Differential effects of exercise intensities in hippocampal BDNF, inflammatory cytokines and cell proliferation in rats during the postnatal brain development

ContentslistsavailableatScienceDirect

Neuroscience

Letters

j o ur na l h o me p a g e :w w w . e l s e v i e r . c o m / l o c a t e / n e u l e t

Differential

effects

of

exercise

intensities

in

hippocampal

BDNF,

inflammatory

cytokines

and

cell

proliferation

in

rats

during

the

postnatal

brain

development

Alexandre

Aparecido

de

Almeida

_,

_Sérgio

_Gomes

_da

_Silva

,

Jansen

Fernandes

,

Luiz

Fernando

Peixinho-Pena

_,

_Fulvio

_Alexandre

_Scorza

_,

Esper

Abrão

Cavalheiro

_,

_Ricardo

_Mario

_Arida

a_{DepartmentofPhysiology,UniversidadeFederaldeSãoPaulo(UNIFESP),SãoPaulo,SP,Brazil}

b_{InstitutodoCérebro(INCE),InstitutoIsraelitadeEnsinoePesquisaAlbertEinstein(IIEPAE),SãoPaulo,SP,Brazil} c_{DepartmentofNeurologyandNeurosurgery,UniversidadeFederaldeSãoPaulo(UNIFESP),SãoPaulo,SP,Brazil}

h

i

g

h

l

i

g

h

t

s

•BDNFlevelsandcellularproliferationinthehippocampusaredependentsofexerciseintensity.

•Itwasnotedthatexerciseintensityisaninflammation-inducingfactor.

•Exercise-inducedinflammatoryresponseisalsorelatedtodevelopmentalstage.

a

r

t

i

c

l

e

i

n

f

o

Articlehistory: Received25May2013

Receivedinrevisedform7August2013 Accepted8August2013

Keywords: Exercise Intensity Development Inflammation Brain Plasticity

a

b

s

t

r

a

c

t

Ithasbeenestablishedthatlowintensitiesofexerciseproducebeneficialeffectsforthebrain,whilehigh intensitiescancausesomeneuronaldamage(e.g.exacerbatedinflammatoryresponseandcelldeath). Althoughtheseeffectsaredocumentedinthematurebrain,theinfluenceofexerciseintensitiesinthe developingbrainhasbeenpoorlyexplored.Toinvestigatetheimpactofexerciseintensityin develop-ingrats,weevaluatedthehippocampallevelofbrainderivedneurotrophicfactor(BDNF),inflammatory cytokines(TNF␣,IL6andIL10)andtheoccurrenceofhippocampalcelldegenerationandproliferation

atdifferentstagesofpostnatalbraindevelopmentofratssubmittedtotwophysicalexerciseintensities. Tothispoint,maleratsweredividedintodifferentagegroups:P21,P31,P41andP51.Eachagegroup wassubmittedtotwoexerciseintensities(lowandhigh)onatreadmillover10consecutivedays,except thecontrolrats.Weverifiedthatthedensityofproliferatingcellswassignificantlyhigherinthedentate gyrusofratssubmittedtolow-intensityexercisefromP21toP30comparedwithhigh-intensityexercise andcontrolrats.Asignificantincreaseofproliferativecelldensitywasfoundinratssubmittedto high-intensityexercisefromP31toP40whencomparedtolow-intensityexerciseandcontrolrats.Elevated hippocampallevelsofIL6weredetectedinratssubmittedtohigh-intensityexercisefromP21toP30 comparedtocontrolrats.FromP41toP50period,higherlevelsofBDNF,TNF␣andIL10werefoundin

thehippocampalformationofratssubmittedtohigh-intensityexerciseinrelationtotheircontrolrats. Ourdatashowthatexercise-inducedneuroplasticeffectsonBDNFlevelsandcellularproliferationinthe hippocampalregionaredependentonexerciseintensityanddevelopmentalperiod.Thus,exercise inten-sityisaninflammation-inducingfactorandexercise-inducedinflammatoryresponseduringthepostnatal braindevelopmentisalsorelatedtodevelopmentalstage.Ourfindingsindicatethatneuroplasticchanges inducedbyexerciseindevelopingratsdependonbothageandtrainingintensity.

