February 2011
Volume 52, Issue 2
Free
Retinal Cell Biology  |   February 2011
Differential Effects of PPARγ Ligands on Oxidative Stress–Induced Death of Retinal Pigmented Epithelial Cells
Author Affiliations & Notes
  • Gerard A. Rodrigues
    From Biological Sciences, Allergan, Inc., Irvine, California; and
  • Florence Maurier-Mahé
    ExonHit Therapeutics SA, Paris, France.
  • Dixie-Lee Shurland
    From Biological Sciences, Allergan, Inc., Irvine, California; and
  • Anne Mclaughlin
    From Biological Sciences, Allergan, Inc., Irvine, California; and
  • Keith Luhrs
    From Biological Sciences, Allergan, Inc., Irvine, California; and
  • Emeline Throo
    ExonHit Therapeutics SA, Paris, France.
  • Laurence Delalonde-Delaunay
    ExonHit Therapeutics SA, Paris, France.
  • Diego Pallares
    ExonHit Therapeutics SA, Paris, France.
  • Fabien Schweighoffer
    ExonHit Therapeutics SA, Paris, France.
  • John Donello
    From Biological Sciences, Allergan, Inc., Irvine, California; and
Investigative Ophthalmology & Visual Science February 2011, Vol.52, 890-903. doi:10.1167/iovs.10-5715
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Gerard A. Rodrigues, Florence Maurier-Mahé, Dixie-Lee Shurland, Anne Mclaughlin, Keith Luhrs, Emeline Throo, Laurence Delalonde-Delaunay, Diego Pallares, Fabien Schweighoffer, John Donello; Differential Effects of PPARγ Ligands on Oxidative Stress–Induced Death of Retinal Pigmented Epithelial Cells. Invest. Ophthalmol. Vis. Sci. 2011;52(2):890-903. doi: 10.1167/iovs.10-5715.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose.: To investigate the role of the peroxisome proliferator-activated receptor (PPAR)-γ in modulating retinal pigmented epithelium (RPE) responses to oxidative stress.

Methods.: ARPE-19 cells were treated with the oxidant, t-butylhydroperoxide (tBH) to induce apoptosis. Cells pretreated with synthetic PPARγ agonists of the antidiabetic thiazolidinediones class before tBH challenge were assessed for viability and, by microarray analysis, for effects on gene expression.

Results.: Treatment of ARPE-19 cells with tBH resulted in a loss of viability and global changes in the pattern of gene expression. PPARγ ligands were found to have differential modulatory effects on tBH-induced apoptosis of RPE cells. Whereas rosiglitazone and pioglitazone potentiated cell death, troglitazone acted as a potent cytoprotective agent. Downregulation of PPARγ expression by an siRNA resulted in enhanced cell death in response to tBH treatment and blocked the cytoprotective effect of troglitazone consistent with a role of PPARγ in mediating this response. Microarray analysis revealed that while rosiglitazone and pioglitazone had little effect on gene changes induced by tBH treatment, troglitazone dramatically reduced the number of changes caused by oxidative stress. A unique subset of genes that were deregulated by tBH and selectively normalized by troglitazone were identified.

Conclusions.: These findings demonstrate that PPARγ agonists can have differential effects on RPE survival in response to oxidative stress. Oxidative stress leads to deregulation of a large set of genes in ARPE-19 cells. A specific subset of these genes can be selectively modulated by troglitazone and represent potential novel targets for cytoprotective therapies.

Peroxisome proliferator-activated receptors (PPARs) comprise a family of nuclear receptor transcription factors that participate in the maintenance of lipid and glucose homeostasis and the control of cell growth and differentiation. 1,2 These receptors include three family members, PPARα, PPARβ/δ, and PPARγ. All three PPARs are activated by fatty acids and fatty acid-derived eicosanoids. 1,2 In addition, oxidized fatty acids, including oxidized metabolites of linoleic acid, arachidonic acid, and docosahexaenoic acid, were identified as PPARγ agonists. 3,4 Pharmacologic activation of PPARγ causes changes in metabolic pathways in adipose tissue, skeletal muscle, and liver that lead to increased insulin sensitivity, lower blood glucose levels, and triglyceride clearance. 1,2 The thiazolidinedione class of PPARγ ligands, which includes troglitazone, rosiglitazone, pioglitazone, and ciglitazone are synthetic agonists that can induce these changes and, as a consequence, are currently in clinical use for the treatment of non–insulin-dependent diabetes mellitus (NIDDM). 
In addition to its role in carbohydrate metabolism, PPARγ has been shown to regulate other cellular functions, including lipid metabolism. 5,6 PPARγ mediates the uptake of oxidized low-density lipoprotein (oxLDL) in macrophages and promotes the changes in gene expression induced by oxLDL generated by oxidative stress. 4,7 Activation of PPARγ in response to oxLDL has been shown to induce expression of the scavenger receptor CD36 and the nuclear receptor LXR. 4,8 11 While CD36 has been genetically linked to lipid accumulation in macrophages, LXR activates expression of ABCA1, a protein involved in cholesterol efflux. 9,12 Importantly, ABCA1 is mutated in patients with Tangier disease, a disorder characterized by the accumulation of cholesterol esters in various tissues. 12 Moreover, transplantation of PPARγ-null bone marrow into LDLR−/− mice results in a significant increase in atherosclerosis, consistent with the notion that the efflux pathway is dominant in vivo. 9 Thus, PPARγ coordinates a complex response to oxLDL that involves uptake, processing, and efflux of cholesterol through ABCA1. 
Cells of the retinal pigmented epithelium (RPE) play a critical role in sustaining the neurosensory retina in the eye. The RPE is a monolayer of cuboidal cells located adjacent to the neural retina. 13 The major function of RPE cells is to support photoreceptors by controlling the exchange of nutrients, waste products, ions, and gases between the choroidal blood vessels and overlying photoreceptors. 13 In addition, the RPE is involved in phagocytosing shed outer segment membrane discs for photoreceptor renewal and in the shuttling of retinoids required for visual pigment synthesis to photoreceptors. 14 16 CD36 has been shown to participate in phagocytosis of rod outer segments by RPE cells. 17  
The loss of cells in the RPE and the of photoreceptors due to oxidative stress and chronic inflammation is believed to play a role in the pathogenesis of age-related macular degeneration (ARMD). 18 20 The retina, with its high proportion of poly-unsaturated fatty acids (PUFAs), intense exposure to light, robust metabolic activity, and high oxygen tension in the macular region is particularly susceptible to oxidative damage. Lipid peroxidation of PUFAs found in the phagocytosed rod outer segments can directly lead to RPE damage. 20 In addition, the phagocytosis of outer segments rich in PUFAs may contribute to age-related increases in the formation of lipofuscin, a lipid–protein aggregate found in RPE that compromises its function. 21 In addition to PUFAs, cholesterol has been shown to accumulate with age in Bruch's membrane in human eyes. 22 Cholesterol esters are prone to oxidation that generates oxysterols that have cytotoxic and proinflammatory properties. 23 Indeed, oxidized low-density lipoprotein (oxLDL) has been shown to accumulate in Bruch's membrane of patients with ARMD. 24 RPE cells have been shown to internalize oxLDL both in vitro and in vivo. 25 The accumulation of oxLDL in the RPE can alter phagosome maturation and impair their ability to process photoreceptor outer segments, ultimately resulting in cell death. 26 28 Moreover, exposure of RPE cells to oxLDL induces transcriptional alterations in genes related to lipid metabolism, oxidative stress, inflammation, and apoptosis. 24  
PPARγ has been shown to be expressed in the RPE, 29 and phagocytosis of photoreceptor outer segments increases PPARγ expression in these cells. 30 Thus, as oxidized lipids derived from PUFAs or cholesterol serve as ligands for PPARγ, we were interested in understanding the role of PPARγ in modulating the response of RPE cells to oxidative stress. The effects of the thiazolidinedione class of PPARγ ligands on RPE cell survival in response to challenge with t-butylhydroperoxide (tBH) were tested. In addition, we conducted microarray analysis of gene changes induced by tBH and similarly measured the effects of PPARγ agonism on these changes. 
Methods
Oxidative Stress Assay
ARPE-19 cells were maintained in DMEM-F12 medium containing 10% fetal bovine serum, and 1% antibiotic-antimycotic. The cells were plated at a density of 1 × 104 cells per well in 100 μL medium in a 96-well plate and incubated overnight at 37°C in 5% CO2. The next day the cells were washed twice with DMEM-F12 containing 0.1% charcoal dextran stripped fetal bovine serum and 1% antibiotic-antimycotic. After they were washed, the cells were starved for 24 hours in the same medium at 37°C. Twenty-four hours after starvation, tertiary-butylhydroperoxide (tBH) was added at the indicated concentrations for 3 or 6 hours. The cells were then washed twice with DMEM-F12 containing 0.1% FBS and 1% antibiotic antimycotic. Fresh medium was added, and the cells were further incubated for a total of 24 hours at 37°C from initial tBH treatment. Cell viability was measured using the MTT assay (Roche Diagnostics, Indianapolis, IN) according to the manufacturer's instructions. Briefly, 10 μL of MTT labeling reagent was added, and the cells were incubated at 37°C for 4 hours. After incubation, 100 μL solubilization solution was added and the plates were read on a plate reader at 550 nm (SpectraMax M2; Molecular Devices, Sunnyvale, CA). 
Detection of Reactive Oxygen Species
Reactive oxygen species (ROS) production was measured by using the fluorescent dye, 2′-7′-dichlorofluorescein diacetate (DCFDA; Invitrogen-Molecular Probes, Eugene, OR), as previously described. 31 ARPE-19 cells were plated at a density of 1 × 104 cells per well in a 96-well plate. The next day, the cells were starved in DMEM-F12 containing 0.1% charcoal dextran-stripped serum and 1% antibiotic-antimycotic. Twenty-four hours later, the cells were washed with HBSS without phenol red. DCF DA was diluted in HBSS and added to the cells for 30 minutes. The cells were washed with HBSS, and tBH diluted in HBSS was added to the cells at concentrations of 0, 100, 200, 300, 500, and 1000 μM for 3 hours. The plates were read in a plate reader (SpectraMax M2; Molecular Devices, Sunnyvale, CA) with an excitation wavelength of 425 nm and an emission wavelength of 520 nm. 
Caspase 3 Assay
ARPE-19 cells were plated at a density of 1 × 104 cells per well in a 96-well plate and incubated overnight at 37°C in 5% CO2. The next day, the cells were starved for 24 hours in DMEM-F12 with 0.1% stripped FBS. After starvation, the cells were treated with tBH for the times indicated in Figure 1. Caspase 3 levels were determined using a homogeneous caspase 3/7 assay (Apo-ONE; Promega, Madison, WI), according to the manufacturer's instructions. Briefly, 100 μL of the caspase-3/7 reagent was added to each well and gently mixed on a plate shaker at 300 to 500 rpm. The plates were incubated at room temperature for 3 hours. Fluorescence was measured with an excitation wavelength of 485 nm and an emission wavelength of 530 nm, after addition of the reagent. 
Figure 1.
 
