November 2015
Volume 56, Issue 12
Free
Nantotechnology and Regenerative Medicine  |   November 2015
Modeling the Dynamic AMD-Associated Chronic Oxidative Stress Changes in Human ESC and iPSC-Derived RPE Cells
Author Affiliations & Notes
  • Thelma Y. Garcia
    Buck Institute for Research on Aging, Novato, California, United States
  • Mark Gutierrez
    Buck Institute for Research on Aging, Novato, California, United States
  • Joseph Reynolds
    Buck Institute for Research on Aging, Novato, California, United States
  • Deepak A. Lamba
    Buck Institute for Research on Aging, Novato, California, United States
    Department of Ophthalmology, University of Washington, Seattle, Washington, United States
Investigative Ophthalmology & Visual Science November 2015, Vol.56, 7480-7488. doi:10.1167/iovs.15-17251
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Thelma Y. Garcia, Mark Gutierrez, Joseph Reynolds, Deepak A. Lamba; Modeling the Dynamic AMD-Associated Chronic Oxidative Stress Changes in Human ESC and iPSC-Derived RPE Cells. Invest. Ophthalmol. Vis. Sci. 2015;56(12):7480-7488. doi: 10.1167/iovs.15-17251.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose: Here we use human embryonic stem cells (hESCs) and human-induced pluripotent stem cell (hiPSC)-derived retinal pigment epithelium (RPE) cells to model chronic oxidative stress in vitro. This model allows us to understand the evolution of chronic stress response in RPE in vivo, as well as to monitor microRNAs changes. Finally, we use this in vitro model to identify a partial agonist of NRF2 that is protective against reactive oxygen species (ROS)-induced cytotoxicity.

Methods: The hESCs and hiPSCs were differentiated toward an RPE fate. Upon maturation, RPE cells were subjected to chronic oxidative stress using Paraquat (PQ). The cells were then analyzed using immunocytochemistry and quantitative RT-PCR to look for changes in gene expression and microRNA changes. Small molecules targeting NRF2 pathways were utilized to look for protection against oxidative stress–induced apoptosis.

Results: We show that 160 μM PQ can be used to generate a model of chronic oxidative stress in RPE cells derived from hESCs and hiPSCs. Using this model, we characterize the NRF2 pathway effectors during the early and late stages of chronic oxidative stress and identify microRNAs changes during oxidative stress. We find that hsa-miR144 modulates NRF2 activity during ROS stress. Lastly, we found a small molecule modulator of NRF2 that plays a protective role against oxidative stress–induced RPE apoptosis.

Conclusions: In summary, pluripotent stem cell–derived retinal cells can be used to model retinal diseases in a dish. This can provide an unprecedented opportunity to understand the evolution of disease processes and allow us to identify novel therapeutics.

