SEARCH

SEARCH BY CITATION

Keywords:

  • glutathione;
  • glycolysis;
  • lactate;
  • oxidative stress;
  • pentose-phosphate pathway

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments and conflict of interest disclosure
  7. References
Thumbnail image of graphical abstract

Vitamin C is an essential factor for neuronal function and survival, existing in two redox states, ascorbic acid (AA), and its oxidized form, dehydroascorbic acid (DHA). Here, we show uptake of both AA and DHA by primary cultures of rat brain cortical neurons. Moreover, we show that most intracellular AA was rapidly oxidized to DHA. Intracellular DHA induced a rapid and dramatic decrease in reduced glutathione that was immediately followed by a spontaneous recovery. This transient decrease in glutathione oxidation was preceded by an increase in the rate of glucose oxidation through the pentose phosphate pathway (PPP), and a concomitant decrease in glucose oxidation through glycolysis. DHA stimulated the activity of glucose-6-phosphate dehydrogenase, the rate-limiting enzyme of the PPP. Furthermore, we found that DHA stimulated the rate of lactate uptake by neurons in a time- and dose-dependent manner. Thus, DHA is a novel modulator of neuronal energy metabolism by facilitating the utilization of glucose through the PPP for antioxidant purposes.

We proposed that the ascorbic acid (AA) taken up by neurons is rapidly oxidized to dehydroascorbic acid (DHA), which inhibits glycolysis and activates the pentose phosphate pathway (PPP), consequently producing NADPH, a critical antioxidant in the recycling of oxidized glutathione (GSSG). In these metabolic conditions, neurons increase lactate uptake, probably using it as an energy source. This data supported the idea that DHA could play a critical role in the modulation of energy metabolism in neurons.

Abbreviations used
AA

ascorbic acid

DHA

dehydroascorbic acid

DOG

deoxyglucose

G6PD

glucose-6-phosphate dehydrogenase

GFAP

glial fibrilary acid protein

GLUT

glucose transporter

PPP

pentose phosphate pathway

SVCT

sodium/vitamin C cotransporters

Vitamin C is an essential micronutrient required for normal metabolic function of the brain. It is also an important soluble antioxidant (Carr and Frei 1999; Rice 2000; Savini et al. 2008) because of its low redox potential, which is capable of neutralizing a wide range of pro-oxidants (May et al. 2003; Linster and Van Schaftingen 2007; Bowman 2012; Lykkesfeldt 2012). Vitamin C can be found in its reduced form, ascorbic acid (AA), as well as its oxidized form, dehydroascorbic acid (DHA) (Englard and Seifter 1986; Wilson, 2002a; Linster and Van Schaftingen 2007). Vitamin C is ubiquitously present in the brain, reaching concentrations of 400 μM in cerebrospinal fluid, 200–300 μM in the cerebral parenchyma, 10 mM in cortical neurons, and 1 mM in cortical astrocytes (Rice and Russo-Menna 1998; Rice 2000). The high concentration of vitamin C within neurons and astrocytes is attributable to the presence of specific and efficient transporters (Wilson 2005; Savini et al. 2008; Corti et al. 2010).

AA, but not DHA is transported by a family of sodium/vitamin C cotransporters (SVCT1 and SVCT2) (Daruwala et al. 1999; Tsukaguchi et al. 1999; Garcia Mde et al. 2005). DHA transport is mediated by facilitated hexose transporters (GLUTs), specifically the GLUT-1, -3, and -4 isoforms (Rumsey et al. 1997, 2000; Astuya et al. 2005). Cortical neurons express both SVCT2 (Castro et al. 2001; Astuya et al. 2005; Caprile et al. 2009) and GLUT3 (Mueckler 1994; Vannucci et al. 1997; Castro et al. 2001) and therefore are capable of AA and DHA uptake, respectively.

