Demonstration by Redox Fluorometry that Sulforaphane Protects Retinal Pigment Epithelial Cells against Oxidative Stress

Abstract

purpose. To quantify the effects of oxidant challenge on the redox state of adult human retinal pigment epithelial cells using microscopic autofluorescence spectroscopy and to determine whether treatment with the isothiocyanate sulforaphane protects these cells against oxidative stress.

methods. Oxidative stress was evoked in ARPE-19 cells by H₂O₂ and tert-butyl hydroperoxide. Reduced nicotinamide nucleotides NAD(P)H were assessed by excitation at 366 nm with measurement of fluorescence at 450 nm. Oxidized flavoproteins were assessed by excitation at 460 nm with measurement of fluorescence at 540 nm. The ratio of these measurements served as the index of cellular redox status.

results. Redox ratio and cell viability decreased in a dose-dependent manner after oxidant exposure. ARPE-19 cells treated with sulforaphane maintained significantly higher redox ratio and cell viability. The ratio for sulforaphane-treated cells after exposure to 0.64 mM H₂O₂ was 2.64 ± 0.19 compared with 1.77 ± 0.16 in untreated cells (P = 0.001). At 1.2 mM H₂O₂, the redox ratio of sulforaphane-treated cells was 2.30 ± 0.18 compared with 1.76 ± 0.13 in untreated cells (P = 0.02). Similar results were observed after insult with tert-butyl hydroperoxide.

conclusions. Redox fluorometry provides quantitative information on the redox status of living cells. Sulforaphane protects ARPE-19 cells from oxidative injury by induction of antioxidant phase 2 genes. The findings in this study describe a useful method for assessing antioxidant effects in live cells and support phase 2 gene induction as a potential treatment strategy for macular degeneration and diseases in which oxidative injury plays a causative role.

Living systems have evolved multiple elaborate mechanisms for defense against the damaging effect of reactive oxygen species (ROS), which are obligatory byproducts of aerobic metabolism. Generation of ROS at rates exceeding cellular capacity for detoxification leads to oxidative stress, a fundamental mechanism for disease development and progression. Oxidative stress and its accompanying inflammatory reactions have been implicated in a broad spectrum of chronic degenerative conditions including cancer, atherosclerosis, Alzheimer disease, cataract, age-related macular degeneration, and aging itself. ^{1 2 3 4 5 6 7 8}  

The retina is one of the most metabolically active tissues. It has high oxygen requirements, is exposed to high levels of light, and contains high concentrations of polyunsaturated fatty acids, all which contribute to oxidative stress. In addition, the phagocytosis and degradation of photoreceptor outer segments by retinal pigment epithelial (RPE) cells further contribute to oxidative stress and potential retinal damage over time. ⁹  

The ROS generated in retinal cells require detoxification. In addition to small exogenous direct antioxidant molecules such as ascorbic acid, tocopherols, and carotenoids, the retina—like other tissues—contains high levels of glutathione, which is the principal endogenous small molecule antioxidant. However, recent evidence suggests that major antioxidant protection of cells arises from an elaborate network of “phase 2” enzymes that are highly inducible by a variety of phytochemicals and synthetic agents. Upregulation of these phase 2 proteins provides broad, long-lasting protection of cells against the damaging effects of ROS, electrophiles, and inflammation. ^{10 11} Transcription of the protective phase 2 response is highly inducible by the isothiocyanate sulforaphane, which we have used in this study to protect against oxidative stress. Sulforaphane is of particular interest because it is a potent inducer derived from its glucosinolate precursor, which is abundant in cruciferous vegetables such as broccoli. Delivery of a potent upregulator of the protective phase 2 response by dietary means is feasible and would facilitate rapid and economical clinical translation. 

