Activated astrocytes increase dehydroascorbic acid uptake, changing intracellular metabolism and vitamin C recycling and emulating neuropathological conditions

TL;DR

This study shows how vitamin C recycling in astrocytes changes with age, affecting brain cell metabolism and antioxidant defenses.

Contribution

The paper introduces a cellular model of reactive astrocytes with impaired vitamin C recycling and altered metabolic functions.

Findings

30-day-old astrocytes show reduced efficiency in recycling vitamin C and altered redox metabolism.

DHA accumulation in activated astrocytes inhibits the pentose phosphate pathway and reduces glutathione levels.

DHA stimulates lactate uptake mainly in younger astrocytes (15 days in vitro).

Abstract

Oxidative damage in neurodegenerative diseases activates astrocytes and perturbs antioxidant defenses. Vitamin C is the principal antioxidant in the brain. Ascorbic acid (AA, reduced form) is taken up by neurons via the sodium/vitamin C transporter 2 (SVCT2). Astrocytes take up only the oxidized form of vitamin C, dehydroascorbic acid (DHA), through glucose transporters (GLUT). AA is recycled between neurons and astrocytes, preserving antioxidant capacity and maintaining physiological DHA levels. We postulate that AA recycling modulates astrocyte energy and redox metabolism. We therefore examined the effects of AA and DHA accumulation on glycolysis, pentose phosphate pathway (PPP) activity, and glutathione (GSH) concentrations in activated astrocytes. Culture time negatively modulated DHA recycling. At 15 days in vitro (DIV), astrocytes efficiently took up physiological DHA and reduced it to AA, enhancing redox metabolism, stimulating PPP activity, and increasing intracellular GSH. At 30 DIV (cells positive for activation markers), astrocytes took up higher amounts of DHA but reduced it inefficiently; at this time point, the glycolytic rate was unchanged, PPP activity was inhibited, and GSH decreased. In both 15- and 30-DIV astrocytes, DHA stimulated lactate uptake. We propose that 30-DIV astrocytes constitute a cellular model of reactive astrocytes with impaired AA recycling, ultimately altering glycolytic and antioxidant function.

Image 1

•Vitamin C recycling modulates astrocyte energy and redox metabolism.•30-DIV astrocytes constitute a cellular model of reactive astrocytes with reduced AA recycling.•DHA accumulation alters glycolysis, PPP activity, and GSH levels in activated (30-DIV) astrocytes.•DHA stimulates lactate uptake predominantly in 15-DIV astrocytes.

Vitamin C recycling modulates astrocyte energy and redox metabolism.

30-DIV astrocytes constitute a cellular model of reactive astrocytes with reduced AA recycling.

DHA accumulation alters glycolysis, PPP activity, and GSH levels in activated (30-DIV) astrocytes.

DHA stimulates lactate uptake predominantly in 15-DIV astrocytes.

Figures

Astrocytes in culture induce functional GLUT1 and GLUT3.A. Immunocytochemical analysis of GFAP and tubulin βIII in 15 and 30 DIV astrocytes. Only GFAP-positive reactions are detected. The nucleus was stained with TOPRO (blue). Scale bar, 50 μm. B. Immunocytochemical analysis of GFAP, vimentin, MAP2, and Nestin in 15 and 30 DIV astrocytes. Increased positive reactions for all proteins are observed in 30 DIV astrocytes. Scale bars, 50 μm. C. Immunocytochemical analysis of GLUT1 and GFAP in primary cultured 7, 15, and 30 DIV astrocytes. The nucleus was stained with TOPRO (blue). After 15 and 30 D

Astrocytes in culture increases DHA uptake.A. The uptake of 100 μM DHA was analyzed over time at 22 °C in primary 15 and 30 DIV astrocytes by HPLC. Previously, for DHA uptake studies, L-[^14^C]-AA was incubated with 1 U/μmol ascorbate oxidase for 15 min at room temperature. B. DHA uptake analysis in the presence of 20 μM cytochalasin B or E in primary 15 DIV astrocytes. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗∗p < 0.005, Bonferroni test. C. DHA uptake analysis in the presence of 20 μM cytochalasin B or E in primary 30 DIV astrocytes. Data represent the

DHA accumulation and the differential metabolic response of astrocytes maintained in vitro*.*A. Analysis of basal glycolytic rates in 15 and 30 DIV astrocytes. Data represent the mean ± SD of three experiments, each performed in triplicate. The absolute value of the glycolytic rate in the control condition was 1.5 ± 0.17 nmol min^−1^ mg^−1^ protein. B, C. Rate of glycolysis in 15 DIV (B) and 30 DIV (C) astrocytes. The cells were incubated with different concentrations of AA or DHA for 45 min, after which the rate of [5-^3^H]-glucose conversion into ^3^H_2_O was assessed. D, E. Rate of glycolys

DHA alters intracellular capacity to maintain reduced glutathione (GSH).A. Basal GSH concentration in 15- and 30-DIV astrocytes. Data represent the mean ± SD of three experiments, each performed in triplicate. B. GSH levels in 15- and 30-DIV astrocytes. Cells were incubated with different concentrations of DHA for 10 min, after which GSH levels were assessed. Data represent the mean ± SD of four experiments, each performed in triplicate. ∗∗p < 0.005, Bonferroni test. C. GSH levels in 15- and 30-DIV astrocytes. Cells were incubated with 1 mM DHA, after which GSH levels were assessed at differen

Full text

Introduction

1

The pathogenesis of several neurodegenerative diseases, including Parkinson disease (PD), Amyotrophic lateral sclerosis (ALS), and Alzheimer's disease (AD), has been linked to oxidative damage driven by reactive oxygen species (ROS). Oxidative damage also increases with astrocytic activation, further exacerbating ROS-mediated cellular injury [1,2].

Vitamin C is the principal antioxidant in the central nervous system (CNS), reaching ∼10 mM in neurons [3,4]. Lower concentrations are found in astrocytes, which mediate vitamin C recycling to maintain the reduced state. We previously showed that neuronal accumulation of dehydroascorbic acid (DHA), the oxidized form of vitamin C, decreases glutathione (GSH), inhibits glycolysis, and increases activity of the pentose phosphate pathway (PPP) [5]. Similar effects have been reported in colon cancer cells following DHA accumulation, including inhibition of glycolysis and GAPDH activity and increased ROS damage [6].