Effects of tBH treatment on ARPE-19 cells. Cells were incubated with DCFDA for 30 minutes and then challenged with increasing concentrations of tBH for 3 hours, after which fluorescence intensity was measured to determine the presence of ROS (A). Cells were similarly challenged with increasing concentrations of tBH for 30 minutes and 1, 3, and 24 hours (B) or 3, 4, 6, 8, and 24 hours (C), and viability was measured 24 hours after the challenge. The percentage of viability was determined by comparison with untreated controls. Increases in caspase 3 activity were measured after 30-minute, 3-hour, and 6-hour tBH challenges (D).
Figure 1.
 
Effects of tBH treatment on ARPE-19 cells. Cells were incubated with DCFDA for 30 minutes and then challenged with increasing concentrations of tBH for 3 hours, after which fluorescence intensity was measured to determine the presence of ROS (A). Cells were similarly challenged with increasing concentrations of tBH for 30 minutes and 1, 3, and 24 hours (B) or 3, 4, 6, 8, and 24 hours (C), and viability was measured 24 hours after the challenge. The percentage of viability was determined by comparison with untreated controls. Increases in caspase 3 activity were measured after 30-minute, 3-hour, and 6-hour tBH challenges (D).
Transfections
ARPE 19 cells were plated at a density of 6 × 105 cells per 10-cm tissue culture dish, in DMEM-F12 containing 10% FBS and 1% antibiotic-antimycotic and incubated at 37°C overnight. The RNAi oligonucleotide or control oligonucleotide was added to the medium (OptiMem; Invitrogen-Gibco, Grand Island, NY). Fifteen microliters of lipophilic transfection reagent (Lipofectamine 2000; Invitrogen, Carlsbad, CA) was similarly added to an equal volume of the medium (OptiMem; Invitrogen-Gibco) and incubated at room temperature for 5 minutes. Both solutions were then mixed together and further incubated for 20 minutes at room temperature. The cells were washed once with the medium and the mixture was added to the cells and left on overnight at 37°C. Twenty-four hours after transfection the transfection mixture was removed, and fresh medium was added. 
RNA Isolation and cDNA Synthesis
Cells were washed once with PBS, lysed (Qiazol; Qiagen, Valencia, CA) and scraped into 2-mL tubes. Chloroform (200 μL/mL Qiazol) was added and mixed, and the tubes were spun (12,000g) for 15 minutes at 4°C. The aqueous phase was removed and applied to a minicolumn (RNeasy; Qiagen), washed, and eluted in water. Five milligrams was used for subsequent RT-PCR reactions. RT-PCR reactions were performed with synthesized first-strand cDNA (Superscript 111 First Strand Synthesis Supermix; Invitrogen). The RNA was combined with reaction mix, RT enzyme mix, and DEPC water and incubated at 25°C for 10 minutes. The mixture was then incubated for 30 minutes at 50°C, the reaction was terminated at 85°C for 5 minutes, and the mixture was chilled on ice. One microliter (2 U) of Escherichia coli RNase H was added to the reaction and incubated at 37°C for 20 minutes. The cDNA was then stored at −20°C. 
Quantitative PCR
Quantitative (q)PCR) was conducted with 5 μL cDNA generated by RT-PCR. qPCR master mix–UDG (Platinum SuperMix; Invitrogen) was used for amplification. Reactions were performed for 45 cycles using the following conditions: 60°C for 30 seconds, 95°C for 2 minutes, 95°C for 15 seconds, and 60°C for 40 seconds. GAPDH was used to normalize expression levels. Reactions and analysis were performed on a PCR system (Applied Biosystems 7500 Real-Time PCR System; Life Technologies Corporation, Carlsbad, CA). 
siRNA-Mediated mRNA and Protein Knockdown
A validated stealth siRNA targeting PPARγ, 5′-GCUUAUCUAUGACAGAUGUGAUCUU-3′ was obtained from Invitrogen. The siRNA was complexed to the transfection reagent (Lipofectamine 2000; Invitrogen), according to the manufacturer's recommendations and added to cell medium to a final concentration of 25 nM for 6 hours, after which the medium was changed, and the cells were further incubated for the indicated times (see Fig. 3). Validation of mRNA knockdown was conducted by qPCR with LUX primers for PPARγ (Invitrogen). Protein knockdown was verified by Western blot analysis using the anti-PPARγ antibody H-100 (sc-7196; Santa Cruz Biotechnology, Inc., Santa Cruz, CA). 
Western Blot Analysis
Cells were lysed with Triton lysis buffer (50 mM HEPES, 150 mM NaCl, 1.5 mM MgCl2, 1 mM EGTA, 10% glycerol, and 1% Triton X-100). Immunocomplexes were collected on protein A-Sepharose, washed with lysis buffer, resuspended in Laemmli sample buffer, boiled for 5 minutes, subjected to sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE), and transferred to nitrocellulose membranes. The membranes were blocked in TBST (10 mM Tris [pH 7.5], 50 mM NaCl, and 0.1% Triton X-100) containing 5% bovine serum albumin for 1 hour. They were then incubated for 1.5 hours with antibody and washed extensively with TBST. Immunoreactive proteins were detected by incubation with horseradish peroxidase-conjugated protein A. Proteins were visualized by enhanced chemiluminescence and autoradiography. 
Microarrays
A gene expression analysis (Two-Color Microarray-based Gene Expression Analysis; Agilent, Santa Clara, CA) used cyanine (Cy)-3- and Cy-5-labeled targets to measure gene expression in experimental and control samples. RNA quality and quantity were determined by spectrophotometry and by examining a microarray chip on a bioanalyzer (RNA 6000 Nanochip kit; 2100 Bioanalyzer; Agilent Technologies). All total RNA that passed the quality-control criteria were used for further analysis. cDNA synthesis, cRNA synthesis, amplification, and labeling were performed (Low RNA Input Linear Amplification kit; Agilent Technologies). Double-stranded cDNA was synthesized from RNA via MMLV reverse transcriptase. cRNA was amplified and labeled by using T7 RNA polymerase, which simultaneously amplifies the target material and incorporates Cy3- or Cy5-labeled CTP with at least a 100-fold RNA amplification rate. In one case, cRNA from treated cells was amplified with the incorporation of Cy5-CTP (fluorescent in the red region), while cRNA from control samples was labeled with Cy3-CTP (fluorescent in the green region) and purified. In another case, the same amplifications were performed but the dyes were swapped, meaning that each comparison was performed in duplicate. The labeled cRNA samples were then purified, quantified, and fragmented in the fragmentation buffer at 60°C for 30 minutes before the microarray hybridization. The treated sample and the control sample (750 ng of both) were hybridized on a whole human genome (1 × 44 K) microarray (Agilent Technologies) overnight at 65°C in a hybridization oven. The slides were then washed, stabilized, dried, and immediately scanned with a microarray scanner (Agilent Technologies). The microarray scan data were extracted with specialized software (Feature Extraction [FE] Software), and the FE files were imported into genomics software (Genomics Suite; Partek Inc., St Louis, MO). After quantile normalization across all arrays, log2 transformation of the data, and probe summarization, an ANOVA comparison of treated versus nontreated samples was used to generate P values, calculate expression changes (x-fold change; FC), and determine the significant probe lists. 
Semiquantitative RT-PCR
Reverse transcription reactions were performed (Transcriptor Reverse Transcriptase; Roche Applied Science) starting with 3 μg of total RNA. The RNA was combined with random primers and DEPC water and incubated 10 minutes at 65°C. RT reaction buffer, protector RNase inhibitor, dNTP mix, and reverse transcriptase were then added, incubated 10 minutes at 25°C and then again for 30 minutes at 55°C. The reaction was inactivated by heating to 85°C for 5 minutes and chilled on ice. The 20-μL final volume of cDNA was diluted to 100 μL with water. PCR reactions were performed using the PCR master mix (M7505; Promega). cDNA template (2 μL) and 2.5 μL of primer mix at 10 μM (final volume, 25 μL) were added to 20.5 μL of master mix, and reactions were performed for 30 cycles: 94°C for 30 seconds, 57°C for 30 seconds, and 72°C for 1 minute. Tubulin was used to normalize expression levels. PCR reactions (5 μL) were run on electrophoresis gel, and 1 μL was quantified on the bioanalyzer (DNA 1000 Nanochip kit; model 2100 Bioanalyzer; Agilent Technologies). 
Primers sequences used were CXCR4-forward: 5′-CACCATCTACTCCATCATCTTC-3′ CXCR4-reverse: 5′-GACAATACCAGGCAGGATAAGG-3′; ATF3-forward: 5′-TGAGGTT TGCCATCCAGAACAAG-3′; ATF3-reverse: 5′-GTCAACAGCCCATATGCAGGTC-3′; ADM-forward: 5′-CCTGGGTTCGCTCGCCTTC-3′; ADM-reverse: 5′-CGTGTGCTTGTGG CTTAGAAGAC-3′; CCL2-forward: 5′-GGAACCGAGAGGCTGAGACTAAC-3′; CCL2-reverse: 5′-TTCTTTGGGACACTTGCTGCTG-3′; AMOTL2- forward: 5′-CGCTACGGCAA CCTGACTGAG-3′; AMOTL2-reverse: 5′-TGTGGGAAGCTGTGGGAGGAG-3′; JUB- 5′-forward: 5′-AGTCGCCTGCTGGAGAAGTTC-3′; JUB-reverse: 5′-CATGCTGATGCCG CTGGTG-3′; Tubulin-forward: 5′-GCTGCCATTGCCACCATC-3′; Tubulin-reverse: 5′-CACACCAACCTCCTCATAATCC-3′. 
ChIP Analysis
Cells were fixed in 1% formaldehyde for 15 minutes to induce DNA and protein cross-linking. Fixation was stopped with 0.125 M glycine, and the cells were washed with PBS containing 0.5% Igepal (Sigma-Aldrich, St. Louis, MO) and collected. Cell samples were submitted to Genpathway, Inc. (San Diego, CA) for ChIP analysis. The cells were homogenized and the lysates sonicated. Generally, 5 to 30 μg of DNA in the form of sonicated chromatin was immunoprecipitated. The PPARγ antibody H-100 (sc-7196; Santa Cruz Biotechnology, Inc.) was used for immunoprecipitation. This antibody was validated for ChIP analysis with 3T3-L1 adipocyte lysates and PCR primers for known PPARγ response elements (PPREs). Protein A-agarose (Invitrogen) was used to isolate the immune complexes, followed by DNA purification. Approximately 2% to 5% of the final immunoprecipitated fraction and 25 ng of the nonimmunoprecipitated sonicated chromatin (input DNA) were subjected to qPCR. A list of primers used for qPCR is shown in Table 1. PCR primers were designed to test binding to specific PPREs in the ChIP assay. Putative PPREs were identified by scanning the sequences of each gene with MatInspector software. 32 MatInspector uses the PPRE consensus sequence CWRAWCTAGGNCAAAGGTCA, where W is A or T, R is A or G, K is G or T, and M is A or C. In addition, two degenerate consensus matrices were used based on the work of Tachibana et al. 33 : (1) WWSNRGGNMAAAGKTCA (two mismatches allowed); (2) RGGNMANAGKTCA (one mismatch allowed). The level of transcription for each DNA region tested was determined from the threshold cycle number by generating a standard curve of three to four 10-fold serial dilutions of input DNA from one of the chromatin samples. A standard curve was included in each multiwell plate and amplified in parallel with a primer pair for one of the reference genes. The average Ct values of triplicate amplifications from immunoprecipitated DNA or input DNA were compared to those from the standard curve and converted to the appropriate values. To normalize for differences in amplification efficiency for various primer-pairs using the same DNA input, we divided the values obtained from amplification of immunoprecipitated DNA by those obtained from amplification with input DNA. The resulting numbers were multiplied by 100 and referred to as normalized transcription units. Positive and negative controls, with well-characterized primer-pairs for constitutively expressed genes and untranscribed genomic regions (Untr), were included in every ChIP experiment. 
Table 1.
 