To assess the antioxidant effects of sulforaphane in live cells, we sought a method that could measure dynamic changes in the redox state that did not require fixation or cell damage. Fluorescence microscopy provides such real-time, noninvasive assessment by measuring changes in the ratios of both reduced to oxidized nicotinamide nucleotides and oxidized to reduced flavins. Intensity of fluorescence emission in the region of 450 nm after excitation at 366 nm is an index of the levels of reduced nicotinamide nucleotides (NADH plus NADPH) in the cytoplasm and mitochondria, with greater quantum yield from the mitochondrial bound species. Fluorescence in the region of 540 nm after excitation at 460 nm is an index of the cellular levels of oxidized flavins that occur mostly as cofactors for flavoproteins, such as lipoamide dehydrogenase and electron transfer flavoprotein, which are involved in redox reactions. These two signals respond reciprocally to changes in the cellular redox state. The ratios of these fluorescence values are subject to minimal interference from absorption of excitation and emission light by other chromophores, light scattering, and variation in mitochondrial density. These ratios have been previously proposed as a noninvasive means to assess cellular metabolic redox status but have not been evaluated in RPE or other retinal cells. ^{12 13 14 15 16 17}  

In this study we exposed human RPE cells to two oxidizing agents, hydrogen peroxide and tert-butyl hydroperoxide, and showed by use of fluorescence spectroscopy that previous treatment of cells with sulforaphane significantly inhibited oxidative stress. Redox spectroscopy is therefore an effective screening method for identifying antioxidant protection in RPE cells. The biological basis of the redox signal provides pathophysiological insight into the effects of various oxidants and compounds that protect from oxidant injury. A potential role for sulforaphane as antioxidant protection against the development of chronic degenerative conditions involving the RPE, such as age-related macular degeneration (AMD), is supported. 

Materials and Methods

Cultured ARPE-19 Cells

Cells were grown in 10% CO₂ at 37°C in DMEM/F-12 media (Gibco, Grand Island, NY) supplemented with 10% FBS, 1 g sodium bicarbonate per 500 mL of media, and 0.4 mL glutamine. The medium was replaced every 2 to 3 days. 

On reaching confluence (at approximately 3 days), the medium was removed, and the cells were rinsed with PBS and were trypsinized with 1× trypsin-EDTA (Gibco) for 1 minute. The presence of cells released from the culture plate was confirmed by microscopic visualization. Two milliliters of culture medium were added to the 1 mL trypsin-EDTA solution to neutralize the trypsin. The solution was then centrifuged for 5 minutes at 2000 rpm, and the cells were resuspended in fresh medium, counted, and either replated for future passage or used for fluorescence imaging. 

Cell Oxidation

Cells were plated on glass surface culture plates (MatTek Cultureware, Ashland, MA) and allowed to attach for 24 hours before the study. Sulforaphane-treated cells were plated in medium containing 4 μM sulforaphane during the attachment period. After 24 hours, the medium was replaced with serum-free medium containing 0.6 mM, 1.2 mM, and 2.4 mM hydrogen peroxide (Sigma, St. Louis, MO) or 4.7, 6.5, 8.5, or 9.4 mM tert-butyl hydroperoxide for 2 hours. The cells were then imaged. 

Autofluorescence Microscopy

All images were obtained using a Zeiss (Thornwood, NY) inverted microscope (Axiovert 200M) with a 100× objective (FLUAR 100×, 1.3 oil). The microscope was equipped with a mercury lamp (HB 103) and a cooled charge-coupled device camera (Axiocam MRc5) for acquiring images. To detect intrinsic reduced nicotinamide nucleotides, a Zeiss DAPI filter set (excitation, G365; emission, bandpass 445/50) was used. Oxidized flavoproteins were identified using a Zeiss FITC filter set (excitation, bandpass 450–490; emission, bandpass 515–565). To minimize photograph bleaching and light stimulation, the transmitted visible illumination source was turned off during fluorescence imaging. All the images and fluorescence ratios were processed and analyzed using Zeiss software (AxioVision). Before autofluorescence microscopy, 5000 cells were plated and expanded on glass surface culture dishes (MatTek). Additionally, cell cultures were equilibrated in PBS solution pH.7.4 before imaging. 

Redox Ratio Calculation

After outlining the cell border, the intensity of cellular fluorescence and background fluorescence was autocalculated (Axiovision 4.3; Zeiss) in a standardized fashion. Ten randomly selected cells in each condition were analyzed. The net value of fluorescence from DAPI signal divided by the net value of fluorescence from the FITC signal. Each 10-cell assay was conducted in triplicate. 