In the CNS, ascorbic acid (AA) enters neurons via the long form of the sodium/vitamin C transporter 2 (SVCT2), a high-affinity transporter [3,4]. A short form has also been described in leukemic cells and cortical neurons [[7], [8], [9]], as well as in glioblastoma cells, where it is predominantly intracellular and localized within the endoplasmic reticulum [10]. It has been proposed that neurons release DHA extracellularly, where it is taken up by astrocytes and reduced to AA. Glucose transporter 3 (GLUT3) would mediate DHA efflux from neurons, whereas GLUT1 would mediate DHA influx into astrocytes. Once DHA accumulates in astrocytes, it is converted to AA and released to the extracellular space, establishing a vitamin C recycling loop between neurons and astrocytes [11]. Glutamate released from neurons is rapidly cleared by astroglial excitatory amino-acid transporters (EAAT1/GLAST), a Na^+^-coupled process that drives astrocytic Na^+^ influx and cell swelling. The ensuing volume increase activates volume-regulated anion channels (VRAC; also termed VSOAC), formed by the LRRC8 complex, through which AA is released to the extracellular space. Pharmacological blockade of VRAC/VSOAC suppresses this AA efflux in astrocyte cultures, and elevations in extracellular AA have been observed in vivo following glutamatergic stimulation[12]. Under oxidative stress and heightened glutamate release, AA is more extensively oxidized to DHA, expanding both extra- and intracellular DHA pools, shifting neuronal metabolic states, and altering astrocyte biology. Consequently, astrocyte activation across diverse brain pathologies is likely to profoundly disrupt the neuron–glia vitamin C recycling cycle [13,14].

Previous studies have demonstrated in situ and in vitro GLUT1 expression in normal astrocytes [15,16]; however, its regulation in activated astrocytes remains unclear. Moreover, data from extended in vitro culture, during which astrocyte activation emerges, are limited [17]. To address this, we examined astrocyte metabolism and vitamin C recycling in cells maintained for 15 and 30 days in vitro (DIV), the latter being a period when activation markers are detected in astrocytes [18]. We show alterations in GLUT expression in activated astrocytes, accompanied by changes in DHA uptake, glycolytic flux, PPP activity, intracellular GSH, and lactate uptake. In summary, 15-DIV astrocytes act as vitamin C–recycling cells, taking up DHA and releasing AA, whereas 30-DIV astrocytes progressively lose these capacities. These metabolic shifts are expected to exacerbate pathological changes across neuropathologies.

Materials and methods

2

Ethics statement

2.1

All experimental procedures and facility use were approved by the Committee on the Ethics of Animal Experiments and conducted in accordance with the University of Concepcion Guidelines for Animal Experiments (protocols 1221147 and 1100396) and the Manual of Biosafety Standards and Associated Risks (CONICYT, 2009).

Animals

2.2

Sprague–Dawley rats at postnatal day 1 (P1) were used. Pregnat rats had ad libitum access to standard rodent chow (LabDiet; Animal Care, St. Louis, MO, USA). In postnatal rats, cervical dislocation was performed before brain dissection.

Primary cultures of cortical astrocytes

2.3

Primary astrocyte cultures were prepared from P1 rats [19]. Cerebral cortices were dissected and incubated for 15 min at 37 °C in 0.25% (w/v) trypsin and 0.20% (w/v) EDTA. Tissue was then mechanically dissociated in Minimum Essential Medium (MEM) supplemented with 10% (v/v) fetal bovine serum (FBS), 2 mM l-glutamine, 100 U/mL penicillin, 100 μg/mL streptomycin, 2 mM glucose, and 2.5 μg/mL Fungizone. Cells were plated at 2 × 10^4^ cells/cm^2^ and maintained for 7, 15, or 30 days in vitro (DIV).

Reagents and antibodies

2.4

The following primary antibodies were used in this study: chicken anti-human vimentin (AB5733, 1:400 dilution, Millipore, Billerica, MA, USA), rabbit anti-human GFAP (MAB360, 1:500 dilution, Millipore), mouse anti-βIII tubulin (G712A, 1:1000 dilution, Promega, Madison, WI, USA), rabbit anti-human GLUT1 (C110491, 1:100 dilution, EMD Millipore, Burlington, MA, USA), rabbit anti-rat GLUT3 (C110491, 1:50 dilution, EMD Millipore, Burlington, MA, USA), mouse anti-MAP2 (1:50 dilution, Chemicon, Temecula, CA, USA), and anti-nestin (1:50 dilution).

Immunofluorescence and confocal microscopy

2.5

Cells were labeled for immunofluorescence in confocal microscopy. The cells were fixed in 4% paraformaldehyde, washed three times with 10 mM Tris-phosphate buffer (10 mM Tris, 120 mM NaCl, 8.4 mM Na_2_HPO_4_, and 3.5 mM KH_2_PO_4,_ pH 7.8), and incubated with the primary antibody (see above) prepared in 10 mM Tris-phosphate buffer (pH 7.8) supplemented with 1% w/v bovine serum albumin (BSA) in a humid chamber for 16 h. After three 10-min washes, the sections were incubated for 2 h at room temperature with anti-rabbit immunoglobulin (IgG) coupled to Cy2/Cy3 (Jackson ImmunoResearch, Baltimore Pike, PA, USA), anti-chicken IgY (H + L) coupled to Alexa Fluor 647 (Jackson ImmunoResearch), anti-mouse IgG (H + L) coupled to Cy2/Cy3 (Jackson ImmunoResearch), or anti-goat IgG (H + L) coupled to Cy2/Cy3 (Jackson ImmunoResearch). All secondary antibodies were prepared in 10 mM Tris-phosphate buffer supplemented with BSA. Additionally, Hoechst 33342 was used for nuclear staining. As a negative control, the primary antibody was omitted. The samples were analyzed with an LSM 780 NLO spectral confocal microscope (Zeiss, Berlin, Germany). Images were acquired with Zen 2011 software (Zeiss).

Uptake assays of different metabolites

2.6

Uptake assays were performed in astrocyte cultures at 15 and 30 DIV. For AA uptake, L-[^14^C]-ascorbic acid was used (specific activity 4-8 mCi/mmol, NEN-DuPont). AA was dissolved in 0.1 mM 1,4-dithiothreitol and kept at 4 °C prior to the assay. For DHA uptake studies, L-[^14^C]-AA was incubated with 1 U/μmol ascorbate oxidase (Sigma) for 15 min at room temperature. Conversion of AA to DHA was confirmed by HPLC analysis (see below). To analyze glucose uptake, the non-metabolizable analog [^3^H]-2-DOG (specific activity 8 Ci/mmol, PerkinElmer) was used. To analyze lactate transport, [^14^C]-lactate (specific activity 150 mCi/mmol, American Radiolabeled Chemicals) was used. [3-^3^H]-glucose (specific activity 20 Ci/mol, PerkinElmer) was used in the glycolytic rate studies. To determine the activity of the PPP, transport studies were performed using [6-^14^C] glucose and [1-^14^C] glucose (specific activity 8 Ci/mol, PerkinElmer). For methodological details, see Cisternas et al. (2013) [5].

