Cigarette smoke is the most important environmental risk factor for developing age-related macular degeneration (AMD). Damage to the retinal pigment epithelium (RPE) caused by cigarette smoke may underlie the etiology of AMD. This study investigated the molecular and cellular effects of cigarette smoke exposure on human RPE cells. ARPE-19 or primary human RPE cells were exposed to cigarette smoke extract (CSE) or hydroquinone (HQ), a component of cigarette smoke. The effect of this exposure on key aspects of RPE vitality including viability, cell size, mitochondrial membrane potential (ΔΨm), superoxide production, 4-hydroxy-2-nonenal (4-HNE), vascular endothelial growth factor (VEGF), and heme oxygenase-1 (HO-1) expression was determined. Exposure of RPE cells to CSE or HQ caused oxidative damage and apoptosis, characterized by a reduction in cell size and nuclear condensation. Evidence of oxidative damage also included increased lipid peroxidation (4-HNE) and mitochondrial superoxide production, as well as a decrease in intracellular glutathione (GSH). Exogenous administration of antioxidants (GSH and N-acetyl-cysteine) prevented oxidative damage to the RPE cells caused by CSE. Cigarette smoke also induced expression of VEGF, HO-1, and the transcription factor nuclear factor erythroid-derived 2, like 2 (NRF2). However, NRF2 was only modestly involved in CSE-induced HO-1 expression, as shown by the NRF2 small interfering RNA studies. These new findings demonstrate that cigarette smoke is a potent inducer of oxidative damage and cell death in human RPE cells. These data support the hypothesis that cigarette smoke contributes to AMD pathogenesis by causing oxidative damage and cell death to RPE cells.
- heme oxygenase-1
- vascular endothelial growth factor
- nuclear factor erythroid-derived 2, like 2
age-related macular degeneration (AMD) is the leading cause of blindness among the elderly in industrialized nations (60). Cigarette smoking is the single most important environmental risk factor for developing AMD. Current smokers are estimated to have a 45% greater probability of developing early AMD and exhibit worse AMD disease progression, compared with nonsmokers (34). Smoking intensity also increases the risk of developing AMD (66), likely through oxidative injury (66). Cigarette smoke yields an estimated 1017 oxidant molecules per puff (12) that can damage retinal pigment epithelial (RPE) cells. In addition to evidence of oxidative damage in AMD patients (16, 59), oxidative stress can alter RPE cells (9) and increase the expression of growth factors conducive to vascular endothelial cell growth (18, 26, 29, 47). Oxidative stress is thought to be important in forming lipofuscin and drusen (33, 61), both of which accumulate in the eye and are thought to be major players in AMD pathophysiology (33, 72). Subsequent loss of RPE cells due to oxidative stress caused by cigarette smoke exposure may underlie AMD pathogenesis and/or contribute to disease progression (16, 21, 30, 59). Antioxidant vitamins also decrease the risk of AMD progression (3), highlighting the importance of oxidative damage in AMD.
While some individual components of cigarette smoke cause oxidative damage and apoptosis in RPE cells (31, 58, 63), there is no information regarding the direct effects of cigarette smoke exposure on human RPE cells. We therefore sought to investigate the molecular and cellular basis for the epidemiological association between cigarette smoke and AMD. Here, we show that cigarette smoke extract (CSE), a widely used in vitro model of cigarette smoke exposure (8, 44, 57), causes oxidative damage to human RPE cells in vitro. The results reported herein strengthen the hypothesis that cigarette smoking contributes to AMD pathogenesis by causing oxidative damage to RPE cells. Understanding key molecular pathways within RPE cells that govern oxidative stress will provide additional therapeutic and preventative targets to treat AMD.
