Cysteine Starvation Activates the Redox-Dependent Mitochondrial Permeability Transition in Retinal Pigment Epithelial Cells

Abstract

purpose. Glutathione (GSH) plays a key role in protection against oxidative stress. l-Cysteine is thought to be rate-limiting for the synthesis of glutathione (GSH) and therefore may be a critical component in protection against oxidative stress. The purpose of this study was to investigate the role of l-cysteine in GSH metabolism and oxidative stress in human retinal pigment epithelial (hRPE) cells.

methods. To identify the role of cysteine in GSH metabolism in hRPE cells, a strategy of cysteine starvation was used to determine (1) GSH levels and oxidative stress by measuring reactive oxygen species (ROS) production, (2) mitochondrial membrane potential (Δψ_m) and mitochondrial ultrastructure by using conventional electron microscopy (EM), and (3) indices of cell viability and apoptosis including analysis of cells containing hypodiploid amounts of DNA.

results. Cysteine starvation resulted in approximately a 95% decrease in GSH concentrations over 24 hours. The GSH Nernst redox potential (E _h) increased approximately 70 mV (E _h = −248 ± 2.9 mV in control cells compared with E _h = −179 ± 2.0 mV in cysteine-starved cells) indicating significant intracellular oxidation. Cysteine starvation increased the production of ROS by mitochondrial respiratory complex III (cytochrome bc ₁), determined using a pharmacological strategy that resulted in the loss of Δψ_m and cell death. The loss of Δψ_m and cell death was prevented with bongkrekic acid, an inhibitor of the adenine nucleotide translocator inhibitor, suggesting activation of the mitochondrial permeability transition (MPT). This conclusion was further supported by electron microscopic studies that showed significant mitochondrial swelling, a hallmark of MPT activation. Cell death was not prevented with either the broad-spectrum caspase inhibitor zVADfmk or the caspase 3–specific inhibitor DEVD-CHO, indicating that cytochrome bc ₁–mediated ROS production results in the MPT and necrosis.

conclusions. These results show that cysteine is a required component for normal GSH metabolism and protection against oxidative stress in hRPE cells.

Several age-related clinical conditions, including Parkinson’s disease, Alzheimer’s disease, and age-related macular degeneration (ARMD) are associated with decreased levels of glutathione (GSH) ^{1 2} and/or increased oxidative stress. ^{1 2 3}  

ARMD is a degenerative condition of the macula, often leading to blindness due to dysfunction of the retinal pigment epithelium (RPE) and overlying retina. ³ The cause of ARMD is unknown, but results of the recent multicenter Age-Related Eye Disease Study (AREDS) clinical trial strongly support an oxidative stress component in ARMD, because a significant reduction in the rate of ARMD progression was observed in subjects taking antioxidant and zinc-containing supplements. ⁴ Moreover, because the RPE is close to the underlying choriocapillaries with a high partial oxygen pressure, and because of the role of the RPE in phagocytosis, this cell type is susceptible to increased physiological risk of oxidative stress. ³ To counter this susceptibility to oxidative stress, the RPE has effective antioxidant defenses, including the antioxidant enzymes superoxide dismutase, catalase, and the glutathione S-transferases; and the compounds ascorbate, vitamin E, and glutathione (GSH.). ^{4 5 6} However, it is thought that these defenses may be significantly reduced with increasing age, predisposing the RPE to increased oxidative-stress–mediated damage. ^{7 8}  