Mitochondrial Staining

Cells cultured on glass bottom plates were stained (MitoTracker Green FM; Molecular Probes, Eugene, OR) 24 hours after seeding in accordance with the manufacturer’s instructions. In brief, a 30-nM probe solution (MitoTracker; Molecular Probes) was prepared in complete culture media and incubated with the cells for 30 minutes at 37°C and 10% CO₂. The samples were then rinsed twice with PBS, and image capture was performed with the Zeiss FITC filter set (excitation, bandpass 450–490; emission, bandpass 515–565). 

Cell Viability Assay

A cell proliferation assay (CellTiter 96 AQ_ueous; Promega, Madison, WI) was used in accordance with the manufacturer’s instructions to assess the viability of the cells. This assay is a colorimetric method for determining the number of viable cells in proliferation. The assay is based on the cellular conversion of a tetrazolium salt (MTS) into a formazan product. The resultant absorbance is directly proportional to the number of living cells in culture. In brief, 5000 cells were cultured in 96-well plates for 24 hours, as previously described. Then 25 μL reagent (CellTiter 96 AQ_ueous One Solution; Promega) was added directly to the culture wells and incubated for 1 to 4 hours, and absorbance was measured at 490 nm with a spectrophotometric plate reader. Cell viability was expressed as a percentage, with 100% representing the signal from untreated cells and 0% representing the background signal from empty wells. 

Determination of Nicotinamide Nucleotide Levels

Levels of reduced nicotinamide nucleotides were determined spectrophotometrically by an enzymatic recycling method, ¹⁸ as described by Rao and Bhat. ¹⁹ Five thousand cells were used for the determination of reduced nucleotides, and 5000 cells were used for the oxidized nucleotides. For the extraction of NADPH and NADH, cells were lysed with 0.2 M KOH for 1 minute in a boiling water bath, chilled, neutralized with an equal volume of 0.23 M KH₂PO₄, and centrifuged at 0°C to 4°C for 30 minutes at 20,000g. Supernatants were used immediately for assay. The order of treatment with KOH and KH₂PO₄ was reversed for NADP⁺ and NAD⁺ extraction. 

Statistical Analysis

Statistical analysis was performed with a paired t-test assuming unequal variances. P < 0.05 was prospectively determined to be statistically significant. 

Results

Using a standardized approach to image acquisition for each channel NAD(P)H (channel 1, pseudocolor green) and flavoprotein (channel 2, pseudocolor red) allows for overlaying images without further processing. The difference between untreated cells and cells exposed to 0.6 mM H₂O₂, 1.2 mM H₂O₂, and 2.4 mM H₂O₂ for 2 hours (Fig. 1)is apparent in photographs and is readily measured by expressing the fluorescence in each channel as a “redox ratio” (channel 1/channel 2). Therefore, photographic shifts toward red coloration result in a declining redox ratio indicating a greater amount of oxidation. Prevention of these shifts under experimental conditions represents an antioxidant or a protective effect. 

Treatment of RPE cells with sulforaphane (4 μM) for 24 hours before H₂O₂ exposure at the same concentrations(0, 0.6, and 1.2 mM) provides photographic evidence of protective effect (less shift from green to red; Figs. 1 2 ). Treatment with sulforaphane also protects the cells from oxidative stress generated by a more physiologic oxidant tert-butyl hydroperoxide (Fig. 3) . This is supported by the measured fluorescence and calculated redox ratio for each group. Redox ratios were calculated and plotted for 0.6 mM, 1.2 mM, 2.4 mM, and 4.7 mM H₂O₂ and showed a dose-dependent decline (Fig. 4) . Redox fluorometry allowed discrimination of redox shifts at the concentrations of H₂O₂ used. Comparing the redox ratio in untreated controls (1.27) to that in cells treated with 2.4 mM H₂O₂ (0.24), the difference was significant (P = 8.36 × 10⁻¹⁰; Fig. 4 ). Cells pretreated with 4 μM sulforaphane for 24 hours had smaller redox shifts at all concentrations of H₂O₂ examined (Fig. 5) . As a control, we tested the effect of sulforaphane alone, and it did not significantly shift redox ratios when used to treat RPE cells (Fig. 5) . We did not evaluate resistance to oxidative injury at H₂O₂ concentrations higher than those shown in Figure 6because 4.7 mM H₂O₂ resulted in a redox value near zero (Fig. 4) , and nearly all cells were dead. The results presented here demonstrate that treatment with sulforaphane protects RPE cells from oxidation (Fig. 5)and preserves viability during oxidative stress (Fig. 6) . 