Under physiological conditions, DHA concentrations are extremely low. However, extracellular AA is massively oxidized to DHA in certain pathological conditions, such as cerebral infarction and deregulation of glutamatergic neurotransmission, thus increasing its uptake and intracellular concentration in neurons (Coyle and Puttfarcken 1993; Kim et al. 1994; Wilson, 2002a,b). Thus, such an increase in intracellular DHA may result in neuronal damage, although the molecular mechanism remains unknown. In Jurkat cells, DHA stimulated the activity of glucose-6-phosphate dehydrogenase (G6PD) in vitro (Puskas et al. 2000), the rate-limiting step of the pentose phosphate pathway (PPP) (Herrero-Mendez et al. 2009). Here, the effects of DHA on the antioxidant and energy metabolism of neurons were investigated. Intracellular DHA transiently decreased reduced glutathine (GSH) concentrations, which were spontaneously recovered in parallel by increased PPP activity and a decrease in that of glycolysis. These results strongly suggest that the oxidation of AA to DHA during certain pathological conditions has profound metabolic consequences for neuronal energy metabolism.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments and conflict of interest disclosure
  7. References

Animals

All animals were handled in strict accordance with the Animal Welfare Assurance (permit number 2010101A), and all animal work was approved by the appropriate Ethics and Animal Care and Use Committee of the University of Concepción (Chile). Sprague–Dawley rats were used for the experiments. Animals were kept in a 12-h light/12-h dark cycle with food and water ad libitum.

Cell cultures

Neurons were obtained from the forebrains of 17-day-old rat embryos. The dissection was carried out with the samples immersed in dissection buffer containing 10 mM HEPES (pH 7.4, 320 mOsm/L). Tissues were incubated with 0.25% trypsin-0.2% EDTA (w/v) for 15 min at 37°C, and then triturated to homogeneity with a fire-polished Pasteur pipette. Cells were seeded at 5 × 105 cells/cm2 in poly-d-Lysine-coated culture dishes and cultured in Dulbecco's modified Eagle's medium (DMEM) (Invitrogen, Grand Island, NY, USA) containing 10% (v/v) fetal bovine serum (Thermo Fisher Scientific Inc., Waltham, MA, USA). After 30 min, the culture medium was changed to Neurobasal medium supplemented with B27 DMEM (Invitrogen), 2 mM l-glutamine, 100 U/mL penicillin, 100 mg/mL streptomycin, and 2.5 mg/mL fungizone (Thermo Fisher Scientific Inc). Astroglial-rich cultures were isolated from rat brain cortex of 1-day-old post-natal rats using the same protocol described above. However, the isolated cells were cultured in minimal essential medium supplemented with 10% fetal bovine serum, 2 mM l-glutamine, 100 U/mL penicillin, and 100 mg/mL streptomycin. Both neuronal and astroglial cell cultures were incubated in 5% CO2 in a humidified environment at 37°C. The cultures reached confluence within 2 weeks, when each 60-mm dish contained approximately 3 × 106 cells. All cells were used for experiments after 2–3 weeks.

Immunocytochemistry

Neurons were grown on coverslips previously coated with poly-d-lysine, and fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) for 30 min at 22°C. After washing with PBS, cells were incubated in PBS containing 1% bovine serum albumin and 0.2% Triton X-100 for 5 min at 22°C to block non-specific antibody binding. Cells were incubated overnight at 22°C with the following primary antibodies: rabbit anti-GLUT1 (1 : 200, Millipore, Billerica, MA, USA), rabbit anti-GLUT3 (1 : 100, Alpha Diagnostic, San Antonio, TX, USA), goat anti-SVCT2 (G19, 1 : 100, Santa Cruz Biotechnology, Santa Cruz, CA, USA), mouse anti-Tubulin βIII (1 : 2000, Promega, Madison, WI, USA), and rabbit anti-glial fibrilary acid protein (1:500, Dako, Via Real Carpinteria, CA, USA). Cells were incubated for 2 h with Cy2- or Cy3-conjugated affinity purified donkey anti-goat IgG, donkey anti-rabbit IgG, or donkey anti-mouse IgG (1 : 200; Jackson Immuno Research, West Grove, PA, USA) at 22°C. The cells were then incubated with the nuclear marker, TOPRO-3 (1 : 1000, Invitrogen). Negative controls consisted of cells incubated with preimmune serum and both primary antibodies pre-absorbed with the respective peptides used to raise them. Preparations were analyzed using confocal laser microcopy (D-Eclipse C1 Nikon, Tokyo, Japan).