Age-related macular degeneration (AMD) is the leading cause of worldwide blindness in the elderly, affecting almost 15 million people in the United States. Retinal changes associated with AMD are present in approximately 10% of people over 65 years of age and as many as one in three people over the age of 80 years. Although the disease was first described in the 1800s, its etiology remains poorly understood, and multiple factors may be involved in the progression of the disorder, including chronic oxidative stress. A number of studies have associated oxidative stress as the key driver to AMD.1,2 The retina is one of the tissues with the greatest consumption of oxygen in the body.3 This results in significant production of reactive oxygen species (ROS) in the retinal pigment epithelium (RPE). Increasing age results in the loss of ability to deal with this excessive ROS, which leads to oxidative damage. 
There are no known effective forms of treatment for the common dry form of AMD. This likely is due to the complex multifactorial etiology of AMD and, very importantly, a lack of accessible naturally occurring animal models of AMD. Fortunately, discovery of human pluripotent stem cells has revolutionized regenerative medicine. Human induced pluripotent stem cells (hiPSCs) generated by reprogramming somatic cells to an embryonic stem cell–like cell46 have now been successfully used to model a number of diseases including retinal degenerations.79 
A number of previous studies have looked to model AMD in vitro using the human RPE cell line ARPE-19,10 RPE from postmortem eye tissue,11,12 and more recently using human pluripotent stem cell–derived RPE cells.8,9 Most of these studies have focused on the ability of these cells to deal with acute ROS stress and studying the mechanisms involved. Here, we use human embryonic stem cells (ESCs) and human iPSC-derived RPE cells to model chronic oxidative stress in vitro. We use this model to understand the evolution of the response to chronic stress as experienced by RPE cells in vivo and to monitor changes in microRNAs secondary to this response. Our data demonstrate dynamic changes in NRF2–KEAP1 pathway effectors in oxidative stress response of RPE cells, as well as confirm an important role of hsa-miR-144. We also identify a small molecule partial agonist that promotes protection to severe oxidant challenge. 
Methods
Human Pluripotent Stem Cells and Human RPE Cell Culture
All stem cell work was approved by the Buck Institute SCRO committee. Human ESCs (WA-01; National Institutes of Health registry #0043) and iPSCs were maintained in Essential 8 medium (Gibco, Grand Island, NY, USA) and 1% penicillin-streptomycin-amphotericin B solution (Lonza, Walkersville, MD, USA). Cells were grown on Matrigel (BD Biosciences, St. Paul, MN, USA)-coated plates and serially passaged using 0.5 mM EDTA solution. Induced PSCs were generated by transfection of fibroblast cells of a 9-year-old male donor and a 17-year-old male donor using episomal vectors as previously described.13 The genotyping information of the key AMD single nucleotide polymorphism (SNP) in the various lines is described in Supplementary Table S1 and was carried out as previously described.14,15 Briefly, the lines did not have any risk alleles. The cells were maintained in the Essential 8 medium until iPSC colony formation. Individual iPSC colonies appearing at 3 weeks following initiation of reprogramming were manually picked and grown in Essential 8 media and serially passaged as described above. Human ESCs and iPSCs were differentiated to retinal lineage using our previously published protocol.1619 The RPE regions were manually picked and expanded. 
The RPE cells generated from both hESCs and iPSCs were cultured in α-MEM medium (Life Technologies, Grand Island, NY, USA) containing 1% fetal bovine serum (Atlanta Biologicals, Flowery Branch, GA, USA), l-glutamine (VWR, Radnor, PA, USA), taurine (Sigma-Aldrich Corp., St. Louis, MO, USA), hydrocortisone (Sigma-Aldrich Corp.), and tri-iodo-thyronine (Sigma-Aldrich Corp.) on Matrigel-coated plates or filter membranes (VWR).20 Cells were subcultured using Accutase (Gibco) in the presence of thiazovivin (1 lM), a Rho-associated protein kinase pathway inhibitor that allows passaging RPE cells for over eight passages.21 The RPE cells were in culture for up to seven passages without any appreciable loss in their ability to mature into polarized RPE cells. The cells were allowed to grow to 100% confluence and used for experiments following maturation. 
The ARPE-19 cell line was obtained from ATCC (Manassas, VA, USA) and cultured per the manufacturer's instructions to 100% confluency prior to oxidative stress studies. 
The RPE Oxidative Stress Treatments
The RPE cells were seeded in either a 24- or 12-well plate and allowed to mature for up to 3 weeks following final passage. Fully confluent plates of cells showing typical cobblestone morphology and presence of pigmentation were treated every other day with various concentrations of Paraquat (PQ; Sigma-Aldrich Corp.) diluted in RPE media described above for durations ranging from 1 day to 3 weeks. For Ai-1 (Sigma-Aldrich Corp.) treatment studies, Ai-1 (10 μM) was added to the cells in RPE cell media independently or simultaneously with PQ. 
Reverse Transcription–Polymerase Chain Reaction
Total RNA was isolated from RPE cells using Direct-zol RNA MiniPrep kit (ZYMO Research Corporation, Irvine, CA, USA) according to the manufacturer's instructions. RNA was quantified by spectrophotometry using a Thermo Scientific NanoDrop 2000. The cDNA was synthesized from 0.5 to 1.0 μg total RNA template using the iScript cDNA Synthesis kit (Bio-Rad, Hercules, CA, USA). Quantitative RT-PCR was performed on the CFX Connect Real-Time PCR Detection System (Bio-Rad) using conditions recommended by the manufacturer. Briefly, cDNA was added to 2× iTaq Universal SYBR Green Supermix (Bio-Rad), specific forward and reverse primers (see Supplementary Table S2 for primer sequences), and water, to a final reaction volume of 20 μL. The complete thermocycling parameters were as follows: 95°C for 30 seconds, followed by amplification for 40 cycles at 95°C for 5 seconds and at 60°C for 25 seconds and was followed by melting curve analysis from 60°C to 95°C. Results were normalized to β-actin levels. 
For microRNA analysis, total RNA was extracted and quantified as above. Complementary DNA was synthesized from 0.25 to 0.50 μg total RNA template using iScript cDNA Synthesis kit (Bio-Rad) in 20-μL reactions in a T100 Thermal Cycler (Bio-Rad) according to the manufacturer's recommended instructions, with the modification that each reaction contained 0.5 μM reverse transcription stem-loop primer. Up to five reverse transcription stem-loop primers were used per reverse transcription reaction. Quantitative RT-PCR analysis for mature miRNA transcripts was performed as above. Each reaction was composed of a miRNA-specific forward primer and a universal reverse primer (see Supplementary Table S3 for primers sequences), each at 0.4 μM concentrations per reaction, 1 μL cDNA diluted fivefold, 10 μL iTaq Universal SYBR Green Supermix (Bio-Rad), and diethylpyrocarbonate-treated water. A PCR was run using the parameters described above, and sample data values were normalized to U6 small nuclear RNA-2 (RNU6-2). 
Immunofluorescence Microscopy
Cells were fixed with 2% paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA, USA). Cells were blocked for 30 minutes in PBS containing Triton X-100 (Sigma-Aldrich Corp.) and 10% donkey serum (vol/vol). Primary and secondary antibodies were also diluted in this blocking solution for immunostaining (see Supplementary Table S4 for the list of antibodies). Phosphate-buffered saline was used for washing between steps, and 4′,6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich Corp.) was used to counterstain nuclei. Coverslips and slides were mounted with Fluoromount-G media (Electron Microscopy Sciences). Images were acquired on an Olympus IX71 fluorescence microscope and Zeiss LSM700 confocal laser scanning microscope. For TUNEL analysis, cells were incubated in the dark for 1 hour at 37°C with 50 μL TUNEL reaction mixture (In situ cell death detection kit, TMR Red; Roche, San Francisco, CA, USA) prepared according to the manufacturer's instructions. Cells were then rinsed three times with PBS and counterstained with DAPI prior to acquiring images on an LSM700 confocal microscope. 
Intracellular ROS Detection
Detection of compartmentalized accumulation of ROS in RPE cells was done using a dihydrodichlorofluorescein diacetate (DCF-DA) staining method (Acros Organics, Morris Plains, NJ, USA). The RPE cells were incubated for 1 to 1.5 hours at 37°C with 5 μM DCF-DA dissolved in Hanks' balanced salt solution (HBSS) medium followed by at least three washes with HBSS prior to imaging on an Olympus IX71 microscope connected to a monochrome camera. Images were captured, and fluorescence intensity was quantified using ImageJ software (http://imagej.nih.gov/ij/; provided in the public domain by the National Institutes of Health, Bethesda, MD, USA). The ROS levels were measured by quantifying fluorescence intensity from 10 to 15 cells from three different regions of the plate. 
Lentiviral Transduction of RPE Cells
Plasmids for hsa-miR-144 (HmiR0275-MR03) and scrambled controls (CmiR0001-MR03) were purchased from GeneCopoeia (Rockville, MD, USA). Lentiviruses were generated using three vector systems in HEK293 cells as previously described.16 A confluent plate of RPE cells was infected with lentiviruses along with 5 μg/mL polybrene (Sigma-Aldrich Corp.) overnight, followed by culture in RPE media for 1 week prior to oxidative stress analysis. Infectivity was confirmed by green fluorescent protein (GFP) expression from the lentivirus plasmid (70%–75% GFP+ cells per field). 
Statistical Analysis
All statistical analysis was carried out using GraphPad Prism 6 software. When comparing two groups, we performed a Student's t-test. For comparison of multiple groups, we performed 1-way and 2-way ANOVA analysis followed by a multiple comparisons test, using n = 3 to 5 for statistical analysis for all data. 
Results
Differentiation to RPE From hESCs and iPSCs
We differentiated hESCs and hiPSCs toward RPE fate using a combination of TGF-β pathway inhibitors along with a Wnt pathway inhibitor and insulin-like growth factor (IGF)-1 as previously described.1619 The cells, upon further differentiation over 4 to 6 weeks, took up RPE fate choice identified by their typical cobblestone morphology and pigmentation. These cells were manually isolated and expanded. Upon maturation over 3 to 4 weeks following passage, the cell displayed typical morphology and pigmentation (Fig. 1A). Immunocytochemistry (ICC) confirmed the expression of typical RPE markers including OTX2, MITF, and ZO-1 (Figs. 1B, 1C, 1F, respectively). Also, quantitative RT-PCR showed the expression of additional markers including BEST1, CRALBP, RPE65, TYR, PMEL17, and MITF (Fig. 1G). The cells were also analyzed by TEM, and Figures 1D and E show typical maturation features, including apical microvilli and melanosomes, and basal nuclei and mitochondria. 
Figure 1
 
Generation and characterization of RPE cells from human iPSCs. (A) Passage 7 human iPSC-derived RPE cells in culture showing typical cobblestone pattern and pigmentation. Cells coexpress OTX2 (B) and MITF (C), as well as RPE tight junction marker ZO-1 (F) on ICC. (G) Quantitative RT-PCR confirms high levels of expression of a number of other RPE cell-specific genes at 3 months of differentiation (at passage 6) compared with undifferentiated iPS cells. (D, E) The transmission electron microscopy (TEM) analysis of RPE cells showing typical morphologic features such as basal nuclei, apical microvilli, and pigmentary granules. Scale bars: 20 μm (AC, F).
Figure 1
 