MATERIALS AND METHODS
The human retinal pigment epithelial cell line, ARPE-19, was purchased from the American Type Culture Collection (Manassas, VA). These cells were grown and maintained in DMEM-F-12 culture media and supplemented with 10% fetal bovine serum, 2 mM l-glutamine, 15 mM HEPES, 100 U/ml penicillin, and 100 μg/ml streptomycin and maintained at 37°C in humidified 7% CO2-93% air. Primary human RPE cells were purchased from ScienCell Research Laboratories (Carlsbad, CA) and cultured according to supplier's recommendations. For all experiments, controls were treated the same as the experimental cultures except for the addition of CSE or hydroquinone. For all experiments, cells were serum-deprived 24 h prior. In parallel experiments, some cells were pretreated with 1 mM N-acetyl-cysteine (NAC) for 1 h, followed by coincubation with the CSE. Cells were also pretreated with 5 mM glutathione (GSH) ethyl ester for 2 h.
Preparation of CSE
Research-grade cigarettes (1R3F) with filter were obtained from the Kentucky Tobacco Research Council (Lexington, KY), and CSE was generated as previously described (7, 11). Briefly, CSE was prepared by bubbling smoke from two cigarettes into 20 ml of serum-free culture medium at a rate of 1 cigarette/min (7). The pH of the culture medium was adjusted to 7.4 and sterile filtered with a 0.45-μm filter (25-mm Acrodisc; Pall, Ann Arbor, MI). An optical density (taken at a wavelength at 320 nm) of 0.65 was considered to represent 10% CSE (2). Dilutions to the appropriate concentrations were carried out in serum-free DMEM-F-12 culture media.
MTT was performed to assess viability after CSE exposure as previously described (7). ARPE-19 cells or primary human RPE cells were cultured in flat-bottomed 96-well plates until they reached ∼70% confluence. After treatment, 5 mg/ml MTT (in PBS) was added to each well for 4 h at 37°C, and the insoluble precipitate was dissolved by adding of 200 μl of dimethyl sulfoxide (DMSO) to each well. The plates were read with a Bio-Rad (Hercules, CA) microplate reader at 510 nm.
Morphological changes were assessed by seeding cells onto eight-well type I collagen-coated glass chamber slides (BD Biosciences, Bedford, MA) at a density of 10,000 cells/well and left undisturbed for 48 h. The culture medium was then replaced with serum-free medium for 24 h, after which cells were exposed to 1% CSE or medium alone (control) for 24 h. Following treatments, cells were fixed with 4% paraformaldehyde for 10 min, stained with hematoxylin and eosin (H&E) and coverslipped in Immuno-Mount (Shandon, Pittsburgh, PA). H&E was visualized with an Olympus BX51 microscope (New Hyde Park, NY), and images were captured with a SPOT camera (New Hyde Park, NY).
For determination of heme oxygenase-1 (HO-1) and nuclear factor erythroid-derived 2, like 2 (NRF2) expression, ARPE-19 cells were treated with CSE for 8 h or 2 h and HO-1 and NRF2 expression was examined, respectively, utilizing antibodies as described previously (8).
Western Blot Analysis
Cells were seeded at a density of 550,000 cells/100-mm tissue culture dish and allowed to grow for ∼72 h in the medium described above. The cells were incubated for another 24 h in the same media but without serum. ARPE-19 cells were then treated with 1% CSE for 2–24 h. Cells were lysed in buffer containing 150 mM NaCl, 1% Igepal, 50 mM Tris, 1 mM EDTA, and 10% protease inhibitor mixture. Protein quantification was performed with the bicinchoninic acid (BCA) method according to the manufacturer's instructions (Thermo Scientific, Rockford, IL). Ten micrograms of total cellular protein was fractionated on 10% SDS-PAGE gels, electroblotted onto Immun-blot polyvinylidene difluoride membrane (Bio-Rad), and blocked with 5% nonfat dry milk in 0.1% Tween 20 (in PBS) overnight at 4°C. Changes in protein expression following CSE exposure were assessed using antibodies against HO-1, HO-2 (1:5,000, Stressgen, Ann Arbor, MI), 4-hydroxy-2-nonenal (4-HNE; 1:200, OxisResearch, Portland, OR), VEGF (1:1,000, Abcam, Cambridge, MA), NRF2 (1:500, Santa Cruz Biotechnology, Santa Cruz, CA), GAPDH (1:5,000, Calbiochem, San Diego, CA), or actin (1:20,000; Oncogene Research Products, San Diego, CA). Protein was visualized by enhanced chemiluminescence (NEN Life Science Products, Boston, MA) and developed on Classic X-ray film (Kodak, Rochester, NY). Densitometric analysis of protein expression was performed with Kodak 1D Imaging Software (Kodak Scientific Imaging Systems, New Haven, CT); values were normalized to total actin or GAPDH.