One of these antioxidants, GSH, is known to play a key role in the regulation of intracellular signaling, the maintenance of redox status and the protection against oxidative stress. ^{9 10} The chemical induction of GSH defenses in human (h)RPE cells has been effective in the protection against exogenous oxidants. ^{11 12} In the case of the RPE, this has led to the proposal of various pharmacological strategies aimed at increasing GSH levels in the RPE to increase its antioxidant potential and thereby block or reduce ARMD progression. ¹³ To design safe and effective clinical strategies aimed at enhancing GSH status, a thorough knowledge of the role of factors involved in the regulation of GSH metabolism is necessary. ¹⁴ One of the critical factors regulating GSH metabolism is substrate availability. ^{14 15 16} In particular, cysteine is limiting for GSH synthesis, because there is only a small pool of the amino acid available to sustain a much larger metabolically active GSH pool. ³ This fact has previously stimulated investigations into the relationship between diet, and plasma and tissue cysteine levels and GSH. ^{10 11 17 18} As expected from these studies, factors that stimulate cysteine uptake also typically increase cellular and tissue GSH, ^{16 17} providing an important link between cysteine availability, GSH, and the protection against oxidative stress. ^{19 20} Because it is known that GSH is critical for the maintenance of the intracellular redox status and the protection against oxidative stress of both cellular and subcellular organelles, including the mitochondrion, ^{20 21 22 23} it is, therefore, also apparent that cysteine availability may also be a key factor involved in protection against mitochondrial oxidative stress. 

To date, the role of oxidative stress in ARMD has been modeled in vitro using relatively high concentrations of exogenous oxidants including tert-butyl hydroperoxide (TBH) and hydrogen peroxide (H₂O₂). ^{24 25 26 27 28} However, it has been suggested that correlating the results of such studies to explain the oxidative stress component of ARMD, which is a chronic progressive disease, may not be relevant to the in vivo condition. ¹ In an attempt to resolve this issue, we used a cysteine-starvation protocol to study the role of cysteine in GSH redox metabolism and protection against mitochondrial oxidative stress in hRPE cells. The consequences of cysteine depletion in hRPE cells included increased reactive oxygen species (ROS) production by mitochondrial respiratory complex III (cytochrome bc ₁), followed by loss of mitochondrial membrane potential (Δψ_m) and mitochondrial swelling due to the activation of the mitochondrial permeability transition (MPT). Because we found that cell death was not inhibited by blocking caspase activation, this indicated that the redox-dependent MPT resulted predominantly in necrotic cell death. These results suggest that the availability of cysteine is a crucial factor for normal GSH metabolism in hRPE cells and is also necessary for protection against oxidative stress-dependent MPT and cell death. Thus, cysteine depletion may be useful in modeling the chronic oxidative stress thought to occur in diseases where there is an oxidative stress component including ARMD. 

Methods

Cell Culture

hRPE cell cultures were established from donor eyes obtained through the Georgia Eye Bank, as previously described. ¹⁰ Methods for securing human tissues complied with the Emory University Human Investigations Review Board and the Declaration of Helsinki. Experiments were performed on hRPE cells cultured between the 4th and 10th passages in DMEM supplemented with fetal calf serum (FCS) and antibiotics at 37°C under 95% air and 5% CO₂. ¹²  

Measurement of Intracellular GSH Redox Change

hRPE cells were incubated for 24 hours in cysteine-free DMEM or in DMEM treated with 500 μM BSO (a pharmacologic inhibitor of γ-glutamyl cysteine ligase [γGCL], the rate-controlling enzyme for GSH synthesis), pelleted by centrifugation and extracted with 5% perchloric acid and saturated boric acid. Intracellular GSH was measured with high-pressure liquid chromatography, as previously described. ²² The amount of acid-insoluble protein was determined by the Bradford method with γ-globulin as a standard. The redox potential (E _h) was calculated using the Nernst equation with E ₀ at pH 7.0 taken as −240 mV. ²⁹  

Measurement of ROS Production

After incubation of hRPE cells in cysteine-free DMEM or in DMEM treated with 500 μM BSO, cells were washed in cold PBS and treated with 10 μM dichlorofluorescein diacetate (DCFDA) for 30 minutes. DCFDA is a nonfluorescent ester of the dye fluorescein that is cleaved by intracellular esterases and entrapped within the cell as the oxidant-sensitive DCF compound. ROS oxidize DCF to the fluorescent product fluorescein. ³⁰ The green fluorescence of fluorescein was determined by confocal fluorescence microscopy. Control hRPE were treated with H₂O₂ (1 mM) for 30 minutes as a positive control for increased ROS production. 