Crown Copyright © 2013 Published by Elsevier Ireland Ltd. All rights reserved.

∗Correspondingauthorsat:DepartamentodeFisiologia,UniversidadeFederal deSãoPaulo(UNIFESP),RuaBotucatu862,Ed.CiênciasBiomédicas,5◦_andar.Vila Clementino,CEP:04023-900,SãoPaulo–SP,Brazil.Tel.:+551155764513; fax:+551155739304.

E-mailaddresses:sergio.gomes@unifesp.br,sgomesilva@hotmail.com(S.Gomes daSilva),arida.nexp@epm.br(R.M.Arida).

1. Introduction

Brain development is a complex process characterizedby a seriesofcritical stages.Eachstagemust beproperly fulfilledin orderforthebraintoconfiguretheirnormalstructure.Although thebrain’sstructureisformedbeforebirth,itsfulldevelopment dependsonpostnatalstimulifromtheenvironment.Stimuliduring

0304-3940/$–seefrontmatter.Crown Copyright © 2013 Published by Elsevier Ireland Ltd. All rights reserved.

2 553 (2013) 1–6

thispostnatal period maymodulatethebrain’s functional mat-uration and determine its lifelong integrity [1]. Evidence from humanstudieshasshownthatsomeenvironmentalstimulisuch as physical activity habits in infancy and adolescence seem to haveafavorableinfluence[3,11,22].Forinstance,apositive cor-relationbetweenphysical activityandlearning andintelligence scoreshasbeenobservedinschool-agechildren[22].In preado-lescentchildren,singleorvariousexercisesessionsmaymodify neuronalactivity,improvetheperformanceinattentiontasks,and enhancereading comprehension and response accuracyin aca-demicachievementtests [3,11].Data from animalstudies have shownthatexercisecanmodulatethefunctionalmaturationofthe developingbrainbymeansofneuroplasticprocesses[8,24].Ina studyconductedbyourgroup,weverifiedthatanexerciseprogram undertakenduringadolescentperiodofratswasabletoincrease axonaldensityofgranulecells ofthedentategyrus,toenhance hippocampalexpressionofneurotrophicfactors,andtoimprove learningand memory[8].Together, theseinteresting data sug-gestthatphysicalexerciseduringpostnataldevelopmentresults inpositivechangesforthebrain,particularlyinthehippocampal formation–ahighlyplasticregionofthebrainlinkedtocognitive andemotionalprocesses.However,itisimportanttopointoutthat exercise-inducedneuroplasticeffectsmayvaryaccordingtoanimal age[14]andtrainingintensity[15].

It is usually accepted that low and moderate intensities of exerciseproducebeneficialeffectsforthebrain,whilehigh inten-sitiescancausesomedamage (e.g.celldeath). Indeed,harmful effectscouldappearinundue conditionsofphysicalor psycho-logicalstress.Ofthefewstudiesconductedwithphysicaltraining injuvenilerats,itwasdemonstratedthatexercise-induced neuro-plasticitywasdependentuponexerciseintensity[15].Thelow-,but notthehigh-,intensityexerciseresultedinenhancedneurogenesis andgeneexpressionofbrainderivedneurotrophicfactor(BDNF),

N-methyl-d-aspartatereceptor type1 (NMDAR1)andfetal liver kinase-1(Flk-1)inthehippocampalformationof35-day-oldrats [15].Nevertheless,itisunclearwhetherthereisadevelopmental stageinwhichexerciseintensitycouldinduceagreater neuroplas-ticinfluence.Moreover,ithasbeendescribedthatexercisecanalter thebrainexpressionofinflammatorycytokines[4,5],small cell-signalingproteinsknowntoaffecttheintegrityoftheblood–brain barrierandinducecelldeathduringdevelopment[6,10].Inviewof theseobservations,weevaluatedthehippocampallevelsofBDNF andinflammatorycytokinesandtheoccurrenceofhippocampal celldegenerationandproliferationatdifferentstagesof postna-talbraindevelopment(prepuberte:P21–P30;juvenile:P31–P40; adolescent:P41–P50;lateadolescent:P51–P60)ofratssubmitted totwophysicalexerciseintensities.

