σ-1 Receptors Protect RGC-5 Cells from Apoptosis by Regulating Intracellular Calcium, Bax Levels, and Caspase-3 Activation

Abstract

purpose. σ-1 Receptor ligands prevent neuronal death associated with glutamate excitotoxicity both in vitro and in vivo. However, the molecular mechanism of the neuroprotective effect remains to be elucidated. The present study was undertaken to determine whether σ-1 receptor agonists provide neuroprotection by decreasing glutamate-induced calcium mobilization and preventing apoptotic gene expression.

methods. Cell death was measured by using a calcein-AM/propidium iodide cell-survival assay. Western blot analysis determined the expression levels of Bax in normal RGC-5 cells. Caspase-3 activation after glutamate treatment was determined with a carboxyfluorescein caspase-3 detection kit. Glutamate-induced intracellular calcium mobilization was measured by using ratiometric calcium imaging.

results. σ-1 Receptor-overexpressing RGC-5 (RGC-5-S1R) cells had lower glutamate-induced intracellular calcium mobilization than did normal RGC-5 cells, and the σ-1 receptor ligand (+)-SKF10047 reduced the glutamate calcium response in normal and RGC-5-S1R cells. (+)-SKF10047 protected RGC-5 cells from glutamate-induced cell death, and the RGC-5-S1R cells showed a significant resistance to glutamate-induced apoptosis compared with the control RGC-5 cells. BD1047, a σ-1 receptor antagonist, blocked the protective effect of (+)-SKF10047. Western blot analysis showed that (+)-SKF10047 inhibited the increase in Bax after glutamate treatments. Glutamate-mediated cell death involved activation of caspase-3, and σ-1 receptor activation prevented an increase in caspase-3 expression.

conclusions. The results suggest that σ-1 receptors regulate intracellular calcium levels and prevent activation of proapoptotic genes, thus promoting retinal ganglion cell survival. The σ-1 ligands appear to be neuroprotective and are a potential target for neuroprotective therapeutics.

The σ-1 receptor was first cloned from guinea pig liver by Hanner et al. ¹ Although first described as one of the subtypes of opioid receptors, further studies have led to the distinction between σ receptors from opioid receptors. ² The biology of σ-1 receptors is poorly understood. The σ-1 receptor has been studied extensively in the central nervous system, ³ but it is also recognized to be overabundant in the eye. ^{4 5 6} Two types of σ receptors have been described: those with a molecular mass of ∼25 kDa and those with a molecular mass of ∼18 to 21 kDa. ^{1 7} Hayashi and Su ⁸ have shown that σ-1 receptors are localized on both the endoplasmic reticulum (ER) and the plasma membrane. In addition, they have shown that σ-1 receptors translocate from the ER-associated reticular network to the cell periphery on stimulation. σ-1 receptors are also associated with lipid raft microdomains and control the ER lipid compartmentalization and export, ⁸ as well as oligodendrocyte differentiation. ⁹  

In vitro exposure of cultured rat brain neurons to selective σ receptor ligands protects cells against glutamate or NMDA exposure. ¹⁰ The σ-2 receptors have been shown to be a biomarker for solid tumor proliferation. ¹¹ In addition, the σ-2 receptors have been shown to be associated with cancer cell proliferation and survival. ^{12 13} Glutamate, an NMDA receptor agonist, when present in excess, can lead to accumulation of intracellular calcium via NMDA receptor activation, which is a well-documented process leading to neuronal death or injury. ^{14 15} Recently, the σ-1 receptors have also been shown to lower intraocular pressure (IOP) ^{16 17} and protect retinal ganglion cells (RGCs) from stress ¹⁸ and, in the lens, to modulate cell growth and pigmentation. ⁶ Knockdown of σ-1 receptors in the human lens cells induces apoptosis. ¹⁹ Caspases, a unique class of aspartate-specific proteases, are the central components of the apoptotic response. The release of cytochrome c from mitochondria into the cytosol leads to the activation of the initiator caspase-9, which in turn activates the effector caspase-3. Once activated, caspase-3 is responsible for the proteolytic cleavage of a broad spectrum of cellular targets, which ultimately leads to cell death. ^{20 21} It has recently been shown that glutamate-induced excitotoxicity is mediated through the activation of caspase-3 ²² and the upregulation of Bax. ²²  

In the current study, the σ-1 receptor agonist (+)-SKF10047 protected RGCs from glutamate-induced apoptosis. The protective effect of the σ-1 receptor agonist (+)-SKF10047 was prevented by the σ-1 receptor-selective antagonist BD1047. Glutamate-mediated RGC death appeared to be signaled by calcium, followed by an upregulation of Bax and caspase-3 activation, leading to cell death. The conclusion is that σ-1 receptors are thus neuroprotective. 

Materials and Methods

Glutamate was purchased from Sigma-Aldrich (St. Louis, MO); (+)-SKF10047, BD1047, and MK801 from Tocris Bioscience (Ellisville, MO); rabbit polyclonal NMDAR1 from Cell Signaling Technology (Danvers, MA); and rabbit polyclonal Bax (N-20; catalog number: sc-493) antibody from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit anti-σ-1 receptor was a gift from Teruo Hayashi (Intramural Research Program, National Institute of Drug Abuse [IRP/NIDA], Baltimore, MD). Rabbit normal IgG was purchased from Sigma-Aldrich. Secondary antibodies, including donkey anti-rabbit IgG and donkey anti-mouse IgG conjugated to HRP, were purchased from GE Healthcare (Piscataway, NJ); fluorescent probes including goat anti-rabbit Alexa 488, goat anti-rabbit Alexa 633, and fura-2-AM from Invitrogen-Molecular Probes (Eugene, OR); and the carboxy fluorescein caspase-3 detection kit (Apologix, catalog number: FAM200-1) from Cell Technology, Inc. (Mountain View, CA). 

