Role of NF-κB and MAPKs in Light-Induced Photoreceptor Apoptosis

Abstract

purpose. To elucidate the role of nuclear factor kappa B (NF-κB) and mitogen-activated protein kinases (MAPKs) in light-induced apoptosis of photoreceptors in culture and to explore the potential inhibitory effect of minocycline and sulforaphane on apoptosis.

methods. Apoptosis of 661W cells was induced by exposure to light and was detected by terminal dUTP transferase nick end labeling (TUNEL). The mRNA expression and protein production of 10 chemokines and noxious factors were examined by reverse transcription polymerase chain reaction (RT-PCR) and enzyme-linked immunosorbent assay (ELISA). The protein expression of the p65 subunit of NF-κB, and the MAPKs p-p38, p-p44/42, and p-JNK were examined by Western blot and immunofluorescence analyses.

results. After exposure to light for 4 hours, 60% to 70% of the 661W cells underwent apoptosis. The expression of five selected chemokines and noxious factors was upregulated. The protein expression of the p65 subunit of NF-κB was downregulated, and the expression of the MAPKs p-p38, p-p44/42, and p-JNK was upregulated. Pretreatment with SB203580 for 1 hour inhibited light-induced upregulation of p-p38 and inhibited photoreceptor apoptosis. Pretreatment with minocycline or sulforaphane for 1 hour inhibited light-induced downregulation of the NF-κB p65 subunit and inhibited photoreceptor apoptosis.

conclusions. Apoptotic photoreceptors secrete chemokines and noxious factors to induce an immunologic response. The NF-κB and MAPK pathways both are involved in light-induced 661W photoreceptor apoptosis. Minocycline and sulforaphane inhibit light-induced photoreceptor apoptosis, partly through an NF-κB-dependent mechanism, but not through the MAPK pathway.

Inherited retinal degenerations are a group of retinal diseases characterized by progressive and selective loss of photoreceptor cells. The hallmark of this group of genetically and phenotypically diverse diseases is photoreceptor death by apoptosis. ¹ The inhibition of apoptosis appears to be a logical approach to the amelioration of the disease process. However, the molecular events that initiate the apoptotic cascade are poorly understood. Studies of pure cultures of photoreceptor cells and their interaction with various therapeutic agents would facilitate our understanding of the molecular and cellular events that underlie the diseases. Retinal cell culture has been a useful tool for ocular research. As mentioned by Tan et al., ² it provides a convenient experimental system for the assessment of isolated cellular processes. There are also potential limitations, including loss of native tissue architecture, lack of functional feedback from other retinal cell types, and a questionable correlation between in vitro and in vivo findings. However, to tackle certain research questions, the advantages offered by in vitro systems outweigh the limitations. The difficulty of obtaining a large number of photoreceptor cells in vitro for pharmacologic studies necessitated our use of 661W photoreceptor cells. The 661W photoreceptor cell line was originally isolated from transgenic mice, expressing an H1T1 construct. ² H1T1 contains an SV40 T-antigen and is driven by the human interphotoreceptor retinol binding protein (IRBP) promoter. ² The 661W cells express the photoreceptor cell markers such as opsin, arrestin, phosphodiesterase, transducin, phosducin, recoverin, and IRBP. ^{2 3} They also possess light-induced cell death pathways, similar to those observed in vivo in photoreceptor cells, ⁴ making them a valuable tool in the study of cellular mechanisms of photoreceptor cell death. 

Studies in RCS rat, retinal degeneration (rd), and retinal degeneration slow (rds) mice showed that during photoreceptor degeneration, microglia were activated and immigrated from the inner retinal layer (IRL) to the outer nuclear layer (ONL). ^{5 6 7 8} The close spatial and temporal relationship between photoreceptor degeneration and microglial migration suggested that the degenerating photoreceptors release stimulating factors to attract microglia to the ONL. ⁸ It is widely accepted that microglial activation is elicited by neuronal damage, and the cytotoxic effects contribute to the pathogenesis of neurodegeneration. ^{5 9} Chemokines are chemotactic cytokines that act through G-protein-coupled receptors. They are subdivided into four groups, known as CXC, CC, C, and CX3C chemokines. ¹⁰ One of the key signaling candidates for microglial recruitment are the chemokines. ¹¹ However, there are currently no reports relating the expression of chemokines or noxious factors on degenerating photoreceptor cells. 

Mitogen-activated protein kinases (MAPKs) are serine/threonine kinases that play an instrumental role in signal transduction from the cell surface to the nucleus. The mammalian MAPKs include p38 (p38), an extracellular signal-regulated kinase (p44/42), and Jun N-terminal kinase (JNK). ¹² It has been shown that a balance between the survival-promoting p44/42 pathway and the death-promoting the p38 and JNK pathways determine the fate of cells. ¹³ Nuclear factor-κB (NF-κB) is a ubiquitous transcriptional factor that regulates a broad range of genes and plays a pivotal role in cell death and survival. ^{14 15 16} The MAPK and NF-κB pathways may have both positive and negative effects on apoptosis, depending on the types of cells and stimuli. Studies by Tang et al. ¹⁷ and Liu et al. ¹⁸ showed that the MAPK and NF-κB pathways are intimately linked and are almost invariably coactivated by cytokines and stress. One of the major molecules for p65 phosphorylation and transactivation may be p38. ¹⁹ A study by Krishnamoorthy et al. ⁴ demonstrated that NF-κB is constitutively expressed in 661W cells and that its activity is progressively downregulated on exposure to photo-oxidative stress. The possibilities that MAPKs play a role in light-induced photoreceptor apoptosis and that the co-relationship between MAPKs and NF-κB pathways regulates photoreceptor apoptosis are important questions for research. 

In the present study, we investigated (1) the expression of immunologic signaling molecules in light-induced apoptotic 661W cells in culture, (2) the cellular pathways regulating the light-induced photoreceptor apoptosis process, and (3) the potential therapeutic effect of minocycline and sulforaphane on light-induced photoreceptor apoptosis. 

Materials and Methods

Materials

All chemicals were reagent grade or better. dl-Sulforaphane was obtained from LKT laboratories (St. Paul, MN). MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide), minocycline hydrochloride, U0126, SB203580, curcumin, and pyrrolidine dithiocarbamate (PDTC) were obtained from Sigma-Aldrich (St. Louis, MO). 