Primers Used for ChIP Analysis
Table 1.
 
Primers Used for ChIP Analysis
Gene Site Primer Sequences (5′–3′)
CXCR4 +1100 CCGCGAGCGTCTTTGAAT
TCCGAGTCCAGTTGTGATGT
CCL2 −1190 TGGGCTAGGAGAATCGAGAG
GGGAATTGGAAAGCCTGAC
CDKN1C −1000 CTGCCCAGTAGATGGAACAG
ACAGCCCCTTCCCCTATGAC
AMOTL2 −1570 CCCTTACGAAGAGAAGCTGTG
−1670* CCCAAATGCCTGAAAGACAG
JUB −1160 TTTCTCATTTCAGGTGACTGTG
TTTCCTCTACACGGCATTTC
Statistical Analysis
The data are expressed as the mean ± SD. Data were analyzed, as required, using a Student's t-test or ANOVA. P < 0.05 was considered statistically significant. 
Results
tBH Generates ROS and Induces Apoptosis of ARPE-19 Cells
Previous results have demonstrated that the chemical oxidant tBH can induce oxidative injury in cultured RPE cells. 34 To determine whether exposure to tBH resulted in accumulation of intracellular ROS, ARPE-19 cells were treated with increasing concentrations of tBH for 3 hours and the presence of ROS measured by incubating the cells with the fluorophor, 2′-7′-dichlorofluorescein diacetate (DCFDA). The results in Figure 1A show that there was a dose-dependent increase in the levels of fluorescence because of oxidation of dichlorofluorescein in ARPE-19 cells indicating an increase in the levels of ROS in the cell. To determine the doses and time course of RPE cell death induced by tBH, ARPE-19 cells were exposed to increasing doses of tBH ranging from 100 μM to 1 mM for between 30 minutes and 24 hours, after which the drug was removed. Viability was measured 24 hours after the initial addition of drug using an MTT assay. Figure 1B shows that there was no significant change in viability after exposure of the cells to tBH for 30 or 60 minutes. However, by 3 hours, only approximately 70% of the cells remained viable. The levels of cell death were at their maximum at a dose of between 200 and 300 μM. The results in Figure 1C show that after a 4-hour or longer exposure at a dose of 200 μM, tBH resulted in complete cell death. The concentrations and time required to observe loss of viability in these experiments are consistent with previous reports using tBH and ARPE-19 cells. 35,36 To determine whether the cell death observed was due to apoptosis, caspase 3 levels were measured in cells exposed to tBH. Figure 1D shows a dose-dependent increase in caspase 3 levels after tBH exposure. Increases in caspase 3 levels were seen with exposure times as short as 30 minutes, suggesting that tBH-treated cells were undergoing apoptosis. Furthermore, pretreatment of cells with the pan-caspase inhibitor, zVAD before and during tBH exposure protected cells from death, whereas a similar treatment with necrostatin, an inhibitor of programmed necrosis had no effect on ARPE-19 viability (data not shown). Thus, it appears that the ARPE-19 cells are committed to undergo apoptosis between 3 and 4 hours after exposure to tBH. 
Differential PPARγ Agonist Effects on Oxidative Stress
As oxLDL is generated through lipid peroxidation and oxLDL can activate PPARγ, we wanted to understand the role of PPARγ in modulating the response of RPE cells to oxidative stress. To address this question, ARPE-19 cells were treated with various thiazolidinedione PPARγ ligands 1 hour before challenge with 200 μM tBH as well as throughout the period of tBH exposure. Figure 2A shows that pretreatment of the cells with 1 or 10 μM troglitazone protected the cells from increasing exposure times to tBH. At the 1 μM dose of troglitazone, 80% of the cells remained viable even after a 6-hour exposure to tBH, at which time almost all the cells were dead (Figs. 1, 2A). This dose is similar to the EC50 (500 nM) for activation of PPARγ. 1 In contrast, no cytoprotection was observed at any concentration with rosiglitazone, pioglitazone, or ciglitazone (Fig. 2A). In fact, there is evidence of potentiation of cell death in the presence of tBH (Fig. 2A). Similar results were obtained when the cells were treated for 24 hours before challenge with tBH (data not shown). A survey of other PPARγ agonists including nTZDpa, telismartan, T0070907, and MCC-555 failed to identify additional cytoprotective agents. To further validate that troglitazone is cytoprotective against oxidative stress, cells were treated with a full dose range of tBH (50–1000 μM) for 3 hours after a 1-hour pretreatment and in the presence of troglitazone. Under these conditions, whereas increasing doses of tBH resulted in increased cell death, 1 μM of troglitazone resulted in an 80% to 90% survival rate throughout the dose range (Fig. 2B). The amount of ROS in the cells after a 3- or 6-hour exposure to 200 μM tBH in the presence of increasing doses of troglitazone was measured, and reductions were observed only at doses of troglitazone significantly higher than that necessary for cytoprotection (data not shown). To test whether troglitazone was required continuously during the period of tBH exposure or whether a pretreatment was sufficient for cytoprotection, we pretreated the cells with 1 μM troglitazone for 24 hours, and then the troglitazone was washed out before exposure of the cells to 300 μM tBH for 3 hours. Under these conditions, only 50% of the cells remained viable. However, a significantly higher percentage (approaching 100%) of the cells remained viable if pretreated with troglitazone (Fig. 2C). 
Figure 2.
 
Differential cytoprotective effects of glitazones on tBH-treated ARPE-19 cells. Cells were pretreated with 0.1, 1, and 10 μM troglitazone, rosiglitazone, or pioglitazone for 1 hour before challenge with 200 μM tBH for 30 minutes, 3 hours, or 6 hours (A). Preincubation of cells for 1 hour with 1 μM troglitazone protected the cells from increasing doses of tBH. An 80% to 90% survival rate was obtained in the presence of up to 1 mM tBH (B). Preincubation of cells for 24 hours with troglitazone followed by withdrawal and tBH challenge was sufficient to achieve cytoprotection (C). ***P < 0 0.001 compared with tBH treatment alone.
Figure 2.
 