RGC-5 Cell Culture and Differentiation

The RGC-5 cell line was developed specifically to establish a permanent rat RGC culture. ²³ RGC-5 cells were grown in Dulbecco’s modified Eagle’s medium (cat no. 23700-040; DMEM; Invitrogen-Gibco, Grand Island, NY), supplemented with 10% heat-inactivated fetal bovine serum (cat no. 26140-079; Invitrogen-Gibco), 100 U/mL penicillin, and 100 μg/mL streptomycin. RGC-5 cells were cultured in growth medium and incubated at 37°C in 5% CO₂ and air. 

In all the experiments, the cells were differentiated by using human nonpigmented ciliary epithelial (HNPE) cell-conditioned medium collected when cells were 100% confluent from T-150 flasks (Dauphin R et al., manuscript in preparation). Briefly, 24 hours after the RGC-5 cells were seeded, the DMEM was removed from the culture dishes and replaced by the HNPE-conditioned medium (50 mL) and 1 mL complete DMEM (10% heat-inactivated fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin). The cultures dishes were then incubated at 37°C in 5% CO₂ and air for 48 hours, to allow the cells to differentiate (Fig. 1) . 

Preparation of Stably Transfected σ-1 Receptor RGC-5 Cells

The plasmid pSPORT-σ-l-R cDNA ²⁴ was the kind gift of Vadivel Ganapathy (Medical College of Georgia, Augusta, GA). 

Large-Scale Plasmid Preparation.

Escherichia coli-competent strain SURE-2 cells were transformed with a plasmid pSPORT-σ-1-R. After transformation, the cells were plated on ampicillin-containing Luria-Bertani (LB) plates. Many colonies were obtained after overnight incubation. Colonies were picked up and grown in LB medium for large-scale preparation of plasmid DNA. Plasmid DNA was purified by the equilibrium cesium chloride gradient method. After plasmid preparation, the purity of the DNA was checked on agarose gels (data not shown). 

Permanent Transfection.

A lipophilic method (Lipofectamine 2000; Invitrogen, San Diego, CA) was used for transfection of RGC-5 cells, as instructed by the manufacturer. Because RGC-5 cells are neomycin (G418)-resistant, we performed a cotransfection using pSPORT-σ-1R and psiRNA-hH1GFPzeo (InvivoGen, San Diego, CA). 

The σ-1 receptor empty vector transfection was performed with psiRNA-hH1GFPzeo alone. After transfection, zeocin-resistant clones were selected, and the expression of σ-1 receptors was assessed by Western blot analysis with anti-σ-1 receptor rabbit polyclonal antibody (1:500 dilution; Fig. 2 ). 

Preparation of Cell Lysates and Western Blot Analysis

Normal RGC-5 cells lysates were prepared as described by Bu et al. ²⁵ Protein concentrations of cell lysates were determined with a DC protein kit (Bio-Rad Laboratories, Hercules, CA). Samples of protein (20 μg) were run on an SDS-polyacrylamide gel and immunoblotted by standard methods, as described by Laemmli ²⁶ and Towbin et al., ²⁷ with rabbit polyclonal NMDAR1 antibody (1:500 dilution; for NMDA receptors expression), and rabbit polyclonal Bax (N-20) antibody (1:500 dilution; for Bax expression). The blots were developed with an ECL kit (GE Healthcare). β-Actin was used as the control for equal loading. 

Primary Rat RGC Isolation

All procedures were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the Institutional Animal Care and Use Committee at the University of North Texas Health Science Center. RGCs were purified by a modification ²⁸ of immunopanning with antibodies against Thy 1.1 specific for RGCs. Briefly, Wistar Kyoto rats (3–4 weeks old, either sex) were purchased from Harlan Sprague-Dawley (Indianapolis, IN) and were anesthetized with isoflurane and then decapitated. The eyes were enucleated and the retinas gently peeled off with fine forceps and placed in sterile phosphate-buffered saline (PBS). The retinas were collected in DMEM and incubated at 37°C in 3 mL of Dispase for 1 hour. After a 1-hour incubation, the retinas were gently washed once with complete neuronal cell growth medium (Neurobasal; Invitrogen). The RGCs were obtained by trituration of the retina in neuronal growth medium, with a fire-polished Pasteur pipette. 

Panning dishes (Falcon; BD Biosciences, Bedford, MA) were incubated with goat anti-mouse IgG antibodies (2 μg/mL; Sigma-Aldrich) in Tris-HCl buffer (pH 9.5) for 24 hours at 4°C. The dishes were then washed three times with phosphate-buffered saline (PBS) before each of the subsequent steps. Incubation with antibodies against Thy1.1 (monoclonal antibody 1.5–2.5 μg/mL; cat no. MAB1406; Chemicon International, Temecula, CA) was performed for at least 2 hours at room temperature in PBS. To prevent nonspecific binding of cells, dishes were then incubated with 0.2% bovine serum albumin in PBS for 20 minutes at room temperature. Approximately 5 mL of cell suspension was added per dish and incubated for 20 minutes at 37°C. The dishes were gently swirled every 5 minutes to ensure access of all RGCs to the surface of the plate. For removal of nonadherent cells, the dishes were washed repeatedly with PBS and swirled moderately until only adherent cells remained. Washing was monitored under the microscope. RGCs were removed from the panning dish with trypsin (0.0625%) solution. After centrifugation at 200g for 5 minutes, purified RGCs were suspended in culture medium. The cell-suspension collected was seeded (500 μL) on coverslips coated first with poly-d-lysine (overnight). The coverslips were covered with 2 mL of complete neuronal medium (Neurobasal, supplemented with B27 and penicillin-streptomycin; Invitrogen-Gibco), brain-derived neurotrophic factor (BDNF), ciliary neurotrophic growth factor (CTNF; PHC 7074; Invitrogen-BioSource), and forskolin (F6886; Sigma-Aldrich). Cell culture dishes were subsequently incubated in a 5% CO₂ incubator at 37°C for up to 7 days. Isolated RGCs were characterized by immunocytochemistry for a normally expressed RGC marker, Thy-1. ²⁹  

