Light Induces Programmed Cell Death by Activating Multiple Independent Proteases in a Cone Photoreceptor Cell Line

Abstract

purpose. Although the apoptotic death of photoreceptor cells in retinal degenerative disorders is well documented, the molecular mechanism is not understood. The objective of this study was to determine the molecular events leading to the death of photoreceptor cells.

methods. An assay was developed wherein 661W cells, a cone photoreceptor cell line, were stressed with light and percentage of surviving cells was determined. The degree of cell death was established using the MTT assay. Western blot analysis was used to confirm the activation of multiple proteases. Amounts of retinaldehydes were determined by extraction and HPLC.

results. 661W cells were more susceptible to light stress only in the presence of the chromophore 9-cis retinal for 4 hours. On exposure to light, 9-cis retinal was converted to all-trans retinal, which was found to be toxic to cells in the presence of light. However, all-trans retinol, which is the product of action by the enzyme retinol dehydrogenase on all-trans retinal, was not toxic. The sensitivity to light increased with serum deprivation. Light stress activated caspases, calpain 2, and cathepsin D independently and led to the demise of the cell. The mitochondria-dependent apoptotic pathway was also activated after the truncation of Bid, the pre-proapoptotic protein. Truncation of Bid led to the release of cytochrome c from the mitochondria and the activation of caspase 9.

conclusions. The activation of multiple proteases by light-induced stress is a relevant finding for studies conducted to investigate the use of pharmaceutical agents to retard or cure the loss of cone photoreceptors observed in age-related macular degeneration and other degenerative retinal diseases.

Apoptosis is the predominant form of cell death by which photoreceptors die in many hereditary retinal degenerations. ^{1 2 3} Several mutations that increase the susceptibility of photoreceptors to light-induced apoptosis have been identified in humans. ^{4 5} Also, that cone photoreceptors undergo apoptosis after life-long exposure to light in age-related macular degeneration (AMD) establishes light as a risk factor. ⁶ AMD is the leading cause of blindness in the elderly and is estimated to affect more than 8 million individuals in the United States alone. ⁷ The prevalence of AMD is expected to increase significantly in the population aged 85 years and over, an age group that is projected to increase by 107% by the year 2020. ⁸ Animal models, primarily in mice, have been established for many disorders of human blindness. Studies on the effects of exposure to light on retinal structure and function have been performed on rat ^{9 10 11 12} and mouse retinas. ^{13 14 15} However, biochemical studies of light-induced apoptosis in vivo are complicated by the fact that the retina is a complex tissue that consists of many cell types. Finally, it is difficult to establish the relevance of results of animal model studies to the disease in patients with AMD, because the retina in most models is devoid of a macula, an area that has a highly concentrated population of cone photoreceptors. 

There is a scarcity of in vitro models of retinal degeneration, especially those that closely resemble in vivo models. The 661W cell line was derived from a mouse retinal tumor and has been characterized as a cone-specific cell line that expresses cone pigments, transducin, and arrestin. ¹⁶ One important property of the 661W cells as an in vitro model system is its response to light, ¹⁷ which makes it valuable in the assessment of light-induced stress and apoptotic cell death in cone photoreceptors. 

Although the role of apoptosis in retinal degenerative disorders and light-induced damage has been well documented, ^{14 18 19} the molecular events that orchestrate this process are not well understood. This shortage of information is particularly true of cone photoreceptors, partly due to the small number of cones in the retinas of murine models. The chromophore 9-cis retinal is present in the retina, where it is converted to all-trans retinal (ATR) in the presence of light. ATR has been shown to cause photooxidative damage to certain retinal proteins ²⁰ ; however, its potential role in the mediation of apoptosis in photoreceptors is not well understood. 

In this study, the cone photoreceptor cell line 661W was used to assess the molecular events involved in light-induced cell death in the presence of 9-cis retinal. This cell line was found to be more susceptible to light-induced death in the presence of 9-cis retinal than in its absence, and its susceptibility was elevated when serum levels in the growth media were reduced. Light sensitivity in the presence of the chromophore resulted from the conversion of 9-cis retinal to ATR, which was toxic to cells in the presence of light. All-trans retinol (ATol), by contrast, was not toxic to cells in either the presence or absence of light. Light stress induced the activation of multiple caspases, calpain 2, and cathepsin D, resulting in apoptosis. The mitochondrial apoptotic pathway was also activated after the truncation of Bid, leading to both cytochrome c translocation and the activation of caspase-9, which are essential for the assembly and activation of the apoptosome. The activation of multiple proteases in light-stressed 661W cells is in agreement with the behavior of photoreceptor cells that are destined to die in certain retinal degenerative disorders. It was recently observed that the ablation of caspase-3 only delayed, rather than inhibited, the death process. ²¹ Moreover, 661W cells challenged with growth factor deprivation caused parallel activation of caspase and calpain pathways. ²²  

Materials and Methods

Immortalization, Culture Conditions, and Characterization of Cell Lines

The 661W cell line was derived from mouse retinal tumors and has been characterized previously and shown to be of cone photoreceptor cell lineage. ¹⁶ B4 cells were derived from 661W cells by permanent transfection of Bcl₂. ²³ RGC-5 is a rat ganglion cell line. ²⁴  

661W, B4, and RGC5 cells were grown in DMEM (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS; Cellgro, Herndon, VA) and a 1% antimycotic-antibiotic cocktail (Invitrogen). All cell cultures were maintained in a humidified atmosphere of 95% air and 5% CO₂ at 37°C. 

Growth of Cells for Light-Induced Apoptosis Experiments

Twenty-four thousand cells were added in each well of a 96-well tissue culture plate (catalog no. 353072; BD Biosciences, Franklin Lakes, NJ) and allowed to grow overnight. The next morning, the medium was replaced with a 100-μL aliquot of fresh medium, to remove excess cells and cellular debris. The medium replacement ensured that the degree of confluence of cells in each well was the same and that results would be more consistent. For experiments that involved light-induced apoptosis in the presence of 9-cis retinal, ATR, or ATol (all from Sigma-Aldrich, St. Louis, MO), the medium was replaced with 100 μL new media containing 10 μM 9-cis retinal, ATR, or ATol, and the cells were returned to the incubator for 4 hours and then exposed to light. 

Dark control cells and light-stressed 661W cells were all from the same stock, eliminating any preexisting bias (such as light or temperature) as a variable. Cell stocks were not exposed to direct light at any time during passaging and feeding, nor were they maintained in any light–dark cycles. Because of the long incubation in retinoids and the duration of exposure to light, experiments were always started at around 10 AM. 

Cells in the absence or presence of retinoids were exposed to either 15,000 lux (1.5 lumen/cm²/4.5 mW/cm²) or 30,000 lux (3 lumen/cm²/9 mW/cm²) of light for the indicated times. To insure that temperature was not a variable, measurements were made by inserting a thermometer in the media before, during, and at the end of the exposure to light. 