Generation and characterization of RPE cells from human iPSCs. (A) Passage 7 human iPSC-derived RPE cells in culture showing typical cobblestone pattern and pigmentation. Cells coexpress OTX2 (B) and MITF (C), as well as RPE tight junction marker ZO-1 (F) on ICC. (G) Quantitative RT-PCR confirms high levels of expression of a number of other RPE cell-specific genes at 3 months of differentiation (at passage 6) compared with undifferentiated iPS cells. (D, E) The transmission electron microscopy (TEM) analysis of RPE cells showing typical morphologic features such as basal nuclei, apical microvilli, and pigmentary granules. Scale bars: 20 μm (AC, F).
Development of a Model of Chronic Oxidative Stress in RPE Cells
One of the important players in AMD pathogenesis is oxidative stress in RPE cells. Most studies have analyzed the role of oxidative stress in an acute setting in human retina-derived cells or ARPE-19 cell lines. We instead sought to study the progression of the response to stress as the cells move from an acute to a chronic stage as occurs in vivo. For this, we used a known chemical inducer of oxidative stress: PQ.22 Paraquat acts as a redox cycler. It reacts with oxygen rapidly to generate superoxide radicals and, over time, hydrogen peroxide and hydroxyl radicals. Additionally, it induces superoxide radical production in mitochondria. To generate a chronic oxidative stress environment, we initially surveyed concentrations of PQ that would induce ROS generation but not cell death. Cells exposed to as much as 160 μM PQ showed minimal TUNEL staining, whereas PQ at 320 μM and above was toxic to the cells within 3 days (Figs. 2A, 2A′; Supplementary Fig. S1A). We confirmed that PQ induces ROS generation using the DCF-DA assay, a fluorescence assay used to quantify intracellular ROS. Fluorescent DCF inside the cytoplasm of the cell in the presence of ROS was quantified (Figs. 2B, 2B′). We detected a dose-dependent increase in ROS-induced fluorescence with PQ. In addition, we tested additional ROS inducers including hydrogen peroxide and tert-butyl hydroquinone (TBHQ). However, we found that PQ was best tolerated and gave the most consistent ROS accumulation (Supplementary Fig. S1B) over weeks of exposure to the cells compared with other compounds. Both hESC-derived and iPSC-derived RPE cells could be maintained in media containing 160 μM PQ for over 3 to 4 weeks without any appreciable cellular loss. Finally, we compared the cellular response between stem cell–derived RPE cells and the ARPE-19 cell line to 160 μM PQ exposure. Over 1 week of oxidative stress, hESC-derived mature RPE cells did not undergo any obvious morphologic changes or cell death (Figs. 2A, 2C), whereas ARPE-19 cells lost their typical epithelial morphology, appearing spindly and showing cytoplasmic retraction (Fig. 2C). To test whether these changes were due to lack of maturation of the ARPE-19 cell line, we exposed confluent but immature unpigmented hESC-derived RPE cells to ROS stress (PQ 160 μM) and found similar morphologic disruption in these cells akin to the ARPE-19 cell line (Fig. 2C). This suggests that ARPE-19 represents a more immature RPE state and shows the advantage of using mature stem cell–derived RPE for these studies. 
Figure 2
 
Paraquat induced ROS accumulation in RPE cells. (A) The TUNEL analysis for dose-dependent toxicity of PQ in passage 6 hESC-derived RPE cells. Cells counterstained with DAPI. The PQ concentrations greater than 320 μM result in significantly higher cell death at day 3 following PQ exposure. Results are quantified and significance is tested using 1-way ANOVA in (A′). (B) The DCF-DA assay for dose-dependant ROS accumulation in passage 6 hESC-derived RPE cells 2 days following PQ exposure. The quantification of DCF fluorescence from 10 to 15 cells per condition from three different regions in the well is shown in (B′). Significance was tested using 1-way ANOVA. (C) Brightfield images comparing cellular morphology of unstressed and 160 μM PQ exposed confluent ARPE-19 cells, mature hESC-derived RPE cells (passage 6), and immature unpigmented confluent hESC-derived RPE cells (passage 7). The ARPE-19 and immature cells showed significant morphologic changes within 7 days of PQ exposure, whereas mature pigmented cells did not (arrowheads highlight spindle-shaped appearance of stressed ARPE-19 lines). Scale bars: 20 μm.
Figure 2
 
Paraquat induced ROS accumulation in RPE cells. (A) The TUNEL analysis for dose-dependent toxicity of PQ in passage 6 hESC-derived RPE cells. Cells counterstained with DAPI. The PQ concentrations greater than 320 μM result in significantly higher cell death at day 3 following PQ exposure. Results are quantified and significance is tested using 1-way ANOVA in (A′). (B) The DCF-DA assay for dose-dependant ROS accumulation in passage 6 hESC-derived RPE cells 2 days following PQ exposure. The quantification of DCF fluorescence from 10 to 15 cells per condition from three different regions in the well is shown in (B′). Significance was tested using 1-way ANOVA. (C) Brightfield images comparing cellular morphology of unstressed and 160 μM PQ exposed confluent ARPE-19 cells, mature hESC-derived RPE cells (passage 6), and immature unpigmented confluent hESC-derived RPE cells (passage 7). The ARPE-19 and immature cells showed significant morphologic changes within 7 days of PQ exposure, whereas mature pigmented cells did not (arrowheads highlight spindle-shaped appearance of stressed ARPE-19 lines). Scale bars: 20 μm.
Characterization of the RPE Cells During Early and Late Stages of Chronic Oxidative Stress: NRF2–KEAP1 Pathway
The NRF2–KEAP1 regulatory pathway plays a central role in the protection of cells against oxidative stress in many systems.23 In the cytoplasm, NRF2 is bound into a complex with KEAP1 and degraded by ubiquitination. Under oxidative stress, NRF2 is released and moves into the nucleus and drives downstream antioxidant response elements (AREs). We confirmed the NRF2 nuclear translocation using ICC. Exposure of mature iPSC-derived RPE cells (passage 5) to 160 μM PQ drove NRF2 from the cytoplasm (Fig. 3A) into the nucleus (Fig. 3B) in nearly all the cells in the plate. 
Figure 3
 
Characterization of the dynamic changes in the NRF2–KEAP1 pathway following chronic oxidative stress. (A, B) The ICC analysis for NRF2 showing nuclear translocation of the protein following PQ exposure in passage 5 iPSC-derived RPE cells. (C) Quantitative RT-PCR analysis of RPE cell cultures following 1 and 3 weeks of constant exposure to 160 μM PQ of either five replicates from passage 4 iPSC-RPE or six replicates from passage 6 hESC-RPE. NRF2 effectors, NQO1, HMOX1, GCLC, GCLM, and KEAP1, have varying degrees of response to the duration of ROS stress, with NQO1 and GCLC signficantly further up-regulated at week 3. (D) A component of the UPR pathway, CHOP, is also up-regulated within a week of ROS accumulation. In addition, p21 is up-regulated following ROS exposure. Significance was tested by two-way ANOVA analysis followed by a multiple comparisons test (*P < 0.05). Scale bars: 20 μm.
Figure 3
 
