Testing the effects of ethinylestradiol and of an environmentally relevant mixture of xenoestrogens as found in the Douro River (Portugal) on the maturation of fish gonads-A stereological study using the zebrafish (Danio rerio) as model

(1)

ContentslistsavailableatSciVerseScienceDirect

Aquatic

Toxicology

j o ur na l ho me p ag e :w w w . e l s e v i e r . c o m / l o c a t e / a q u a t o x

Testing

the

effects

of

ethinylestradiol

and

of

an

environmentally

relevant

mixture

of

xenoestrogens

as

found

in

the

Douro

River

(Portugal)

on

the

maturation

of

ﬁsh

gonads—A

stereological

study

using

the

zebraﬁsh

(Danio

rerio)

as

model

Paula

Silva

a,b

,

Maria

J. Rocha

b,c

,

Catarina

Cruzeiro

b

_,

_Fernanda

_Malhão

a

_,

_Bruno

_Reis

b

_,

_Ralph

_Urbatzka

b

_,

Rogério

A.F.

Monteiro

a,b

, Eduardo

Rocha

a,b,∗

a_Laboratory_of_Histology_and_Embryology,_Institute_of_Biomedical_Sciences_Abel_Salazar_(ICBAS),_University_of_Porto_(U.Porto),_Portugal

b_Laboratory_of_Cellular,_Molecular_and_Analytical_Studies,_{Interdisciplinary}_Centre_of_Marine_and_{Environmental}_Research_(CIIMAR),_CIMAR_Associated_Laboratory_(CIMAR_LA,

UniversityofPorto(U.Porto),Portugal

c_Superior_Institute_of_Health_{Sciences-North}_(ISCS-N),_CESPU,_Gandra,_Paredes,_Portugal

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received7February2012 Receivedinrevisedform1July2012 Accepted4July2012 Keywords: Zebraﬁsh Estrogens Testis Ovary Stereology Endocrinedisruptors

a

b

s

t

r

a

c

t

Innaturalenvironmentsfishpopulationsareexposedtomanypotentialxenoestrogens,whereby under-standingtheimpactsofmixturescontinuetobeofgreatinterest.Themainobjectiveofthisstudywas, therefore,tounderstandwhetherandhowanenvironmentallyrelevantmixtureofxenoestrogensfound intheDouroRiverestuarycandisruptthenormalgametogenesisinfish.Forthispurpose,adultzebrafish ofbothsexeswereexposedfor21daystoanenvironmentalmixture(MIX)of11xenoestrogensfrom diversesources.A100ng/Lethinylestradiol(EE2)positivecontrolwasadded.Aquantitative (stereolog-ical)analysiswithsystematicsamplingwasmadeinthegonads,andusinglightmicroscopyboththe relativeandtheabsolutevolumesofthegametogenicstageswereestimated.DatapointthattheEE2 stimulusinducedchangesinstructuralcompartments;withdecreasingtrendsfortheadvanced mat-urationstagesbothinmalesandfemales.Therewasalsoatrendforagreateramountofinterstitial tissueinmales.Alongwithaninterstitialfibrosisincreasedetected,thepresenceofaproteinaceousfluid wasobservedinbothsexesandexperimentalgroups(EE2andMIX).Otherhistopathologicalterations wereobservedintheEE2femalegroup,suchasthepresenceoffociofgranulomatousinflammationand follicularmineralizationinthegerminalparenchymaandluminalareas.Themostinterestingfinding ofthisstudywasthattheexposuretotheMIXcausedadecreaseoftherelativevolumeof spermato-zoainzebrafish.Thiskindofestrogeniceffecthasnotearlierbeenstructurallyquantifiedinsuchafine detailwithunbiasedstereologyinfishgonads.Despitetheultimateconsequencesofsuchdisruptions beingunknown,itcouldbelogicallyarguedthatreductionorslowing-downoftheappearanceofthe mostmaturecohortsand/oreventualinterstitialfibrosisandotherpathologicchangescanadversely affectbreeding.Thefindingsaddfurtherexplanatorybasesforunderstandingthenegativeimpactsof xenoestrogens.

1. Introduction

Overthepasttwentyyears,considerableandgrowingattention hasbeengiventothefactthatchemicalcompoundscanbe biolog-icallyactivebydisruptingthenormalfunctionsoftheendocrine systemsofanimals.Thosesubstances,calledendocrinedisrupting chemicals(EDCs),arewidespreadthroughouttheaquatic ecosys-temsandtheireffectsandmodesofactionhavebeendocumented

∗ Correspondingauthorat:LaboratoryofHistologyandEmbryology,ICBAS– Insti-tuteofBiomedicalSciencesAbelSalazar,RuadeJorgeViterboFerreira228,4050-313 Porto,Portugal.Tel.:+351220428000.

E-mailaddress:erocha@icbas.up.pt(E.Rocha).

(DepledgeandBillinghurst,1999;Porteetal.,2006;Segneretal., 2003). Aquatic species are exposed to EDCs through a variety of sources,but wastewater isthe mainsourceof EDCssuchas thosethatmimictheeffectsofnaturalestrogens(xenoestrogens) (Falconeretal.,2006;Kolpinetal.,2002;Metcalfeetal.,2001;Ying etal.,2002).

Underlaboratoryconditions,xenoestrogenshavebeenreported tocauseavarietyofeffectsinﬁshreproductionsuchaschanges in plasma hormone concentrations (Khan and Thomas, 1998), gonadalsize(Ashﬁeldetal.,1998;GrayandMetcalfe,1997;Jobling etal.,1996),developmentofintersexgonads(ovotestis)(Grayand Metcalfe,1997)andinductionofthefemaleyolkprecursor pro-tein,vitellogenin(VTG)(Joblingetal.,1996).Althoughtherehas beenanincreasedefforttounderstandthemechanismsbywhich