Uptake analysis

After 5 days in culture, neurons were carefully selected under the microscope to ensure that only plates showing uniformly grown cells were used. The cells were washed with incubation buffer (15 mM HEPES, 135 mM NaCl, 5 mM KCl, 1.8 mM CaCl2, and 0.8 mM MgCl2) and incubated in the same medium for 5 min at 22°C. Uptake assays were carried out in 500-μL incubation buffer containing one of the following substrates: 1–1.2 μCi 2-deoxy-d-[1,2-(N)3H]glucose (2-DOG) (26.2 Ci/mmol; Dupont NEN, Boston, MA, USA), 1-4 μCi L-[14C(U)]lactic acid sodium salt (> 100 mCi [3.70 GBq]/mmol; PerkinElmer-NEN, Boston, MA, USA) or 0.1 μCi 14C-AA/DHA (specific activity 4 mCi/mmol), to a final concentration of 100–500 μM. DHA was obtained from AA by adding 0.02 U ascorbate oxidase/mL (Sigma, St. Louis, MO, USA) to 1 mM AA. Uptake was stopped by washing the cells with ice-cold PBS. Cells were lysed in 0.5 mL lysis buffer (10 mM Tris-HCl, pH 8.0, and 0.2% sodium dodecyl sulfate), and the incorporated radioactivity was assayed by liquid scintillation spectrometry (Castro et al. 2001; Astuya et al. 2005). For the pharmacological experiments, the inhibitors were incubated for 10 min with a radioactive substrate in the following concentrations: 20 μM cytochalasin B, 20 μM cytochalasin E, 100 μM quercetin, and 1 mM α-cyano-4-hydroxycinnamate (4αCIN, Sigma-Aldrich, St. Louis, MO, USA).

Determination of the glycolytic rate

Glycolytic rates were determined using the methods described previously (Trus et al. 1981). Neurons, cultured as described above and 1 mM glucose, were incubated with different concentrations of DHA (0–1 mM) for several time periods (15–60 min). Cells (1 × 106) were placed into tubes containing 5 mM glucose, and then washed twice in Krebs Henseleit solution (11 mM Na2HPO4, 122 mM NaCl, 3.1 mM KCl, 0.4 mM KH2PO4, 1.2 mM MgSO4, and 1.3 mM CaCl2, pH 7.4) containing the appropriate concentration of glucose. After equilibration in 0.5-mL Hank's balanced salt solution/glucose at 37°C for 10 min, 0.5 mL Hank's balanced salt solution containing [5-3H]glucose (GE Healthcare Life Sciences, Cleveland, OH, USA) at various concentrations and with a final specific activity of 1–3 disintegrations/min/pmol (approximately 1 mCi/mmol), was added. After 15–60 min, the incubation was stopped by adding 250 μL of 10% perchloric acid. Aliquots of 100 μL were then transferred to another tube, placed inside a capped scintillation vial containing 0.5 mL water, and incubated at 45°C for 48 h. After this vapor-phase equilibration step, the tube was removed from the vial, a scintillation mixture was added to the vial, and 3H20 content was determined by counting over a 5-min period (Herrero-Mendez et al. 2009; Rodriguez-Rodriguez et al. 2012).

Measurement of glucose oxidation through the pentose phosphate pathway

Glucose oxidation via the PPP was determined as described previously (Hothersall et al. 1979), which is based on determining the difference in 14CO2 production from [1-14C]glucose (decarboxylated in the 6-phosphogluconate dehydrogenase-catalyzed reaction and in the Krebs cycle) and [6-14C]glucose (only decarboxylated in the Krebs cycle). Neurons, cultured as described above and 1 mM glucose, were incubated with 1 mM DHA for 30 min, after which the medium was removed, and cells were washed with ice-cold PBS and collected by trypsinization. Cell pellets were resuspended in O2-saturated Krebs Henseleit buffer (11 mM of Na2HPO4, 122 mM of NaCl, 3.1 mM of KCl, 0.4 mM of KH2PO4, 1.2 mM of MgSO4, and 1.3 mM of CaCl2, pH 7.4), and 500 μL of this suspension (~ 106 cells) was placed in Erlenmeyer flasks with another 0.5 mL of the Krebs Henseleit solution containing 0.5 μCi d-[1-14C]glucose or 2 μCi d-[6-14C]glucose and 5.5 mM d-glucose (final concentration). The Erlenmeyer flasks were equipped with a central well containing an Eppendorf tube with 500-μL benzethonium hydroxide. The flasks were flushed with O2 for 20 s, sealed with rubber caps, and incubated for 60 min at 37°C in a water bath with shaking. Incubations were stopped by injecting 0.2 mL 1.75 M HClO4 into the main well, although shaking was continued for a further 20 min to facilitate trapping 14CO2 by benzethonium hydroxide. Radioactivity was assayed by liquid scintillation spectrometry (Konagaya et al. 1990; Larrabee 1990).