AA (0.2-1 mM), DHA (0.2-1 mM), lactate, or 2-DOG (0.01-5 mM) assays were performed in 15 mM HEPES, pH 7.4, 135 mM NaCl, 5 mM KCl, 1.8 mM CaCl_2_, and 0.8 mM MgCl_2_. The assay was stopped with 1 mL of transport buffer containing 0.2 mM HgCl_2_ at 4 °C. Cells were then lysed in 10 mM Tris-HCl, 0.2% SDS, pH 8.0, and analyzed in an LS 6500 scintillation counter (Beckman Coulter). For Km analyses of 2-DOG, cells were incubated for 15 s in increasing concentrations of 2-DOG (0-14 mM), and 0.1 μCi of 2-deoxy-D- [ 1,2-(N)^3^H]-glucose (26.2 Ci/mmol, Dupont-NEN, Boston, MA, USA) was used. In some assays, inhibitors were used: Cytochalasin B/E (0.001-0.01 mM for 10 min) or α-cynohydroxy- cinnamate (5 mM 4α-CIN for 20 min). Assays were performed in triplicate, and the results represent the average of at least three experiments. The data were analyzed using the Prism 4.06 program, and statistical analysis was performed using the GraphPad InStat 3.06 program.

Glycolytic flux analysis

2.7

Glycolytic flux in astrocytes was determined by measuring the production of ^3^H_2_O from [3-^3^H]-glucose. The cells were incubated with varying concentrations of AA and DHA and then analyzed in Krebs-Henseleit buffer (11 mM Na_2_HPO_4_, 122 mM NaCl, 3.1 mM KCl, 0.4 mM K_2_HPO_4_, 1.2 mM MgSO_4_, and 1.3 mM CaCl_2_, pH 7.4), supplemented with 5 mM d-glucose and 5 μCi [3-^3^H]-glucose (PerkinElmer). The tests were carried out in an oxygen-saturated atmosphere to allow oxidative metabolism. The method is described in detail in Ref. [5].

PPP activity analysis

2.8

Glucose oxidation via the PPP was measured essentially as described in Ref. [20]. The difference in the production of ^14^CO_2_ from [6-^14^C] glucose (decarboxylated by 6-phosphogluconate dehydrogenase and in the Krebs cycle) and from [1-^14^C] glucose (decarboxylated only in the Krebs cycle) was analyzed. To assess the effect of AA and DHA on the PPP in astrocytes, cells were treated with different concentrations of AA or DHA. The method is described in detail in Ref. [5].

HK and G6PDH activity quantification

2.9

To quantify enzymatic activity, the cells were pretreated with 1 mM AA or DHA for 45 min at 37 °C. The cells were then washed with PBS, scraped into 0.25% trypsin-0.2% EDTA, and centrifuged at 500x g for 5 min at 4 °C. The cells were resuspended in isolation medium (250 mM sucrose, 20 mM HEPES buffer, 10 mM KCl, 1.5 mM MgCl, 1 mM EDTA, 1 mM dithiothritol, 2 μg/mL aprotinin, 1 μg/mL pepstatin A and 2 μg/mL leupeptin), sonicated at 4 °C, and centrifuged at 1500 g for 5 min at 4 °C. The supernatant was then centrifuged at 13,000×g for 30 min at 4 °C. Finally, the supernatant was used to quantify enzymatic activity. To quantify HK activity, the purified fraction was mixed with 25 mM Tris-HCl, 1 mM dithiothreitol, 0.5 mM NADP/Na^+^, 2 mM MgCl_2_, 1 mM ATP, 2 U/mL G6PDH, 10 mM glucose, and incubated for different times at 37 °C. The reaction was stopped by adding 10% trichloroacetic acid, and the generation of NADPH was measured at 340 nm. To quantify G6PDH activity, glucose was replaced with 10 mM glucose-6-phosphate, and G6PDH was omitted from the reaction medium [21,22].

Glutathione determination in astrocytes

2.10

Cells were pre-incubated with AA or DHA (1 mM), then washed with PBS and detached from culture plates using 1% v/v sulfosalicylic acid. Cell lysates were centrifuged at 13,000×g for 5 min at 4 °C. The supernatant was used to determine total GSH levels (GSH + 2xGSSG). Quantification was performed in 96-well plates containing the reaction buffer (0.1 M phosphate, 0.2 M EDTA, 0.3 mM 6,6′-Dithionitro-3,3′-benzoic acid, 0.4 mM β-Nicotinamide adenine dinucleotide 2′-phosphate, and 1 U/mL GSH reductase). Absorbance was measured at 405 nm for 2.5 min (every 15 s). To quantify GSSG, equal volumes of the cell supernatant and 2-vinylpyridine were mixed. After 1 h, 1 mL of reaction buffer was added, and absorbance was measured at 405 nm for 5 min (every 30 s).

Identification of the redox form of intracellular vitamin C using HPLC

2.11

To identify the redox form of vitamin C present in astrocytes, the separation of AA and DHA was standardized by HPLC. Thus, 20 μL of radioactive AA or DHA standards was dissolved in 60% methanol, 1 mM EDTA, pH 7.8, and injected for HPLC. Fractions were collected every 3 min and analyzed by liquid scintillation to identify the fractions with AA or DHA effluxes. To determine the redox form present intracellularly, astrocytes were incubated with 100 μM radioactive AA or DHA for different times at 37 °C, after which they were lysed using a 60% v/v methanol, 1 mM EDTA solution, pH 7.8. The lysate was filtered (20 μm filters), and 20 μL was injected for HPLC using a Partisil 10 SAX anion exchange column (Whatman) and a silica precolumn. The separation was carried out over 39 min, with the following elution conditions: flow rate of 0.5 mL/min, mobile phase buffer 7 mM KH_2_PO_4_, 7 mM KCl (pH 4.0). Fractions were collected every 3 min and analyzed by liquid scintillation.

