Mechanisms of Neuroprotection by Glucose in Rat Retinal Cell Cultures Subjected to Respiratory Inhibition

Abstract

Purpose.: Previous experiments have demonstrated that short-term hyperglycemia in rats renders the retina resistant to subsequent metabolic insults. The present study aimed to elucidate putative mechanisms involved in this protective response.

Methods.: Retinal cultures comprising neurons and glia were treated with the mitochondrial complex I inhibitor, rotenone, at a range of concentrations, for up to 24 hours. In some cases, glucose or the alternative energy substrates, pyruvate or lactate, and/or inhibitors of glycolysis or the pentose phosphate pathway (PPP) were also applied. Cell viability was assessed using complementary techniques: immunocytochemistry, immunoblotting, cytotoxicity assay, and TUNEL. Cellular levels of ATP, reactive oxygen species (ROS), and nicotinamide adenine dinucleotide phosphate (NAD[P]H) were also assessed.

Results.: Rotenone caused the preferential loss of neurons from retinal cultures in a concentration-dependent manner; glial cells were also affected, but only at a higher concentrations (10 μM). Cell loss was by apoptosis, and was preceded by a reduction of both cellular ATP and NAD(P)H levels and an increase in the production of ROS. Glucose counteracted the detrimental effects of rotenone. This involved a reduction in ROS levels and an increase in the cellular ATP/NAD(P)H ratio. The protective effect of glucose was partially reversed by either PPP or glycolysis inhibition.

Conclusions.: Glucose rescued cultured rat retinal cells from rotenone-induced toxicity. Glucose acted via both the PPP and the glycolytic pathway, maintaining cellular ATP and NAD(P)H levels and reducing ROS production. These data have implications for treatment of retinal diseases that involve metabolic compromise to neurons.

Introduction

Cellular ATP is produced by two related processes: cytoplasmic glycolysis and mitochondrial oxidative phosphorylation (OXPHOS). In glycolysis, glucose is converted to pyruvate, forming 2 ATP molecules. In the presence of oxygen, pyruvate is usually converted to acetyl CoA, which then enters the Krebs cycle, forming electron donors for OXPHOS and subsequently generating approximately 32 ATP molecules. When oxygen is scarce, pyruvate is converted to lactate via the process of anaerobic respiration. 

Although glucose is a fundamental energy requirement of the central nervous system (CNS), including the retina ¹ and optic nerve, it has become dogma within clinical medicine that elevated blood glucose (hyperglycemia) is deleterious to neurons under conditions of energy deprivation. ² Indeed, we have previously shown that cerebral ischemic injury is severely exacerbated by hyperglycemia. ³ In contrast, we showed that providing glucose to retinal neurons in vivo, during both prolonged ³ and acute ⁴ periods of ischemia afforded a profound neuroprotective effect. We also recently reported a similar neuroprotective response in a rodent model of experimental glaucoma in which energy deprivation may be a pathogenic component. ⁵ The robustness of the neuroprotective effect of glucose against ischemic retinal injury and the contrast with its detrimental effect on ischemic cerebral neurons was striking, and we hypothesized that the observed difference between the responses of the two tissues was likely to be related to the unusual retinal metabolism. 

The retina has an exceptionally high energy demand, ⁶ and a propensity for “aerobic glycolysis” whereby pyruvate is converted to lactate despite the presence of abundant oxygen. This is similar to the metabolism of many tumors and has become known as the Warburg effect. Understanding the mechanism of the Warburg effect has recently become an explosive area of cancer research. ^7,8 A related phenomenon, also displayed by the retina is the Pasteur effect. This describes the upregulation of glycolytic lactate production when oxygen tension is reduced, ^9,10 and is considered to be an adaptive response. ^9,11–14 In fact, Pasteur's original observation described the reciprocal effect, namely inhibition of glycolysis by O₂. ¹⁵  

We hypothesized that the neuroprotective effect of glucose against ischemic retinal injury was due to the Pasteur effect: in the absence of physiological levels of oxygen, retinal neurons produce their ATP via upregulated anaerobic glycolysis. We also considered an alternative explanation for the protective effect of glucose: its role in the pentose phosphate pathway (PPP). The PPP uses glucose to make pentose sugars and the reducing agent, nicotinamide adenine dinucleotide phosphate (NAD[P]H), which keeps the powerful antioxidant, glutathione, (GSH) reduced. This pathway, therefore, could also conceivably contribute to the neuroprotective effect of glucose against ischemic injury. In the current study, we sought to determine the cellular mechanisms by which glucose was able to protect retinal cells. For these studies we used a mitochondrial respiratory injury model. 

