Brain Research Bulletin 65 (2005) 443–450

Acute and repeated restraint stress influences cellular damage in rathippocampal slices exposed to oxygen and glucose deprivation

Fernanda Urruth Fontella, Helena Cimarosti, Leonardo Machado Crema, Ana Paula Thomazi,Marina Concli Leite, Christianne Salbego, Carlos Alberto Saraiva Gonc¸alves,

Susana Wofchuk, Carla Dalmaz∗, Carlos Alexandre NettoPPG-Fisiologia e Departamento de Bioqu´ımica, Instituto de Ciˆencias B´asicas da Sa´ude, UFRGS, Rua Ramiro Barcelos,

2600 anexo, 90035-003 Porto Alegre, RS, Brazil

Received 17 August 2004; received in revised form 24 January 2005; accepted 17 February 2005Available online 14 March 2005

Abstract

Several studies have shown that high corticosteroid hormone levels increase neuronal vulnerability. Here we evaluate the consequenceso . Cellulari e stress didn xposure toO , on LDHa cytic protein.A s increases thev exposuret ponsible forr©

K

1

stemeipe

b

henia,

caf-roid

pone

s ofmin-

is in-t

t ef-cam-erved

0d

f in vivo acute or repeated restraint stress on cellular viability in rat hippocampal slices suffering an in vitro model of ischemianjury was quantified by measuring lactate dehydrogenase (LDH) and neuron-specific enolase released into the medium. Acutot affect cellular death when oxygen and glucose deprivation (OGD) was applied both immediately or 24 h after restraint. The eGD, followed by reoxygenation, resulted in increased LDH in the medium. Repeated stress potentiated the effect of OGD bothnd neuron-specific enolase released to the medium. There was no effect of repeated stress on the release of S100B, an astrodditionally, no effect of repeated stress was observed on glutamate uptake by the tissue. These results suggest that repeated stresulnerability of hippocampal cells to an in vitro model of ischemia, potentiating cellular damage, and that the cells damaged by theo repeated stress + OGD are mostly neurons. The uptake of glutamate was not observed to participate in the mechanisms resendering the neurons more susceptible to ischemic damage after repeated stress.

2005 Elsevier Inc. All rights reserved.

eywords: Ischemia; Hippocampus; LDH; Neuron-specific enolase; Glutamate uptake; Glucocorticoids

. Introduction

Evidence from the literature suggests that exposure to per-istently high levels of corticosterone or repeated exposureo stressors present adverse effects on various brain regions,specially the hippocampus[15,16,40,52,55,56]. These hor-ones exacerbate the toxicity of other insults, possibly accel-rating the decline in ATP levels in neurons during periods of

ncreased energy demand[4,15,16,29,59,64,65]. For exam-le, acute, short exposure to glucocorticoids can significantlyxacerbate postischemic outcome[47,48].

Ischemia is defined as a severe reduction or completelockage of blood flow[68], and is a pathophysiological event

∗ Corresponding author. Tel.: +55 51 33165540; fax: +55 51 33165540.E-mail address:carladalmaz@yahoo.com.br (C. Dalmaz).

that frequently results in cerebral damage. Therefore, whigher glucocorticoid levels are combined with ischemthere is an aggravation in cell loss caused by in vitro[47]or in vivo models of ischemia[27,53]. In addition, synaptifunction along with cellular integrity can be preservedter hypoxia–ischemia by preventing the rise in corticostelevels by the use of the steroid synthesis inhibitor metyra[1,2,27]or by adrenalectomy[3,43].

Most of the available studies concerning the effectglucocorticoids on ischemic events have used acute adistration of, or in vitro exposure to these hormones[6,64],or reduced glucocorticoid release by the use of syntheshibitors or adrenalectomy[25,27,55,58]. Chronic treatmenwith glucocorticoids has been shown to have differenfects on ischemia-induced neuronal damage in hippopus; some behavioral and histological studies have obs

361-9230/$ – see front matter © 2005 Elsevier Inc. All rights reserved.oi:10.1016/j.brainresbull.2005.02.026

444 F.U. Fontella et al. / Brain Research Bulletin 65 (2005) 443–450

aggravated damage[24,51,61], while others suggest a reduc-tion in brain damage following hypoxia–ischemia[26,33].Chronic exposure to stressors, however, presents several dif-ferent aspects in relation to chronic administration of glu-cocorticoids: first, it is well known that repeated exposureto the same stressor can lead to a process of adaptation tothat stimulus[4,5,7,8,21,28,34]. Accordingly, it is importantto note that the levels of corticosterone observed follow-ing acute restraint stress are higher than those that followthe same restraint when repeated several times[65]. Sec-ondly, besides releasing glucocorticoids, the stress responseinvolves a large number of neurotransmitters, neuropeptidesand neuromodulators in various brain regions (for a reviewsee[46]).

Damage induced by both procedures, stress and ischemiahas been suggested to be due, at least in part, to increased glu-tamatergic tonus[30,32,41,63,67]. In such a situation, trans-porter proteins which uptake glutamate play an importantrole in the termination of glutamatergic neurotransmissionand prevention of excitotoxicity[12].

Our hypothesis is that the exposure to stress will possiblyincrease susceptibility to ischemic injury. Both acute and in-termittent stress will be used, since adaptation to stress maylead to a different outcome in this condition. The presentstudy aimed to evaluate the consequences of acute or re-peated exposure to restraint stress on cellular function aftero delo ls them onse[ resse dg ctedt d 24 ha ondi-t

2

2

be-g sed.T ps off erial( -d ycle( era-t ardl re-s s per-f entsw fol-l il forL de

to minimize animal suffering as well as to reduce the numberof animals.