2. Methods

2.1. Exerciseparadigm

Wistarratswereusedinthisstudy.Allexperimentalprotocols describedbelowwereapprovedbytheethicscommitteeofthe Uni-versidadeFederaldeSãoPaulo(#0607/09).Thecolonyroomwas maintainedat21±2◦_{Cwitha12hlight/darkschedule,andad} libi-tumfoodandwaterthroughouttheexperiments.Maleratswere dividedintodifferentagegroups:P21,P31,P41andP51(n=44in eachgroup).Eachagegroupwassubmittedtotwoexercise inten-sities(lowandhigh)onatreadmill(Columbusinstruments)over 10consecutivedays,exceptthecontrolrats.Exerciseintensities weredeterminedinaccordancewiththedevelopmentstageofeach group:theratsranfromP21toP30atlow-intensityof12m/min andhigh-intensityof16m/min,fromP31toP40atlow-intensity of14m/minandhigh-intensityof18m/min,fromP41toP50at

low-intensity of 16m/min and high-intensity of 20m/min, and fromP51toP60atlow-intensityof18m/minandhigh-intensity of22m/min.Ratssubmittedtolow-andhigh-intensityexercise ranonatreadmillfor 30minperday.Exercise sessionsstarted witha3minwarm-upat8m/min,andelectricshockswereused sparinglytomotivatetheratstorun[7].Wechosethetreadmill runningbecausetheintensityanddurationofexercisecaneasily becontrolled,differentiallyofthevoluntarywheelrunning.

2.2. Methodsofproteinimmunodetection

2.2.1. Tissuepreparation

The hippocampal formations of twenty-seven animals from each agegroup(n=9in thecontrol,low-intensity exerciseand high-intensity exercise subgroups) were removed immediately afterdecapitationandhomogenizedin0.01MTris–hydrochloride (pH7.6)containing5.8%ofsodiumchloride,10%ofglycerol,1% ofNonidetP40(NP-40),0.4%ofethylenediaminetetraaceticacid (EDTA) and protease inhibitors. Animals submitted to physical trainingwerekilled1hafterthelastexercisesession.Sampleswere sonicatedandstoredat−80◦_C.

2.2.2. Enzyme-linkedimmunosorbentassay(ELISA)

HippocampalBDNFlevelswereexaminedbyELISAkitE-max® (Promega).Samplespreviouslystoredat−80◦_{Cwerecentrifuged} for5minat14,000rpmat4◦_{Candthesupernatanttransferredto} a96-wellplate(CorningCostar)coatedwithanti-BDNF(1:1000) thenincubatedfor2hatroomtemperature.Afterthisperiod,the platewaswashedwithTris-bufferedsalineTween-20(TBS-T)and incubatedwiththefollowingantibodies:anti-human(1:500)for 2h,andconjugateanti-IgYHRP(1:200)for1h.Then,color reac-tionwithtetramethylbenzidinewasquantifiedinaplatereaderat 450nm(QuickElisa).

2.2.3. Immunoassay

Hippocampal levels of inflammatory cytokines were deter-minedbymeansofimmunoassay.Samplespreviouslystoredat

−80◦_{Cwerecentrifugedfor5minat10,000×gat4}◦_Candcytokine concentrationsmeasuredwithaMilliporemultiplexRatCytokine KitontheLuminex®_xMAP®_{platform.Thelevelsoftumornecrosis} factoralpha(TNF␣),interleukin6(IL6)andinterleukin10(IL10)

weredetermined.