NRF2 Small Interfering RNA-Mediated Gene Silencing
ARPE-19 cells were plated in 100-mm cell culture dishes at 550,000 cells/dish in DMEM-F-12 culture media supplemented with 10% fetal bovine serum (without antibiotic) and allowed to reach 70% confluence. After ARPE-19 cells were washed with sterile PBS, a NRF2 small interfering RNA (siRNA) or control siRNA (67) (5′-UUCUCCGAACGUGUCACGU-3′) was added to the cells. The NRF2 siRNA was prepared by combining NRF2 siRNA (Santa Cruz Biotechnology) with Opti-Mem media and Lipofectamine LTX (Invitrogen, Carlsbad, CA) for a final concentration of 10 nM. This mixture was incubated at room temperature for 30 min before being added to ARPE-19 cells. Dishes were mixed gently and incubated for 24 h. After 24 h, 2 ml of DMEM-F-12 culture media were added and cells were allowed to incubate for another 24 h. Following this, samples were exposed to 1% CSE for 2 h or 8 h. Finally, cells were harvested for Western blot analysis as described above.
Measurement of Glutathione
Following exposure to CSE, cells were washed with ice-cold PBS and scraped into ice-cold extraction buffer (0.1% Triton X-100, 0.6% sulfosalicylic acid in 0.1 M phosphate buffer with 5 mM EDTA, pH 7.5). Determination of total intracellular levels of GSH was performed as described by Rahman et al. (51) with DTNB-GSSG/glutathione reductase recycling method (51).
Inhibition of Heme Oxygenase Activity
ARPE-19 cells were cultured in flat-bottomed 96-well plates at a density of 8,000 cells/well until they reached ∼70% confluence. After serum starving for 24 h, ARPE-19 cells were pretreated with 5 μM tin protoporphyrin-IX (snPPIX) for 3 h and then coexposed to 0.5% CSE for 24 h. Following exposure, the MTT assay was used to assess cell viability (as described above).
Measurement of VEGF Production by ELISA
ARPE-19 cells (20,000 cells/well) were cultured with 0.1 ml of cell culture medium in 96-well plates for 48 h, serum starved for 24 h, and then treated with 1% CSE. Cells were incubated at 37°C in 7% CO2, and media were collected after 2, 6, 8, 16, and 24 h of CSE exposure. Levels of VEGF were assessed by ELISA (R&D Systems, Minneapolis, MN) according to the manufacturer's protocol.
Measurement of Apoptosis by Flow Cytometry
Alterations in cell size, indicative of apoptosis (7, 46), were determined by flow cytometry. Equivalent numbers of ARPE-19 cells were either exposed or not exposed to 1% CSE for 3 or 24 h. Cells were then washed, trypsinized, and resuspended in PBS. Flow cytometric analysis was performed with a Becton-Dickinson FACSCalibur flow cytometer (BD Biosciences, Mountain View, CA); 20,000 events were acquired for each sample.
Mitochondrial membrane potential.
DiOC6 (Molecular Probes, Carlsbad, CA) is a dye that labels active mitochondria; a decrease in fluorescence is indicative of cell death. Here, cells were grown to confluence and treated with control media or 1% CSE for 6 h. After treatments, DiOC6 was added at a final concentration of 40 nM for 15 min at 37°C as described previously (7). Cells were then resuspended in PBS and analyzed by flow cytometry. Controls included cells that were treated with control media without DiOC6 and cells that were treated with hydroquinone (500 μM), an inducer of oxidative stress in ARPE-19 cells (4).