Materials and Methods

Rat Retinal Cell Cultures

This research was approved by both the University of Adelaide (M-2011-070) and the Animal Ethics Committees of SA Pathology and conforms to the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes, seventh edition, 2004, as well as the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Rat retinal cell cultures comprising both neurons and glia were prepared using a trypsin-mechanical digest procedure previously described. ^1,16 Briefly, retinas were enucleated from 1- to 2-day-old rat pups and incubated in physiological buffer (Solution 1; 120 mM NaCl, 5.4 mM KCl, 24 mM NaHCO₃, 0.1 mM NaH₂PO₄, 3 g/L BSA, 20 mM Glucose, 0.15 mM MgSO4, and 28 μM phenol red) containing 0.1 mg/mL trypsin (Sigma Aldrich, Castle Hill, New South Wales [NSW], Australia), at 37°C for 8 minutes. After the reaction was stopped, cells were resuspended in minimal essential medium (MEM) containing 10% fetal bovine serum (FBS), 10 mg/mL gentamicin sulfate, 200 μM glutamine, and 25 mM glucose, and applied to 13-mm diameter borosilicate glass coverslips (immunocytochemistry), 6-well plates (Western blot), or 12-well plates (NAD(P)H assay, ATP assay, and reactive oxygen species [ROS] determination), all of which had previously been coated with 10 μg/mL poly-D-lysine, for 2 hours. Cells were maintained in saturating humidity with 5% CO₂ at 37°C and were used 7 days post culture. 

Retinal Explant Culture

Retinal explants cultures were produced essentially according to Johnson. ¹⁷ Briefly, adult Sprague–Dawley rats were euthanized by CO₂ asphyxiation and their eyes enucleated. Retinas were removed and then were separately transferred into 6-well plates with the retinal ganglion cell side facing up. Explants were incubated in neuronal growth medium (Neurobasal A supplemented with 2% B27, 1% N2, 0.8 mM L-glutamine, and 100 U/mL penicillin/streptomycin; all reagents were from Invitrogen, Mulgrave, VIC, Australia) and maintained at 37°C with 5% CO₂ and saturating humidity. Explants were maintained untreated, for 24 hours before initiation of any experiments. 

Treatment of Cultured Rat Retinal Cells

For mitochondrial inhibition, rat retinal cultures or explants were incubated for up to 24 hours in MEM (with 200 μM glutamine) containing rotenone (100 nM, 1 μM, and 10 μM) in the presence or absence of 25 mM glucose. In some instances, 2 mM lactate or 2 mM pyruvate were present as alternative potential energy substrates. In order to determine the effects of blocking catabolic pathways, cultured retinal cells were cotreated in the same MEM with the glycolytic inhibitor, iodoacetic acid (IOA; 1 μM, 10 μM) or the PPP inhibitor, 6-aminonicotinamide (6-AN; 10 μM, 100 μM). Both treatments were applied for 24 hours and were in the presence of 25 mM glucose. To rule out the possibility that observed effects were derived from serum contaminants, all experimental incubations were performed in serum-free medium. 

Immunocytochemistry

Cells were fixed with neutral-buffered formalin for 15 minutes, washed in PBS (137 mM NaCl, 5.4 mM KCl, 1.28 mM NaH₂PO4, 7 mM Na₂HPO4; pH 7.4) and then permeabilized with PBS containing 0.1% Triton X-100 (PBS-T). Cells were then blocked in normal horse serum (NHS, 3.3 % vol/vol in PBS; PBS-HS) and labeled with a range of neuronal- and glia-specific antibodies, diluted in PBS-HS, at 4°C, overnight. Labelling was visualized by consecutive incubations with appropriate biotinylated second antibodies (1:250 in PBS-HS, 30 minutes; Vector Laboratories, Abacus ALS, Brisbane, Australia) and streptavidin-AlexaFluor 488 or streptavidin-AlexaFluor 594 (1:500 in PBS-HS, 1 hour; Invitrogen); nuclei were counterstained using 4′,6-diamidino-2-phenylindole (DAPI, 500 ng/mL in PBS, 5 minutes). Finally, cells on coverslips were mounted using antifade mounting medium (DAKO, Botany, NSW, Australia) and examined under a confocal fluorescence microscope (Olympus, Mount Waverly, Australia). 