2.3. Histologicalmethods

2.3.1. Tissuepreparation

Seventeen animals from each age group (n=5 for control subgroup, n=6 for low-intensity exercise subgroup and n=6 for high-intensity exercisesubgroup)were deeply anesthetized (Tionembutal, 50mgkg, i.p.) and perfused transcardially with solutionof0.01Mphosphate-bufferedsaline(PBS), followedby solutioncontaining4%formaldehydein0.1Mphosphate-buffered, pH7.4.Animalssubmittedtophysicaltrainingwerekilled1hafter thelastexercisesession.Afterperfusion,theanimals’brainswere removedfromtheskull,cutcoronallywithavibratome(Leica)in 50␮m-thickslicesandstoredat−20◦_{Cinthebiologicaltissuebank} inourlaboratory(forpreservationoftissue).Toinhibitthe forma-tionoficecrystalsthatdamagethestructureofcells,thesliceswere maintainedinanantifreezesolutioncontaining30%ofsucrose,1% ofpolyvinylpyrrolidone40(PVP-40)and30%ofethyleneglycolin PB(pH7.2).

2.3.2. Ki-67immunohistochemistry

553 (2013) 1–6 3

storedinthetissuebankwereselectedinordertoanalyzethe cel-lularproliferationrateinthehippocampalregionofthedentate gyrus(bregma−2.8/−3.6mm)[19].Fiveslicesperanimalwere pre-treatedwith3%ofH2O2for10mintoblockendogenousperoxidase activity,rinsedinPBS,pre-incubatedfor45mininPBScontaining 10%of normal serumand 0.2%of TritonX-100,and then incu-batedinprimaryantibodyagainsttheKi-67proliferationmarker (1:1000;Abcam)at4◦_{Covernight.Pairedslicesofeachsubgroup} wereprocessedinthesamevialinordertominimizethedifferences duringtheimmunohistochemicalprocedure.Afterasequenceof procedures(similartothatdescribedbyGomesdaSilvaetal.[8]), thesliceswerefinallywashedinPBS,mountedongelatin-coated slides,dehydrated,coverslippedwithEntellan(Merk).Afterwards, thehippocampalregionofthedentategyrusofeachanimalwas digitizedwithabright-fieldmicroscope(NikonEclipse6600)for quantitativeanalysis.ToanalyzetheKi-67staining,weuseda sim-ilarmethodtothatdescribedbyScorzaetal.[21]andGomesda Silvaetal.[8].Briefly,thesignificantpixelswerethenconverted intobinarymatrix(blackandwhite)andquantifiedbytheblack pixelssumperarea(i.e.densityofKi-67staining).The quantifi-cationof pixelswascarriedout bysoftwarethatallows matrix manipulations(Matlab)andinimageswiththesameresolution.

2.3.3. Fluoro-JadeB

A sequence of five hippocampal slices per animal (bregma

−2.8/−3.3mm) [19] previously stored in the tissue bank was selectedtoobservetheimmunohistochemicalstainingof Fluoro-Jade B (FJB), a derivative of fluorescein anionic tribasic that selectivelylabelsdegeneratingneurons[20].ToperformtheFJB histochemicalstaining,weusedasimilarmethodtothatdescribed byGomesdaSilva etal.[8].Asapositive control,weaddedto theimmunohistochemicalprocedureaslice ofthehippocampal region(fromanotherproject)ofananimalinjectedwith350mg/kg of pilocarpine(a potent cholinergic agonistthat induces status epilepticus and leads to severe widespread cell loss in several brain areas). Then, hippocampal slices were rinsed in distilled water,mountedonslides,coverslippedandanalyzedqualitatively inconfocalmicroscope(NikonEclipse6600)bytwoindependent investigators.

2.4. Statisticalanalyses

StatisticalanalyseswerecarriedoutusingaSPSSsoftware ver-sion17.0 (SPSS Inc.,Chicago,IL).The Shapiro–Wilkand Levene testswereusedtoverifydatanormalityandhomogeneity, respec-tively.Datawithnormaldistributionorhomogenyvarianceswere comparedby two-wayANOVAfollowedbyBonferroni posthoc test.Differenceswereconsideredsignificantwhenp<0.05.Data withnon-normaldistributionwerecomparedbynonparametric Kruskal–WallistestfollowedbyMann–Whitneytestsand Bonfer-ronicorrectionwhenp<0.05.Resultsarepresentedasmeanand standarderrorofthemean(±SEM).