Differential cytoprotective effects of glitazones on tBH-treated ARPE-19 cells. Cells were pretreated with 0.1, 1, and 10 μM troglitazone, rosiglitazone, or pioglitazone for 1 hour before challenge with 200 μM tBH for 30 minutes, 3 hours, or 6 hours (A). Preincubation of cells for 1 hour with 1 μM troglitazone protected the cells from increasing doses of tBH. An 80% to 90% survival rate was obtained in the presence of up to 1 mM tBH (B). Preincubation of cells for 24 hours with troglitazone followed by withdrawal and tBH challenge was sufficient to achieve cytoprotection (C). ***P < 0 0.001 compared with tBH treatment alone.
Troglitazone's Protective Effects Are Mediated by PPARγ
To determine whether the cytoprotective effect we observed with troglitazone is dependent on and mediated by PPARγ, we tested the effects of the compound in cells where PPARγ expression had been knocked down using RNA interference. Treatment of the cells with an siRNA to PPARγ resulted in an almost complete knockdown of mRNA expression 48 hours after transfection. Moreover, an 80% knockdown was sustained up to 8 days after transfection (Fig. 3A). To confirm knockdown of the PPARγ protein, we performed Western blot analyses 72 hours after transfection. Figure 3B shows that protein expression was knocked down by 70%. To test the effects of the PPARγ knockdown on responses to oxidative stress, we exposed the cells to increasing doses of tBH 72 hours after siRNA transfection. As shown in Figure 3C, increasing doses of tBH caused increased cell death in the untransfected cells. In the cells that had the PPARγ knockdown, a potentiation of cell death was observed (Fig. 3C). This result is consistent with the protective role of PPARγ receptor activity and with the notion that PPARγ activation by troglitazone would be protective. To confirm that this is the case, naïve and siRNA-transfected cells were pretreated with 1 μM troglitazone for 1 hour and during a 6-hour exposure to 200 μM tBH. In the naïve cells, troglitazone treatment resulted in a significant protective effect, with 80% of the cells remaining viable compared with 10% of the cells not pretreated with troglitazone (Fig. 3D). In contrast, compared with the naïve cells treated with tBH, no significant increase in viability was observed with troglitazone in the cells treated with tBH but lacking PPARγ expression (Fig. 3D). These data provide evidence that the cytoprotective effect of troglitazone is mediated by PPARγ. 
Figure 3.
 
Cytoprotective effect of troglitazone is mediated by PPARγ. mRNA levels of PPARγ were measured 48 hours, 72 hours, 4 days, and 8 days after transfection of ARPE-19 cells with an siRNA to PPARγ or a control siRNA (A). Analysis of protein knockdown in ARPE-19 cells 72 hours after transfection with a PPARγ siRNA. The graph was generated from densitometric scans of the gel bands of a Western blot (B). Effect of PPARγ knockdown on ARPE-19 cell viability after exposure to increasing doses of tBH (C). Cell viability was assessed after exposure of ARPE-19 cells to tBH, with or without troglitazone pretreatment, and demonstrates that knockdown of PPARγ blocks the cytoprotective effect of troglitazone (D). **P < 0.01 compared with naïve cells treated with tBH alone.
Figure 3.
 
Cytoprotective effect of troglitazone is mediated by PPARγ. mRNA levels of PPARγ were measured 48 hours, 72 hours, 4 days, and 8 days after transfection of ARPE-19 cells with an siRNA to PPARγ or a control siRNA (A). Analysis of protein knockdown in ARPE-19 cells 72 hours after transfection with a PPARγ siRNA. The graph was generated from densitometric scans of the gel bands of a Western blot (B). Effect of PPARγ knockdown on ARPE-19 cell viability after exposure to increasing doses of tBH (C). Cell viability was assessed after exposure of ARPE-19 cells to tBH, with or without troglitazone pretreatment, and demonstrates that knockdown of PPARγ blocks the cytoprotective effect of troglitazone (D). **P < 0.01 compared with naïve cells treated with tBH alone.
Differential Effects of PPARγ Agonists on tBH-Induced Changes in Gene Expression
The fact that troglitazone, but not other thiazolidinedione agonists, is cytoprotective suggests that it acts as a selective modulator of PPARγ. Selective modulation by agonists is a well-described phenomenon for some nuclear receptors, including PPARγ. 37,38 One feature of selective modulation of nuclear receptors is that the unique biological consequences of receptor activation are the result of the regulation of distinct as well as overlapping sets of genes. To determine the effects of the various thiazolidinedione agonists on gene expression changes induced by tBH, RNA from untreated ARPE-19 cells or cells treated with tBH, with or without troglitazone, rosiglitazone, or pioglitazone was isolated and used for microarray analysis. We chose to compare treatment groups after a 6-hour exposure to 200 μM tBH, as at this dose and treatment time these cells are committed to undergo apoptosis and can be protected by troglitazone (Fig. 2). Two independent experiments were designed, one comparing the effects of tBH plus troglitazone with those of tBH plus rosiglitazone and the other comparing the effects of tBH plus troglitazone with those of tBH plus pioglitazone. All the treatment groups were compared pair-wise to naïve controls. As anticipated and consistent with a previous report using similar conditions, 39 tBH treatment of ARPE-19 cells alone led to a large number of changes in gene expression. Using a 1.8-fold change in expression and P ≤ 0.001 as our significance level, we found 3205 genes to be upregulated and 2830 to be downregulated by tBH treatment in the first experiment comparing troglitazone to rosiglitazone. A similar number of genes (2377 up- and 3750 downregulated) were found to be deregulated by tBH treatment alone in the second experiment comparing troglitazone to pioglitazone. Of those genes significantly deregulated, 780 upregulated genes and 553 downregulated genes were common across both experiments. Global functional analyses and canonical pathway analyses were generated through the use of pathway analyses (Ingenuity Pathway Analysis [IPA], Ingenuity Systems, Redwood, CA). Figure 4A shows the top 10 major biological functions, according to significance, that were deregulated by tBH treatment. These included functions such as cell death (195 genes upregulated), cell growth and proliferation (200 genes upregulated), cell cycle (81 genes upregulated, 59 downregulated), cancer (192 genes upregulated, 62 downregulated), cellular development (174 genes upregulated, 28 downregulated), and cell morphology (61 genes upregulated, 32 downregulated), as well as others (Fig. 4A). Pathway analysis also identified the major canonical pathways affected by t-BH treatment (Fig. 4B). Figure 4B shows the pathway analysis of the top 10 most deregulated canonical signaling pathways. Interestingly, there was very little overlap in the signaling pathways containing genes that were upregulated versus those that were downregulated. 
Figure 4.
 
Top 10 tBH-induced biological functions and canonical pathways. Graphs of the top 10 biological functions (A) and canonical pathways (B) identified by pathway analysis of genes upregulated (top) and downregulated (bottom) in response to tBH treatment and common to two independent experiments. The y-axis is the −log (P value) and the horizontal line is the threshold of P, which is equivalent to 0.05.
Figure 4.
 
Top 10 tBH-induced biological functions and canonical pathways. Graphs of the top 10 biological functions (A) and canonical pathways (B) identified by pathway analysis of genes upregulated (top) and downregulated (bottom) in response to tBH treatment and common to two independent experiments. The y-axis is the −log (P value) and the horizontal line is the threshold of P, which is equivalent to 0.05.
In contrast to tBH treatment alone, tBH+troglitazone treatment resulted in far fewer deregulated genes. In the first experiment, only 764 genes were upregulated and 1499 genes were downregulated. By comparison, tBH+rosiglitazone treatment resulted in the upregulation of 2635 genes and the downregulation of 2586—results more consistent with tBH treatment alone. Moreover, 92% of the upregulated genes and 85% of the downregulated ones were common between the tBH and the tBH+rosiglitazone treatments. A similar phenomenon was observed in the second experiment, where tBH+troglitazone caused the upregulation of 1238 genes and the downregulation of 1517. Again, in contrast to tBH+troglitazone, tBH+pioglitazone resulted in significantly more changes, with 2072 genes upregulated and 2762 downregulated, albeit with somewhat less overlap with the tBH-alone treatment (55% upregulated and 65% downregulated genes). Thus, while the addition of rosiglitazone or pioglitazone has little effect on the tBH gene profile, troglitazone treatment appears to significantly ameliorate the gene expression changes induced by tBH treatment. 
To identify those genes selectively regulated by troglitazone in the presence of tBH, we flagged the genes that were regulated by tBH treatment and whose expression was robustly changed by the addition of troglitazone. Among the genes significantly deregulated by tBH treatment, the expression of several of them was found to be strongly modulated by troglitazone treatment. From the 3205 genes found to be upregulated and 2830 genes downregulated by tBH (using a 1.8-fold change in expression and P ≤ 0.001) in the first experiment, 1487 of them were significantly modulated to some extent by troglitazone (P ≤ 0.001 and correction >20%). Of those, 749 genes were found to be corrected or normalized ≥50%. Similarly, in the second experiment from the 6127 genes (2377 upregulated and 3750 downregulated) found to be deregulated by tBH, 1978 were also found to be significantly changed on the addition of troglitazone. Of those, 430 genes were found to be corrected or normalized ≥50%. Genes uniquely regulated by troglitazone were identified from among these genes by filtering out genes whose expression was changed less than 20% by either rosiglitazone or pioglitazone or genes in which expression changes caused by tBH treatment were exacerbated by these compounds. This yielded 490 genes that were uniquely corrected by troglitazone but not rosiglitazone in the first experiment and 128 genes in the second experiment comparing troglitazone to pioglitazone. In contrast, using the same methodology, no genes were found to be selectively corrected by rosiglitazone treatment, and only two genes were selectively corrected by pioglitazone treatment. This result is consistent with the fact that the overall total number of genes deregulated by tBH was reduced by troglitazone but not by rosiglitazone or pioglitazone treatment and suggests that troglitazone tends to normalize expression of many of the genes deregulated by tBH treatment. A total of 56 genes that were normalized by ≥50% were common in both experiments (listed in Table 2). Of the 56 genes normalized ≥50% by troglitazone in both experiments, the expression of 29 changed less than 20%, and 34 were changed less than 30% by either rosiglitazone or pioglitazone treatment. 
Table 2.
 
List of Genes Normalized ≥50% by Troglitazone Treatment and Common to Both Experiments
Table 2.
 