Statistical analysis

2.12

Data are presented as mean ± SD. In this study, N denotes biological replicates (N = 3). Two-group comparisons were performed using a one-tailed, unpaired Student's t-test. For comparisons involving more than two groups, one-way ANOVA followed by a Bonferroni post hoc test was used. Statistical significance was set at p < 0.05. All analyses were conducted in Prism 6.0 (GraphPad Software, San Diego, CA, USA).

Results

3

Effect of the primary astrocyte culture period on GLUT1 and GLUT3 expression

3.1

First, we determined that primary astrocyte cultures at 15 and 30 DIV exhibit high purity, showing positive reactivity for GFAP (astrocytic marker) but not for tubulin ẞIII (neuronal marker) (Fig. 1a). Furthermore, only 30 DIV astrocytes showed positive reactivity for vimentin, MAP2, and Nestin, markers of astrocytic activation (Fig. 1b).Fig. 1. Astrocytes in culture induce functional GLUT1 and GLUT3.A. Immunocytochemical analysis of GFAP and tubulin βIII in 15 and 30 DIV astrocytes. Only GFAP-positive reactions are detected. The nucleus was stained with TOPRO (blue). Scale bar, 50 μm. B. Immunocytochemical analysis of GFAP, vimentin, MAP2, and Nestin in 15 and 30 DIV astrocytes. Increased positive reactions for all proteins are observed in 30 DIV astrocytes. Scale bars, 50 μm. C. Immunocytochemical analysis of GLUT1 and GFAP in primary cultured 7, 15, and 30 DIV astrocytes. The nucleus was stained with TOPRO (blue). After 15 and 30 DIV, astrocytes show intense immunoreactions for GLUT1 (white arrows). Scale bar, 50 μm. D. Immunocytochemical analysis of GLUT3 and GFAP in 7, 15, and 30 DIV astrocytes. The nucleus was stained with TOPRO (blue). After 15 and 30 DIV, astrocytes show intense immunoreactions for GLUT3 (white arrows). Scale bar, 50 μm. E. Functional analysis of glucose transporters in primary 15 and 30 DIV astrocytes. Uptake of 250 μM 2-DOG was analyzed over time at 22 °C**.** To analyze glucose uptake, the non-metabolizable analog [^3^H]-2-DOG was used. F. Analysis of 2-DOG uptake in the presence of 20 μM cytochalasin B or E in 15 DIV astrocytes. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗∗p < 0.005, Bonferroni test. G. Analysis of 2-DOG uptake in the presence of 20 μM cytochalasin B or E in primary 30 DIV astrocytes. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗∗p < 0.005, Bonferroni test. H. Dose-dependence of 2-DOG uptake using 15-s assays in 15 DIV astrocytes. I. Lineweaver-Burk analysis of the data presented in panel J. J. Dose-dependence of 2-DOG uptake using 15-s assays in 30 DIV astrocytes. K. Lineweaver-Burk analysis of the data presented in panel L.Fig. 1

Normal astrocytes express only GLUT1, but they can induce expression of GLUT3, a high-affinity transporter, under certain conditions [[23], [24], [25], [26]]. Using immunocytochemical analysis, we found a weak GLUT1 expression in 7 DIV astrocytes, concentrated in the perinuclear zone (Fig. 1c, arrows). In 15 DIV astrocytes, a weak increase in the reaction was observed, uniformly distributed throughout the cell body (Fig. 1c, arrows). An increase in the positive reaction for GLUT1 was detected in 30 DIV astrocytes (Fig. 1c). GLUT3 expression was low at 7 DIV (Fig. 1d, arrows) and 15 DIV (Fig. 1d, arrows) but increased in intensity in 30 DIV astrocytes (Fig. 1d, arrows). The cells analyzed were also positive for GFAP; however, 7 DIV astrocytes showed low immunoreactivity (Fig. 1c and d).

2-Deoxyglucose uptake in astrocytes cultured over different periods

3.2

We next analyzed the effect of culture period on GLUT function in astrocytes. Uptake experiments were performed using the non-metabolizable glucose analog 2-deoxyglucose (2-DOG). At 250 μM 2-DOG, an initial linear uptake over 45 s was observed in both 15 and 30 DIV astrocytes (Fig. 1e).

Uptake of 6 ± 0.9 nmol × 10^6^ cells in astrocytes at 15 DIV and 7 ± 0.5 nmol × 10^6^ cells in astrocytes at 30 DIV was achieved within 90 s. To confirm that uptake occurs through GLUTs, we analyzed the effect of 20 μM cytochalasin B. Uptake of 2-DOG was inhibited by 88 ± 5% in 15 DIV astrocytes and by 74 ± 9% in 30 DIV astrocytes (Fig. 1f and g). We demonstrate that cytochalasin B inhibits both transporters (Fig. 1f and g) (uptake in control: 7 nmol × 10^6^ cells). As a control, 20 μM cytochalasin E, a structural analogue with no effect on GLUT function, was used. The presence of cytochalasin E did not alter the transport of 2-DOG in any of the culture periods analyzed (Fig. 1f and g) (uptake in control: 7.1 nmol × 10^6^ cells).

To determine the contribution of each transporter to 2-DOG uptake, we measured the kinetic parameters (Km and Vmax) for 2-DOG uptake. Saturation curves for 2-DOG were generated at 15 s using increasing concentrations of 2-DOG (0-14 mM) (Fig. 1h and j). Results from 15 to 30 DIV astrocytes showed a single kinetic component (Fig. 1h and j) and a Km of 4.5 ± 1.3 mM and a Vmax of 10 ± 6 nmol x10^6^ cells/min in 15 DIV astrocytes (Fig. 1i) and an apparent Km of 5.8 ± 1 mM and a Vmax of 12 ± 3 nmol x10^6^ cells/min in 30 DIV astrocytes (Fig. 1k). In both models, a single kinetic component was observed, indicating that the Kms of GLUT1 and GLUT3 are very similar in astrocytes.