2.2. Stress procedure

2.2.1. Acute treatmentThe animals were submitted to a single exposure (1 h re-

straint) and the control group was not stressed. Restraint wascarried out by placing the animal in a 25 cm× 7 cm plasticbottle, adjusting it with plaster tape on the outside, so that theanimal was unable to move. There was a 1 cm hole in one farend for breathing.

2.2.2. Repeated stress treatmentRepeated restraint stress was based on the procedure de-

scribed by Gamaro et al.[18]. Animals were separated intwo groups: stressed and control. Rats were stressed 1 h/day,5 days a week for 8 weeks. Control animals were kept intheir homecages. About 2 weeks before the biochemical ex-periments, controls were gently handled for a few momentsat the same time the stressed group was being stressed. Ratswere killed 24 h after the last restraint section.

2.3. Preparation and incubation of slices

Rats were decapitated, their hippocampi were quickly dis-s -t acedi red:c rein-c tion( MK1a

2

wasb rw d,c eleits ing(2 rre-s other6 the“ cu-b asp Dm withN ound3 cedw edo r 3 ha

xygen and glucose deprivation (OGD), an in vitro mof hypoxic–ischemic events[9,42,44], in rat hippocampalices. This structure was chosen since it is one ofost important brain structures mediating the stress resp

36–39,56], being also a region particularly sensitive to stffects, as mentioned earlier (e.g.[52]). We also evaluatelutamate uptake by hippocampal slices from rats subje

o repeated exposure to restraint stress, and sacrificefterwards, when these slices were exposed to OGD c

ions.

. Materials and methods

.1. Subjects

Sixty-four adult male Wistar rats (60 days at theinning of the treatment) weighing 180–230 g were uhe experimentally naive animals were housed in grou

our to five rats, in homecages made of Plexiglas mat65 cm× 25 cm× 15 cm) with the floor covered with sawust. They were maintained on a standard dark/light clights on between 7:00 and 19:00 h), with a room tempure of 22± 2◦C. Animals had free access to food (standab rat chow) and water, except during the period whentraint stress was applied. The restraint procedure waormed between 13:00 and 15:00 h. All animal treatmere in accordance with the institutional guidelines and

owed the recommendations of the International Councaboratory Animal Science (ICLAS); all efforts were ma

ected out and transverse sections (400�m) were rapidly obained using a McIlwain tissue chopper. One slice was plnto each well of a 24-well culture plate (plates were paiontrol and oxygen and glucose deprived-OGD) and pubated for 15 min in a modified Krebs–Henseleit solupreincubation solution) containing 120 mM NaCl, 2 mCl, 0.5 mM CaCl2, 10 mM MgSO4, 26 mM NaHCO3,.18 mM KH2PO4 and 11 mM glucose (pH 7.4) at 37◦C inn atmosphere of 5% CO2 (95% O2/5% CO2) [9].

.4. Oxygen and glucose deprivation (OGD)

The exposure to oxygen and glucose deprivationased on the method described by Strasser and Fische[60],ith some modifications[11]. After the preincubation perioontrol slices were incubated in a modified Krebs–Hensolution (incubation solution-control medium) containmM): NaCl 120, KCl 2, CaCl2 2, MgSO4 1.19, NaHCO36, KH2PO4 1.18 and glucose 11 (pH 7.4). The slices coponding to the control group were then incubated for an0 min, in the same conditions. Slices corresponding toischemic” experimental group were rinsed once with ination solution without glucose (OGD medium), which wreviously bubbled with N2 for 30 s, and incubated in OGedium for 60 min in an anaerobic chamber, saturated2. During the procedure the temperature was kept ar7◦C. After this period, the incubation solution was replaith fresh control medium (“ischemic” slices were first rinsnce with control medium) and slices were incubated fot 37◦ C in an atmosphere of 5% CO2 (95% O2/5% CO2)

F.U. Fontella et al. / Brain Research Bulletin 65 (2005) 443–450 445

to simulate a “reperfusion” period[9]. In a second experi-ment, this same methodology was used, but 10 mM lactatewas added to the incubation medium instead of glucose.

2.5. Assessment of neural injury—LDH assay

Neural cell injury was quantified by the measurement oflactate dehydrogenase (LDH) released from damaged cellsinto the extracellular fluid[23]. LDH efflux occurs from eithernecrotic or apoptotic cells, and is proportional to the numberof damaged cells[22,23,31].

LDH activity was determined using a commercial kit(Doles Reagentes); the activity of this enzyme was assessedin the bathing fluid. Following the conversion of exogenouslyadded lactate to pyruvate, 1,10-phenantroline is converted toa colored complex, after a chain of reactions resulting fromthe NADH formed by the enzymatic reaction, which is mea-sured using a spectrophotometric method (490 nm).

Protein content was measured in aliquots of the dissociatedslices following the method described by Peterson[49].

2.6. Neuron-specific enolase (NSE) determination

NSE was determined by ELISA (CanAg Diagnostics AB).The incubation medium was collected after 3 h of reperfusionand stored at−20◦ C. Samples (25�L) were applied on mi-c SE,a hing,p ationc sub-s in int otiterp

2

cytesi e-swS vi-o car-b min.A onalD uedf sub-s batedf ureda

2

2tely

r olu-t

4.17 NaHCO3; 5.36 KCl; 0.44 KH2PO4; 1.26 CaCl2; 0.41MgSO4; 0.49 MgCl2 and 1.11 glucose, in pH 7.2. Hip-pocampi were dissected onto Petri dishes with HBSS andslices (0.4 mm) were obtained using a McIIwain tissue chop-per. Slices of hippocampus were separated with the help ofa magnifying glass and transferred to 24-well culture plates:one plate was maintained at 35◦C and the other on ice. Theslices from the first plate were washed once with 1 mL of35◦C HBSS and the second with 1 mL of 4◦C sodium-freeHBSS for the analysis of non-specific uptake (see below).