Pretreatment of RPE cells with sulforaphane protects against oxidizing injury from various agents. Protection from injury induced by tert-butyl hydroperoxide (Fig. 7)is also present. At lower concentrations, the less reactive tert-butyl hydroperoxide induced smaller redox shifts. The significant protective effect of sulforaphane was evident at tert-butyl hydroperoxide concentrations of 8.5 mM (P = 0.03) and at 9.4 mM (P = 0.00002). 

As has been previously reported, ¹⁶ the NADH/NAD(P)H measure by autofluorescence is primarily of mitochondrial origin. The status of the mitochondrial NAD system reflects the rate of oxidative phosphorylation. The oxidation of cells results in a decrease in the NAD(P)H/flavoprotein ratio and is consistent with our data. Figure 8represents further evidence that the autofluorescence of NAD(P)H correlated with the mitochondrial fluorescence when this was stained with probe (MitoTracker; Molecular Probes). 

Treatment with sulforaphane alters the concentration of reduced glutathione as well as NAD(P)H, and quinone reductase 1 increases are present in the cytosol. ¹ To determine whether redox fluorescence in sulforaphane-treated cells was significantly altered by the production of intermediates with competing fluorescence, we evaluated cells that had been treated with sulforaphane for only 30 minutes, when potentially fluorescent intermediates are formed but no protection is expected. The lack of fluorescence change during this period indicates that production of early intermediates did not significantly alter RPE cell redox ratios and that treatment of cells with sulforaphane for a period of 30 minutes did not sufficiently alter redox ratios in response to oxidant challenge (Fig. 9) . When the levels of NADH and NADPH were measured under the experimental conditions described here, they did correlate strongly with those measured by redox fluorometry (correlation; R = 0.92). 

Discussion

Redox fluorometry is now an established technique in a variety of cell types, but it has not yet been evaluated in RPE or other retinal cells. ^{12 14 17 20 21} In this study we demonstrate the usefulness of redox fluorometry for the quantitative assessment of oxidant injury in RPE cells. The principal attributes of redox fluorometry are that it is noninvasive, can be performed in live cells, provides data in real time, and permits serial measurements on single cells. Other available methods measure redox status in dead cells and require staining techniques ^{22 23} or chemical assays ²⁴ that measure byproducts of redox reactions. 

To our knowledge, this is the only available and generally accessible technique that allows real-time and serial assessment in living cells. This study indicates that redox fluorometry is sufficiently sensitive and reproducible to allow measurement of RPE cell redox state changes, induced by small changes in applied concentrations of hydrogen peroxide or tert-butyl hydroperoxide. By evaluating RPE cells of early and late passages with corresponding differences in melanin content, we have determined that melanin autofluorescence did not contribute significantly to the fluorescent signal measured in this study. We have also shown that sulforaphane intermediates did not contribute significantly to the redox signal. This novel ocular application of redox fluorometry may be useful for the screening of antioxidant compounds with potential therapeutic benefit for diseases in which oxidant injury contributes to RPE pathology. 

In RPE cells, the flavoproteins and nicotinamide nucleotides are near reduction-oxidation equilibrium, and the ratio of fluorescent intensities between these two molecules approximates the redox status of the cell. Previous work indicates that redox status is highly correlated with biochemical measures of NADH and NADPH and with fluorometric measurement of NADH/NADPH. ^{16 17 20} This study extends this finding by demonstrating a high correlation between fluorometric measures of nicotinamide nucleotides and chemical measures of nicotinamide nucleotides in RPE cells. It also localizes the fluorescence of NADH/NAD(P)H primarily to the RPE mitochondria, consistent with previous studies conducted in mesenchymal stem cells. ¹⁶ Localization of the measured signal to the mitochondria and the identification of an antioxidant protective effect based on measurement of a mitochondria-dominated redox signal has clear pathophysiological and therapeutic implications for diseases involving the RPE, such as AMD. 