Characterization of the dynamic changes in the NRF2–KEAP1 pathway following chronic oxidative stress. (A, B) The ICC analysis for NRF2 showing nuclear translocation of the protein following PQ exposure in passage 5 iPSC-derived RPE cells. (C) Quantitative RT-PCR analysis of RPE cell cultures following 1 and 3 weeks of constant exposure to 160 μM PQ of either five replicates from passage 4 iPSC-RPE or six replicates from passage 6 hESC-RPE. NRF2 effectors, NQO1, HMOX1, GCLC, GCLM, and KEAP1, have varying degrees of response to the duration of ROS stress, with NQO1 and GCLC signficantly further up-regulated at week 3. (D) A component of the UPR pathway, CHOP, is also up-regulated within a week of ROS accumulation. In addition, p21 is up-regulated following ROS exposure. Significance was tested by two-way ANOVA analysis followed by a multiple comparisons test (*P < 0.05). Scale bars: 20 μm.
We followed this up by expression analysis of the key effectors of NRF2 including NQO1, HMOX1, GCLC, GCLM, MT1A, and PRDX1. Although these effectors are known to work downstream of NRF2, we sought to look at dynamic changes in the effector response with time. We compared changes in the pathway in both passage 6 hESC-derived RPE and passage 4 iPSC-derived RPE. Upon analyzing the acute phase of the response at 24 hours after exposure to 160 μM PQ, NQO1 and HMOX1 show significant up-regulation (3- to 4-fold) (Supplementary Fig. S2). Following exposure over 1 week, during the early stages of PQ-induced chronic stress, NQO1 and HMOX1 are still the first line of effectors with a 9- to 16-fold increase in expression. This was associated with a 3- to 4-fold up-regulation of the modifier subunit of glutamate cysteine ligases (GCLM) and KEAP1 at week 1 (Fig. 3C). We did not observe any significant difference in the cellular response between the ESC and iPSC lines. As oxidative stress is continued over 3 weeks, we observed a further increase in the above genes to levels as high as 30-fold (for NQO1). In addition, the catalytic subunit of glutamate cysteine ligases (GCLC) is up-regulated as a part of chronic stress response suggesting a role in glutathione in the later stage of chronic oxidative stress (Fig. 3C). Under none of the conditions above did we see any up-regulation in either MT1A or PRDX1 (data not shown), suggesting these NRF2 effectors do not have a role to play in human RPE cell oxidative stress response. 
The p21 pathway has recently been shown to be protective during oxidative stress by competing with KEAP1 for NRF2.24 We see a consistent up-regulation of P21 in our postmitotic RPE cultures within 1 day of ROS exposure and continuing over 3 weeks (Fig. 3D; Supplementary Fig. S2), suggesting a similar role in human RPE cells. Finally, we analyzed endoplasmic reticulum (ER) stress-associated unfolded protein response (UPR) by analyzing expression of CHOP, an ATF4 effector that modulates BCL2.25 We found a significant increase in the expression at 1 and 3 weeks as the cells cope with the constant stress (Fig. 3D). 
microRNA Changes During Chronic Oxidative Stress in RPE
Small RNAs likely modulate the oxidative stress response in cells. Work by a few groups has associated certain microRNAs with RPE stress including miR-23a26 and miR-9,27 using the ARPE-19 cell line. We sought to look at microRNA changes under low (PQ 40 μM) and high (PQ 160 μM) chronic stress conditions in passage 4 human iPSC-derived RPE cells based on conditions that resulted in the lowest and highest ROS levels without significant cell death (Figs. 2A, 2A′, 2B, 2B′). We carried out microRNA profiling using quantitative RT-PCR for previously described stress regulated microRNAs (Supplementary Fig. S2). The RPE cells were either exposed to PQ 40 μM or PQ 160 μM over 3 weeks, and miRNA expression was compared with unstressed controls. We discovered a core group of miRNAs (Fig. 4A) that changes under both low and high PQ conditions. These include hsa-miR-146a and hsa-miR-29a, which were up-regulated with oxidative stress; hsa-miR-144, hsa-miR-200a, and hsa-miR-21, which were down-regulated upon oxidative stress; and hsa-miR-27b, which showed a biphasic response (i.e., up-regulated upon low chronic stress and down-regulated following high chronic stress conditions; Fig. 4B). Additionally, the degree to which the expression changed was also dependent on the severity of chronic stress (Fig. 4A). One microRNA, hsa-miR-144, was most significantly down-regulated (12- to 16-fold) under high chronic oxidative stress conditions (PQ 160 μM), and we followed up on its role in oxidative stress response. 
Figure 4
 
microRNA changes to oxidative stress in human RPE cells. (A) Heat map representing changes in microRNA expression by quantitative RT-PCR in passage 4 iPSC-derived RPE cells following 3 weeks of exposure to either low chronic stress (40 μM PQ) or high chronic oxidative stress (160 μM PQ). (B) Venn diagram representing the key up-regulated and down-regulated microRNAs under both high and low oxidative stress conditions, with hsa-miR-27b showing a biphasic response to ROS stress. (C) Quantitative RT-PCR analysis of the NRF2–KEAP1 pathway in RPE cells following overexpression of hsa-miR-144 or scrambled control. hsa-miR-144 dampens the NRF2 effector response. Significance tested by t-test (**P < 0.005 and ***P < 0.0005).
Figure 4
 
microRNA changes to oxidative stress in human RPE cells. (A) Heat map representing changes in microRNA expression by quantitative RT-PCR in passage 4 iPSC-derived RPE cells following 3 weeks of exposure to either low chronic stress (40 μM PQ) or high chronic oxidative stress (160 μM PQ). (B) Venn diagram representing the key up-regulated and down-regulated microRNAs under both high and low oxidative stress conditions, with hsa-miR-27b showing a biphasic response to ROS stress. (C) Quantitative RT-PCR analysis of the NRF2–KEAP1 pathway in RPE cells following overexpression of hsa-miR-144 or scrambled control. hsa-miR-144 dampens the NRF2 effector response. Significance tested by t-test (**P < 0.005 and ***P < 0.0005).
We carried out miRNA bioinformatics predictor analysis on multiple sites, including miRanda, PicTar, Targetscan, and miRDB, for potential targets of hsa-miR-144. These sites predicted the presence of up to four putative hsa-miR-144 binding sites in the 3′-untranslated region (UTR) of the NRF2 gene. To look into the role of hsa-miR-144 in the regulation of the NRF2 pathway in human RPE cells, we overexpressed it (or a scrambled microRNA sequence as a control) using lentiviral vectors. In the absence of PQ, the cells did not show any difference in NRF2 effector expression. However, upon PQ exposure, the hsa-miR-144–overexpressing cells had a smaller effector response (GCLC, GCLM, and NQO1) to oxidative stress compared with scrambled controls (Fig. 4C), confirming the hsa-miR-144 regulation of the NRF2 pathway in RPE cells. 
Modulation by Small Molecule Ai-1
Because the NRF2 pathway plays a central role in the antioxidant response, activating this pathway should promote survival upon severe oxidant challenge. In order to identify potential NRF2 activators, we treated the cultures with a few previously characterized compounds that may drive the NRF2 pathway. These included N-acetyl-cysteine,28,29 Oltipraz,30,31 and Ai-1.32 Of these, Ai-1 resulted in maximal NRF2 effector activation in passage 6 hESC-derived RPE cells (Fig. 5C; Supplementary Fig. S2). Ai-1 is known to react with Cys151 of KEAP1, resulting in NRF2 stabilization, nuclear translocation, and transcriptional activation.32 In unstressed RPE cells, Ai-1 drove nuclear translocation of NRF2 protein (Figs. 5A, 5B) and expression of all assayed effector genes (Fig. 5C). Upon cotreating the cells with Ai-1 and PQ, Ai-1 interestingly repressed hyper-activation of the NRF2 pathway (Fig. 5D) while promoting survival of RPE cells upon treatment with a previously described toxic dose (PQ 640 μM) (Figs. 5E, 5E′, 5F, 5F′). Ai-1 did not affect cellular ROS levels induced by PQ nor did it induce them on its own when assessed by DCF assay (data not shown). The above data confirm a primary role of the NRF2–KEAP1 pathway in the oxidative stress response in RPE cells and that the pathway is tightly regulated to promote cell survival. 
Figure 5
 