Light Source

The light box (Model MS1417-C-DIM-T83) was obtained from Aristo Lighting Technologies (Roslyn, NY). Three phosphors are used in manufacturing the light source in this model. They peak at 453 nm (red), 546 nm (blue), and 611 nm (green). UV light was blocked by a Plexiglas filter. To maintain temperature, the light box was fitted with two small fans. 

Viability Assay

To test the viability of cells, 10 μL 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich) from a 5-mg/mL stock was added to the media, and the cells were placed in an incubator overnight. Mitochondrial succinate dehydrogenase (E.C.1.3.9.9) from viable cells reduces MTT to form a dark blue formazan crystal. The next day, the media were removed, the formazan precipitate was dissolved in dimethyl sulfoxide (DMSO), and the absorbance was read at 570 nm. The viability of the cells was expressed either as absorbance at 570 nm or as percentage of cells surviving. Data were graphed with commercial software (Prism; GraphPad Software, Inc., San Diego, CA). Each experiment was performed at least three times, with 12 replicates for each treatment. Cells that underwent light treatment were compared with dark control cells. The percentage of viable cells was calculated by averaging the ratios of absorbance readings of cells in the light to the dark control cells, assuming that the dark control cells were 100% viable, and the average percentage was determined. Standard Student’s t-tests were performed by using absorbance readings (minimum n = 36) and statistically significant differences (P < 0.05) from dark control cells are denoted by a symbol in the figures. For 1 2 3 4 5 6 Figure 7C , statistical analysis was performed on the samples with Student’s t-tests and one-way ANOVA followed by Bonferroni post test. All the groups were compared to each other, and statistical significance was set at P ≤ 0.05. 

To determine the linearity of the MTT assay, an increasing amount of cells between 200 and 32,000 were used in a viability assay. The MTT assay was found to be linear for cell counts between 4,000 and 32,000 (data not shown). 

In experiments involving retinoids, their addition neither affected the absorbance readings of the MTT assay nor induced changes in the temperature of the media. Furthermore, control samples of media and the additives (no cells) were always prepared, and readings were subtracted from those containing cells. 

Uptake of Retinoids in 661W Cells

The optimal amount of 9-cis retinal used in these experiments was determined empirically by incubating cells in various amounts of the chromophore. Concentrations between 6 and 13 μM were observed to lack cytotoxic effects in the dark. Further analysis was performed to determine the concentration that resulted in both a lack of cytotoxicity in the dark and induction of considerable cell death in the presence of light. A concentration of 10 μM was found to be optimal for both requirements (Supplementary Fig. S1). 

For studies involving the uptake of 9-cis retinal and isomerization to ATR in 661W cells, 5.6 × 10⁶ cells were grown overnight in T-75 tissue culture flasks (BD Biosciences) in DMEM. The following day, the medium was replaced with fresh medium containing 10 μM 9-cis retinal, and the cells were allowed to incubate for an additional 4 hours in the dark. Afterward, the cells were washed three times, medium devoid of 9-cis retinal was added, and the cells were exposed to 30,000-lux light for 30 minutes. Retinoids were extracted by a modification of a previously described method. ^{25 26} Briefly, 661W cells were collected with a scraper followed by centrifugation at 500g for 5 minutes. The cell pellet was resuspended in 300 μL PBS and sonicated twice for 20 seconds each. Cold methanol (300 μL at 4°C) and hydroxylamine (60 μL, 1 M) in 0.2 M sodium phosphate buffer (pH 7.4) were added, and samples were mixed for 30 seconds and then allowed to stand for 5 minutes at room temperature. Subsequently, 300 μL methylene chloride was added, and the samples were mixed again for 30 seconds and centrifuged (16,000g for 5 minutes). The lower phase was collected and dried under argon. Samples were dissolved in the HPLC mobile phase (11.2% ethyl acetate, 2.0% dioxane, 1.4% octanol in hexane, 180 μL) and retinoids were separated on a 5-μm column (Lichrosphere SI-60; Alltech Associates, Deerfield, IL). The retinal peaks were identified by comparison with pure retinoid isomeric standards. All procedures were performed under dim red light (Filter GBX-2; Eastman Kodak, Rochester, NY). 

Growth of 661W Cells under Low FBS Concentrations

Cells were initially grown in DMEM supplemented with 10% FBS and a 1% antimycotic-antibiotic cocktail. Approximately 24,000 cells were plated in 96-well plates and allowed to grow overnight. The following day, the medium was replaced with medium supplemented with either 1% or no FBS, 10 μM 9-cis retinal was added, and the cells were allowed to incubate for an additional 4 hours in the dark. Cells were then exposed to either 15,000- or 30,000-lux light for the indicated times, and viability analyses were performed by MTT assay, as described earlier. 

Studies with Protease Inhibitors

Cells were plated at a concentration of approximately 12,000 cells per well in 96-well tissue culture plates and allowed to adhere overnight. On day 2, the medium was replaced with medium containing inhibitors, and the cells were incubated overnight. On day 3, the medium containing inhibitors was removed and replaced with one containing inhibitors and 9-cis retinal, and the cells were allowed to incubate in the dark (in the incubator) for 4 hours. One plate was placed on a light box that emitted 30,000-lux white fluorescent light for 4 hours and one plate was placed for 4 hours in the dark, after which viability tests were performed. The pancaspase inhibitor, benzyloxycarbonyl-Val-Ala-Asp fluoromethyl ketone (zVAD-fmk; R&D Systems, Minneapolis, MN), was used at concentrations ranging between 10 and 100 μM, and only data obtained with 10 μM are presented. The cathepsin D inhibitor, pepstatin A (Sigma-Aldrich), was used at concentrations between 10 and 50 μM, and only data obtained with 25 μM are presented. MDL 28170, the calpain 2 inhibitor (Calbiochem, La Jolla, CA), was used at concentrations between 1 and 4 μM, and only results obtained with 2 μM are presented. 

