Caspase-8, -12, and -3 Activation by 7-Ketocholesterol in Retinal Neurosensory Cells

Abstract

purpose. To determine the caspase pathways involved with 7-ketocholesterol (7kCh)–induced apoptosis in rat R28 cells.

methods. R28 cells were exposed to 7kCh with or without low-density lipoprotein (LDL) and z-VAD-fmk, a pan-caspase inhibitor. Cell viability was measured by a trypan blue dye exclusion assay. Caspase-3, -8, -9, and -12 activities were measured by fluorochrome caspase assays. ARPE-19 cells were used as control for caspase-3 inhibition experiments.

results. R28 cultures showed decreased cell viability on 7kCh exposure compared with controls (P < 0.001), and this was reversed with LDL and LDL + z-VAD-fmk (P < 0.001). The 7kCh-treated R28 cultures had increased caspase-8 activity compared with controls (P < 0.001). This activity was blocked partially with LDL (P < 0.01) or LDL + z-VAD-fmk (P < 0.001) but not with z-VAD-fmk alone. Caspase-12 activity was increased after 7kCh treatment compared with controls (P < 0.01), and this activity was increased further with the addition of LDL. Caspase-3 activity in R28 cultures increased with 7kCh treatment compared with controls (P < 0.001). In R28 cultures, the z-VAD-fmk treatment did not blocked 7kCh-induced caspase-3 activity but did block activity in ARPE-19 cultures (P < 0.001). Caspase-9 was not activated by 7kCh treatment.

conclusions. In R28 cells, 7kCh-induced apoptosis involves the caspase-3 along with the caspase-8 and caspase-12 pathways. LDL partially blocked 7kCh-induced caspase-8 activity but increased caspase-12 activities, suggesting that caspase-8 and caspase-12 pathways are independent of each other. The z-VAD-fmk inhibitor blocked caspase-3 activities in the homogeneous ARPE-19 cultures but not in the heterogeneous R28 cultures.

Retinal degeneration is the major event responsible for the loss of vision in various inherited and acquired retinal diseases. The apoptosis of photoreceptors and other neurosensory cells is responsible not only for the degeneration but for the selective elimination of inappropriate connections and cells during normal development of the retina. Many parallel pathways are available for apoptotic or necrotic cell death, which may be triggered by factors such as ischemia, heat shock proteins, deprivation of growth factors, and oxysterols. 

Caspases, a family of cysteine proteases, is one of the best-studied apoptotic pathways. Caspases are synthesized in the cytosol as inactive proenzymes and, in response to severe stress, become activated in a particular sequential cascade. Once activated, executioner caspases contribute to dismantle the cell through direct proteolysis of cell structures and repair enzymes and regulatory proteins through initiator caspases (i.e., caspase-2, -8, -9, or -12) or effector (downstream) caspases (i.e., caspase-3, -6, or -7). Activation of caspase-3 represents a commitment for cell disassembly and is a hallmark for apoptosis. Three different pathways are recognized as the initiator caspases involved in activating apoptosis. ^{1 2} The extrinsic (receptor-mediated) pathway includes the Fas/Fas-ligand cell-surface receptors, which are activated and subsequently lead to caspase-8 activation. ³ Alternatively, the intrinsic (mitochondrial) pathway involves the activation of caspase-2 and -9, is associated with mitochondrial stress, and leads to cytochrome c release. ^{4 5} The newly recognized caspase-12 is activated by endoplasmic reticulum stress and induces the cleavage of caspase-3 in a cytochrome c–independent manner. ⁶ This caspase-12 pathway can cause pathologically relevant apoptosis and is implicated in neurodegenerative disorders. ⁷  