Determination of reduced glutathione

Neurons were incubated in DMEM (1 mM glucose) in the absence (control) or presence of 1 mM AA or 0.1–1 mM DHA. After the incubation period, cells were washed with ice-cold PBS and immediately collected by scraping with 0.5 mL 1% (wt/vol) sulfosalicylic acid. Cell lysates were transferred to 1.5-mL Eppendorf tubes and underwent centrifugation at 13 000 g for 5 min at 4°C, and the supernatants were used for glutathione determination. The glutathione was measured as described previously (Tietze 1969; Dringen and Hamprecht 1996). In brief, 10 μL of the supernatant was transferred into microtiter plate wells and diluted with 90 μL water. After the addition of 100 μL-reaction mixture [1 mM EDTA, 0.3 mM 5,59-dithio-bis(2-nitrobenzoic acid), 0.4 mM NADPH, and 1 U/mL glutathione reductase in 0.1 M sodium phosphate buffer (pH 7.5)], the increase in absorbance at 405 nm was monitored at 15-s intervals for 2.5 min using a microtiter plate reader (Tecan, Mannedorf, Switzerland). GSx concentrations (GSx = GSH+2xGSSG) were calculated with the software provided by the plate reader using GSSG standard solutions (0–50 μM), which were treated in exactly the same way as the samples.

Determination of glucose-6-phosphate dehydrogenase activity

Cells were pre-treated with 1 mM AA or DHA for 30 min at 37°C after which cells were washed with PBS, collected by scraping in 0.25% trypsin-0.2% EDTA (w/v), and pelleted at 500 g for 5 min at 4°C. The cell pellet was resuspended in isolation medium (250 mM sucrose, 20 mM HEPES, 10 mM KCl, 1.5 mM MgCl2, 1 mM EDTA, 1 mM dithiothreitol, 2 mg/mL aprotinin, 1 mg/mL pepstatin A, 2 mg/mL leupeptin), sonicated at 4°C, and pelleted by centrifugation at 1500 g for 5 min at 4°C. Subsequently, the pellet was discarded, and the supernatant was further separated by centrifugation at 13 000 g for 30 min at 4°C. Finally, the supernatant was used to quantify G6PD activity in a reaction buffer consisting of 25 mM Tris-HCl, 1 mM dithiothreitol, 0.5 mM NADP/Na+, 2 mM MgCl2, 1 mM ATP, and 10 mM glucose-6-phosphate for 45 min at 37°C. The reaction was stopped by adding 10% trichloroacetic acid, and the generation of NADPH was measured at 340 nm (Tsai and Chen 1998).

Measurement of intracellular AA

Neurons and astrocytes were pre-loaded with 1 mM 14C-AA or 1 mM 14C-DHA for 5 min at 37°C in incubation buffer (15 mM HEPES, 135 mM NaCl, 5 mM KCl, 1.8 mM CaCl2, and 0.8 mM MgCl2). AA and DHA levels were measured essentially as described by Guaiquil et al. (1997). The cells were lysed in 60% methanol, 1 mM EDTA (pH 8.0). HPLC analyses were performed using a Whatman strong anion exchange reverse phase column (Partisil 10 SAX, 4.6 mm 3 25 cm, 10-mm particle) with a silica-pre-conditioning column (Whatman, Pittsburgh, PA, USA). The HPLC system was equipped with a radioactivity detector arranged in series. The elution conditions were as follows: temperature, 25°C; flow, 1.0 mL/min from 0 to 20 min, and 2.0 mL/min from 20 to 60 min; buffers, buffer A (0.007 M KH2PO4, 0.007 M KCl, pH 4.0) and buffer B (0.25 M KH2PO4, 0.5 M KCl, pH 5.0); mobile phase, isocratic buffer A (0–5 min), linear gradient of buffer A 100%, buffer B (5–20 min), isocratic buffer B (20–37 min), linear gradient of buffer B 100%, buffer A (37–52 min), and isocratic buffer A (52–60 min).

