream of luciferase (abbreviated as pPPREx3-tk-Luc, a gift from Ronald M. Evans, Salk Institute), and 20 ng pRL-SV40 Renilla luciferase reporter (Promega, Inc., Madison, WI). The plasmid mixtures were incubated with the commercial reagent (Lipofectamine 2000) at room temperature for 20 minutes, transferred to the cells for 3 hours, and then washed away with DMEM and grown in DMEM/F12/10% FBS for 18 hours. Subsequently, MG-132 or clasto-lactacystin-β-lactone (LA) were used to treat the transfected cultures in serum-free media. Firefly and Renilla luciferase activities were measured 24 hours later, using a dual-luciferase assay system (Promega) according to the manufacturer's instruction. 

Electrophoretic Mobility Shift Assay.

For nuclear extractions, all procedures were performed on ice. Culture dishes (10-cm diameter) were washed once with 10 mM ice-cold PBS. Cells were harvested by centrifugation at 200g for 10 minutes at 4°C. The cell pellets were washed twice with ice-cold PBS, followed by centrifugation at 200g for 5 minutes at 4°C, and resuspended with 5 vol of Buffer A in mM: 20 HEPES, 1.5 MgCl₂, 10 KCl, 1 EDTA, 1 EGTA, 250 sucrose, 0.1 PMSF, 1 dithiothreitol (DTT), and 1× protease inhibitor cocktail, pH 7.9. ^25,26 After a 10-minute incubation on ice, cells were homogenized with a minipestle. The lysates were centrifuged at 750g for 15 minutes at 4°C; the nuclear pellets were washed twice with the same lysis buffer, resuspended in 45 μL of Buffer B in mM: 20 HEPES, 1.5 MgCl₂, 20 KCl, 0.2 EDTA, 0.5 DTT, 0.2 PMSF, 1× protease, and phosphatase inhibitor cocktail, pH 7.9, on ice for 30 minutes; and 15 μL of Buffer C in mM: 20 HEPES, 1200 KCl, 0.2 EDTA, 0.5 DTT, 0.2 PMSF, 1× protease and phosphatase inhibitor cocktail, pH 7.9, were added and mixed. The samples were placed on ice for 30 minutes and centrifuged at 15,000g. Supernatants containing nuclear protein were transferred and stored at −80°C until analysis. Protein concentrations were determined using the Bradford method. For DNA-binding reactions, 5 μg of nuclear protein was diluted in binding buffer (10 mM Tris-HCl, 50 mM KCl, 1 mM DTT, 2.5% glycerol, 5 mM MgCl₂, 0.05% Nonidet P-40, 1 mM β-mercaptoethanol, 50 μg/mL poly[dI:dC]; pH 7.5). After a 30-minute incubation on ice, a biotin-labeled oligonucleotide probe (50 fmol) was added, and the reaction was continued for another 20 minutes. The probes were commercially available PPRE probes labeled at the 5′ end with biotin (F: 5′-CAA AAC TAG GTC AAA GGT CA-3′; R: 5′-TGA CCT TTG ACC TAG TTT TG-3′, the consensus sequence for PPRE was underlined) (synthesized by Shanghai Biotechnology, Inc.). The probes were heated in Tris-HCl buffer at 85°C for 10 minutes and then sat in room temperature for 2 hours. For competition assays, an unlabeled oligonucleotide probe was added to the prebinding reaction at 100-fold molar excess over the amount of biotin-labeled probes. Protein-DNA complexes were resolved by nondenaturing polyacrylamide (6%) gel electrophoresis, buffered by 0.25× Tris-borate/EDTA (TBE). The complexes consisting of biotin-labeled DNA and nuclear proteins were electrophoretically transferred to nylon membrane (Hybond-N+; GE Healthcare Life Sciences) and the membrane was irradiated for 5–30 minutes with a 245 nM illuminator and detected by chemiluminescence (Lightshift Chemiluminescent EMSA Kit; Thermo Fisher Scientific). 

Statistical Analysis

Data were analyzed for significant difference (P < 0.05) by ANOVA and Bonferroni post hoc test for multiple comparisons (SPSS 15.0.1; SPSS, Inc., Chicago, IL). 