Oxidative processes contribute to aging and to the progression of diseases such as neurodegeneration, some cancers, heart disease, and ocular diseases such as cataract and AMD. ^{6 8 9} The pathogenesis of AMD is multifactorial, including photo-oxidative and oxidative stress, toxic electrophiles from smoking and other sources of inflammation, immune factors, and genetic contribution. There is strong evidence that protection from phototoxic and oxidizing injury is important to the health of the RPE and the photoreceptors. ^{8 9 25 26 27 28 29 30} Consumption of direct-acting antioxidants to provide protection of the retina and RPE is supported by previous work, ^{31 32 33 34} including the AREDS clinical trial, which has added an antioxidant formulation to the routine care of patients with dry AMD. ⁴ The frequent use of high-dose, direct-acting antioxidants for the prevention of AMD is widely accepted ³¹ and leads to the slowing of AMD progression. A treatment strategy to induce long-acting, intracellular antioxidant protection through the endogenous phase 2 response is an attractive strategy for evaluation. A means to assess intracellular redox changes in the target cells involved in disease is integral to the discovery of potentially therapeutic approaches. Previous work predicts that the protective responses induced by sulforaphane may be a general phenomenon. ¹¹ It would therefore be predicted that photoreceptors and other retinal cells would be similarly protected. Redox fluorometry may be an ideal approach for investigating this question. 

Indirect antioxidants such as sulforaphane can provide potent and sustained protection against oxidant injury by inducing the broad phase 2 enzyme response in cells. ^{10 35} Upregulation of the phase 2 response protects RPE and retina from the damaging effects of photo-oxidant ROS and electrophiles. ^{10 11} In this study, sulforaphane protected RPE cells from chemical oxidant stress, as evidenced by the diminished intracellular redox shifts and the increased RPE cell viability. The protective effect conferred to RPE cells in this study is consistent with earlier work. Cytoplasmic and mitochondrial redox changes contribute to the fluorescent signal. That protection against redox shifts and increased RPE cell viability demonstrated by the mitochondria-dominated fluorescent signal is a new and intriguing finding with pathophysiological implications. 

Redox fluorometry is a potentially useful tool for the assessment of redox status in RPE cells. The measured signal reflects mitochondrial and cytoplasmic redox changes. It is not significantly affected by melanin autofluorescence. The technique is applicable to a variety of cell types and appears useful as a tool for the screening of potentially protective antioxidant compounds. In this study, quantitative assessment of redox shifts was shown to associate with significant protection against chemical oxidant injury of RPE conferred by sulforaphane. 

Methods are available for induction of the phase 2 response by administration of a dietary phytochemical, such as sulforaphane and its precursor, glucoraphanin. The findings presented here further support the development of phase 2 induction strategies as a therapeutic approach for AMD and other diseases in which oxidative injury plays a causative role. Moreover, redox fluorometry may be useful in the preclinical development of current and future antioxidant compounds. 

 Supported by The JG Foundation (PLG); an unrestricted grant from Research to Prevent Blindness (Wilmer Eye Institute) and a Research to Prevent Blindness Career Development Award (PLG); the Wilmer Gene Therapy Vector Core; William Weis Award (PLG); Jack and Gail Baylin Philanthropic Fund; The JHU Fund for Medical Discovery (PLG); and a gift from Kenneth and Brenda Richardson (PLG).

 Submitted for publication July 27, 2007; revised December 19, 2007; accepted March 31, 2008.

 Disclosure: M. del V. Cano, None; J.M. Reyes, None; C.Y. Park, None; X. Gao, None; K. Mori, None; R.S. Chuck, None; P.L. Gehlbach, 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.

 

Two-dimensional redox fluorometric photomicrographs of ARPE-19 cells treated with increasing concentrations of H₂O₂. Two-dimensional redox fluorometry photomicrographs of ARPE-19 cells. Top: control cell (no H₂O₂). Left column: cells are excited in the region of 366 nm and emission detected in the region of 450 nm (channel 1, green). Middle column: cells are excited at 460 nm and detected at 540 nm (channel 2, red). Right column: unprocessed overlaid images of the previous image sets. We observed an increase in the pseudocolor switch as the cells were exposed to increasing concentrations of H₂O₂.