Impact of astrocytes culture period on DHA uptake

3.3

To study the uptake of DHA, we first determined the time course of 100 μM DHA at 22 °C in 15 and 30 DIV astrocytes (Fig. 2a). 15 DIV astrocytes incorporated DHA linearly for up to 4 min, reaching 2000 ± 200 pmol x10^6^ cells at 8 min. In 30 DIV astrocytes, uptake was linear until 3 min. This result shows that 30 DIV astrocytes can incorporate three times more DHA than 15 DIV astrocytes, an unexpected finding, given that 2-DOG uptake was similar in both types of astrocytes.Fig. 2. Astrocytes in culture increases DHA uptake.A. The uptake of 100 μM DHA was analyzed over time at 22 °C in primary 15 and 30 DIV astrocytes by HPLC. Previously, for DHA uptake studies, L-[^14^C]-AA was incubated with 1 U/μmol ascorbate oxidase for 15 min at room temperature. B. DHA uptake analysis in the presence of 20 μM cytochalasin B or E in primary 15 DIV astrocytes. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗∗p < 0.005, Bonferroni test. C. DHA uptake analysis in the presence of 20 μM cytochalasin B or E in primary 30 DIV astrocytes. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗p < 0.01, Bonferroni test. D. Dose-dependent effects of cytochalasin B and cytochalasin E on the uptake of 0.1 mM DHA in 15 DIV astrocytes (5 min). E. Dose-dependent effects of cytochalasin B and cytochalasin E on the uptake of 100 mM DHA in 30 DIV astrocytes (5 min). F. Dose-dependent inhibition of DHA transport (at 0.1 mM DHA concentration) by glucose. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗∗p < 0.005, Bonferroni test.Fig. 2

Incubation with cytochalasin B inhibited DHA uptake by 80 ± 7% in 15 DIV astrocytes (Fig. 2b), whereas in 30 DIV astrocytes, the inhibition was only 55 ± 7%, an unexpected result, since cytochalasin B is a powerful inhibitor of class I GLUTs (Fig. 2f). The difference in cytochalasin B inhibition could be explained by a change in affinity for the inhibitor in 30 DIV astrocytes. To analyze this aspect, we determined the IC_50_ values for the uptake of 100 μM DHA in 15 and 30 DIV astrocytes: 0.8 ± 0.2 mM in 15 DIV astrocytes (Figs. 2d) and 1 ± 0.3 mM in 30 DIV astrocytes (Fig. 1e). No significant differences were observed.

To better describe the transport of DHA and the possible differences between the cells studied, we analyzed the impact of glucose on DHA uptake in 15 and 30 DIV astrocytes. To carry out these experiments, glucose was co-incubated with DHA, simulating an in vivo condition. In 15 DIV astrocytes, glucose did not cause a significant change in 100 mM DHA uptake. In 30 DIV astrocytes, glucose significantly inhibited DHA uptake, an effect that was dose-dependent, reaching an inhibition of 50 ± 9% in the presence of 4 mM glucose and a maximum inhibition of 85 ± 5% with 10 mM glucose (Fig. 2f).

Taken together, the astrocyte culture period does not affect 2-DOG uptake but does affect DHA incorporation. GLUT3 expression in 30 DIV astrocytes could explain the increase in DHA uptake. Thus, the culture period affects the expression of DHA and glucose transporters in astrocytes. We postulate that astrocytes at 30 DIV would have a reactive phenotype, overexpressing GLUT3 and increasing their capacity to capture DHA.

Effect of DHA accumulation on the redox and energy metabolism of astrocytes in culture

3.4

We initially investigated astrocytes' potential to reduce DHA to AA. To efficiently separate and identify both redox forms of vitamin C, we standardized a high-pressure liquid chromatography (HPLC) protocol in which AA or DHA standards were injected into the column. When analyzing AA, we found that 95 ± 3% of the injected AA remained unchanged, indicating that passage of the sample through the HPLC column does not cause significant AA oxidation (Fig. 3a). When analyzing DHA, we observed that 98 ± 2% of the injected sample remained as DHA, indicating that the HPLC method does not reduce DHA in the column (Fig. 3b).Fig. 330 DIV astrocytes show decreased DHA reduction and increased oxidative capacity.A. L-[^14^C]-AA or DHA detection after AA was injected for HPLC analysis (N = 5). For DHA analysis, L-[^14^C]-AA was incubated with 1 U/μmol ascorbate oxidase for 15 min at room temperature. B. AA or DHA detection after DHA was injected for HPLC analysis (N = 5). C. 15 DIV astrocytes were incubated with 0.1 mM ^14^C-DHA for 20 min at 37 °C. Intracellular AA was analyzed by HPLC. D. 30 DIV astrocytes were incubated with 0.1 mM ^14^C-DHA for 20 min at 37 °C. Intracellular AA was detected by HPLC. E. 15 or 30 DIV astrocytes were incubated with 0.1 mM ^14^C-AA for different times at 37 °C. Intracellular AA was detected at different times by HPLC. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗p < 0.01, Bonferroni test.Fig. 3

We next analyzed the reducing capacity of astrocytes incubated with 100 μM DHA for 20 min at 37 °C, and the percentage of AA inside the astrocytes was measured at different times. In 15 DIV astrocytes, intracellular AA levels increased in a time-dependent manner from 17 ± 5% at 15 min to 60 ± 8% at 2 h, a level that was maintained until 3 h after incubation (Fig. 3c). In 30 DIV astrocytes, 50 ± 10% of the incorporated DHA was found as AA after 30 min; however, AA levels decreased as the post-incubation time increased, reaching 26 ± 5% after 3 h (Fig. 3d).

To assess astrocytic ability to maintain AA reduction, we measured intracellular AA levels over time after incubating astrocytes with 0.1 mM AA. In 15 DIV astrocytes, 70 ± 9% of the AA remained reduced after 60 min. In 30 DIV astrocytes, only 31 ± 10% of the AA was reduced at 60 min (Fig. 3e). Thus, 30 DIV astrocytes may exhibit rapid intracellular oxidation.

Effect of DHA accumulation on glycolytic rate, hexokinase (HK) activity, and PPP activity in cultured astrocytes

3.5

For glycolytic rate studies, astrocytes were previously incubated with AA or DHA for different times and concentrations. The cells were then resuspended and incubated with H^3^-glucose for 1 h. After lysis, the cells were incubated at 37 °C to allow exchange of ^H^H_2_O, and the radioactive label generated was quantified.