Ai-1 modulates the NRF2 pathways and promotes cytoprotection against ROS challenge. (A, B) The ICC analysis for NRF2 showing nuclear translocation of the protein following Ai-1 exposure in the absence of ROS stress. (C) Quantitative RT-PCR analysis showing that Ai-1 drives NRF2 effectors. (D) Graph representing the quantitative RT-PCR changes in Ai-1–treated cells exposed to PQ160. The Ai-1–treated cells have a lower activation of the NRF2–KEAP2 pathway and the UPR pathway. (E, E′) Brightfield images of RPE cells treated to a severe oxidant challenge (640 μM PQ for 2 days). The Ai-1–treated cells survive the stress and maintain morphology and pigmentation. (F, F′) The TUNEL analysis comparing untreated, Ai-1–treated, PQ 640 μM–exposed, and Ai-1– and PQ 640 μM–exposed passage 6 hESC-derived RPE cells at 3 days following exposure. The result are quantified in (F′), and significance was tested using 1-way ANOVA. Scale bars: 20 μm.
Figure 5
 
Ai-1 modulates the NRF2 pathways and promotes cytoprotection against ROS challenge. (A, B) The ICC analysis for NRF2 showing nuclear translocation of the protein following Ai-1 exposure in the absence of ROS stress. (C) Quantitative RT-PCR analysis showing that Ai-1 drives NRF2 effectors. (D) Graph representing the quantitative RT-PCR changes in Ai-1–treated cells exposed to PQ160. The Ai-1–treated cells have a lower activation of the NRF2–KEAP2 pathway and the UPR pathway. (E, E′) Brightfield images of RPE cells treated to a severe oxidant challenge (640 μM PQ for 2 days). The Ai-1–treated cells survive the stress and maintain morphology and pigmentation. (F, F′) The TUNEL analysis comparing untreated, Ai-1–treated, PQ 640 μM–exposed, and Ai-1– and PQ 640 μM–exposed passage 6 hESC-derived RPE cells at 3 days following exposure. The result are quantified in (F′), and significance was tested using 1-way ANOVA. Scale bars: 20 μm.
Discussion
In this report, we present a very useful in vitro model system to understand the evolution of changes in response to chronic oxidative stress using human pluripotent stem cell (ESC and iPSC)-derived RPE. Using this system, we demonstrate previous undescribed dynamic changes in the NRF2–KEAP1 pathway components in human RPE cells, as well as microRNA changes in response to oxidative stress over weeks of exposure (Fig. 6). Finally, we show protective activity using a novel NRF2 pathway partial agonist following severe oxidant challenge. The model has potential value both for disease modeling and in drug discovery as highlighted in the results above. 
Figure 6
 
Model of NRF2–KEAP1 pathway and regulators involved in the oxidative stress response in human RPE cells.
Figure 6
 