Quantification of γ, aminobutyric acid (GABA)-immunoreactive(-IR) neurons was performed by manually counting positively-labelled cells on five different randomly chosen fields per coverslip from four to six independent experiments. Glial acidic fibrillary protein (GFAP)-IR glia, and cells, PGP9.5-IR, and Tau-IR neurons were quantified by using ImageJ (National Institutes of Public Health, Bethesda, MD; http://rsb.info.nih.gov/ij/index.html, in the public domain). 

Western Blotting

Cells were harvested from plates by scraping into PBS; samples were then sonicated in homogenization buffer (20 mM Tris-HCl, pH 7.4; containing 2 mM EDTA, 0.5 mM EGTA, 1 mM dithiothreitol, 50 μg/mL leupeptin, 50 μg/mL pepstatin A, 50 μg/ mL aprotinin, and 0.1 mM phenylmethylsulfonyl fluoride). An equal volume of sample buffer (62.5 mM Tris-HCl, pH 7.4, containing 4% wt/vol SDS, 10% vol/vol glycerol, 10% vol/vol β-mercaptoethanol, and 0.002% wt/vol bromophenol blue) was added and samples were boiled for 3 minutes. Electrophoresis was performed as reported previously ¹⁸ using 10% or 12% polyacrylamide gels (as appropriate) containing 0.1% (wt/vol) SDS. Proteins were transferred to polyvinylidene difluoride membrane and blots labelled as previously described for the presence of β-actin (1:10,000; Sigma-Aldrich, Castle Hill, Australia), Tau (1:10,000; DAKO, Botany Bay, NSW, Australia), PGP9.5 (1:2500; Sigma-Aldrich), vimentin (1:1000; DAKO), GFAP (1:10,000; DAKO), LC-3 (1:500; Abgent, Sydney, NSW, Australia). β-actin was assessed in cell extracts as a positive control and quantification was performed with Adobe Photoshop (version CS2; Adobe Systems, Inc., San Jose, CA). 

ROS Measurement

The redox-sensitive cell-permeable fluorophore, dihydroethidium (DHE; Molecular Probes; Invitrogen), was used to quantify levels of cellular/mitochondrial ROS in cultures exactly as described previously. ¹⁸ Incubations were carried out as described except that for the final 30 minutes DHE (5 μM) was added to wells. Dihydroethidium is a non-fluorescent, reduced form of ethidium, which can passively cross plasma membranes of live cells. When oxidized to ethidium by ROS, it can bind to DNA and yield ‘red' fluorescence (excitation 475 nm/emission 610 nm). After incubation, cells were fixed in neutral-buffered formalin containing 1% methanol for 15 minutes and then washed in PBS for direct visualization by using a fluorescence microscope or were just washed three times in PBS and fluorescence quantified using a Typhoon fluorimeter (GE Healthcare Life Science, Piscataway, NJ) and related to total protein content according to the method of Bradford. ¹⁹  

NAD⁺/NADH Ratio

Cells harvested from 12-well plates were immediately extracted into 200 μL of Extraction Buffer (ab65349; Abcam, Cambridge, UK) by homogenization. After vortexing for 10 seconds, samples were spun at 14,000 rpm for 5 minutes. The supernatants were collected and assayed for levels of both NAD and NADH, using a thiazolyl blue-based cycling assay ²⁰ modified to be used with Fluostar Optima (OD 450 nm; BMG Labtech, Mornington, VIC, Australia). The amount of NADP⁺/NAD(P)H ratio was calculated as: [NADP⁺-NAD(P)H]/NAD(P)H and was expressed in nanogram per milligram protein, as equalized by performing protein analysis according to the method of Bradford. ¹⁹  

ATP Measurements

The experiments were performed using a standard bioluminescence assay kit obtained from Sigma-Aldrich. Briefly, after treatment, cells were permeabilized by 50 μL somatic cell ATP-releasing reagent (FL-SAR) and extracts allowed to react with 50 μL ATP assay mix reagent (FLAA) containing luciferin and luciferase. After incubation at room temperature for 10 minutes, luminescence was assessed by luminometry (Fluostar Optima; BMG Labtech). The ATP levels were also compared with a standard curve in each measurement for actual quantification. Data were collected from four independent experiments and related to total protein content as assessed according to the method of Bradford. ¹⁹  