Caspase Activity Assays

Caspase activity assays were performed with a caspase family colorimetric substrate kit II (Biovision, Mountain View, CA). Briefly, cells were plated in T-75 tissue culture flasks, 10 μM 9-cis retinal was added to the medium, and the cells were allowed to incubate in the dark at 37°C for 4 hours. Because all previous experiments were performed in 96-well plates, exposure times in T-75 flasks that produced equivalent results were determined. It was found that exposure to light for 4 hours in T-75 flasks produced cell death equivalent to that produced by 3 hours of exposure in 96-well plates. This exposure time was used for all experiments in T-75 flasks. After the cells were incubated in medium containing 9-cis retinal, the plates were either exposed to 30,000-lux light or kept in the dark for 3 hours. Cells were harvested and subjected to a freeze-thaw cycle in lysis buffer (10 mM Tris [pH 7.5], 100 mM NaCl, 1 mM EDTA, and 0.01% Triton X-100). Samples were sonicated for 10 seconds and centrifuged at 5000 rpm for 5 minutes. Assay buffer (10 mM PIPES [piperazine-N-N′-bis(2-ethanesulfonic acid], pH 7.4; 2 mM EDTA; and 0.1% CHAPS [ 3-[3-cholamidopropyl ]dimethylammonio-2-hydroxy-1-propanesulfonate ]) was added to 70 μg of protein, to a final volume of 100 μL. Finally, 5 μL of paranitroanilide (pNA)-conjugated substrate was added, to a final concentration of 200 μM, and allowed to incubate overnight at 37°C. Absorbance was measured at 405 nm with a plate reader (Thermomax; Molecular Devices Corp., Sunnyvale, CA). 

Western Blot Analysis

661W cells were grown to 80% confluence in T-75 flasks, 9-cis retinal was added, and the plates were exposed to either dark or light for the indicated times indicated in the figures. The cells were then washed with Hanks’ balanced salt solution (Invitrogen), and lysates were prepared (NE-PER Extraction kit; Pierce, Rockford, IL). The mitochondria-enriched fraction was prepared according to a previously described method. ²⁷  

Protein quantification was performed using the Bradford protein assay kit (Pierce). Antibodies used were caspase-8 rabbit polyclonal antibody (1:200; BD PharMingen, San Jose, CA), cleaved caspase-3 rabbit polyclonal antibody (1:500; Cell Signaling Technology, Beverly, MA), caspase-9 rabbit polyclonal antibody (1:1000), caspase-10 rabbit polyclonal antibody (1:1000; Cell Signaling Technology), Bid rabbit polyclonal antibody (1:200), PARP rabbit polyclonal antibody (1:500; Santa Cruz Biotechnology, Santa Cruz, CA), β-actin mouse monoclonal antibody (1:2000; Abcam, Cambridge, MA), cytochrome c sheep whole antiserum (1:1000; Sigma-Aldrich), cathepsin D goat polyclonal antibody (1:200; Santa Cruz Biotechnology), and calpain-2 rabbit polyclonal antibody (1:5000; Calbiochem). 

Results

Light Sensitivity of 661W Cells

To analyze their response to light stress, 661W cells were subjected to 30,000-lux light, and viability was assessed with the MTT assay, after 2, 3, 4, and 6 hours of light-induced stress. 

The 661W cell line showed no reduction in viability for up to 4 hours, but after 6 hours, only 68% of cells were viable (Fig. 1A , media). In comparison, 86% of B4 cells (661W cells overexpressing Bcl-2; Fig. 1B , media) and 98% of RGC-5 (rat retinal ganglion) cells (Fig. 1C , media) were still viable after 6 hours of light stress. 

Examination of cell morphology after exposure to light showed that the cells were rounding up and detaching from the plastic surface, a behavior typical of cells undergoing apoptosis (Supplementary Fig. S2). 

To determine the role of 9-cis retinal in light-induced death, the cells were incubated in the presence of the chromophore and subjected to light-induced stress, and viability was again assessed at 2, 3, 4, and 6 hours. Because 9-cis retinal was dissolved in DMSO, control experiments were performed in which the cells were placed in media containing DMSO before exposure to light. Although cells in media alone or in media containing DMSO did not show any response to light-induced stress until after 6 hours of exposure (Fig. 1A) , the cells in the presence of 9-cis retinal showed an average of a 26% reduction in viability as early as 2 hours after exposure to light. More than 50% of the cells died after 3 hours, 75% died after 4 hours, and almost all cells were dead after 6 hours of exposure (Fig. 1A , media/9 cis). 

Although B4 cells also responded to light stress in the presence of the chromophore, the overexpression of Bcl-2 seemed to afford the cells with a protective effect since only 5% and 18% of the cells died after 3 and 4 hours of light stress, respectively (Fig. 1B , media/9 cis). RGC-5 cells showed no response to light-induced stress after exposure for up to 4 hours (Fig. 1C) . However, after 6 hours of exposure to light, only 40% of RGC-5 cells survived (Fig. 1C , media/9 cis). Exposure to 6 hours of light in the presence of the chromophore induced cell death in all cell types, albeit to various degrees, with 661W cells being the most susceptible (Figs. 1A 1B 1C , media/9 cis). 

661W Cell Uptake of 9-cis Retinal from Media

To determine the extent of cellular uptake of the chromophore, 661W cells were lysed and analyzed for retinoids by HPLC. As shown in Figure 2A , the 661W cells did not contain any endogenous chromophore. However, they were capable of taking up exogenous 9-cis retinal from the medium. After 4 hours of incubation in medium containing 10 μM 9-cis retinal followed by its removal, these cells contained a considerable amount of 9-cis retinal (Fig. 2B) . 

On exposure to light, 9-cis retinal is converted to ATR. To verify that the 9-cis retinal to ATR photoconversion takes place in this cell line, 661W cells were incubated for 4 hours in medium containing 9-cis retinal, the medium was removed, the cells were washed with Hanks’ salt solution, fresh medium lacking the chromophore was added, and the cells were exposed to light for 30 minutes. Analysis of the lysed cells by HPLC showed that both 9-cis retinal and ATR were present in the cell lysates, indicating that photoisomerization had occurred (Fig. 2C) . Also, a small amount of ATol was formed from the reduction of ATR by the enzyme retinol dehydrogenase. Integration of the area under the HPLC curve revealed that these cells took up and retained 432 picomoles of 9-cis retinal (Fig. 2B) . The cells also contained approximately 45 picomoles of ATR. After exposure to light, the amount of ATR increased to 371 picomoles, whereas the amount of 9-cis retinal decreased to 110 picomoles (Fig. 2C) . Therefore, within 30 minutes of exposure to light, 75% of the chromophore was converted from 9-cis retinal to ATR. 

ATR and Light-Induced Cell Death

Because light causes photoisomerization of 9-cis to ATR, light-induced cytotoxicity of retinoids was assessed in the presence of ATR. Cells were incubated with 10 μM ATR for 4 hours at 37°C, after which some cells were exposed to light for either 4 or 6 hours. Controls were kept in the dark for 6 hours. All cell types were viable after 6 hours in the dark in the presence of ATR (Fig. 3) . However, exposure of cells to 4 hours of light in the presence of ATR was sufficient to cause death in all different cell types, albeit to various degrees. Although almost all RGC-5 cells survived (Fig. 3C , media/ATR), less than 20% of 661W cells endured (Fig. 3A , media/ATR). Again, Bcl-2 provided some protection: After 4 hours of exposure to light, 77% of B4 cells were viable (Fig. 3B , media/ATR). These results indicate that ATR is cytotoxic only in the presence of light and that 661W cells are more susceptible to ATR cytotoxicity than are the other cell lines studied. 

