Müller Cell Protection of Rat Retinal Ganglion Cells from Glutamate and Nitric Oxide Neurotoxicity

Abstract

purpose. Low concentrations of excitotoxic agents such as glutamate and nitric oxide decrease survival rates of purified retinal ganglion cells (RGCs). In the retina, RGCs are ensheathed by retinal Müller glial (RMG) cell processes. The purpose of this study was to determine whether RMG cells could protect RGCs from these excitotoxic injuries.

methods. RGCs were purified from 7- or 8-day-old Long Evans rats and cultured on polylysine/laminin-coated coverslips in serum-free medium for 2 days. The coverslips were then moved to dishes containing either confluent RMG monolayers or no glial cells in glutamate-free medium. Some dishes with confluent RMG cells were exposed to d,l-threo-β-hydroxyaspartate (THA), a blocker of glutamate uptake. Three days after exposure to various concentrations of glutamate or the NO donor, 2,2′-(hydroxynitroso-hydrazino)bisethanamine, survival rates of RGCs were measured by calcein-acetoxymethyl ester staining. Glutamate concentrations in the medium were measured using amino acid analysis.

results. Without RMG cells, the application of increasing concentrations (5–500μ M) of glutamate caused a dose-dependent increase in RGC death after 3 days. The neurotoxic effects of glutamate were blocked in the RMG cell cocultures, even when there was no direct contact between the cell types. The protective effect of RMG cells was weakened by THA treatment. NO also had toxic effects on RGC. RMG cells prevented this toxicity but only when in direct contact with the RGCs.

conclusions. RMG cells can protect RGCs from glutamate and NO neurotoxicity. We suggest that functional disorders of glutamate uptake in RMGs might be one of the etiologies of glaucoma.

Retinal ganglion cell (RGC) death is the final common pathway of virtually all diseases of the optic nerve, including glaucomatous optic neuropathy. Glaucoma is a leading cause of blindness and a clear association exists between it and certain risk factors, such as high intraocular pressure (IOP) or blood-flow dysregulation. Nevertheless, many cases of glaucoma do not correlate with these risk factors, and it has been proposed that some of these involve glutamate-mediated excitotoxicity. ¹  

A number of studies have been published on the relationship between RGCs and glutamate in vitro, but the different methods used and the rapid and extensive RGC death noted in many control cultures make interpretation and comparison of these difficult. Recently, we showed that a low concentration of glutamate, 25 μM, decreased survival of RGCs that could otherwise survive in culture for several weeks. ² In this study the RGCs were cultured in the absence of other cells, but in the retina they are ensheathed by Müller glial cell processes. We believed it important, therefore, to determine whether RGCs in the presence of glial cells are as sensitive to excitotoxins as the purified cells. 

Glial cells are thought to protect neurons from various neurologic insults. The glial cells in the vertebrate retina are grouped into microglia, astrocytes, and Müller cells. Microglia are found predominantly in the outer retina in the adult and have little contact with RGCs. Astrocytes surround the RGC axons in the optic nerve layer of the retina, but the most extensive glial contact is with Müller cells whose processes surround ganglion cell bodies and dendrites. Many functions have been postulated for Müller glial cells, including structural and nutritional roles and removal of ions and neurotransmitters from the extracellular space. ^{3 4} Several glutamate transporters have been cloned: l-glutamate/l-aspartate transporter (GLAST), ^{5 6} GLT-1, ⁷ EAAC1, ⁸ EAAT4. ⁹ The primary glutamate transporter expressed by retinal astrocytes and Müller cells is GLAST, ¹⁰ which has been postulated to contribute to the clearance of glutamate and protect RGCs from glutamate neurotoxicity. ^{11 12 13 14}  