Mitochondrial superoxide production.
ARPE-19 cells were cultured in 25-cm2 cell culture flasks at a density of 250,000 cells/flask. After 3 h of exposure with 1% CSE, cells were washed, trypsinized, and resuspended in 2 μM MitoSOX reagent. Cells were then incubated at 37°C for 10 min, resuspended in warm HBSS-Ca2+-Mg+ buffer and analyzed by flow cytometry.
Statistical analysis was performed with Statview version 5.0 software (SAS Institute, Cary, NC), and analysis of variance and Fisher's post hoc test were used to assess differences between multiple treatment groups. P < 0.05 was considered to be statistically significant.
CSE Decreases Viability in Human RPE Cells
CSE exposure decreased ARPE-19 cell viability in a dose-dependent manner (Fig. 1). There was a significant loss in viability after ARPE-19 cells were exposed to 0.5% CSE (P < 0.01), and the cell viability loss further decreased with increasing concentrations of CSE (1%, 2%, and 4% CSE; Fig. 1). The approximate median lethal dose value (LD50) for ARPE-19 cells was 1% CSE. Primary human RPE cells exhibited a similar sensitivity (data not shown). We also used lactate dehydrogenase (LDH) release to measure cell viability. The LDH assay showed similar results in loss of viability as the MTT assay (data not shown). Therefore, we used ARPE-19 cells for the remaining experiments, because these cells are a commonly used cellular model for investigating the physiology of the RPE (4, 6). Because the LD50 for both the primary RPE and ARPE-19 cells was 1% CSE, we used 1% CSE for the remainder of the experiments.
CSE Induces Morphological Changes Characteristic of Apoptosis in ARPE-19 Cells
ARPE-19 cells treated with CSE exhibited many of the morphological features of apoptotic cells, including fragmented nuclei and nuclear condensation (Fig. 2A, arrows). Note the decrease in the number of cells within the field of the CSE-treated RPE cells (Fig. 2A, asterisk).
ARPE-19 cells also exhibited a significant reduction in cell size, an early, prerequisite hallmark of cells that are undergoing apoptosis (7, 42, 46). Flow cytometric analysis of ARPE-19 cells that were unexposed to CSE showed that cells were uniform in size and granularity, exhibiting normal cell volume, with ∼93% of cells being found in this region, designated region 3 (Fig. 2B, unexposed, R3). There were two additional regions, R1 and R2, which likely consisted of cell debris (R1) and ARPE-19 cells of slightly smaller size (R2) (7). Exposure to 1% CSE caused a dramatic decrease in cell size, as evidenced by a decrease in forward scatter (Fig. 2B). The reduction in cell size significantly decreased the percentage of cells within region R3 (92.6 ± 0.6% vs. 56.8 ± 2.6%; P < 0.0001; Fig. 2C). This decrease in cell size correlated with a significant increase in the percentage of cells within the intermediate R2 region (from 2.1 ± 0.1% to 12.7 ± 1.1%; P < 0.01; Fig. 2C). There was also a significant increase in the R1 region, the region corresponding to the smallest size (from 2.7 ± 0.03% to 14.5 ± 0.5%; P < 0.01; Fig. 2C).
CSE Reduces Mitochondrial ΔΨm in ARPE-19 Cells
Mitochondrial membrane permeabilization (ΔΨm) irreversibly commits cells to undergo death and is an early marker of apoptosis (39, 50). Reduced mitochondrial ΔΨm can be measured by a diminished incorporation of the dye DiOC6 into the mitochondria of cells. Exposure of ARPE-19 cells with 1% CSE for 6 h resulted in a decrease in DiOC6 incorporation compared with cells exposed to media alone (Fig. 3, compare red and green histograms, respectively). Hydroquinone, a component of cigarette smoke and known inducer of oxidative damage and apoptosis in ARPE-19 cells (63) also decreased mitochondrial ΔΨm (Fig. 3, blue histogram). Together with the decrease in cell size and reduction in viability after CSE exposure, these data indicate that ARPE-19 cells are undergoing apoptosis in response to cigarette smoke exposure.