Toxicity of ATol to 661W Cells

On isomerization of 11-cis retinal to ATR, photoreceptors rapidly reduce ATR to ATol by means of retinol dehydrogenases. To determine whether the cytotoxicity of the retinoids results from the formation of ATol, cells were exposed to light in the presence or absence of ATol. Figure 4shows that all cell lines, irrespective of exposure to dark or light, were viable after 4 hours of exposure to ATol. Cell death was observed after cells were exposed to light for 6 hours; however, death occurred to the same extent in either the presence or absence of ATol (Fig. 4) , indicating that ATol was not responsible for the cytotoxicity. 

Data presented in both Figures 1 and 3show that RGCs died after 6 hours of exposure to light, suggesting that a secondary pathway that is different from the one that caused death in 661W in 4 hours, may be responsible. Because our study focused on cell death in cone photoreceptors that results primarily from pigment photoisomerization, exposure to light in subsequent experiments was limited to 4 hours. 

Sensitivity to Light Stress in Reduced-Serum Conditions

Fetal bovine serum (FBS), a major component of cell culture medium, contains growth factors and albumin. Growth factors are needed for survival, and albumin binds retinoids. Both factors may play a critical role in modulating the light intensity required to cause stress. Consequently, cells were grown under reduced-serum conditions and their response to light-induced stress was recorded. When FBS levels were reduced from 10% to 1% (Fig. 5A)or to 0% (Fig. 5B) , cells underwent death faster in response to 30,000-lux light, especially in the presence of 9-cis retinal. Although 74% and 45% of cells grown in 10% serum and exposed to light for 2 and 3 hours, respectively, survived (Fig. 1A) , reducing the serum to 1% allowed only 31% and 2% of the cells to survive under a similar light intensity (Fig. 5A) . In addition, complete elimination of serum from the medium made the cells even more susceptible to light-induced stress (Fig. 5B) . Furthermore, lowering FBS levels made the 661W cells more susceptible to stress at lower light intensities. Whereas 100% of cells in 10% FBS survived 3 hours of exposure to 15,000 lux (data not shown), only 46% and 29% of cells tolerated exposure to 15,000 lux at 1% FBS (Fig. 5C)and 0% FBS (Fig. 5D) , respectively. The 661W cells grown under reduced FBS conditions were sensitive to 15,000-lux light in the presence of 9-cis retinal after a 1-hour exposure (Figs. 5C 5D) . Whereas 46% of cells survived exposure to 15,000 lux for 3 hours, only 2% of the cells survived when the light intensity was increased to 30,000 lux (∼ 96% reduction). Furthermore, eliminating the serum from the medium reduced the survivability of the cells by only ∼37%. This reflects a direct correlation between amount of light and the ensuing cell death. 

Activation of Multiple Caspases in Response to Light-Induced Stress

Although the cells exhibited morphologic changes indicative of apoptotic cell death in response to light-induced stress in the presence of 9-cis retinal (Supplementary Fig. S2), the activation of caspases was also investigated. As shown in Table 1 , most caspases investigated were activated after 3 hours of exposure to light. Some activation was observed as early as 1 hour after exposure to light (data not shown), but maximum activation was observed after a 3-hour exposure. Except for caspase-3, all other caspases tested showed moderate activation (Table 1) . However, caspase-3 showed more than 1800% activation, suggesting that it is the primary caspase responsible for the execution of light-induced cell death in 661W cells. Data in Table 1also suggest that multiple initiator caspases activate caspase-3. These data also provide an indication of the degree of amplification of the death signal downstream from these initiator caspases. For example, if activation of caspase-8 was the upstream event that led to activation of caspase-3, there would be a greater than fivefold signal amplification. 

Cleavage of Bid and Activation of the Mitochondrial Death Pathway

Activation of caspases was also observed by Western blot analyses of caspase-10, -8, and -3 (Fig. 6) . With an antibody to the uncleaved (inactive) form of caspase-10, it was clear that there was a reduction in the amount of procaspase-10 after 2 hours of light-induced stress. This result suggests that the activation of caspase-10 is an early event in the commitment to cell death. Procaspase-8, a 55-kDa protein (Fig. 6B , gray arrowhead), was cleaved to an active fragment of 40 kDa (Fig. 6B , black arrowhead) as early as 1 hour after stress. At 3 and 4 hours after stress, levels of this cleaved fragment increased (Fig. 6B) . Because all cells were still viable 1 hour after stress, these data suggest that cleavage of procaspase-10 and -8 were the initial triggers for light-induced cell death in the presence of 9-cis retinal. With an antibody to cleaved caspase-3, ample amounts of activated caspase-3 were detected by Western blot analysis (Fig. 6C) , in agreement with data obtained from the enzymatic assays and presented in Table 1 . 

In addition to its activation by initiator caspases, caspase-3 can be activated by the mitochondria-dependent apoptotic pathway via the cleavage of Bid, a proapoptotic protein. Truncated (t)Bid translocates to the mitochondria, allowing the release of multiple components of the apoptosome, such as cytochrome c, and resulting in the activation of caspase-9. As seen in Table 1 , the activity of caspase-9 increased by greater than 150% in response to light. This result is supported by the appearance of two fragments of activated caspase-9 on Western blot analysis as seen in Figure 6D(black arrows). To determine whether tBid was both produced and responsible for the activation of the mitochondrial apoptotic machinery, Western blot analysis was performed with an antibody to full-length Bid. As shown in Figure 6E , levels of Bid were reduced with increased durations of exposure to intense light, whereby after 4 hours of exposure, full-length Bid was barely detectable. Release of cytochrome c from the mitochondria was assessed by analyzing mitochondria-enriched protein preparations by Western blot analysis. As shown in Figure 6F , 3 hours of exposure to light were sufficient to deplete the mitochondrial fraction of cytochrome c almost entirely. Taken together, our data suggest that caspase-9 activation is mediated through the truncation of Bid, its translocation to the mitochondria, and the subsequent release of mitochondrial cytochrome c. Our data also suggest that the mitochondrial death pathway is secondary to the activation of caspase-8 and -10. 

Cleavage of Survival Proteins

To investigate whether specific survival proteins are targets for caspase proteolytic activity in light-induced apoptosis, two potential substrates, poly-(ADP-ribose) polymerase (PARP) and nuclear factor-κB (NF-κB), were investigated. After 2 hours of exposure to light, full-length PARP, which is 113 kDa (gray arrow, Fig. 6G ), was cleaved to produce two fragments: 86 kDa (black arrow, Fig. 6G ) and 25 kDa (not shown in Fig. 6G ). By 4 hours, almost all the intact PARP was cleaved (Fig. 6G) . 