In addition to the direct actions of glutamate, there is increasing evidence that it can exert an indirect excitotoxic action. One of these indirect actions may be stimulation of synthesis and release of nitric oxide (NO). NO has been implicated in a number of retinal diseases, including glaucoma. ¹⁵ Preinjection with a NOS inhibitor partially protected against RGC degeneration induced by intravitreal injection of NMDA into an nNOS-deficient mouse. ¹⁶ Nitric oxide synthase (NOS) is present in a few cells in the disorganized lamina cribrosa of the glaucomatous eye but is not present at all in normal tissue. ¹⁷ Treatment of a rat model of chronic glaucoma for 6 months with aminoguanidine, a relatively specific inhibitor of NOS-2, produced normal eyes, compared with an untreated group that developed pallor and cupping of the optic disks in the eyes with elevated IOP. ¹⁸ In contrast to these deleterious effects of NO, there are some reports showing that NO can be supportive. ^{19 20 21} The mechanisms of action of NO within the complex milieu of the intact retina are not fully understood but may include nitrosylation of membrane proteins and activation of ion channels through the cGMP pathway. ^{22 23}  

Though many articles have discussed the relationship between RGC death and glial cells, little is known about interaction of Müller cells and RGC in the absence of other cells. In mixed retinal cell cultures, addition of drugs may stimulate release of factors from other cells that act on RGCs or on Müller cells. We have cultured purified RGCs and Müller cells separately and combined them in different ways. Our results show that Müller cells can protect retinal ganglion cells from both glutamate and NO neurotoxicity. 

Materials and Methods

Experimental animals were treated in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Cell culture reagents were obtained from Gibco (Grand Island, NY). Papain was obtained from Worthington Biochemical (Freehold, NJ). Recombinant neurotrophic factors were obtained from R&D Systems (Minneapolis, MN; human brain-derived neurotrophic factor [BDNF]) or Peprotech (Rocky Hill, NJ; rat ciliary neurotrophic factor [CNTF]). Unless noted, all other reagents were obtained from Sigma (St. Louis, MO). 

Preparation of Retinal Suspensions

Retinal ganglion cells were purified, as previously described. ² Briefly, 7- to 8-day-old Long Evans rats were euthanatized by dry ice inhalation, and eyes were dissected. Retinas were incubated at 37°C for 30 minutes in 10 U/ml papain and 70 U/ml collagenase in Hanks’ balanced salt solution containing 0.2 mg/ml bovine serum albumin (BSA) and 0.2 mg/ml d,l-cysteine. To yield a suspension of single cells, the tissue was then triturated sequentially through a narrow-bore Pasteur pipette in a solution containing 2 mg/ml ovomucoid, 0.004% DNase, and 1 mg/ml BSA. After centrifugation at 600 rpm for 5 minutes, the cells were rewashed in another ovomucoid-BSA solution (10 mg/ml of each). After centrifugation, the cells were resuspended in 0.1% BSA in phosphate-buffered saline (PBS). 

Panning Procedure

The preparation of tubes coated with MAC1 or 2G12 (anti-Thy1) antibodies and the panning procedure has been described previously. ² Adherent cells on 2G12-coated tubes were washed with serum-free culture medium (described below). After centrifugation at 600 rpm for 5 minutes, the cells were seeded on 12-mm glass coverslips that had been coated, first with 50 μg/ml poly-l-lysine and then with 10 μg/ml laminin. 

Culture of Purified RGCs

Purified RGCs were plated at a low density of approximately 200 cells/cm² of growth substrate. This plating density provided cultures in which most RGCs grew in physical isolation from other cells. The purified RGCs were cultured for 2 days in 400μ l serum-free medium containing Neurobasal (Gibco) with 1 mM glutamine; 10 μg/ml gentamicin; B27 supplement (1:50); and 40 ng/ml each BDNF, CNTF, and 5 μM forskolin. (RGC-culture medium). Each test substance was diluted in the serum-free medium described above. Cultures were maintained at 37°C in a humidified atmosphere containing 5% CO₂ and 95% air. 