661W Cell Culture

The photoreceptor 661W cell line was a generous gift from Muayyad R. Al-Ubaidi, University of Oklahoma (Oklahoma City). 661W cells were grown in DMEM (Invitrogen Corp., Carlsbad, CA) containing 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin and were incubated at 37°C in a humidified atmosphere of 5% CO₂ and 95% air. Depending on the experiment, the cells were seeded in either 6- or 12-well culture plates. 

Cell-Exposure Experiments

Exposure to Light of 661W Cells.

The 661W cells were exposed to light, according to the protocol published by Al-Ubaidi et al. ³ The cells were seeded in a six-well culture plate for 24 hours. Exposure to light was performed in a cylindrical chamber of green Plexiglas (catalog no. 2092; Dayton Plastics, Dayton, OH) at an illuminance level of 4.5 mW/cm² (PR-650; Photo Research, CA). Because green Plexiglas has an effective band pass of 490 to 580 nm, ²⁰ the cells received only green light. The cells were 18 cm away from the light source. The chamber was surrounded by six 12-inch-long circular, 40-W, cool-white fluorescent tubes (General Electric, Fairfield, CT). The illumination inside the chamber was equidirectional, and the medium temperature during exposure to light was 35 ± 2.0°C. The cell incubator was placed in a dark room, and the cells were cultured in darkness. We examined the cells daily at 8 AM. The light treatment was always performed at the same time of day: 9 AM to 1 PM. The accompanying control cells were shielded from light and were kept under conditions similar to those for the cells in the light-exposure paradigm. 

Treatment with Therapeutic Agents.

In separate experiments, 661W cell cultures were exposed to the following agents: minocycline ²¹ (0.2 μM), sulforaphane ^{22 23} (5.0 μM), U0126 ²⁴ (10.0 μM, a p44/42 inhibitor), SB203580 ²⁵ (10.0 μM, a p38 inhibitor), curcumin ²⁶ (1.0 μM, a JNK inhibitor), or PDTC ²⁷ (60.0 μM, an NF-κB p65 subunit inhibitor). The cells were exposed for 1 hour to these agents, and the agents were removed and fresh medium was added to the cell culture. The stated concentrations were selected in studies with various concentrations of these agents. In addition, a dose–response curve was used to examine the effects of different doses of minocycline on photic injury. Cultures were treated with 20, 5, 2.5, 1.25, 0.625, 0.32, 0.16, 0.08, 0.04, 0.02, 0.01, 0.005, 0.0025, 0.0012, or 0.0006 μM minocycline for 1 hour, followed by exposure to photic injury. Similarly, for a sulforaphane dose–response curve, cultures were treated with 80, 40, 20, 10, 5, 2.5, 1.25, 0.625, and 0.32 μM sulforaphane for 1 hour, followed by exposure to photic injury. For the dose–response curve of different concentrations of SB203580, cultures were treated with 640, 320, 160, 80, 40, 20, 10, 5, 2.5, 1.25, 0.625, 0.32, and 0.16 μM SB203580 for 1 hour, followed by exposure to photic injury. Sulforaphane, U0126, SB203580, and curcumin were dissolved in dimethyl sulfoxide (DMSO) and diluted with serum-free medium before addition to the microtiter plate well. The final concentration of DMSO was <0.1% (by volume). Minocycline and PDTC were dissolved in phosphate-buffered saline (PBS) and diluted with serum-free medium. 

Immunofluorescence Studies

For immunofluorescence, the cultures were fixed with 4% paraformaldehyde in 0.01 M PBS for 30 minutes and then rinsed with PBS. The cells were permeabilized with cold (−20°C) methanol for 10 minutes. After they were washed with PBS, the cells were incubated in blocking buffer (1% bovine serum albumin) for 3 hours at room temperature. Subsequently, they were incubated with primary antibodies to green-sensitive opsin (1:100, sc-14358; Santa Cruz Biotechnology, Santa Cruz, CA), the p65 subunit of NF-κB (1:100, sc-375; Santa Cruz Biotechnology), or the MAPKs p-p38 (1:1000, V1211, Promega, Madison, WI), p-p44/42 (1:1000, V8031; Promega), or p-JNK (1:1000, V7931, Promega) overnight at 4°C. After they were rinsed with PBS, the cultures were incubated with the appropriate secondary antibodies (code 305-165-003 for opsin; code 111-165-003 for the NF-κB, p-p38, and p-JNK MAPKs; and code 111-095-003 for the p-p44/42 MAPK; Jackson ImmunoResearch Laboratories, West Chester, PA) for 45 minutes and examined by fluorescence microscopy. 

3′-End Labeling of Fragmented DNA by Fluorometric TUNEL Labeling

The 661W cells, pretreated with or without minocycline, sulforaphane, or SB203580, were exposed to light for 4 hours. After exposure, the cells were immediately fixed with 4% paraformaldehyde for 30 minutes and processed for a TUNEL assay. The TUNEL procedure, as described by Gavrieli et al., ²⁸ was performed with a commercially available apoptosis kit (DeadEnd Fluorometric TUNEL System; Promega), according to the supplier’s instructions. 

Cell-Viability Assays

Cell viability was determined by spectroscopic measurement of the reduction of MTT. ²⁹ The MTT assay is based on the ability of a mitochondrial dehydrogenase enzyme from viable cells to cleave the tetrazolium rings of the pale yellow MTT and form dark blue formazan crystals to which cell membranes are largely impermeable, thus resulting in its accumulation within healthy cells. Solubilization of the cells by the addition of a detergent results in the liberation of the crystals that are solubilized. The viability of cells is directly proportional to the level of the formazan product created. After exposure to light for 4 hours, the 661W cell culture medium was discarded, and the cells were washed three times with PBS. Each well received 1 mL of MTT solution (0.5mg/mL) in serum-free medium. The plate was incubated for 2 hours at 37°C, the MTT solution was discarded, 500 μL of DMSO was added to each well, and the plate was shaken at 200 rpm on an orbital shaker for 5 minutes. The absorbances of the wells were determined at 548 nm with a microplate reader (Bio-Rad, Hercules, CA). In each experiment, three identical six-well plates were used. The means of the absorbance values, the standard deviations of the means, and their coefficients of variation were calculated. 