Model of NRF2–KEAP1 pathway and regulators involved in the oxidative stress response in human RPE cells.
Our study is one of the first to look at the RPE cellular response to weeks of chronic stress. The NRF2–KEAP1 pathway is a master regulator of the protective response to oxidative stress in a number of systems. Following NRF2's nuclear translocation during ROS stress, it binds to the ARE in the promoter of genes coding for various effectors and promoting cell survival under these conditions (Fig. 6). The Nrf2−/− mice are viable and show age-associated RPE degeneration and some of the hallmarks of AMD including drüsen formation and choroidal neovascularization.33 A recent report in aging mice shows that NRF2 response to oxidative stress is impaired in aged mice even though there was higher NRF2 effector gene expression by 15 months of age.34 Our results shows that response to chronic ROS accumulation, as experienced by aging RPE, is evolving within the first phase due a subset of effectors, namely NQO1 and HMOX1, whereas a later phase response following weeks of stress involves driving the glutathione pathway (Fig. 3). Interestingly, the pathway does not drive certain effectors including MT1A and PRDX1 in RPE cells. 
Additionally, using hESC- and iPSC-derived RPE provides us with the advantage of access to clinically relevant samples without the issues associated with paucity of adequate primary human RPE tissue and its limited proliferation potential. This gets us around extrapolating data from transformed RPE cell lines that may not respond in the same way as naïve cells35 (Fig. 2C). Although stem cell–derived tissue may not show all the typical hallmarks of the age-associated diseases, these cells are still extremely useful for studying basic mechanisms and pathways involved in age-associated disorders (e.g., chronic ROS stress described here). Understanding of these mechanisms is critical to discovery of therapeutic agents to stall the disorder or preferably reverse it. A case in point is a recent report using iPSC-derived RPE from different ARMS2/HTRA haplotypes that could show increased susceptibility to N-retinylidene-N-retinylethanolamine stress in lines from the AMD-associated risk haplotype.8 Our study using healthy donor iPSC-derived RPE provide us with a benchmark for future studies comparing response to chronic stress in iPSC lines from AMD patients. 
Recent papers have suggested a role of microRNAs, small RNA processing enzyme Dicer1, and Alu RNAs in the pathogenesis of AMD.36,37 Nuclear factor-κB–regulated microRNAs are up-regulated in both AMD and Alzheimer's disease, and certain miRNAs are known to target the 3′-UTR of CFH, thereby down-regulating CFH.10,38 Mir23a has also been found to be decreased in RPE cells from AMD donor eyes and has been associated with cellular resistance to oxidative stress.26 A reduction in Dicer1 and a concomitant rise in Alu RNA has also been linked to RPE degeneration in AMD.36 There have been a couple of studies profiling microRNA in the RPE in response to stress.26,27,39 Most of these studies were carried out using the ARPE-19 line, which does not fully mature and pigment as native or stem cell–derived RPE and may respond differently to the stressors, as we show above. Indeed, we identified a unique microRNA signature under low and high oxidative stress conditions. We report the detection of hsa-miR-144 as a highly down-regulated microRNA in human RPE cells under high chronic oxidative stress; hsa-miR-144 has multiple NRF2 binding sites in the 3′-UTR and has previously been described to directly regulate its expression in erythrocytes.40 In our cells, it is likely that the regulation of NRF2 is post-translational repression as described in a previous report in blood cells.40 
Another interesting finding in our study was the up-regulation of p21 in the absence of any appreciable apoptosis or proliferation. The p21 pathway has been shown to be protective during oxidative stress in other systems.24,41 A study by Chen et al.24 suggests that p21 does this by up-regulating NRF2 effectors in the human cancer cell line HCT116, and p21 competes with KEAP1 for NRF2 binding, thereby preventing ubiquitination of NRF2. Thus, it is likely that in the postmitotic RPE cells, p21 up-regulation serves to modulate NRF2 oxidative stress response (Fig. 6). Similarly, PQ-induced oxidative stress drives UPR-associated CHOP expression. Here again, CHOP, which ordinarily functions as a proapoptotic gene during ER stress, is required here for NRF2 up-regulation as recently observed in cigarette smoke–induced ROS stress experiments in ARPE-19 cells.42 
Our results also identify a NRF2 activator, Ai-1, as an effective protective compound against severe oxidative challenge. Other NRF2 activators have previously been shown to have protective effects in RPE cell lines.4345 These are thought to act either by boosting the GSH pathway43 or the HMOX1 pathway.45 Our data suggest that, although Ai-1 drives common effectors of NRF2, it acts as a partial agonist in the presence of ROS stress. By doing so, it prevents hyper-activation of the pathway that requires tight regulation. Although NRF2 is a critical cytoprotective pathway against various stressors, including oxidative and xenobiotic, dysregulation of NRF2 signaling is also associated with oncogenic potential and malignant tumors.46,47 Thus, it follows that a therapeutic agent that restores a more physiologic level of NRF2 activation may serve as a good drug for ROS-associated disorders including age-related macular degeneration. 
Acknowledgments
Supported by grants from the California Institute for Regenerative Medicine (CIRM RB4-05785) and the Foundation for Retinal Research. 
Disclosure: T.Y. Garcia, None; M. Gutierrez, None; J. Reynolds, None; D.A. Lamba, None 
References
Jarrett SG, Boulton ME. Consequences of oxidative stress in age-related macular degeneration. Mol Aspects Med. 2012; 33: 399–417.
Cai J, Nelson KC, Wu M, Sternberg P,Jr, Jones DP. Oxidative damage and protection of the RPE. Prog Retin Eye Res. 2000; 19: 205–221.
Yu DY, Cringle SJ. Oxygen distribution and consumption within the retina in vascularised and avascular retinas and in animal models of retinal disease. Prog Retin Eye Res. 2001; 20: 175–208.
Okita K, Ichisaka T, Yamanaka S. Generation of germline-competent induced pluripotent stem cells. Nature. 2007; 448: 313–317.
Takahashi K, Okita K, Nakagawa M, Yamanaka S. Induction of pluripotent stem cells from fibroblast cultures. Nat Protoc. 2007; 2: 3081–3089.
Yu J, Vodyanik MA, Smuga-Otto K, et al. Induced pluripotent stem cell lines derived from human somatic cells. Science. 2007; 318: 1917–1920.
Singh R, Shen W, Kuai D, et al. iPS cell modeling of Best disease: insights into the pathophysiology of an inherited macular degeneration. Hum Mol Genet. 2013; 22: 593–607.
Yang J, Li Y, Chan L, et al. Validation of genome-wide association study (GWAS)-identified disease risk alleles with patient-specific stem cell lines. Hum Mol Genet. 2014; 23: 3445–3455.
Chang YC, Chang WC, Hung KH, et al. The generation of induced pluripotent stem cells for macular degeneration as a drug screening platform: identification of curcumin as a protective agent for retinal pigment epithelial cells against oxidative stress. Front Aging Neurosci. 2014; 6: 191.
Dunn KC, Aotaki-Keen AE, Putkey FR, Hjelmeland LM. ARPE-19, a human retinal pigment epithelial cell line with differentiated properties. Exp Eye Res. 1996; 62: 155–169.
Johnson LV, Forest DL, Banna CD, et al. Cell culture model that mimics drusen formation and triggers complement activation associated with age-related macular degeneration. Proc Natl Acad Sci U S A. 2011; 108: 18277–18282.
Rabin DM, Rabin RL, Blenkinsop TA, Temple S, Stern JH. Chronic oxidative stress upregulates Drusen-related protein expression in adult human RPE stem cell-derived RPE cells: a novel culture model for dry AMD. Aging (Albany NY). 2013; 5: 51–66.
Okita K, Matsumura Y, Sato Y, et al. A more efficient method to generate integration-free human iPS cells. Nat Methods. 2011; 8: 409–412.
Grassmann F, Fritsche LG, Keilhauer CN, Heid IM, Weber BH. Modelling the genetic risk in age-related macular degeneration. PLoS One. 2012; 7: e37979.
Mello E, Falsini B, Zuppi C, Giardina B, Concolino P, Capoluongo E. Rapid detection of CFH (p.Y402H) and ARMS2 (p.A69S) polymorphisms in age-related macular degeneration using high-resolution melting analysis. Clin Chem Lab Med. 2012; 50: 1031–1034.
Lamba DA, Gust J, Reh TA. Transplantation of human embryonic stem cell-derived photoreceptors restores some visual function in Crx-deficient mice. Cell Stem Cell. 2009; 4: 73–79.
Lamba DA, Karl MO, Ware CB, Reh TA. Efficient generation of retinal progenitor cells from human embryonic stem cells. Proc Natl Acad Sci U S A. 2006; 103: 12769–12774.
Lamba DA, McUsic A, Hirata RK, Wang PR, Russell D, Reh TA. Generation, purification and transplantation of photoreceptors derived from human induced pluripotent stem cells. PLoS One. 2010; 5: e8763.
Lamba DA, Reh TA. Microarray characterization of human embryonic stem cell–derived retinal cultures. Invest Ophthalmol Vis Sci. 2011; 52: 4897–4906.
Maminishkis A, Miller SS. Experimental models for study of retinal pigment epithelial physiology and pathophysiology. J Vis Exp. 2010; 45: 2032.
Croze RH, Buchholz DE, Radeke MJ, et al. ROCK inhibition extends passage of pluripotent stem cell-derived retinal pigmented epithelium. Stem Cells Translat Med. 2014; 3: 1066–1078.
Blanco-Ayala T, Anderica-Romero AC, Pedraza-Chaverri J. New insights into antioxidant strategies against paraquat toxicity. Free Radic Res. 2014; 48: 623–640.
Suzuki T, Motohashi H, Yamamoto M. Toward clinical application of the Keap1-Nrf2 pathway. Trends Pharmacol Sci. 2013; 34: 340–346.
Chen W, Sun Z, Wang XJ, et al. Direct interaction between Nrf2 and p21(Cip1/WAF1) upregulates the Nrf2-mediated antioxidant response. Mol Cell. 2009; 34: 663–673.
Hetz C. The unfolded protein response: controlling cell fate decisions under ER stress and beyond. Nat Rev Mol Cell Biol. 2012; 13: 89–102.
Lin H, Qian J, Castillo AC, et al. Effect of miR-23 on oxidant-induced injury in human retinal pigment epithelial cells. Invest Ophthalmol Vis Sci. 2011; 52: 6308–6314.
Kutty RK, Samuel W, Jaworski C, et al. MicroRNA expression in human retinal pigment epithelial (ARPE-19) cells: increased expression of microRNA-9 by N-(4-hydroxyphenyl)retinamide. Mol Vis. 2010; 16: 1475–1486.
Zhang L, Zhu Z, Liu J, Zhu Z, Hu Z. Protective effect of N-acetylcysteine (NAC) on renal ischemia/reperfusion injury through Nrf2 signaling pathway. J Recept Signal Transduct Res. 2014; 34: 396–400.
Ji L, Liu R, Zhang XD, et al. N-acetylcysteine attenuates phosgene-induced acute lung injury via up-regulation of Nrf2 expression. Inhal Toxicol. 2010; 22: 535–542.
Merrell MD, Augustine LM, Slitt AL, Cherrington NJ. Induction of drug metabolism enzymes and transporters by oltipraz in rats. J Biochem Mol Toxicol. 2008; 22: 128–135.
Yu Z, Shao W, Chiang Y, et al. Oltipraz upregulates the nuclear factor (erythroid-derived 2)-like 2 [corrected](NRF2) antioxidant system and prevents insulin resistance and obesity induced by a high-fat diet in C57BL/6J mice. Diabetologia. 2011; 54: 922–934.
Hur W, Sun Z, Jiang T, et al. A small-molecule inducer of the antioxidant response element. Chem Biol. 2010; 17: 537–547.
Zhao Z, Chen Y, Wang J, et al. Age-related retinopathy in NRF2-deficient mice. PLoS One. 2011; 6: e19456.
Sachdeva MM, Cano M, Handa JT. Nrf2 signaling is impaired in the aging RPE given an oxidative insult. Exp Eye Res. 2014; 119: 111–114.
Ablonczy Z, Dahrouj M, Tang PH, et al. Human retinal pigment epithelium cells as functional models for the RPE in vivo. Invest Ophthalmol Vis Sci. 2011; 52: 8614–8620.
Kaneko H, Dridi S, Tarallo V, et al. DICER1 deficit induces Alu RNA toxicity in age-related macular degeneration. Nature. 2011; 471: 325–330.
Lukiw WJ, Surjyadipta B, Dua P, Alexandrov PN. Common micro RNAs (miRNAs) target complement factor H (CFH) regulation in Alzheimer's disease (AD) and in age-related macular degeneration (AMD). Int J Biochem Mol Biol. 2012; 3: 105–116.
Lukiw WJ, Alexandrov PN. Regulation of complement factor H (CFH) by multiple miRNAs in Alzheimer's disease (AD) brain. Mol Neurobiol. 2012; 46: 11–19.
Howell JC, Chun E, Farrell AN, et al. Global microRNA expression profiling: curcumin (diferuloylmethane) alters oxidative stress-responsive microRNAs in human ARPE-19 cells. Mol Vis. 2013; 19: 544–560.
Sangokoya C, Telen MJ, Chi JT. microRNA miR-144 modulates oxidative stress tolerance and associates with anemia severity in sickle cell disease. Blood. 2010; 116: 4338–4348.
O'Reilly MA, Staversky RJ, Watkins RH, et al. The cyclin-dependent kinase inhibitor p21 protects the lung from oxidative stress. Am J Respir Cell Mol Biol. 2001; 24: 703–710.
Huang C, Wang JJ, Ma JH, Jin C, Yu Q, Zhang SX. Activation of the UPR protects against cigarette smoke-induced RPE apoptosis through up-regulation of Nrf2. J Biol Chem. 2015; 290: 5367–5380.
Nelson KC, Armstrong JS, Moriarty S, et al. Protection of retinal pigment epithelial cells from oxidative damage by oltipraz, a cancer chemopreventive agent. Invest Ophthalmol Vis Sci. 2002; 43: 3550–3554.
Gao X, Talalay P. Induction of phase 2 genes by sulforaphane protects retinal pigment epithelial cells against photooxidative damage. Proc Natl Acad Sci U S A. 2004; 101: 10446–10451.
Zhang H, Liu YY, Jiang Q, et al. Salvianolic acid A protects RPE cells against oxidative stress through activation of Nrf2/HO-1 signaling. Free Radical Biol Med. 2014; 69: 219–228.
Shelton P, Jaiswal AK. The transcription factor NF-E2-related factor 2 (Nrf2): a protooncogene? FASEB J. 2013; 27: 414–423.
Geismann C, Arlt A, Sebens S, Schafer H. Cytoprotection “gone astray”: Nrf2 and its role in cancer. Onco Targets Ther. 2014; 7: 1497–1518.
Figure 1
 