Assessment of DNA Breakdown by TUNEL

For the TUNEL procedure, treated cells on coverslips were fixed as described for immunocytochemistry, washed in PBS containing 0.1% Triton X-100 (PBS-T) for 10 minutes and immersed in PBS. The labelling procedure was carried out as described previously ¹⁸ using the enzyme TdT to add the d-UTP label; final cell labelling was visualized in the present study, however, using streptavidin AlexaFluor 488. Some cells were treated after fixation but before TUNEL staining with DNAse I (0.1 mg/mL; Sigma-Aldrich) for 15 minutes at 37°C in order to determine that the labelling procedure correctly identified DNA breakdown in nuclei. In order to quantify the numbers of TUNEL-labelled nuclei, counts were obtained and averaged from 5 different randomly-selected fields of at least four coverslips from six separate cultures. 

Live/Dead Viability Assay

To determine the proportion of live versus total dead cells, a live/dead assay kit was employed (Invitrogen). In this assay, live cells were distinguished by the presence of cell permeant calcein-AM while ethidium homodimer (EthD-1) was used to produce fluorescence in dead cells. After treatment, adherent sample cells on glass coverslips were washed with PBS and then incubated as described in the kit protocol for 30 minutes at room temperature. The resulting fluorescence was determined using a fluorescence microplate reader (Fluostar Optima; excitation 485–530 nm, emission 530–645 nm; BMG Labtech). In some cases, cells were imaged by fluorescence microscopy. The percentage of dead cells was calculated from the fluorescence readings. Counts were obtained and quantified from four separate experiments in triplicate. 

Statistical Analysis

Experiments were carried out with n = 4 for the Western blotting densitometry and n = 6 to 8 for immunohistochemistry. Between–group comparisons were made using an ANOVA with commercially available statistical software (Stata/IC 12.1; College Station, TX); P less than 0.05 was considered significant. Data are expressed as mean ± SEM. 

Results

Cell Culture

After 7 days in culture, neurons and glia could be discerned by fluorescence microscopy (Fig. 1). These cultures showed positive labelling for Tau- and PGP 9.5-immunoreactive (IR) neurons as well as for the Müller cell marker, Vimentin and the astrocytic marker, GFAP. 

Figure 1

 

Characterization of the mixed rat retinal cell cultures. (A, D) Seven-day cell cultures immunolabeled for neuronal markers (PGP9.5 and Tau). (B, E) Seven-day cell cultures immunolabeled for glial cell markers, (GFAP and vimentin). (C, F) Merged images. Scale bar: 20 μm.

Attenuation of Rotenone-Induced Toxicity to Rat Retinal Cells by Glucose (25 mM)

Immunocytochemical examination of rat retinal cell cultures showed reduced numbers of neurons (GABA-IR, Tau-IR, and PGP 9.5-IR) after a 24-hour treatment with either 1 μM or 10 μM rotenone. This loss was markedly attenuated in the presence of 25 mM glucose (Figs. 2A–C). Glial fibrillary acidic protein–positive glial cells were only affected by the higher concentration of rotenone (10 μM); glucose was also able to protect these cells (Fig. 2D). 

Figure 2

 

Quantification of neuronal and glial cell counts from mixed retinal cell cultures by immunocytochemistry in the presence of different concentrations of rotenone and glucose. (A–D) Rotenone caused loss of GABA, Tau, PGP9.5, and GFAP-immunoreactivity in a dose-dependent manner. Glucose rescued neurons and glial cells (D) in a dose dependent manner. *P < 0.05, **P < 0.01, comparing with control cells in medium containing without glucose (n = 6 individual experiments comprising five determinations per experiment). Images were taken with a fluorescence microscope (200×).

Western blot analyses were also performed to confirm these immunocytochemical results (Figs. 3, 4). Cultures treated with either 1 μM or 10 μM rotenone for 24 hours displayed reduced labelling of the neuronal markers, Tau and PGP9.5 (Fig. 3). Glial loss, as determined by analysis of protein expression levels in the cultures for the glial-specific proteins Vimentin and GFAP (Fig. 4), was only seen when 10 μM rotenone was applied; 1 μM rotenone had no effect. In all cases, glucose was again seen to markedly reduce rotenone-induced neuronal and glial toxicity. 