3. Results

3.1. HippocampalBDNFlevels

We investigatedthehippocampalBDNFlevelsin ratsat dif-ferent stages of postnatal braindevelopmentsubmitted totwo physicalexerciseintensities.Ratssubmittedtohigh-intensity exer-cise from P41 to P50 presented a significant increase of the hippocampalBDNFlevels(35.51±3.76pg/ml)whencomparedto low-intensityexercise(24.05±1.38pg/ml;p=0.003)andcontrol rats(26.57±2.75pg/ml;p=0.026)(Fig.1).Nosignificantdifference wasfoundinotheragegroups(p>0.05).

3.2. Hippocampalcytokinelevels

A significant increase of the hippocampal levels of pro-inflammatory cytokine IL6 was detected in rats submitted to high-intensityexercisefromP21toP30(36.11±3.79pg/ml) com-paredtocontrolrats(5.84±2.10pg/ml,p=0.001)(Fig.1).From P41toP50period,elevatedlevelsofpro-inflammatorycytokine TNF␣ were found in the hippocampal formation of rats

sub-mitted to high-intensity exercise (TNF␣=15.19±5.50pg/ml) in relationtotheircontrolrats(TNF␣=2.58±0.39pg/ml,p=0.014). Elevatedhippocampallevelsofanti-inflammatorycytokineIL10 werealso detectedin ratssubmittedto high-intensity exercise fromP41toP50(1309.68±459.13pg/ml)comparedtotheir con-trols(56.72±41.38pg/ml,p=0.009)(Fig.1).Inotheragegroups,no significantdifferencewasfoundinthehippocampalinflammatory cytokinelevels(p>0.017;p-valueaftercorrectionofBonferroni).

3.3. Cellularproliferationanddegenerationinthedentategyrus

TheKi-67markerwasusedtoanalyzethecellularproliferation rateinthehippocampalformationofdevelopingratssubmittedto twoexerciseintensities.ProliferativecellsmarkedwithKi-67were observedinsubgranularzoneandhilarregionofthedentategyrus inallagegroups(Fig.2A).However,quantitativeanalysisrevealed thatthedensityofproliferatingcells wassignificantlyhigherin thedentategyrusofratssubmittedtolow-intensityexercisefrom P21toP30comparedwithhigh-intensityexerciseandcontrolrats (p<0.001forboth)(Fig.2B).FromP31toP40,asignificantincrease ofproliferativecelldensitywasfoundinratssubmittedto high-intensityexercisewhencomparedtolow-intensityexerciseand controlrats(p<0.001).Inotheragegroups,nosignificantdifference inthedensityofproliferatingcellswasdetectedamongcontrol, low-andhigh-intensityexerciserats(p>0.05)(Fig.2B).Toverify thepossibilityofcellulardegeneration,weusedFJBhistochemical staininginthedentategyrusofratssubmittedtotwophysical exer-ciseintensitiesatdifferentstagesofpostnatalbraindevelopment. Inthepositivecontrol,theFJBstainingwasobservedinneurons locatedinthehilarregionofthedentategyrus(Fig.3).However, nodegeneratingneuronwasfoundinthedentategyrusofcontrol ratsorthosesubmittedtotwoexerciseintensitiesduringpostnatal braindevelopment(Fig.3).

4. Discussion

Ourstudyexaminedtheinfluenceofexerciseintensityat differ-entstagesofpostnatalbraindevelopmentinrats.Wefoundthat exercise-inducedneuroplasticeffectsonBDNFlevelsandcellular proliferationinthehippocampalregionaredependentsof exer-ciseintensityanddevelopmentalperiod.Moreover,itwasnoted thatexerciseintensityisaninflammation-inducingfactorandthat exercise-induced inflammatory response duringpostnatal brain developmentisalsorelatedtodevelopmentalstage.