Müller Glial Cell Culture

Müller cells were obtained by a previously described method. ²⁴ Briefly, enucleated eyes from Long-Evans rats at postnatal (PN) day 12 to PN16 under sterile conditions were soaked as intact eyeballs in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10 μg/ml gentamicin overnight at room temperature in the dark. They were then incubated in DMEM containing 0.1% trypsin and 70 U/ml collagenase for 60 minutes at 37°C. The retinas were removed and dissociated by trituration with a narrow-bore Pasteur pipette into small aggregates in culture medium. The culture medium was DMEM with low glucose supplemented with 10% fetal bovine serum (FBS; HyClone, Logan, UT) and 10 μg/ml gentamicin (Glia-culture medium). Cells were seeded into 60-mm culture dishes at the density of 16 retinas. After 5 to 6 days the cultures were washed extensively with medium until only a strongly adherent flat cell population remained. Cultures were maintained at 37°C in humidified atmosphere containing 5% CO₂ and 95% air. The culture medium was changed three times per week until confluence. 

Immunocytochemistry

Retinal Müller glial (RMG) cells subcultured onto coverslips were fixed in 2% paraformaldehyde in PBS for 10 minutes. After rinsing the coverslips, cells were permeabilized in 0.05% Triton X-100, 1% normal goat serum (NGS), and 1% BSA followed by preincubation for 10 minutes in PBS containing 5% NGS. The primary antibodies were used at dilutions of 1:100 for RET-G2, ²⁵ S-100, MAP2, and anti-GFAP; 1:10 for RET-G1. ²⁵ Coverslips were incubated overnight at 4°C, washed five times in PBS, and then incubated for 30 minutes with anti-mouse IgG-fluorescein isothiocyanate (anti-FITC; Jackson, West Grove, PA) 1:150 or anti-mouse IgG-Texas Red. Coverslips were washed, mounted in 50% PBS-50% glycerol, and viewed by epifluorescence illumination on a Zeiss photomicroscope (Thornwood, NY). Control experiments were processed at the same time using PBS containing 0.05% Triton X-100, 1% NGS, and 1% BSA, with RET-P1 (1:2) in the place of primary antibody. 

Cultures for RGC Survival

Culture conditions for the four groups were as follows: (1) RGC + no RMG: RGC coverslips were moved into 60-mm dishes without RMG with the RGC facing up; (2) RGC + RMG—no contact: RGC coverslips were moved into RMG-confluent monolayer dishes with the RGCs facing up; (3) RGC + RMG—cell contact: RGC coverslips were moved into RMG-confluent monolayer dishes with the RGCs facing down; (4) RGC + RMG—cell intermixing: trypsinized RMG cells were added to coverslips containing RGCs. After 2 days of culture, these mixed coverslips were added to dishes containing confluent monolayers of RMG cells. 

Cultures were maintained in dialyzed FBS medium for 3 days. This consisted of Neurobasal medium containing 1 mM glutamine; 10 μg/ml gentamicin; B27 supplement (1:50); and 40 ng/ml each BDNF, CNTF, and 5μ M forskolin. It also contained 10% FBS that had been extensively dialyzed against Neurobasal medium to reduce the glutamate concentration to negligible levels. Glutamate at concentrations from 0 to 500 μM or 2,2′-(hydroxynitroso-hydrazino) bisethanamine (NOC18; Dojindo, Kumamoto, Japan) at concentrations of 0 to 100 μM were added at the beginning of the experiment. NOC18 has a half-life for NO release of 21 hours. Controls used NOC18 that had been incubated for at least 30 days to release essentially all its NO (spent NOC18). 

d,l-Threo-β-hydroxyaspartate Application

RMG-confluent 60-mm dishes were exposed to DMEM with 10% FBS including 1 mM d,l-threo-β-hydroxyaspartate (THA), a blocker for EAAC, GLAST, and GLT-1, for 1 day. After changing the medium into dialyzed FBS medium containing glutamate at defined concentrations, RGC coverslips were moved into the dishes upside-up. 

Assay of Retinal Ganglion Cell Survival

Three days after exposure to various concentrations of glutamate, cell viability was determined using 1 μM calcein-AM. In this study, a surviving RGC was defined as a cell with a calcein-stained cell body and a process extending at least two cell diameters from the cell body (Fig. 1) . Approximately 200 cells were counted in the no-treatment experiment. The percentage of surviving RGCs was determined for each condition on each experiment and was normalized to control specimens examined in parallel under the same conditions. The average relative percentage of cell survival in at least 10 experiments conducted under each condition is expressed in the text and figure as the mean ± SD. Statistical comparisons were made with Student’s t-test analysis of distributed data. 