Loss of retinal structure and function has been reported in apolipoprotein E (ApoE)–deficient mice fed high-fat diets. ^{8 9} Oxidation of cholesterol yields oxysterols such as 7-ketocholesterol (7kCh), which was found to be one of the most toxic and predominant elements within the oxidized low-density lipoprotein (oxLDL). ¹⁰ Levels of 7kCh are elevated in diabetic rats ¹¹ and atherosclerotic plaques. ¹² The sterol 27-hydroxylase (CYP27A1), which has a substrate specificity for 7kCh, has been localized to the photoreceptor cells, RPE, and choriocapillaris. ¹³ In vitro, 7kCh and 25-hydroxycholesterol are toxic to the neuroretinal cells, retinal pigment epithelial cells, macrophages, and vascular cells, causing increased formation of reactive oxygen species and loss of cell viability. ¹⁴ Oxysterol toxicity may have implications in various retinal diseases such as age-related macular degeneration. Understanding the mechanisms of the apoptotic pathways is critical to develop interventional steps to reverse cell loss. Many studies examine the responses of single-cell types in culture. However, cell response may differ when multiple populations are cultured together. R28 cultures offer a well-characterized population of precursors to multiple neuroretinal cell types to investigate the mechanisms of caspase pathways. 

R28 retinal cells were subcloned from the immortalized E1A-NR3 retinal cell line, which is not tumorigenic. ¹⁵ Extensive characterization of the R28 cells showed that these cultures contained subpopulations of neuronal cells with many properties of central nervous system neurons and retinal cells. ¹⁶ These neuronal/retinal markers include microtubule-associated protein 2 (MAP2), syntaxin, calbindin, neuron-specific enolase, internexin, and nestin, many of which have been associated with retinal ganglion cells, horizontal cells, amacrine cells, glial cells, photoreceptors, and bipolar cells. Functionally, the R28 cells express neurotransmitter receptors ¹⁷ and are capable of responding to neurotransmitters such as dopamine, acetylcholine, serotonin, and glycine. ¹⁶ Therefore, the R28 cells represent a model system that is well studied for gene expression, ¹⁶ electrophysiology, ¹⁷ and apoptosis. ^{14 18 19} Although it is known that 7kCh can damage R28 cells in vitro, the specific caspase pathways have not been investigated. 

The present study was designed to determine which caspase pathways were activated in R28 cells after 7kCh treatment. Our results showed that 7kCh treatment decreased R28 cell viability and that the caspase-3, caspase-8, and caspase-12 pathways were involved. Our findings support the hypothesis that oxysterols can damage retina cells through apoptosis and may play a role in retinal diseases. In addition, the R28 cell line contains various neuroretinal precursors and thereby represents a multiple population of heterogeneous cells, similar to what is found in the retina. Our data suggest that these multiple population cultures respond differently to the caspase inhibitors than do the single-cell population cultures, such as ARPE-19 cultures. 

Materials and Methods

Cell Culture

R28 rat neurosensory cells were cultured in Dulbecco modified Eagle medium, high glucose (DMEM high glucose; Invitrogen-Gibco, Carlsbad, CA) with 10% fetal bovine serum, 10 mM nonessential amino acids, and 10 μg/mL gentamicin. They were obtained from postnatal day 6 rat retinas and are known for functional neuronal properties similar to those of various human neurosensory cells. ¹⁶  

Cells were plated in 12-well plates (Becton Dickinson Labware, Franklin Lakes, NJ) at 2 × 10⁵ cells per well and incubated at 37°C in 5% CO₂ until confluence. Before they were exposed to 7kCh, the cells were incubated in serum-free medium for 24 hours to make them relatively nonproliferating. This simulates the natural human retinal neurosensory cells, which remain in a nonproliferating phase and which are not exposed to the circulation because of the blood-retinal barrier. 

Similarly, ARPE-19 cells (ATCC, Manassas, VA) were grown in a 1:1 mixture (vol/vol) of DMEM and Ham nutrient mixture F-12; (Invitrogen-Gibco), 10 mM 1× nonessential amino acids, 0.37% sodium bicarbonate, 0.058% l-glutamine, 10% fetal bovine serum, and antibiotics (100 U/mL penicillin G, 0.1 mg/mL streptomycin sulfate, 10 μg/mL gentamicin, 2.5 μg/mL amphotericin-B). 