Measurement of Δψ_m

After incubation in cysteine-free DMEM medium for 0, 24, 48, or 72 hours, or for 72 hours with or without bongkrekic acid (100 μM; BA), Δψ_m was measured with the fluorescent lipophilic cationic dye tetra-methyl rhodamine methyl ester (TMRM; 250 nM) which accumulates within mitochondria in a manner that is dependent on the Δψ_m. ²⁰ hRPE cells were loaded with TMRM for 15 minutes after incubation in cysteine-free medium and red fluorescence was measured by flow cytometry (FACScan; BD Biosciences, San Diego, CA) in the FL-2 mode. In each analysis, 10,000 events were recorded. 

Assessment of hRPE Cell Viability

hRPE cell viability was monitored by trypan blue exclusion after incubation of hRPE in cysteine-free DMEM for 0, 24, 48, and 72 hours, with or without BA (50 μM) and by PI staining as previously described. ²⁰ Briefly, after incubation of hRPE cells in cysteine-free medium for 0, 24, 48, and 72 hours, cells were fixed in ethanol, stained with PI and analyzed by flow cytometry using the FL-3 channel to determine the percentage of apoptotic cells (cells with a sub-G₁ content of DNA). Trypan blue-stained hRPE cells were observed by light microscope. 

TEM Examination of hRPE Cells

hRPE cells, in logarithmic proliferation, were incubated in cysteine-free DMEM for 0 and 72 hours. Cells were fixed with 2.5% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4) at room temperature for 1 hour. The cells were washed with 0.1 M cacodylate buffer and postfixed with 1% osmium tetroxide in 0.1 M cacodylate buffer. Finally, the cells were dehydrated in a graded series of ethanol, and embedded (LX112; Millipore, Bedford, MA). Thin sections were prepared and stained with uranyl acetate. Specimens were examined on by electron microscope operating at 80 kV (1000×; JEOL, Tokyo, Japan). 

hRPE Cell Cytochrome bc ₁ Activity

hRPE cell cytochrome bc ₁ activity was determined before and after cells were incubated for 24 hours in cysteine-free DMEM, with or without stigmatellin (1 μM) or in DMEM treated with 500 μM BSO, with or without stigmatellin (1 μM). Cytochrome bc ₁ activity measurements were made according to the method of Trounce et al., ³¹ with modifications. Briefly, hRPE cells were permeabilized with PBS-digitonin (∼ 0.01%), using a volume of PBS-digitonin that caused >95% hRPE trypan blue positivity. hRPE cell cytosolic fractions were removed and the membrane fractions washed two times in cold PBS. Membranes were sonicated and stored at −80°C until time of enzyme assay. Cytochrome bc ₁ complex activity was assayed in 50 mM potassium phosphate (pH 7.0), 250 mM sucrose, 0.2 mM EDTA, 1 mM NaN₃, 0.1% (wt/vol) and 0.01% Tween-20 at 23°C, using 50 μM 2,3-dimethoxy-5-methyl-decyl-1,4-benzoquinol (decylubiquinol [dUb]as substrate and 50 μM cytochrome c. dUb was synthesized in the laboratory from decylubiquinone by reduction with sodium borohydride (NaBH₄). ³¹ Reduction of cytochrome c was monitored in a spectrophotometer at 550 versus 539 nm in dual-wavelength mode. Data are expressed as a percentage of control activity and were determined from five individual isolations that were assayed in triplicate. 

Statistical Analysis

Statistical analyses were performed using Student’s t-test for unpaired data, and P < 0.05 were considered significant. Data are presented as the mean ± SEM. 