4 553 (2013) 1–6

Fig.1.HippocampallevelsofBDNFandanti-(IL10)andpro-inflammatory(IL6andTNF-␣)cytokinesinratssubmittedtotwoexerciseintensities(lowandhigh)atdifferent stagesofpostnatalbraindevelopment.*p<0.05comparedtocontrolgroup;#p<0.05comparedtorespectiveexercisegroup.BDNFstatisticalanalysiswasperformedby two-wayANOVAfollowedbyBonferroni’sposthoctest.CytokinesstatisticalanalyseswereconductedbyKruskal–WallistestfollowedbyMann–WhitneytestsandBonferroni correction(0.05/3=0.017).

couldalsoincreasethehippocampalBDNFatdifferentstagesof postnatalbraindevelopment.However,theresultsdidnotshow thiseffect.Moreover,elevatedlevelsof inflammatorycytokines werealso foundin the hippocampalformation of rats submit-ted to high-intensity exercise from P41 to P50. Inflammatory cytokinescanexert dualeffects (positive or negative) onbrain functions.Theseeffectsaredependentonnumerousfactors,such asthetypeof cytokineproduced, thefunctionalstateand type ofstimulatedcells, theconcentrationandthedurationof expo-suretothecytokines.The over-expressionof pro-inflammatory cytokinesin thebrain may contribute to unsuccessful mainte-nance of neuronal communication and brain functions. It has

beenobservedthatpro-inflammatorycytokinesenhanceactivity ofstress-activatedproteinkinasesandimpairthebrain’sabilityto maintainhippocampallong-termpotentiation(LTP)[18],a physi-ologicalprocessinvolvedinmemoryconsolidation.Inopposition, it was demonstrated that inflammation-linked LTP impairment canbereversed byintracerebroventricularinfusionof the anti-inflammatorycytokineIL10[16].Althoughoftheseinterestingdata revealanantagoniceffectoftheanti-inflammatorycytokineson deleteriouseventsproduced bypro-inflammatorycytokines,we cannotensurethatourresultsinratssubmittedtohigh-intensity exercise from P41 to P50 are linked to beneficial neuroplastic effects.

553 (2013) 1–6 5

Fig.3.FJBhistochemicalstaininginthedentategyrusofdevelopingratssubmittedtotwoexerciseintensities.FJBstaininginpositivecontrolwasobservedinneurons (fluorescentgreen)locatedinthepyramidalcelllayerofCA1(A)andCA3(B)andinthehilusofthedentategyrus(C).Nodegeneratingneuronwasfoundinthehippocampal regionofdevelopingratssubmittedtolow-andhigh-intensityofexercise.Toensurethereliabilityofresults,somesamplesreceivedagreatertimeexposurethantheother samples.Scalebar=150␮m.

As mentioned above, over-expression of inflammatory cytokinescouldresult in negativeand positive changes forthe developingbrain.Inourstudy,high-intensityexerciseincreased both pro- and anti-inflammatory cytokines (TNF␣ and IL10,

respectively)in thehippocampalformation ofratsexercised in theP41toP50period.Nevertheless,nohippocampaldegenerating neuron wasfoundin these animals.A possibleexplanation for theseresultscouldbeattributedatleasttotheexercise-induced BDNF.Indeed,apromisinginteractionbetweengrowthfactorsand cytokineshasbeendescribed[17,23].It hasbeenreportedthat BDNFadministrationstimulatesIL10 secretion[17]and reduces theinflammation-inducedbrainlesionsextension[23].Takinginto

accounttheinterplaybetweengrowthfactorsandcytokines,itis possiblethattheincreaseinIL10levelsinourstudycouldalsobe relatedtotheincreaseofBDNFlevelsdetectedinthehippocampal formationofratssubmittedtohigh-intensityexercisefromP41to P50.Alteredneuronalactivityandgenicexpressionmightbeother factorsinvolvedinthisprocess.