(2)

xenoestrogensexertfinaleffects,fundamentaldoubtsstillremain, suchastheimpactonthekineticsofgametogenesis.Several stud-iessuggestthatexposuretoxenoestrogensmaycausesuppression ofgametogenesisinboth maleand femalefish(Blázquez etal., 1998;Christiansenetal.,1998;Drèzeetal.,2000;Gimenoetal., 1998;Papouliasetal.,2000;PiferrerandDonaldson,1992;Tanaka andGrizzle,2002;VanDenBeltetal.,2002;Weberetal.,2003). Allthosehistologicalinvestigationsinvolvedatissuemicroscopic analysis,whichresultedinvisualqualitativedescriptionsand/or semi-quantitativedataorinquantificationsbytwo-dimensional measurements,whicharetobeavoidedastheyarewell-known nowtosufferfromuncontrolledbiases(HowardandReed,2005). Theapplicationofstereologytosuchissues,however,canresult inunbiasedquantitativedatasets,andthuscouldbeafarmore adequateapproachforinvestigatingtheimpactofxenoestrogen exposure on gametogenesis kinetics. Most research on effects ofxenoestrogensonfish,doneunderlaboratoryconditions,has focusedprimarilyonexposuretosinglechemicals,despitethefact thattheaquaticenvironmentreceivesinfluxesofdifferent chemi-cals.Thiswiderangeofxenoestrogensintheaquaticenvironment highlightstheimportanceofimprovingourunderstandingofthe combinatoryeffectsofthechemicals,atleastinmodelorganisms. Theexistenceofinteractiveeffectsimpliesthattheestrogeniceffect ofamixturemaysomehowdeviatefromwhatwouldbeexpected ofeachsingleagentactingonitsown.

Inviewoftheabove,wewantedtostudywhetherand how oneenvironmentallyrelevantmixtureofxenoestrogensasfound intheDouroRiver(Rochaetal.,2011)coulddisruptthenormal cel-lularkineticsofbothspermatogenesisandoogenesisinfish.Asa positivecontrolweexposedfishto17␣-ethinylestradiol(EE2)ata highconcentration(100ng/L)topromotetheappearanceofmore markedeffectsthatcouldserveasreference.Anotherstudywith thesameexperimentaldesign,andtestingboththesameMIXand EE2,disclosedchangesinsexsteroidogenesisatthemolecularlevel andadifferentbehavioroftheMIXandEE2onspecific transcrip-tionfactorsinthezebrafishbrain(Urbatzkaetal.,2012).Herein,we specificallytestedthehypothesisthatsuchexposurescouldevoke volumechangesindifferentgermcells,detectableonlywith ste-reologicaltools.Thezebrafish(Daniorerio,Hamilton)waschosen becauseithasbeenrecommendedforassessingeffectsoftoxicants ondevelopmentandreproduction(Andersenetal.,2000;Segner etal.,2003).

2. Materialandmethods

2.1. Testcompoundsandtheirnominalandmeasured concentrations

StocksolutionsofEE2(Sigma–Aldrich)andtheDouro’s environ-mentalmixture(MIX)werepreparedinethanolatthebeginning oftheexposure. Thexenoestrogensthat composedthemixture andtheirchosennominalconcentrationswerebasedonwhatwas foundbychemical screeningin theDouroRiverestuary(Rocha et al.,2011)and are shown on Table1. Thisalso providesthe measuredconcentrationsintheaquaria(analyticaldetailsbelow). Measuredconcentrationsweresimilartothenominalones,and variationswereminimal.TheestrogenicpotencyoftheMIXwas estimatedtobe20.73ng/LEE2-equivalents(Urbatzkaetal.,2012). AsthefocushereinwastheMIX,theestrogenicpotencyofeach chemicalwasnotevaluated.Also,itiswellknownthattheydiffer inpotencies(Lawsetal.,2000;GutendorfandWestendorf,2001; Preussetal.,2006).

During the trials, thestock solutions were diluted, daily, in water,andusedforrenewalofthetestsolutionsintheaquaria. Ethanol(p.a.grade)wasusedasvehicle,anditsﬁnalconcentration

Table1

XenoestrogensintheDouro’senvironmentalmixture(MIX)usedintheexposure assay.Measuredconcentrationisapooledvalue(±standarddeviation)between times0h(justafterwaterrenewal)and24h(justbeforethenextwaterrenewal).

Xenoestrogen Nominal Concentration (ng/L) Measured Concentration (ng/L) Estrone(E1) 2.8 2.8±0.0 17␤-Estradiol(E2) 20.0 20.0±0.0 17␣-Ethinylestradiol(EE2) 4.3 4.2±0.2 4-t-Octylphenol(4-t-OP) 17.9 16.9±1.5 4-n-Octylphenol(4-n-OP) 6.0 6.0±0.0 BisphenolA(BPA) 25.7 25.7±0.0

4-Octylphenolmonoethoxylate(OP1EO) 10.5 10,0±0.7 4-Octylphenoldiethoxylate(OP2EO) 16.0 16.0±0.0 4-Nonylphenolmonoethoxylate(NP1EO) 21.9 21.9±0.0 4-Nonylphenoldiethoxylate(NP2EO) 1608.0 1608.0±0.0

4-n-Nonylphenol(4-n-NP) 3.5 3.1±0.6

intheaquaria,inbothtreatmentandcontrols,was≈0.001%.This

wastheminimumamountnecessarytodissolvethechemicalsof

themixture.Thenominalusedconcentrationwas10timeslower

thanthemaximumsolventconcentrationrecommendedbyOECD

(100␮l/L).Additionally,theconcentrationusedwashalfthe

con-servativesafe-guardlimitof20␮l/LproposedbyHutchinsonetal.

(2006),whenreviewingacuteandchroniceffectsofcarriersolvents inaquaticanimals.

Watersamples(1L)wereweeklycollectedafterrenewingthe testsolutionsandjustbeforethenextrenewalfromeachofthe experimentalaquaria(seebelowSection2.3).Thesesampleswere thenaddedwithinternaldeuteratedsurrogates(E2-d2 and

