Hormone Actions on the Brain

Neuroendocrinology 2000;72:179–186

Effect of Oxidative Stress on the Uptake ofGABA and Glutamate in Synaptosomes Isolatedfrom Diabetic Rat Brain

Ana Duartea,b Maria Santosa,b Raquel Seiçaa,c

Catarina Resende de Oliveiraa,c

aCenter for Neurosciences of Coimbra, and bDepartment of Zoology and cFaculty of Medicine, University of Coimbra, Portugal

Received: February 15, 2000Accepted after revision: June 29, 2000

Catarina Resende de OliveiraCenter for Neurosciences of CoimbraUniversity of CoimbraP–3004-517 Coimbra (Portugal)Tel. +351 239 820190, Fax +351 239 826798, E-Mail croliveira@cmc.cj.uc.pt

ABCFax + 41 61 306 12 34E-Mail karger@karger.chwww.karger.com

© 2000 S. Karger AG, Basel0028–3835/00/0723–0179$17.50/0

Accessible online at:www.karger.com/journals/nen

Key WordsStress W Oxidative stress W GABA W Diabetes W Excitatoryamino acids W Synaptosomes

AbstractIt has been suggested that increased oxidative stressmight be involved in the pathophysiology of diabeticcomplications. In this study, we investigated the effect ofdiabetes on the susceptibility of synaptosomes to oxida-tive stress (induced by the oxidizing pair ascorbate/Fe2+)and on the uptake of the amino acid neurotransmittersÁ-aminobutyric acid (GABA) and glutamate. We found alower susceptibility of synaptosomes isolated fromGoto-Kakizaki (GK) rats, a model of non-insulin-depen-dent diabetes mellitus, to lipid peroxidation as comparedwith synaptosomes isolated from Wistar control rats(6.40 B 1.05 and 12.14 B 1.46 nmol thiobarbituric acidreactive substance/mg protein, respectively). The lowersusceptibility of GK rat synaptosomes to membrane lipidperoxidation correlates with an increase in synaptosom-al vitamin E levels (835 B 58.04 and 624.26 B 50.26pmol/mg protein in diabetic and normal rats, respective-ly). In the absence of ascorbate/Fe2+, no significant differ-ences were observed between the levels of lipid peroxi-dation of synaptosomes isolated from diabetic and nor-mal rats. Studies of neurotransmitter uptake show thatthe [3H]glutamate uptake was decreased by about 30% in

diabetic GK rats as compared with control Wistar rats,whereas the [3H]GABA uptake was not significantly dif-ferent from controls. Under oxidizing conditions, the glu-tamate uptake in diabetic rats was unaffected, and adecreased GABA uptake (41.39 B 4.41 and 60.96 B 6.4%of control in GK and Wistar rats, respectively) was ob-served. We conclude that the increased resistance to oxi-dative stress in GK rat synaptosomes may be due to theincreased vitamin E content and that diabetic state andoxidative stress conditions differentially affected the up-take of the neurotransmitters GABA and glutamate.

Copyright © 2000 S. Karger AG, Basel

Introduction

Increased oxidative damage and changes in antioxi-dant capacity have been described in experimental dia-betes and have been associated with the development ofdiabetic complications [1–4], such as cataracts and cen-tral nervous system dysfunction [5].

The evidence of increased oxidative stress under dia-betic conditions, both in patients and experimentally dia-betic rats, includes increased generation of reactive oxy-gen species (ROS), decreased antioxidant content, and/orimpairment of the generation of the reduced forms ofthese antioxidants, resulting in high lipid peroxidationlevels [6–8]. Hyperglycemia can also induce modifica-

Dow

nloa

ded

by:

UC

LA B

iom

edic

al L

ibra

ry

128.

97.9

0.22

1 -

5/8/

2014

10:

25:0

5 A

M

180 Neuroendocrinology 2000;72:179–186 Duarte/Santos/Seiça/Resende de Oliveira

tions in the balance between prooxidants/antioxidants,increasing the production and decreasing the scavengingof ROS [9]. The brain is particularly affected by ROS,since its membranes are rich in polyunsaturated, highlyoxidizable fatty acids, and it has a high oxygen consump-tion, a high transition metal (such as Fe2+) content, anddecreased antioxidant defenses [10–12].

It was shown that diabetes causes alterations of specifcamino acid transport systems and/or alterations of somecell populations. Vilchis and Salceda [13] reported anincrease in taurine and GABA uptake, while glycine andglutamate uptakes were unaffected in diabetic rat retinaand retinal pigment epithelium. The amino acid trans-porter systems are also affected by oxidative stress, al-though the molecular mechanisms involved are not total-ly understood [14, 15]. Some authors suggest that directcarrier oxidation [16] or indirect impairment of mem-brane ion-motive ATPases caused by oxidative stress [17]may be involved.