Results

Depletion of GSH and Increase in E _h

Figure 1A shows that after incubation in cysteine-free DMEM for 24 hours or in DMEM containing 500 μM, BSO reduced the hRPE GSH level by approximately 95% compared with control cells. The E _h in control hRPE cells was −248 ± 2.9 mV compared with −179 ± 2.0 mV after incubation in cysteine-free medium for 24 hours and −182 ± 2.3 mV after incubation in 500 μM BSO for 24 hours (Table 1) . No significant difference in either GSH levels or E _h was found between hRPE cells incubated in cysteine-free DMEM or in DMEM containing 500 μM BSO (Fig. 1A ; Table 1 ). 

Increase in ROS Production

Figure 1B shows an increase in hRPE cell ROS production (determined by an increase in DCF fluorescence measured by confocal microscopy) after incubation for 24 hours in DMEM treated with 500 μM BSO (Fig. 1Bd) and in hRPE incubated in cysteine-free medium (Fig. 1Bb) compared with control (Fig. 1Ba) . Control hRPE cells treated with H₂O₂ (1 mM) for 30 minutes were used as a fluorescence comparison for increased ROS production (Fig. 1Bf) . 

Mitochondrial Respiratory Site Responsible for Increased ROS Production

To determine the hRPE cellular site that regulates ROS production, cells were coincubated with BSO, BSO+rotenone (1 μM), or BSO+stigmatellin (1 μM), or in cysteine-free medium, cysteine-free media+rotenone (1 μM), or cysteine-free media+stigmatellin (1 μM). Rotenone selectively blocks respiratory complex I, ³² stigmatellin blocks the Qo site of cytochrome bc ₁. ³³ Rotenone did not prevent an increase in ROS causing significant cell death (data not shown), whereas, stigmatellin prevented the increased DCF fluorescence (ROS production) in hRPE cells after incubation of cells in cysteine-free DMEM for 24 hours (Fig. 1Bc) and after incubation of hRPE cells in DMEM and 500 μM BSO for 24 hours (Fig. 1Be) . To confirm the role of cytochrome bc ₁ in mitochondrial ROS production, hRPE cell membrane fractions were processed from cells incubated for 24 hours in cysteine-free DMEM or in DMEM treated with 500 μM BSO and cytochrome bc ₁ activity measurements were made. Figure 1C shows that (1) GSH depletion, induced by incubating cells in cysteine-free DMEM or in DMEM treated with 500 μM BSO, did not significantly alter cytochrome bc ₁ enzyme activity and (2) cytochrome bc ₁ activity was reduced ∼95% in hRPE cells incubated in cysteine-free DMEM with or without 1 μM stigmatellin or in hRPE cells incubated in DMEM treated with 500 μM BSO with or without 1 μM stigmatellin. Figure 1D shows that stigmatellin also prevented loss of hRPE cell viability induced by GSH depletion due to incubating hRPE cells in cysteine-free DMEM for 24 hours. Coincubation of hRPE with the cytochrome bc ₁ inhibitor stigmatellin (1 μM) with or without cysteine-free DMEM or stigmatellin (1 μM) with or without DMEM treated with 500 μM BSO for periods longer than 24 hours (i.e., 48–96 hours) showed variable results but in general did not significantly protect hRPE cells against loss of cell viability (data not shown). 

Loss of Δψ_m

The Δψ_m in hRPE cells was determined by monitoring the fluorescence of TMRM with flow cytometry (Fig. 2) as previously described. ^{22 23} Figure shows representative examples of the Δψ_m in hRPE cells after incubation in cysteine-free medium for 0, 24, 48, and 72 hours and for 72 hours, with or without BA (Fig. 2A) . The percentage of hRPE cells with decreased TMRM fluorescence increased as a function of time, after incubation of cells in cysteine-free DMEM (0–70 hours), indicating that this treatment caused loss of Δψ_m (Fig. 2B) . The loss of TMRM fluorescence induced in hRPE by incubation in cysteine-free DMEM was completely blocked by cotreatment of cells with BA (50–100 μM) suggesting that cysteine starvation activates the MPT in hRPE cells (Figs. 2A 2B) . To confirm the activation of the MPT in hRPE cells as a consequence of cysteine starvation, TEM was performed before and after cells were incubated in cysteine-free DMEM to determine hRPE mitochondrial ultrastructure. In Figure 2C , representative electron micrographs show mitochondrial ultrastructure in control hRPE cells (Fig. 2Ca) , and cells treated with cysteine-free medium for 72 hours (Fig. 2Cb) . The images suggest that cysteine starvation induces activation of the MPT, since the mitochondria appeared swollen and the internal cristae structure lacked the organization found in mitochondria of untreated hRPE cells (magnification, ×7250). 