6 553 (2013) 1–6

supporttheidea that beneficial effects in thedeveloping brain dependuponlowintensitiesofexercise[15].Ontheotherhand, thehigh-,butnotthelow-,intensityexerciseparadigmresulted inasignificantincreaseofproliferativecelldensityinratstrained fromP31toP40.Thesefindingsrevealthatnewcellformationmay varyaccordingtoexerciseintensityanddevelopmentalstageofthe brain.Itisimportanttopointoutthatnewcellformationinduced byexerciseintheseearlystagescouldhaveasignificantimpact onbrainstructureand functionaldevelopment.Studiesinadult animalshaveshownthatphysicalexerciseenhancescell prolifera-tionandsurvivalinthedentategyrusandincreasesthemagnitude ofhippocampalLTPand improvesspatiallearning andmemory [14,25]. Considering that new cell formation in the hippocam-palregionismostprevalentinearlylife[14],wespeculatethat exercise-inducedcellularproliferationintheseanimalsmayalso beaccompaniedbyimprovedcognitivecapability.Nevertheless, furtherstudiesareneededtoestablishthisrelationshipbetween exerciseintensity,cellproliferationandcognitivefunctionsduring postnatalbraindevelopment.

Inconclusion,thepresentstudyhasshownthattwo intensi-tiesofexercisemayresultinpositiveandnegativechangesforthe developingbrain.Dependingonthedevelopmentalstage,low-or high-intensityofexercisecanenhancethecellproliferationratein thedentategyrusandincreaseinflammatorycytokinesand neu-rotrophicfactorlevelsinthehippocampalformationofdeveloping rats.These findings indicate that exercise-induced neuroplastic effectsindevelopingratsdependonbothageandtrainingintensity.

Conflictofinterest

Theauthorsdeclarethattheyhavenoconflictsofinterest.

Acknowledgements

ThisstudywassupportedbyCAPES,CNPq,FAPESPandINNT (Brazil).

References

[1]S.L.Andersen,Trajectoriesofbraindevelopment:pointofvulnerabilityor win-dowofopportunity?Neurosci.Biobehav.Rev.27(2003)3–18.

[2]P.Bernd,Theroleofneurotrophinsduringearlydevelopment,GeneExpr.14 (2008)241–250.

[3]S.M.Buck,C.H.Hillman,D.M.Castelli,Therelationofaerobicfitnesstostroop taskperformanceinpreadolescentchildren,Med.Sci.SportsExerc.40(2008) 166–172.

[4]M.Chennaoui,C.Drogou,D.Gomez-Merino,Effectsofphysicaltrainingon IL-1beta,IL-6andIL-1raconcentrationsinvariousbrainareasoftherat,Eur. CytokineNetw.19(2008)8–14.

[5]L.H.Colbert,J.M.Davis,D.A.Essig,A.Ghaffar,E.P.Mayer,Tissueexpression andplasmaconcentrationsofTNFalpha,IL-1beta,andIL-6followingtreadmill exerciseinmice,Int.J.SportsMed.22(2001)261–267.

[6]B.E.Deverman,P.H.Patterson,Cytokines,CNSdevelopment,Neuron64(2009) 61–78.

[7]S.GomesdaSilva,F.Dona,M.J.daSilvaFernandes,F.A.Scorza,E.A. Caval-heiro,R.M.Arida,Physicalexerciseduringtheadolescentperiodoflifeincreases hippocampalparvalbuminexpression,BrainDev.32(2010)137–142. [8]S.GomesdaSilva,N.Unsain,D.H.Masco,M.Toscano-Silva,H.A.deAmorim,B.H.

SilvaAraujo,P.S.Simoes,G.Naffah-MazzacorattiMda,R.A.Mortara,F.A.Scorza, E.A.Cavalheiro,R.M.Arida,Earlyexercisepromotespositivehippocampal plas-ticityandimprovesspatialmemoryintheadultlifeofrats,Hippocampus22 (2012)347–358.