BPA-d16)andsubmittedtosolidphaseextraction(SPE)(Rochaetal.,

2010).Brieﬂy,thewatersampleswereloadedontoSPEOasisHLBTM

(hydrophilic–lipophilicbalance)cartridgesataconstantflowrate of 5mL/min followed by a washing step of 10mLof ultrapure waterandmethanol(9:1).Cartridgesweredriedundervacuumfor 60minandthenelutedwith10mLofethylacetate,at1mL/min. Theresulting extractswereevaporated todryness in a heating blockat40◦Cunderagentlestreamofnitrogenandreconstituted in500␮Lofanhydrousmethanol.Duetothelowvolatilityofthe majorityofthepresentcompounds,50␮Lofeachextracted frac-tionweretransferredintoGCvialsandevaporatedtodrynessunder agentlenitrogenstream.Fifty␮Lofpyridinewereaddedtothe dryresidueswhichwerederivatizedbytheadditionof50␮Lof N,O-bis(trimethylsilyl)trifluoroacetamide(BSTFA)added with1% trimethylchlorosilane(1%TMCS).Thevialsweremixedusinga vor-texsystemandheated,inaheatingblock,for30minat65◦C.Then eachsamplewasanalyzedinagaschromatograph(TraceGCultra, ThermoFinniganElectronCorporation)coupledwithaniontrap massspectrometer(Thermo Scientific ITQTM ₁₁₀₀ _GC–MSn_), _an

autosampler(ThermoScientiﬁcTriPlusTM₎_and_a_TR5MS_capillary

column(30m×0.25mmi.d.,0.25␮mﬁlmthickness).Helium car-riergas(99.999%purity)wasmaintainedataconstantﬂowrateof 1.0mL/minandthecolumnoventemperatureswereprogrammed accordingtoRochaetal.(2010).Finally,thequantitativeanalysis wasmadeinaselectedionmonitoringmode(SIM)usingexternal calibration.Afterrenewingthetestsolutionsthemeasured concen-trationsmatchedthenominalones.Immediatelybeforethenext renewaltheactualconcentrationsaveraged95% ofthenominal ones,withnodetectablevariationsinmostcompounds(Table1). 2.2. Experimentalanimals

Adultzebraﬁshwereobtainedfromalocalspecializeddealer. FishwerereceivedattheInterdisciplinaryCentreforMarineand Environmental Research(Porto) bioterium and werekeptfor 4 weeks,separatedbysex,ina600Ltankwithdechlorinatedwater.

(3)

Therewasnotspecificspawningsubtract.Aftertheabove speci-fiedperiod,theanimalswereallowedtoacclimatetoexperimental conditionsfor15days.Forthatpurposeandalsotoinduce inter-fishsynchronization,homogeneousgroupsofzebrafish(16males and16 females –toprovidea sexratioof 1:1) wererandomly distributedamong6glasstanks(25Leach),atatemperatureof 27±1◦C.Fishwereexposedtoa14-hlight/10-hdark photope-riod.Thewateroxygenconcentrationwasnearairsaturationlevel. Beforetheexperiment,fishwerefedadlibitum,twiceaday,with dryflakefood(Seravipan®₎_and_once_with_artemia_nauplii_(Artemia

salina,OceanNutrition). 2.3. Experimentaldesign

Fish wereexposed toone of the following treatments (two replicatespercondition):solventcontrol(C),17␣-ethinylestradiol positivecontrol(EE2)(nominalandmeasuredconcentrationsof 100ng/L) and Douro’s environmentalmixture(MIX). The expo-suretookplaceintheglassaquariaundersemi-staticconditions withonedailytotalrenewalofthewater,forrestorationofthe nominalconcentrations.Tominimizestress,waterchangeswere gentlymade by slowlysiphoning,temperature wasalwaysthe same,andtheanimalswereexposedtolowwaterlevelsonlyfor shortperiods(intheorderofseconds).Duringtheexperiment,the fishwerefedtwotimesadaywiththesameflakedryfoodused duringtheacclimatizationperiod.MeasurementsofpH,dissolved oxygen,temperature,ammoniaandnitritelevelswerecarriedout daily.After21daysofexposure,eachfishwasdeeplyanesthetized byimmersioninabathof2-phenoxyethanolsolution(0.5mL/L) (Sigma–Aldrich)andimmediatelysacrificedbycuttingthespinal cord.Forthisstudy,6fishpersexandpertreatmentwereanalyzed (3perreplica).Thesefishwereunbiasedlyselectedbya system-aticsamplingscheme.Theremainingfishweresampledforother studies.Itisimportanttopointoutthatwithinthescopeofthis study,nodifferenceswerefoundbetweenreplicatedtanks.After beingweighedandmeasured(totallength)thefishweredissected andtheirgonadswereweighed.Thegonadosomaticindices(GSI) werecalculatedasfollows:

GSI= (weight of the gonads in mg)

(weight of the total body in mg)×100. 2.4. Histologicalpreparation

Gonadsfixedincommercial4%bufferedformalin(J.T.Baker) wereprocessedforhistologicalexaminationbylightmicroscopy. Afterfixation(48h),thetissuesweredehydratedthroughaseriesof gradedethanol(AGA)solutions(70–99.8%),clearedinxylene(BDH –Prolabo),andimpregnatedinparaffin(Merck)inasemi-enclosed benchtoptissueprocessor(LeicaTP1020).Finally,thetissueswere embeddedinparaffininamodulartissueembeddingcenter(Leica EG1140C).Eachgonadwasentirelysectioned(Microtome–Leica RM2155)intothinsections(4␮mthick).Thesectionswere sys-tematicallysampledduringsectioningsothatatleastaminimum of10sectionsperanimalwereanalyzed.For improvingsection adhesion,the sampledsectionswere mounted in slides coated withaminopropyltriethoxysilane(APES)(Sigma–Aldrich),andthen stainedwithhaematoxylin-eosinbeforebeingcoversliped. 2.5. Histologicalandstereologicalanalysis

For estimating the relative volume densities of the gonadal structures,astereologicalapproachwasmadeusingan interac-tiveimageanalysisworkstation(CAST-Grid,Olympus,Version1.6), workingwithalive-imagecapturedbyacolorCCD-videocamera (Sony).Thelightmicroscope(BX50,Olympus)wasequippedwitha

Fig.1. Differentgermcellcystsobservedduringthespermatogenicprocessin zebraﬁsh.Ser– Sertolicell;S– Spermatogonia;ES– Earlyspermatocytes;LS– Late spermatocytes;SPs–spermatidsandSZ–Spermatozoa.

fully-motorizedstage(PriorScientiﬁc),thusallowingmeander sys-tematicsamplingwithan(x–yaxis)accuracyof1␮m.Themeander samplingwascontrolledbythesoftware.