Loss of Cell Viability

Next, we investigated whether the loss of hRPE cell viability due to cysteine starvation resulted from the induction of apoptosis or necrosis. We found that, although cysteine starvation increased the number of hypodiploid hRPE cells (apoptotic fraction) in a time-dependent fashion (Figs. 3A 3B) , this fraction was always significantly less than the fraction of cells that were trypan blue positive (Fig. 3C) . Because cysteine starvation caused cell death that was completely blocked with BA (Fig. 3C) but was not prevented with either the broad-spectrum caspase inhibitor zVADfmk or the caspase 3 inhibitor ac-DEVD-CHO (Fig. 3C) , this indicated that the loss of cell viability induced by activation of the MPT in hRPE cells was predominantly necrotic in nature. 

Discussion

In this study, the consequences of altered cysteine metabolism in hRPE cells were investigated by using a strategy of cysteine starvation. This provided us with a model of cellular mitochondrial oxidative stress resulting from the loss of GSH that may be useful in modeling the oxidative stress thought to occur in vivo in the aged RPE. ³ Cysteine starvation resulted in the loss of ∼95% GSH in hRPE cells over a 24-hour period. The intracellular E _h significantly increased from −248 ± 2.9 mV in control hRPE cells to −179 ± 2.0 mV in cells grown in cysteine-free DMEM for 24 hours. This E _h was not significantly different from that observed after inhibition of γGCL (the rate controlling enzyme for GSH) with BSO, a pharmacologic inhibitor of γGCL (−182 ± 2.3 mV; Table 1 ). This result indicates that cysteine starvation resulted in hRPE cell oxidative stress of a magnitude similar to that observed with complete inhibition of GSH synthesis by using BSO, which has been shown to result in cell death by apoptosis. ²⁰ Therefore, this result suggested to us that cysteine availability was not only a rate-limiting factor for GSH synthesis in hRPE cells but was also crucial for protection against oxidative stress, as previously suggested. ^{20 21 22 23} The level of intracellular oxidation observed in hRPE cells, as a consequence of either cysteine starvation or inhibition of γGCL, was determined by monitoring levels of DCF fluorescence, a method that has been used to monitor intracellular ROS production. ^{20 21 22 23} Our results indicate that a similar level of oxidative stress occurred subsequent to either cysteine starvation or inhibition of γGCL, since relative levels of DCF fluorescence after incubation of hRPE cells with either cysteine-free DMEM or DMEM+BSO were of a magnitude similar to those in control hRPE cells treated for 30 minutes with 1 mM H₂O₂ (Fig. 1Bf) . These results are in agreement with previous reports showing that depletion of GSH results in increased cellular ROS production. ^{20 21 22 23}  