Immunofluorescence Microscopy

Primary RGCs (7 DIV) were grown on coverslips and subsequently fixed with 4% paraformaldehyde and blocked with blocking solution (5% BSA+5% normal goat serum) for 1 hour at room temperature. The blocking buffer was discarded, and the coverslips were washed three times with 1× PBS, before the addition of the Thy1.1 monoclonal antibody (1/500 dilution) with 1% normal goat serum and 5% bovine serum albumin (Sigma-Aldrich). The incubation was performed overnight at 4°C. Coverslips were washed three times, 1:500 of secondary antibody (Alexa Fluor 633; Invitrogen-Molecular Probes) was added, and the coverslips were incubated for 1 hour in the dark. After the incubation, the coverslips were washed again three times with 1× PBS. The coverslips were mounted on glass slides (Prolong Gold antifade reagent with DAPI; Invitrogen) and allowed to dry for 20 minutes in the dark. The cells were viewed with a confocal laser scanning microscope (LSM 510; Carl Zeiss Meditec, Inc., Dublin, CA). 

Intracellular Calcium ([Ca²⁺]_i) Measurement

Glutamate-induced [Ca²⁺]_i mobilization in normal RGC-5, green fluorescence protein transfected RGC-5 (RGC-5-GFP) cells (σ-1 receptor empty vector), and σ-1 receptor-overexpressing RGC-5 cells (RGC-5-S1R) was measured at 37°C by the ratiometric technique with fura-2-AM (excitation at 340 and 380 nm, emission at 510 nm) according to Prasanna et al. ³⁰ (Diaphot microscope; Nikon, Tokyo, Japan, using Metafluor software; Universal Imaging, West Chester, PA). [Ca²⁺]_i in nanomolar was calculated by using the Grynkiewicz equation. ³¹ In calcium imaging studies, the cells were pretreated with the NMDA receptor antagonist, MK801, and σ-1 receptor antagonist for 30 minutes before stimulating them with glutamate. 

Calcein-AM/Propidium Iodide Cell-Survival Assay

Cell viability was determined by using a calcein/propidium iodide (cat. no. C3099; Invitrogen-Molecular Probes) dual-staining assay. The calcein/propidium iodide assay relies on the intracellular esterase activity within living cells, through which the calcein-AM, a cell-permeable fluorogenic esterase substrate, hydrolyzes to a green fluorescent product, calcein. Living cells show green fluorescence, and dead cells are stained red by the cell-impermeant propidium iodide. Propidium iodide is a nucleic acid stain normally used as a counterstain in multicolor fluorescence techniques to identify nuclei showing apoptotic changes. RGC-5 cells were seeded on coverslips and treated with glutamate (1 mM) and glycine (10 μM) for 4 days, with or without the σ-1 receptor ligands (+)-SKF10047, 1 μM; BD1047, 3 μM; and MK801, 10 μM. After the treatment, the culture medium was removed, and the coverslips were rinsed with 1× PBS. Then, 2 mL of 2 μM calcein and 2 μg/mL propidium iodide in 1× PBS was added to each culture well. The culture dishes with the cells were incubated at 37°C for 30 minutes, and fluorescence was measured (Microphot FXA digital fluorescent microscope; Nikon). 

A 1-mM concentration of glutamate was chosen after a concentration dose-response cell death study by using the calcein/propidium iodide dual-staining survival assay. The following concentrations of glutamate (50, 100, 250, and 500 μM and 1 mM) were used to determine the maximum number of dead RGC-5 cells. The 1 mM glutamate dose caused the highest rate of cell death (Fig. 3) . Three different types of RGC-5 cells (normal, GFP-transfected, and σ-1 receptor overexpressing) were used in this part of the study. 

Caspase-3 Activation by Glutamate Treatment

Conversion of procaspase-3 to active caspase-3 was assessed by the carboxyfluorescein caspase detection kit (Apologix; cat. no. FAM200-1; Cell Technology, Inc.). Caspase-3 activation was assessed as instructed by the manufacturer. This kit is based on carboxyfluorescein labeled fluoromethyl ketone (FMK)-peptide inhibitors of caspases. These inhibitors are cell permeable and noncytotoxic. Once inside the cell, the inhibitor binds covalently to the active caspase. ³² Cells that contain bound inhibitor can be analyzed by fluorescence microscopy. In this experiment both the RGC-5 line and primary rat RGCs were used. The cells were analyzed using a confocal laser scanning microscope (LSM 410; Carl Zeiss Meditec, Inc.). 

Results

Expression of σ-1 Receptors in RGC-5 Cells

The σ-1 receptors are expressed in many organs, including the eye. ⁴ In this study, we used Western blot analysis to examine the expression of σ-1 receptors. The Western blot results showed the expression of σ-1 receptors in normal RGC-5 cells, RGC-5-GFP transfected (σ-1 receptor empty vector) cells, and σ-1 receptor-overexpressing RGC-5 cells (clone 4, see the Methods section for the preparation of stably transfected RGC-5 cells; Fig. 2 , lane 3). In addition, the Western blot data showed that normal RGC-5 cells and RGC-5-GFP-transfected cells expressed basal σ-1 receptors (Fig. 2 , lanes 1, 2). 