Results

Proteasome Inhibitors Protected against Oxidative Injuries in ARPE-19 Cells

Previous studies in primary murine neocortical neuronal cultures indicated that pretreatment with proteasome inhibitors attenuates oxidative injuries induced by menadione and paraquat; this effect may be mediated by an increase in proteasome activities. ²¹ We tested the possibility that proteasome inhibitors MG-132 or LA could similarly reduce oxidative injury in RPE cells. An 18-hour pretreatment with MG-132 at doses from 30 to 300 nM attenuated the toxicity of 4-HNE or VK3 in ARPE-19 cells, with near-complete attenuation of injury at 30 nM MG-132 (Fig. 1A). The 18-hour treatment with MG-132 resulted in a dose-dependent reduction in the chymotrypsin-like proteasome activity in ARPE-19 cells (Fig. 1B). To confirm that the protective effects of MG-132 are also present with other proteasome inhibitors, we tested another proteasome inhibitor, LA, and found that LA similarly reduced the toxicity of 4-HNE or VK3 at 300 to 1000 nM (Fig. 2A). The LA treatment also repressed proteasome activity in ARPE-19 cells (Fig. 2B). Shorter pretreatment intervals (e.g., 1 hour with MG-132) did not reduce the toxicity of 4-HNE (data not shown). 

Figure 1. 

 

Pretreatment with reversible proteasome inhibitor, MG-132, protected against oxidative injuries in ARPE-19 cells. (A) Cultures were pretreated with the indicated concentrations of MG-132 for 18 hours before 18-hour exposure to HNE (15 μM) or VK3 (20 μM). MTS assay was used to measure cell viability at the end of the 18-hour HNE or VK3 treatment. The value in the sham-washed control cultures was set at 100% and survivals in treated cultures were normalized to the sham-washed control value. (B) ARPE-19 cell cultures were treated with different concentrations of MG-132 (3–1000 nM) for 18 hours, the cultures were harvested, and chymotrypsin-like proteasome activity was measured. The values from treated cultures were normalized to those in sham-washed control cultures (proteasome activity 100%). The results shown are mean (± SEM) of at least three independent experiments in quadruplicate cultures (*P < 0.05 versus control).

Figure 2. 

 

Pretreatment with irreversible proteasome inhibitor, LA, protected against oxidative injuries in ARPE-19 cells. (A) Cultures were pretreated with the indicated concentrations of LA for 18 hours before 18-hour exposure to HNE (15 μM) or VK3 (20 μM). At the end of the HNE or VK3 treatment, cell viability was assayed by MTS, as described in Figure 1. (B) ARPE-19 cell cultures were treated with the indicated concentration of LA (10–3000 nM) for 18 hours. At the end of the treatment period, chymotrypsin-like proteasome activity was measured. The values shown are the mean (± SEM) percentage of activity observed in sham-washed control cultures. All values shown are derived from at least three independent experiments in quadruplicate cultures (*P < 0.05 versus control).

PPARα, but Not PPARγ Antagonist, Reversed the Protective Effects of MG-132; in Contrast, Both Antagonists Could Not Reverse the Protective Effects of LA.

Previous studies have shown that MG-132 increases PPRE transactivation in some cell types. ²⁷ Therefore, we tested whether PPAR antagonists could attenuate the protective effects of MG-132 on oxidant-induced injury in RPE cells. The PPARα antagonist GW6471, at doses of 10 to 20 μM, partially reversed the protective effects of MG-132 (Figs. 3A, 3C); in contrast, the PPARγ antagonist GW9662 did not have any effect on the protective effects of MG-132 (Figs. 3B, 3D). Somewhat surprisingly, GW6471 did not reverse the protective effect of LA on oxidant-induced injury (Fig. 4). Application of GW9662 or GW6471 alone (without MG-132) had no effect on cell viability (Supplemental Fig. S1; link to supplemental material) or proteasome activity; GW9662 did not alter proteasome activity in the presence of MG-132; addition of 10 μM GW6471 to 30 nM MG-132 caused a greater reduction in proteasome activity than MG-132 alone (Supplemental Fig. S2). 

Figure 3. 