Statistical analysis

Data represent mean ± SD of three experiments with each determination done in triplicate. Statistical comparison between two or more groups of data was carried out using anova (followed by Bonferroni post-test). p > 0.05 was considered to be statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments and conflict of interest disclosure
  7. References

Immunocytochemical and functional characterization of AA and hexose transporters in rat primary cortical neurons

The neuronal phenotype of the 5-DIV primary neurons was first characterized using immunocytochemical analysis of the neuronal marker, tubulin βIII, and the glial marker, glial fibrilary acid protein. Analysis of the resultant images revealed a neuronal enrichment of ~ 98% (Fig. 1a, b). In addition, the expression of the AA transporter, SVCT2, was higher in the neuronal soma (Fig. 1c) than in the processes (Fig. 1d, arrows). Immunocytochemical analysis of GLUT1 and GLUT3 transporters showed a weakly positive immunoreaction for GLUT1 transporter, which was exclusively present in the neuronal soma (Fig. 1e, f, arrows). In contrast, an intense immunoreaction for GLUT3 was observed in the neuronal soma (Fig. 1g) and, to a lesser degree, in the cellular extensions (Fig. 1h). Thus, the primary cultures of cortical neurons were differentiated and expressed the transporters involved in vitamin C uptake.

image

Figure 1. Immunocytochemical analysis of sodium/vitamin C cotransporter (SVCT)2, glucose transporter (GLUT)1 and GLUT3 transporters in primary neurons. Immunofluorescence analysis was performed using secondary antibodies labeled with Cy2 (green) or Cy3 (red). The nucleus was stained with TOPRO (blue). After 5 days in culture, the neuronal soma and neuronal processes showed intense immunoreaction with anti-Tubulin βIII, a classic neuron marker (a); only 2% of the cells was positive for the glial marker, glial fibrilary acid protein (GFAP) (b). An intense expression for the SVCT2 transporter was detected in neurons, especially in the neuronal soma (c–d). Diffuse GLUT1 expression was detected in the neuronal soma (e and f). An intense immunoreaction against GLUT3 was found mainly in neuronal soma as well as in some cell extensions (g and h).

Download figure to PowerPoint

Next, SVCT2 transporter function was analyzed by determining AA uptake. In a preliminary experiment, the incorporation of AA was linear up to 40 min, showing a maximum uptake of 127 ± 25 pmol/106 cells at 120 min (Fig. 2a). AA uptake was inhibited by 77 ± 7% after Na+ was replaced with choline, and by 57 ± 11% in the presence of 100 μM quercetin (Fig. 2b). However, it was not affected by 20 μM cytochalasin B, an inhibitor of GLUT transporters (Fig. 2b), ruling out the presence of DHA in the transport mixture. Transport of 2-DOG by GLUT transporters was linear up to 45 s, reaching a maximum value of 7.93 ± 0.1 pmol/106 cells at 120 s (Fig. 2c). 2-DOG uptake was inhibited by 83 ± 5% in the presence of 20 μM cytochalasin B, but was not affected by cytochalasin E (Fig. 2d) or by replacing Na+ with choline (not shown). The transport of DHA was also linear up to 2 min, with a maximum uptake value of 287 ± 30 pmol/106 cells (Fig. 2e). In the presence of 20 μM cytochalasin B, DHA transport was inhibited by 78 ± 9%; it was not inhibited with 20 μM cytochalasin E (Fig. 2f) or choline replacement (data not shown). These results confirm that, in our hands, primary neurons in culture incorporate both redox forms of vitamin C.

image

Figure 2. Functional analysis of sodium/vitamin C cotransporter (SVCT)2, glucose transporter (GLUT)1 and GLUT3 transporters in primary neurons. The uptake of 100 μM ascorbic acid (AA) (a), 250 μM 2-DOG (b), and 100 μM dehydroascorbic acid (DHA) (c) was analyzed over time at 22°C. Analysis of AA uptake in the presence of 20 μM cytochalasin B (Cyt B), choline, and 100 μM Quercetin was also assessed (d). In the pharmacological analysis of 2-DOG (e) and DHA (f) uptake, cells were pre-incubated with 20 μM cytochalasin B or E (Cyt B or Cyt E). Data represent the mean ± SD of three experiments, each performed in triplicate. *p < 0.01; **p < 0.005, Bonferroni test.