To identify the intracellular site of ROS production in hRPE cells, we adopted a pharmacologic strategy previously used in HL60 cells using mitochondrial site-selective inhibitors. ^{22 23} Inhibition of the Qo site of respiratory complex III (cytochrome bc ₁) with stigmatellin blocked ROS production in cysteine-starved hRPE cells (Fig. 1B) . The respiratory complex I inhibitor rotenone (1 μM) did not block ROS production in cysteine-starved or γGCL-inhibited hRPE cells (data not shown). This indicates that mitochondrial respiratory complex I is unlikely to be the predominant site of ROS generation in cysteine-starved hRPE cells, but cytochrome bc ₁ is a key ROS-producing site in these cells. Similar results have been obtained with the HL60 myeloid leukemic cell line ^{22 23} and in the leukemic lymphoid CEM cell line (Armstrong JS, Whiteman M, unpublished observations, 2004). The role of a functional cytochrome bc ₁ in ROS production in hRPE cells was further supported by the observation that cytochrome bc ₁ was fully active in cysteine-starved hRPE cells compared with control cells (Fig. 1C) , whereas it was completely inhibited in stigmatellin-treated, cysteine-starved hRPE cells (Fig. 1C) . This indicates that cytochrome bc ₁ activity is essential for ROS production after cysteine starvation, as in other cell systems. ^{22 23} In agreement with our results, Sun and Trumpower ³³ recently found that cytochrome bc ₁ complexes from bovine heart and Saccharomyces cerevisiae mitochondria generated significant levels of ROS which were blocked with stigmatellin. Taken together, these results indicate that (1) cytochrome bc ₁ is a critical mitochondrial respiratory site involved in ROS production in cysteine-starved hRPE cells and (2) cytochrome bc ₁ activity (i.e., electron flow through the complex) is crucial for ROS production. We found that stigmatellin preserved hRPE cell viability after cysteine starvation, as reported for other cells (Fig. 1D) . ^{22 23}  

It is known that mitochondrial GSH is a critical factor regulating the redox status of protein thiols that regulate MPT, ^{22 23 34 35 36} and, because cysteine starvation significantly reduced GSH levels, we determined the effects of incubating hRPE cells in cysteine-free DMEM on Δψ_m and on mitochondrial ultrastructure, by using electron microscopy, as previously reported. ²³ We found that cysteine starvation caused a significant loss in Δψ_m, determined by a reduction in TMRM fluorescence, but that this was significant only after approximately 48 hours of treatment, indicating that loss of Δψ_m occurred after increased ROS production (Figs. 2A 2B) . Loss of Δψ_m in hRPE cells was therefore taken to be the direct consequence of cysteine-dependent intracellular oxidation. The loss of Δψ_m was blocked by incubation of cysteine-starved hRPE cells with BA, an inhibitor of the ANT. Because the ANT may be involved in MPT regulation, this result indicated that the observed decrease in Δψ_m was a consequence of activation of the MPT. ^{37 38 39} However, both the loss of Δψ_m and cell death induced by cysteine starvation was not blocked with the classic inhibitor of the MPT cyclosporine A (1–10 μM), suggesting that this MPT may be an unregulated form of MPT (data not shown). ^{40 41} TEM was used to determine mitochondrial ultrastructure in cysteine-starved hRPE cells which showed that the classic hallmarks of the MPT, including mitochondrial swelling and the presence of electron opaque mitochondria with a loss of cristae organization (Fig. 2C) . ^{23 42 43}  

The consequences of the MPT induced in hRPE cells by cysteine starvation were determined by monitoring cell viability with trypan blue dye exclusion and by determining the induction of apoptosis by measuring the percentage of hypodiploid cells (i.e., cells with increased sub-G₁ DNA). We found that cysteine starvation increased the number of hRPE cells with increased sub-G₁ fraction (Figs. 3A 3B) , but this fraction was significantly lower than the percentage of nonviable (trypan blue positive cells; Fig. 3C ) indicating that the apoptotic program was activated in only a small number of hRPE cells subsequent to cysteine starvation. Also, although the loss of hRPE cell viability induced by cysteine starvation was effectively blocked by cotreatment of hRPE cells with BA, it was not inhibited with the broad-spectrum caspase inhibitor zVADfmk or the caspase 3–specific inhibitor ac-DEVD-CHO (Fig. 3C) . This indicated that cysteine starvation induced ROS-dependent MPT and caspase-independent hRPE cell death. 