We previously demonstrated that synaptosomes iso-lated from Goto-Kakizaki (GK) diabetic rats are less sus-ceptible to Aß-induced toxicity, suggesting an adaptiveresponse of the brain during the onset of diabetes [18].The GK rat is a nonobese, spontaneously diabetic animalmodel of non-insulin-dependent diabetes mellitus pro-duced by repeated selective breeding of Wistar nondiabet-ic rats, using glucose intolerance as a selective index [19].This animal model of non-insulin-dependent diabetes hasbeen used to predict the development of diabetes, beingan appropriate model to study the events at the onset ofthe disease as compared with obese diabetic rats whichpresent severe hyperglycemia and hyperlipidemia [20].

The aim of the present work is to evaluate the interac-tions between abnormal glucose metabolism and the up-take of the amino acid trasmitters Á-aminobutyric acid(GABA) and glutamate in synaptosomes isolated from thecerebral cortex of normal Wistar and GK diabetic ratsunder oxidative stress conditions. The ascorbate/Fe2+ in-cubation procedure was used to induce oxidative injury insynaptosomes, and the extent of lipid peroxidation wasfollowed by measuring the thiobarbituric acid reactivesubstance (TBARS) production.

Materials and Methods

AnimalsMale spontaneously diabetic GK rats were obtained from our

breeding colony established in 1995 with breeding couples from thecolony at the Tohoku University School of Medicine (Sendai, Japan;courtesy of Dr. K.I. Susuki). Control animals were nondiabetic male

Wistar rats of similar age, obtained from our local colony (AnimalResearch Center Laboratory, University Hospitals, Coimbra). Theanimals were kept under controlled light and humidity conditionsand had free access to powdered rodent chow (diet CRF 20; CharlesRiver France) and water.

MaterialsAll chemicals used were of the highest grade of purity commer-

cially available. [3H]GABA (65 Ci/mmol) and L-[G-3H]glutamate(49 Ci/mmol) were from Amersham International (Little Chalfont,UK).

Preparation of SynaptosomesCrude synaptosomes were prepared from brains of male Wistar

rats (30.75 B 1.44 weeks of age) and GK rats (32.61 B 1.28 weeks ofage) according to a preestablished method [21], with some modifica-tions. After decapitation, the whole cerebral cortices were rapidlyremoved and homogenized in 10 vol of 0.32 M sucrose buffered withTris at pH 7.4. The homogenate was centrifuged at 1,000 g for10 min, and the synaptosomal fraction was isolated from the super-natant by centrifugation at 12,000 g for 20 min. The white and fluffysynaptosome layer was then resuspended, respun, and resuspendedin the sucrose medium at a protein concentration of 15–20 mg/ml, asdetermined by the biuret method. Experiments were carried outwithin 3 h of preparation.

Blood Glucose and Glycated Hemoglobin (HbA1c)DeterminationsBlood glucose levels were determined immediately after sacrifice

through the glucose oxidation reaction, using a glucometer (Gluco-meter Elite, Bayer, Portugal) and compatible reactive tests. HbA1c lev-els were determined through ionic exchange chromatographic assay(Abbott IMx Glicohemoglobin, Abbott Laboratories, Portugal).

Induction of Oxidative StressThe oxidizing agents, ascorbic acid and iron (Fe2+, ferrous sul-

fate), were used to induce oxidative stress [22]. Synaptosomes (1 mg/ml) were peroxidized by incubation at 30°C in a standard mediumcontaining (in mM ): 132 NaCl, 3 KCl, 1 MgCl2, 1 NaH2PO4, 1.2CaCl2, 10 glucose, and 20 Hepes, adjusted to pH 7.4 with Tris, sup-plemented with 0.8 mM ascorbic acid and 2.5 ÌM iron. Ascorbic acidand iron solutions were prepared immediately before use and pro-tected from light.

Evaluation of Lipid PeroxidationThe extent of lipid peroxidation was determined by measuring

TBARS, using the thiobarbituric acid assay [23]. The amount ofTBARS formed was calculated using a molar extinction coefficient of1.56 ! 105 mol–1 cm–1 and expressed as nanomoles TBARS per mil-ligram protein.

Determination of Vitamin E Content in SynaptosomesExtraction and separation of synaptosomal membrane vitamin E

(·-tocopherol) were performed by following a previously describedmethod [24]. 1.5 ml sodium dodecyl sulfate (10 mM ) was added to1.5 mg synaptosomal protein, followed by the addition of 2 ml etha-nol. Then 2 ml hexane and 50 Ìl of KCl 3 M were added, and themixture was vortexed for about 3 min. The extract was centrifuged at2,000 rpm, and 1 ml of the upper phase, containing n-hexane (n-hex-ane layer), was recovered and evaporated to dryness under a stream of

Dow

nloa

ded

by:

UC

LA B

iom

edic

al L

ibra

ry

128.

97.9

0.22

1 -

5/8/

2014

10:

25:0

5 A

M

Diabetes and Oxidative Stress Neuroendocrinology 2000;72:179–186 181

Table 1. Characterization of Wistar and GK rats at 6 months of age

Wistar rats GK rats

Body weight, g 643.2B80.01 (n = 5) 371.0B23.38** (n = 5)Glycemia, mg/dl 95.3B2.88 (n = 10) 197.0B15.16*** (n = 10)HbA1c, % 5.05B0.16 (n = 4) 8.27B0.24** (n = 4)

Data are mean B SEM of the number of animals indicated. Bloodglucose was determined by glucose oxidase reaction (GlucometerElite). HbA1c was determined using an ionic change chromatographicassay (Abbott IMx Glicohemoglobin kit). Statistically different fromnormal Wistar rats: ** p ! 0.01; *** p ! 0.001.