Download figure to PowerPoint

AA is transformed into DHA in primary neurons

The redox status of intracellular vitamin C in primary neurons was next investigated. After the neurons were pre-incubated with 1 mM 14C-DHA or 1 mM 14C-AA for 5 min, the intracellular AA concentrations were analyzed thereafter by ion exchange HPLC, and expressed as the percentage of AA with respect total vitamin C. As shown in Fig. 3a, neurons pre-incubated with 14C-DHA only slightly converted it into AA, with a maximum reduction of 34.6 ± 6% at 60 min; however, these cells were no longer capable of reducing 14C-DHA, at least up to 180 min. In contrast, 14C-AA oxidation was 77 ± 11% at 60 min (Fig. 3b), suggesting that the ability of neurons to oxidize AA is much higher than their ability to reduce DHA. In primary cultures of cortical astrocytes, which are known to express a prominent reductive capacity (Siushansian et al. 1997; Dringen 2000; Dringen et al. 2000), as well as a high expression level of SVCT2 transporter (Berger and Hediger 2000), AA was maintained in its reduced form, at least up to 60 min (Fig. 3c). Together, these results indicate that neurons in primary culture efficiently uptake and oxidize AA into DHA. Finally, we measured the efflux of DHA from neurons and astrocytes after administration of AA. While DHA efflux was not observed in astrocytes, neurons released 38% DHA after 60 min (Fig. 3d).

image

Figure 3. Ascorbic acid (AA) is transformed into dehydroascorbic acid (DHA) in primary neurons. Cultured neurons were incubated with 1 mM 14C-DHA (a) or 1 mM 14C-AA (b) at 37°C in incubation buffer. After 5 min, the cells were washed and maintained in incubation buffer without radiolabeled AA or DHA. Then, the intracellular AA levels were quantified after lysing the cells at different time points and analyzing the lysates by HPLC using a Whatman Partisil 10 SAX column and a radioactive detector. The DHA eluted at 4.4 min and AA eluted at 10.4 min. (c) The astrocytes were treated with the same protocol. (d) DHA efflux in neurons and astrocytes. After cultured neurons and astrocytes were incubated with 1 mM 14C-AA, DHA efflux was determined over 60 min. Data represent the mean ± SD of three experiments, each performed in triplicate *p < 0.001; **p < 0.005. Bonferroni test.

Download figure to PowerPoint

DHA transiently oxidizes GSH in primary neurons

Because most intracellular vitamin C was converted into its oxidized form by neurons and, given the known pro-oxidizing ability of DHA, the effect of DHA on the redox status of the antioxidant, glutathione, was determined in primary neurons. The concentrations of both reduced (GSH) and oxidized (GSSG) glutathione were analyzed after incubation of the neurons with DHA. As shown in Fig. 4a, a concentration-dependent decrease in GSH concentration was observed after a 45-min incubation with DHA, which was exhausted (by 97%) at 1 mM DHA. Analysis at further time points revealed a significant recovery of GSH concentration in neurons treated with 1 mM DHA, reaching 57 ± 9% of the initial value at 90 min (Fig. 4b). These results indicate that DHA induces transient oxidation of GSH, suggesting the presence of a metabolic switch able to provide the necessary reduction equivalents to restore GSH from GSSG.

image

Figure 4. Dehydroascorbic acid (DHA) transiently oxidizes GSH in primary neurons. Neurons were incubated with different concentrations of DHA (0–1 mM) for 45 min, and GSH concentrations were determined. (b) To maintain a relatively constant concentration of DHA, fresh DHA was added every 10 min. The absolute GSH value in the control condition was 17.2 ± 3.2 nmol/mg protein. The data represent the mean ± SD of three experiments, each performed in triplicate. *p < 0.01; **p < 0.005, Bonferroni test.

Download figure to PowerPoint

DHA switches glycolysis to PPP in primary neurons

It has been widely reported that the PPP provides the necessary equivalents, in the form of NADPH(H+), to support GSH regeneration from GSSG (Garcia-Nogales et al. 2003). Moreover, upon oxidative stress, neurons respond by stimulating glucose consumption through the PPP to sustain reduced GSH (Ben-Yoseph et al. 1996). Thus, the effects of DHA, which initially triggered GSH oxidation, on PPP activity were assessed. As shown in Fig. 5a, incubation of neurons with DHA triggered a rapid (within 30 min) and considerable increase in the rate of 14C-glucose oxidation through the PPP (by ~ 2.5-fold). Furthermore, DHA promoted a ~5-fold increase in the activity of G6PD, the rate-limiting step of the PPP (Fig. 5b). Because glucose consumption through the PPP may limit the availability of glucose-6-phosphate for glycolysis, the rate of [5-3H]glucose consumption through the glycolytic pathway was next determined. As shown in Fig. 5c, DHA induced a concentration-dependent decrease in the glycolytic rate, reaching a 50 ± 8% reduction at 1 mM DHA. Furthermore, the decrease in glycolysis by DHA occurred rapidly, as a significant reduction time-dependent decrease was observed after 30 min (Fig. 5d). Together, these data indicate that DHA accumulation promotes a decrease in GSH, thereby stimulating glucose consumption through the PPP at the expense of the glycolytic pathway as an attempt to restore GSH.