Characterization of the Isolated Primary RGCs

Isolated RGCs were characterized by immunocytochemistry for detection of a normally expressed RGC marker, Thy-1 (Fig. 4) . 

Low Calcium Response to Glutamate of σ-1 Receptor-Overexpressing RGC-5 Cells Compared with σ-1 Receptor Empty Vector RGC-5 Cells

σ-1 receptors have been shown to inhibit NMDA receptors in neurons. ^{3 33 34} We first examined the effect of glutamate and the σ-1 receptor agonist (+)-SKF10047 on [Ca²⁺]_i mobilization. The glutamate-mediated increase in [Ca²⁺]_i was concentration dependent (Table 1) . The σ-1 receptor agonist (+)-SKF10047, dose-dependently significantly inhibited glutamate-induced [Ca²⁺]_i mobilization (Table 2) . To investigate further the effect of σ-1 receptors on NMDA receptors in the RGCs, we stably overexpressed σ-1 receptors in RGC-5 cells by using the σ-1 receptor expression construct pSPORT-σ-l-R plasmid and compared the glutamate-induced [Ca²⁺]_i mobilization between transfected and nontransfected RGC-5 cells. [Ca²⁺]_i mobilization in RGC-5 cells was measured by a ratiometric technique with fura- 2-AM (Invitrogen-Molecular Probes). The σ-1 receptor agonist (+)-SKF10047 (1 μM) inhibited the glutamate-induced calcium mobilization peak (Table 3)from 1086 ± 102 to 502 ± 43 nM (P < 0.05) in σ-1 receptor empty vector RGC-5 cells and from 630 ± 40 to 199 ± 8 nM (P < 0.05) in σ-1 receptor-overexpressing RGC-5 cells (Table 3) . In σ-1 receptor-overexpressing RGC-5 cells, the glutamate response was lower than in the σ-1 receptor empty vector RGC-5-GFP cells glutamate response (Table 3) . The σ-1 receptor antagonist BD1047 did not have an inhibitory effect on its own with respect to glutamate-induced calcium mobilization, but was able to block the inhibitory effect of (+)-SKF10047 on glutamate-induced calcium mobilization (Table 4) . 

To test whether the lower glutamate response of σ-1 receptor-overexpressing RGC-5 cells was due to the downregulation of NMDA receptors, we performed a Western blot analysis with a rabbit polyclonal NMDAR1 antibody. The results of the Western blot data showed that the NMDA receptor levels were not affected (Fig. 5) . This finding suggests that σ-1 receptor-mediated NMDA receptor inhibition occurred by other mechanisms but not by downregulation of the NMDA receptor expression level. 

σ-1 Receptor Agonist (+)-SKF10047 Inhibition of Glutamate-Induced Excitotoxicity in RGC-5 Cells

The σ-1 receptor ligands have been shown to protect neurons by preventing the neurotoxicity associated with central nervous system (CNS) injury and neurodegenerative disorders. ³⁵ In addition, the neuroprotective effects of σ-1 receptor ligands are thought to include modulation of NMDA receptors, ^{36 37} attenuation of postsynaptic glutamate-evoked calcium influx, ^{35 38} depression of neuronal responsivity to NMDA receptor stimulation, and reduction of nitric oxide production. ¹⁸ We therefore studied the effect of high glutamate concentrations on RGC-5 cell death. 

Calcein-AM/Propidium Iodide Cell-Survival Assay

To assess the consequences of acute exposure to high levels of glutamate, normal RGC-5, RGC-5-GFP (σ-1 receptor empty vector), and σ-1 receptor-overexpressing RGC-5 cells, were exposed to 1 mM glutamate, with or without the σ-1 receptor agonist (+)-SKF10047 and antagonist BD1047 and the NMDA receptor antagonist MK801. Glycine (10 μM), an NMDA receptor coactivator, was used in each experiment. The cells were then processed for determination of apoptosis by using calcein/propidium iodide double staining. There were very few propidium iodide-positive cells in the control cultures of the three types of RGC-5 cells (Figs. 6I 7I 8I) . However, treatment with 1 mM glutamate increased the number of propidium-positive cells considerably in normal RGC-5, RGC-5-GFP, and RGC-5-S1R (91.19% ± 1%, 90.00% ± 1%, and 40.08% ± 2%, respectively; Figs. 6II 7II 8II ). The σ-1 receptor agonist (+)-SKF10047 reduced the glutamate-induced cell death in normal RGC-5, RGC-5-GFP, and RGC-5-S1R cells to only 8.31% ± 1%, 12.75% ± 2%, and 13.53% ± 5% (Figs. 6III 7III 8III) , respectively. To test whether cell protection was being mediated specifically by σ-1 receptor activation, we exposed the RGC-5 cells to BD1047, a σ-1 receptor antagonist. BD1047 had little effect on RGC-5 cell death on its own (Figs. 6V 7V 8V) , but was able to block the protective effect of (+)-SKF10047 (Figs. 6VI 7VI 8VI) . The reduction in the percentage of RGC-5 cells labeled with propidium iodide by MK801, a specific antagonist of NMDA receptors also confirmed that glutamate-induced RGC-5 cell death was mediated through the activation of the NMDA receptor (Figs. 7VIII 8VIII) . These data suggest that acute exposure to high levels of glutamate is toxic to retinal ganglion RGC-5 cells. In addition, the calcein/propidium iodide survival assay showed that σ-1 receptor-overexpressing RGC-5 cells had a significant increased resistance to glutamate-induced RGC-5 cell death (Figs. 8II 9C) . The calcein/propidium iodide fluorescence image quantitative data are shown for normal RGC-5 (Fig. 9A) , RGC-5-GFP (Fig. 9B) , and σ-1 receptor-overexpressing RGC-5 (Fig. 9C)cells. Data are shown as the mean percentage ± SEM in six different fields of cells, where each field contained ∼60 cells. 