Figure 3

 

Effect of rotenone (1 μM and 10 μM) and glucose on retinal neuronal cells markers. (A) Western blot analysis after 24 hours of rotenone (1 μM and 10 μM) demonstrating the protective effect of 25-mM glucose on neuronal protein expression. (B) Quantification of the immunoblotting data normalized for the β-actin levels in each sample and expressed as the ratio compared with control cell cultures. *P < 0.05, **P < 0.01 compared with control cultures. The results shown represent the mean ± SD of three independent experiments.

Figure 4

 

Effect of rotenone (1 μM and 10 μM) and glucose on different retinal glia cells markers. (A) Western blot analysis demonstrating reduced expression of glial cell proteins, GFAP, and vimentin after 24-hours treatment with rotenone (10 μM); the addition of 25 mM glucose to the media recovered protein expression to control levels. (B) Quantification of the immunoblotting data normalized for the β-actin levels in each sample and expressed as the ratio compared with control cell cultures. *P < 0.05, **P < 0.01 compared with control cultures. The results shown represent the mean ± SD of three independent experiments.

In order to verify that these data were also applicable to adult retinal cells and not just those from developing neonatal retinas, an adult rat retinal explant system was subsequently employed. Supplementary Figure S1 shows that there was an obvious dose-dependent loss of neuron-specific proteins, as ascertained by immunoblotting experiments, after treatment of explants with rotenone. As in the mixed retinal cell cultures, such rotenone-dependent neuron loss in adult rat retinal explants was also reversed in the presence of glucose (Supplementary Fig. S2). 

Effect of Different Glucose Concentrations and Other Energy Substrates

Western blotting and immunocytochemistry were employed to determine whether either lower glucose levels (5 mM) or the monocarboxylates, lactate and pyruvate, could also protect neurons from rotenone-induced toxicity (Fig. 5). Treatment with rotenone (10 μM) for 24 hours, in the absence of glucose, led to a substantial loss of PGP9.5 protein (83% reduction; Fig. 5A). The presence of glucose (25 mM) increased PGP9.5 protein expression 4.6-fold; glucose at 5 mM was also found to increase both Tau and PGP9.5 expression by 200% and 150%, respectively (Figs. 5B, 5C). In addition, the presence of pyruvate (2 mM) was also able to partially counteract the effect of rotenone: pyruvate reduced the rotenone-induced loss of Tau and PGP9.5 levels in cultures by 2.0-fold and 4.4-fold, respectively. Lactate (2 mM), which may act in brain neurons as an alternative energy substrate, ²¹ failed to elicit any protective effect on neurons against rotenone (Fig. 5). 

Figure 5

 

Neuroprotective effect of different concentrations of glucose and comparison with other energy substrates in retinal cultures treated with rotenone. (A) Immunocytochemical analysis of retinal cultures after incubation of cells in media containing different metabolic substrates. In order to visualize neurons, PGP 9.5-immunoreactive cells were labeled and quantified and compared with the control group (n = 5 individual experiments comprising five determinations per experiment). (B) Immunoblots of neuronal protein expression form rotenone-impaired (10 μM, 24 hours) cultures in the presence of glucose (5 mM, 25 mM), lactate (2 mM) or pyruvate (2 mM). (C) Quantification of the immunoblotting data normalized for the β-actin levels in each sample and expressed as the ratio compared with control cell cultures. *P < 0.05, **P < 0.01 compared with control cultures. †P < 0.05, ††P < 0.01 compared with rotenone-treated group. The results shown represent the mean ± SD of three independent experiments.

Metabolic Inhibition

In this set of experiments, the effects of the glycolytic inhibitor, iodoacetate (IOA) and the PPP inhibitor, 6-aminonicotinamide (6-AN), were investigated (Fig. 6). Iodoacetate completely abolished the protective effect of glucose to both neurons and glial. Inhibition of the PPP by 6-AN caused a partial loss of the protective effect of glucose against rotenone toxicity (Fig. 6). Pyruvate can act as an antioxidant ²² and so we also tested the effects of the well-characterized antioxidant, trolox, as a positive control. From our results, mitochondria impaired retinal neurons and glia with PPP inhibition were partially protected by 2 mM pyruvate (Fig. 6). 