image

Figure 5. Dehydroascorbic acid (DHA) shifts glycolysis to pentose phosphate pathway (PPP) in primary neurons. Neurons were incubated with 1 mM DHA for 30 min, after which PPP activity, assessed as the difference in 14CO2 release from [1-14C]- and [6-14C]glucose (a) and glucose-6-phosphate dehydrogenase (G6PD) activity (b) were determined. The absolute value of PPP activity in the control condition was 0.7 ± 0.14 nmol min−1 mg protein−1. To assess the rate of glycolysis, neurons were incubated with several concentrations of DHA for 45 min, after which the rate of [5-3H]glucose conversion into 3H2O was assessed (c), or at different times points in the presence of 1 mM DHA (d). The absolute value for the glycolytic rate in the control condition was 1.51 ± 0.17 nmol min−1 mg−1 protein. Data represent the mean ± SD of three experiments, each performed in triplicate. *p < 0.01; **p < 0.005, Bonferroni test.

Download figure to PowerPoint

DHA stimulates lactate uptake in primary cultures of neurons

Although decreased glycolysis might limit neuronal energy support, we did not observe signs of neuronal death (data not shown). Several authors have demonstrated that lactate is an alternative energy source for neurons (Magistretti 2006; Schurr 2006; Suzuki et al. 2011). Therefore, the influence of DHA on the rate of lactate uptake for mitochondrial oxidation in primary neurons was examined. As shown in Fig. 6a, the uptake of 100 μM lactate was linear with time up to 45 s; a maximum uptake of 492 ± 35 pmol/106 cells was observed. Uptake was inhibited by the monocarboxylate carrier inhibitor, α-cyano-4-hydroxycinnamate (4αCIN, 1 mM), by 67 ± 9% but not with 20 μM cytochalasin B (Fig. 6b). Neurons were then pre-incubated with DHA for 45 min at different concentrations, and the rate of lactate uptake was determined. As shown in Fig. 6c, DHA induced a dose-dependent increase in lactate uptake, reaching a ~ 2-fold increase at 1 mM DHA. Furthermore, DHA-stimulated lactate uptake significantly increased with the incubation time (Fig. 6d). These data indicate that DHA stimulates the uptake of lactate by cortical neurons in primary culture.

image

Figure 6. Dehydroascorbic acid (DHA) stimulates lactate uptake in primary neurons. Uptake of 100 μM lactate over time by neurons (a). Lactate uptake (100 μM) analysis in neurons pre-incubated for 10 min with 20 μM cytochalasin B (Cyt B) or 1 mM 4αCIN (b). To analyze the effect of vitamin C on the accumulation of lactate (100 μM), cells were incubated with DHA at different concentrations (0–1 mM) and different times (0–60 min) (c, d). The results are the mean ± SD of three experiments, each performed in triplicate. *p < 0.01, Bonferroni test.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments and conflict of interest disclosure
  7. References

Primary cultures of cortical neurons prominently expressed the SVCT2 and GLUT3 transporters responsible for the efficient uptake their cognate substrates, AA and DHA, respectively. Moreover, neuronal uptake of AA was rapidly followed by its oxidation to DHA, whereas astrocytes maintain AA in its reduced form. These data are consistent with previous studies describing an astrocyte–neuronal vitamin C recycling system whereby astrocytes uptake extracellular DHA and reduce it to AA, which is then released and taken up by neighboring neurons in which it is oxidized back to DHA and serves as a neuroprotective antioxidant (Hediger 2002; Korcok et al. 2002; Astuya et al. 2005). While this cycle would account for physiological vitamin C metabolism in the nervous system, under certain pathological conditions, such as stroke or inflammation, extracellular AA is oxidized to DHA (Coyle and Puttfarcken 1993; Wilson, 2002a,b; Kim et al. 2005), which in turn can be taken up by neurons, thus contributing to neurotoxicity. The data of this study that show the efficient uptake of DHA by neurons indeed support this notion. The Km for DHA uptake was ~ 5-fold lower than that for glucose uptake by either GLUT1 or GLUT3. Thus, at physiological concentrations of cerebral glucose (~ 2 mM), very low competition would be expected to occur between glucose and DHA (Nualart et al. 2003; Montel-Hagen et al. 2008).