List of Genes Normalized ≥50% by Troglitazone Treatment and Common to Both Experiments
  List of Genes Normalized ≥50% by Troglitazone Treatment and Common to Both Experiments
Validation of Microarray Data by RT-PCR
To validate the microarray data, RT-PCR was performed on selected genes. Deregulation by tBH of selected genes and their modulation by troglitazone treatment was confirmed by PCR. Three genes upregulated and three genes downregulated by tBH treatment were selected for RT-PCR analysis. The PCR products for each of these genes are shown and the relative mRNA expression levels of the genes are plotted in Figure 5. Not only did the RT-PCR analysis confirm the direction of expression changes for each of the genes with the different treatments, but in many cases the expression changes were also consistent with the microarray data (Table 2). 
Figure 5.
 
RT-PCR confirmation of microarray results. Products of the RT-PCR reactions for selected genes deregulated by tBH treatment were run out on agarose gels and visualized with ethidium bromide staining. A graph of the relative mRNA expression levels of these genes after normalization with tubulin is shown.
Figure 5.
 
RT-PCR confirmation of microarray results. Products of the RT-PCR reactions for selected genes deregulated by tBH treatment were run out on agarose gels and visualized with ethidium bromide staining. A graph of the relative mRNA expression levels of these genes after normalization with tubulin is shown.
Pathway Analysis of Genes Selectively Modulated by Troglitazone
To explore the functional relationship between PPARγ and those genes regulated by troglitazone, we conducted a pathway analysis (IPA software; Ingenuity Systems). The relationships between PPARγ and those 56 genes normalized by ≥50% with troglitazone treatment in both experiments are illustrated in Figure 6. This analysis generated a network containing 10 of the possible 56 genes. These genes included activating transcription factor 3 (ATF3), adrenomedullin (ADM), chemokine (C-C motif) ligand 2 (CCL2), chemokine (C-X-C motif) receptor 4 (CXCR4), connective tissue growth factor (CTGF), cyclin-dependent kinase inhibitor 1C (CDKN1C), integrin alpha V (ITGAV), Kruppel-like factor 2 (KLF2), N-myc downstream regulated gene 1 (NDRG1), and vascular endothelial growth factor (VEGF). Six of these 10 genes were modulated less than 20% by rosiglitazone or pioglitazone. Direct relationships were confirmed between PPARγ and CDKN1C, VEGF, KLF2, and ADM
Figure 6.
 
Pathway analysis of genes deregulated by tBH and selectively modulated by troglitazone. Interactions between PPARγ and genes deregulated by tBH and selectively modulated by troglitazone are illustrated. The nature of the interaction is described with the following abbreviations: A, activation; E, expression; PD, protein-DNA interactions; PP, protein–protein interactions; RB, regulation of binding; TR, translocation.
Figure 6.
 
Pathway analysis of genes deregulated by tBH and selectively modulated by troglitazone. Interactions between PPARγ and genes deregulated by tBH and selectively modulated by troglitazone are illustrated. The nature of the interaction is described with the following abbreviations: A, activation; E, expression; PD, protein-DNA interactions; PP, protein–protein interactions; RB, regulation of binding; TR, translocation.
Differential Effects of PPARγ Agonists on tBH-Induced PPARγ Binding to PPREs
To confirm that the tBH-induced changes in gene expression selectively corrected with troglitazone treatment are in fact mediated by PPARγ, we measured PPARγ binding to PPAR response elements (PPREs) found in the promoter regions of these genes. Five genes were chosen from among those 29 genes in which expression was significantly deregulated at the transcriptional level upon tBH treatment but were normalized more than 50% by troglitazone but less than 20% by either rosiglitazone or pioglitazone treatment in both microarray experiments (Table 2). Two of the selected genes were upregulated (CDKN1C, CXCR4) and three downregulated (AMOTL2, JUB, and CCL2) on tBH treatment. The locations of potential PPREs in the different promoters of these genes were identified by using the MatInspector algorithm 32 and consensus PPRE sequences 33 to search gene sequences. PCR primers (Table 1) were then designed to probe those PPREs in chromatin immunoprecipitation (ChIP) experiments. Chromatin isolated from ARPE-19 cells treated with tBH with or without troglitazone, rosiglitazone, or pioglitazone for 3 hours was immunoprecipitated with antibodies to PPARγ and the presence of promoters of different genes assessed by qPCR using primers flanking predicted PPREs. The presence of predicted PPREs in anti-PPARγ immunoprecipitates, indicating PPARγ binding, was detected for four of the five genes probed, including CDKN1C, CXCR4, AMOTL2, and CCL2 (Fig. 7). It is possible for JUB, in which no binding was detected (data not shown), that PPARγ binds a site other than that predicted by the algorithm used to select potential PPREs. While AMOTL2 showed similar binding in the presence of all the glitazones, the amount of PPARγ binding seen for tBH was reduced in the presence of troglitazone but not of rosiglitazone or pioglitazone for CDKN1C, CXCR4, and CCL2 (Fig. 7). Thus, tBH can activate PPARγ, leading to direct DNA binding. Moreover, troglitazone is able to selectively modulate PPARγ binding to the promoters of select genes and thereby alter their expression levels. 
Figure 7.
 
ChIP analysis of PPARγ binding to putative PPREs in genes selectively modulated by troglitazone. Chromatin from naïve, tBH-treated cells, tBH+troglitazone, tBH+rosiglitazone, or tBH+pioiglitazone was immunoprecipitated with anti-PPARγ antibodies. The binding of PPARγ to associated DNA was then determined by qPCR, using primer pairs shown in Table 1 flanking putative PPREs for CDKN1C, AMOTL2, and CCL2 (A) and CXCR4 (B).
Figure 7.
 