2.8.2. Total uptakeGlutamate uptake was performed according to Frizzo

et al. [17], adjusted for hippocampus in accordance withincubation times. Slices were incubated at 35◦C with0.66�Ci mL−1 l-[3H]glutamate and 100�M (final concen-tration) unlabeled glutamate in glucose–HBSS solution. In-cubation was stopped after 5 min with two ice-cold washes of1 mL HBSS, immediately followed by the addition of 0.2 mL0.5N NaOH; this material was then kept overnight.

2.8.3. Sodium-independent uptake (non-specific uptake)To measure sodium-independent uptake, the same proto-

col described above was used, though with differences inthe temperature and the medium used. Sodium-independentuptake was determined on ice (4◦C), usingN-methyl-d-g sub-t Botht trip-l

2uid

s

2de-

s

2

dw n-c

3

3

di-a

s af-f LDHa sin-g iately

rotiter plates previously coated with monoclonal anti-Nnd incubated overnight at room temperature. After waseroxidase-conjugated anti-NSE was added and incubontinued for 1 h. The plate was washed and peroxidasetrate was added and the plate incubated for a further 5 mhe dark. Absorbance was measured at 620 nm on a micrlate reader.

.7. S100B determination

The assay for S100B, a protein released by astron conditions of injury, was carried out by ELISA as dcribed by Tramontina et al.[66]. Briefly, incubation mediumas collected after 3 h of reperfusion and stored at−20◦C.amples (50�L) were applied on microtiter plates preusly coated with monoclonal anti-S100B (Sigma) inonate buffer and blocked with 1% bovine serum albufter washing, peroxidase-conjugated anti-S100 (polyclAKO) diluted 1:1000 was added and incubation contin

or 1 h. The plate was washed and 0.2 mL of peroxidasetrate (Sigma Fast OPD) was added and the plate incuor a further 30 min in the dark. Absorbance was meast 492 nm on a microtiter plate reader.

.8. Glutamate uptake

.8.1. Slices preparationThe animals were decapitated, their brains immedia

emoved and humidified with Hank’s balanced salt sion (HBSS) containing (in mM): 137 NaCl; 0.63 Na2HPO4;

lucamine instead of sodium chloride. The results wereracted from the total uptake to obtain the specific one.he specific and non-specific uptakes were performed inicate.

.8.4. Radioactivity quantificationIncorporated radioactivity was measured using a liq

cintillation counter (Wallac, model 1409).

.8.5. Protein measurementProtein content was measured following the method

cribed by Peterson[49].

.9. Statistical analysis

Data are expressed as mean± standard error (S.E.M.) anere analyzed using a two-way ANOVA, followed by Duan’s multiple range test when indicated.

. Results

.1. Experiment 1

Effect of OGD on hippocampal slices obtained immetely and 24 h after 1 h of restraint stress.

In order to evaluate whether exposure to acute stresects cellular damage induced by OGD, we assessedctivity in hippocampal slices from rats submitted to onele exposure of restraint stress, and sacrificed immed

446 F.U. Fontella et al. / Brain Research Bulletin 65 (2005) 443–450

Fig. 1. Effect of acute stress on LDH activity released to the medium(mean± S.E.M.) in control slices or in slices exposed to OGD for 60 min andsubmitted to a 3 h reoxygenation period, immediately after acute restraint(A) or 24 h after acute restraint. (B) (n= 4–5/group). (*) Significantly dif-ferent from control and acute stress groups, not submitted to OGD (Duncantest,P< 0.05).

or 24 h afterwards.Fig. 1A shows that, in slices obtainedfrom control rats and from animals sacrificed immediatelyafter acute restraint, OGD caused increased LDH release intothe medium [two-way ANOVA,F(1, 15) = 70.54,P< 0.001],but no difference was observed after stress exposure whencompared to control [F(1, 15) = 2.42,P> 0.05]. Acute stressdid not interfere with the effects of OGD [F(1, 15) = 0.57,P> 0.05 when tested for interaction].

Similar results were found after 24 h: there was no signif-icant difference in LDH release between stressed and controlgroups (Fig. 1B); the only significant effect was that causedby OGD [two-way ANOVA,F(1, 14) = 20.19,P< 0.001 forOGD andF(1, 14) = 0.75,P> 0.05 for the stress treatment].No interaction between stress and OGD was observed [F(1,14) = 0.02,P> 0.05].

3.2. Experiment 2

Effect of OGD in hippocampal slices from rats previouslysubmitted to repeated restraint stress.

This experiment aimed to evaluate whether previous expo-sure to repeated restraint stress interferes with the neuronaldamage induced by OGD. After 3 h of reoxygenation fol-lowing 60 min of OGD, LDH activity was assayed in themedium.Fig. 2shows that both treatments (OGD and stress)a A,F2 m

Fig. 2. Effect of repeated stress on LDH activity released to the medium(mean± S.E.M.) in control slices or in slices exposed to OGD for 60 minand submitted to a 3 h reoxygenation period (n= 8/group), 24 h after the lastrestraint stress section. (*) Significantly different from control and chronicstress groups, not submitted to OGD (Duncan test,P< 0.05). (**) Signifi-cantly different from all other groups (Duncan test,P< 0.05).

repeatedly stressed animals had LDH levels similar to thoseof controls, however, the group submitted to both treatmentspresented higher levels of LDH in the medium than those ofall other groups.

3.3. Experiment 3

Effect of OGD on hippocampal slices from rats previouslysubmitted to repeated restraint stress, when lactate was usedas substrate.