Thevolumedensities (laterexpressedintheresultsas%)of thegonadalstructures[VV(structure,gonad)]wereestimatedbyan

unbiased,stereologicaltechniquebasedonpoint-countingbythe observeroperatingtheCAST-Grid(HowardandReed,2005): VV(structure,gonad)=

P(structure) R

P(gonad)

where

P(structure)and

P(gonad)correspondtothetotal num-berofgridpointsinasectionhittingthestructureofinterestor itsreferencespace(gonad), respectively,andRistheinverseof thepoint-ratiousedforeachparticularVVestimation.Inmalesa

1:4twolatticepoint-systemgridwasoverlaidonthelive video-imagedisplayedona17monitor.Forthetestisandspermatozoa the1:1pointratiowasused.Thetighter1:4pointratiowasused foralltheotherstructuralcompartments.Infemalesa1:9two lat-ticepointsystemgridwasused.The1:9pointratiowasusedfor allthestructureswiththeexceptionofmatureoocytestage. Point-countingwasmadeinsystematicallysampledfieldsworkingeither withthe100×(males)or10×(females)objectivelens,givingthe respectivefinalworkingmagnificationoftheliveimageinthe mon-itorof4050×and399×,respectively.Afterpilottrials,thestepwise stagex–ymovementwassetaccordingtothegonadsizeinorder tocountanoptimalnumberofcellspersection;theadoptedsteps variedfrom200to300␮minmales,andfrom700to950␮min females.

Histologicalalterationsinthetestiswereevaluatedbasedonthe presenceofdifferentstagesofgermcells,ofinterstitialtissueandof Sertolicells.Fivetesticularpopulationswereidentiﬁedasfollows (Fig.1):Spermatogonia(thelargestofthespermatogeniccells; aci-dophiliccytoplasmwithrelativelylargernucleuscontainingpoorly condensedchromatinandoneortwocompactnucleoli);Earlyand latespermatocytes(earlyspermatocytesarelargerthanlate sper-matocyteswiththread-likeorcondensedchromatin,respectively, withrelativelysmallercytoplasmwithnoafﬁnityfordye); Sper-matids(densenucleiandnarrowrimsofeosinophiliccytoplasm) andSpermatozoa(dark,roundnucleiandminimalornoapparent cytoplasm).Thelatterarethesmallestspermatogeniccellsandthey existasscatteredindividualcellswithinthelumenofthe anasto-mosingseminiferoustubules.

(4)

Fig.2.Differentgermcelltypesobservedduringtheoogenesisinzebraﬁsh.A–Oogonium;B–Primarygrowthstage;C–Corticalalveolusstage;D–Vitellogenicstage;E– Maturestage.

Fivefollicularstagesofmaturationwereidentifiedandcounted fromtheovaries(Fig.2):Oogonium(highrationucleus/cytoplasm; acidophilicnuclearmembraneandnucleolus;cytoplasmweakly acidophilic);Primarygrowthstage(enlargedbasophiliccytoplasm; pale nucleus with many nucleoli distributed along the nuclear membrane);Corticalalveolusstage(presenceofthecortical alve-oliinthecytoplasm,firstintheperipheralzoneandafterinall cytoplasm;irregularnucleishapewithnucleoliattachedtoinner borderofnuclearmembrane);Vitellogenicstage(cytoplasmwith yolk bodies which areheterogeneous in appearance;migration ofthe nucleus totheanimal pole; zona radiataprominent and granulosa-thecacellseasilyidentified);Maturestage(oocytefilled withyolkbodies;smallareaoccupiedbythecytoplasm;oocyte layersfoldedirregularly).Atretic(degenerated)folliclesand inter-stitialtissuewerealsocounted.

Forestimatingtotalvolumes,weassumedinthisstudythatthe gonadaldensitywas≈1mg/mm3_both_for_testis_and_ovaries,_so_that

thegonadvolume[V(gonad)](mm3₎_was_deemed_herein_as_equal

tothegonadweight(GW).Theabsolutevolumes(mm3₎_of_all_the

gonadalstructuresconsideredabovewerethusestimatedbasedon theformula:

V(structure)=V(gonad)VV(structure,gonad)

Bothtestisandovarysectionswereexaminedunderlight micro-scopetoinvestigatethepresenceofpathologicalmodiﬁcationsin thegonads.Thediagnosisofsuchchangeswasdoneinaccordance withtheOECDguidelines(OECD,2009).

2.6. Statistics

DatawereanalyzedusingthesoftwareStatistica(version10, StatSoft).Thesignificancelevel wassetat˛=0.05. Allvariables were checked for normality and homogeneity of variance by usingtheKolmogorov–SmirnovandtheLevenetests,respectively. Dataweresubmittedtoone-wayanalysisofvariance (ANOVA). AfterasignificantANOVA,pairsofmeanswerecomparedbythe Newman–Keulstest. Inthespecificcases wheretheparametric ANOVAassumptionsfailed,dataweresubmittedtoKruskal–Wallis test and to Mann–Whitney U non-parametric test. The non-parametric results were alike those of the parametric ANOVA analysis,strengthening theconfidence inthesignificanceofthe differences.

3. Results

3.1. Biometricparameters

Attheendofexperimentthebodyweightofzebrafishmales andfemalesexposedtoEE2wassignificantlylowerthanthatin theothertwogroups(Table2).Nosignificanteffects(p>0.05)on

totallengthwereobservedafter21daysofexposuretoany treat-mentineithersex.Thetestisandovaryweightfollowedthesame trendasbodyweight,asalowervaluewasfoundinfishexposed toEE2withnodifferencesbetweenthecontrolandtheMIXgroup observed(Table2).Asignificantreduction(p<0.05)inGSIwasalso observedinbothzebrafishmalesandfemalesofthepositive con-trolgroupwhencomparedtothecontrolone.Femalesexposed tothexenoestrogenmixture,however,presentedthehighestGSI (Table2).

3.2. Histologicalandstereologicalanalysis

Inaccordancewiththegonadweights,theabsolutevolumesof thegonadsfollowedthesametrendasthebodyweight:adecrease inthetestisandovaryvolumeswasonlyobservedinmalesand femalesexposedtoEE2,respectively(Tables 3and4). Histolog-icalexamination of gonadsections fromcontrol and treatment groupsrevealednormaltesticularandovarianorganizational archi-tectureandprogression ofgametogenesis.Leydigcells couldbe identifiedwithinthestroma,althoughtheirpositiveidentification wasnotconsistent.Thegermcelltypesselectedfor quantifica-tionwereaveragedandrepresentedasarelative volumeofthe wholetestis(Fig.3)and ovary (Fig. 4).The stereologicalstudy revealeda reduction in theabsolutevolumesofthegerm cells in both maleand female gonadsof fishfromthe positive con-trolgroup(Tables3and4).Theanalysisoftherelativevolumes oftestiscellsrevealedmaturationsuppressionbothwiththeEE2 andthemixtureofxenoestrogens(Fig.3).Thenumberof sperma-tozoadecreasedwiththetwotreatments,thelowervaluesbeing observedinthepositivecontrolgroup.Thenumberofspermatids decreasedonlyinmalesexposedtoEE2(Fig.3).Boththerelative andtheabsolutevolumesofSertolicellsweresimilarinallgroups Table2

Bodyweight(BW),totallength(TL),gonadweight(GW)andgonadosomaticindex (GSI)ofzebraﬁshexposedtoxenoestrogensfor21days(solventcontrol–C;positive control– EE2andDouro’senvironmentalmixture–MIX).