 

PPARα antagonist GW6471, not PPARγ antagonist GW9662, reversed the protective effects of pretreatment with the reversible proteasome inhibitor, MG-132. Cultures were pretreated with MG-132 (30 nM) and the indicated concentrations of GW6471 (A, C) or GW9662 (B, D) for 18 hours before exposure to HNE (15 μM; A, B) or VK3 (20 μM; C, D) for 18 hours. At the end of HNE or VK3 treatment, cell viability was measured by MTS assay. The results shown are mean (± SEM) of at least three independent experiments in quadruplicate cultures (*P < 0.05 versus control, **P < 0.05 indicated that the three combinatorial treatment including 4-HNE, MG-132, and GW6471 differed significantly from cultures treated by 4-HNE plus MG-132). An 18-hour treatment with GW6471 or GW9662 alone did not alter the viability of ARPE-19 cells at doses applied (Supplemental Fig. S1).

Figure 4. 

 

PPARα antagonist GW6471 and PPARγ antagonist GW9662 did not reverse the protective effects of pretreatment with the irreversible proteasome inhibitor, LA. Cultures were pretreated with LA (1 μM) and the indicated concentrations of GW6471 (A, C) or GW9662 (B, D) for 18 hours before exposure to HNE (15 μM; A, B) or VK3 (20 μM; C, D) for 18 hours. At the end of HNE of VK3 treatment, cell viability was assayed by MTS assay. The results shown are mean (± SEM) of at least three independent experiments in quadruplicate cultures (*P < 0.05 versus control).

MG-132 but Not LA-Induced PPRE Transactivity in ARPE-19 Cells.

As shown in Figure 3, the protective effect by MG-132 (30, 100 nM) could be reversed by a PPARα antagonist. We next tested whether MG-132 could induce the transactivation of PPRE in ARPE-19 cells. MG-132 increased PPRE-driven reporter gene expression in a dose-dependent manner; by contrast, LA did not have an effect (Fig. 5). We also tested PPRE binding activity in ARPE-19 cells treated by MG-132. Nuclear proteins isolated from ARPE-19 cells treated with MG-132 for 1 to 6 hours (especially at 6 hours) has increased binding to the PPRE probe; this binding was reduced by incubating the nuclear protein extract with an excess of unlabeled PPRE probe (competitor) before applying biotin-labeled probe (Fig. 6A). To determine which PPARs played a role in the MG-132–induced neuroprotective effects, we tested the effect of adding antibodies PPARα or PPARγ to the nuclear protein binding reaction. Both antibodies reduced the intensity of the PPARs/PPRE band in the EMSA assay but no “supershifted” band was detected, similar to a previous study of PPRE binding in adipocytes. ²⁸  

Figure 5. 

 

MG-132, but not LA, increased PPRE-driven luciferase reporter gene expression. ARPE-19 cells grown on 96-well plates were transfected with plasmid mixtures of pPPREx3-tk-Luc and pRL-SV40, followed by treatment of the indicated concentrations of MG-132 or LA for 24 hours. The detailed procedures are described in Materials and Methods. Values shown are mean (± SEM) ratio of firefly luciferase normalized to renilla luciferase performed in at least six experiments in quadruplicate cultures (*P < 0.05 versus control).

Figure 6. 

 

EMSA analysis of PPRE DNA binding in response to MG-132 treatment. (A) Canonical PPRE cis-element labeled by biotin was used as probe. ARPE-19 cells were treated with MG-132 (30 nM) for 1, 3, and 6 hours and nuclear proteins were extracted. “Cold” condition is identical to 6-hour condition except that 100-fold molar excess of unlabeled PPRE cis-element (competitor) was included in the reaction. Predicted location for migration of PPRE sequence and free probe are indicated as PPARs and FP, respectively. The binding was also decreased by preincubating antibodies PPARα or PPARγ with nuclear proteins isolated from ARPE-19 cells treated by MG-132 (B).