Neurons preferentially utilize lactate both in vivo and invitro, as substrate[62]. In order to evaluate whether differ-ent substrates may result in distinct cellular damage whenslices from animals subjected to repeated stress are submit-ted to OGD, we assessed LDH activity in hippocampal slicesfrom rats subjected to repeated exposure to restraint stress,and sacrificed 24 h afterwards, when lactate is added to themedium instead of glucose.Fig. 3 shows that OGD causedincreased LDH release into the medium [two-way ANOVA,F(1, 15) = 33.43,P< 0.001], and the same effect was observedafter stress exposure [F(1, 15) = 10.94,P< 0.001; no inter-action between the variables was observed:F(1, 15) = 1.53,

F dium( mina bstrate( essgd

ffected LDH release into the medium [two-way ANOV(1, 28) = 5.49,P< 0.05 for the stress treatment andF(1,8) = 22.60,P< 0.001 for the OGD treatment]. Slices fro

ig. 3. Effect of repeated stress on LDH activity released to the memean± S.E.M.) in control slices or in slices exposed to OGD for 60nd submitted to a 3 h reoxygenation period, using lactate as a sun= 4–5/group). (*) Significantly different from control and chronic strroups, not submitted to OGD (Duncan test,P< 0.05). (**) Significantlyifferent from all other groups (Duncan test,P< 0.05).

F.U. Fontella et al. / Brain Research Bulletin 65 (2005) 443–450 447

Fig. 4. Effect of repeated stress on (A) NSE and (B) S100B levels in themedium (mean± S.E.M.) in control slices or in slices exposed to OGD for60 min and submitted to a 3 h reoxygenation period (n= 5–6/group). (*)Significantly different from all other groups (Duncan test,P< 0.05).

P> 0.05]. The group submitted to both treatments presentedhigher levels of LDH released into the medium than all othergroups.

3.4. Experiment 4

Effect of OGD on hippocampal slices from rats previouslysubmitted to repeated restraint stress on neuron-specific eno-lase and on S100B levels in the medium.

In order to evaluate whether cellular damage induced byOGD reflects neuronal or astroglial death, we measured bothNSE content and the levels of S100B, a protein produced andreleased by astrocytes[45,50]in the medium.Fig. 4A showsthat OGD caused increased NSE released into the medium[two-way ANOVA,F(1, 17) = 10.86,P< 0.001], and the sameeffect was observed after stress exposure [F(1, 17) = 5.51,P< 0.05; a significant interaction between the OGD and stresswas observed:F(1, 17) = 5.36,P< 0.05]. Here again, thegroup submitted to both treatments presented higher NSE ac-tivity in the medium than all other groups. No effect of OGDor repeated stress was observed (Fig. 4B) on S100B levels inthe medium [two-way ANOVA,F(1, 14) = 0.29,P> 0.05 forOGD;F(1, 14) = 1.33,P> 0.05 for the stress treatment; therewas no interaction:F(1, 14) = 1.19,P> 0.05].

3.5. Experiment 5

pals raints

Fig. 5. Effect of repeated stress in hippocampal slices exposed to OGD onglutamate uptake (mean± S.E.M.) immediately (A) or after 3 h of reperfu-sion (B) (n= 4–5/group). (*) Significantly different from groups not exposedto OGD (Duncan test,P< 0.05).

Fig. 5A shows that, immediately after OGD, there was adecrease in glutamate uptake by hippocampal slices [two-way ANOVA, F(1, 15) = 20.13,P< 0.001]. No effect ofstress exposure was observed [F(1, 15) = 0.21,P> 0.05]. Thesame pattern of effect was observed with 3 h of reperfu-sion [two-way ANOVA,F(1, 15) = 22.97,P< 0.001 for OGD;F(1, 15) = 0.04,P> 0.05 for stress treatment;F(1, 15) = 0.43,P> 0.05 for the interaction] (Fig. 5B).

4. Discussion

ATP produced by aerobic metabolism is the major sourceof energy in the brain, but it may be compromised by the in-terruption of oxygen and substrate delivery and disturbancesin cerebral metabolism, such as the condition resulting fromischemia. The exposure of hippocampal slices to an in vitromodel of hypoxia–ischemia, followed by reoxygenation, re-sulted in increased LDH in the medium, which is a conse-quence of cell damage or death.

The consequences of acute or repeated exposure to stresson cellular integrity after OGD in rat hippocampal sliceswere investigated. We observed that acute and intermittentrepeated stress induced different degrees of susceptibility toischemic injury. Acute stress did not interfere with the ef-fects of OGD exposure, i.e. when the slices were exposed toO ffecto hip-

Effect of OGD on glutamate uptake by hippocamlices from rats previously submitted to repeated resttress.

GD immediately or 24 h after acute stress, only the ef OGD was observed. On the other hand, slices from

448 F.U. Fontella et al. / Brain Research Bulletin 65 (2005) 443–450

pocampus of repeatedly stressed animals presented a highersusceptibility to OGD. An increased damage was observedafter OGD in repeatedly stressed animals when comparedto controls. It is important to consider that the repetition ofa stress phenomenon decreases the corticosterone increaseper stress episode. While rats restrained during 1 h present a216% increase in plasmatic corticosterone, after 40 days ofrestraint this increase is much smaller (about 50% when com-pared to handled controls). Besides, these levels returned tonormal 24 h after restraint, when these experiments were done[65]. Consequently, since enhanced cell death, as measuredby LDH, occurs with the repeated stress paradigm, whileno effect was observed after acute stress; these results couldsuggest that corticosterone does not have a significant rolein the observed cell death. Conversely, there was a marginalincrease in baseline levels in the repeatedly stressed animalsprior to OGD, which may be related to a continued stressresponse, including corticosterone.