N2 and kept at –80 °C. The extract was redissolved in n-hexane, andthe vitamin E content was analyzed by reverse-phase high-perfor-mance liquid chromatography. A Spherisorb S10w column (4.6 !200 nm) was eluted with n-hexane modified with 0.9% methanol, at aflow rate of 1.5 ml/min. Detection was performed by an ultravioletdetector at 287 nm. The levels of vitamin E in synaptosomal mem-branes were expressed as picomoles per milligram protein.

Determination of [3H]GABA and [3H]Glutamate UptakeThe uptake of [3H]GABA and [3H]glutamate was determined

according to a previously described method [25]. Synaptosomes at aconcentration of 0.5 mg protein/ml were equilibrated at 30°C in aNa+ medium containing (in mM ): 125 NaCl, 3 KCl, 1.2 MgSO4, 1.2CaCl2, 3 glucose, and 10 Hepes-Tris at pH 7.4 in the absence or pres-ence of ascorbic acid (0.8 mM ) plus iron (2.5 ÌM). In order to deter-mine whether the amino acid neurotransmitter uptake was due to theactivity of carriers, similar experiments were done in a Na+-freemedium, containing Cl-choline (125 mM ) instead of NaCl (125mM ) in the absence of ascorbate/iron. Experiments were done inwhich synaptosomes were incubated in both media at 0°C. Amino-oxyacetic acid (10 ÌM ) was included in all media when studying theuptake of [3H]GABA to pevent GABA metabolism. After 15 min ofincubation, the uptake reaction was started by adding [3H]GABA (fi-nal concentration 0.5 ÌM, 0.25 ÌCi/mol) or [3H]glutamate (final con-centration 10 ÌM, 49 Ci/mmol). The uptake was stopped at 10 minby rapid filtration of the samples (0.2 ml) under vacuum throughglass fiber filtres Whatman GF/B, followed by a wash with 10 ml ofthe Na+-containing glucose-free medium (washing medium). The fil-ters were placed in scintillation vials with 3 ml of Universol scintilla-tion cocktail (ICN, Irvine, Calif., USA), and the radioactivity wascounted in a liquid scintillation analyzer Packard Tri-Carb 2500 TR.The quenching of radioactivity in the samples was corrected automa-tically by using an efficiency correlation curve obtained for 3H-quenched standards by the external standardization method. Theresults were presented as percentage of control, considering the meanof controls as 100%.

Data AnalysisThe results are presented as mean B SEM of the indicated num-

ber of experiments. Statistical significance was determined using theStudent t test or the one-way Anova test for multiple comparisons.p ! 0.05 was considered significant.

Fig. 1. Induction of lipid peroxidation by ascorbate/iron in synapto-somes isolated from Wistar and GK rat brains. Oxidative stress wasinduced with ascorbate (0.8 mM ) and iron (2.5 ÌM ) for 15 min at30°C, whereas control synaptosomes were incubated in the absenceof the oxidizing agents. The extent of lipid peroxidation was deter-mined by the thiobarbituric acid method and expressed as nano-moles per milligram protein. Data are mean B SEM of six indepen-dent experiments. Statistical significance: ** p ! 0.01 or *** p !0.001 as compared with respective controls not treated with ascor-bate/iron; ### p ! 0.01 as compared with Wistar synaptosomestreated with ascorbate/iron.

Results

General Characteristics of GK RatsBody weight, blood glucose levels, and HbA1c of dia-

betic rats and of age-matched controls are presented intable 1. At the time of sacrifice, the mean body weight ofthe diabetic GK rats was 371.0 B 23.38 g, and that ofcontrol rats was 643.2 B 80.01 g. Blood glucose levels andHbA1c were significantly higher in the diabetic GK group(197.0 B 15.16 mg/dl, 8.27 B 0.24%) as compared withcontrol Wistar rats (95.3 B 2.88 mg/dl, 5.05 B 0.16%,respectively).

Synaptosomal Susceptibility to Oxidative StressInduced by Ascorbate IronIn the present study, oxidative stress was induced with

0.8 mM ascorbate and 2.5 ÌM Fe2+. Under these condi-tions, an increase in lipid peroxidation of synaptosomes,as determined by the formation of TBARS, was observed,[26, 27]. Figure 1 shows that, in the absence of ascorbate/iron, the levels of TBARS were similar in synaptosomalpreparations isolated from Wistar or GK rats (1.03 B0.27 and 0.82 B 0.25 nmol TBARS/mg protein, respec-

Dow

nloa

ded

by:

UC

LA B

iom

edic

al L

ibra

ry

128.