CSE Induces Oxidative Stress in Human RPE Cells
A consequence of oxidative stress is membrane lipid peroxidation (7). 4-HNE is a specific product of lipid peroxidation that occurs in response to oxidative stress (13), including cigarette smoke exposure (7). Therefore, we examined the ability of CSE to induce lipid peroxidation in ARPE-19 cells. Western blot analysis demonstrated that there were numerous 4-HNE-modified proteins in ARPE-19 cells exposed to cigarette smoke. Here, intense 4-HNE adduct formation was evident, at a mass of ∼96 kDa (Fig. 4A), starting at 2 h post CSE exposure and lasting through 16 h. Immunocytochemical staining also revealed intense 4-HNE cellular staining (data not shown). These data suggest that there is an increase in lipid peroxidation, consistent with the ability of CSE to cause oxidative stress.
GSH is an important antioxidant that aids in eliminating toxic chemicals (7). Importantly, a decrease in intracellular levels of GSH is indicative of oxidative stress (10). Therefore, we measured the levels of GSH in CSE-exposed ARPE-19 cells. There was a significant decrease in intracellular GSH when ARPE-19 cells were exposed to 0.5% CSE compared with cells that were cultured in control media (Fig. 4B; 15 ± 3.3 nmol/mg protein vs. 35 ± 8.4 nmol/mg protein, respectively, P < 0.05). This drop in GSH levels after exposure to 0.5% CSE corresponded to a significant decrease in viability (P < 0.01; compare to Fig. 1). Higher percentages of CSE (1%, 2%, and 4%) also decreased the amount of GSH in ARPE-19 cells (Fig. 4B; 12 ± 4.6, 9 ± 4.4, and 9 ± 4.3 nmol/mg protein, respectively). There were no significant differences between exposure to 0.5% and 1% CSE for 24 h for viability or GSH levels (P > 0.05). Buthionine sulfoximine (BSO), an irreversible inhibitor of GSH synthesis (27), also significantly reduced the amount of intracellular GSH (Fig. 4B; P < 0.01). To ascertain whether the decrease in GSH was the result of a concomitant rise in GSSG, the oxidized form of GSH, we assayed for GSSG levels but did not find any significant increases (data not shown). A previous study has shown that CSE forms conjugates with GSH, which may explain why increases in GSSG levels were not found after CSE exposure (37).
Treatment With GSH and NAC Prevents CSE-Induced Oxidative Damage in Human RPE Cells
We also examined whether NAC, a precursor of GSH as well as GSH, could attenuate the oxidative damage associated with exposure to cigarette smoke. We pretreated ARPE-19 cells with NAC (1 mM) or GSH (5 mM) in the presence or absence of CSE and assessed mitochondrial ΔΨm, mitochondrial superoxide production, and lipid peroxidation (4-HNE). Exposure to 1% CSE caused a dramatic decrease in mitochondrial ΔΨm, (Fig. 5A, green histogram; compare with Fig. 3, red histogram). Treatment with either GSH or NAC prevented this decline in mitochondrial ΔΨm (Fig. 5A). Similarly, treatment with NAC or GSH dramatically attenuated lipid peroxidation (data not shown) as well as mitochondrial superoxide production (Fig. 5B). These data highlight the importance of reducing oxidative stress caused by cigarette smoke in RPE cells.