In 15 DIV astrocytes, a basal glucose consumption rate of 5 ± 0.4 nmol x10^6^ cells/h was observed. In 30 DIV astrocytes, a basal rate of 26 ± 4 nmol x10^6^ cells/h was detected, a value 5-fold higher than that observed in 15 DIV astrocytes (Fig. 4a). To analyze the effect of AA or DHA dose on the glycolytic rate, the cells were incubated with different concentrations of AA or DHA for 45 min. In 15 DIV astrocytes, incubation with 1 mM AA slightly affected the basal glycolytic rate. In contrast, incubation with DHA stimulated the glycolytic rate in a dose-dependent manner, reaching a maximum of 147 ± 6% with 1 mM DHA (Fig. 4). In 30 DIV astrocytes, incubation with AA or DHA did not sustainably modify the glycolytic rate (Fig. 4c). Additionally, we analyzed the effect of 1 mM AA or DHA at different incubation times. As shown in Fig. 4d, AA did not cause significant changes in the glycolytic rate of 15 or 30 DIV astrocytes. In contrast, DHA stimulated the glycolytic rate in a time-dependent manner in 15 DIV astrocytes, reaching a maximum stimulation of 155 ± 9% after 90 min (Fig. 4e). In 30 DIV astrocytes, a slight decrease was observed in the presence of DHA (Fig. 4e). In summary, DHA stimulated the glycolytic rate only in 15 DIV astrocytes.Fig. 4DHA accumulation and the differential metabolic response of astrocytes maintained in vitro*.*A. Analysis of basal glycolytic rates in 15 and 30 DIV astrocytes. Data represent the mean ± SD of three experiments, each performed in triplicate. The absolute value of the glycolytic rate in the control condition was 1.5 ± 0.17 nmol min^−1^ mg^−1^ protein. B, C. Rate of glycolysis in 15 DIV (B) and 30 DIV (C) astrocytes. The cells were incubated with different concentrations of AA or DHA for 45 min, after which the rate of [5-^3^H]-glucose conversion into ^3^H_2_O was assessed. D, E. Rate of glycolysis in 15 DIV (D) and 30 DIV (E) astrocytes. The cells were incubated with 1 mM AA, and the rate of [5-^3^H]-glucose conversion into ^3^H_2_O was assessed at different times. F. HK activity in 15 DIV astrocytes. The cells were pretreated with 1 mM AA or DHA for 45 min at 37 °C. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗∗∗p < 0.001, Bonferroni test. G. HK activity in 30 DIV astrocytes. The cells were pretreated with 1 mM AA or DHA for 45 min at 37 °C. H. HK activity in 15 and 30 DIV astrocytes. The cells were pretreated with 1 mM DHA at 37 °C. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗p < 0.01, ∗∗p < 0.005, Bonferroni test. I. 15 or 30 DIV astrocytes were analyzed to evaluate basal PPP rate, assessed as the difference in ^14^CO_2_ release from [1-^14^C]- and [6-^14^C]-glucose. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗∗∗p < 0.001, Bonferroni test. The absolute value of PPP activity in the control condition was 0.7 ± 0.14 nmol min^−1^ mg protein^−1^. J. Rate of ^14^CO2 production from PPP in 15 DIV astrocytes treated with 1 mM AA or DHA for 45 min. K. Rate of ^14^CO2 production from PPP in 30 DIV astrocytes treated with 1 mM AA or DHA for 45 min. L. G6PDH activity in 15 DIV astrocytes. The cells were pretreated with 1 mM AA or DHA for 45 min at 37 °C. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗∗∗p < 0.001, Bonferroni test. M. G6PDH activity in 30 DIV astrocytes. The cells were pretreated with 1 mM AA or DHA for 45 min at 37 °C. Data represent the mean ± SD of three experiments, each performed in triplicate.Fig. 4

The effect of DHA previously described in 15 DIV astrocytes could be explained by a direct effect on the activity of enzymes that participate in glycolysis. Thus, we determined HK activity in cultured astrocytes. At 15 DIV, AA did not change HK activity. However, 1 mM DHA (30 min) stimulated HK activity by 8-fold (Fig. 4f), an effect that was directly proportional to incubation time (Fig. 4h). In 30 DIV astrocytes, AA or DHA did not significantly change HK activity (Fig. 4g).

We also analyzed the effect of vitamin C accumulation on PPP activity. Cells were treated with 1 mM AA or DHA for 45 min and then incubated with 1-^14^C-glucose or 6-^14^C-glucose for 1 h. The amount of ^14^CO_2_ released via the PPP and glycolysis was then measured. At baseline, 15 DIV astrocytes showed activity of 480 ± 100 nmol x10^6^ cells/h, indicating high PPP activity in these cells. In 30 DIV astrocytes, we determined a basal activity of 10 ± 3 nmol x10^6^ cells/h, which is 300-fold lower than that observed in 15 DIV astrocytes (Fig. 4i). In 15 DIV astrocytes, incubation with 1 mM AA inhibited PPP activity by 48 ± 11% (Fig. 4j), whereas incubation with 1 mM DHA significantly stimulated PPP activity (Fig. 4j). In 30 DIV astrocytes, DHA inhibited basal PPP activity by 35 ± 10%, whereas AA did not produce a statistically significant inhibition (Fig. 4k).

To determine how DHA affects the PPP pathway, we analyzed the effect of vitamin C on the activity of glucose-6-phosphate dehydrogenase (G6PDH), the first enzyme in the pathway and its most important control point. In 15 DIV astrocytes, DHA stimulated G6PDH activity by 1110 ± 223%, a highly significant change consistent with the increase in PPP activity (Fig. 6l). In 30 DIV astrocytes, the presence of DHA or AA did not affect G6PDH activity (Fig. 4m). In summary, the data indicate that DHA significantly increased HK and PPP activity only in 15 DIV astrocytes, cells that represent a “normal” astrocyte and are eventually involved in active vitamin C recycling.