Figure 6

 

Quantification of immunocytochemical analysis of 7-day-old cultures grown in 25-mM glucose exposed to the outlined treatments (all for 24 hours). PGP9.5-positive neurons (A) and GFAP-positive glial cells (B) were selected as being typical of their general cell type. Iodoacetic acid led to a loss of both neurons and glia, whereas 6-AN mainly caused a loss of neurons. Data are expressed as percentage of remaining labeled cells as related to control cells (without glucose). **P < 0.01, compared with untreated control cells by two-way ANOVA test followed by a Bonferroni correction. †P < 0.05, ††P < 0.01 compared with cells treated with 10 μM rotenone without glucose by two-way ANOVA test followed by a Bonferroni correction. §P < 0.05, §§P < 0.01 compared with group treated with 10 μM rotenone with 25 mM glucose by two-way ANOVA test followed by a Bonferroni correction (n = 6 individual experiments comprising five determination per experiment).

Oxidative Stress and the Cellular NADP⁺/NAD(P)H Balance

Mitochondrial ROS production was qualitatively and quantitatively assessed using the fluorophore, DHE (Figs. 7A, 7B). Incubation of rat retinal cultures treated with rotenone (100 nM–1 μM) for 24 hours led to a concentration-dependent increase in the production of ROS in cultured cells (4.0-fold, 5.2-fold, and 3.9-fold, respectively compared with the control group). This was observed up to a concentration of rotenone of 1 μM, but at higher concentrations (e.g., 10 μM, Fig. 7B), many cells had died and so the recorded value was relatively decreased. Glucose significantly blunted the observed increase in ROS production (Figs. 7A, 7B); there was no significant difference between ROS levels in cells with or without glucose when the concentration of rotenone applied was 10 μM. 

Figure 7

 

Analysis of ROS formation in 25-mM glucose-protected retinal cultures and the effect of 6-AN. (A) Twenty-five millimoles per liter of glucose reduced the rotenone-induced production of ROS; this effect was abolished in the presence of the PPP pathway inhibitor, 6-AN. The redox-sensitive, cell-permeable fluorophore dihydroethidium was imaged in rat retinal cultures using a fluorescence microscope (200×) to further confirm the mitochondrial production of ROS, as shown in untreated cultures and cells exposed to rotenone (100 nM, 1 μM, and 10 μM) with or without 25-mM glucose and the administration of 6-AN. (B) Quantitative analysis of the mitochondrial ROS production, assessed with the DHE (expressed as fluorophore dihydroethidium units/μg cell protein), in retinal cultures treated with rotenone (100 nM, 1 μM), glucose (25 mM) and 6-AN (10 μM) for 24 hours. Data were obtained from four separate experiments (n = 4 individual experiments comprising three determination per experiment). *P < 0.05 compared with cells without glucose group. †P < 0.05 compared with rotenone with 25 mM glucose-treatment group. ††P < 0.01 compared with rotenone with 25-mM glucose-treatment group.

We next investigated the role of the PPP in the protective effect of glucose against rotenone-stimulated ROS production in retinal cultures. In the presence of 6-AN, the protective effect of glucose against rotenone-induced ROS production was completely lost (Figs. 7A, 7B). Furthermore, by examining the ratio of NADP⁺ to NAD(P)H in treated cells, it was determined that rotenone induced a significant decrease (33 ± 6%) in cellular reducing capacity (an increased NADP⁺ to NAD(P)H ratio), which was completely reversed in the presence of glucose (25 mM; Table). Again, co-incubation with 6-AN (10 μM, 100 μM) partially, but significantly blocked this effect. 

Table

 

Analysis of NADP⁺/NADPH Ratio in 25-mM Glucose-Protected Rat Retinal Cells and the Effect of 6-AN

Treatment	NADP+/NADPH Ratio of Control Group
Control	100
25 mM Glu	87 ± 17
1 μM Rot	327 ± 3†
25 mM Glu+1 μM Rot	90 ± 16
25 mM Glu+1 μM Rot+10 μM 6-AN	155 ± 21*‡
25 mM Glu+1 μM Rot+100 μM 6-AN	208 ± 34†‡

 The cellular NADP⁺/NADPH total ratio after a 24 h/37°C incubation of glucose (25 mM), rotenone (1 μM) or 6-AN (10,100 μM) was tested by using a using chemiluminescence assay (n = 3 individual experiments comprising three determination per experiment). Data are expressed as percentage of remaining labeled cells as related to control cells. Glu, glucose.

*   P < 0.05.

†   P < 0.01 compared with cells without glucose group by two-way ANOVA test followed by a Bonferroni correction.

‡   P < 0.05 compared with rotenone with 25-mM glucose treatment group by two-way ANOVA test, followed by a Bonferroni correction.