Cell Viability Studies

In the first set of experiments, the R28 cells were incubated for 24 hours with different concentrations—2.5, 5, 10, 20, and 40 μg/mL—of 7kCh, and then cell viability was measured. The vehicle, dimethyl sulfoxide (DMSO), was used to dissolve the 7kCh crystals to make a 10 mg/mL stock solution. ¹⁴ In a second set of experiments, 7kCh was used in 2 different concentrations, 20 and 40 μg/mL. In addition, some wells were pretreated for 1 hour with 20 μM z-VAD-fmk, a pan-caspase inhibitor, or LDL (32 μg/mL) ¹⁰ before exposure to 7kCh. Cells were treated in triplicate in the following groups: (1) R28 cells with 20 and 40 μg/mL 7kCh alone; (2) R28 cells with 20 and 40 μg/mL 7kCh + LDL; (3) R28 cells with 20 and 40 μg/mL 7kCh + z-VAD-fmk; (4) R28 cells with 20 and 40 μg/mL 7kCh, + LDL + z-VAD-fmk; (5) R28 cells with DMSO; (6) untreated R28 cells; and (7) R28 cells with a negative control for z-VAD-fmk. 

Cell Viability Assay

Cell viability assay was performed as described by Narayanan et al. ²⁰ The cells were incubated for 5 minutes with 0.2% trypsin-EDTA after washing with PBS-EDTA; this was followed by centrifugation at 228g for 5 minutes. These were resuspended in 1 mL culture medium, and automated trypan blue dye exclusion assay for cell viability was performed (ViCell Analyzer; Beckman Coulter Inc., Fullerton, CA). Cell viability was measured in two sets of experiments, R28 cells treated with varying concentrations of 7kCh and R28 cells treated with 7kCh with and without LDL and z-VAD-fmk. 

DNA Fragmentation Assay

R28 cells (5 × 10⁶) were plated overnight in 100-mm dishes and then incubated for another 24 hours with 20 μg/mL 7kCh in serum-free medium. DNA was extracted (QIAamp DNA Micro kit; Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Samples were separated by electrophoresis on 2% agarose gels and stained with 5% ethidium bromide. Images were captured with a fluorescence image scanning instrument (FMBIO III; Hitachi, Yokohama, Japan). 

Caspase Detection

R28 cells were grown in 24-well plates in the groups outlined. Caspase-3, -8 and -9 were detected with carboxyfluorescein apoptosis detection kits (FLICA; Immunochemistry Technologies LLC, Bloomington, MN). The FLICA reagent has an optimal excitation range from 488 to 492 nm and an emission range from 515 to 535 nm. Caspase activities were measured with fluorescence image scanning instrument (FMBIO III; Hitachi). Apoptosis was quantified as the amount of green fluorescence emitted from FLICA probes bound to caspases. Nonapoptotic cells appeared unstained, whereas cells undergoing apoptosis fluoresced brightly. The caspase-12 assay was used as described by the manufacturer’s protocol (BioVision, Inc., Mountain View, CA). 

At 24 hours, the wells were rinsed briefly with fresh culture medium, followed by incubation with 300 μL/well FLICA solution in culture medium for 1 hour and washed with phosphate-buffered saline (PBS). In addition to the experimental groups, we analyzed the following control groups: (1) untreated R28 cells without FLICA to exclude autofluorescence from cells; (2) untreated R28 cells with FLICA for comparison of caspase activity of treated cells; (3) untreated R28 cells with negative control (z-FA-fmk) and FLICA as a control for z-VAD-fmk; (4) tissue culture plate wells without cells with buffer alone to represent the background levels; (5) tissue culture plate wells without cells with culture media and DMSO to exclude a cross-reaction of culture media, DMSO with the plastic material of the tissue culture plate or cross-reaction of FLICA with DMSO, and culture media; and (6) R28 cells with DMSO and FLICA to account for any cross-fluorescence resulting from the cross-reaction of untreated cells with DMSO. Additionally, we used 5 μM epicatechin (Sigma-Aldrich, St. Louis, MO), an antioxidant, for caspase-3 inhibition studies on R28 cells and analyzed the following groups: (1) R28 cells with DMSO; (2) R28 cells with 7kCh; and (3) R28 cells with 7kCh + epicatechin. Similarly, ARPE-19 cells were treated in triplicate as the following groups: ARPE-19 cells with DMSO; ARPE-19 cells with 7kCh; and ARPE-19 cells with 7kCh + z-VAD-fmk. 