CSE Induces VEGF in Human RPE Cells
VEGF is an important component in the pathology of wet AMD, with anti-VEGF pharmacotherapy now considered an important standard of care (5, 68). To determine whether CSE influences VEGF expression in ARPE-19 cells, we assessed VEGF levels following exposure of ARPE-19 cells to CSE. When ARPE-19 cells were exposed to 1% CSE, there was a time-dependent increase in VEGF protein expression (Fig. 6A). We also used an ELISA-based approach and found that there was a significant induction in VEGF production after 24-h CSE exposure (139.03 ± 16.7 pg/ml) compared with control (25.12 ± 4.2 pg/ml; Fig. 6B; P < 0.01).
CSE Activates Intracellular Pathways Associated With Oxidative Stress in ARPE-19 Cells
Heme oxygenases are cytoprotective enzymes that catalyze the rate-limiting step in the degradation of heme. HO-1 is the inducible isozyme whose expression increases during conditions of oxidative stress (8, 49). Western blot analysis demonstrates that ARPE-19 cells constitutively express both HO-1 and HO-2 (Fig. 7A). When ARPE-19 cells were exposed to 1% CSE, there was a significant increase in HO-1 protein expression after 8 h of exposure (4.4 ± 0.85 difference compared with control, Fig. 7A; P < 0.05). A similar pattern of HO-1 induction was observed in primary human RPE cells (data not shown). Immunofluorescence confirmed that some ARPE-19 cells have basal HO-1 expression (Fig. 7B; control) that was induced after 8 h of 1% CSE exposure (Fig. 7B). This CSE-induced HO-1 expression in ARPE-19 cells was localized both in the cytoplasm and in the nucleus (Fig. 7B, inset). In contrast, densitometric analysis revealed that there were no significant increases in HO-2 after CSE exposure (2 h: 0.82 ± 0.01; 8 h: 1.03 ± 0.25; 16 h: 1.14 ± 0.31; 24 h: 0.68 ± 0.02 ± SE; P > 0.05).
To examine whether HO activity was important in preventing damage caused by CSE, ARPE-19 cells were pretreated with snPPIX, a well-characterized inhibitor of HO activity (19, 45). Neither DMSO nor snPPIX alone significantly reduced viability (Fig. 7C). Similar to results presented in Fig. 1, exposure to 0.5% CSE reduced viability of ARPE-19 cells. This decrease in viability with a low percentage of CSE was further reduced when ARPE-19 cells were treated with snPPIX (Fig. 7C). These data indicate that HO activity is a critical component of the antioxidant defense system of human RPE cells.
The molecular mechanisms governing the induction of HO-1 are diverse, and a recent report highlights the importance of the redox-sensitive transcription factor NRF2 in CSE-induced HO-1 in human epithelial cells (8). Therefore, we examined whether NRF2 was involved in the ability of CSE to increase HO-1 expression. Exposure to CSE increased NRF2 protein expression, compared with levels in media-treated cells (Fig. 8, A and B). We then used a siRNA-based approach to genetically decrease NRF2 expression in ARPE-19 cells. This resulted in a significant decrease in CSE-induced NRF2 expression after 2 h (Fig. 8C, compare lanes 7 and 8; P < 0.01). Quantification showed significant decreases in NRF2 expression in NRF2 siRNA-treated cells exposed to 1% CSE for 8 h compared with cells treated with control siRNA exposed to 1% CSE for 8 h (P < 0.01) The remarkable reduction in NRF2 expression resulted in only a modest attenuation of HO-1 expression caused by 1% CSE (Fig. 8C, compare lanes 7 and 8). Densitometric analysis showed that HO-1 was not significantly decreased with NRF2 knockdown at either 2 h or 8 h. These results are the first to demonstrate the molecular regulation of HO-1 by NRF2 in response to cigarette smoke in human RPE cells.