ChIP analysis of PPARγ binding to putative PPREs in genes selectively modulated by troglitazone. Chromatin from naïve, tBH-treated cells, tBH+troglitazone, tBH+rosiglitazone, or tBH+pioiglitazone was immunoprecipitated with anti-PPARγ antibodies. The binding of PPARγ to associated DNA was then determined by qPCR, using primer pairs shown in Table 1 flanking putative PPREs for CDKN1C, AMOTL2, and CCL2 (A) and CXCR4 (B).
Discussion
Oxidative stress and chronic inflammation is believed to play a role in the pathogenesis of ocular diseases such as ARMD. 18,19 Progression of the atrophic form of ARMD leads to the loss of cells of the RPE and, ultimately, loss of photoreceptors. Thus, the preservation of RPE from oxidative and inflammatory insults represents a potential therapeutic strategy. To better understand the response of RPE cells to oxidative stress in vivo we examined the effects of tBH treatment on RPE cells in vitro. In addition, we wanted to understand the role of PPARγ in modulating the response of RPE cells to oxidative stress. 
The best-known ligands of PPARγ, the thiazolidinediones, were tested for their effects on RPE cell survival in response to challenge with tBH. Rosiglitazone and pioglitazone both potentiated cell death, whereas troglitazone protected RPE cells from apoptosis induced by tBH (Fig. 2). The role of PPARγ in mediating this effect was confirmed using RNA interference (Fig. 3). While there are reports that the PPARγ agonist, 15-deoxy-Δ12,14-prostaglandin J2 (dPGJ2) can protect RPE cells from oxidative stress, this protection was shown to be PPARγ independent. 40,41 Thus, this is the first report to demonstrate that selective modulation of PPARγ can alter RPE responses to oxidative stress. Selective receptor modulation is a common feature of many nuclear receptors. The most notable example is that of the estrogen receptor for which several selective ER modulators (SERMs) such as tamoxifen and raloxifene have been described. 42 However, there is also precedence for selective PPARγ modulators (SPPARMs). 37,38 In particular, several thiazolidinedione-like and non–thiazolidinedione-like partial agonists of PPARγ have been described that have insulin-sensitizing activity but lower levels of the adipogenesis-stimulating activity associated with full agonists. 43 Even among the known thiazolidinediones, differential effects have been observed. For example, troglitazone has been shown to act as a partial agonist for PPARγ in C2C12 and HEK 293T cells compared with rosiglitazone, but as a full agonist in 3T3-L1 adipocytes. 44  
SPPARMS are believed to effect differential gene changes through the recruitment of a unique complement of co-activators(s) to PPARγ bound to DNA. ChIP experiments revealed that, whereas the tBH-treatment leads to enhanced binding of PPARγ to PPREs, the presence of troglitazone causes a diminution of PPARγ binding to DNA (Fig. 6). Thus, troglitazone appears to act in a unique fashion by preventing binding of PPARγ itself to PPREs, leading to differential gene expression. In contrast, despite the potentiation of cell death by rosiglitazone and pioglitazone, we did not observe enhanced PPARγ binding to PPREs of genes selectively regulated by troglitazone in the presence of these agonists. Consistent with our data, troglitazone has been shown to antagonize rosiglitazone-stimulated PPARγ transcriptional activity. 44 This differential binding in the presence of the thiazolidinediones may account for the different cytoprotective activities in response to oxidative stress. Interestingly, specific biological differences between troglitazone and rosiglitazone in vivo have been observed. For example differential vasoactive properties between these two agonists have been described. 45 Similarly, troglitazone was shown to inhibit thrombin-induced platelet aggregation, whereas pioglitazone was ineffective. 46  
One prediction is that a bona fide SPPARM would induce a unique gene expression pattern that mediates its differential activity. Microarray analysis confirmed that this is indeed the case with troglitazone, as we observed both quantitative and qualitative changes in gene expression when cells were treated with tBH in the presence of different thiazolidinediones. Both rosiglitazone and pioglitazone appear to have minimal effects on tBH-induced changes, as a similar number of genes were deregulated by tBH treatment alone or with tBH plus rosiglitazone or pioglitazone. Moreover, a high percentage of the genes with expression that is altered by tBH are similarly altered in expression in the presence of rosiglitazone or pioglitazone. In contrast, the addition of troglitazone dramatically reduced the number of genes that were deregulated by tBH and selectively normalized the expression of a number of genes. Not surprisingly, most changes in gene expression induced by tBH involve cellular functions related to stress responses including cell death, cell cycle, cancer, and cell growth and proliferation (Fig. 4A). Most notably, the NRF2-mediated oxidative stress pathway is upregulated in response to tBH. Nuclear factor E2-related factor 2 (Nrf2) is a transcription factor that controls the expression of antioxidant genes and cytoprotective phase II detoxifying enzymes. 47,48 It has even been proposed that Nrf2 can activate a cytoprotective response in ARMD. 49 Thus, the upregulation of NRF2 signaling maybe an attempt by the cell to suppress oxidative stress. 
Interesting to us was that the genes in the hypoxia and HIF-1α signaling pathways were upregulated in response to oxidative stress. It is known that HIF-1α protein undergoes rapid prolyl hydroxylation and degradation under normoxic conditions and that levels of ROS, especially H2O2, may directly and indirectly regulate HIF-1α hydroxylation and thereby upregulate its expression. 50 52 One of the hypoxic genes we found upregulated by tBH was vascular endothelial growth factor (VEGF), the major driver of the advanced neovascular form of ARMD. Moreover, VEGF expression was corrected more than 50% by troglitazone treatment and less than 30% by either rosiglitazone or pioglitazone (Table 2). The fact that oxidative stress induces VEGF expression in RPE cells in culture is consistent with the proposed role of oxidative stress in the initiation and progression of ARMD. 
When we used a 50% change or greater relative to tBH treatment alone we found that 56 genes were identified that were strongly and selectively affected by troglitazone in both experiments (Table 2). About half these genes fell into distinct functional categories, including apoptosis (CDKN1C, HSPA1L, NDRG1, CRYAB, DUSP10, and CRYAB), chemokines (CXCR4 and CCL2), hypoxia (HIG2, NDRG1, and VEGF), angiogenesis (VEGF, AMOTL2, and CTGF), and transcription (ATF3, KLF2, and ID2). The fact that all these processes have been implicated in some aspect of the initiation and progression of ARMD further supports a role for oxidative stress on RPE in this disease. CDKN1C, the gene encoding the cell cycle inhibitor, p57Kip2 was found to be potently upregulated in response to tBH (Table 2). Pathway analysis (Fig. 5) revealed that PPARγ had been shown to directly regulate the expression of CDKN1C. 53 Significantly, CDKN1C expression was corrected 65% by troglitazone but not at all by either rosiglitazone or pioglitazone (Table 2). In addition, PPARγ binding to a PPRE in the CDKN1C gene was selectively inhibited by troglitazone and not by rosiglitazone or pioglitazone (Fig. 7). p57Kip2 is a cyclin-dependent kinase inhibitor (CKI) that inhibits the activity of cyclin-CDK2 or -CDK4 complexes, and thus functions as a regulator of cell cycle progression at G1. 54 Our canonical pathway analysis revealed that more than 76 genes involved in the cell cycle were upregulated and 48 were downregulated (Fig. 4A). p57Kip2 can modulate apoptosis in different ways depending on cellular context. While CKIs can protect against apoptosis via their CDK inhibitory activity, they can also promote or inhibit apoptosis independent of CDK. 55,56 It is unclear what the role is for p57Kip2 in our model. However, its selective modulation by troglitazone suggests that it may be proapoptotic rather than part of a protective response. 
Another class of genes strongly deregulated by tBH and selectively modulated by troglitazone are the chemokines, including CXCR4 and CCL2. Troglitazone not only modulated the expression of these genes (Table 2), but ChIP experiments confirmed that troglitazone also directly affected PPARγ binding to PPREs in these genes (Fig. 7). The expression of CXCR4, a chemokine receptor involved in leukocyte and stem cell trafficking, had been shown to be strongly upregulated in APRE-19 cells in response to oxLDL treatment. 24 In vivo, CXCR4 was found to be upregulated in the RPE/choroid of aged mice. 57 Moreover, CXCR4 levels were found to be increased in ischemic retina and human CNV lesions. 58 60 Importantly, antibodies to SDF-1 the ligand for CXCR4 or inhibitors of CXCR4 block incorporation of stem cells in choroidal neovascular lesions and inhibit neovascularization. 58 Together, these results demonstrate that expression of CXCR4 is upregulated in the eye in response to different stresses and is implicated in pathologic neovascularization. Consistent with this notion, troglitazone has been shown to inhibit CNV in animal models. 29 In contrast to CXCR4, the chemokine CCL2 was found to be strongly downregulated by tBH treatment. Significantly, CCL2-deficient mice have an AMD-like phenotype including drusen-like deposits below the RPE. 61,62  
tBH treatment leads to increased binding of PPARγ to PPREs (Fig. 7). This finding suggests that tBH induces the production of a ligand for PPARγ. The most likely candidate(s) are oxidized lipids of which there are several potential candidates. Several endogenous oxidized lipids, including 9S-hydroxy-10E,12S-octadecadienoic acid (9-HODE), 13S-hydroxy-9Z,11E-octadecadienoic acid (13-HODE), and 15S-hydroxy-5Z,8Z,11Z,13E-eicosatetraenoic acid (15-HETE) have been described as ligands of PPARγ. 4 While 9-HODE and 13-HODE are oxidized metabolites of linoleic acid, 15-HETE is a metabolite of arachidonic acid. More recently, oxidized docosahexaenoic acid derivatives were also identified as PPARγ agonists. 3 Treatment of monocytes with oxLDL or specific oxidized lipid components of oxLDL such as 9-HODE and 13-HODE reduces expression of CCR2, the receptor for CCL2. 63 Moreover, this regulation of CCR2 expression may be mediated by PPARγ. 63 CCL2 was found to be upregulated in the RPE/choroid of aged mice. 57 It has also been reported that there is an increase in macrophages in the RPE/choroid of aged mice, 62 suggesting that during normal ageing, increased expression of CCL2 by the RPE may lead to the recruitment of macrophages to help handle age-related increases in lipid accumulation. The PPARγ-dependent reduction of CCL2 expression that we observed in RPE cells in response to tBH and the fact that the CCR2 knockout develops ARMD-like features similar to those of the CCL2 knockout is consistent with the hypothesis that impaired macrophage recruitment may play a role in ARMD. 62  
Together, these results demonstrate that oxidative stress can induce gene changes in RPE cells in vitro, consistent with many of the described phenotypic changes associated with ARMD. This result underscores the importance of oxidative stress in this disease and validates the use of in vitro models to help understand its effects. Furthermore, we have shown that selective modulation of PPARγ can alter RPE responses to oxidative stress through suppression of a subset of these changes. A greater understanding of the nature of gene changes induced by cytoprotective agents such as troglitazone has the potential to lead to the identification of novel therapeutic targets. 
Footnotes
 Supported by Allergan, Inc., and ExonHit Therapeutics SA.
Footnotes