Glutamate Concentration Measurement

RGC coverslips were moved into 60-mm RMG-confluent monolayer dishes in dialyzed-FBS medium, in which glutamate was applied at defined concentrations. Aliquots of medium were collected at defined times, dried, dissolved in 20 μl sample loading buffer (sodium citrate with 2 nmol homoserine internal standard), and analyzed on a Beckman 6300 Amino acid Analyzer (Fullerton, CA). 

Results

Glial Culture Characterization

Ganglion cells after 3 to 5 days in culture showed a 10- to 17-μm cell body with multiple processes that extended for many cell body diameters (Fig. 1) . After 14 days in vitro, glial cultures were labeled with monoclonal antibodies RET-G1, RET-G2, and polyclonal S-100, all markers of Müller cells. Almost all cells present in these cultures showed positive staining with either RET-G1 or RET-G2 (Fig. 2) . In addition, S-100 labeling of cells showed intense cytoplasmic filament staining. By this immunocytochemical labeling, the cultured cells were thought to be RMG cells. There were some contaminating cells in the cultures, but neurons with neurites were not seen under phase contrast illumination, and no MAP2-positive cells were detected by immunocytochemistry. Some of the contaminant cells were stained by anti-GFAP, suggesting that they were astrocytes. 

Neurotoxic Effect of Glutamate on RGCs Is Blocked by Müller Glial Cells

Application of increasing concentrations (0–500 μM) of glutamate caused a dose-dependent increase in cell death in purified RGC cultures (culture condition 1) measured after 3 days. Twenty-five micromolar glutamate reduced cell survival to 67.0 ± 8.0% (n = 10) and 500 μM glutamate to 18.0 ± 7.8% (n = 10) (Fig. 3) . This neurotoxic effect of glutamate was alleviated in cocultures of RGCs and in Müller cells such that cell survival was 96.7 ± 5.0% (n = 10) in 25 μM and 57.2 ± 9.1% (n = 10) in 500 μM glutamate when the RGC had no contact with the RMG (culture condition 2) and 96.8 ± 5.8% (n = 10) in 25 μM and 59.6 ± 6.4% (n = 10) in 500 μM glutamate when the coverslips were placed such that the RGC contacted the RMG (culture condition 3). Coverslips that had RMG directly plated onto the RGC coverslips (culture condition 4) were slightly less effective (88.8 ± 8.4%[ n = 10] and 49.8 ± 8.1% [n = 10] cell survival, respectively) at overcoming the glutamate toxicity, but this difference was not statistically significant. 

Depletion of Glutamate in Medium by RMG Cells

The simplest explanation for the observed protective effect of Müller cells is that they removed glutamate from the medium. To test this, glial cultures were incubated with RGC coverslips upside down in medium (culture condition 3) with different concentrations of glutamate, and the residual levels were tested at different time points. Glutamate concentrations (5 and 25 μM) in the medium showed a rapid initial decline; within 30 minutes concentrations had dropped to nontoxic levels, and within 1 hour to less than 1 μM (Fig. 4) . These low levels were maintained for at least 72 hours, indicating that RMG cells can remove toxic levels of glutamate from the vicinity of RGCs. 

Protective Effect of Müller Cells Was Inhibited by THA Pretreatment

To provide further evidence for the mechanism of the protective effect of RMGs, we also carried out mixed culture experiments in which the RMG glutamate transporter had been blocked by preincubation with THA. In preliminary experiments we added THA to cocultures of RGCs and RMG cells, but THA itself was toxic to RGC cells at concentrations needed to inhibit GLAST (data not shown). We found that treating RMG cells with THA and then washing out the drug still inhibited glutamate uptake for several hours (Fig. 5) . When RMGs were preincubated with THA, the protective effects of RMGs on RGCs were lessened (Fig. 6) . 