Nitric Oxide and Cytokine Assays

The production of NO was quantified by measuring the released NO metabolites (nitrates and nitrite) with Griess Reagent (Sigma-Aldrich). After exposure to photo-oxidative stress for 4 hours, the cells were incubated at 37°C for another 24 hours. The culture medium samples were then collected and rendered cell-free by centrifugation. The medium (100 μL) was incubated with the same volume of Griess reagent at room temperature for 15 minutes and was measured at 570 nm in a microplate reader with an appropriate standard. 

Fractalkine (CX3CL1), monocyte chemotactic protein (MCP)-1, MCP-3, macrophage inflammatory protein (MIP)-1α, MIP-1β, eotaxin, regulated on activation normal T-cell expressed and secreted (RANTES) protein, interleukin-1 beta (IL-1β), and tumor necrosis factor (TNF)-α samples were prepared in manner similar to that used for the NO samples. The samples were assayed with mouse ELISA kits (R&D Systems, Minneapolis, MN). The experiments were repeated four times. 

Total RNA Extraction and Semiquantitative Reverse Transcription

The 661W cells, pretreated with or without minocycline, sulforaphane, or SB203580, were exposed to light for 4 hours; incubated at 37°C for 1 hour or 4, 12 or 24 hours; and harvested. The cells were seeded in six-well culture plates with 1 × 10⁶ cells per well. Total RNA was extracted (TRIzol reagent; Invitrogen-Gibco, Grand Island, NY). Reverse transcription was performed with oligonucleotide primers and reverse transcriptase (Superscript II; Invitrogen), and PCR was performed. The primers and annealing temperatures are shown in Table 1 . Each PCR product was separated on a 2% agarose gel and analyzed (Quantity One 1-D Analysis Software; Bio-Rad). From our experiments with series diluted template, amplification for 30 to 35 cycles was in the linear range of detection for the examined genes. Because of this, we chose to use 35 cycles of amplification for all the genes in the study. In the RT-PCR assays, the mRNA level for every sample was diluted to the same concentration, to avoid the influence of changes in the number of cells. The PCR experiments were repeated four times from separate cultures. Both β-actin and GAPDH were chosen as the housekeeping controls. In 10 experiments, β-actin was shown to be a satisfactory control and did not change in all experiments with various treatments. Because the expression level of GAPDH was parallel to the expression of β-actin, we showed only β-actin as the control for normalizing the data. The level of expression was compared within the time course for one particular gene mRNA. In a particular series of experiments (between 0 and 24 hours), the highest expression measured was normalized to 100%. 

Western Blot Analysis

The 661W cells, pretreated or not with minocycline, sulforaphane, or SB203580, were exposed to light for 4 hours. They were then incubated at 37°C for 0, 0.5, or 1 hour or 2 or 4 hours and were harvested. The cells were seeded in six-well culture plates with 1 × 10⁶ cells per well. Protein was extracted with protein lysis buffer (50 mM Tris-Cl [pH 8.0] 0.02% sodium azide, 1 μg/mL aprotinin, 1% NP-40, and 100 μg/mL phenylmethylsulfonyl fluoride [PMSF]), and the final protein concentrations were determined with a bicinchoninic acid (BCA) protein assay kit (Pierce, Rockford, IL). Western blot analysis was then performed as previously described. ³⁰ Polyclonal antibodies specific for the p65 subunit of NF-κB (sc-372; Santa Cruz Biotechnology), p-p38 (Anti-ACTIVE p38, V1211; Promega), p-p44/42 (Anti-ACTIVE MAPK, V8031; Promega), and p-JNK (Anti-ACTIVE JNK, V7931, Promega) were used for immunodetection. In some cases an anti-ERK2 polyclonal antibody (sc-154, Santa Cruz Biotechnology) was used to assay total (phospho-independent) ERK and served as a control to ensure that equivalent quantities of proteins were used for SDS-PAGE. Next, the blots were incubated with donkey anti-rabbit IgG (H+L) horseradish peroxidase (HRP), which is an affinity-purified HRP-conjugated secondary antibody. Blots were visualized using an enhanced chemiluminescent technique (SC-2048; Santa Cruz Biotechnology). To evaluate other proteins, the blot was immersed in stripping buffer (62.5 mM Tris-Cl [pH 6.8], 2% SDS, and 100 mM β-mercaptoethanol) for 30 minutes at 50°C and reprobed. The Western blot experiments were repeated four times on separate cultures. 

Statistical Analysis

Data are presented as the mean ± SD. Statistical comparisons were made by single-factor ANOVA. P <0.01 was considered significant. 

Results

Characterization of 661W Cells

To characterize the photoreceptor origin of the cultured cells, we tested the 661W cells for their expression of opsin, a photoreceptor cell marker. Positive labeling by immunofluorescence techniques was observed in the cytoplasm and nucleus of all cultured cells (Fig. 1) . Omission of the primary antibody resulted in the absence of labeling. 

Apoptosis in Light-Exposed 661W Cells

To determine the effect of photo-oxidative stress on the 661W cells, we exposed the cells to green light, as described in the Methods section, and studied them with a phase-contrast microscope, with the TUNEL technique. The cultured cells grown in the dark were flattened (Fig. 2A)and did not show any TUNEL-positive labeling (Fig. 3A) . After exposure to light for 4 hours, the cells assumed a spindle shape, and large intercellular spaces developed (Fig. 2B) . Sixty to 70% of light-exposed cells underwent apoptosis, as seen by incorporation of fluoresceinated dUTP in the nuclei of apoptotic cells containing fragmented DNA (Fig. 3B) . 

For studies with therapeutic agents, the cells were exposed to minocycline (0.2 μM), sulforaphane (5 μM), SB203580 (10 μM), U0126 (10 μM), PDTC (60 μM), or curcumin (1.0 μM) in the dark for 24 hours. The cultured cells exhibited no morphologic change and no apoptosis. Pretreatment with minocycline (0.2 μM; Fig. 2C ), sulforaphane (5 μM; Fig. 2D ), or SB203580 (10 μM; Fig. 2E ) for 1 hour before exposure to light for 4 hours caused the 661W cells to appear more flattened than those that did not receive treatment (Fig. 2B) , but no TUNEL-positive staining was detected (Figs. 3C 3D 3E) . Cultured cells that were pretreated with U0126 (10 μM) or PDTC (60 μM) for 1 hour followed by exposure to light for 4 hours were quickly detached and died, whereas those that did not receive treatment remained intact. There was no difference with or without curcumin treatment. 