σ-1 Receptor Ligand (+)-SKF10047 Downregulation of Bax in RGC-5 Cells

Increases in Bax expression may lead to mitochondrial depolarization and cytochrome c release, resulting in the downstream activation of executioner caspase to augment apoptosis. ^{20 39} In addition, the expression level of Bax has been shown to be affected by glutamate treatment in PC12 cells. ²² Treatment of RGC-5 cells with glutamate led to an increase in Bax protein expression compared with the control (Fig. 10) . To determine whether the glutamate-induced increase in Bax levels was mediated by NMDA receptor activation, RGC-5 cells were incubated with the NMDA-specific antagonist MK801+glutamate. MK801 reduced Bax protein levels. Treatment of RGC-5 cells with glutamate plus the σ-1 receptor agonist (+)-SKF10047 also resulted in a decrease in Bax protein levels. The effect of (+)-SKF10047 was blocked by BD1047, a σ-1 receptor antagonist. 

The increase of Bax after treatment of BD1047, a σ-1 receptor antagonist, may be because σ-1 receptor antagonists are capable of inducing apoptosis. Overall, the Western blot data showed that σ-1 receptor agonist protective effect involved the regulation of Bax cellular levels in RGC-5 cells. 

Detection of Activated Caspase-3 in RGC-5 Cells Treated with Glutamate

The conversion of pro-caspase-3 to active caspase-3 is generally accepted as one of the most reliable indicators of apoptosis. ^{20 21} To test whether the activation of caspase-3 is associated with glutamate-induced RGC-5 cell apoptosis, we treated RGC-5 cells with 1 mM glutamate for 24 hours, with or without σ-1 receptor agonist (+)-SKF10047 (1 μM), and σ-1 receptor antagonist BD1047 (3 μM). Activation of caspase-3 in RGC-5 cells in response to glutamate was confirmed by the carboxyfluorescein caspase detection kit (Apologix; Cell Technology, Inc.; Fig. 11A ) and suggested that glutamate-treated cells were dying by apoptosis. The σ-1 receptor ligand (+)-SKF10047 was again able to reduce the activation of caspase-3 after glutamate treatment. BD1047, a σ-1 receptor antagonist blocked the effect of (+)-SKF10047. In addition, MK801, The NMDA receptor antagonist, inhibited caspase-3 activation significantly. The same results were obtained with rat primary RGCs (Fig. 11B) . 

Discussion

The cellular transduction events mediated by σ-1 receptors are still unclear. Changes in the concentration of free [Ca²⁺]_i ions are recognized to be linked to the induction of apoptosis, but the relationship between calcium and its linkage to the apoptotic program is complex, because calcium can be a signal for both life and death pathways. ⁴⁰ The NMDA ionotropic glutamate receptor, plays an important role in both the physiology and pathology of neurons. ^{38 40} It has been shown that repeated NMDA receptor activation is one of the main causes of neurodegenerative disorders. ^{41 42 43} The major finding reported in this study is that σ-1 receptors protect RGCs from glutamate-induced apoptosis by regulating the [Ca²⁺]_i levels, and the subsequent signaling of apoptotic events including decreasing the expression of Bax levels, as well as limiting caspase-3 activation. The important calcium influx resulting from the hyperactivation of the NMDA receptor complex has been shown to be critically associated with delayed excitotoxic neuronal death. ^{15 40 42 44} These findings are consistent with our current observations that the toxic effect of glutamate was mediated through the activation of the glutamate ionotropic NMDA receptors and involved increases in [Ca²⁺]_i mobilization in RGC-5 cells. In addition, our results showed that the σ-1 receptor agonist (+)-N-allylnormetazocine ((+)-SKF10047) inhibited glutamate-induced [Ca²⁺]_i mobilization. Similarly, Renaudo et al. ⁴⁵ reported that the activation rate of volume-regulated chloride channels (VRCCs) was dramatically delayed in σ-1 receptor HEK293-transfected cells in the absence of ligands. These findings are in agreement with those in our studies that showed σ-1 receptor-overexpressing cells to have a lower response to glutamate-induced calcium mobilization than did the nonoverexpressing σ-1 receptor RGCs. 

Changes in [Ca²⁺]_i levels appears to be associated with neuroprotective properties of σ-1 receptors, similar to that reported in cancer cells by Spruce et al. ⁴⁶ Most σ ligands that have been used in different studies also demonstrate that there is altered calcium mobilization ^{38 44} in the presence of these σ-1 receptor ligands. Similar results have been reported by other studies in cells of the central nervous system. ³⁵ Our data also confirm the reported data that showed that the σ-1 receptor agonist (+)-pentazocine inhibits significantly the apoptosis induced by homocysteine or glutamate. ⁴⁷ The receptor mechanisms mediating σ-1 receptor neuroprotection comprise complex interactions involving the ionotropic, and even the voltage-gated, calcium signaling processes. ⁴⁸  

In addition to showing that calcium regulation is involved in the neuroprotective mechanism of σ-1 receptor activation in RGC-5 cells, we showed that the expression levels of Bax and caspase-3 activation were decreased by the activation of σ-1 receptors. Glutamate, the principle excitatory amino acid in the central nervous system, is considered to play an important role in neurotransmission, neuronal development, synaptic plasticity, and neuronal degeneration. Glutamate has been shown to upregulate Bax levels in PC12 cells and cortical neurons ^{21 22} and to activate caspase-3 in primary cortical neurons. ⁴⁹ Activation of caspase-3, induced by various triggers of apoptosis, is an early biochemical marker of general apoptosis in certain types of cells. The activation of caspase-3 was observed in the RGC-5 cells after 24 hours of glutamate treatment, and σ-1 receptor agonists can prevent this activation. The same result was reproduced in primary rat RGCs with low glutamate concentration and short time exposure (6 hours). 