97.9

0.22

1 -

5/8/

2014

10:

25:0

5 A

M

182 Neuroendocrinology 2000;72:179–186 Duarte/Santos/Seiça/Resende de Oliveira

Fig. 2. Effect of oxidation induced by ascorbate/iron on the uptake of[3H]GABA in synaptosomes isolated from Wistar or GK rat brains.Synaptosomes were equilibrated in Na+ medium for 15 min at 30°Cin the absence (control) or in the presence of ascrobate/iron, andthe [3H]GABA uptake reaction was started by the addition of[3H]GABA. [3H]GABA uptake was also measured in the absenceof Na+ (choline medium). Data, expressed as a percentage of[3H]GABA uptake in control synaptosomes isolated from Wistar rats(considering 180.5 B 16.02 pmol/mg protein as 100%), are the meanvalues B SEM of seven to twelve different experiments. Statisticalsignificance: ### p ! 0.001 as compared with respective controls incu-bated in Na+-medium; ** p ! 0.01 or *** p ! 0.001 as compared withrespective controls not treated with ascorbate/iron; # p ! 0.05 as com-pared with respective controls treated with ascorbate/iron.

Table 2. Vitamin E (·-tocopherol) content (pmol/mg protein) in syn-aptosomes of Wistar and GK rats

Wistar rats GK rats

624.26B50.26 (n = 9) 835B58.04* (n = 9)

Vitamin E was evaluated by HPLC analysis. Data are mean BSEM from nine different experiments. Significantly different fromcontrol Wistar rats: * p ! 0.05.

tively). However, the induction of oxidation with ascor-bate/Fe2+ for 15 min resulted in a significant increase inthe production of TBARS to 12.14 B 1.46 nmol/mg pro-tein in control rats and to 6.40 B 1.05 nmol/mg protein inGK rats. It was observed that preparations from synapto-

Fig. 3. Effect of oxidation induced by ascorbate/iron on the uptake of[3H]glutamate in synaptosomes isolated from Wistar and GK ratbrains. Synaptosomes were equilibrated in Na+ medium for 15 minat 30°C in the absence (control) or in the presence of ascorbate/iron,and the [3H]glutamate uptake reaction was started by the addition of[3H]glutamate. The control was incubated in the absence of ascor-bate/iron. [3H]glutamate uptake was also measured in the absence ofNa+ (choline medium). Data, expressed as a percentage of [3H]gluta-mate uptake in control synaptosomes isolated from Wistar rats (con-sidering 5.07 B 0.79 nmol/mg protein as 100%), are the mean valuesB SEM of seven to twelve different experiments. Statistical signifi-cance: ### p ! 0.001 as compared with respective controls incubatedin Na+ medium; *** p ! 0.001 as compared with respective controlsnot treated with ascrobate/iron; ## p ! 0.01 as compared with Wistarsynaptosomes treated with ascorbate/iron.

somes of diabetic GK rats were more resistant to oxida-tion than synaptosomes isolated from Wistar rats.

Table 2 shows the levels of endogenous vitamin E insynaptosomes of diabetic and normal rats. Diabetic GKrats had significantly higher levels of synaptosomal vita-min E (835.0 B 58.04 pmol ·-tocopherol/mg protein)than normal Wistar rats (624.26 B 50.26 pmol ·-tocophe-rol/mg protein).

Effect of Oxidation on the Uptake of [3H]GABA and[3H]GlutamateThe effect of oxidative conditions on the accumulation

of [3H]GABA and [3H]glutamate was analyzed in synap-tosomes isolated from Wistar and GK rats. Figures 2 and3 show that the accumulation of [3H]GABA and [3H]glu-tamate by synaptosomes isolated from Wistar and GKrats is mediated by a Na+-dependent carrier, because in aNa+-free medium (Cl-choline medium), there was only a

Dow

nloa

ded

by:

UC

LA B

iom

edic

al L

ibra

ry

128.

97.9

0.22

1 -

5/8/

2014

10:

25:0

5 A

M

Diabetes and Oxidative Stress Neuroendocrinology 2000;72:179–186 183

residual uptake of these neurotransmitters (2.53 B 0.55and 2.67 B 0.25% of control for GABA and 4.96 B 0.86and 8.08 B 0.96% of control for GABA and glutamate inWistar and GK rats, respectively). Figure 2 shows that theuptake of [3H]GABA by Wistar rat synaptosomes was notsignificantly different from that observed in GK rats.Under oxidizing conditions, the uptake of [3H]GABA wassignificantly decreased in synaptosomes isolated fromWistar rats (from 102.32 B 9.08 to 60.96 B 6.41% ofcontrol). This effect was also observed in synaptosomesisolated from GK rats: the [3H]GABA uptake decreasedfrom 104.71 B 9.66% to 41.39 B 4.41% of that of controlWistar rats. In synaptosomes isolated from GK rats, a sig-nificant decrease in the uptake of [3H]glutamate to 82.36B 7.71% of control was observed. This value of uptakewas not significantly affected after ascorbate/Fe2+ treat-ment. In synaptosomes isolated from Wistar rats, theincubation with ascorbate/Fe2+ significantly decreasedthe accumulation of [3H]glutamate observed under con-trol conditions to values of 40.16 B 5.19% of the respec-tive control. The amounts of [3H]GABA and [3H]gluta-mate accumulated in control synaptosomes isolated fromWistar rats were 180.5 B 16.02 pmol/mg protein and 5.07B 0.79 nmol/mg protein, respectively.