Since glucose can be potentially taken up and utilized byneurons as well as astroglia and neurons might preferentiallyutilize lactate as a substrate[62], LDH activity in the mediumwas also assessed when lactate is added to the medium in-stead of glucose, to evaluate whether different substrates mayresult in distinct cellular damage when slices from animalssubjected to repeated stress are submitted to OGD. The sameresults were observed after incubation with glucose (utilizedb izedb e ob-s ittedt in ad-d afterO .

nota it hasb raint wed blishd be-t cts ofc ringi ny-w cedp f thes ittedp

neu-r corti-c ens gea emia[ thef reasei -e n them id noi the

differences observed between effects of corticosterone on invivo hypoxia–ischemia and those of acute stress on in vitrohypoxia–ischemia is the different time used to observe thedamage induced by these processes. Additionally, the stressresponse comprises more than the release of glucocorticoids;rather, other hormones and neurotransmitters are released,which may result in a different effect.

The measurement of LDH released into the medium in-dicates that repeated stress increases the effect of OGD onthis parameter. This result may be interpreted as an increasedvulnerability of these cells to ischemia. This effect is fur-ther evidenced by the increased NSE present in the medium,which is much higher when the slices are submitted to bothsituations, stress and OGD. These results are in agreementwith previous reports, which suggest that repeated stress mayincrease vulnerability of hippocampal neurons to several in-sults, including hypoxia–ischemia[37]. Chronic exposure toelevated titers of glucocorticoids has been demonstrated toresult in cell loss as well as in reduced neuronal plasticityand regeneration[13,36,57]. Some mechanisms suggested toplay a role in this increased vulnerability of hippocampal cellsinclude decreased glucose uptake[19], increased glutamaterelease[37–39], and increased production of oxygen reactivespecies (ROS;[54]).

On the other hand, based on the finding that S100B wasreleased in the medium, no astrocyte damage seems to occura rkerfis de-p ions,S rvivalo lev-e siono a re-v elso henm ationi fici im-mr at-t pairm ls ofi imeso

tresso luta-m tracel-l dingt es duet t,h rterp hasb n of

y neurons and astroglia) or lactate (preferentially utily neurons), suggesting that the increase in the damagerved after OGD when animals were previously submo repeated stress is possibly related to neuronal death;ition, neuron-specific enolase released to the mediumGD was higher in slices from repeatedly stressed ratsIn spite that in this model of ischemic injury we can

ccess long-term alterations in response to OGD insult,een used to investigate the effect on vulnerability of b

issue to ischemic injury of many drugs and agents. Asemonstrate in our experiments, it was possible to estaifferences in brain tissue response to OGD condition

ween stressed and control animals. Studies of the effehronic stress using in vivo ischemic models would bnteresting information concerning long-term effects. Aay, this finding may be relevant for the ischemia-induathophysiological sequelae in vivo, and the effects otress conditions to which the patient was possible submreviously to the ischemic event should be considered.

The same mediators suggested to be involved inonal damage caused by stress exposure, e.g. glucooids, EAA, NMDA receptor activation, and reactive oxygpecies production[54], participate in the neuronal damand death that is caused in pyramidal neurons by isch

10,30,35,51]. This is believed to be one explanation foract that stress and acute corticosterone treatment incschemic damage[3,37,38]. In this model of OGD, howver, no effect of acute restraint stress on LDH release iedium was observed, suggesting that acute stress d

ncrease in vitro cell death. A possible explanation for

t

fter 3 h of reperfusion. This protein is an important maor monitoring damage and activation of astrocytes[50], andt is also a glial modulator of neuronal synaptic plasticity[45],ince secreted glial S100B exerts trophic or toxic effectsending on its concentration: at nanomolar concentrat100B stimulates neurite outgrowth and enhances suf neurons during development. In contrast, micromolarls of extracellular S100B in vitro stimulate the expresf proinflammatory cytokines and induce apoptosis (foriew, see[14]). Clinical studies have shown that serum levf S100B correlate with the degree of brain ischemia weasured many hours later, e.g. 48 h, while no correl

s observed until 15 h[20]. In contrast to NSE, no specincrements of peripheral levels of S100B were observed

ediately after brain damage in stroke[69]. This delayedelease of S100B, found in in vivo situations has beenributed to the putative role of this protein in the brain reechanism. Thus, it is possible that, using in vivo mode

schemia, altered levels may be observed with longer tf reperfusion.

The changes in hippocampal function caused by sr by ischemia have been attributed to an increased gatergic tonus. Exposure to acute stress increases ex

ular levels of glutamate in some brain structures, incluhe hippocampus[32,41]. Accumulation of glutamate in thynaptic cleft may lead to excitotoxic neuronal damageo over stimulation of these receptors. The Na+-dependenigh affinity uptake of glutamate is mediated by transporoteins located mainly in astrocytes and this mechanismeen proved to play an important role in the terminatio

F.U. Fontella et al. / Brain Research Bulletin 65 (2005) 443–450 449

glutamatergic neurotransmission and prevention of excito-toxicity [12]. A compromised glutamate uptake in animalssubmitted to chronic stress could contribute to the higherdamage induced by OGD exposure in these animals. This isnot the case, however, since glutamate uptake was decreasedonly by OGD, with no effect of chronic stress in this param-eter. Therefore, although exposure to acute stress may alterglutamatergic neurotransmission[32,41], the ability of thetissue, mainly carried out by astrocytes, in reuptaking glu-tamate from the medium is not compromised after chronicstress, independently of exposure of cells to OGD. It mightbe, however, that the glutamate uptake is already so low afterOGD, that a further decrease in glutamate uptake cannot bedetected after repeated restraint and OGD.