 Disclosure: G.A. Rodrigues, Allergan, Inc. (F, I, E), P; F. Maurier-Mahé, ExonHit Therapeutics SA (F, I, E); D.-L. Shurland, Allergan, Inc. (F, I, E); A. Mclaughlin, Allergan, Inc. (F, I, E); K. Luhrs, Allergan, Inc. (F, I, E); E. Throo, ExonHit Therapeutics SA (F, I, E); L. Delalonde-Delaunay, ExonHit Therapeutics SA (F, I, E); D. Pallares, ExonHit Therapeutics SA (F, I, E); F. Schweighoffer, ExonHit Therapeutics SA (F, I, E), P; J. Donello, Allergan, Inc. (F, I, E), P
References
Willson TM Brown PJ Sternbach DD Henke BR . The PPARs: from orphan receptors to drug discovery. J Med Chem. 2000;43:527–550. [CrossRef] [PubMed]
Rosen ED Spiegelman BM . PPARγ : a nuclear regulator of metabolism, differentiation, and cell growth. J Biol Chem. 2001;276:37731–37734. [CrossRef] [PubMed]
Yamamoto K Itoh T Abe D . Identification of putative metabolites of docosahexaenoic acid as potent PPARγ agonists and antidiabetic agents. Bioorg Med Chem Lett. 2005;15:517–522. [CrossRef] [PubMed]
Nagy L Tontonoz P Alvarez JG Chen H Evans RM . Oxidized LDL regulates macrophage gene expression through ligand activation of PPARγ. Cell. 1998;93:229–240. [CrossRef] [PubMed]
Walczak R Tontonoz P . PPARadigms and PPARadoxes: expanding roles for PPARγ in the control of lipid metabolism. J Lipid Res. 2002;43:177–186. [PubMed]
Lee CH Evans RM . Peroxisome proliferator-activated receptor-γ in macrophage lipid homeostasis. Trends Endocrinol Metab. 2002;13:331–335. [CrossRef] [PubMed]
Tontonoz P Nagy L Alvarez JG Thomazy VA Evans RM . PPARγ promotes monocyte/macrophage differentiation and uptake of oxidized LDL. Cell. 1998;93:241–252. [CrossRef] [PubMed]
Chawla A Barak Y Nagy L Liao D Tontonoz P Evans RM . PPAR-γ dependent and independent effects on macrophage-gene expression in lipid metabolism and inflammation. Nat Med. 2001;7:48–52. [CrossRef] [PubMed]
Chawla A Boisvert WA Lee CH . A PPAR γ-LXR-ABCA1 pathway in macrophages is involved in cholesterol efflux and atherogenesis. Mol Cell. 2001;7:161–171. [CrossRef] [PubMed]
Feng J Han J Pearce SF . Induction of CD36 expression by oxidized LDL and IL-4 by a common signaling pathway dependent on protein kinase C and PPAR-γ. J Lipid Res. 2000;41:688–696. [PubMed]
Moore KJ Rosen ED Fitzgerald ML . The role of PPAR-γ in macrophage differentiation and cholesterol uptake. Nat Med. 2001;7:41–47. [CrossRef] [PubMed]
Chinetti G Lestavel S Bocher V . PPAR-alpha and PPAR-γ activators induce cholesterol removal from human macrophage foam cells through stimulation of the ABCA1 pathway. Nat Med. 2001;7:53–58. [CrossRef] [PubMed]
Strauss O . The retinal pigment epithelium in visual function. Physiol Rev. 2005;85:845–881. [CrossRef] [PubMed]
McBee JK Palczewski K Baehr W Pepperberg DR . Confronting complexity: the interlink of phototransduction and retinoid metabolism in the vertebrate retina. Prog Retin Eye Res. 2001;20:469–529. [CrossRef] [PubMed]
Thompson DA Gal A . Vitamin A metabolism in the retinal pigment epithelium: genes, mutations, and diseases. Prog Retin Eye Res. 2003;22:683–703. [CrossRef] [PubMed]
Nguyen-Legros J Hicks D . Renewal of photoreceptor outer segments and their phagocytosis by the retinal pigment epithelium. Int Rev Cytol. 2000;196:245–313. [PubMed]
Ryeom SW Sparrow JR Silverstein RL . CD36 participates in the phagocytosis of rod outer segments by retinal pigment epithelium. J Cell Sci. 1996;109):387–395.
Donoso LA Kim D Frost A Callahan A Hageman G . The role of inflammation in the pathogenesis of age-related macular degeneration. Surv Ophthalmol. 2006;51:137–152. [CrossRef] [PubMed]
Zarbin MA . Current concepts in the pathogenesis of age-related macular degeneration. Arch Ophthalmol. 2004;122:598–614. [CrossRef] [PubMed]
Cai J Nelson KC Wu M Sternberg PJr Jones DP . Oxidative damage and protection of the RPE. Prog Retin Eye Res. 2000;19:205–221. [CrossRef] [PubMed]
Beatty S Koh H Phil M Henson D Boulton M . The role of oxidative stress in the pathogenesis of age-related macular degeneration. Surv Ophthalmol. 2000;45:115–134. [CrossRef] [PubMed]
Curcio CA Millican CL Bailey T Kruth HS . Accumulation of cholesterol with age in human Bruch's membrane. Invest Ophthalmol Vis Sci. 2001;42:265–274. [PubMed]
Joffre C Leclere L Buteau B . Oxysterols induced inflammation and oxidation in primary porcine retinal pigment epithelial cells. Curr Eye Res. 2007;32:271–280. [CrossRef] [PubMed]
Yamada Y Tian J Yang Y . Oxidized low density lipoproteins induce a pathologic response by retinal pigmented epithelial cells. J Neurochem. 2008;105:1187–1197. [CrossRef] [PubMed]
Gordiyenko N Campos M Lee JW Fariss RN Sztein J Rodriguez IR . RPE cells internalize low-density lipoprotein (LDL) and oxidized LDL (oxLDL) in large quantities in vitro and in vivo. Invest Ophthalmol Vis Sci. 2004;45:2822–2829. [CrossRef] [PubMed]
Hoppe G O'Neil J Hoff HF Sears J . Accumulation of oxidized lipid-protein complexes alters phagosome maturation in retinal pigment epithelium. Cell Mol Life Sci. 2004;61:1664–1674. [CrossRef] [PubMed]
Hoppe G Marmorstein AD Pennock EA Hoff HF . Oxidized low density lipoprotein-induced inhibition of processing of photoreceptor outer segments by RPE. Invest Ophthalmol Vis Sci. 2001;42:2714–2720. [PubMed]
Rodriguez IR Alam S Lee JW . Cytotoxicity of oxidized low-density lipoprotein in cultured RPE cells is dependent on the formation of 7-ketocholesterol. Invest Ophthalmol Vis Sci. 2004;45:2830–2837. [CrossRef] [PubMed]
Murata T He S Hangai M . Peroxisome proliferator-activated receptor-γ ligands inhibit choroidal neovascularization. Invest Ophthalmol Vis Sci. 2000;41:2309–2317. [PubMed]
Ershov AV Bazan NG . Photoreceptor phagocytosis selectively activates PPARγ expression in retinal pigment epithelial cells. J Neurosci Res. 2000;60:328–337. [CrossRef] [PubMed]
Rosenkranz AR Schmaldienst S Stuhlmeier KM Chen W Knapp W Zlabinger GJ . A microplate assay for the detection of oxidative products using 2′,7′-dichlorofluorescein-diacetate. J Immunol Methods. 1992;156:39–45. [CrossRef] [PubMed]
Cartharius K Frech K Grote K . MatInspector and beyond: promoter analysis based on transcription factor binding sites. Bioinformatics. 2005;21:2933–2942. [CrossRef] [PubMed]
Tachibana K Kobayashi Y Tanaka T . Gene expression profiling of potential peroxisome proliferator-activated receptor (PPAR) target genes in human hepatoblastoma cell lines inducibly expressing different PPAR isoforms. Nucl Recept. 2005;3:3. [CrossRef] [PubMed]
Sternberg PJr Davidson PC Jones DP Hagen TM Reed RL Drews-Botsch C . Protection of retinal pigment epithelium from oxidative injury by glutathione and precursors. Invest Ophthalmol Vis Sci. 1993;34:3661–3668. [PubMed]
Nelson KC Armstrong JS Moriarty S . Protection of retinal pigment epithelial cells from oxidative damage by oltipraz, a cancer chemopreventive agent. Invest Ophthalmol Vis Sci. 2002;43:3550–3554. [PubMed]
Cai J Wu M Nelson KC Sternberg PJr Jones DP . Oxidant-induced apoptosis in cultured human retinal pigment epithelial cells. Invest Ophthalmol Vis Sci. 1999;40:959–966. [PubMed]
Rangwala SM Lazar MA . The dawn of the SPPARMs? Sci STKE. 2002;2002:PE9. [PubMed]
Smith CL O'Malley BW . Coregulator function: a key to understanding tissue specificity of selective receptor modulators. Endocr Rev. 2004;25:45–71. [CrossRef] [PubMed]
Weigel AL Handa JT Hjelmeland LM . Microarray analysis of H2O2-, HNE-, or tBH-treated ARPE-19 cells. Free Radic Biol Med. 2002;33:1419–1432. [CrossRef] [PubMed]
Qin S McLaughlin AP De Vries GW . Protection of RPE cells from oxidative injury by 15-deoxy-delta12,14-prostaglandin J2 by augmenting GSH and activating MAPK. Invest Ophthalmol Vis Sci. 2006;47:5098–5105. [CrossRef] [PubMed]
Garg TK Chang JY . Oxidative stress causes ERK phosphorylation and cell death in cultured retinal pigment epithelium: prevention of cell death by AG126 and 15-deoxy-delta 12, 14-PGJ2. BMC Ophthalmol. 2003;3:5. [CrossRef] [PubMed]
Dutertre M Smith CL . Molecular mechanisms of selective estrogen receptor modulator (SERM) action. J Pharmacol Exp Ther. 2000;295:431–437. [PubMed]
Zhang F Lavan BE Gregoire FM . Selective modulators of PPAR-γ activity: molecular aspects related to obesity and side-effects. PPAR Res. 2007;2007:32696. [CrossRef] [PubMed]
Camp HS Li O Wise SC . Differential activation of peroxisome proliferator-activated receptor-γ by troglitazone and rosiglitazone. Diabetes. 2000;49:539–547. [CrossRef] [PubMed]
Walker AB Naderali EK Chattington PD Buckingham RE Williams G . Differential vasoactive effects of the insulin sensitizers rosiglitazone (BRL 49653) and troglitazone on human small arteries in vitro. Diabetes. 1998;47:810–814. [CrossRef] [PubMed]
Ishizuka T Itaya S Wada H . Differential effect of the antidiabetic thiazolidinediones troglitazone and pioglitazone on human platelet aggregation mechanism. Diabetes. 1998;47:1494–1500. [CrossRef] [PubMed]
Kaspar JW Niture SK Jaiswal AK . Nrf2:INrf2 (Keap1) signaling in oxidative stress (review). Free Radic Biol Med. 2009;47:1304–1309. [CrossRef] [PubMed]
Niture SK Kaspar JW Shen J Jaiswal AK . Nrf2 signaling and cell survival (review). Toxicol Appl Pharmacol. 2010;244;37–42.
Cano M Thimmalappula R Fujihara M . Cigarette smoking, oxidative stress, the antioxidant response through Nrf2 signaling, and age-related macular degeneration. Vision Res. 2010;50:652–664. [CrossRef] [PubMed]
Qutub AA Popel AS . Reactive oxygen species regulate hypoxia-inducible factor 1alpha differentially in cancer and ischemia. Mol Cell Biol. 2008;28:5106–5119. [CrossRef] [PubMed]
Brunelle JK Bell EL Quesada NM . Oxygen sensing requires mitochondrial ROS but not oxidative phosphorylation. Cell Metab. 2005;1:409–414. [CrossRef] [PubMed]
Gerald D Berra E Frapart YM . JunD reduces tumor angiogenesis by protecting cells from oxidative stress. Cell. 2004;118:781–794. [CrossRef] [PubMed]
Yu S Matsusue K Kashireddy P . Adipocyte-specific gene expression and adipogenic steatosis in the mouse liver due to peroxisome proliferator-activated receptor γ1 (PPARγ1) overexpression. J Biol Chem. 2003;278:498–505. [CrossRef] [PubMed]
Besson A Dowdy SF Roberts JM . CDK inhibitors: cell cycle regulators and beyond. Dev Cell. 2008;14:159–169. [CrossRef] [PubMed]
Vlachos P Nyman U Hajji N Joseph B . The cell cycle inhibitor p57(Kip2) promotes cell death via the mitochondrial apoptotic pathway. Cell Death Differ. 2007;14:1497–1507. [CrossRef] [PubMed]
Watanabe H Pan ZQ Schreiber-Agus N DePinho RA Hurwitz J Xiong Y . Suppression of cell transformation by the cyclin-dependent kinase inhibitor p57KIP2 requires binding to proliferating cell nuclear antigen. Proc Natl Acad Sci U S A. 1998;95:1392–1397. [CrossRef] [PubMed]
Chen H Liu B Lukas TJ Neufeld AH . The aged retinal pigment epithelium/choroid: a potential substratum for the pathogenesis of age-related macular degeneration. PLoS One. 2008;3:e2339. [CrossRef] [PubMed]
Lima e Silva R Shen J Hackett SF . The SDF-1/CXCR4 ligand/receptor pair is an important contributor to several types of ocular neovascularization. FASEB J. 2007;21:3219–3230. [CrossRef] [PubMed]
Guerin E Sheridan C Assheton D . SDF1-alpha is associated with VEGFR-2 in human choroidal neovascularisation. Microvasc Res. 2008;75:302–307. [CrossRef] [PubMed]
Bhutto IA McLeod DS Merges C Hasegawa T Lutty GA . Localisation of SDF-1 and its receptor CXCR4 in retina and choroid of aged human eyes and in eyes with age related macular degeneration. Br J Ophthalmol. 2006;90:906–910. [CrossRef] [PubMed]
Chan CC Ross RJ Shen D . Ccl2/Cx3cr1-deficient mice: an animal model for age-related macular degeneration. Ophthalmic Res. 2008;40:124–128. [CrossRef] [PubMed]
Ambati J Anand A Fernandez S . An animal model of age-related macular degeneration in senescent Ccl-2- or Ccr-2-deficient mice. Nat Med. 2003;9:1390–1397. [CrossRef] [PubMed]
Han KH Chang MK Boullier A . Oxidized LDL reduces monocyte CCR2 expression through pathways involving peroxisome proliferator-activated receptor γ. J Clin Invest. 2000;106:793–802. [CrossRef] [PubMed]
Figure 1.
 