The transcription factor NF-κB regulates gene expression during immune and inflammatory responses and has been shown to protect from apoptosis. It was shown that apoptotic cells undergo caspase-mediated cleavage of the NF-κB p65 subunit leading to the loss of the C-terminal transactivation domains. ^{28 29} The truncated p65 promoted apoptosis via a dominant negative inhibition of NF-κB. ^{28 29} In light-stressed 661W cells, NF-κB was also gradually cleaved from a 65-kDa fragment (Fig. 6H , gray arrowhead) to a 50-kDa fragment (Fig. 6H , black arrowhead). However, by 4 hours of light stress, both fragments were degraded, and the total amount of NF-κB was drastically reduced, as shown by the intensities of these bands compared with that of actin (Fig. 6H) . The introduction of Bcl-2 into 661W cells afforded the cells a protective mechanism, since, after 4 hours of light-induced stress, the B4 cells still exhibited moderate levels of NF-κB (Fig. 6H)and more than 80% of the cells were still alive (Fig. 1B) . Of note is that untreated B4 cells endogenously exhibited much higher amounts of NF-κB than did untreated 661W cells (Fig. 6H , compare lane 4D 661W to lane 4D B4). To determine whether stabilization of NF-κB in B4 cells was a result of its association with Bcl-2, immunoprecipitation experiments were performed using anti- Bcl-2 and NF-κB antibodies. However, no association was found between the two proteins (data not shown). Thus, our data suggest that Bcl-2 plays an indirect role in the stabilization of NF-κB and is worthy of investigation as a potential survival mechanism for photoreceptors in retinal degenerative disorders. 

Activation of Multiple Death-Inducing Proteases

Growth factor deprivation of 661W cells leads to apoptotic cell death with activation of both caspases and calpains. ³⁰ To determine whether calpains are activated after light-induced stress in 661W cells, Western blot analysis was performed using anti-calpain 2 antibodies that recognize the intact form of the protein. As shown in Figure 7A , approximately 50% of calpain 2 was activated after 1 hour of light stress and no further activation was observed with continued exposure to light (Fig. 7A) . 

Activation of cathepsin D and apoptotic cell death were observed in the rd/rd mouse, which undergoes rapid rod photoreceptor degeneration at an early age ³¹ and in 661W cells after treatment with sodium nitroprusside. ^{31 32} To determine whether cathepsin D plays a role in light-induced cell death, Western blot analysis was performed using anti-cathepsin D antibodies to both the intact and cleaved forms of the protein. Cathepsin D was activated as early as 1 hour after stress. The level of the active form was maintained for up to 4 hours of stress, whereas the intact form decreased with increased exposure time (Fig. 7B) . 

Because multiple proteases were activated, it was possible that they were activated either sequentially or simultaneously. To differentiate between these possibilities, various protease inhibitors were used to modulate their specific corresponding pathways. First, a broad-range caspase inhibitor (z-VAD-fmk) was used to modulate caspase-dependent cell death. No significant protection at increasing concentrations of the inhibitor, added both before and during light-induced stress, was observed (data for 10 mM are shown in Fig. 7C ), raising the possibility that caspases were still activated in the presence of the inhibitor. However, Western blot analysis demonstrated that the addition of the pan-caspase inhibitor zVAD-fmk reduced the amount of active caspase-3 by more than eightfold (Fig. 7D) , suggesting that, although caspases were activated in the system, they were not the sole executioners of the apoptotic pathway in these cone-specific cells. 

Because calpain 2 was also activated on light stress, the next logical step was to inhibit calpain 2 and determine whether significant protection was provided. The calpain inhibitor III (MDL 28170) provided significant protection at a concentration of 2 μM. Although only 24% of the cells were viable after 4 hours of light stress in the absence of this calpain inhibitor, 47% of cells were viable in the presence of 2 μM MDL 28170 (Fig. 7C) . These data suggest that although calpain 2 plays a role in cell death, its activation (similar to caspases) does not bear full responsibility in the execution of the process. To determine whether calpain 2 participates in the caspase activation pathway, Western blot analysis was performed for activated caspase-3 in both the absence and presence of calpain inhibitor. However, the addition of MDL did not prevent or decrease the extent of caspase-3 cleavage, suggesting that calpain 2 was not responsible for the activation of caspase-3 but rather constituted a distinct apoptotic pathway. Furthermore, the addition of zVAD-fmk to MDL-treated cells did not increase the degree of survival, providing further indication that calpain did not act through caspases but rather through its own separate apoptotic pathway (Fig. 7C) . 

Because cathepsin D was also activated after light stress (Fig. 7B) , we reasoned that incubation with the cathepsin D inhibitor pepstatin A at 25 μM may provide significant protection against 9-cis retinal mediated apoptosis. Whereas 24% of the cells were viable in the presence of 9-cis retinal after 4 hours of exposure to light, 36% of the cells were viable when 25 μM of pepstatin A was added (Fig. 7C) . The addition of zVAD-fmk did not increase the degree of survival of the cells, suggesting that the role of cathepsin D in cell death was not via the activation of a downstream caspase such as caspase-3. 

Taken together, our data suggest that caspase-3 activation in light-induced apoptosis in the presence of 9-cis retinal is an event that is independent of calpain 2 and cathepsin D activation. To test this hypothesis, all three inhibitors were used in a light-induced stress experiment. As shown in Figure 7C , the addition of all three inhibitors afforded 73% protection from light stress, the highest observed, suggesting that the multiple proteases were acting simultaneously through separate pathways. 

Discussion

This study describes an assay in which light-induced stress activated multiple proteases and apoptosis in an established 661W cone photoreceptor cell line in the presence of the chromophore 9-cis retinal. This chromophore was used instead of 11-cis retinal because it is a more stable isomer and is also present in the normal retina, albeit at much lower concentrations. Its presence in the retina is independent of RPE65, an enzyme required for both the synthesis and regeneration of 11-cis retinal. Like 11-cis retinal, 9-cis retinal can bind opsin to generate a photoactivatable molecule. Parry et al. ³³ succeeded in reconstituting visual pigments in intact retinas of goldfish by delivering 9-cis retinal that was incorporated into phospholipid vesicles. Finally, low levels of 9-cis retinal were shown to mediate rod function in RPE65^−/− mice. ³⁴ Like 11-cis retinal, 9-cis retinal is taken up by 661W cells and converted to ATR after exposure to light. 