In conclusion, cysteine starvation causes depletion of GSH in hRPE cells resulting in oxidative stress mediated by cytochrome bc ₁ which activates an unregulated form of the MPT and causes caspase-independent cell death. These results show that the amino acid cysteine is crucial for normal GSH metabolism and for protection against mitochondrial oxidative stress in hRPE cells, as has been shown in other mammalian cell types. ^{20 21 22} Similar results in lower eukaryotes, such as the yeast S. cerevisiae and prokaryotes, including Escherichia coli, have led to similar conclusions and the almost universal concept that GSH is crucial for protection against oxidative stress. ^{10 44 45} The present data indicate a link between cysteine availability and normal GSH homeostasis in hRPE cells and suggest that cysteine depletion, as often occurs in the elderly undernourished individual, may be a contributing factor in the progression of ARMD. This hypothesis could be further tested in a suitable animal model. For example, Vaziri et al., ⁴⁶ administered buthionine sulfoximine (BSO, 30 mM in drinking water) to rats for 2 weeks to deplete GSH. It is plausible that similar models could be used to determine whether our in vitro findings translate to the in vivo situation, in which one could administer either BSO or feed a cysteine-free diet to animals and monitor ocular indices of oxidative stress including genomic and mitochondrial DNA–based modifications and protein carbonylation. 

A caveat is that because ARMD is a chronic disease and our experiments merely show the acute effects of cysteine depletion, our results should be considered with caution, because they may not completely reflect the chronic situation, where cell-protective mechanisms, including the antioxidant response, might be expected to prevent ROS-mediated toxicity. To summarize, the results of this study may have implications for many age-related diseases with a known oxidative stress component and may aid in the design of safe and effective clinical antioxidant strategies aimed at maintaining or enhancing GSH status. 

 Supported by National Eye Institute Grants EY07892 and EY06360; National Institute of Environmental Health Sciences Grant ES09047; NUS Office of Life Science Grants ARF R-183-000-103-112, R-183-000-606-101 (JSA), and R183000603712 (MW); National Medical Research Council Grants NMRC/0474/2000, NMRC/0481/2000, and NMRC/0635/2002 (MW); the Foundation Fighting Blindness; and an unrestricted grant from Research to Prevent Blindness, Inc.

 Submitted for publication May 21, 2004; revised July 22, 2004; accepted July 23, 2004.

 Disclosure: J.S. Armstrong, None; M. Whiteman, None; H. Yang, None; D.P. Jones, None; P. Sternberg, Jr, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

(A) GSH levels and E _h values in hRPE cells after incubation in cysteine-free DMEM or DMEM+BSO for 24 hours. Aliquots of hRPE cells (approximately 4 × 10⁶/mL) were incubated in cysteine-free DMEM, DMEM+BSO (500 μM), or DMEM alone (control) for 24 hours. GSH levels were determined on these samples by HPLC assay. **Significant difference (P < 0.01). (B) ROS levels in hRPE cells after incubation for 24 hours in cysteine-free DMEM or DMEM+BSO (with or without 1 μM stigmatellin). Cells were incubated in DMEM (Ba), cysteine-free DMEM (Bb), cysteine-free DMEM+1 μM stigmatellin (Bc), DMEM+500 μM BSO (Bd), DMEM+500 μM BSO+1 μM stigmatellin (Be) for 24 hours and loaded with 10 μM dichlorofluorescein diacetate (DCFDA) for 30 minutes. The green fluorescence of fluorescein was determined by confocal fluorescence microscopy. For a positive control of increased ROS production, hRPE cells were treated with H₂O₂ (1 mM) for 30 minutes and loaded with 10 μM DCFDA for 30 minutes (Bf). Figure shows representative example of at least 3 experiments. (C) Cytochrome bc ₁ activity in hRPE cells after incubation in cysteine-free DMEM or DMEM+BSO (with or without 1 μM stigmatellin) for 24 hours. Cytochrome bc ₁ activity was determined in mitochondrial fractions from hRPE cells that were incubated for 24 hours in DMEM (control), cysteine-free DMEM, or DMEM+500 μM BSO, in the presence or absence of 1 μM stigmatellin. Cytochrome bc ₁ activity measurements were then made. Results are expressed as a percentage of control cytochrome bc ₁ activity ± SEM and were determined using three individual mitochondrial isolations that were each assayed in triplicate. **Significant difference (P < 0.01). (D) Viability of hRPE cells after incubation for 24 hours in cysteine-free DMEM, with or without 1 μM stigmatellin. hRPE cells were incubated in DMEM (control), cysteine-free DMEM, or cysteine-free DMEM+1 μM stigmatellin for 24 hours. Cell viability was determined by trypan blue analysis. Data are expressed as mean ± SEM (n = 3).