Statistical Analysis

Data were subjected to statistical analysis by ANOVA (GraphPad Prism, version 3.0; GraphPad Software Inc., San Diego, CA). Newman-Keuls multiple comparison tests were used to compare the data within each experiment. Data were presented as mean ± SEM. Experiments were performed in triplicate. P < 0.05 was considered statistically significant. 

Results

Cell Viability Studies

R28 cells showed a concentration-dependent decrease in cell viability after exposure to 7kCh for 24 hours. Untreated and DMSO-treated cells showed cell viabilities of 85.7% ± 1.4% and 86.7% ± 2.9%, respectively (Fig. 1A) . Cell viabilities were 53.4% ± 4.7% (P < 0.001) and 43.4% ± 3.9% (P < 0.001) at doses of 20 and 40 μg/mL 7kCh, respectively. At the lower concentrations of 2.5, 5, and 10 μg/mL 7kCh, cell viabilities were 89.9% ± 0.3% (P > 0.05), 84.0% ± 1.2% (P > 0.05), and 76.1% ± 2.3% (P < 0.01), respectively. 

R28 cultures showed decreased cell viability after 7kCh treatment, but this loss was partially reversed by LDL and z-VAD-fmk treatments. R28 cells treated with 20 μg/mL 7kCh + LDL showed increased cell viability of 65.1% ± 3.8% compared with R28 cells treated with 7kCh alone (51.7% ± 4.7%; P < 0.01). Similarly, the viability of R28 cells with 20 μg/mL 7kCh + z-VAD-fmk was increased to 59.1% ± 2.1% (P < 0.05) compared with 7kCh alone. R28 cells treated with 7kCh + LDL + z-VAD-fmk showed the greatest increase in cell viability (77.3% ± 1.2%; P < 0.001) compared with 7kCh-treated cultures (51.7% ± 4.7%). 

Caspase-8 Activity

Caspase-8 activity was increased significantly after 7kCh treatment. Moreover, LDL had a protective effect against this increase (Fig. 2) . Untreated R28 cells and DMSO-treated cells showed low caspase-8 activities (6030 ± 996 and 3646 ± 485, respectively). After 7kCh treatment, caspase-8 activity increased significantly (18,570 ± 3399; P < 0.001) compared with untreated samples. When LDL was added to the 7kCh cultures, caspase-8 activity decreased significantly (12,070 ± 1525; P < 0.01). 7kCh-induced caspase-8 activity was significantly decreased when LDL + z-VAD-fmk was added (10,195 ± 3746; P < 0.001) but not when z-VAD-fmk alone was added (18,846 ± 1943; P > 0.05). After treatment with the negative control for z-VAD-fmk, caspase-8 activity was 3386 ± 340. 

Caspase-12 Activity

Significant caspase-12 activity was seen at 4 hours in R28 cell cultures after 7kCh treatment (Fig. 3) . When R28 cultures were examined at 24-hour treatment with 20 μg/mL 7kCh, there was minimal caspase-12 activity (DMSO-treated cultures [10,730 ± 444] vs. 7kCh-treated R28 cultures [8832 ± 461]). This was surprising because at 24 hours in human ARPE-19 cell cultures, 7kCh treatment induced caspase-12 activity. ²¹ We suspected that in R28 cultures, caspase-12 might be activated earlier than 24 hours so we decreased the culture incubation time to 4 hours. In these DMSO-treated cultures, R28 cells had minimal caspase-12 activity (14,079 ± 1080), but it was increased 1.6-fold with 7kCh treatment (22,083 ± 2163; P < 0.01). Interestingly, adding LDL alone, z-VAD-fmk alone, or the combination of LDL + z-VAD-fmk actually further increased caspase-12 activity (29,256 ± 882 [P < 0.001]; 33,046 ± 5701 [P < 0.001]; and 31,412 ± 1187 [P < 0.001]) compared with DMSO-treated controls (14,079 ± 1080). 