Overall, we have shown that σ-1 receptor activation protects RGC-5 cells from glutamate-induced excitotoxicity by regulating calcium mobilization in RGCs, decreasing Bax expression levels, and inhibiting caspase-3 activation. The regulation of glutamate-induced calcium mobilization in RGC-5 cells by σ-1 receptors can activate cell-survival signaling pathways and decrease pathways linked to cell death. 

 Supported by a grant from the Texas Higher Education Coordinator Board and an National Science Foundation (NSF) SCORE (Schools’ Cooperative Opportunities for Resources) training grant.

 Submitted for publication August 24, 2007; revised January 3 and February 8, 2008; accepted April 16, 2008.

 Disclosure: K.T. Tchedre, None; T. Yorio, 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.

 

Morphologic differences between undifferentiated and differentiated RGC-5 cells. (I) Morphology of RGC-5 cells: (A) undifferentiated and (B) differentiated RGC-5 cells. (II) Pseudo color of images of RGC-5 cells: (A) undifferentiated and (B) differentiated RGCs after 30 minutes of incubation with fura-2 AM in HBSS that reflect the levels of [Ca²⁺]_i. Colors reflect relative [Ca²+]_i concentration as follows: blue, low; green/yellow, intermediate; red, high (see calibration bar) in (B).

Figure 2.

 

Western blot analysis of σ-1 receptor protein expression in normal, GFP, and σ-1 receptor-overexpressing RGC-5 cells. Cell lysates (20 μg of protein) from normal RGC-5, RGC-5-GFP, and σ-1 receptor-overexpressing RGC-5 cells cultured in 10% FBS-DMEM to complete confluence were subjected to 15% SDS-PAGE followed by immunoblot analysis with specific antibodies for the σ-1 receptor. Lane 1: normal RGC-5 cells; lane 2: RGC-5 cells transfected with GFP vector (empty σ-1 receptor vector); and lane 3: RGC-5 cells overexpressing σ-1 receptor (clone 4). The σ-1 receptor was detected as a protein band of ∼25 kDa.

Figure 3.

 

Dose-response of glutamate-induced excitotoxicity in RGC-5 cells. Cell survival was monitored by using the calcein-AM/propidium iodide cell-survival assay. Normal RGC-5 cells were treated as follows: (A1) control (no glutamate); (A2) 50 μM glutamate; (A3) 100 μM glutamate; (A4) 250 μM glutamate; (A5) 500 μM glutamate; and (A6) 1 mM glutamate for 4 days. Glycine (10 μM) was added in each well. The cells were incubated with 2 μM calcein (green, live cells) and 2 μg/mL propidium iodide (PI; red, dead cells) for 30 minutes. The coverslips were mounted on microscope slides and analyzed by fluorescence microscopy. (B) Summary of glutamate dose-response calcein-AM/propidium iodide cell survival. The quantitative data collected from the fluorescence images are expressed as the mean percentage ± SEM or six different fields of cells, where each field contained ∼60 cells). Image J software (National Institutes of Health, Bethesda, MD) was used for the quantification. *Statistically significant compared with the control (no glutamate treatment). Scale bar, 200 μm.

Figure 4.

 

Primary RGCs were grown on 35-mm coverslips for 7 days, fixed, and subjected to immunofluorescent staining for detection of the normally expressed RGC marker, Thy-1 (A). The cells were incubated with primary Thy-1.1 monoclonal antibody and σ-1 receptor polyclonal antibody followed by incubation with the secondary antibodies (Alexa Fluor 633 and 488; Invitrogen-Molecular Probes, Eugene, OR). Confocal laser scanning microscopy was used to detect Thy-1 (red) DIC (B).

Table 1.

 

Concentration-Dependent Elevation of [Ca²⁺]_i

Treatment	[Ca2+]i (Mean nM ± SEM)	Cells (n)
Baseline	68 ± 11	15
Glutamate 50 μM	299 ± 63*	15
Baseline	102 ± 6	24
Glutamate 250 μM	1194 ± 177*	24
Baseline	178 ± 24	11
Glutamate 500 μM	4660 ± 1254*	11
Baseline	112.9 ± 14	19
Glutamate 1 mM	6011.98 ± 657*	19

 Glutamate-induced calcium mobilization was measured by fura-2-AM imaging. The cells were preincubated for 30 minutes with fura-2 before stimulation with glutamate. Data were analyzed by one-way repeated-measures ANOVA, and pair-wise multiple-comparison (Student-Newman-Keuls method) of levels at baseline, peak, and 1 minute after peak.

*  Statistically significant (P < 0.05) compared with the baseline.

Table 2.

 

Concentration-Dependent Inhibition of [Ca²⁺]_i by (+)-SKF10047

Treatment	[Ca2+]i (Mean ± nM SEM)	Cells (n)
Baseline	64 ± 5	44
Glutamate 250 μM	1675 ± 226*	44
Baseline	80 ± 5	22
Glutamate 250 μM+ (+)-SKF10047 (10 nM)	1090 ± 83*	22
Baseline	137 ± 56	41
Glutamate 250 μM+ (+)-SKF10047 (1 μM)	639 ± 67* , †	41
Baseline	182 ± 22	13
Glutamate 250 μM+ (+)-SKF10047 (10 μM)	268 ± 47, †	13
Baseline	190 ± 17	25
Glutamate 250 μM+ (+)-SKF10047 (100 μM)	260 ± 23, †	25

 Calcium mobilization was determined as described in Table 1 .