Effects of tBH treatment on ARPE-19 cells. Cells were incubated with DCFDA for 30 minutes and then challenged with increasing concentrations of tBH for 3 hours, after which fluorescence intensity was measured to determine the presence of ROS (A). Cells were similarly challenged with increasing concentrations of tBH for 30 minutes and 1, 3, and 24 hours (B) or 3, 4, 6, 8, and 24 hours (C), and viability was measured 24 hours after the challenge. The percentage of viability was determined by comparison with untreated controls. Increases in caspase 3 activity were measured after 30-minute, 3-hour, and 6-hour tBH challenges (D).
Figure 1.
 
Effects of tBH treatment on ARPE-19 cells. Cells were incubated with DCFDA for 30 minutes and then challenged with increasing concentrations of tBH for 3 hours, after which fluorescence intensity was measured to determine the presence of ROS (A). Cells were similarly challenged with increasing concentrations of tBH for 30 minutes and 1, 3, and 24 hours (B) or 3, 4, 6, 8, and 24 hours (C), and viability was measured 24 hours after the challenge. The percentage of viability was determined by comparison with untreated controls. Increases in caspase 3 activity were measured after 30-minute, 3-hour, and 6-hour tBH challenges (D).
Figure 2.
 
Differential cytoprotective effects of glitazones on tBH-treated ARPE-19 cells. Cells were pretreated with 0.1, 1, and 10 μM troglitazone, rosiglitazone, or pioglitazone for 1 hour before challenge with 200 μM tBH for 30 minutes, 3 hours, or 6 hours (A). Preincubation of cells for 1 hour with 1 μM troglitazone protected the cells from increasing doses of tBH. An 80% to 90% survival rate was obtained in the presence of up to 1 mM tBH (B). Preincubation of cells for 24 hours with troglitazone followed by withdrawal and tBH challenge was sufficient to achieve cytoprotection (C). ***P < 0 0.001 compared with tBH treatment alone.
Figure 2.
 
Differential cytoprotective effects of glitazones on tBH-treated ARPE-19 cells. Cells were pretreated with 0.1, 1, and 10 μM troglitazone, rosiglitazone, or pioglitazone for 1 hour before challenge with 200 μM tBH for 30 minutes, 3 hours, or 6 hours (A). Preincubation of cells for 1 hour with 1 μM troglitazone protected the cells from increasing doses of tBH. An 80% to 90% survival rate was obtained in the presence of up to 1 mM tBH (B). Preincubation of cells for 24 hours with troglitazone followed by withdrawal and tBH challenge was sufficient to achieve cytoprotection (C). ***P < 0 0.001 compared with tBH treatment alone.
Figure 3.
 
Cytoprotective effect of troglitazone is mediated by PPARγ. mRNA levels of PPARγ were measured 48 hours, 72 hours, 4 days, and 8 days after transfection of ARPE-19 cells with an siRNA to PPARγ or a control siRNA (A). Analysis of protein knockdown in ARPE-19 cells 72 hours after transfection with a PPARγ siRNA. The graph was generated from densitometric scans of the gel bands of a Western blot (B). Effect of PPARγ knockdown on ARPE-19 cell viability after exposure to increasing doses of tBH (C). Cell viability was assessed after exposure of ARPE-19 cells to tBH, with or without troglitazone pretreatment, and demonstrates that knockdown of PPARγ blocks the cytoprotective effect of troglitazone (D). **P < 0.01 compared with naïve cells treated with tBH alone.
Figure 3.
 
Cytoprotective effect of troglitazone is mediated by PPARγ. mRNA levels of PPARγ were measured 48 hours, 72 hours, 4 days, and 8 days after transfection of ARPE-19 cells with an siRNA to PPARγ or a control siRNA (A). Analysis of protein knockdown in ARPE-19 cells 72 hours after transfection with a PPARγ siRNA. The graph was generated from densitometric scans of the gel bands of a Western blot (B). Effect of PPARγ knockdown on ARPE-19 cell viability after exposure to increasing doses of tBH (C). Cell viability was assessed after exposure of ARPE-19 cells to tBH, with or without troglitazone pretreatment, and demonstrates that knockdown of PPARγ blocks the cytoprotective effect of troglitazone (D). **P < 0.01 compared with naïve cells treated with tBH alone.
Figure 4.
 
Top 10 tBH-induced biological functions and canonical pathways. Graphs of the top 10 biological functions (A) and canonical pathways (B) identified by pathway analysis of genes upregulated (top) and downregulated (bottom) in response to tBH treatment and common to two independent experiments. The y-axis is the −log (P value) and the horizontal line is the threshold of P, which is equivalent to 0.05.
Figure 4.
 
Top 10 tBH-induced biological functions and canonical pathways. Graphs of the top 10 biological functions (A) and canonical pathways (B) identified by pathway analysis of genes upregulated (top) and downregulated (bottom) in response to tBH treatment and common to two independent experiments. The y-axis is the −log (P value) and the horizontal line is the threshold of P, which is equivalent to 0.05.
Figure 5.
 
RT-PCR confirmation of microarray results. Products of the RT-PCR reactions for selected genes deregulated by tBH treatment were run out on agarose gels and visualized with ethidium bromide staining. A graph of the relative mRNA expression levels of these genes after normalization with tubulin is shown.
Figure 5.
 
RT-PCR confirmation of microarray results. Products of the RT-PCR reactions for selected genes deregulated by tBH treatment were run out on agarose gels and visualized with ethidium bromide staining. A graph of the relative mRNA expression levels of these genes after normalization with tubulin is shown.
Figure 6.
 
Pathway analysis of genes deregulated by tBH and selectively modulated by troglitazone. Interactions between PPARγ and genes deregulated by tBH and selectively modulated by troglitazone are illustrated. The nature of the interaction is described with the following abbreviations: A, activation; E, expression; PD, protein-DNA interactions; PP, protein–protein interactions; RB, regulation of binding; TR, translocation.
Figure 6.
 
Pathway analysis of genes deregulated by tBH and selectively modulated by troglitazone. Interactions between PPARγ and genes deregulated by tBH and selectively modulated by troglitazone are illustrated. The nature of the interaction is described with the following abbreviations: A, activation; E, expression; PD, protein-DNA interactions; PP, protein–protein interactions; RB, regulation of binding; TR, translocation.
Figure 7.
 
ChIP analysis of PPARγ binding to putative PPREs in genes selectively modulated by troglitazone. Chromatin from naïve, tBH-treated cells, tBH+troglitazone, tBH+rosiglitazone, or tBH+pioiglitazone was immunoprecipitated with anti-PPARγ antibodies. The binding of PPARγ to associated DNA was then determined by qPCR, using primer pairs shown in Table 1 flanking putative PPREs for CDKN1C, AMOTL2, and CCL2 (A) and CXCR4 (B).
Figure 7.
 
ChIP analysis of PPARγ binding to putative PPREs in genes selectively modulated by troglitazone. Chromatin from naïve, tBH-treated cells, tBH+troglitazone, tBH+rosiglitazone, or tBH+pioiglitazone was immunoprecipitated with anti-PPARγ antibodies. The binding of PPARγ to associated DNA was then determined by qPCR, using primer pairs shown in Table 1 flanking putative PPREs for CDKN1C, AMOTL2, and CCL2 (A) and CXCR4 (B).
Table 1.
 
Primers Used for ChIP Analysis
Table 1.
 
Primers Used for ChIP Analysis
Gene Site Primer Sequences (5′–3′)
CXCR4 +1100 CCGCGAGCGTCTTTGAAT
TCCGAGTCCAGTTGTGATGT
CCL2 −1190 TGGGCTAGGAGAATCGAGAG
GGGAATTGGAAAGCCTGAC
CDKN1C −1000 CTGCCCAGTAGATGGAACAG
ACAGCCCCTTCCCCTATGAC
AMOTL2 −1570 CCCTTACGAAGAGAAGCTGTG
−1670* CCCAAATGCCTGAAAGACAG
JUB −1160 TTTCTCATTTCAGGTGACTGTG
TTTCCTCTACACGGCATTTC
Table 2.
 
List of Genes Normalized ≥50% by Troglitazone Treatment and Common to Both Experiments
Table 2.
 
List of Genes Normalized ≥50% by Troglitazone Treatment and Common to Both Experiments
  List of Genes Normalized ≥50% by Troglitazone Treatment and Common to Both Experiments
×
×

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.

×