Neurotoxic Effects of Nitric Oxide on RGC Are Blocked by Müller Glial Cells

NOC18 (2,2′-(hydroxynitroso-hydrazino) bisethanamine) is a slow-release NO donor that has a half-life of NO release of 21 hours. Freshly prepared NOC18 reduced RGC survival in a dose-dependent manner such that 10 μM led to the survival of 33.6 ± 10.2% (n = 8) of cells with neurites in purified RGC cultures (Fig. 7) . In controls using NOC18 that had released its NO the compound caused minimal toxicity (Fig. 8) . When RGCs and RMG cells were cocultured in the same dish but with no contact between the cell types, the survival rate increased but was not significantly different from RGCs alone. When RGCs and RMG cells were cocultured such that they contacted each other, the survival rate increased significantly (P < 0.001). This suggests that the neuroprotective effect of RMG cells requires contact with RGCs. 

Discussion

We have previously used a panning procedure to prepare stable RGC cultures of high purity. Because some freshly isolated RGCs seem supersensitive to glutamate, we modified the way we measured glutamate toxicity, and this has produced small differences in survival at particular glutamate concentrations. Previously, glutamate was applied at the time of culture; here we applied glutamate after 2 days of culture. In each case we measured cell survival using calcein fluorescence. The use of this dye, plus the criterion of good process growth, has provided an accurate and sensitive measure of cell survival. 

The neurotoxic effects of glutamate were significantly reduced by Müller cells. It has previously been shown that Müller cells can protect against the excitotoxic effects of even 1 mM glutamate in the whole retina and increase survival of ganglion cells in culture. ^{26 27} Our findings extend these studies in two ways. First, we have shown that this protective effect is mediated directly by Müller cells because we eliminated other cell types from our cultures. Second, we have shown that Müller cells can reduce glutamate concentrations to nontoxic levels within 30 minutes. Several glutamate transporters have been cloned (GLAST, GLT-1, EAAC1, and EAAT4), but GLAST is thought to be the most important in retinal glia. ^{10 11 12} We cultured Müller cells from Long-Evans rats at PN12 to PN16, and at this age the cells should express GLAST, which has been detected in Müller cells at PN7 to PN10. ²⁸ We were able to confirm the importance of Müller cell glutamate uptake on RGC survival by using the GLAST uptake inhibitor THA. We were not able to obtain complete reversal of the glutamate-induced cell death because we could not keep THA in the medium during the experiment. 

It has been reported that retinal neurons in a GLAST-deficient mutant mouse were more susceptible to ischemia-induced degeneration than those of wild-type mice, ²⁹ but a GLAST knockout mouse did not appear to show excessive neuronal degeneration. ³⁰ Glutamate transporters in addition to GLAST are also involved in glutamate uptake. In the genetically altered animals it is possible that other glutamate transporters compensate for the lack GLAST but that such mechanisms would not be induced in the short-term suppression induced by THA in our experiments. 

As would be expected for a protective effect involving uptake from the medium, there was no difference whether or not the glial cells and RGCs were in contact. A previous report suggested that Müller glial cells could protect against excitoxic damage to retinal neurons but that this effect needed cell–cell contact. ³¹ The major difference between this finding and the present results is that we have used purified cells of high viability. It is possible that glutamate can have a rapid action on other types of retinal neurons that do require Müller cell contact for protection. For example, it has been shown that glutamate stimulation of some neurons can lead to the production of nitric oxide. ³² As our results show, protection against NO toxicity does need Müller cell contact. Further experiments are needed to determine whether such indirect effects of glutamate occur in the retina. 

There seems to be the possibility that the weight of coverslips may affect the underlying RMG cells and may induce calcium waves. ³³ When we removed the coverslips after 3-day exposure to glutamate, the underlying RMG cells were still present and appeared intact with no detectable scars. The diameters of a coverslip and culture dish are 12 and 60 mm, respectively. The ratio of the areas is 1:25. Therefore the majority of RMGs had no contact with RGCs, and we suggest that the main protective effects were not induced by contact. 