Intracellular ATP Analysis

Incubation with rotenone (1 μM and 10 μM; 24 hours) in the absence of glucose decreased the intracellular ATP level to 75 ± 9% and 50 ± 8%, respectively, with respect to the concentration measured in control cells. The presence of glucose (25 mM) significantly counteracted the rotenone-induced depletion of cellular ATP levels; the reversal of the rotenone-induced loss was by 90 ± 9% (1 μM rotenone) and 70 ± 8% (10 μM rotenone) of control levels, respectively. Neither glucose nor pyruvate significantly attenuated the decrease in ATP in the presence of IOA (Fig. 8). 

Figure 8

 

Effect of 25-mM glucose and glycolysis pathway inhibition on the ATP concentration in rat retinal cells treated with rotenone. Cultured cells were incubated with or without 1 μM, 10 μM rotenone in the presence of the indicated compounds: glucose (25 mM), lactate (2 mM), or pyruvate (2 mM). The ATP concentration was determined by the firefly luciferase assay (see Methods section). This result represents the mean ± SEM of four independent experiments. *P < 0.05, **P < 0.01 compared with cells without glucose group. ††P < 0.05 compared with rotenone-treated group with 25 mM glucose.

Determination of Cell Death

Rotenone-induced cell death in retinal cultures was determined by both the live/dead cell assay and the TUNEL assay (Fig. 9A). As the rotenone concentration was elevated from 1 μM to 10 μM, there was an obvious increase in cell death as shown by both assays. When 1 μM rotenone was applied, glucose significantly prevented cell death as determined in both assays. When the higher concentration of rotenone (10 μM) was applied, however, glucose only partially prevented the EthD-1–positive cell death (Fig. 9B). 

Figure 9

 

The mode of rat retinal cell death after exposure to rotenone and/or 25 mM glucose. Cultured cells were incubated for 24 hours in 1 μM or 10 μM rotenone in Dulbecco's modified Eagle's medium (DMEM) supplemented with the indicated substrates. In live and dead assay, live cells were distinguished by the presence of cell permeant calcein-AM (green) while the EthD-1 was used to produce red fluorescence in dead (necrotic) cells. Fluorescence was imaged in cells (A) using a fluorescence microscope (200×); total dead (EthD-1) and apoptotic (TUNEL) cells were illustrated in (B). Results represent the mean ± SEM of four independent experiments. *P < 0.05, **P < 0.01 compared with untreated cells; †P < 0.05 compared with rotenone-treated group without glucose. Scale bar: 20 μM.

Discussion

In the current study, we have demonstrated that glucose attenuates rotenone-induced neuronal death in retinal cultures. The mechanisms of glucose-induced neuroprotection comprised both the glycolytic synthesis of ATP and a reduction in the production of ROS via glucose entry into the PPP. 

The importance of glycolysis with regard to energy metabolism in the vertebrate retina was first demonstrated by Noell in 1951. ²³ Later, Winkler and others ²⁴ showed that the retina was capable of upregulating glycolytic ATP production in the presence of reduced oxygen tension (the Pasteur effect); when O₂ is unavailable, 50% to 70% of glucose taken up by the mammalian retina is metabolized via anaerobic glycolysis and preferentially converted to lactate. ^25,26 Moreover, unlike the brain, this tissue also tends to undertake glycolytic energy production even in the presence of oxygen (“aerobic glycolysis”; the Warburg effect). ²⁷  

In rats, the mean glucose concentration in blood, vitreous, and retina is 7.15 ± 0.82 mM, 3.81 ± 0.42 mM, 5.3 ± 0.24 μg/mg, respectively. ²⁸ Hence, 5-mM glucose corresponds approximately to a physiological vitreous concentration and 25 mM corresponds to severe hyperglycemia. 

It is noteworthy that despite the presence of 10 μM rotenone, the retinal cultures can maintain approximately 50% of their basal ATP levels, suggesting that OXPHOS was impaired, but not abolished in this model. With the inhibition of the glycolytic enzyme glyceraldehyde-3-phosphate dehydrogenase by IOA, there was a dramatic loss of cells, concurrent with a marked reduction in cellular ATP content, consistent with Noell's original experiments. ²³ In our mitochondria-compromised retina model, GFAP-positive glia cells were relatively resistant to rotenone (1 μM) cytotoxity (Fig. 2), but extremely vulnerable in the presence of IOA (Fig. 6). Collectively, these data indicate that, in the present model, retinal cells rely more on glycolytic ATP production than oxidative respiration, a finding that is consistent with previous work. ^29–31  

In the present study, pyruvate was able to confer some protection to retinal cells. Since pyruvate did not increase ATP levels in the presence of rotenone (data not shown), and so was not acting to produce ATP via its own catabolism, then its protective effect may have been due to inherent antioxidant properties (Table). ³² As a positive control, trolox, a widely used antioxidant, was also shown to inhibit rotenone-induced retinal cell death in this study (Fig. 6). 