In vitro studies have shown that ATR can mediate photoxidative damage to proteins present in the retina such as rod outer segment membrane protein (Rom-1), retinal degeneration slow protein (also known as peripherin II; Rds) and ABCA-4 when rod outer segments preparations are incubated with ATR and exposed to light. ²⁰ Moreover, ATR has been shown to play a role in light-mediated damage by generating singlet oxygen molecules ³⁵ and by oxidizing proteins ³⁶ lipids ³⁷ and DNA. ³⁸ However, the actual mechanism through which ATR causes retinal damage is not well understood. In the normal retina, ATR is removed from the photoreceptor discs by ABCA-4. ³⁰ However, ATP-binding cassette transporter (ABCA-4, ABCR) mutations have been found in a subpopulation of patients with certain retinal diseases such as RP, ³⁹ rod–cone dystrophy, ⁴⁰ AMD, ⁴¹ and Stargardt’s disease. ⁴² In these patients, there is delayed removal and thereby increased stores of ATR, which condenses with phosphatidylethanolamine (PE) in the photoreceptor outer segment disc membrane to form N-retinylidene-phosphatidylethanolamine (NRPE). ⁴³ In ABCA-4-knockout mice, there is delayed clearance of ATR/NRPE from the retina ⁴³ and these animals accumulate a threefold excess of NRPE compared with wild-type animals. Both the accumulation of NRPE and the condensation of NRPE with an additional molecule of ATR to form 2-(2,6-dimethyl-8-(2,6,6-trimethyl-l-cyclohexen-1-yl)-1E,3E,4E,7E-octatetraenyl)-1-(2-hydroxylethyl)-4-(4-methyl-6-(2,6,6-trimethyl-1-cyclohexen-1-yl)-1E,3E,5E-hexatrienyl)-pyridinium (A2E) were also found to be increased by 6.22- to 11.5-fold in the retinas of patients with certain forms of retinal dystrophies. ⁴⁴ A2E irradiation with light results in the generation of reactive oxygen species (ROS), which can also cause oxidative damage to lipids and proteins and thus inhibit several antioxidant and lysosomal enzymes. 

In the current investigation, it was shown that 9-cis retinal and ATR are toxic to 661W cells only in the presence of light while ATol is not cytotoxic. Because 9-cis retinal is photoconverted to ATR, it is safe to conclude that the observed cytotoxicity is a result of the presence of ATR. ATR shares the same ring structure with ATol and only differs in the nature of the side-chain functional group making it quite puzzling as to why Atol is not toxic. One possibility is that the toxic nature of ATR results from the fact that its aldehyde side group is capable of Schiff base formation with proteins, which under light conditions can result in protein oxidative damage. Because ATR was only toxic on exposure to light, it is safe to assume that the toxic factors are either directly downstream of ATR such as NRPE or A2E or much further downstream such as ROS formation. In fact, chemical modification of the proteins indicative of oxidative stress in 661W cells was observed (data not shown). One such modification was the formation of the reactive aldehyde 4-hydroxy-2-nonenal (HNE), which forms adducts with proteins. ⁴⁵  

The light intensity used in the current experiments was 15,000 to 30,000 Lux. These values are well within the range of the amounts of light that cone photoreceptors receive daily since the eye receives approximately 50,000 lux when a person is standing under the autumn sun. Light intensities as high as 15,000 lux have been used to stress and damage rod photoreceptors in in vivo studies. Exposure of wild-type (C57BL/6) mice to 15,000 lux has resulted in photoreceptor death after 24 hours. ¹³ Under this chronic light condition, all the chromophore 11-cis retinal gets converted to ATR and therefore plays an important role in photoreceptor damage. However, cones are 25 to 100 times less sensitive to light than are rods and therefore require strong light for stress induction. ⁴⁶ Furthermore, the medium we used in this study contains FBS with albumin, which binds retinoids, ⁴⁷ and growth factors that allow the cells to survive under strong light. Because these factors may have influenced our experimental conditions, we also conducted studies under reduced serum conditions. When FBS was removed, the cytotoxic effect of ATR was even more prominent, because cells were much more susceptible to light-induced cell death. 

Bcl-2 expression afforded only short-term protection against light-induced ATR stress by delaying apoptosis. Apoptosis in 661W cells occurred via multiple pathways, including the activation of multiple caspases. Caspases-8 and -10 are apoptotic initiators that were activated within 2 hours of light-induced stress. Mitochondria-dependent apoptotic pathways were also activated as indicated by the activation of caspase-9. There was evidence of crosstalk between the extrinsic and intrinsic pathways through the cleavage of Bid, as has been previously shown. ⁴⁸ Activated caspase-2 was shown to be induced in light-stressed 661W cells (Table 1) , which can also stimulate the release of cytochrome c, independent of Bid cleavage, from the mitochondria. ⁴⁹ It is currently unknown how caspase-2 is activated and whether caspase-8 and -10 have direct roles in its activation. Therefore, it is possible that caspase-9 was activated by more than one mechanism. Although it has been suggested that caspase-4 is involved in ER stress-induced apoptosis, ⁵⁰ there is also evidence to suggest that caspase-4 is unimportant in ER stress and that apoptosis can occur independently of caspase-4. ⁵¹ Therefore, the functional significance of caspase-4 activation in 661W cells is also unclear. The same argument applies to caspase-2 and -6, since their degree of activation was not as high as those of caspase-8 and -10. Alternatively, all these caspases may have been activated by the apoptosome as a final event in cellular degradation. 

The activation of caspases leads to the degradation of specific targets such as PARP and NF-κB. Both of these targets were initially cleaved but their overall amounts (the sum of both cleaved and intact fragments) remained almost the same. However, longer exposure to light led to a decrease in their overall amounts. This result is not very surprising since multiple proteases are activated and such a reduction may be the result of overall protein degradation. 

The use of a broad-range caspase inhibitor did not afford the cells with protection against apoptosis, suggesting that although caspases are activated, they are either not the sole effectors of apoptosis or their activation is the result of the initiation of other upstream proteases. The first of these possibilities is supported by our findings that calpain 2 and cathepsin D are also activated in 661W cells after light-induced stress. However, our data did not indicate that either calpain 2 or cathepsin D initiated any caspase activation pathways. Our data also do not exclude the possibility that other calpains or cathepsins are activated. One of the limitations we faced is that certain protease inhibitors are themselves light sensitive, which would thus complicate the interpretation of data. An interesting finding that emerged from our analysis is that cell death execution continued to the same degree even in the absence of caspase-3 activation, suggesting that other proteases, likely calpains and cathepsins, are involved in the execution process. 

The combination of any two of the three types of protease inhibitors tested generally led to a protection level of ∼50%, suggesting the involvement of other proteases in independent, nonsequential pathways. This observation was confirmed by the use of all three inhibitors tested, leading to the >70% survival of light-stressed cells. Our inability to achieve 100% protection with all three inhibitors may reflect the involvement of still other proteases in the cell death pathway, or more interesting, a role for the inhibited proteases in cell survival. One surprising finding was that zVAD-fmk showed protection only in the presence of the other inhibitors. 