Cell-Viability Assay

To determine cell viability in the different treatment conditions, we performed the MTT test. The viability of the cells grown in the dark was approximately three times higher than that of the light-exposed cells. The viability of the cells grown in the dark and pretreated with minocycline, sulforaphane, or SB203580 was statistically nonsignificant when compared with that of control cells grown in the dark. However, the viability of the light-exposed cells pretreated with minocycline, sulforaphane, or SB203580 was approximately three times greater than that of light-exposed cells without therapeutic treatment, and the difference was statistically significant (Table 2) . 

Minocycline, sulforaphane, or SB203580 had a protective effect against light-induced injury in the 661W cells (Fig. 4) . The maximum protection was achieved by treatment with 20 nM to 1.25 μM minocycline, 2.5 to 40 μM sulforaphane, or 40 to 640 μM SB203580. Higher doses did not further increase photoreceptor survival. As a result, 0.2 μM minocycline, 5 μM sulforaphane, or 10 μM SB203580 were used for the next set of experiments. 

mRNA Expression of Several Chemokines and Noxious Factors in 661W Cells

The mRNA expression of chemokines and noxious factors in the 661W cells grown in the dark or after exposure to light were investigated. Representative gel images were selected from four separate experiments (Fig. 5) . We quantified all four separate gel images, and the statistical results were based on the quantitative data (mean ± SD). The 661W cells grown in the dark constitutively expressed modest quantities of mRNA transcripts for CX3CL1, MCP-1, and MCP-3, and their expression was mildly but insignificantly upregulated after exposure to light. The 661W cells grown in the dark did not express mRNA transcripts for MIP-1α and -1β. After the cells were exposed to light for 4 hours, MIP-1α and -1β expression was increased, peaking at 12 hours after exposure to light. The 661W cells grown in the dark expressed modest quantities of eotaxin and RANTES mRNA transcripts. After 4 hours of exposure to light, eotaxin and RANTES expression in the cells decreased. The expression of RANTES was lowest at 4 hours and that of eotaxin was lowest at 12 hours after exposure (Fig. 5A) . 

The 661W cells grown in the dark expressed low levels of mRNA transcripts for iNOS, IL-1β, and TNFα. After exposure to light for 4 hours, their expression was upregulated. The expression of IL-1β peaked at 1 hour after exposure. However, the expression of iNOS and TNFα was delayed and peaked at 12 hours after exposure (Fig. 5A) . 

Minocycline, sulforaphane, and SB203580 had no effect on the mRNA expression of chemokines and noxious factors in cell culture in the dark. Their modulatory effect on the examined genes was most evident at 12 hours after exposure, except for IL-1β, which was most evident at 1 hour after exposure, these two time points were selected to demonstrate in Figures 5B and 5C . Minocycline (0.2 μM) decreased the light-induced expression of MIP-1α, iNOS, IL-1β, and TNFα mRNAs by 75%, 69%, 100%, and 50%, respectively, after exposure to light for 12 hours, and markedly decreased the expression of RANTES. Sulforaphane (5 μM) decreased the light-induced expression of MIP-1α, iNOS, IL-1β, and TNFα mRNAs by 69%, 42%, 100%, and 65%, respectively, after exposure to light for 12 hours, but had no effect on the expression of RANTES. Both minocycline and sulforaphane almost completely blocked the light-induced downmodulation of eotaxin, but had no significant effect on the expression of MIP-1β. Pretreatment with SB203580 (40 μM) for 1 hour decreased light-induced expression of MIP-1α, IL-1β, and TNFα mRNAs by 70%, 100%, and 92%, respectively, after exposure to light for 12 hours, but had no significant effect on the expression of MIP-1β, eotaxin, RANTES, and iNOS (Figs. 5B 5C) . 

NO and Chemokine Protein Production Assays

The production of NO, CX3CL1, MCP-1, MCP-3, MIP-1α, MIP-1β, eotaxin, RANTES, IL-1β, and TNFα in the culture media of 661W cells grown in the dark or after exposure to light were examined by using Griess reagent and ELISA. The 661W cells grown in the dark produced modest quantities of CX3CL1, MCP-1, and MCP-3. After exposure of the cells to light for 4 hours, production was mildly upregulated, but the difference was statistically insignificant. The 661W cells grown in the dark produced low levels of MIP-1α, MIP-1β, NO, IL-1β, and TNFα. Their production was upregulated after the cells were exposed to light for 4 hours. The 661W cells grown in the dark produced modest quantities of eotaxin and RANTES. Their production was downregulated after 4 hours (Table 3) . 

Minocycline, sulforaphane, and SB203580 had no effect on the production of CX3CL1, MCP-1, MCP-3, MIP-1α, MIP-1β, NO, IL-1β, and TNFα under the dark experimental condition (data not shown). Pretreatment with minocycline or sulforaphane for 1 hour, followed by exposure of the cells to light for 4 hours, significantly decreased the light-induced production of MIP-1α, NO, IL-1β, and TNFα, but had no significant effect on the production of CX3CL1, MCP-1, MCP-3, MIP-1β, eotaxin, and RANTES. Pretreatment of the cells with SB203580 for 1 hour, followed by exposure to light for 4 hours, significantly decreased the light-induced production of MIP-1α, IL-1β, and TNFα, but had no significant effects on the production of CX3CL1, MCP-1, MCP-3, MIP-1β, eotaxin, RANTES, and NO (Table 3) . 

NF-κB and MAPKs Activity in 661W Cells

Photo-oxidative stress has been reported to downmodulate the expression of NF-κB and may lead to photoreceptor apoptosis. ⁴ The MAPK pathways link extracellular stimuli to transcriptional activation, but their roles in light-induced photoreceptor apoptosis have not been determined. The protein expression of the NF-κB p65 subunit and the MAPKs p-p38, p-p44/42, and p-JNK in dark conditions and after exposure to light was investigated. The 661W cells expressed a high level of the NF-κB p65 subunit in the dark, and its expression decreased after exposure to light (Fig. 6A) . The 661W cells expressed very low levels of p-p38, p-p44/42, and p-JNK in the dark. Their expression was upregulated after 4 hours of exposure to light (Fig. 6A) . Pretreatment with SB203580, a p-p38 specific inhibitor, completely inhibited the light-induced upregulation of p-p38 (Fig. 6B) . 