Discussion

In the present study, synaptosomes isolated from thecerebral cortex of nondiabetic Wistar and diabetic GKrats were used to evaluate the effect of diabetes and oxida-tive stress on the accumulation of the amino acid neuro-transmitters GABA and glutamate. Synaptosomes iso-lated from mammalian brain have been currently used asan in vitro model for studying several nerve functions.These isolated nerve terminals are metabolically activeand retain many of the functional properties of nerve end-ings, supporting a wide range of energy-requiring pro-cesses, including the maintenance of membrane potential,the extrusion of calcium, and the accumulation of neuro-transmitters [28]. Our results indicated that the formationof TBARS was significantly decreased in GK rats relativeto controls, and this lower susceptibility of GK rats tomembrane lipid peroxidation seems to be correlated withthe higher synaptosomal vitamin E content measured inthese rats. We also observed that diabetes differentiallyaffected the uptake of neurotransmitters: the uptake ofGABA was not modified whereas that of glutamate de-creased in GK-derived synaptosomes as compared withWistar control rats. However, glutamate uptake, in dia-

betic rats, was not susceptible to oxidative stress, whereasit was significantly reduced in Wistar rats.

Many authors have associated the untreated diabeticstate with oxidative stress in the etiology of diabetic com-plications. Evidence for oxidative stress in diabetes in-cludes reports of increased free radical generated lipidperoxidation [4, 9, 29, 30], protein oxidation [8, 31],DNA oxidation [32], and changes in antioxidant capacity,observed both in clinical and experimental diabetes [4].Recent studies reported an increase in total and oxidizedglutathione (GSSG) [4] and catalase activity and a re-duced glutathione content in sciatic nerve and liver mito-chondria from diabetic rats [4, 33], although Obrosova etal. [34] reported a decrease in glutathione levels in diabet-ic rat lens.

The significant increase in both HbA1c and glycemialevels in diabetic GK rats is consistent with the resultsobtained by Ferreira et al. [33]. These values indicate thatdiabetic rats are exposed to constant hyperglycemic levelsthat have been described to cause increased oxidativestress [35]. The increased formation of reducing sugarsthrough glycolysis and polyol pathways and the subse-quent nonenzymatic glycation of lipids and proteins [36]have been shown to increase the production of ROS [37].

Despite the enhancement of ROS production, inducedby hyperglycemia, the higher content in vitamin E in GKrat synaptosomes is consistent with the lower oxidationlevels observed. These results are in agreement with theobservations of other investigators [4] and suggest that themetabolism or storage of vitamin E in diabetic rats mightbe modified [38]. Recently, it has been reported that dia-betic heart, kidney, and liver were more resistant to lipidperoxidation than those from controls, suggesting thatthese tissues developed a protective mechanism againstoxidative stress [3, 38, 39].

We observed that diabetes and oxidative stress differ-entially modified the uptake of glutamate and GABA.The most important observation appeared to be the inhi-bition of glutamate uptake in the GK-derived synapto-somal preparation, a result that is consistent with the pos-sible effect of ROS in the functioning of the glutamatetransporter. It has been reported that ROS decrease bothglutamate uptake affinity and velocity [40–43] and thatthis effect is neutralized by a mixture of the scavengerenzymes superoxide dismutase and catalase [40, 41, 43]or by some antioxidants [16]. These authors postulat-ed that oxidation of sulfhydryl groups in the transport-er (rather than lipid peroxidation) might play a centralrole in the molecular mechanism by which ROS in-hibit glutamate uptake. Harris et al. [44] hypothesized

Dow

nloa

ded

by:

UC

LA B

iom

edic

al L

ibra

ry

128.

97.9

0.22

1 -

5/8/

2014

10:

25:0

5 A

M

184 Neuroendocrinology 2000;72:179–186 Duarte/Santos/Seiça/Resende de Oliveira

that the inhibition of glutamate uptake might be causedby three potential mechanisms: (1) the inhibition ofNa+,K+ATPase and subsequent dissipation of the trans-membrane Na+ and K+ gradients, essential for glutamatetransport; (2) direct oxidation/damage of the transportermolecule that contains sulfhydryl groups (the modifica-tion of which inhibits the uptake), while oxidation of ami-no acids like histidine and lysine (essential for transporterfunction) may inhibit its activity, and (3) membrane lipidperoxidation may remove structural requirements neces-sary for proper transporter function. Madl and Burgesser[45] hypothesized that moderate ATP depletion mightincrease extracellular excitatory amino acids (EAAs) bydecreasing the uptake, while severe ATP depletion mightfurther increase extracellular EAAs by reversing the trans-porter function, thereby releasing the cytosolic neuronalpools of EAAs.

Keller et al. [17, 46] observed that 4-hydroxynonenal,an aldehydic product of lipid peroxidation, mediated oxi-dation-induced impairment of glutamate uptake and ledto excessive accumulation of extracellular glutamate, re-sulting in a sustained activation of glutamate receptorsand, ultimately, in neuronal death. The vulnerability ofglutamate transporters to oxidants further contributed tothe idea that excitotoxic and oxidative mechanisms mightbe interdependent and act synergistically to induce neu-ronal damage [41].