*  Statistically significant (P < 0.05) compared with baseline.

†  Statistically significant (P < 0.05) compared with glutamate (250 μM).

Table 3.

 

Comparison of Glutamate-Induced [Ca²⁺]_i Mobilization between RGC-5 Cells Tansfected with GFP Vector and σ-1 Receptor-Overexpressing RGC-5 Cells

Treatment	[Ca2+]i (Mean ± nM SEM)	Cells (n)
RGC-5 transfected with GFP/sigma-1 receptor empty vector
Baseline	154 ± 8	21
Glutamate, 250 μM	1086 ± 102*	21
Baseline	90 ± 10	19
Glutamate (250 μM) + (+)-SKF10047 (1 μM)	502 ± 43* , †	19
Baseline	83 ± 16	19
Glutamate (250 μM) + MK801 (10 μM)	83 ± 14, †	19
GFP/Sigma-1 receptor overexpressing RGC-5 cells
Baseline	67 ± 6	16
Glutamate, 250 μM	630 ± 40* , ‡	16
Baseline	74 ± 4	43
Glutamate (250 μM) + (+)-SKF10047 (1 μM)	199 ± 8, †	43
Baseline	57 ± 2	17
Glutamate (250 μM) + MK801 (10 μM)	60 ± 2, †	17

 Calcium mobilization was determined as described in Table 1 .

*  Statistically significant (P < 0.05) compared with the baseline.

†  Statistically significant (P < 0.05) compared with the control (glutamate 250 μM).

‡  Statistically significance (P < 0.05) versus RGC-5 transfected with GFP vector (σ-1 receptor empty vector) control (glutamate, 250 μM).

Table 4.

 

Inhibition of Glutamate-Induced Calcium Influx in RGC-5 Cells Mediated through σ-1 Receptor

Treatment	[Ca2+]i (Mean ± nM SEM)	Cells (n)
RGC-5 transfected with GFP/σ-1 receptor empty vector
Baseline	122.7 ± 7	28
Glutamate (250 μM)	1571 ± 433	28
Baseline	109.9 ± 8	32
Glutamate (250 μM) + (+)-SKF10047 (1 μM)	666.9 ± 56*	32
Baseline	100.1 ± 12	32
Glutamate (250 μM) + BD1047 (3 μM)	1520.2 ± 225	32
Baseline	49 ± 4	43
Glutamate (250 μM) + BD1047 (3 μM) + (+)-SKF10047 (1 μM)	1584 ± 237	43
GFP/σ-1 receptor-overexpressing RGC-5 cells
Baseline	67.9 ± 3	29
Glutamate (250 μM)	671.4 ± 68	29
Baseline	65.4 ± 3	43
Glutamate (250 μM) + (+)-SKF10047 (1 μM)	254.4 ± 14*	43
Baseline	47.9 ± 3	42
Glutamate (250 μM) + BD1047 (3 μM)	602.9 ± 68	42
Baseline	63.1 ± 4	36
Glutamate (250 μM) + BD1047 (3 μM) + (+)-SKF10047 (1 μM)	633.8 ± 92	36

 Glutamate induced calcium mobilization was measured by fura-2-AM imaging. Cells were preincubated for 30 minutes with fura-2 and BD1047 before stimulation with glutamate. Data were analyzed by one-way, repeated-measures ANOVA and pair-wise multiple-comparison (Student-Newman-Keuls method) between baseline, peak, and 1 minute after peak.

*  Statistically significant (P < 0.05) compared with the control (glutamate 250 μM).

Figure 5.

 

NMDA receptor expression was not affected by σ-1 receptor overexpression. Cell lysates (20 μg of protein) from normal RGC-5, RGC-5 cells transfected with GFP vector, and σ-1 receptor-overexpressing RGC-5 cells were subjected to 7% SDS-PAGE, followed by immunoblot analysis with a specific antibody for the NMDAR1 receptor subunits. Lane 1: normal RGC-5 cells; lane 2: RGC-5 transfected with GFP vector; lane 3: RGC-5 overexpressing σ-1 receptor. NMDAR1 can be detected as a protein band of ∼120 kDa.

Figure 6.

 

The σ-1 receptor agonist (+)-SKF10047 protected RGC-5 cells from glutamate-induced excitotoxicity. Cell survival was monitored with the calcein-AM/propidium iodide cell-survival assay. Normal RGC-5 cells were treated as follows: (I) control (no glutamate); (II) 1 mM glutamate; (III) 1 μM (+)-SKF10047; (IV) 1 mM glutamate+1 μM (+)-SKF10047; (V) 3 μM BD1047; and (VI) 1 mM glutamate+3 μM BD1047+1 μM (+)-SKF10047 for 4 days. Glycine (10 μM) was added to each well. The cells were incubated with 2 μM calcein (green, live cells) and 2 μg/mL propidium iodide (PI; red, dead cells) for 30 minutes. The coverslips were mounted on microscope slides and analyzed by fluorescence microscopy. (A) Calcein staining; (B) propidium iodide staining; (C) merged image of (A) and (B). Scale bar, 20 μm.

Figure 7.