GSH levels in astrocytes maintained in culture

3.6

In addition to serving as an alternative pathway for glucose consumption, the PPP is essential for maintaining cellular redox balance by generating the cofactor NADPH, which is used by antioxidant enzymes and for GSH recycling. For this reason, we set out to analyze whether the effect of DHA on the PPP directly alters intracellular GSH levels in 15 and 30 DIV astrocytes. Astrocytes were incubated with DHA or AA for 10 min, and basal GSH levels were determined. 15 DIV astrocytes contained 25 ± 4 nmol x10^6^ cells of GSH, while 30 DIV astrocytes contained 19 ± 5 nmol x10^6^ cells, indicating similar GSH levels in both cell types (Fig. 5a). The effect of DHA on GSH levels was then analyzed. In 15 DIV astrocytes, DHA produced a dose-dependent increase, reaching 47 ± 10% with 1 mM DHA (Fig. 5b). In 30 DIV astrocytes, low concentrations of DHA (0.1-0.5 mM) increased GSH levels by 24 ± 9%; however, at higher concentrations, GSH levels decreased significantly to 9 ± 4% (Fig. 5b).Fig. 5DHA alters intracellular capacity to maintain reduced glutathione (GSH).A. Basal GSH concentration in 15- and 30-DIV astrocytes. Data represent the mean ± SD of three experiments, each performed in triplicate. B. GSH levels in 15- and 30-DIV astrocytes. Cells were incubated with different concentrations of DHA for 10 min, after which GSH levels were assessed. Data represent the mean ± SD of four experiments, each performed in triplicate. ∗∗p < 0.005, Bonferroni test. C. GSH levels in 15- and 30-DIV astrocytes. Cells were incubated with 1 mM DHA, after which GSH levels were assessed at different times. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗p < 0.01, ∗∗p < 0.005, Bonferroni test.Fig. 5. Fig. 6DHA stimulates lactate uptake in astrocytes maintained in culture for a short time.A. Functional analysis of lactate transporters in primary astrocytes at 15 and 30 DIV. Uptake of 100 mM lactate was measured over time at 22 °C**.** Data represent the mean ± SD of three experiments, each performed in triplicate. B. Analysis of lactate uptake in the presence of 20 μM cytochalasin B or 4-αCIN in primary astrocytes at 15 DIV. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗p < 0.01, Bonferroni test. C. Analysis of lactate uptake in the presence of 20 μM cytochalasin B or 4-αCIN in primary astrocytes at 30 DIV. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗p < 0.01, Bonferroni test. D. To analyze the effect of vitamin C on the accumulation of lactate (100 mM), 15 DIV astrocytes were incubated with DHA at different concentrations (0–1 mM) for 45 min. Data represent the mean ± SD of three experiments, each performed in triplicate. ∗p < 0.01, Bonferroni test. E. To analyze the effect of DHA on the accumulation of lactate (100 mM), 30 DIV astrocytes were incubated with DHA at different concentrations (0–1 mM). F. To analyze the effect of vitamin C on the accumulation of lactate (100 mM), 15 DIV astrocytes were incubated with 1 mM DHA at different times (0–60 min). Data represent the mean ± SD of three experiments, each performed in triplicate. ∗∗p < 0.005, Bonferroni test. G. To analyze the effect of vitamin C on the accumulation of lactate (100 mM), 15 DIV astrocytes were incubated with 1 mM DHA at different times (0–60 min). H. Left (green): schematic of a non-activated astrocyte with normal vitamin C recycling; PPP activity and GSH levels are increased, and intracellular lactate also rises. Right (red): schematic of an activated astrocyte with impaired vitamin C recycling, with decreased PPP activity and compromised GSH generation. AA, ascorbic acid; DHA, dehydroascorbic acid; GSH, reduced glutathione; GSSG, oxidized glutathione; GLUT, glucose transporter; NADPH, nicotinamide adenine dinucleotide phosphate (reduced); NADP+, nicotinamide adenine dinucleotide phosphate (oxidized); PPP, pentose phosphate pathway.Fig. 6

We also analyzed the effect of time on GSH levels during DHA treatment. Treating 15 DIV astrocytes with 1 mM DHA resulted in a time-dependent increase, reaching 54 ± 9% by 90 min. In 30 DIV astrocytes, DHA markedly decreased GSH levels, reaching 90 ± 4% after 90 min (Fig. 5c). In summary, DHA uptake by 15 DIV astrocytes increases GSH concentrations, thereby enhancing the reducing capacity of these cells.

Effect of intracellular vitamin C accumulation on lactate utilization in astrocytes maintained in culture

3.7

Our results suggest that intracellular DHA accumulation in astrocytes affects their rate of glucose consumption, which would decrease ATP levels, either by inhibiting glycolysis or by stimulating the PPP. Thus, we analyzed whether DHA accumulation in astrocytes increases the incorporation of other metabolic substrates, such as lactate. We first analyzed lactate transport in 15 and 30 DIV astrocytes using uptake studies over time with 100 mM C^14^-lactate. In 15 DIV astrocytes, transport reached 450 ± 50 pmol x10^6^ cells at 45 s. In 30 DIV astrocytes, transport behaved similarly, reaching 375 ± 50 pmol x10^6^ cells after 45 s (Fig. 6a).

Inhibition analyses using 4α-CIN (an inhibitor of monocarboxylate transporters) revealed 55 ± 10% inhibition in 15 DIV astrocytes and 70 ± 15% inhibition in 30 DIV astrocytes (Fig. 6b and c), demonstrating functional expression of these transporters. Subsequently, astrocytes were incubated with varying concentrations of DHA for 45 min, and transport of 100 mM C^14^-lactate was measured for 15 s (the initial velocity condition). In 15 DIV astrocytes, increasing concentrations of DHA increased lactate uptake, reaching a maximum of 66 ± 9% at 1 mM DHA (Fig. 6d). In 30 DIV astrocytes, the presence of DHA did not affect lactate transport (Fig. 6e).

To analyze the effect of time, cells were incubated with 1 mM DHA for varying periods, and then C^14^-lactate transport was measured for 15 s. In 15 DIV astrocytes, a progressive increase in lactate incorporation was observed, reaching a maximum of 79 ± 12% at 60 min (Fig. 6f). In 30 DIV astrocytes, the presence of DHA did not significantly change lactate incorporation (Fig. 6g).

Discussion

4

Astrocyte activation is a pathophysiological response to brain stress. Accordingly, astrocytes alter the expression of multiple proteins to limit the spread of injury throughout the brain [17,[26], [27], [28], [29]]. We hypothesize that 15-DIV astrocytes represent a mildly activated state, whereas 30-DIV astrocytes correspond to reactive astrocytes. Immunocytochemistry showed that 30-DIV astrocytes upregulate GFAP, MAP2, vimentin, and nestin markers associated with reactive astrocytes [26,27,30]. We also observed changes in GLUT1/3 expression. As noted above, GLUT3 induction is linked to astrocytes under stress [23].

Vitamin C is concentrated in the brain because of its continuous recycling between astrocytes and neurons [3,4,11,31,32]. Although classically regarded as an antioxidant, DHA can irreversibly modulate HK in erythrocytes in vitro in a redox-independent manner [[33], [34], [35]]. We previously showed that DHA affects oxidative metabolism and PPP function in neurons and modifies GSH biosynthesis and lactate transport [5]. Similar findings have been reported in colon cancer with BRAF and KRAS mutations, where GAPDH inhibition and ROS generation are implicated [6].