High-affinity glutamate uptake systems, present insynaptic terminals and astrocytes, represent the majormechanism to control the extracellular levels of EAA andto keep them below neurotoxic values [40]. These gluta-mate transporters seem to be regulated by the surroundingredox environment [41], and defects in glutamate uptakehave been reported both in acute and long-term neurode-generative pathologies associated with oxidative stress[40]. In control Wistar rats a decrease in the uptake ofboth GABA and glutamate, under oxidative stress condi-tions (induced by ascorbate/Fe2+), was observed, whichagrees with the results obtained by other investigators [17,40, 41, 46, 47]. In diabetic GK rat brain synaptosomes,and under oxidizing conditions, the uptake of [3H]GABAwas decreased, and the uptake of [3H]glutamate was unaf-fected, suggesting a protective effect against glutamate-induced neurotoxicity.

Diabetes has been associated with the deterioration ofcognitive performance and modification of cerebral neu-rochemistry, metabolism, and vasculature, leading tolong-term deleterious effects on the brain [5, 48, 49].Recently, it has been suggested that diabetes mellituscould be associated with the pathogenesis of Alzheimer’s

disease [50, 51]. This hypothesis is based on increasedprotein glycation and advanced glycated end productsreported in plaques and tangles of patients with Alzhei-mer’s disease [52], suggesting that, as occurs in diabetes,advanced glycated end product and ROS-induced dam-ages are involved in the pathogenesis of the disease [3,50–52]. A recent study demonstrated that the intrahippo-campal administration of ß-amyloid, in hyperglycemicstreptozotocin diabetic rats, increased neuronal death ascompared with normoglycemic conditions [53].

In summary, we reported that diabetic GK rat brainsynaptosomal preparations were less susceptible to oxida-tive stress than those of control Wistar rats, suggesting aprotective mechanism against free radical induced dam-age. The higher content in vitamin E, observed in thebrain of diabetic rats, further supports this hypothesis.The lower levels of glutamate uptake, reported in diabeticrats relatively to control rats, suggest that the diabeticstate in GK rats induces a selective decrease in glutamatetransporter activity, with no changes in GABA transport-er. However, under oxidizing conditions, the glutamateuptake was not modified, suggesting the development ofdefense systems against brain oxidative injury in GKrats.

From our results, we can conclude that diabetes andhyperglycemia modify glutamate transport function andincrease neuronal damage occurring under pathologicalconditions, namely, when the energy metabolism is com-promised (such as by ischemia, hypoxia, or ß-amyloiddeposition) [50]. Taking into account that glutamate is themain EAA in the brain, the decreased glutamate uptakemay be assumed to contribute to cerebral injury observedin diabetic patients.

Acknowledgments

We wish to thank Dr. Luis Rosario (Department of Biochemistry,University of Coimbra) for all the facilities conceded in the use ofGK rats and Prof. Joao Patrıcio and coworkers (Laboratory of Ani-mal Research Center, University Hospitals, Coimbra) for all the helpin the maintenance of animals. We also wish to thank Dr. TeresaProença (Laboratory of Neurochemistry, University Hospitals,Coimbra) for her technical assistance in the determination of vita-min E contents. This work was supported by FCT (PortugueseResearch Council). Ana Duarte is a recipient of a fellowship fromFCT (Praxis XXI/BTI/16713/98).

Dow

nloa

ded

by:

UC

LA B

iom

edic

al L

ibra

ry

128.

97.9

0.22

1 -

5/8/

2014

10:

25:0

5 A

M

Diabetes and Oxidative Stress Neuroendocrinology 2000;72:179–186 185

References

1 Oberley LW: Free radicals and diabetes. FreeRadic Biol Med 1988;5:113–124.

2 Baynes JW: Role of oxidative stress in develop-ment of complications in diabetes. Diabetes1991;40:405–412.

3 Wolff SP, Jiang ZY, Hunt JV: Protein glycationand oxidative stress in diabetes and ageing.Free Radic Biol Med 1991;10:339–352.

4 Van Dam PS, Van Asbeck BS, Bravenboer B,Van Oirschot JFLM, Gispen WH, Marx JJM:Nerve function and oxidative stress in diabeticand vitamin E-deficient rats. Free Radic BiolMed 1998;24:18–26.

5 Mooradian AD: Diabetic complications of thecentral nervous system. Endocr Rev 1988;9:346–356.

6 Singh S, Melkani GC, Rani C, Gaur SPS, Agra-wal V, Agrawal CG: Oxidative stress and meta-bolic control in non-insulin dependent diabetesmellitus. Indian J Biochem Biophys 1997;34:512–517.

7 Palmer AM, Thomas CR, Gopaul N, Dhir S,Änggård EE, Poston L, Tribe R: Dietary an-tioxidant supplementation reduces lipid perox-idation but impairs vascular function in smallmesenteric arteries of the streptozotocin-dia-betic rat. Diabetologia 1988;41:148–156.

8 Traverso N, Menini S, Cosso L, Odetti P, Alba-no E, Pronzato MA, Marinari UM: Immuno-logical evidence for increased oxidative stressin diabetic rats. Diabetologia 1998;41:265–270.