Figure 2.

 

Two-dimensional redox fluorometric photomicrographs of ARPE-19 treated with increasing concentrations of H₂O₂ after exposure to sulforaphane for 24 hours. Two-dimensional redox fluorometric photomicrographs of ARPE-19 cells, treated with sulforaphane for 24 hours. Top: control cell (no H₂O₂). Left column: cells are excited in the region of 366 nm and emission detected in the region of 450 nm (channel 1, green). Middle column: cells excited at 460 nm and detected at 540 nm (channel 2, red). Right column: unprocessed overlaid images of the previous image sets.

Figure 3.

 

Two-dimensional redox fluorometric photomicrographs of ARPE-19 cells treated with increasing concentrations of Tert-butyl hydroperoxide after exposure to sulforaphane for 24 hours. Two-dimensional redox fluorometric photomicrographs of ARPE-19 cells, treated with sulforaphane for 24 hours. Top: control cell (no Tert-butyl hydroperoxide). Left column: cells are excited in the region of 366 nm, and emission is detected in the region of 450 nm (channel 1, green). Middle column: cells are excited at 460 nm and detected at 540 nm (channel 2, red). Right column: unprocessed overlaid images of the previous image sets.

Figure 4.

 

Redox ratio (reduced nicotinamide nucleotides/flavoproteins) of ARPE-19 cells treated for 2 hours with increasing concentrations of H₂O₂. The redox ratio decreases as the H₂O₂ concentration is increased. Each value is the mean ± SD of 30 measurements.

Figure 5.

 

Redox ratio (reduced nicotinamide nucleotides/flavoproteins) of ARPE-19 cells treated for 2 hours with increasing concentrations of H₂O₂ after exposure to sulforaphane for 24 hours. (□) Cells were treated with sulforaphane for 24 hours before exposure to hydrogen peroxide. (♦) Control cells were treated for 2 hours with increasing concentrations of H₂O₂ (0.6 mM, 1.2 mM, and 2.4 mM). Each value is the mean ± SD of 30 measurements.

Figure 6.

 

Viability of ARPE-19 cells treated for 2 hours with a range of increasing concentrations of H₂O₂ after exposure to sulforaphane for 24 hours. (□) Cells treated with sulforaphane for 24 hours before exposure to hydrogen peroxide. (▪) Control cells. Cells were treated for 2 hours with increasing concentrations of H₂O₂ (0.6 mM, 1.2 mM, 2.4 mM). The percentage of viable cells was measured.

Figure 7.

 

Redox ratio (reduced nicotinamide nucleotides/flavoproteins) of ARPE-19 cells treated for 2 hours with increasing concentrations of tert-butyl hydroperoxide after exposure to sulforaphane for 24 hours. (□) Cells treated with sulforaphane for 24 hours before exposure to tert-butyl hydroperoxide. (♦) Control cells. Cells were treated for 2 hours with increasing concentrations of tert-butyl-hydroperoxide (4.7 mM, 6.5 mM, 8.5 mM, 9.4 mM). Each value is the mean ± SD of 30 measurements.

Figure 8.

 

Two-dimensional photomicrographs of ARPE-19 cells marked with probe. Two-dimensional redox fluorometric photomicrographs of ARPE-19 cells, stained with probe. Left column: cells are stained with probe. Middle column: cells are excited in the region of 366 nm, and emission is detected in the region of 450 nm (channel 1, red). Right column: unprocessed overlaid images of the previous image sets.

Figure 9.

 

Redox ratio (reduced nicotinamide nucleotides/flavoproteins) of ARPE-19 cells treated for 2 hours with increasing concentrations of H₂O₂ after exposure to sulforaphane for 30 minutes. (□) Cells treated with sulforaphane for 30 minutes before exposure to hydrogen peroxide. (♦) Control cells. Each value is the mean ± SD of 30 measurements.

The authors thank Paul Talalay (Department of Pharmacology, Johns Hopkins University School of Medicine) for his expert advice, and Jason Rosenzweig (Wilmer Gene Therapy Vector Core, Johns Hopkins University School of Medicine) and Tinghua Wu (Department of Ophthalmology, Johns Hopkins University School of Medicine) for their technical assistance.