To examine whether minocycline or sulforaphane protect 661W cells from photic injury via MAPK- or NF-κB-dependent pathways, the expression of the NF-κB p65 subunit and the expression of the MAPKs p-p38, p-p44/42, and p-JNK after exposure of the cells to light in the presence of minocycline or sulforaphane were determined. Minocycline or sulforaphane had no effect on the expression of the p65 subunit of NF-κB or p-p38, p-p44/42, and p-JNK in the dark. Pretreatment with minocycline or sulforaphane for 1 hour followed by exposure to light completely blocked the light-induced downregulation of NF-κB, but had no effect on the upregulation of p-p38, p-p44/42, and p-JNK (Fig. 6B) . 

Immunofluorescence Studies

To confirm further the downregulation of the p65 subunit of NF-κB and the upregulation of the MAPKs p-p38, p-p44/42, and p-JNK in the 661W cells, the protein levels were studied by using specific antibodies. The 661W cells expressed a high level of NF-κB in the dark (Fig. 7A) . After the cells were exposed to light for 4 hours, the expression decreased both in the cytoplasm and in the nuclei (Fig 7B) . Pretreatment with minocycline (Fig. 7C)or sulforaphane (Fig. 7D)for 1 hour completely blocked the light-induced downregulation of NF-κB. The 661W cells did not express p-p38 (Fig. 8A) , p-p44/42, or p-JNK in the dark (data not shown). In cells exposed to light for 4 hours, their expression was upregulated (Figs. 8B 8C 8D) . Pretreatment with SB203580 for 1 hour followed by exposure of the cells to light for 4 hours completely inhibited light-induced upregulation of p-p38 MAPK (Fig. 8E) . 

Discussion

In the present study, the 661W cells in culture were exposed to light for 4 hours and were studied as a model to investigate photoreceptor apoptosis. Previous studies have established the expression of mRNA and/or proteins of photoreceptor cell markers such as opsin, arrestin, phosphodiesterase, transducin, phosducin, recoverin, and IRBP in the 661W cells. These cellular expressions have provided us a valuable tool for the study of photoreceptor cell death mechanism. ^{2 3} However, when compared with the in vivo photoreceptor cells, the 661W cells have limitations: (1) They are proliferative in culture, unlike photoreceptor cells in the differentiated retinas; (2) they do not exhibit cone photoreceptor morphology, such as formation of outer-segment–like membranes; (3) they do not express outer segment structural proteins, such as peripherin/rds and ROM1; and (4) they do not express RPE65, an important determinant in rhodopsin regeneration kinetics and light-damage susceptibility in mice. 

It has been demonstrated that rhodopsin and its ability to regenerate is an essential factor in light-induced photoreceptor apoptosis. ^{31 32} In a previous study, green light was reported to be absorbed by the green-sensitive cone photopigment. ³³ In the present study, immunolocalization studies showed that green opsin was distributed all over the 661W cell, including the nucleus. An intriguing question is how these cells regenerate their pigment and respond to light stress in the absence of RPE65. Our explanation was as follows: (1) The cone cells may evolve an additional pathway for regeneration of 11-cis retinal. ³⁴ It has been reported that cone cells regenerate visual pigment in isolated retinas detached from the RPE. ^{35 36} (2) The cone cells may express their own RPE65. ³⁷ Therefore, unless the cone cells, or the 661W cells specifically, have an alternative pathway for pigment regeneration, it is difficult to speculate as to why the 661W cells are light sensitive. One possibility is that the 661W cells express RGR (retinal G-protein-coupled receptor), a protein that structurally resembles visual pigments and other G-protein-coupled receptors. ³⁸ RGR may play a role as a photoisomerase in the production of 11-cis retinal, the chromophore of the visual pigments. ²  

However, in the present study, after exposure to light for 4 hours, most of the 661W cells underwent apoptosis. Apoptotic 661W cells expressed mRNA transcripts and secreted proteins of chemokine and noxious factors, to induce an immunologic response. The amounts of protein production were consistent with the expressed mRNA levels. Eotaxin and RANTES were the exception. The mRNA expression of eotaxin and RANTES was markedly downregulated (Fig. 5A) . However, the protein production in the culture medium was only mildly lowered (Table 3) . Photic injury suppressed the mRNA transcription, but it was possible that the remaining mRNA in the cytoplasm may continue to translate their proteins and secret them into the media. The CC chemokines, including MCP_1-5, MIP-1α and -1β, eotaxin, and RANTES predominantly acts on the microglia, monocytes, and macrophages. MIP-1α and -1β are involved in acute inflammation, ³⁹ and their expression is upregulated in retinal degeneration. ^{40 41} The MCP-1 and MCP-3 are undetectable in the normal retina, but their expression is upregulated in light-induced or inherited retinal degeneration. ^{8 41 42} There is also evidence that mice lacking either MCP-1 or its receptor CCR2 undergo pathologic changes comparable to those in age-related macular degeneration (ARMD), demonstrating photoreceptor atrophy, drusen accumulation, and lipofuscin in the retinal pigment epithelium (RPE). ⁴³ Consistent with the in vivo studies, the 661W photoreceptor cells constitutively expressed MCP-1 and -3, and after exposure to light for 4 hours, their expression was mildly upregulated (Fig. 5A , Table 3 ). A study in our laboratory ⁴¹ demonstrated that the normal retina expresses modest quantities of RANTES and eotaxin and that their expression is markedly upregulated in retinal degeneration. The 661W cells constitutively expressed modest quantities of RANTES and eotaxin; but, inconsistent with the in vivo studies, after exposure of the cells to light, their expression was downregulated (Fig. 5A , Table 3 ). CX3CL1 (fractalkine) is a relatively new member of the chemokine family and is the sole member of the CX3C chemokine class. In contrast to many other chemokines, CX3CL1 binds to only one receptor (CX3CR1). ⁴⁴ Expression of CX3CL1 is localized in the neurons, ⁴⁵ whereas CX3CR1 is expressed by the microglia ⁴⁶ in the brain. The constitutive expression of CX3CL1 and its receptor CX3CR1 implies a role in mediating neuronal–microglial cross-talk under normal and pathologic conditions. ⁴⁴ However, a recent in vivo study demonstrated that CX3CL1 was constitutively expressed in the retina, and its expression did not change significantly during retinal degeneration. ⁸ Consistent with this study, our observation demonstrated that CX3CL1 was constitutively expressed in 661W photoreceptor cells, and after the cells were exposed to light, its expression was not significantly altered (Fig. 5A , Table 3 ). 