Caspase-9 Activity

Caspase-9 activity was not increased after 7kCh treatment (Fig. 4) . Values for the control untreated cells, control DMSO-treated cells, and negative controls for z-VAD-fmk were 3366 ± 78, 2184 ± 1088, and 579 ± 70, respectively. Values for 7kCh-treated cultures, 7kCh + z-VAD-fmk–treated cultures, and 7kCh + z-VAD-fmk + LDL–treated cultures were –1250 ± 882, 4348 ± 561, –4407 ± 718, respectively, and were similar to control cultures. 

Caspase-3 Activity

Caspase-3 activity in R28 cells increased significantly after treatment with 7kCh. Values for untreated cells and DMSO-treated cultures were 5988 ± 623 and 2575 ± 862, respectively (Fig. 5A) . Cells treated with 7kCh alone showed mean fluorescence of 34,747 ± 4005, but values were 29,304 ± 3321, 30,493 ± 6621, and 30,333 ± 2060 for 7kch + LDL, 7kCh + z-VAD-fmk, and 7kCh + LDL + z-VAD-fmk, respectively. Although caspase-3 activation is a hallmark of apoptosis, we wanted to verify with the DNA fragmentation assay that the R28 cells were indeed undergoing apoptosis. Figure 5Bshows that after treatment with 7kCh for 24 hours, the DNA laddering patterns seen on the 2% agarose gel electrophoresis had the distinct pattern of 200- to 500-bp fragmented DNA characteristics of apoptotic cells. In contrast, untreated R28 cells lacked this fragmented DNA pattern. In Figure 5A , we were surprised that neither LDL nor z-VAD-fmk decreased caspase-3 activity significantly in R28 cultures. Therefore, to demonstrate that the z-VAD-fmk inhibitor was working effectively, human ARPE-19 cell cultures were established, and experiments were repeated (Fig. 5C) . 7kCh-treated ARPE-19 cells had increased caspase-3 activity compared with DMSO-treated cultures (12,311 ± 998 vs. 6086 ± 1576). Caspase-3 activity decreased in the 7kCh + z-VAD-fmk cultures (5512 ± 243) compared with 7-kCh–treated cultures (P < 0.001). In another experiment, 5 μM epicatechin significantly inhibited the 7kCh-induced caspase-3 activity in R28 cells (Fig. 5D) . With the addition of epicatechin, mean fluorescence was reduced to 26,805 ± 1693 compared with mean fluorescence of 35,719 ± 6895 with 7kCh alone (P < 0.05). 

Discussion

The present study examined the pathways involved in 7kCh-induced apoptosis in the R28 cell line, a heterogeneous population of neuroretinal progenitor cells. Our data showed that the endoplasmic reticulum pathway (caspase-12) and the extrinsic Fas/Fas ligand (Fas-L) pathway (caspase-8) were activated by 7kCh treatment, whereas the mitochondrial (caspase-9) pathway was not involved. In addition, pretreatment with LDL partially blocked caspase 8 activity (P < 0.001) but not caspase-3 or caspase-12 activities. Additionally, DNA fragmentation that is characteristics of apoptosis was seen in the 7kCh-treated R28 cultures. The findings support our hypothesis that oxysterols such as 7kCh can lead to decreased cell viability through caspase-dependent apoptosis and can be environmental oxidative stressors that lead to retinal cell atrophy. 

Cell viability assays demonstrated that 7kCh was cytotoxic to the R28 cells in a concentration-dependent fashion. This agrees with 7kCh-induced loss of cell viability found in human aorta smooth muscle cells, ²² human umbilical vein endothelial cells, ²³ and human ARPE-19 cells. ^{14 21} In general, R28 cells seem to be more sensitive to 7kCh treatment than ARPE-19 cells because an earlier study showed that R28 cells responded to oxysterols with increased levels of reactive oxygen species (ROS), whereas no immediate production of ROS was detected with ARPE-19 cells after oxysterol-induced toxicity. ¹⁴ Loss of cell viability can be attributed to apoptosis or necrosis. Caspase-3 is a key effector in the apoptosis pathway, and its activation signals the full commitment to disassembly of the cell. Therefore, we used caspase-3 activation analyses and the DNA fragmentation assay to confirm the presence of apoptosis in R28 cells after exposure to 7kCh. Our findings were similar to those of Ong et al., ¹⁴ who used DNA fragmentation assay and Hoechst staining to demonstrate that the loss of cell viability in 7kCh-treated R28 cells resulted in part from apoptosis. 