In conclusion, the present results suggest that: (1) the ef-fects of OGD on cell death in hippocampal slices, as evaluatedby LDH released to the medium, are not altered both imme-diately or 24 h after acute stress; (2) repeated restraint stressseems to worsen cellular damage induced by OGD, at leastas evaluated by LDH and NSE released into the medium; (3)the cells damaged by exposure to repeated stress + OGD aremainly neurons; and (4) the mechanisms responsible for ren-dering the neurons more susceptible to damage after repeatedstress have not been shown to involve glutamate uptake. Themechanisms involved in these effects of repeated stress in asituation of OGD remain to be explored.

A

R

lle-ur. J.

onee on-llularlow.

ionopic

or-r de-

uteation,

ratmin-

veltye re-

n-Brain

[9] A. Cardenas, M.A. Moro, O. Hurtado, J.C. Leza, P. Lorenzo, A.Castrillo, O.G. Bodelon, O.G. Bosca, I. Lizasoain, Implication ofglutamate in the expression of inducible nitric oxide synthase afteroxygen and glucose deprivation in rat forebrain slices, J. Neurochem.74 (2000) 2041–2048.

[10] P.H. Chan, Reactive oxygen radicals in signaling and damage in theischemic brain, J. Cereb. Blood Flow Metab. 21 (2001) 2–14.

[11] H. Cimarosti, R. Rodnight, A. Tavares, R. Paiva, L. Valentim, E.Rocha, C. Salbego, An investigation of the neuroprotective effect oflithium in organotypic slice cultures of rat hippocampus exposed tooxygen and glucose deprivation, Neurosci. Lett. 315 (2001) 33–36.

[12] N.C. Danbolt, Glutamate uptake, Prog. Neurobiol. 65 (2001) 1–105.[13] N.W. Daw, H. Sato, K. Fox, T. Carmichael, R. Gingerich, Cortisol

reduces plasticity in the kitten visual cortex, J Neurobiol. 22 (1991)158–168.

[14] R. Donato, S100: a multigenic family of calcium-modulated proteinsof the EF-hand type with intracellular and extracellular functionalroles, Int. J. Biochem. Cell Biol. 33 (2001) 637–668.

[15] E. Elliot, R. Sapolsky, Corticosterone impairs hippocampal neuralcalcium regulation: possible mediating mechanisms, Brain Res. 602(1993) 84–90.

[16] E. Elliott, M. Mattson, P. Vanderklish, G. Lynch, I. Chang, R. Sapol-sky, Corticosterone exarcebates kainate-induced alterations in hip-pocampal tau immunoreactivity and spectrin proteolysis in vivo, J.Neurochem. 61 (1993) 57–67.

[17] M.E. Frizzo, D.R. Lara, A.S. Prokopiuk, C.R. Vargas, C.G. Salbego,M. Wajner, D.O. Souza, Guanosine enhances glutamate uptake inbrain cortical slices at normal and excitotoxic conditions, Cell. Mol.Neurobiol. 22 (2002) 353–363.

[18] G. Gamaro, M.H. Xavier, J.D. Denardin, J.A. Pilger, D.R. Ely,M.B.C. Ferreira, C. Dalmaz, Effect of restraint stress on nociceptive

[ hibitNeu-

[ y, S.after

[ rgicn: the

ur. J.

[ role992)

[ me-nase

[ mentetab.

[ ma,effectereb.

[ orf,ucesemia

[ euced000)

[ ain-ss isaxis,

cknowledgment

This research was supported by CNPq.

eferences

[1] N. Adachi, J. Chen, K. Liu, T. Nagaro, T. Arai, Metyrapone aviates ischemic neuronal damage in the gerbil hippocampus, EPharmacol. 373 (1999) 147–152.

[2] N. Adachi, J. Chen, K. Liu, S. Tsubota, T. Arai, Dexamethasaggravates ischemia-induced neuronal damage by facilitating thset of anoxic depolarization and the increase in the intraceCa2+ concentration in the gerbil hippocampus, J. Cereb. Blood FMetab. 18 (1998) 274–280.

[3] F.J. Antonawich, G. Miller, D.C. Rigsby, J.N. Davis, Regulatof ischemic cell death by glucocorticoids and adrenocorticotrhormone, Neuroscience 88 (1999) 319–325.

[4] M.P. Armanini, C. Hutchins, B.A. Stein, R.M. Sapolsky, Glucocticoid endangerment of hippocampal neurons is NMDA-receptopendent, Brain Res. 532 (1990) 7–12.

[5] A. Armario, J. Martı, M. Gil, The serum glucose response to acstress is sensitive to the intensity of the stressor and to habituPsychoneuroendocrinology 15 (1990) 341–347.

[6] C.S. Barr, L.A. Dokas, Regulation of pp60(c-src) synthesis inhippocampal slices by in vitro ischemia and glucocorticoid adistration, J. Neurosci. Res. 65 (2001) 340–345.

[7] J.R. Bassett, K.D. Cairncross, M.G. King, Parameters of noshock predictability and response contingency in corticosteronlease in the rat, Physiol. Behav. 10 (1973) 901–907.

[8] L.M. Cancela, C. Bregonzio, V.A. Molina, Anxiolytic-like effect iduced by chronic stress is reversed by naloxone pretreatment,Res. 36 (1995) 209–213.

response, Physiol. Behav. 63 (1998) 693–697.19] H.C. Horner, D.R. Packan, R.M. Sapolsky, Glucocorticoids in

glucose transport in cultured hippocampal neurons and glia,roendocrinology 52 (1990) 57–64.

20] H. Jonsson, P. Johnsson, M. Birch-Iensen, C. Alling, S. WestabBlomquist, S100B as a predictor of size and outcome of strokecardiac surgery, Ann. Thorac. Surg. 71 (2001) 1433–1437.