Table 1.

 

GSH Nernst Redox Values in hRPE Cells

	GSH (μM ± SEM)	GSSG (μM ± SEM)	E h ± SEM
Control hRPE	8558 ± 248	260 ± 102	−248 ± 2.9
hRPE + BSO (500 μM) for 24 hours	287 ± 20	41 ± 1.2	−182 ± 2.3
hRPE + cysteine-free medium for 24 hours	259 ± 32	45 ± 1.8	−179 ± 2.0

 Aliquots of hRPE cells (approximately 4 × 10⁶/mL) were incubated in cysteine-free DMEM, DMEM+BSO (500 μM), or DMEM alone (control) for 24 hours. GSH and GSSG concentrations were determined by HPLC assay. The redox potential (E _h) was calculated using the Nernst equation with E ₀ at pH 7.0 taken as −240 mV. ⁴⁰ n = 3.

Figure 2.

 

(A) Δψ_m of hRPE cells after incubation in cysteine-free DMEM in the presence and absence of 100 μM BA. hRPE cells (∼ 2 × 10⁶/mL) were incubated with either DMEM (control) or cysteine-free DMEM for 0, 24, 48, and 72 hours, and for 72 hours with 100 μM BA. Data are results of a representative flow cytometric analysis of hRPE cells treated as described, loaded with 250 nM TMRM for 15 minutes and analyzed using the FL2-H channel. The experiment was repeated at least three times on different days. (B) Δψ_m of hRPE cells after incubation in cysteine-free DMEM in the presence or absence of 100 μM BA. hRPE cells (∼2 × 10⁶/mL) were incubated with either DMEM (control) or cysteine-free DMEM for 0, 24, 48, and 72 hours and for 72 hours with 100 μM BA. Data are the percentage of cells with intact Δψ_m determined by flow cytometry and are expressed as the mean ± SEM (n = 3). *Significant difference (P < 0.05). (C) TEM showing mitochondrial ultrastructure in hRPE cells before and after incubation in cysteine-free DMEM for 72 hours. Representative transmission electron micrographs of hRPE mitochondria showing the effects of cysteine starvation on mitochondrial ultrastructure. hRPE mitochondria were incubated for 72 hours in DMEM (control; Ca) or cysteine-free DMEM (Cb). Magnification, ×7250.

Figure 3.

 

(A) Induction of apoptosis in hRPE cells by cysteine starvation. Representative cell cycle analysis of hRPE cells stained with PI and analyzed using the FL2-A channel. hRPE cells were incubated in cysteine-free DMEM for 0, 24, 48, and 72 hours and analyzed by flow cytometry (FL-2A channel). In each analysis, 10,000 events were recorded. (B) Induction of apoptosis in hRPE cells by cysteine starvation. Data represent results of a mean cell cycle analysis of hRPE cells stained with PI and analyzed by flow cytometry (FL2-A channel). Data are the percentage of cells with sub-G₁ DNA content and are expressed as the mean ± SEM (n = 3). *Significant difference (P < 0.05). (C) Viability of hRPE cells after incubation in cysteine-free DMEM, with or without caspase inhibitors and BA. hRPE cells were incubated with cysteine-free DMEM for 0, 24, 48, and 72 hours in the presence or absence of zVADfmk (50 μM), DEVD-CHO (50 μM), or BA (100 μM). Cell viability was determined by trypan blue analysis. Data are expressed as the mean ± SEM (n = 3).