In our study, LDL blocked the loss of cell viability and caspase-8 activity in 7kCh-treated R28 cultures. Our findings of an LDL protective effect against apoptosis agree with other reports of similar blocking by other oxysterols, ²⁴ the hydroxylated form of 7kCh, 27OH7kCh, ¹³ and minimally oxidized LDL. ²⁵ The protective effects of LDL may be attributed to competitive inhibition of 7kCh with nonoxidized LDL at the LDL receptor sites on R28 cells. It is also possible that high levels of LDL in the culture medium would offer a decoy protein to be oxidized and therefore protect the cells. However, Boullier et al. ²⁵ showed that minimally oxidized LDL offsets the apoptotic effects of extensively oxidized LDL in macrophages not through involvement of the LDL receptor but rather through the PI3K/Akt-dependent pathway, which promotes cell survival. This PI3K pathway is also involved in antiapoptotic protection by the unoprostone isopropyl metabolite in serum-deprived R28 cell cultures. ²⁶ Further studies are required to determine the mechanism of protection from 7kCh-induced apoptosis in R28 cell cultures. 

Caspase 8

Caspase-8 activity was increased in 7kCh-treated R28 cells compared with DMSO-treated cultures. This is in agreement with previous studies showing that Fas-mediated apoptosis played a role in oxidant-induced cell death in human RPE cells. ^{21 27} Treatment with LDL alone (P < 0.01) or LDL + z-VAD-fmk (P < 0.001) decreased the caspase-8 activity compared with the cultures treated with 7kCh alone. However, the blocking of caspase-8 activation was not complete and represented only partial inhibition. 

Treatment with z-VAD-fmk did not block caspase-8 activation. This finding is contrary to other human RPE cell culture studies showing reversal of caspase-8 activation with z-VAD-fmk or an anti–Fas antibody (ZB4). ²⁷ Our data suggest that in R28 cultures, which have heterogeneous cell populations, ¹⁶ the pan-caspase inhibitor alone is not adequate to block caspase-8 activation; rather, LDL treatment is more effective in blocking the caspase-8 activity. R28 cultures are retinal progenitor cells that have markers for retinal ganglion cells, horizontal cells, amacrine cells, glial cells, photoreceptors, and bipolar cells. It may be that, unlike single-cell type cultures blocked by the pan-caspase inhibitor z-VAD-fmk, the R28 multiple population cultures may have unique properties that make it more difficult to inhibit 7kCh-induced apoptosis. 

Caspase 12

7kCh-induced cell death also involved endoplasmic reticulum stress and, hence, caspase-12 activation. Inactive caspase-12 is associated with the endoplasmic reticulum cytosolic face that, once activated, can trigger caspase-3 activation. ²⁸ In R28 cultures, caspase-12 activation occurred within 4 hours of 7kCh treatment and was minimal at 24 hours. The finding of caspase-12 activation by 7kCh treatment has also been found in ARPE-19 cultures ²¹ and human promonocytic U937 cells. ²⁹ Surprisingly, caspase-12 activation was not blocked by LDL, with or without z-VAD-fmk. This suggests that caspase-12 may have unique pathways for inhibition that are unknown at this time. It also suggests that caspase-8 and caspase-12 pathways may be independent of each other because LDL treatment and LDL + z-VAD-fmk treatment partially blocked caspase-8 activation but had the opposite effect with increased caspase-12 activation. This is similar to 661W retinal photoreceptor cells, which showed different inhibition responses for caspase-12 compared with caspase-3 and caspase-9. ³⁰ In that study, caspase-12 activation was mediated by m-calpain activation, whereas other caspase activities were not. The interactions between caspase-8 and caspase-12 pathways in R28 multiple population cultures must be investigated further. 