The 21-hour half-life of release of NO from 2,2′-(hydroxynitroso-hydrazino) bisethanamine allows the generation of a moderate and constant NO concentration in the medium compared with usual NO donor that provides a rapid pulse of NO but then no NO for the remainder of the experiment. NO had toxic effects on RGCs, but they were ameliorated by the contact of Müller cells. For glial cells to protect the RGCs, it was necessary for the cells to be in direct contact. There are two possible explanations for this finding. The first is that the protective effect of RMG cells is exerted through an interaction between cell surface molecules of the two cell types that is transduced into the RGCs. The second is that proximity is required because of the short half-life of NO in solution. It is possible that the only effective NO is released from NOC18 close to RGCs, and so only those RMG cells in close proximity or contact can exert any protective effect. In either case, the molecular basis of the protective effect is not known. We think that cell surface interactions were the most important protective mechanism, but we cannot exclude local buffering effects provided by RMG cells. If RMG cells are close or touching the RGCs, it is likely that their high endogenous levels of glutathione may serve as a local sink to lower the levels of NO around the adjacent RGCs to provide some protection. 

The actions of NO in the retina are still only incompletely understood. A low concentration of NO may play a protective role in glutamate neurotoxicity by closing the NMDA-receptor–gated ion channel. ²¹ However, elevated concentrations of NO, interacting with oxygen radicals, become toxic and mediate glutamate-induced neurotoxicity in the cultured retinal neurons. ²¹ In addition, the most consistent action of NO on many cell types is to stimulate the production of cGMP. We have previously shown that RGCs possess a Ca²⁺-permeable ion channel that can be activated by cGMP. ³ In addition these cells are likely to possess one or more cGMP-dependent protein kinases. 

Our results suggest several approaches that may be of benefit in patients with glaucoma who have progressive visual field loss, despite satisfactory control of IOP. First, because functional disorders of glutamate uptake in Müller glial cells might be one of the etiologies of glaucoma, stimulation of glial glutamate uptake might directly remove an excitotoxin and might prevent subsequent generation of other excitotoxins such as NO. Second, because NO can be produced by both neurons and glia via several different stimulatory pathways, use of NO blockers or selective inhibition of NOSs might contribute to a clinically significant level of neuroprotection. 

 Supported by the Kemper Foundation and grants from the National Institutes of Health and Research to Prevent Blindness, Inc.

 Submitted for publication April 19, 2000; revised June 22, 2000; accepted July 5, 2000.

 Commercial relationships policy: N.

 

Figure 1.

 

Phase contrast (A) and fluorescence (B) images of retinal ganglion cells (RGCs) purified from a 7-day-old rat. The cell bodies and neurites of RGCs were stained in (B) by calcein-AM, which becomes fluorescent when activated by an intracellular esterase. These cells were maintained in vitro for 5 days. Scale bar, 50 μm.

Figure 2.

 

Immunocytochemical labeling of PN14 retinal cultures (retinal Müller glial cells [RMG]). (A) Phase contrast image of primary cells after 14 days in vitro. (B) Phase contrast image of RMG sheet in vitro. (C) S-100 labeling of cells showing intense staining. (D) Cells labeled with RET-G1. (E) Cells labeled with RET-G2. (F) Nonrelevant antibodies such as RET-P1 show no binding. Scale bar, 50 μm.

Figure 3.

 

Dose-dependent effect of glutamate on RGC survival. Purified RGCs were cultured on coverslips for 2 days in serum-free medium containing 40 ng/ml each of BDNF, CNTF, and 5 μM forskolin. The coverslips were moved into each 60-mm dishes. Medium was changed into dialyzed FBS medium with RMG cells in dishes. Culture conditions 1 to 4 are defined as follows: (1) RGC + no RMG: RGC coverslips were moved into 60-mm dishes without RMG cells, upside–up. (2) RGC + RMG—no contact: RGC coverslips were moved into RMG-confluent monolayer dishes, upside–up. (3) RGC + RMG—cell contact: RGC coverslips were moved into RMG-confluent monolayer dishes, upside down. (4) RGC + RMG—cell intermixing: Trypsinized RMG cells were added to coverslips containing RGCs. After 2 days of culture, these mixed coverslips were added to dishes containing confluent monolayers of RMG cells. After 3 days’ exposure to 25 μM glutamate (left 4 bars) and 500 μM glutamate (right 4 bars), survival rates of RGCs are calculated as a percentage of surviving cells, compared with surviving cells in parallel untreated cultures. Approximately 200 cells were counted in the control experiment. Each data point is the mean ± SD (n = 10). *^,#Significant difference compared with control (P < 0.001).