DHA is taken up by astrocytes via GLUT1 [[36], [37], [38], [39]]. Our immunocytochemistry data confirm GLUT1 expression at 15-DIV, which increases over time in culture (30-DIV) (Fig. 6H). We also observed early GLUT3 expression at 15-DIV, which rises in 30-DIV astrocytes. Previous studies in culture and in situ showed that astrocytes induce GLUT3 under stress (e.g., LPS or ischemia–reperfusion) [23,40]. Surprisingly, the increase in GLUT3 does not raise 2-DOG uptake in 30-DIV astrocytes; however, DHA uptake increases and is efficiently competed by glucose. This can be explained by the fact that, in 30-DIV astrocytes, the Km values for DHA and 2-DOG transport are essentially identical for GLUT3, increasing substrate competition at this transporter. In contrast, GLUT1 exhibits distinct Km values for DHA and glucose [41]. Our data indicate that DHA enters 15-DIV astrocytes through a GLUT1-dependent, glucose-insensitive mechanism, whereas 30-DIV astrocytes increase DHA uptake via a glucose-sensitive, GLUT3-dependent mechanism (Fig. 6H).

Fifteen-DIV astrocytes show a high capacity to reduce DHA (Fig. 6H, green cell). This is favored by low oxidative metabolism, stemming from a low glycolytic rate and high PPP activity, which permits efficient GSH recycling. A stimulatory effect of DHA on PPP activity has been reported in human myeloid cells [42]. GSH reduces intracellular DHA; therefore, DHA enhances the reducing capacity of these cells by increasing PPP and G6PDH activities, thereby promoting GSH recycling. DHA accumulation in 15-DIV astrocytes also increases ATP production by stimulating glycolysis and lactate uptake. Fifteen-DIV astrocytes exhibited a lower glycolytic rate than 30-DIV astrocytes, and DHA increased this rate in 15-DIV cells. Although irreversible HK inhibition by DHA has been reported [[33], [34], [35]], we did not observe inhibition at physiological DHA concentrations up to 1 mM. The reported effects required pharmacological DHA doses (2.5–10 mM). Similarly, DHA inhibition of G6PDH has been described using purified erythrocyte enzymes at ≈10 mM DHA. Mechanistically, HK inhibition has been attributed to covalent modification of a cysteine thiol by DHA, leading to loss of HK activity. Thus, DHA would inhibit HK and G6PDH only at concentrations unlikely to occur in astrocytes under normal or most pathological conditions. At lower DHA concentrations, glycolytic flux is stimulated and the PPP enhanced, thereby promoting GSH biosynthesis (as observed 30 min after treatment with 1 mM DHA). We posit a hormetic (biphasic) dose–response [29,43,44], whereby across a DHA gradient and varying GLUT expression. We postulate that the changes observed in astrocytes kept in culture for more than 25 days are not simply an artifact of their time outside native tissue, but rather deeper physiological changes present in reactive astrocytes.

Thirty-DIV astrocytes are reactive astrocytes typical of brain injury. They exhibit high glycolytic rates and low PPP activity, increasing susceptibility to oxidative stress (Fig. 6H, red cell). DHA accumulation further heightens this susceptibility by inhibiting PPP activity and efficient GSH reduction, directly lowering the capacity to recycle DHA. In these cells, the glycolytic rate was high and unchanged during exposure to 1 mM DHA for up to 30 min, with slight inhibition between 45 and 90 min. These astrocytes show enhanced DHA uptake. When intracellular DHA exceeds ∼1.5 mM, neither HK nor G6PDH is activated; at higher levels, both enzymes are inhibited, consistent with prior reports [33,34]. Concomitantly, 30-DIV astrocytes display low PPP flux; accordingly, cellular GSH drops sharply with 1 mM DHA. Across a DHA dose–response, intracellular GSH falls rapidly above ∼0.5 mM. These data suggest that 30-DIV astrocytes have limited capacity to reduce DHA to AA, leading to DHA accumulation, suppression of glycolysis and PPP activity, and consequent GSH depletion. This metabolic state appears partially compensated by increased lactate uptake, which could support the energetic demands of activated astrocytes. In conclusion, activated astrocytes efficiently internalize DHA, which accumulates intracellularly; notably, intracellular DHA suppresses glycolysis and elicits an oxidative response in which GSH is rapidly oxidized. Unlike neurons [5], the reduced GSH status is not restored within 30 min of DHA exposure in 30-DIV astrocytes.

Together, these results indicate that 15-DIV astrocytes are predominantly DHA-reducing, AA-accumulating cells. In 15-DIV astrocytes, DHA (≥45 min) increases GSH levels, whereas in 30-DIV astrocytes, DHA (≥30 min) causes a time-dependent decline in GSH. Thus, DHA-driven modulation of PPP activity directly determines intracellular GSH levels and the ability to cope with oxidative stress, either by recycling DHA or by limiting reducing capacity.

High basal PPP activity in 15-DIV astrocytes suggests that incoming glucose is preferentially shunted through this pathway, increasing by ∼200% in the presence of DHA. A pertinent question is how astrocytes meet their energetic needs if glycolysis is reduced. Our laboratory previously showed that AA accumulation in neurons can stimulate lactate uptake [[45], [46], [47]]. By analyzing the effect of DHA on lactate uptake in 15- and 30-DIV astrocytes, we found that DHA stimulates lactate uptake in both, with a greater effect in 15-DIV cells (∼80% increase). We propose that glucose is used mainly via the PPP to maintain anabolic potential, while lactate is incorporated as an energy substrate [[48], [49], [50]].

Overall, activated astrocytes exhibit dynamic changes in GLUT expression. Fifteen-DIV astrocytes show a high DHA-reducing capacity, which diminishes in 30-DIV astrocytes that concomitantly consume GSH (Fig. 6H). In 15-DIV astrocytes, DHA accumulation stimulates HK and PPP activity, leading to a robust rise in intracellular GSH (Fig. 6H). In summary, 15-DIV astrocytes function as vitamin C–recycling cells, taking up DHA and releasing AA, whereas this capability declines at 30 DIV.

CRediT authorship contribution statement

Pedro Cisternas: Conceptualization, Investigation, Software, Supervision, Writing – original draft, Writing – review & editing. Katterine Salazar: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Supervision, Writing – original draft. Eder Ramírez: Conceptualization, Formal analysis, Funding acquisition, Investigation, Software, Writing – original draft. Sebastián Elgueta: Conceptualization, Formal analysis, Investigation, Methodology, Software, Visualization. Isabelle de Lima: Conceptualization, Formal analysis, Investigation, Methodology, Software. Valentina Muñoz: Conceptualization, Formal analysis, Investigation, Methodology, Software, Writing – original draft, Writing – review & editing. Francisco Nualart: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Validation, Writing – original draft, Writing – review & editing.

Declaration of competing interest

None.