Injured photoreceptors secrete noxious factors to induce apoptosis. A low level of NO is an important mediator of homeostatic processes in the normal eye, such as regulation of aqueous humor dynamics, ⁴⁷ retinal neurotransmission, ⁴⁸ and phototransduction. ⁴⁹ However, the production of NO at high concentrations activates microglia and is also toxic to neurons. ^{41 50 51} Normal retina expresses a very low level of TNFα, and its expression is upregulated in retinal degenerative diseases. ⁸ Consistent with these in vivo studies, the 661W photoreceptor cells expressed very low levels of iNOS, IL-1β, and TNFα in the dark, but after exposure to light, their expression was upregulated (Fig. 5A , Table 3 ). We hypothesized that the degenerative photoreceptors not only secrete chemokines but also other noxious factors, to activate and attract the microglia and to accelerate the degenerative process through an autocrine positive feedback loop. 

The 661W cells constitutively expressed a high level of the NF-κB p65 subunit in the dark, and its expression decreased after exposure to light (Figs. 6A 7A 7B) . Further pretreatment with PDTC, an inhibitor of the NF-κB p65 subunit, exaggerated the light-induced photoreceptor apoptosis. Krishnamoorthy et al. ⁴ have demonstrated that exposure of culture photoreceptor cells to light generates photo-oxidative stress and causes the production of reactive oxygen intermediates (ROIs). The ROI-activated interleukin 1β-converting enzyme (ICE) produced proteolytic cleavage of NF-κB proteins, which led cells to apoptosis. ⁴ Our observation supported that the p65 subunit of NF-κB was an important survival-promoting transcriptional factor for 661W cells. 

Incubation in cell culture medium (such as DMEM or RPMI 1640) exposed to light generates ROIs in the medium. The main component responsible for the generation of ROIs is riboflavin. The production of ROIs is light dependent. ⁵² The possibility that the response of 661W cells to light resulted from ROIs in the medium must be excluded. To assess the specificity of response of 661W cells to photo-oxidative stress, Krishnamoorthy et al. ⁴ performed two experiments. First, to confirm the role of ROIs in this process, they treated the cells with H₂O₂, and found that ROIs alone are not sufficient for light-induced downregulation of NF-κB and activation of apoptosis. In the second experiment, to assess the specific response of the 661W cells, they studied the effect of exposure to light on Madin-Darby Canine Kidney (MDCK) cells as an unrelated control. Although the exposure to light of the 661W cells led to a decrease in NF-κB binding activity and apoptosis, the same stimulus did not greatly alter the nuclear NF-κB activity and did not lead to cell death in the MDCK cells. These results suggest that apoptosis is a cell-specific response of 661W cells to light. 

The 661W cells constitutively expressed very low levels of the MAPKs p-p38, p-p44/42, and p-JNK in the dark. After exposure to light for 4 hours, the expression of p-p38, p-p44/42, and p-JNK was upregulated (Figs. 6A 8) . Pretreatment with SB203580, a specific inhibitor of p-p38, completely inhibited the light-induced activation of p-p38 and photoreceptor apoptosis. However, pretreatment with U0126, a specific inhibitor of p-p44/42, exaggerated the light-induced photoreceptor apoptosis. In contrast, pretreatment with curcumin, an inhibitor of p-JNK, had no effect on light-induced photoreceptor apoptosis. In studies by Kikuchi et al. ⁵³ and Roth et al., ⁵⁴ inhibition of p38 reduced apoptosis of glutamate-regulated retinal ganglion cells (RGCs) in vitro and reduced axotomy-induced RGC death in vivo. Consistent with their results, in our study, p38 was a key proapoptosis transcriptional factor in light-induced photoreceptor apoptosis. Pretreatment with SB203580 inhibited photo-oxidative stress-induced upregulation of MIP-1α, IL-1β, and TNFα (Figs. 5B 5C ; Table 3 ), and the expression of these chemokines and noxious factors in the 661W cells after exposure to light may partly be due to p38 MAPK activation. The p-p38 MAPK pathway was shown to be involved in the induction of several proinflammatory genes and was the first step in initiating the inflammatory process. SB203580, the specific inhibitor of the p38 MAPK, therefore, protected photoreceptors through a direct antiapoptotic effect, as well as an indirect immunologic suppressive mechanism. The p38 MAPK inhibitors could be useful for the treatment of retinal diseases. Previous studies have shown that the p44/42 is a major proapoptosis transcriptional factor for RGCs ⁵⁴ and that inhibition of the p44/42 pathway results in protection of the brain from ischemic injury. ⁵⁵ In contrast to the findings in these studies, our observation suggest that the activation of p44/42 plays a protective role in photoreceptor apoptosis. Our observation supports a role for the p44/42 MAPK signaling pathway in mediating cell division, migration, and survival. ⁵⁶ The different roles of this pathway may be attributed to differences in cell types. JNK is an important factor controlling programmed cell death or apoptosis, ⁵⁷ but Roth et al. ⁵⁴ found that JNK does not play an important role in RGC death. In the present study, JNK did not play a significant role in light-induced 661W photoreceptor cell apoptosis. 

Minocycline, a semisynthetic long-acting tetracycline derivative, capable of passage through the blood–brain barrier, has recently been shown to exhibit strong neuroprotective properties in models of neurodegeneration. ⁵⁸ Intraperitoneal administration of minocycline ameliorated photoreceptor loss in inherited or light-induced retinal degeneration. ^{59 60 61} The neuroprotective effect of minocycline may occur through inhibition of microglial activation and proliferation ⁶⁰ or alternatively by a direct antiapoptotic effect on the caspase cascade ⁵⁸ or on the p38 pathway. ⁶² In the present study, pretreatment of 661W cells with minocycline inhibited light-induced photoreceptor apoptosis and the downmodulation of the NF-κB p65 subunit (Fig. 6B , Fig. 7C ). However, minocycline had no direct effect on the expression of the MAPKs. Minocycline inhibited the mRNA expression of IL-1β (Fig. 5C) , inhibited ICE, and caused subsequent inhibition of proteolytic cleavage of NF-κB proteins. These observations suggest that minocycline inhibits light-induced photoreceptor apoptosis partly through an NF-κB-dependent mechanism. 