C EE2 MIX Males BW(g) 0.406±0.059a _0.317_±_0.033b _0.427_±_0.063a TL(cm) 3.522±0.209 3.585±0.143 3.569±0.204 GW(g) 0.006±0.001a _0.003_±_0.001b _0.006_±_0.002a GSI 1.448±0.166a _0.938_±_0.300b _1.410_±_0.453a FEMALES BW(g) 0.535±0.032a _0.343_±_0.096b _0.497_±_0.083a TL(cm) 3.588±0.118 3.586±0.153 3.682±0.161 GW(g) 0.047±0.006a _0.010_±_0.004b _0.058_±_0.017a GSI 8.842±0.874a _2.887_±_1.120b _11.346_±_2.043c Valuesarepresentedasmean±standarddeviation(SD),n=6/treatmentgroup. Meanswithin a rowwithouta commonsuperscriptletterdiffersigniﬁcantly (p<0.05).Absenceofsuperscriptindicatesnosigniﬁcantdifferencebetween treat-ments.

(5)

Table3

Absolutevolumes(mm3₎_of_the_germ_cell_present_in_zebraﬁsh_testis_exposed_to xenoestrogensfor21days(solventcontrol–C;positivecontrol–EE2andDouro’s environmentalmixture–MIX).

C EE2 MIX Testis 5.75±1.49a _2.93_±_0.83b _5.98_±_1.86a Spermatogonia 0.20±0.08a _0.09_±_0.08b _0.22_±_0.10a,b Earlyspermatocyte 0.85±0.36a _0.44_±_0.18b _1.06_±_0.39a Latespermatocyte 0.59±0.16a _0.30_±_0.09b _0.64_±_0.32a Spermatid 0.27±0.11a _0.09_±_0.06b _0.27_±_0.08a Spermatozoa 2.28±0.64a _0.45_±_0.15b _1.93_±_0.52a Lumen 1.09±0.33 0.68±0.27 1.34±0.64 Sertolicells 0.03±0.02 0.02±0.02 0.03±0.02 Interstitialtissue 0.44±0.11a _0.87_±_0.35b _0.49_±_0.22a Valuesarepresentedasmean±standarddeviation(SD),n=6/treatmentgroup. Meanswithin a rowwithouta commonsuperscriptletterdiffersigniﬁcantly (p<0.05).Absenceofsuperscriptindicatesnosigniﬁcantdifferencebetween treat-ments.

(Table3andFig.3).ExposuretoEE2alsocausedsuppressionof ovarianmaturation,inducingovariesthatweremainlycomposed offolliclesattheearlieststagesofdevelopment(primarygrowth) and withlower volumesof vitellogenic and of maturefollicles (Fig.4).Thexenoestrogenmixturehadnostatisticallysigniﬁcant effectonfemalegametogenesis(Fig.4).

Theparametertermed“interstitialtissue”includesthe connec-tivetissue, bloodvesselsandallhistologicalterationsobserved, suchasthepresenceoffibrosis,inflammationtissueand proteina-ceousfluid.Astothetestis,Leydigcellswerealsoincludedinthat structuralcompartment.Inmales,theinterstitialtissueabsolute volumewashigherintheEE2group(Table3).Thecorresponding relativevolumewashigherbothinmalesandinfemalesexposed toEE2(Figs.3and4).Theinterstitialfibrosiswasthemajor con-tributortothisincreaseinbothsexes.Besidesfibrosis,thepresence ofinterstitialproteinaceousfluidwastheotherpathological alter-ationobservedinthetestisof50%ofthemales(Fig.5)andinthe ovaryofallfemales(Fig.6)exposedtoEE2.Aloweredamountof proteinaceousfluidwasalsoobservedin50%ofthefemalesand

Table4

Absolutevolumes(mm3₎_of_the_germ_cell_present_in_zebraﬁsh_ovary_exposed_to

xenoestrogensfor21days(solventcontrol–C;positivecontrol–EE2andDouro’s environmentalmixture–MIX).

C EE2 MIX Ovary 47.35±6.18a _9.92_±_4.15b _57.53_±_17.43a Oogonium 0.08±0.06 0.10±0.18 0.12±0.11 Primarygrowth 7.63±2.06 5.29±1.76 8.44±2.42 Corticalalveolus 5.77±1.99a _0.76_±_0.93b _4.51_±_1.67a Vitellogenic 13.86±4.39a _0.26_±_0.55b _13.42_±_8.10a Mature 11.14±5.38a _0.03_±_0.06b _15.60_±_7.39a Atretic 6.19±3.67a _0.34_±_0.45b _13.25_±_6.19a Interstitialtissue 2.68±1.61 3.14±2.33 2.21±0.60 Valuesarepresentedasmean±standarddeviation(SD),n=6/treatmentgroup. Means withina rowwithouta commonsuperscriptletterdiffersigniﬁcantly (p<0.05).Absenceofsuperscriptindicatesnosigniﬁcantdifferencebetween treat-ments.

50%ofthemalesoftheMIXexperimentalgroup.Intheovaryof

allfemales exposedtoEE2,epithelioidmacrophagegranulomas

wereobserved(Fig.6).In3ofthe6ovariesanalyzed,

presump-tivelymineralizedstructureswerefoundnearthosegranulomatous inﬂammation areas (Fig. 6). Thosemineralized structures were commonlyassociatedwithatreticoocytesunderresorption,and forthisreasonwereincludedinthatcategory(Fig.4).However, theabsoluteandrelativevolumesoftheatreticfollicleswerelower infemalesexposedtoEE2whencomparedwithcontrolandMIX groups(Table4andFig.4).Nopathologicalalterationswerefound ingonadsfromthecontrolgroup.