Caspase 3

7kCh-treated R28 cultures showed a 5.1-fold increase of caspase-3 activity compared with DMSO-treated cultures, consistent with findings of human ARPE-19 cells and promonocytic leukemia U937 cells that showed increased caspase-3 activation after treatment with 7kCh. ^{21 31} Interestingly, not all cells respond to 7kCh by apoptosis, but some, such as the human bladder cancer ECV304 cell line, undergo necrotic cell death without caspase 3 activation. ³² We were surprised to find that adding LDL with or without z-VAD-fmk did not significantly decrease caspase-3 activity in the R28 cultures. To verify that the z-VAD-fmk inhibitor was working, we repeated the experiment using human ARPE-19 cells and indeed showed that z-VAD-fmk blocked the 7kCh-induced caspase-3 activity in the ARPE-19 cells but not in the R28 cultures. Sanvicens et al. ³⁰ also found that in cone photoreceptor cultures, the inhibition of caspases did not prevent apoptosis but only resulted in a temporary delay. They showed m-calpain activation was in parallel with caspase activities, indicating that more than one execution pathway was available to cone photoreceptors. In human U937 promonocytic leukemia cells, z-VAD-fmk treatment partially blocked apoptosis but not cell death. ³³ Similarly, in R28 cultures, unidentified alternative pathways may be involved with 7kCh-induced apoptosis. In addition, we speculate that this failure to block the 7kCh-induced caspase-3 activity may be related to the multiple population properties of the R28 cultures in which some cell types might respond to z-VAD-fmk but others would not be affected. The net effect would be a slight decline in caspase-3 activation but not robust inhibition, which is what we found in our experiments. Moreover, progenitor properties of the R28 cells may play a role in the lower than expected inhibition of the caspase-3 activation. In another experiment, the antioxidant flavonoid compound epicatechin significantly inhibited 7kCh-induced caspase-3 activity in R28 cells, suggesting a causative role for oxidative stress in apoptosis caused by 7kCh. This finding was similar to a report on murine macrophage J774A.1 cells. ³⁴ Clearly, further studies are needed to clarify the lack of caspase-3 inhibition with either LDL or z-VAD-fmk and the antiapoptotic effect of antioxidant flavonoids in R28 cells. 

In summary, in R28 cell cultures, 7kCh-induced apoptosis involves the caspase-3 along with the receptor-mediated caspase-8 and the endoplasmic reticulum stress-induced caspase-12 pathways. This supports our hypothesis that oxysterols can damage retina cells through apoptosis and may play a role in retinal diseases. LDL partially blocked 7kCh-induced caspase-8 activity but increased caspase-12 activities, suggesting that the caspase-8 and caspase-12 pathways are independent of each other. The z-VAD-fmk inhibitor blocked caspase-3 activity in ARPE-19 cultures but not in R28 cultures, suggesting that the inhibition of apoptosis appears more complex when multiple cell populations (i.e., R28 cell line cultures) are involved compared with the single-cell population of ARPE-19 cultures. 

 Supported by the Discovery Eye Foundation, the Henry L. Guenther Foundation, the Iris and B. Gerald Cantor Foundation, the Skirball Molecular Ophthalmology Program, the Nike Foundation, and Research to Prevent Blindness Foundation.

 Submitted for publication August 1, 2006; revised October 22, 2006; accepted January 23, 2007.

 Disclosure: A. Neekhra, None; S. Luthra, None; M. Chwa, None; G. Seigel, None; A.L. Gramajo, None; B.D. Kuppermann, None; M.C. Kenne y, None

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

 

Figure 1.

 

(A) R28 cells showed a dose-dependent decrease in cell viability after 7kCh treatment for 24 hours compared with DMSO-treated cultures. Cell viabilities for cultures treated with 2.5 and 5 μg/mL 7kCh were similar to DMSO-treated and untreated cultures. Cell viability was decreased significantly with 10 μg/mL (P < 0.01), 20 μg/mL (P < 0.001), and 40 μg/mL (P < 0.001) 7kCh. 7kCh, 7 ketocholesterol; DMSO, dimethyl sulfoxide; z-VAD-fmk, pan-caspase inhibitor. **Statistically significant (P <0.01). ***Statistically significant (P <0.001). (B) LDL alone and in combination with z-VAD-fmk partially blocked the 7kCh-induced loss of cell viability at 24 hours. R28 cells treated with 20 μg/mL 7kCh showed decreased cell viability compared with the DMSO-treated cultures (P < 0.01). This decrease was partially blocked by the addition of LDL (P < 0.01), z-VAD-fmk (P < 0.05), and the combination of LDL+ z-VAD-fmk (P < 0.001) compared with the 7kCh-treated cultures. *Statistically significant (P < 0.05). **Statistically significant (P <0.01). ***Statistically significant (P <0.001).