Sulforaphane is a naturally occurring isothiocyanate found in broccoli. ²² It inhibits phase I enzymes such as cytochrome P450 and induces phase II detoxification enzymes. ^{63 64 65} Intraperitoneal pretreatment with sulforaphane attenuates light-induced retinal damage via the antioxidant response element (ARE)/thioredoxin (Trx) signaling cascade. ⁶⁶ In the present study, pretreatment of 661W cells with sulforaphane inhibited light-induced photoreceptor apoptosis and inhibited down-modulation of the p65 subunit of NF-κB (Figs. 6B 7D) , but it had no effect on the expression of the MAPKs. Like minocycline, sulforaphane also inhibited the proteolytic cleavage of NF-κB proteins through ICE. It is noteworthy that sulforaphane treatment also activated the ARE in photoreceptor cells and led to expression of Trx reductase, which may eliminate oxidative stress. ⁶⁶ These observations suggest that sulforaphane inhibits light-induced photoreceptor apoptosis partly through an NF-κB-dependent mechanism. Broccoli is a widely consumed vegetable, ²² and it may be useful as a therapeutic agent for retinal degenerative disease. 

In our study, apoptotic photoreceptors secreted chemokines and noxious factors to induce an immunologic response. The NF-κB and MAPK pathways were both involved in light-induced 661W photoreceptor apoptosis. Minocycline and sulforaphane inhibited light-induced photoreceptor apoptosis, partly through an NF-κB-dependent mechanism, but not through the MAPK pathway. 

 Supported by the Michael Panitch Research Fund; Research to Prevent Blindness, Inc.; the Oliver Birckhead Research Fund; Grant 30600689 from the National Natural Science Foundation of China; and the Research Fund of the Peking University Third Hospital.

 Submitted for publication July 26, 2006; revised January 9 and March 12, 2007; accepted August 16, 2007.

 Disclosure: L.-p. Yang, None; X.-a. Zhu, None; M.O.M. Tso, 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.

 

Table 1.

 

Oligonucleotides Used for RT-PCR

Target Gene	Sequences (5′–3′)	Location	Annealing Temperature (°C)	Product Size (bp)
β-Actin	Sense: CTG GAG AAG AGC TAT GAG CTG	NT 786–1031	62	245
	Antisense: AAT CTC CTT CTG CAT CCT GTC
CX3CL1	Sense: GGC TCC CAT CTC CTC TGA AGA	NT 897–1222	59	325
	Antisense: CTG GCA CCA GGA CGT ATG AGT
MCP-1	Sense: CCC CAC TCA CCT GCT GCT ACT	NT 177–556	63	379
	Antisense: GGC ATC ACA GTC CGA GTC ACA
MCP-3	Sense: ATA GCC GCT GCT TTC AGC A	NT 107–335	62	228
	Antisense: CTA AGT ATG CTA TAG CCT CCT CGA
MIP-1α	Sense: CCA AAG AGA CCT GGG TCC AAG	NT 307–627	63	320
	Antisense: GGG TTG AGG AAC GTG TCC TGA
MIP-1β	Sense: CCA TGA AGC TCT GCG TGT CTG	NT 76–463	62	387
	Antisense: GGG CAG GAA ATC TGA ACG TG
Eotaxin	Sense: CCT GCT GCT TTA TCA TGA CCA	NT 139–419	61	280
	Antisense: CCC TCA GAG CAC GTC TTA GGA
RANTES	Sense: TGC CCT CAC CAT CAT CCT CA	NT 57–366	57	309
	Antisense: AAG CGA TGA CAG GGA AGC GTA
iNOS	Sense: CGA CCC GTC CAC AGT ATG T	NT 402–818	57	416
	Antisense: TAC AGT TCC GAG CGT CAA AG
IL-1β	Sense: AAG CTC TCC ACC TCA ATG GAC AG	NT 499–837	61	338
	Antisense: GAC CAC TGT TGT TTC CCA GGA AG
TNF-α	Sense: AGC CGA TGG GTT GTA CCT TGT	NT 546–875	63	329
	Antisense: ACC CAT TCC CTT CAC AGA GCA

Figure 1.

 

Immunofluorescence characterization of the culture 661W cells. (A) Culture 661W cells were reactive to opsin, a cell-type–specific marker for photoreceptors. (B) After exposure to light for 4 hours, the cells were spindle shaped (arrow) and appeared to have large intercellular spaces (arrowhead).

Figure 2.

 

Phase-contrast photomicrographs of the culture 661W photoreceptor cells. (A) 661W cells grown in the dark. (B) After exposure to light for 4 hours, the cells were spindle shaped (arrow) and appeared to have large intercellular spaces (arrowhead). After pretreatment with (C) minocycline, (D) sulforaphane, or (E) SB203580 for 1 hour, followed by exposure to light for 4 hours, the 661W cells appeared more flattened than those that did not receive treatment (B).

Figure 3.

 

TUNEL labeling of 661W cells in culture. The 661W cells pretreated with or without minocycline, sulforaphane, or SB203580 were exposed to light for 4 hours. The cells were immediately fixed with 4% paraformaldehyde and processed for the TUNEL assay. (A) No TUNEL-positive labeling was shown in 661W cells grown in the dark. (B) After exposure to light for 4 hours, 60% to 70% of the 661W cells underwent apoptosis (arrow). After pretreatment with (C) minocycline, (D) sulforaphane, or (E) SB203580 for 1 hour, followed by exposure to light, no TUNEL-positive labeling was seen.

Table 2.

 

Effects of Minocycline, Sulforaphane or SB203580 on MTT Cell-Viability Assays

Treatment	Exposure to Light	Grown in Dark
Culture grown in dark	—	0.965 ± 0.103
Light-exposed culture	0.323 ± 0.056	—
Minocycline pretreatment	0.856 ± 0.132*	0.973 ± 0.138
Sulforaphane pretreatment	0.783 ± 0.096*	0.956 ± 0.169
SB203580 pretreatment	0.987 ± 0.064*	0.962 ± 0.098

 The data are expressed as the mean ± SD of four individual experiments. Statistical comparisons were made by single-factor ANOVA.