Generation and characterization of RPE cells from human iPSCs. (A) Passage 7 human iPSC-derived RPE cells in culture showing typical cobblestone pattern and pigmentation. Cells coexpress OTX2 (B) and MITF (C), as well as RPE tight junction marker ZO-1 (F) on ICC. (G) Quantitative RT-PCR confirms high levels of expression of a number of other RPE cell-specific genes at 3 months of differentiation (at passage 6) compared with undifferentiated iPS cells. (D, E) The transmission electron microscopy (TEM) analysis of RPE cells showing typical morphologic features such as basal nuclei, apical microvilli, and pigmentary granules. Scale bars: 20 μm (AC, F).
Figure 1
 
Generation and characterization of RPE cells from human iPSCs. (A) Passage 7 human iPSC-derived RPE cells in culture showing typical cobblestone pattern and pigmentation. Cells coexpress OTX2 (B) and MITF (C), as well as RPE tight junction marker ZO-1 (F) on ICC. (G) Quantitative RT-PCR confirms high levels of expression of a number of other RPE cell-specific genes at 3 months of differentiation (at passage 6) compared with undifferentiated iPS cells. (D, E) The transmission electron microscopy (TEM) analysis of RPE cells showing typical morphologic features such as basal nuclei, apical microvilli, and pigmentary granules. Scale bars: 20 μm (AC, F).
Figure 2
 
Paraquat induced ROS accumulation in RPE cells. (A) The TUNEL analysis for dose-dependent toxicity of PQ in passage 6 hESC-derived RPE cells. Cells counterstained with DAPI. The PQ concentrations greater than 320 μM result in significantly higher cell death at day 3 following PQ exposure. Results are quantified and significance is tested using 1-way ANOVA in (A′). (B) The DCF-DA assay for dose-dependant ROS accumulation in passage 6 hESC-derived RPE cells 2 days following PQ exposure. The quantification of DCF fluorescence from 10 to 15 cells per condition from three different regions in the well is shown in (B′). Significance was tested using 1-way ANOVA. (C) Brightfield images comparing cellular morphology of unstressed and 160 μM PQ exposed confluent ARPE-19 cells, mature hESC-derived RPE cells (passage 6), and immature unpigmented confluent hESC-derived RPE cells (passage 7). The ARPE-19 and immature cells showed significant morphologic changes within 7 days of PQ exposure, whereas mature pigmented cells did not (arrowheads highlight spindle-shaped appearance of stressed ARPE-19 lines). Scale bars: 20 μm.
Figure 2
 