9 Van Dam PS, Bravenboer B: Oxidative stressand antioxidant treatment in diabetic neuropa-thy. Neurosci Res Commun 1997;21:41–48.

10 LeBel CP, Bondy SC: Oxygen radicals: Com-mon mediators of neurotoxicity. NeurotoxicolTeratol 1991;13:341–346.

11 Halliwell B: Reactive oxygen species and thenervous system. J Neurochem 1992;59:1609–1623.

12 Olanow CW: A radical hypothesis for neurode-generation. Trends Neurosci 1993;16:439–444.

13 Vilchis C, Salceda R: Effect of diabetes on lev-els and uptake of putative amino acid neuro-transmitters in rat retina and retinal pigmentepithelium. Neurochem Res 1996;21:1167–1171.

14 Agostinho P, Duarte CB, Carbalho AP, Olivei-ra CR: Effect of oxidative stress on the releaseof [3H]GABA in cultured chick retina cells.Brain Res 1994;655:213–221.

15 Rego AC, Santos MS, Oliveira CR: Oxidativestress, hypoxia and ischemia-like conditions in-crease the release of endogenous amino acidsby distinct mechanisms in cultured retinalcells. J Neurochem 1996;66:2506–2516.

16 Agostinho P, Duarte CB, Oliveira CR: Impair-ment of excitatory amino acid transporter ac-tivity by oxidative stress conditions in retinalcells: Effect of antioxidants. FASEB J 1997;11:154–163.

17 Keller JN, Mark RJ, Bruce AJ, Blanc E, Roth-stein JD, Uchida K, Waeg G, Mattson MP: 4-Hydroxynonenal, an aldehydic product ofmembrane lipid peroxidation, impairs gluta-mate transport and mitochondrial function insynaptosomes. Neuroscience 1997;80:685–696.

18 Pereira C, Moreira P, Seiça R, Santos MS,Oliveira C: Susceptibility to ß-amyloid-in-duced toxicity is decreased in Goto-Kakizakirats: Involvement of oxidative stress. Exp Neu-rol 2000;161:383–391.

19 Portha B, Giroix MH, Serradas P, Morin L,Saulnier C, Bailbe D: Glucose refractoriness ofpancreatic ß-cells in rat models of non-insulindependent diabetes. Diabète Métab 1994;20:108–115.

20 Duhault J, Koening-Berard E: Le diabète et sesmodèles animaux. Thérapie 1997;52:375–384.

21 Hajos F: An improved method for the prepara-tion of synaptosomal fractions in high purity.Brain Res 1975;93:485–489.

22 Wills ED: Lipid peroxide formation in micro-somes: General considerations. Biochem J1969;113:315–324.

23 Ernster L, Nerdenbrand K: Microsomal lipidperoxidation. Methods Enzymol 1967;10:574–580.

24 Vatassery GT, Younoszai R: Alpha tocopherollevels in various regions of the central nervoussystems of the rat and guinea pig. Lipids 1978;13:828–831.

25 Santos MS, Gonçalves PP, Carvalho AP: Effectof ouabain on the Á-[3H]aminobutyric acid up-take and release in the absence of Ca++ and K+

depolarization. J Pharmcol Exp Ther 1990;253:620–627.

26 Cardoso SM, Pereira C, Oliveira CR: The pro-tective effect of vitamin E, idebenone, andreduced glutathione on free radical mediatedinjury in rat brain synaptosomes. Biochem Bio-phys Res Commun 1998;246:703–710.

27 Santos MS, Duarte AI, Moreira PI, OliveiraCR: Synaptosomal response to oxidative stress:Effect of vinpocetine. Free Radic Res 2000;32:57–66.

28 Nicholls DG: Release of glutamate, aspartate,and Á-aminobutyric acid from isolated nerveterminals. J Neurochem 1989;52:331–341.

29 Gallou G, Ruelland A, Legras B, Maugendre D,Allannic H, Cloarec L: Plasma malondialde-hyde in type 1 and type 2 diabetic patients.Clin Chim Acta 1993;214:227–234.

30 Tiedge M, Lortz S, Drinkgern J, Lenzen S:Relation between antioxidant enzyme gene ex-pression and antioxidative defense status ofinsulin-producing cells. Diabetes 1997;46:1733–1742.

31 Cinaz P, Hasanoglu A, Bideci A, Bibenoglu G:Plasma and erythrocyte vitamin E levels inchildren with insulin dependent diabetes melli-tus. J Pediatr Endocrinol Metab 1999;12:193–196.

32 Dandona P, Thusu K, Cook S, Snyder B, Ma-kowski J, Armstrong D, Nicotera T: Oxidativedamage to DNA in diabetes mellitus. Lancet1996;347:444–445.

33 Ferreira FML, Palmeira CM, Matos MJ, SeiçaR, Santos MS: Decreased susceptibility to lipidperoxidation of Goto-Kakizaki rats: Relation-ship to mitochondrial antioxidant capacity.Life Sci 1999;65:1013–1025.