*   P < 0.01 was considered significant compared with the light-exposed culture.

Figure 4.

 

Cell viability assays. Pretreatment with minocycline, sulforaphane, or SB203580 for 1 hour protected 661W cells against light-induced apoptosis. The maximum and most consistent protection was achieved by treatment with (A) 20 nM to 1 μM minocycline, (B) 2.5 to 40 μM sulforaphane, or (C) 40 to 640 μM SB203580. Higher doses did not increase photoreceptor survival. The results were expressed as the mean ± SD of four individual experiments.

Figure 5.

 

RT-PCR analysis demonstrated the expression of the mRNA of several chemokines and noxious factors in culture 661W photoreceptor cells. (A) Expression of mRNA of chemokines and noxious factors in 661W cells in the dark or after exposure to light for 4 hours. Exposure to light induced the upregulation of MIP-1α, MIP-1β, iNOS, IL-1β, and TNFα and also induced the downregulation of eotaxin and RANTES. The expression levels of CX3CL1, MCP-1, and MCP-3 were little changed. (B, C) Pretreatment with minocycline or sulforaphane significantly reduced the light-induced upregulation of MIP-1α, iNOS, IL-1β, and TNFα, inhibited the downregulation of eotaxin, but had no significant effect on the expression of MIP-1β. Minocycline inhibited the basal expression of RANTES, but sulforaphane had no effect. Pretreatment with SB203580 significantly reduced the light-induced upregulation of MIP-1α, IL-1β, and TNFα, but had no effect on the expression of MIP-1β, eotaxin, RANTES, and iNOS.

Table 3.

 

Effects of Minocycline, Sulforaphane or SB203580 on Light-Induced Production of CX3CL1, MCP-1, MCP-3, MIP-1α, MIP-1β, Eotaxin, RANTES, NO, IL-1β and TNFα

Groups	CX3CL1	MCP-1	MCP-3	MIP-1α	MIP-1β	Eotaxin	RANTES	NO	IL-1β	TNFα
Culture grown in the dark	80.06 ± 5.42	150.56 ± 15.08	189.24 ± 25.13	11.04 ± 2.96	9.87 ± 3.46	22.08 ± 2.63	21.34 ± 5.89	21.86 ± 3.09	15.04 ± 1.87	24.68 ± 3.64
Light-exposed culture	95.93 ± 9.01	168.42 ± 10.78	201.46 ± 26.24	19.83 ± 3.65*	46.35 ± 5.89*	18.06 ± 3.81	15.73 ± 4.46	58.49 ± 8.72*	70.86 ± 4.98*	80.93 ± 11.04*
Minocycline pretreatment and light-exposed	93.48 ± 8.96	158.96 ± 7.84	180.39 ± 16.20	8.73 ± 2.34, †	49.32 ± 6.78	21.64 ± 4.89	16.08 ± 3.71	15.29 ± 2.34, †	16.03 ± 3.96, †	50.73 ± 5.08, †
Sulforaphane pretreatment and light-exposed	86.54 ± 6.79	164.32 ± 5.98	194.76 ± 15.23	9.56 ± 3.89, †	54.83 ± 2.78	18.79 ± 2.14	17.16 ± 3.84	26.84 ± 3.04, †	10.08 ± 3.12, †	42.43 ± 5.88, †
SB203580 pretreatment and light-exposed	91.45 ± 10.31	160.29 ± 12.35	184.25 ± 19.78	5.74 ± 1.67, †	50.63 ± 7.34	19.08 ± 4.79	18.69 ± 4.06	49.87 ± 7.63	7.96 ± 1.24, †	9.46 ± 3.09, †

 661W cells were pretreated with minocycline (0.2 μM), sulforaphane (5.0 μM), or SB203580 (10.0 μM) for 1 hour, exposed to light for 4 hours, and incubated at 37°C for 24 hours. Culture media were collected and assayed. Data are expressed as the mean ± SD of results in four individual experiments.

 Statistical comparisons were made by single-factor ANOVA.

*   P < 0.01 was considered significant compared with cultures grown in the dark.

†   P < 0.01 was considered significant compared with light-exposed cultures.

Figure 6.

 

Western blot analysis demonstrated the protein expression of the NF-κB and MAPKs in cultured 661W photoreceptor cells. (A) The protein expression of the p65 subunit of NF-κB and the MAPKs p-p38, p-p44/42, and p-JNK in 661W cells in the dark or after exposure to light for 4 hours. The 661W cells expressed high levels of the p65 subunit of NF-κB, and expressed low levels of p-p38, p-p44/42, and p-JNK under dark conditions. After exposure to light for 4 hours, the expression of the NF-κB p65 subunit was downregulated; however, the expression of p-p38, p-p44/42, and p-JNK was upregulated. (B) Pretreatment with minocycline or sulforaphane for 1 hour completely inhibited light-induced downregulation of the NF-κB p65 subunit but had no significant effect on the expression of p-p38, p-p44/42, and p-JNK. Pretreatment with SB203580 for 1 hour completely inhibited light-induced upregulation of p-p38, but had no effect on the expression of p-p44/42 and p-JNK or the p65 subunit of NF-κB.

Figure 7.

 

Immunofluorescence localization of the p65 subunit of NF-κB in cultured 661W photoreceptor cells. (A) The p65 subunit of NF-κB was predominantly present in the nuclei (arrow) and cytoplasm (arrowhead) of dark-exposed control cells. (B) After exposure of the cells to light for 4 hours, the p65 subunit labeling was significantly reduced in the nuclei (arrow) and cytoplasm (arrowhead). (C) Pretreatment with minocycline or (D) sulforaphane for 1 hour followed by exposure to light completely inhibited light-induced downmodulation of the NF-κB p65 subunit. Positive labeling of the p65 subunit was present in both the nuclei and cytoplasm.

Figure 8.

 

Immunofluorescence localization of MAPKs in cultured 661W photoreceptor cells. (A) The cells did not express p-p38 in the dark. After exposure to light for 4 hours, (B) p-p38 was present in the nuclei (arrow); (C) p-p44/42 was present in the nuclei (arrow) and cytoplasm (arrowhead); and (D) p-JNK was present in the nuclei (arrow). (E) Pretreatment with SB203580 for 1 hour completely inhibited light-induced upregulation of p-p38 in 661W cells.