 

The σ-1 receptor agonist (+)-SKF10047 protected RGC-5-GFP (σ-1 receptor empty vector) cells from glutamate-induced excitotoxicity. Cell survival was monitored with the calcein-AM/propidium iodide cell-survival assay. RGC-5-GFP transfected (σ-1 receptor empty vector) cells were treated as follows: (I) control (no glutamate); (II) 1 mM glutamate; (III) 1 μM (+)-SKF10047; (IV) 1 M glutamate+1 μM (+)-SKF10047; (V) 3 μM BD1047; (VI) 1 mM glutamate+3 μM BD1047+1 μM (+)-SKF10047; (VII) 10 μM MK801; and (VIII) 1 mM glutamate+10 μM MK801 for 4 days. Glycine (10 μM) was added to each well. The cells were incubated with 2 μM calcein (green, live cells) and 2 μg/mL propidium iodide (PI; red, dead cells) for 30 minutes. The coverslips were mounted on microscope slides and analyzed by fluorescence microscopy. MK801, a specific NMDA receptor antagonist reduced the glutamate-induced increase in RGC-5-GFP cell death. These data suggest that glutamate-induced apoptosis is mediated by NMDA receptor activation. (A) Calcein staining; (B) propidium iodide staining; (C) merged image of (A) and (B). Scale bar, 20 μm.

Figure 8.

 

The σ-1 receptor agonist (+)-SKF10047 protected σ-1 receptor-overexpressing RGC-5 cells from glutamate-induced excitotoxicity. Cell survival was monitored with the calcein-AM/propidium iodide cell-survival assay. The σ-1 receptor-overexpressing RGC-5 cells were treated as follows: (I) control (no glutamate) (II) 1 mM glutamate; (III) 1 μM (+)-SKF10047; (IV) 1 mM glutamate+1 μM (+)-SKF10047; (V) 3 μM BD1047; (VI) 1 mM glutamate+3 μM BD1047+1 μM (+)-SKF10047; (VII) 10 μM MK801; and (VIII) 1 mM glutamate+10 μM MK801 for 4 days. Glycine (10 μM) was added in each well. The cells were incubated with 2 μM calcein (green, live cells) and 2 μg/mL propidium iodide (PI; red, dead cells) for 30 minutes. The coverslips were mounted on microscope slides and analyzed by fluorescence microscopy. MK801, a specific NMDA receptor antagonist reduced the glutamate-induced increase in RGC-5 cell death. The data suggest that glutamate-induced apoptosis was mediated by NMDA receptor activation. In addition, σ-1 receptor-overexpressing RGC-5 cells presented a significant resistance to glutamate-induced apoptosis. (A) Calcein staining; (B) propidium iodide staining; (C) merged image of (A) and (B). Scale bar, 20 μm.

Figure 9.

 

Summary of calcein-AM/propidium iodide cell-survival assay data. Quantitative data from fluorescence images derived from the calcein/propidium iodide cell-survival assay (Figs. 3 5 6)are expressed as the mean percentage ± SEM of results in six different fields of cells, where each field contained ∼60 cells. (A) Normal RGC-5 cells; (B) RGC-5-GFP (σ-1 receptor empty vector) cells; and (C) σ-1 receptor-overexpressing RGC-5 cells. Image J Software (National Institutes of Health, Bethesda, MD) was used for the quantification. * P < 0.05 compared with the control (no treatment); **P < 0.05 compared with glutamate (1 mM); &P < 0.05 compared with 1 mM glutamate treatment in normal RGC-5 cells; and RGC-5-GFP (σ-1 receptor empty vector) cells.

Figure 10.

 

After glutamate treatment, Western blot analysis of Bax levels in normal RGC-5 cells was used to detect the cytosolic levels of Bax protein. RGC-5 cells were treated for 24 hours as follows: (A) Lane 1: control (no glutamate); lane 2: 1 mM glutamate; lane 3: 10 μM MK801; lane 4: 1 mM glutamate+10 μM MK801; lane 5: 1 μM (+)-SKF10047; lane 6: 1 mM glutamate+1 μM (+)-SKF10047; lane 7: 3 μM BD1047; and lane 8: 1 mM glutamate+3 μM BD1047+1 μM (+)-SKF10047. Cell lysates were prepared, and 20 μg protein was separated by SDS 12% polyacrylamide gel electrophoresis. Immunoblot analysis of Bax was performed with rabbit polyclonal Bax (N-20) antibody. β-Actin was used as a control for equal loading. (B) Densitometric analysis of the bands was performed with Scion image analysis software (National Institutes of Health, Bethesda, MD). The quantification of band intensity is represented as a percentage of the value of its corresponding control band on the same membrane and the intensities of staining for Bax are presented as the mean percentage of results in three separate experiments. *P < 0.05 of mean (%) Bax density versus that of the control (no treatment). **Statistical significance (P < 0.05) of mean (%) Bax density versus that of 1 mM glutamate treatment.

Figure 11.

 

(A) Glutamate-induced caspase-3 activation in RGC-5 cells. Normal RGC-5 cells were treated with 1 mM glutamate, as shown in the key in the illustration. After the carboxyfluorescein caspase-3 substrate was added to the cell culture medium for 30 minutes, the coverslips were subsequently washed once with 1× PBS and mounted on microscope slides. Caspase-3 activation was detected by confocal laser scanning microscopy. * P < 0.05 compared with the control (no treatment). **P < 0.05 compared with glutamate (1 mM). (B) Glutamate-induced caspase-3 activation in rat primary RGCs. Primary RGCs were treated with 250 μM glutamate for 6 hours and the coverslips washed and mounted and caspase-3 detected as in (A). The quantitative data summaries for both experiments are shown as the mean fluorescence ± SEM in six fields. Image J Software (National Institutes of Health, Bethesda, MD) was used for the quantifications. * P < 0.05 compared with the control (no treatment); **P < 0.05 compared with glutamate (250 μM).

The authors thank Sarkar Saumyendra for suggestions and I-fen Chen Chang for technical support with the confocal microscopy.