Whether or not lactate is the end product of glycolysis in the brain rather than pyruvate is controversial, ³³ and to our knowledge, this has not been elucidated clearly in the retina. To test this idea, lactate was applied in the present study instead of glucose to determine whether this metabolite could itself act to protect neurons. For lactate to act as a protecting influence to neurons via its metabolism, it would have to be converted to pyruvate by the action of lactate dehydrogenase. Pyruvate was not metabolized directly, because it did not reverse rotenone-induced cellular ATP reduction. Pyruvate was, however, able to produce some protection against rotenone, but likely as a result of antioxidant properties. Lactate, on the other hand, could be converted to pyruvate via lactate dehydrogenase; this reaction would produce NADH, which would increase cellular reducing power. ³⁴ In the current study, although lactate did not significantly protect cells against rotenone toxicity, it should be borne in mind that, conversion of lactate to pyruvate may have partially contributed to the protective effect of the latter compound. In a previous in vivo study, however, intraocular lactate delivery failed to protect retinal neurons against an ischemic injury. ³⁵ Therefore, the role of this compound in retinal cell metabolism remains unclear. 

Aside from the positive role that glucose plays in enhancing cellular energy production in the form of ATP, studies have also demonstrated that this sugar can directly provide cytoprotection through its oxidation via the PPP. The PPP acts as a key component of the cytosolic NAD(P)H regenerating system ^36,37 ; NAD(P)H itself represents a crucial cellular reducing equivalent, which assists neurons in coping with situations of oxidative stress, such as the accumulation of either ROS ³⁸ or peroxynitrite anions. ³⁷ In neurons, for example, it is well known that the reducing power of the PPP can prevent apoptotic cell death, ^36,39 via inhibition of cytochrome c-mediated apoptosis by a mechanism strictly dependent on glucose metabolism. ³⁸ Furthermore, NAD(P)H is known to be produced via the PPP in the retina, where, amongst other roles, it is involved in the recycling of photopigments in photoreceptors. ^29,40 In the present study, we found that inhibition of this pathway by co-application of the 6-phosphogluconate dehydrogenase inhibitor, 6-AN, reduced the protective effect of glucose against rotenone-induced retinal cell toxicity. The NADP⁺/NAD(P)H balance was also affected by administration of 6-AN (Table) in the glucose-protected group. 

All of the above data indicate that the PPP plays a partial role in the glucose-induced protective effect. Previous research, however, has indicated that ATP production rather than PPP-derived antioxidants represent the most likely neuroprotective mechanism for glucose in the intact retina. For example, Winkler et al. ¹³ showed that glucose entry into the PPP is not significantly elevated in the isolated retina when the glucose concentration was elevated from 5 mM to 30 mM. Furthermore, Casson et al. ⁴ showed that delivery of intraocular glucose was protective during a period of ischemia, but not if delivered in the reperfusion period. The present demonstration of the partial protection by the PPP in our study may be due to differences in experimental models. In the present model, for example, rotenone is used to promote metabolic stress. This compound not only inhibits the OXPHOS pathway, but also enhances the production of ROS (Fig. 7). It is therefore likely that in maintaining cellular redox potential in cells in the current study, glucose was also metabolized via the PPP in order to counteract the detrimental effects of ROS. 

Overall, the present evidence suggests that the neuroprotective effect of glucose against mitochondrial toxicity is predominantly mediated via glycolytic ATP production, but that entry of this sugar into the PPP also contributes to its neuroprotective action, by increasing cellular reducing power. 

Supplementary Materials

pdf S1 

Acknowledgments

The authors thank Mark Daymon and Jim Manavis for their helpful advice and skilled technical assistance. 

Supported by a grant from the National Health and Medical Research Council (626964), the Council Scholarship of China (CSC: 2010627027; GH). 

Disclosure: G. Han, None; J.P.M. Wood, None; G. Chidlow, None; T. Mammone, None; R.J. Casson, None