Multiple death execution pathways have also been observed in vivo models of photoreceptor cell death. Calpains and cathepsin D have been shown to be activated in the rd mouse model, ³¹ whereas retinal atrophy in the cathepsin D-deficient mouse was mediated by caspase-9 and -3 in the outer nuclear layer and by nitric oxide (NO) released from activated microglial cells in the inner nuclear layer. ⁵² In an independent study, three models of retinal degeneration namely, the rd mouse, the rds mouse, and a light-induced photoreceptor death model, were studied. Markers of apoptosis, autophagy, and complement activation were upregulated in all the three models, demonstrating that cell death of photoreceptors is a complicated process. ⁵³ However, it is important to remember that the mouse retina is composed mostly of rod photoreceptors, whereas the present study involved a cell culture system of cone photoreceptors. Nevertheless, these studies in conjunction with the current one demonstrate that cones, like rods, activate multiple death pathways. 

The use 661W cells to study photoreceptor death by serum deprivation and SNP treatment has also shown multiple pathways to be involved in cell death. Serum deprivation of 661W cells activates caspases ^{3 9 12} and m-calpain ²² and SNP treatment of 661W activated caspases ^{3 9 12} and cathepsin D. ^{31 22} These studies, in conjunction with those prior studies, demonstrate that 661W cells, like the in vivo system, activate multiple death pathways regardless of the death inducer. 

One can potentially argue that, in light-stressed 661W cells in the presence of the chromophore, caspases, calpain 2, and cathepsin D were activated independently in different retinal cell types that altogether constitute the 661W cell line. If true, these different cell types must also respond to light-Induced stress and activate death pathways. However, data this study or previously published argue against the assumption of a heterogeneous cell line. First, we have shown that the 661W cells do not express any rod-specific antigens ¹⁶ and lack the presence of any pigmentation or the expression of RPE65. ¹⁶ So it is safe to assume that 661W cells, if heterogeneous, do not contain any rods or RPE cells. Second, expression of green and blue opsins was detected in every cell (see representative fields in Fig. 4 of Ref. ¹⁶ ). Third, the 661W cell line was established from a retinal tumor resulting from the expression of SV40 T antigen (Tag) in rods and cones driven by the human IRBP promoter. Those tumors arise only in the photoreceptor cells and the rest of the cell types in the retina are not involved (see Fig. 4A in Ref. ⁵⁴ ). Fourth, even if some retinal cells contaminated the initial culture, they did not have the growth advantage that the Tag-expressing photoreceptor had and as a result will be completely outgrown, in few passages, by cells expressing Tag. The 661W cells continue to express Tag. ¹⁶ Fifth, we showed that a ganglion cell line does not respond to light-induced stress like 661W, which suggests that even if some RGCs exist within the population of 661W, they will not respond to light stress by initiating cell death (Figs. 1 3 4) . Sixth, the 661W cells did not express any of the markers for other retinal cell types, such as chx10 for bipolar cells, Bcl-2 for Müller cells, HPC1 for amacrine cells, calbindin for horizontal cells and Thy-1 for ganglion cells (data not shown). 

We have previously shown that light-stressed 661W cells in absence of a chromophore (Ref. ¹⁷ and Fig. 1 ) undergo apoptosis after depletion of NF-κB mediated through involvement of caspase-1. ¹⁷ However, when cells are light stressed in the presence of the chromophore, caspase-1 activity is absent. This finding suggests that the death signals are different in the two cases, and alternative death pathways are activated, adding more to the complexity of these detrimental cellular events. 

In summary, our study demonstrates the activation of multiple independent proteases after light-induced stress of the cone photoreceptor cell line 661W. These findings are relevant for studies conducted to investigate the use of pharmaceutical agents to delay, prevent, or reverse the loss of cone photoreceptors observed in AMD and other degenerative retinal diseases. The simultaneous activation of proteases suggests the presence of a potentially common signal upstream that triggers their activation, and the identification of such a signal would be imperative for future therapeutic intervention. 

 Supported in part by Grants EY14052 (MRA), EY012231, and EY015650 (JXM) from the National Eye Institute, COBRE grant RR017703 from the National Institutes of Health (NCRR), National Eye Institute Vision Core Grant EY012190 to the Department of Ophthalmology, the Knights Templar Eye Foundation, Inc. (YK), and a grant from the American Health Assistance Foundation-National Glaucoma Program (NA).

 Submitted for publication June 1, 2006; revised July 27 and September 1, 2006; accepted November 20, 2006.

 Disclosure: Y. Kanan, None; G. Moiseyev, None; N. Agarwal, None; J.-X. Ma, None; M.R. Al-Ubaidi, 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 7.

 

(A) Western blot analysis of calpain 2 activation in 661W cells after light stress. Percent activation was determined by dividing the densitometric readings of the calpain band by the corresponding actin band. Lane 4D: 4 hours of dark incubation; lanes 1L, 2L, 3L, and 4L: are 1, 2, 3, and 4 hours of exposure to light, respectively. (B) Western blot analysis of cathepsin D in 661W cells after light stress. A procathepsin D band was observed in samples incubated with 9-cis retinal in the dark for 4 hours (black arrow). Light caused the processing of the procathepsin D band to a mature form (gray arrow) within 1 hour of exposure to light (lane 1L). This processed band was also observed at 2, 3, and 4 hours after exposure to light, indicating cathepsin D activation in the system. Numbers below the actin blot only, to indicate lane numbers. (C) Protection from light stress with protease inhibitors, individually and in combination. The cathepsin D inhibitor pepstatin A (2 μM), the pancaspase inhibitor zVAD-fmk (10 μM), and/or the calpain inhibitor MDL 28170 (25 μM) were used to determine the roles of different proteases in the apoptotic cell death of light-stressed 661W cone photoreceptor cells. Viability was assessed by MTT assay by measuring absorbance of the blue formazan product at 570 nm. The graph was plotted by calculating the percentage of viable cells incubated with different concentrations of inhibitors and 10 μM 9-cis retinal after exposure to light versus the dark control, with dark control cells being 100% viable. A DMSO control (without 9-cis retinal) was included because 9-cis retinal was dissolved in DMSO. Statistical analysis using the Bonferroni multiple comparison was performed and significant probabilities are: A versus B-I, <0.001; B versus D, <0.01; B versus E–I, <0.001; C versus D, <0.05; C versus E–I, <0.001; D versus G, <0.01; D versus H and I, <0.001; E versus G, <0.05; E versus H, <0.01; E versus I, <0.001; F versus I, <0.001; G versus I, <0.001; and H versus I, <0.001. (D) Inhibitory effects of the pancaspase inhibitor zVAD-fmk and/or the calpain 2 inhibitor MDL 28170 on caspase-3 cleavage after 4 hours of exposure to light (4L). There was a reduction of cleaved caspase-3 in the presence of the pancaspase inhibitor compared with no inhibitor or MDL alone. Samples incubated with 9-cis retinal in the dark (4D) showed no caspase activation. The ratio was obtained by dividing the densitometric reading of the band for cleaved caspase-3 by that for actin. Actin controls are shown below each blot.