[9]J.A.Gorski,S.A.Balogh,J.M.Wehner,K.R.Jones,Learningdeficitsin forebrain-restrictedbrain-derivedneurotrophicfactormutantmice,Neuroscience121 (2003)341–354.

[10]H.Hagberg,C.Mallard,Effectofinflammationoncentralnervoussystem devel-opmentandvulnerability,Curr.Opin.Neurol.18(2005)117–123.

[11]C.H.Hillman,M.B.Pontifex,L.B.Raine,D.M.Castelli,E.E.Hall,A.F.Kramer,The effectofacutetreadmillwalkingoncognitivecontrolandacademic achieve-mentinpreadolescentchildren,Neuroscience159(2009)1044–1054. [12]M.E.Hopkins,R.Nitecki,D.J.Bucci,Physicalexerciseduringadolescenceversus

adulthood:differentialeffectsonobjectrecognitionmemoryandbrain-derived neurotrophicfactorlevels,Neuroscience194(2011)84–94.

[13]E.J.Huang,L.F.Reichardt,Neurotrophins:rolesinneuronaldevelopmentand function,Annu.Rev.Neurosci.24(2001)677–736.

[14]Y.P.Kim,H.Kim,M.S.Shin,H.K.Chang,M.H.Jang,M.C.Shin,S.J.Lee,H.H.Lee, J.H.Yoon,I.G.Jeong,C.J.Kim,Age-dependenceoftheeffectoftreadmillexercise oncellproliferationinthedentategyrusofrats,Neurosci.Lett.355(2004) 152–154.

[15]S.J.Lou,J.Y.Liu,H.Chang,P.J.Chen,Hippocampalneurogenesisandgene expres-siondependonexerciseintensityinjuvenilerats,BrainRes.1210(2008) 48–55.

[16]A.M. Lynch,C. Walsh, A. Delaney,Y. Nolan, V.A. Campbell,M.A. Lynch, Lipopolysaccharide-inducedincreaseinsignallinginhippocampusis abro-gatedbyIL-10–aroleforIL-1beta?J.Neurochem.88(2004)635–646. [17]O.Noga,M.Peiser,M.Altenahr,H.Knieling,R.Wanner,G.Hanf,R.Grosse,

N.Suttorp,Differentialactivationofdendriticcells bynervegrowth fac-tor and brain-derived neurotrophic factor, Clin. Exp. Allergy 37 (2007) 1701–1708.

[18]E.O’Donnell,E.Vereker,M.A.Lynch,Age-relatedimpairmentinLTPis accom-paniedbyenhancedactivityofstress-activatedproteinkinases:analysisof underlyingmechanisms,Eur.J.Neurosci.12(2000)345–352.

[19]G.Paxinos,C.Watson,TheRatBraininStereotaxicCoordinate,AcademicPress, NewYork,1996.

[20]L.C.Schmued,K.J.Hopkins,Fluoro-JadeB:ahighaffinityfluorescentmarkerfor thelocalizationofneuronaldegeneration,BrainRes.874(2000)123–130. [21]C.A. Scorza,B.H.Araujo,R.M.Arida,F.A.Scorza,L.B.Torres, H.A.Amorim,

E.A.Cavalheiro,DistinctivehippocampalCA2subfieldoftheAmazonrodent Proechimys,Neuroscience169(2010)965–973.

[22]B.A.Sibley,J.L.Etnier,Therelationshipbetweenphysicalactivityandcognition inchildren:ameta-analysis,Pediatr.Exerc.Sci.15(2003)243–256. [23]M.Spedding,P.Gressens,Neurotrophinsandcytokinesinneuronalplasticity,

NovartisFound.Symp.289(2008)222–233,discussion233–240.

[24]N.Uysal,K.Tugyan,B.M.Kayatekin,O.Acikgoz,H.A.Bagriyanik,S.Gonenc, D.Ozdemir,I.Aksu,A.Topcu,I.Semin,Theeffectsofregularaerobicexercise inadolescentperiodonhippocampalneurondensity,apoptosisandspatial memory,Neurosci.Lett.383(2005)241–245.