4. Discussion

4.1. Noveltyofthestudy

To our knowledge, this is the ﬁrst quantitative histological investigationtoevaluatetheeffectsof anenvironmentally rele-vantandcomplexmixtureofxenoestrogensontherelativeand

Fig.3. Relativevolumesofeachdevelopmentalcysttype,includingspermatogonia,earlyandlatespermatocytes,spermatids,spermatozoa,aswellasSertolicellsand interstitialtissueinzebrafishmalestreatedwithaxenoestrogenmixture(MIX)andwith17-ethinylestradiol(EE2)(6specimenspertreatmentgroup).Valuesarepresented asmean+standarddeviation(SD).Meanswithinacellstagewithoutacommonsuperscriptletterdiffersignificantly(p<0.05).Absenceofsuperscriptindicatesnosignificant differencebetweentreatments.

(6)

Fig.4.Relativevolumesofeachdevelopmentaloocytestage,includingoogonium,primarygrowth,corticalalveolus,vitellogenicandmatureoocytestages,aswellasatretic oocytesandinterstitialtissueinzebrafishfemalestreatedwithaxenoestrogenmixture(MIX)andwith17-ethinylestradiol(EE2)(6specimenspertreatmentgroup).Values arepresentedasmean+standarddeviation(SD).Meanswithinacellstagewithoutacommonsuperscriptletterdiffersignificantly(p<0.05).Absenceofsuperscriptindicates nosignificantdifferencebetweentreatments.

absolutevolumesofgermcells in ﬁshgonads.Moststudieson the effects of xenoestrogens on gonad histology have focused onsingle chemicals,regardless of thefact that ecosystems are impactedbylowtohighconcentrationsofdifferentchemicals.This isthecaseoftheDouroRiverestuarywhereRochaetal.(2011) founda mixture of both natural and synthetic steroidal estro-gensandnon-steroidalsyntheticestrogeniccompounds.Undera sub-acuteexposure,weevaluateddetailedeffectsofamixtureof

Fig.5. Pathologicalalterationsobservedintestisofzebraﬁshtreatedwith 17-ethinylestradiol(EE2)(arrows–proteinaceousﬂuid).

xenoestrogens,asobservedinthecitedstudy,inzebraﬁsh game-togenesis.Weappliedastereologicalapproachneverperformedin previousstudiesthathavetackledsucheffects.

4.2. Effectsonspermatogenesis

Despiteourhistologicalexaminationshowedon-going game-togenesis,onceallgermcelltypeswerepresentingonadsofboth controlandexposedanimals,thefactisthatsignificantdisruptions wererevealed.Onemajorfindingwasthatexposuretothe xenoe-strogenmixturecauseddisruptionofgametogenesisinmalesonly. Stereologyrevealedadecreaseinrelativevolumeofspermatozoa infishexposedtotheMIXandadditionallytoEE2,withthelowest valuenoticedinthelattergroup.Theabsolutevolumeof sperma-tozoawas,however,thesameintheMIXandinthecontrolgroup. Theseresultshighlighttheimportanceofcarryingoutacomplete (multi-level)stereologicalanalysisinthistypeofstudies,sincethe examinationoftheabsolutevolumescouldnottranslatethefine gametogenesisdynamics.Theanalysisoftheabsolutevolumes esti-matedforthetestisalsorevealedthatexposuretoEE2inhibited testiculargrowth.Thereductionofthetesticularvolumeofthose fishwascausedbyadecreaseofthetotalvolumeofalltypesofgerm cells.Themechanism(s)behindtheinhibitionoftesticulargrowth byestrogenicchemicalsis(are)notyetclear.Theycouldeitheract directlyonthetestis,inducingthelocalsuppressionoftesticular androgen,orinterfereatoneormorelevelsinthehormonalcascade thatregulatethematuration,ultimatelyinhibitingthe spermatoge-nesis(Joblingetal.,1996).Anotherpossibleexplanationisaneffect ontheSertolicells.Adirect(orindirect)effectofestrogenic com-poundsonSertolicellsisplausiblesinceitwasfoundthatthese cellsexpressestrogenreceptors(Legleretal.,2000;Andreassen etal.,2003).AsevereeffectonthecytologyofSertolicells dur-ingspermatogenesiswasfoundineelpout(Zoarcesviviparous)after

(7)

Fig.6.Pathologicalalterationsobservedinzebrafishfemalestreatedwitha 17-ethinylestradiol(EE2).Theovaryismainlycomposedbyfolliclesintheearlystages ofdevelopment.Intheinterstitialtissueispossibletoobservewithinfibrotictissue theexistenceofapinkproteinaceousfluid(arrows),granulomatousinflammation areas(GI)andnearbythesesuggestivelymineralizedstructures(asterisk).

exogenousadministrationofE2and4-nonylphenol(Christiansen etal.,1998).Intheeel,Anguillaanguilla,EE2inhibitedthe devel-opmentorcauseddegenerationofSertolicells,thusindicatingthat estrogensmayaffectthesomaticcomponentsofthegonadﬁrstand thegermcellslater(ColomboandGrandi,1995).Asinourwork theSertolicellvolumesweresimilarinallgroups,wecannot sug-gestthatthosecellswerestronglyimpacted.However,functional changescannotbeexcludedandadisruptionoftheSertolicells couldwellimpactongermcellmaturation;thiseventwouldhelp toexplainthereducedspermproduction notedintheMIXand EE2groups.Regardlessofthemechanism(s)involved,itis obvi-ousthattheMIXfoundinDouroRiverhadnegativeeffectsonthe testisstructureofzebraﬁsh,resultingindisturbance(decrease)of spermatogenesis.