Figure 1.

 

Light-induced cell death in the presence and absence of 9-cis retinal. Cell death was assessed by MTT assay, and data were plotted by measuring the absorbance of the blue formazan product at 570 nm. Cells that underwent light treatment were compared with dark control cells and three independent experiments were performed with 12 replicates in each experiment. Data are presented as the average percent viability ± SEM for each time point. Standard student’s t-tests were performed; *statistically significant differences (P < 0.05) compared to dark control cells. (A) 661W cells were incubated in medium without 9-cis retinal (Media), with DMSO (Media/DMSO), or with DMSO and 9-cis retinal (Media/9 cis). DMSO served as the carrier for the chromophore. (B) Similar conditions as in (A) but applied to B4 cells (661W+Bcl-2). (C) RGCs were used as the control and were subjected to the same conditions as described in (A).

Figure 2.

 

Retinoid analysis of 661W cell lysates with HPLC. Retinoids were extracted from cells and analyzed by HPLC. (A) 661W cells incubated without retinoids. (B) 661W cells incubated with 9-cis retinal. (C) 661W cells incubated with 9-cis retinal and exposed to 30,000-lux light for 30 minutes.

Figure 3.

 

Light-induced apoptosis in the presence of ATR. Cell death was assessed by MTT assay, and data were plotted by measuring the absorbance of the blue formazan product at 570 nm. Cells that underwent light treatment were compared with dark control cells, and three independent experiments were performed with 12 replicates in each experiment. Data are presented as the average percent viability ± SEM for each time point. Standard Student’s t-tests were performed; *statistically significant differences (P < 0.05) compared with dark control cells. (A) 661W cells were incubated in media without ATR (Media), with DMSO (Media/DMSO), or with DMSO and ATR (Media/ATR). DMSO served as the carrier for ATR. (B) Similar conditions as in (A) but applied to B4 cells (661W+Bcl-2). (C) RGCs were used as the control, where they were subjected to the same conditions described in (A).

Figure 4.

 

Light-induced apoptosis in the presence of ATol. Cell death was assessed by MTT assay, and data were plotted by measuring the absorbance of the blue formazan product at 570 nm. Cells that underwent light treatment were compared with the dark control cells, and three independent experiments were performed with 12 replicates in each experiment. Data are presented as the average percent viability ± SEM for each time point. Standard Student’s t-tests were performed; *statistically significant differences (P < 0.05) compared with dark control cells. (A) 661W cells were incubated in media alone (Media), with DMSO (Media/DMSO), or with DMSO and ATol (Media/ATol). DMSO served as the carrier for ATol. (B) Similar conditions as in (A ) but applied to B4 cells (661W+Bcl-2). (C) RGCs were used as controls where they were subjected to the same conditions described in (A).

Figure 5.

 

Light-induced apoptosis under reduced FBS concentrations. Cell death was assessed by MTT assay and data were plotted by measuring the absorbance of the blue formazan product at 570 nm. Cells that underwent light treatment were compared with the dark control cells, and three independent experiments were performed with 12 replicates in each experiment. Data are presented as the average percent viability ± SEM for each time point. Standard Student’s t-tests were performed; *statistically significant differences (P < 0.05) compared with dark control cells. (A) 661W cells were incubated in media without 9-cis retinal at 1% FBS and at 30,000 lux (Media), with DMSO in media with 1% FBS (Media/DMSO) or with DMSO and 9-cis retinal in media containing 1% FBS (Media/9-cis). DMSO served as the carrier for the chromophore. (B) Similar experiment as described in (A) but without FBS supplementation. (C) Similar experiment as described in (A) but with 15,000-lux light. (D) Similar experiment as described in B but with 15,000-lux light.

Table 1.

 

In Vitro Caspase Activity Assay in Light-Stressed 661W Cells in the Presence of 9-cis Retinal

Caspase	n	661W/D3 ± SEM	661W/L3 ± SEM	% Increase
2	4	0.041 ± 0.010	0.115 ± 0.052	180
3	3	0.012 ± 0.002	0.232 ± 0.006	1833
4	2	0.029 ± 0.010	0.123 ± 0.059	324
6	2	0.053 ± 0.017	0.094 ± 0.010	77
8	4	0.029 ± 0.002	0.131 ± 0.004	352
9	5	0.046 ± 0.019	0.118 ± 0.018	157
10	5	0.048 ± 0.018	0.159 ± 0.050	231

 Assays were performed using pNA-conjugated substrates. Data are the mean ± SEM of absorbance readings at 405 nm. n, number of times the assay was performed, with two replicates for each experiment; % Increase, percentage increase of caspase activity in samples exposed to light compared with that in the dark. In the dark, there were no significant differences between 661W cells alone and 661W cells in the presence of 9-cis retinal. Caspase-1 and -5 were assayed once, with two replicates, and were found not to be activated; hence, these data are not included.

Figure 6.

 

Western blot analyses of caspases and their potential targets in 661W cells after light stress in the presence of 9-cis retinal. Detected bands after 1 to 4 hours’ exposure to light (1–4L) and samples left for 4 hours in the dark (4D) served as the control. (A) Western blot analysis of procaspase 10. (B) Western blot analysis of caspase-8: (gray arrowhead) the procaspase band; (black arrowhead) the caspase-8 band. (C) Protein blot showing the progression of activation of caspase-3. Numbers below the blots in (A–C) indicate lane sequence only. (D) Western blot analysis of caspase-9 activation: (black arrows) active forms of the enzyme. (E) Western blot analysis showing the progressive degradation of Bid the proapoptotic protein. (F) Western blot analysis of mitochondria-enriched fractions wherein 3 hours of exposure to light (3L) was sufficient to deplete cytochrome c. (G) Western blot analysis of PARP after 1 to 4 hours of exposure to light. Gray arrow: intact form; black arrow: cleaved form of PARP. (H) Analysis of fate of NF-κB after exposure to light. The left three lanes are from 661W cells and the right three lanes are from B4 cells. Actin controls are shown for each blot.

Supplementary Materials

Supplementary Figure S1 - higher-res - 18.1 MB (.tif) 

Supplementary Figure S1 - lower-res - 186 KB (.gif) 

Supplementary Figure S2 - higher-res - 13.8 MB (.tif) 

Supplementary Figure S2 - lower-res - 1.8 MB (.gif) 

The authors thank Robert E. Anderson, Michael Ihnat, and Scot Plafker (Department of Cell Biology, Oklahoma University Health Sciences Center, Oklahoma City) for providing anti-caspases antibodies and to Kenneth Hensley (Free Radical Biology and Aging Research Program, Oklahoma Medical Research Foundation, Oklahoma City) for the anti-cytochrome c antibody, Kjell Sawyer for providing technical help, and Rafal Farjo for comments on the manuscript.