34 Obrosova I, Cao X, Greene DA, Stevens MJ:Diabetes-induced changes in lens antioxidantstatus, glucose utilization and energy metabo-lism: Effect of DL-·-lipoic acid. Diabetologia1998;41:1442–1450.

35 Kaneto H, Fujii J, Myint T, Miyazawa N,Islam KN, Kawasaki Y, Suzuki K, NakamuraM, Tatsumi H, Yamasaki Y, Taniguchi N: Re-ducing sugars trigger oxidative modificationand apoptosis in pancreatic beta-cells by pro-voking oxidative stress through the glycationreaction. Biochem J 1996;320:855–863.

36 Flatt PR, Abdel-Wahab YHA, Boyd AC, Bar-nett CR, O’Harte FPM: Pancreatic B-cell dys-function and glucose toxicity in non-insulin-dependent diabetes. Proc Nutr Soc 1997;56:243–262.

37 Hunt JV, Dean RT, Wolff SP: Hydroxyl radicalproduction and autoxidative glycosylation:Glucose autoxidation as the cause of proteindamage in the experimental glycation model ofdiabetes mellitus and ageing. Biochem J 1988;256:205–212.

38 Sukalski KA, Pinto KA, Berntson JL: De-creased susceptibility of liver mitochondriafrom diabetic rats to oxidative damage andassociated increase in ·-tocopherol. Free RadicBiol Med 1993;14:57–65.

39 Parinandi NL, Thompson EW, Schmid HHO:Diabetic heart and kidney exhibit increasedresistance to lipid peroxidation. Biochim Bio-phys Acta 1990;1047:63–69.

40 Volterra A, Trotti D, Tromba C, Floridi R,Racagni G: Glutamate uptake inhibition byoxygen free radicals in rat cortical astrocytes. JNeurosci 1994;14:2924–2932.

41 Trotti D, Danbolt NC, Volterra A: Glutamatetransporters are oxidant -vulnerable: A molecu-lar link between oxidative and excitotoxic neu-rodegeneration? Trends Pharmacol Sci 1998;19:328–334.

42 Rothstein JD, Martin LJ, Kunkl RW: De-creased glutamate transport by the brain andspinal cord in amyotrophic lateral sclerosis. NEngl J Med 1992;326:1464–1468.

43 Volterra A: Inhibition of high-affinity gluta-mate transport in neuronal and glial cells byarachidonic acid and oxygen-free radicals. Re-nal Physiol Biochem 1994;17:165–167.

44 Harris ME, Wang Y, Pedigo NW, Hensley K,Butterfield DA, Carney JM: Amyloid ß peptide(25–35) inhibits Na+-dependent glutamate up-take in rat hippocampal astrocyte cultures. JNeurochem 1996;67:277–286.

Dow

nloa

ded

by:

UC

LA B

iom

edic

al L

ibra

ry

128.

97.9

0.22

1 -

5/8/

2014

10:

25:0

5 A

M

186 Neuroendocrinology 2000;72:179–186 Duarte/Santos/Seiça/Resende de Oliveira

45 Madl JE, Burgesser K: Adenosine triphosphatedepletion reverses sodium-dependent, neuron-al uptake of glutamate in rat hippocampalslices. J Neurosci 1993;13:4429–4444.

46 Keller JN, Pang Z, Geddes JW, Begley JG, Ger-meyer A, Waeg G, Mattson M: Impairment ofglucose and glutamate transport and inductionof mitochondrial oxidative stress and dysfunc-tion in synaptosomes by amyloid ß-peptide.Role of the lipid peroxidation product 4-hy-droxynonenal. J Neurochem 1997;69:273–284.

47 Rafalowska U, Liu GJ, Floyd RA: Peroxida-tion induced changes in synaptosomal trans-port of dopamine and Á-aminobutyric acid.Free Radic Biol Med 1989;6:485–492.

48 Biessels GJ, Kappelle AC, Bravenboer B, Er-kelens DW, Gispen WH: Cerebral function indiabetes mellitus. Diabetologia 1994;37:643–650.

49 McCall AL: The impact of diabetes on theCNS. Diabetes 1992;41:557–570.

50 Messier C, Gagnon M: Clucose regulation andcognitive functions: Relation to Alzheimer’sdisease and diabetes. Behav Brain Res 1996;75:1–11.

51 Sasaki N, Fukatsu R, Tsuzuki K, Hayashi Y,Yoshida T, Fujii N, Koike T, Wakayama I,Yanagihara R, Garruto R, Amano N, MakitaZ: Advanced glycation end products in Alz-heimer’s disease and other neurodegenerativediseases. Am J Pathol 1998;153:1149–1155.

52 Smith MA, Sayre LM, Vitek MP, MonnierVM, Perry G: Early ageing and Alzheimer’s.nature 1995;374:316.

53 Smith MD, Kesslak JP, Cummings BJ, Cot-man CW: Analysis of brain injury followingintrahippocampal administration of beta-amy-loid in streptozotocin-treated rats. NeurobiolAging 1994;15:153–159.

Dow

nloa

ded

by:

UC

LA B

iom

edic

al L

ibra

ry

128.

97.9

0.22

1 -

5/8/

2014

10:

25:0

5 A

M