21] G.A. Kennett, S.L. Dickinson, G. Curzon, Central serotonineresponses and behavioral adaptation to repeated immobilisatioeffect of the corticosterone synthesis inhibitor metyrapone, EPharmacol. 119 (1985) 143–152.

22] J.-Y. Koh, C.W. Cotman, Programmed cell death: its possiblein calcium channel antagonist neurotoxicity, Brain Res. 587 (1233–240.

23] J.-Y. Koh, D.W. Choi, Quantitative determination of glutamatediated cortical neural injury in cell culture by lactate dehydrogeefflux assay, J. Neurosci. Methods 20 (1987) 83–90.

24] T. Koide, T. Wieloch, B. Siesjo, Chronic dexamethasone pretreataggravates ischemic neuronal necrosis, J. Cereb. Blood Flow M6 (1996) 395–403.

25] H.J. Krugers, R.H.A. Kemper, J. Korf, G.J. Ter Horst, S. KnolleMetyrapone reduces rat brain damage and seizures after HI: anindependent of modulation of plasma corticosterone levels? J. CBlood Flow Metab. 18 (1998) 386–390.

26] H.J. Krugers, S. Knollema, R.H.A. Kemper, G.J. Ter Horst, J. KDown-regulation of the hypothalamo–pituitary–adrenal axis redbrain damage and number of seizures following hypoxia/ischain rats, Brain Res. 690 (1995) 41–47.

27] H.J. Krugers, S. Maslam, J. Korf, M. Joels, The corticosteronsynthesis inhibitor metyrapone prevents hypoxia/ischemia-indloss of synaptic function in the rat hippocampus, Stroke 31 (21162–1172.

28] J. Lachuer, I. Delton, M. Buda, M. Tappaz, The habituation of brstem catecholaminergic groups to chronic daily restraint strestress specific like that of the hypothalamo–pituitary–adrenalBrain Res. 638 (1994) 196–202.

450 F.U. Fontella et al. / Brain Research Bulletin 65 (2005) 443–450

[29] M.S. Lawrence, R. Sapolsky, Glucocorticoids accelerate ATP lossfollowing metabolic insults in cultured hippocampal neurons, BrainRes. 646 (1994) 303–306.

[30] J.-M. Lee, M.C. Grabb, G.J. Zipfel, D.W. Choi, Brain tissue re-sponses to ischemia, J. Clin. Invest. 106 (2000) 723–731.

[31] J. Li, J. Zhang, Inhibition of apoptosis by ginsenoside Rg1 in cul-tured cortical neurons, Chin. Med. J. (Engl) 110 (1997) 535–539.

[32] M. Lowy, L. Gault, B. Yamamato, Adrenalectomy attenuates stress-induced elevation in extracellular glutamate concentration in hip-pocampus, J. Neurosci. 61 (1993) 1957–1960.

[33] J.L. Madrigal, J.R. Caso, J. de Cristobal, A. Cardenas, J.C. Leza, I.Lizasoain, P. Lorenzo, M.A. Moro, Effect of subacute and chronicimmobilisation stress on the outcome of permanent focal cerebralischaemia in rats, Brain Res. 979 (2003) 137–145.

[34] O. Martı, A. Gavalda, T. Jolın, A. Armario, Effect of regularity ofexposure to chronic immobilization stress on the circadian pattern ofpituitary adrenal hormones, growth hormone, and thyroid stimulatinghormone in the adult male rat, Psychoneuroendocrinology 18 (1993)67–77.

[35] M.P. Mattson, W. Duan, W.A. Peterson, C. Culmsee, Neurodegen-erative disorders and ischemic brain diseases, Apoptosis 6 (2001)69–81.

[36] B.S. McEwen, R.M. Sapolsky, Stress and cognitive function, Curr.Opin. Neurobiol. 5 (1995) 205–216.

[37] B.S. McEwen, Stress and hippocampal plasticity, Annu. Rev. Neu-rosci. 22 (1999) 105–122.

[38] B.S. McEwen, Effects of adverse experiences for brain structure andfunction, Biol. Psychiatry. 48 (2000) 721–731.

[39] B.S. McEwen, Plasticity of the hippocampus: adaptation to chronicstress and allostatic load, Ann. N. Y. Acad. Sci. 933 (2001) 265–277.

[40] B.S. McEwen, E.R. De Kloet, W. Tostene, Adrenal steroid recep-986)

[ levelship-1657.

[ M.L.n of

ation

[ ageear-

[ mic

[ 00Bcad.

[ rineev. 22

[ uronal001)

[ rebral971

[ wry997)

[ re-t pat-

[51] A. Rami, A. Rabie, J. Winckler, Synergy between chronic corti-costerone treatment and cerebral ischemia in producing damage innoncalbindinergic neurons, Exp. Neurol. 149 (1998) 439–446.

[52] R.M. Sapolsky, A mechanism for glucocorticoid toxicity in the hip-pocampus: increased neural vulnerability to metabolic insults, J.Neurosci. 5 (1985) 1228–1232.

[53] R.M. Sapolsky, Glucocorticoids, stress, and their adverse neurologi-cal effects: relevance to aging, Exp. Gerontol. 34 (1999) 721–732.

[54] R.M. Sapolsky, The possibility of neurotoxicity in the hippocampusin major depression: a primer on neuron death, Biol. Psychiatry 48(2000) 755–765.

[55] R.M. Sapolsky, W.A. Pulsinelli, Glucocorticoids potentiate ischemicinjury to neurons: therapeutic implications, Science 229 (1985)1397–1401.

[56] R.M. Sapolsky, H. Uno, C.S. Rebert, C.E. Finch, Hippocampal dam-age associated with prolonged glucocorticoids exposure in primates,J. Neurosci. 10 (1990) 2897–2902.