Discussion

Our results demonstrate that pretreatment with proteasome inhibitors can attenuate the injury induced by oxidative stress in ARPE-19 cells; the reversible proteasome inhibitor MG-132 attenuated injury at least in part via a PPARα-dependent pathway, whereas the irreversible inhibitor LA did so via a PPAR-independent pathway. Previous reports have shown that proteasome inhibitors, at relatively low doses, can increase proteasome activity and play antiinflammatory and antioxidation roles in cultured neurons and fibroblasts. ^21,22 Here, we demonstrated that MG-132 or LA had such protective effects in RPE cells at relatively low doses. For example, the greatest protective effect was observed at 30 nM MG-132; the protective effect gradually decreased with increasing MG-132 concentrations and disappeared at 10 μM MG-132. For LA, the largest protective effect was seen at 1 μM; there was no protective effect at the highest concentration tested (30 μM; data not shown). The concentrations of MG-132 and LA that reduced oxidative injury also inhibited proteasome activity in ARPE-19 cells; this is in contrast to previous reports where cytoprotective doses of proteasome inhibitors increased proteasome activity. ²¹ Consistent with previous studies, the protective effects occurred at relatively low doses. The injury-reducing effects of MG-132 could be partially but not completely reversed by the PPARα antagonist, GW6471, suggesting that the protective effect of MG-132 was mediated in part via the PPARα pathway. We did not see a change in the amount of proteasome subunit protein in cultures treated with MG-132 at the protective concentrations (not shown), suggesting that proteasome inhibitors reduce oxidative injury in ARPE-19 cells via pathways that do not involve alterations in proteasome subunit expression. We also probed the possible protective mechanisms provided by LA; the results indicated that LA increased the reduced glutathione (GSH) level in ARPE-19 cells (not shown); increasing the reduced glutathione level was also reported to counteract oxidation and inflammation in RPE cells. ²⁹  

Our data indicated that MG-132 counteracted oxidative injuries in ARPE-19 cells partially via a PPARα-dependent but PPARγ-independent pathway. This is consistent with the observation that the PPARγ agonist, 15-deoxy-delta-12,14-prostaglandin J2 (15d-PGJ2), reduces H₂O₂-induced RPE injuries via a PPARγ-independent pathway. ^30–32 The cytoprotective effect by 15d-PGJ2 relies on induction of reduced GSH, which is similar to the effect by LA in our experiment. The induction of reduced GSH depends on c-Jun N-terminal kinase (JNK) and p38 pathway because inhibitors of these pathways greatly reduce its production. ³¹ Earlier reports indicated that activation of both JNK and nuclear factor-κB (NFκB) simultaneously may have beneficial/cytoprotective effects, whereas activation of JNK with concurrent inhibition on NFκB is more likely to promote cell death. ^33,34 In our unpublished data, LA relieved oxidative injuries by increasing reduced glutathione, whose production is probably dependent on JNK or p38 activation. Therefore, we will determine whether LA activates JNK and NFκB pathways simultaneously. The protective effects of proteasome inhibitors occurred at relatively low doses. For example, maximal protective effects with MG-132 were observed at 30 nM and less robust protective effects were seen at higher concentrations (e.g., 300 nM), even though these higher concentrations provided greater inhibition of proteasome activity (Figs. 1A, 1B). This is consistent with a model in which the protective effects, mediated by activation of the PPARα pathway, are counterbalanced by the toxicity of higher doses of proteasome inhibitors, which are believed to compromise cell survival due to long-term inhibition of proteasome activities. ^35,36 As shown in Figures 1 and 2, MG-132 and LA protect against oxidative injuries even at doses that reduce proteasome activity. It is possible that short-term proteasome inhibition is beneficial to cell survival, whereas long-term inhibition is detrimental to cell integrity, similar to the effects of preconditioning with brief ischemia on reducing ischemic injuries in models of stroke. ³⁷ Our data suggest that pretreatment with MG-132 and LA reduces oxidative injury via different mechanisms. Since MG is a reversible, less specific proteasome inhibitor than LA, it is possible that the injury-reducing effects of MG occur through other “off-target” mechanisms rather than via proteasome inhibition. 

In summary, these studies suggest that there may be other mechanisms to increase the resistance of RPE to oxidative injury; better understanding of these mechanisms could provide new strategies to attenuate the death of RPE in age-related ocular diseases, especially AMD. 

Supplementary Materials

pdf S1 

Acknowledgments

The authors thank their colleague, Hou Ling, for providing the ARPE-19 cell line, which was obtained from the American Type Culture Collection. 

Footnotes

 Supported in part by Zhejiang Province Natural Science Foundation Grant Y2110086, by Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry of China, by Start-up Funding Grant 89210001 from Wenzhou Medical College (SW), and the Talented College Students of Zhejiang Province Grant 2011R413007.

Footnotes

⁴  These authors contributed equally to the work presented here and should therefore be regarded as equivalent authors.

Footnotes

 Disclosure: J. Cai, None; L. Sun, None; B. Lin, None; M. Wu, None; J. Qu, None; B.J. Snider, None; S. Wu, None