WhencomparedtoEE2theMIXhadsimilarconsequenceson thetestis,buttoalowerextent,indicatingthattheMIXeffectsmay haveoccurredviaestrogenicsignaling.Theestrogeniceffectsofthe MIXonzebraﬁshmaleswerealsodemonstratedinaparallelstudy conductedbyourgroup,whichshowedthattheexposuretoEE2 andMIXwasphysiologicallyactiveandinducedvitellogeninmRNA inbothtreatmentgroupsandinasimilarrangedespiteofthelower

estrogenic potencyof theMIX(Urbatzkaetal., 2012).Actually, zebraﬁshishighlyresponsivetovitellogeninmRNAinductionand thiseffectwasobservedinmalesexposedto9.5ng/LofEE2,amuch lowerconcentrationthantheoneusedinourstudy(Langeetal., 2012).Somescientistspointedout,however,thatbiomarkersof estrogenexposurecouldbeevokedbysome“environmental estro-gens”withantiandrogenicactivity(Sohoniand Sumpter,1998). SuchactivitywasreportedfortwocompoundsoftheMIXused herein,namelybisphenolA(BPA)(Parisetal.,2002;Sohoniand Sumpter,1998;Sunetal.,2006)and4-ter-octylphenol(Parisetal., 2002).Suchactions,however,wereobtainedwithinvitroandgene expressionassaysand theantiandrogenicactivitywasobserved only at concentrations much higher thanthe onesused in our MIX.Moreover,linksbetweenalteredgeneexpressionpatternsand physiologicalimplicationswerenotestablished,sinceno pheno-typeeffectwasrecordedinthosestudies.However,somedegree ofanantiandrogeniceffectoftheMIXcannotbediscarded,and morestudiesareneededtodiscoverthephysiologicaleffectsofthe EDCswithantiandrogenicactivity.

4.3. Effectsonoogenesis

Infemales,nodifferencesineithertherelativeortheabsolute volumesoftheovaryandofeachdifferentoocytestagewerefound betweentheMIXandthecontrolgroup.Ashypothesized,andin linewithwhathavebeenpublishedforzebraﬁsh(LinandJanz, 2006;VanDenBeltetal.,2002;Weberetal.,2003),theovaryof thefemalesexposedtoEE2hadahigherrelativevolumeofprimary growthstageoocytesandalowerrelativevolumeofvitellogenic and matureoocytes thanthe control group,thus showingthat ovariesrespondedsensitivelytoexposuretoxenoestrogens.Asthe ovaryoftheﬁshexposedtoEE2wasthesmallest,theabsolute vol-umeoftheprimarygrowthstageoocyteswassimilarinthevarious groups.Thisisoffundamentalimportance,meaningthatthe base-linepoolofprimarygrowthoocytesdidnotactuallyincreaseunder EE2exposure,butratheritstayedstable–estrogenicimpactswere thustargetingthemorematurestages.Again,theseresults high-lighttheneedtolookbothforabsoluteandforrelativevolumesof thegermcellstoreallyunderstandthemorphofunctionalchanges occurringinthegonads–partialapproachescanindeedpromote misinterpretations.

Onepossibleexplanationforthedifferentresultsobtainedin malesandfemalesisthattheestrogenicthresholdfor suppress-ingmalegametogenesisislowerthanthatforfemales.Therefore, the xenoestrogens in the mixture (20.73ng/L EE2-equivalents, Urbatzkaetal.,2012)allowedforoogenesistooccurbutsuppressed spermatogenesis.Atthehigherestrogenicexposure(100ng/LEE2) therewasgametogenesisdisruptioninbothmalesandfemales.In linewithourconclusions,HillandJanz(2003)conducteda par-tiallife-cycle(2–60dph)studywithzebraﬁshconcludingthatlow estrogenconcentrations(1ng/Lnominal)affectonlymale gameto-genesis,whereasdisruptionoffemalematurationrequirehigher estrogenlevels(100ng/Lnominal). Thisdifferencebetweenthe sexescouldbelinkedtothedistinctive natureof thehormonal regulationinmalevs.femalegametogenesis.Furtherstudiesare neededtoconﬁrmthishypothesis.

4.4. Effectsonbiometricparameters

Similarbodyweight,gonadweightandGSIwerefoundinmales oftheMIXandthecontrolgroup.Thesebiometricparameterswere lowerfor malesexposed toEE2,inagreementwithother stud-ieswhere a GSI decrease wasfoundin various adultmale ﬁsh speciesexperimentallyexposedtochemicalswithestrogenic activ-ity(Gimenoetal.,1998;Joblingetal.,1996;Komenetal.,1989), including zebraﬁsh (Van Den Belt et al., 2002).The absenceof

(8)

differencesinGSIbetweenmalesoftheMIXandcontrolgroups, revealedthat the useof the index alone wasnot, by itself, an appropriateproxyforevaluatingthesubtlespermatogenesis dis-turbancesandhighlightedtheimportanceofusingstereological microscopictechniquestostudytheEDCeffectsongonads.Also inarecentstudywithzebrafishaimedatdisclosingthe qualita-tiveandquantitativeeffectsofsub-acuteexposurestofivedistinct pharmaceuticals(alsofoundintheDouroRiver)inthefishgonadal maturationdynamics,Madureiraetal.(2011)foundthatthe sen-sitivityof GSItothose substances didnot correspondtogonad volumepartitioning,whichreinforcestheideathatGSIdoesnot alwaysshowtherealtoxicanteffects ofcompoundsonthefish gonads.ThebiometricparameterswerelowerfortheEE2female groupwhencomparedtotheMIXandcontrolgroups.Thelower GSIforEE2is,probably,aresultofareductionintheovarysizedue tothelossofmaturegermcells.Noeffectofthexenoestrogens mix-turewasobservedonfemalesregardingbodyandgonadweights. TheGSIvalue,however,washigherintheMIXgroupwhen com-paredtotheothergroups.Asaresult,similartomales,thefemale GSIaftertheexposuretothexenoestrogensmixturedidnotreflect thestereologicalresults.Inshort,theGSIshouldnotbeasingle endpointtostudyxenoestrogenseffectsonfishgametogenesis. 4.5. Abouttheeffectsofsinglechemicalsvs.mixtures

Overthelastdecades,fieldandlaboratorystudieshaveshown inductionofadverseeffectsinfishgonadsunderexposureto sin-glexenoestrogens.ExposureofzebrafishtoEE2(≥1ng/Lnominal) hasbeenreportedtoinhibitgametogenesisbothintheearly(Lin andJanz,2006;Weberetal.,2003)andlatelifestages(VanDen Beltetal.,2002).Repressionofgametogenesiswasalsoobserved inzebrafishexposedtononylphenol(≥100␮g/Lnominal)(Linand Janz,2006;Weber etal.,2003).Asomewhatsimilarresultwas obtainedinrainbowtrout(Oncorhynchusmykiss)malesinastudy includingEE2 (_≈1.8ng/L), nonylphenol and other alkylphenols, suchas:4-tert-octylphenol (4-t-OP,≈38.5␮g/L), nonylphenoxy-carboxylicacid(NP1EC,≈31.8␮g/L)andnonylphenoldiethoxylate (NP2EO,≈38.3␮g/L)(Joblingetal.,1996).Gonadaldisruptionwas alsoseenunder estrone(E1,≈31.8ng/L) and17␤-estradiol(E2, >100ng/L) exposures in an experiment with fathead minnows (Pimephalespromelas)(Panter etal., 1998).Lin andJanz (2006) determinedtheeffectsofexposuretobinarymixturesofbothEE2 andnonylphenolongametogenesisofzebrafish,andtheyfound thatsuchexposuresupressedtheoogenesisinanadditivewayin femalesat60dph.Basedonourresultsitisnotpossibleto pre-dictthewaythesechemicalsinteractwitheachotherandthiswas beyondourscope.Thenoteddisruptionofspermatogenesisinour studycouldthushavebeencausedbyadditiveand/ornon-additive effects.