[57] S.W. Scheff, L.S. Bernardo, C.W. Cotman, Hydrocortisone adminis-tration retards axon sprouting in the rat dentate gyrus, Exp. Neurol.69 (1980) 195–201.

[58] V.L. Smith-Swintoski, L.C. Pettigrew, R.M. Sapolsky, C. Phares,S.D. Craddock, S.M. Brooke, M.P. Mattson, Metyrapone, an inhibitorof glucocorticoid production, reduces brain injury induced by focaland global ischemia and seizures, J. Cereb. Blood Flow Metab. 16(1996) 585–598.

[59] B.A. Stein-Behrens, W.J. Lin, R.M. Sapolsky, Physiological eleva-tions of glucorticoids potentiate glutamate accumulation in the hip-pocampus, J. Neurochem. 63 (1994) 596–602.

[60] U. Strasser, G. Fischer, Quantitative measurement of neuronal de-generation in organotypic hippocampal cultures after combined oxy-gen/glucose deprivation, J. Neurosci. Methods 57 (1995) 177–186.

[ .C.mice,

[ denceglia

[ matedium-

[ oxia-Res.

[ ski,ated

l and11–

[ L.V.s in

by

[ toryeely

[ eu-and

urol.

[ M.ed toe 30

tors and actions in the nervous system, Physiol. Rev. 66 (11121–1188.

41] B. Moghaddam, Stress preferentially increases extraneuronalof excitatory amino acids in the prefrontal cortex: comparison topocampus and basal ganglia, J. Neurochem. 60 (1993) 1650–

42] M.A. Moro, J. De Alba, J.C. Leza, P. Lorenzo, A.P. Fernandez,Bentura, L. Bosca, J. Rodrigo, I. Lizasoain, Neuronal expressioinducible nitric oxide synthase after oxygen and glucose deprivin rat forebrain slices, Eur. J. Neurosci. 10 (1998) 445–456.

43] J.K. Morse, J.N. Davis, Regulation of ischemic hippocampal damin the gerbil: adrenalectomy alters the rate of CA1 cell disappance, Exp. Neurol. 110 (1990) 86–92.

44] G.C. Newman, F.E. Hospod, P. Wu, Glucose utilization of ischehippocampal slices, J. Neurosci. Methods 28 (1989) 23–34.

45] H. Nishiyama, T. Knopfel, S. Endo, S. Itohara, Glial protein S1modulates long-term neuronal synaptic plasticity, Proc. Natl. ASci. U.S.A. 99 (2002) 4037–4042.

46] K. Pacak, M. Palkovits, Stressor specificity of central neuroendocresponses: implications for stress-related disorders, Endocr. R(2001) 502–548.

47] R.S. Payne, A. Schurr, Corticosterone-aggravated ischemic nedamage in vitro is relieved by vanadate, Neuroreport 12 (21261–1263.

48] R.S. Payne, M.T. Tseng, A. Schurr, The glucose paradox of ceischemia: evidence for corticosterone involvement, Brain Res.(2003) 9–17.

49] G.L. Peterson, A simplification of the protein assay method of Loet al. which is more generally applicable, Anal. Biochem. 83 (1346–356.

50] A. Petzold, G. Keir, D. Lim, M. Smith, E.J. Thompson, Cebrospinal fluid (CSF) and serum S100B: release and wash-outern, Brain Res. Bull. 61 (2003) 281–285.

61] N. Sugo, P.D. Hurn, M.B. Morahan, K. Hattori, R.J. Traystman, ADeVries, Social stress exacerbates focal cerebral ischemia inStroke 33 (2002) 1660–1664.

62] S. Takahashi, T. Abe, J. Gotoh, Y. Fukuuchi, Substrate-depenof reduction of MTT: a tetrazolium dye differs in cultured astroand neurons, Neurochem. Int. 40 (2002) 441–448.

63] C.P. Taylor, S.P. Burke, M.L. Weber, Hippocampal slices: glutaoverflow and cellular damage from ischemia are reduced by sochannel blockade, J. Neurosci. Methods 59 (1995) 121–128.

64] G.C. Tombaugh, R. Sapolsky, Corticosterone accelerates hypinduced ATP loss in cultured hippocampal astrocytes, Brain588 (1992) 154–158.

65] I.L. Torres, G.D. Gamaro, S.N. Silveira-Cucco, M.B. MichalowJ.B. Correa, M.L. Perry, C. Dalmaz, Effect of acute and reperestraint stress on glucose oxidation to CO2 in hippocampacerebral cortex slices, Braz. J. Med. Biol. Res. 34 (2001) 1116.

66] F. Tramontina, J. Karl, C. Gottfried, A. Mendez, D. Goncalves,Portela, C.A. Goncalves, Digitonin-permeabilization of astrocyteculture monitored by trypan blue exclusion and loss of S100BELISA, Brain Res. Protoc. 6 (2000) 86–90.

67] C. Venero, J. Borrell, Rapid glucocorticoid effects on excitaamino acid levels in the hippocampus: a microdialysis study in frmoving rats, Eur. J. Neurosci. 11 (1999) 2465–2473.

68] B.C. White, J.M. Sullivan, D.J. DeGracia, B.J. O’Neil, R.W. Nmar, L.I. Grossman, J.A. Rafols, G.S. Krause, Brain ischemiareperfusion: molecular mechanisms of neuronal injury, J. NeSci. 179 (2000) 1–33.

69] M.T. Wunderlich, A.D. Ebert, T. Kratz, M. Goertler, S. Jost,Herrmann, Early neurobehavioral outcome after stroke is relatrelease of neurobiochemical markers of brain damage, Strok(1999) 1190–1195.