While the loss of vision associated with AMD results from damage to the photoreceptors, the primary insult involves the RPE (62, 69). Although there are numerous environmental risk factors for AMD, cigarette smoking remains the single most important environmental factor associated with developing AMD (14). RPE dysfunction from oxidative stress (22) caused by cigarette smoke may contribute to some key components of AMD, such as choroidal neovascularization (17, 24, 25, 29, 32), the main cause of vision loss (70). Cigarette smoke is a complex mixture of more than 5,000 chemicals, and several individual components contained within cigarette smoke have been shown to induce oxidative stress in RPE cells (4, 31). However, chronic smokers are continuously exposed to a myriad of cigarette smoke components. Therefore, we chose to assess the effects of CSE on human RPE cells. CSE is a widely accepted model system for studying in vitro effects of tobacco smoke (36, 44, 48) and contains most of the chemicals inhaled by smokers (57). A recent study by Fujihara et al. (23) has shown the results of cigarette smoke exposure in an in vivo mouse model. Although this study (23) demonstrates damage to mouse retina from cigarette smoke, it is important to keep in mind the important differences between mice and humans. Our laboratory has published that, despite being described as a key regulator of oxidative stress in most cell types, nuclear factor erythroid-derived 2, like 2 (Nrf2) is not a dominant transcription factor in the induction of HO-1 in human lung fibroblasts (8). Yet, this pathway is robustly inducible in mouse lung fibroblasts. This strongly suggests that there are important species-specific differences in oxidative stress pathways between human and mouse cells. Also noteworthy are important anatomical differences between mouse and human eyes. The mouse eye, for example, does not contain a macula, the major anatomical region affected in AMD (43). Our use of human RPE cells, in conjunction with cigarette smoke extract, rather than individual components of CSE, strengthens epidemiological data linking cigarette smoke and AMD. Herein, we have shown that exposure of human RPE cells to CSE results in oxidative stress and cell death, and the results of our study strengthen the epidemiological evidence associating cigarette smoke, RPE dysfunction, and AMD.
CSE significantly reduced viability in both ARPE-19 and primary RPE cells (Fig. 1). CSE may reduce viability via alterations in mitochondrial integrity. Mitochondrial integrity is crucial for cell survival, and decreased mitochondrial ΔΨm is an early sign of apoptosis (39, 50). Our results are in agreement with previous studies demonstrating that RPE cells have mitochondrial dysfunction and undergo apoptosis when exposed to acrolein, a component of cigarette smoke (31). This decline in mitochondrial ΔΨm caused by CSE (Fig. 3) could be prevented by boosting intracellular antioxidant levels via exogenous GSH (Fig. 5). GSH is an important endogenous antioxidant in the eye (10), and previous studies have shown that plasma GSH levels are significantly lower in AMD patients (15, 56). Thus, a decrease in GSH may render the RPE sensitive to the oxidative effects of cigarette smoke. Our data clearly indicate that there is a significant decrease in GSH after exposure to CSE (Fig. 4). This decrease in GSH correlated with viability. After ARPE-19 cells were exposed to 0.5% CSE, there was a significant decline in viability (P < 0.01) as well as a significant decrease in GSH levels (P < 0.05). This suggests that cell viability tracks with GSH content. In addition, there was no significant difference in viability or GSH levels between 0.5% and 1% CSE (P > 0.05), implying that without major changes in GSH levels, there are little changes in cell viability. Therefore, a decline in GSH, elicited by the oxidant properties of cigarette smoke, followed by mitochondrial dysfunction, may irreversibly commit RPE cells to an apoptotic fate.
Mitochondrial injury in response to cigarette smoke may also be hastened by lipid peroxidation. 4-HNE is a toxic product of lipid peroxidation that is increased by smoke exposure (7) and can cause both mitochondrial (54) as well as lysosomal dysfunction. Abnormal lysosome functioning can cause accumulation of large amounts of undigested material that may facilitate the progression of AMD (38). Our study shows increased 4-HNE levels after ARPE-19 cells were exposed to CSE (Fig. 4). In addition, 4-HNE can increase levels of VEGF (71), which is constitutively produced by RPE cells (1). Indeed, we were able to detect increases in VEGF expression in ARPE-19 cells after exposure to CSE (Fig. 6). VEGF is an important stimulator of neovascularization and is considered a key cytokine in the pathophysiology of wet AMD (5, 28). Thus, in addition to the cellular damage caused by increased 4-HNE, induction of VEGF may further heighten the progression of AMD via chronic overproduction by RPE cells.