In this study, DHA uptake by neurons triggered oxidative stress, as revealed by a decline in GSH. However, following the initial phase of DHA-induced glutathione oxidation, a spontaneous recovery of reduced glutathione concentration was observed, suggesting the activation of a metabolic process aimed at supplying the necessary reducing equivalents. GSH is regenerated from GSSH by glutathione reductase, an enzyme that requires NADPH(H+) as the electron donor. Among the different metabolic systems responsible for NADPH(H+) regeneration, the PPP represents one of the most active and amenable to regulation by oxidative stress (Ben-Yoseph et al. 1996; Garcia-Nogales et al. 2003). Consistently, the recovery of glutathione redox status coincided with a dramatic increase in the activity of the PPP, suggesting that DHA-induced oxidative stress triggered a neuronal protective response aimed to restore GSH at the expense of glucose consumption through the PPP. This may have taken place through the activation of G6PD, an enzyme that is activated by oxidative stress, after DHA treatment (Garcia-Nogales et al. 2003).

The increased neuronal PPP activity in response to DHA was accompanied by a paralleled decrease in the rate of glycolysis, which is consistent with a shift of glucose-6-phosphate consumption from glycolysis to the PPP, possibly as a consequence of strong G6PD activation. This is consistent with previous data reporting such a metabolic switch by G6PD overexpression in primary cortical neurons (Herrero-Mendez et al. 2009) and PC12 cells (Garcia-Nogales et al. 2003). Interestingly, the shift in glucose consumption away from glycolysis did not seem to produce neurotoxic effects in our hands, which suggests that either the remaining use of glucose through glycolysis is enough to satisfy neuronal energy needs, or that an alternative energy source is being used. Noticeably, our data show that DHA rapidly and persistently stimulated the rate of lactate uptake by primary neurons. Because neurons express the lactate dehydrogenase isoform that preferentially converts lactate into pyruvate, the intracellular lactate may be used for mitochondrial oxidation (Magistretti et al. 1993; Magistretti and Pellerin 1996; Izumi et al. 1997; Barros and Deitmer 2010; Jolivet et al. 2010) and could contribute to the energy needs of neurons. These data are consistent with those reporting that astrocyte-released lactate is taken up by neurons as an energy source essential for neurotransmission (Magistretti 2006; Schurr 2006; Suzuki et al. 2011). Whether the extracellular increase in DHA observed during stroke (Coyle and Puttfarcken 1993; Wilson, 2002a,b; Kim et al. 2005) promotes an increase in lactate uptake by neurons remains unknown; however, this mechanism would provide a transient protection to neurons during ischemic episodes.

In conclusion, neurons efficiently take up both forms of vitamin C, and readily oxidize AA into DHA, which accumulates within the cell. Interestingly, intracellular DHA triggers a biphasic oxidative response in which glutathione is first rapidly oxidized, but thereafter the reduced glutathione status is considerably restored. This takes place at the expense of shifting glucose consumption from glycolysis to the PPP, a metabolic route that efficiently provides the reducing equivalents in the form of NADPH(H+) for GSH regeneration from its oxidized, GSSG form. In addition, DHA stimulated the uptake of lactate, a metabolic substrate that likely contributes to satisfy the high energetic needs of neurons. Together, these results provide, for the first time, the notion that the redox metabolism of vitamin C modulates energy metabolism in neurons. Further studies will assess whether the increased conversion of AA into DHA during stroke might represent a potential novel therapeutic window worth exploring in the future.

Acknowledgments and conflict of interest disclosure

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments and conflict of interest disclosure
  7. References

This work was funded by a FONDECYT grant (1100396) to FN as well as grants from the Ministerio de Economía y Competitividad (SAF2013-41177-R), European Regional Development Fund, Instituto de Salud Carlos III (RETICEF, RD12/0043/0021), and Junta de Castilla y Leon (SA112A12-2) to JPB. DIUC 211031109-1.0 to KS. Centro de Microscopía Avanzada CMA BIOBIO, ACT12 PIA Grant. The authors have declared that no competing interests exist.

All experiments were conducted in compliance with the ARRIVE guidelines. The authors have no conflict of interest to declare.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments and conflict of interest disclosure
  7. References