Figure 2.

 

R28 cells treated for 24 hours with 20 μg/mL 7kCh had increased caspase-8 activity compared with DMSO-treated cultures. DMSO-treated and untreated R28 cultures showed minimal caspase-8 activity. After treatment with 20 μg/mL 7kCh, R28 cultures had an increased mean fluorescence compared with DMSO-treated cultures (P < 0.001). Adding LDL partially blocked caspase-8 activation (P < 0.01), as did the combination of LDL + z-VAD-fmk (P < 0.001) compared with 7kCh-treated cultures. z-VAD-fmk alone did not block caspase-8 activity. *Statistically significant (P < 0.05). **Statistically significant (P <0.01). ***Statistically significant (P <0.001).

Figure 3.

 

Caspase-12 activity was increased by R28 cell cultures after 7kCh treatment. DMSO-treated R28 cultures had minimal caspase-12 activity. After treatment with 7kCh, caspase-12 activity increased significantly. Adding LDL alone, z-VAD-fmk alone, or the combination of LDL+ z-VAD-fmk led to elevated caspase-12 activity compared with DMSO-treated control cultures (P < 0.001). At 24 hours, minimal caspase-12 activity was seen in the 7kCh-treated R28 cultures. *Statistically significant (P < 0.05). **Statistically significant (P <0.01). ***Statistically significant (P <0.001).

Figure 4.

 

Caspase 9 was not activated by 20 μg/mL 7kCh treatment for 24 hours in R28 cultures. After treatment with 20 μg/mL 7kCh, caspase-9 activity was similar to control, as was 7kCh+ LDL, 7kCh + z-VAD-fmk, and 7kCh + z-VAD-fmk + LDL. *Statistically significant (P < 0.05). **Statistically significant (P <0.01). ***Statistically significant (P <0.001).

Figure 5.

 

Caspase-3 activity increased in the R28 and ARPE-19 cultures after 7kCh treatment. z-VAD-fmk blocked this increase in ARPE-19 cultures but had little effect in R28 cell cultures. However, 5 μM epicatechin inhibited 7kCh-induced caspase-3 activity in R28 cells significantly at 24 hours. (A) R28 cultures treated with 7kCh significantly increased caspase-3 activity compared with DMSO-treated cultures. Untreated and DMSO-treated R28 cultures had minimal caspase-3 activity. After 20 μg/mL 7kCh treatment, caspase-3 activity increased significantly (P < 0.001) and remained high after the addition of LDL, z-VAD-fmk, and the combination of LDL + z-VAD-fmk. No significant inhibition of 7kCh-induced caspase-3 activity was seen in any of these groups. *Statistically significant (P < 0.05). **Statistically significant (P <0.01). ***Statistically significant (P <0.001). (B) DNA extracts from untreated and 7kCh-treated cultures were separated on a 2% agarose gel and stained with ethidium bromide. The amount of DNA fragmentation was greater in the 7kCh-treated cultures than in the untreated samples. Ctl, untreated; M, 100-bp DNA ladder. (C) Caspase-3 activity was increased by ARPE-19 cell cultures after 7kCh treatment. DMSO-treated ARPE-19 cultures had minimal caspase-3 activity compared with 7kCh-treated cultures (P < 0.001). In 7kCh + z-VAD-fmk–treated cultures, caspase-3 activity was decreased compared with 7kCh alone (P < 0.001). (D) In R28 cell cultures, increased caspase-3 activity 24 hours after treatment with 7kCh alone was significantly reduced in the wells, which were incubated with 5 μM epicatechin for 1 hour before 7kCh was added. *Statistically significant (P < 0.05).