Cigarette smoke-induced decrease in GSH is associated with induction of HO-1 (8). HO-1 is the rate-limiting enzyme in the oxidative degradation of heme into equimolar concentrations of iron, bilirubin, and carbon monoxide. We speculated that HO-1 expression would increase in response to CSE in ARPE-19 cells. Our study demonstrates that there is an increase in HO-1 protein expression in ARPE-19 cells after CSE exposure (Fig. 7). The cytoprotective properties of HO-1 are well documented in many organ systems (55, 64) including the retina (40). Inhibition of HO activity synergized with CSE to augment cell death (Fig. 7), suggesting that HO-1 may protect against oxidative stress caused by cigarette smoke. While many studies show HO-1 to be protective, a study by Lee et al. (41) demonstrated cellular toxicity from high levels of free iron via enhanced HO-1 catalytic activity/expression. Given the evidence that iron accumulation within the eye may account for pathological progression of AMD (20), we felt it vital to determine the role of HO-1 in cigarette smoke-induced cell death in RPE cells. Indeed, our data suggest that HO activity is beneficial and may be protective against cigarette smoke-induced oxidative stress and cell death in human RPE cells.
The ability of CSE to increase HO-1 may involve many transcription factors. We began by assessing whether NRF2 was a key factor involved in this induction. NRF2 is activated by cigarette smoke in human lung epithelial cells (8), and its expression increases in mouse NIH 3T3 cells on exposure to CSE (35). Indeed, we observed an increase in NRF2 expression on exposure of ARPE-19 cells to CSE (Fig. 8). Using a siRNA-based approach to knock down NRF2 expression, we observed that a near complete loss in NRF2 protein resulted in only a modest attenuation of CSE-induced HO-1 expression (Fig. 8C). We also found that a reduction in NRF2 expression did not adversely affect viability (data not shown). Despite strong evidence that NRF2 is a key transcription factor involved in protection against oxidative stress (including HO-1 induction) in many mouse models of cigarette smoke-induced disease (i.e., emphysema) (52, 53, 65), our data strongly suggest that other transcription factors play a more dominant role in modulating HO-1 expression and survival in human RPE cells. It has been reported in human lung fibroblasts that both activator protein-1 and nuclear factor-κB can also influence HO-1 expression caused by cigarette smoke (8). Indeed, NRF2 was not activated in this cell type, and the authors speculated that important cell and species differences exist in signaling pathways between human and mouse cells (8). Further study is needed to better understand the molecular mechanisms governing oxidative stress-induced genes in human RPE cells.
Cigarette smoke is an etiological factor in AMD pathogenesis. Here, we have shown that exposure to cigarette smoke causes profound oxidative damage to human RPE cells. This damage was associated with increased VEGF levels, mitochondrial dysfunction, altered GSH levels, increased lipid peroxidation, and increased stress-associated proteins, namely, HO-1. The combined effects of these parameters can account for the morphological and physiological changes associated with AMD progression. Antioxidants can halt AMD progression (3), and, in agreement, our results show that exogenous administration of antioxidants, including GSH, can prevent damage to the RPE cells caused by cigarette smoke. Thus, preventive strategies aimed at smokers, such as augmenting GSH levels, may prove useful in combating cigarette smoke-related diseases such as AMD.
This research was supported by a Parker B. Francis Fellowship and American Thoracic Society Research Grant (to C. J. Baglole); Research to Prevent Blindness Career Development Award (to R. T. Libby); University of Rochester Department of Environmental Medicine Pilot Project Grant (to R. T. Libby); Toxicology Training Grant T32ES07026 (to K. M. Bertram); National Institutes of Health Grants ES01247, HL075432, and EY017123; and a Challenge Grant from The Research to Prevent Blindness (Department of Ophthalmology at University of Rochester).
- Copyright © 2009 American Physiological Society