Figure 4.

 

Depletion of glutamate in medium by RMG cells. Purified RGC coverslip was moved to the 60-mm dishes with confluent RMG monolayer in dialyzed 10% FBS in serum-free medium containing 40 ng/ml each of BDNF and CNTF, and 5 μM forskolin. Glutamate was applied at the concentrations of 5 (▴) or 25 μM (□). Samples of medium were taken at the time points indicated, and the glutamate content was determined. The concentrations of glutamate dropped below 1 μM within 1 hour after RMGs positively depleted glutamate under the presence of purified RGCs.

Figure 5.

 

Glutamate removal by THA-treated Müller glial cells. RMG-confluent 60-mm dishes were exposed to DMEM with 10% FBS including 1 mM THA, a blocker for EAAC, GLAST, and GLT-1, for 1 day. After changing the medium into dialyzed FBS medium containing glutamate at defined concentrations, RGC coverslips were moved into the dishes, upside-up. Glutamate was applied at a concentration of 25 μM. Samples of medium were taken at the time points indicated, and the glutamate content was determined (•). For comparison the concentrations over time after application of 25 μM glutamate to untreated RMG cells are replotted from Figure 4 (□). Treatment of RMGs with THA and then washing out the drug continued to inhibit glutamate uptake for several hours. The concentrations of glutamate dropped below 1 μM within 24 hours.

Figure 6.

 

Effects of THA, EAAC, GLAST, and GLT-1 inhibitor, on RGC survival with RMG cells. Purified RGCs were cultured for 2 days in serum-free medium containing 40 ng/ml each of BDNF, CNTF, and 5 μM forskolin. Medium was changed to dialyzed 10% FBS in serum-free medium containing 40 ng/ml each of BDNF and CNTF, and 5 μM forskolin. The RGC coverslips were moved into (1) dishes without RMG cells (culture condition 1), (2) dishes with confluent RMG cell monolayer (culture condition 2), or (3) dishes with RMG cell monolayer preexposed to 1 mM THA. Glutamate was applied at the concentrations of 5 to 500 μM for 3 days (n = 10). *Significant difference (P < 0.001).

Figure 7.

 

Dose-dependent effect of NOC18 on RGC survival. Purified RGCs were cultured in serum-free medium containing 40 ng/ml each of BDNF, CNTF, and 5 μM forskolin. After 3 days’ exposure to 1, 10, and 100 μM 2,2′-(Hydroxynitroso-hydrazino) bisethanamine (NOC18), a slow release NO donor that has a half-life of NO release of 21 hours, survival rates of RGCs were calculated as a percentage of surviving cells, compared with surviving cells in untreated cultures. Freshly prepared NOC18 reduced RGC survival in a dose-dependent manner. Each data point is the mean ± SD (n = 8). Approximately 200 cells were counted in the no-treatment experiment.

Figure 8.

 

The effects of RMG on RGC death induced by a nitric oxide donor, NOC18 (10 μM). Each data point is the mean ± SD. Approximately 200 cells were counted in the no-treatment experiment. Controls used 10μ M NOC18 that had already released nitric oxide. NO treatment was 10μ M fresh NOC18 and that reduced RGC viability (*P < 0.001). Addition of 10 μM fresh NOC18 to RGCs in contrast with RMG cells caused significantly less cell death (^# P < 0.001). RMG cells not in contact with the RGCs were unable to exert such a protective effect. Results are given as mean ± SD (n = 8).

The authors thank Keely Bumsted and Ming-Hu Han for helpful discussions and Stephen Viviano for excellent technical assistance.