Paraquat induced ROS accumulation in RPE cells. (A) The TUNEL analysis for dose-dependent toxicity of PQ in passage 6 hESC-derived RPE cells. Cells counterstained with DAPI. The PQ concentrations greater than 320 μM result in significantly higher cell death at day 3 following PQ exposure. Results are quantified and significance is tested using 1-way ANOVA in (A′). (B) The DCF-DA assay for dose-dependant ROS accumulation in passage 6 hESC-derived RPE cells 2 days following PQ exposure. The quantification of DCF fluorescence from 10 to 15 cells per condition from three different regions in the well is shown in (B′). Significance was tested using 1-way ANOVA. (C) Brightfield images comparing cellular morphology of unstressed and 160 μM PQ exposed confluent ARPE-19 cells, mature hESC-derived RPE cells (passage 6), and immature unpigmented confluent hESC-derived RPE cells (passage 7). The ARPE-19 and immature cells showed significant morphologic changes within 7 days of PQ exposure, whereas mature pigmented cells did not (arrowheads highlight spindle-shaped appearance of stressed ARPE-19 lines). Scale bars: 20 μm.
Figure 3
 
Characterization of the dynamic changes in the NRF2–KEAP1 pathway following chronic oxidative stress. (A, B) The ICC analysis for NRF2 showing nuclear translocation of the protein following PQ exposure in passage 5 iPSC-derived RPE cells. (C) Quantitative RT-PCR analysis of RPE cell cultures following 1 and 3 weeks of constant exposure to 160 μM PQ of either five replicates from passage 4 iPSC-RPE or six replicates from passage 6 hESC-RPE. NRF2 effectors, NQO1, HMOX1, GCLC, GCLM, and KEAP1, have varying degrees of response to the duration of ROS stress, with NQO1 and GCLC signficantly further up-regulated at week 3. (D) A component of the UPR pathway, CHOP, is also up-regulated within a week of ROS accumulation. In addition, p21 is up-regulated following ROS exposure. Significance was tested by two-way ANOVA analysis followed by a multiple comparisons test (*P < 0.05). Scale bars: 20 μm.
Figure 3
 
Characterization of the dynamic changes in the NRF2–KEAP1 pathway following chronic oxidative stress. (A, B) The ICC analysis for NRF2 showing nuclear translocation of the protein following PQ exposure in passage 5 iPSC-derived RPE cells. (C) Quantitative RT-PCR analysis of RPE cell cultures following 1 and 3 weeks of constant exposure to 160 μM PQ of either five replicates from passage 4 iPSC-RPE or six replicates from passage 6 hESC-RPE. NRF2 effectors, NQO1, HMOX1, GCLC, GCLM, and KEAP1, have varying degrees of response to the duration of ROS stress, with NQO1 and GCLC signficantly further up-regulated at week 3. (D) A component of the UPR pathway, CHOP, is also up-regulated within a week of ROS accumulation. In addition, p21 is up-regulated following ROS exposure. Significance was tested by two-way ANOVA analysis followed by a multiple comparisons test (*P < 0.05). Scale bars: 20 μm.
Figure 4
 
microRNA changes to oxidative stress in human RPE cells. (A) Heat map representing changes in microRNA expression by quantitative RT-PCR in passage 4 iPSC-derived RPE cells following 3 weeks of exposure to either low chronic stress (40 μM PQ) or high chronic oxidative stress (160 μM PQ). (B) Venn diagram representing the key up-regulated and down-regulated microRNAs under both high and low oxidative stress conditions, with hsa-miR-27b showing a biphasic response to ROS stress. (C) Quantitative RT-PCR analysis of the NRF2–KEAP1 pathway in RPE cells following overexpression of hsa-miR-144 or scrambled control. hsa-miR-144 dampens the NRF2 effector response. Significance tested by t-test (**P < 0.005 and ***P < 0.0005).
Figure 4
 
microRNA changes to oxidative stress in human RPE cells. (A) Heat map representing changes in microRNA expression by quantitative RT-PCR in passage 4 iPSC-derived RPE cells following 3 weeks of exposure to either low chronic stress (40 μM PQ) or high chronic oxidative stress (160 μM PQ). (B) Venn diagram representing the key up-regulated and down-regulated microRNAs under both high and low oxidative stress conditions, with hsa-miR-27b showing a biphasic response to ROS stress. (C) Quantitative RT-PCR analysis of the NRF2–KEAP1 pathway in RPE cells following overexpression of hsa-miR-144 or scrambled control. hsa-miR-144 dampens the NRF2 effector response. Significance tested by t-test (**P < 0.005 and ***P < 0.0005).
Figure 5
 
Ai-1 modulates the NRF2 pathways and promotes cytoprotection against ROS challenge. (A, B) The ICC analysis for NRF2 showing nuclear translocation of the protein following Ai-1 exposure in the absence of ROS stress. (C) Quantitative RT-PCR analysis showing that Ai-1 drives NRF2 effectors. (D) Graph representing the quantitative RT-PCR changes in Ai-1–treated cells exposed to PQ160. The Ai-1–treated cells have a lower activation of the NRF2–KEAP2 pathway and the UPR pathway. (E, E′) Brightfield images of RPE cells treated to a severe oxidant challenge (640 μM PQ for 2 days). The Ai-1–treated cells survive the stress and maintain morphology and pigmentation. (F, F′) The TUNEL analysis comparing untreated, Ai-1–treated, PQ 640 μM–exposed, and Ai-1– and PQ 640 μM–exposed passage 6 hESC-derived RPE cells at 3 days following exposure. The result are quantified in (F′), and significance was tested using 1-way ANOVA. Scale bars: 20 μm.
Figure 5
 
Ai-1 modulates the NRF2 pathways and promotes cytoprotection against ROS challenge. (A, B) The ICC analysis for NRF2 showing nuclear translocation of the protein following Ai-1 exposure in the absence of ROS stress. (C) Quantitative RT-PCR analysis showing that Ai-1 drives NRF2 effectors. (D) Graph representing the quantitative RT-PCR changes in Ai-1–treated cells exposed to PQ160. The Ai-1–treated cells have a lower activation of the NRF2–KEAP2 pathway and the UPR pathway. (E, E′) Brightfield images of RPE cells treated to a severe oxidant challenge (640 μM PQ for 2 days). The Ai-1–treated cells survive the stress and maintain morphology and pigmentation. (F, F′) The TUNEL analysis comparing untreated, Ai-1–treated, PQ 640 μM–exposed, and Ai-1– and PQ 640 μM–exposed passage 6 hESC-derived RPE cells at 3 days following exposure. The result are quantified in (F′), and significance was tested using 1-way ANOVA. Scale bars: 20 μm.
Figure 6
 
Model of NRF2–KEAP1 pathway and regulators involved in the oxidative stress response in human RPE cells.
Figure 6
 
Model of NRF2–KEAP1 pathway and regulators involved in the oxidative stress response in human RPE cells.
×
×

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.

×