ExposuretoMIXcaused changesinzebraﬁshgametogenesis, especiallyinmales.Localeffectsofreproductiontoxicityattributed toxenoestrogens foundin theDouroRiverwere demonstrated before,suchastheconsistentpresenceofabnormaltestis-ovain greymullet(Mugilcephalus)males(Carrolaetal.,2009;Ferreira etal., 2004).The effectsof thexenoestrogens onthegermcell maturationasseeninourstudywereobtainedwithrealistic envi-ronmental concentrations (Rocha et al., 2011), and it must be emphasizedthatthesewereevenlowerthantheonesveryrecently reportedalsofortheDouroestuarybyRochaet al.(2012).The authorsconcludedthatanthropogenicpollutionintheDouro estu-aryis a continuous processand that theEDCs werepresent in quantitiesthat shouldbeable toproduceendocrine disruption under diverse conditions and in different species. The present studysupportsthisideaandagreeswithothersthathavefound ﬁshreproductionimpairmentatlowxenoestrogenconcentrations. Thelowestobservableeffectconcentration(LOEC)foreffectson

fecundity and fertilitywasfound inzebrafish, where a lifelong exposureto5ng/LEE2caused56%reductioninfecundityand com-pletepopulationfailurewithnofertilization(Nashetal.,2004).The effectsofEE2 atsmallconcentrations infishreproductionwere alsoreportedinotherstudies:(i)0.1–15ng/Lcanaffectthe nor-malsexualdevelopmentanddifferentiation(Andersenetal.,2003; Metcalfeetal.,2001;Weberetal.,2003);(ii)2–10ng/Lcanaffect fecundity(Längeetal.,2001;ScholzandGutzeit,2000;VanDen Beltetal.,2002);and(iii)1–10ng/Lcanreducethefertilization suc-cessortheviabilityofembryosfromexposedadults(HillandJanz, 2003;Längeetal.,2001;Segneretal.,2003).Summarizing,thedata supportthenotionthattheconcentrationsofthechemicalswith estrogenicactivityfoundintheDouroRiverestuaryaresufficient toimpairfishgametogenesis,withyetnotclearconsequencesto thefertilityoffishinhabitingthosewaters.

Thisisoneofthefirststudiesonsub-acuteeffectsofan environ-mentallyrelevantmixtureofxenoestrogensinthegametogenesis kinetics of zebrafish using stereological tools, directed at rela-tiveandabsolutevolumes.Previousstudiesonestrogenexposure haveusedvarioustechniques,suchasvitellogenindetectionand commonhistology.Thelattermainlyinvolvedqualitative tissue analyses, resulting in descriptions of structural alterations (as detailed,e.g.,bySegneretal.,2003).Theapplicationof stereol-ogyinthisstudyallowedustoquantify,inanunbiasedway,some ofthosegonadalterations,namelytherelativevolumeoccupied bythedifferentgermcelltypespresentinthegonads,leadingto soundconclusionsaboutestrogeniceffectsongametogenesis.This stereologicalapproachisalsoaveryusefultooltostudytheimpact ofnon-estrogeniccompoundsinzebrafishgonads(Madureiraetal., 2011).Asaresult,itisouropinionthathistologiceffectsonfish gonadsasevaluatedbystereologymaybeoneofthemostsensitive andrelevantendpointstolookatinreproductivetoxicology. 4.6. Histopathologicalfindings

PathologicalchangesintestisexposedtoEE2andtotheDouro Rivermixturewere seenin ourstudy.The alterationsincluded presenceofinterstitialfibrosisandproteinaceousfluid.Increased interstitialfibrosiswasearlierfoundafterestrogenicexposureof adultmalefish(Grayetal.,1999;Karelsetal.,2003;Rasmussen etal.,2005).Rasmussenetal.(2005)demonstratedthatan anti-estrogencanabolisheffectsonthehistologyofthetestis,proving thatsomeeffectsaremediatedbytheestrogenreceptor.Halfof themalespossessedproteinaceousfluidinbothassaygroups(EE2 andMIX);thiskindoffluidispresumablyaderivatefrom vitel-logenin,thepresenceofwhich inthetestisisusually linkedto exposuretoestrogenicsubstances(OECD,2009).Theproteinaceous fluidinthetestisoffishexposedtothemixturerelateswellwiththe aboveproposal–i.e.,thatthexenoestrogenmixtureeffectswere viaestrogenicsignaling.

Pathologicalchangesoftheovarieswerealsopresent, appear-ingasfollicularatresia,fibrosis,proteinaceousfluidpresenceand infiltrationwithmacrophages.Ovarianfollicleatresiawaslower infishexposedtoEE2thaninfishexposedtothexenoestrogens mixtureorinfishofthecontrolgroup.Thiscouldbeexplainedby thefactthatovariesoffemalesofbothcontrolandMIXgroupshad morevitellogenicandmaturefolliclesthantheonesexposedto EE2.Folliclesinanadvancedmaturitystagehadahigher probabil-ityofbecomingatreticthereforeoccupyinghighervolumeinthe ovary.Despitenothavingbeenunbiasedlyquantified,increased ovarianfollicleatresia wasreportedin studieswhere fishhave beenexposedtoestrogens(Grayetal.,1999;Längeetal.,2001; Papouliasetal.,1999;Zilliouxetal.,2001).Allfemalesexposed tothehigherestrogenicstimulus(EE2)possessedgranulomatous inflammationandproteinaceousfluidintheovaryandhalfofthem hadalsomineralizedstructures.Fromallpathologicalalterations

(9)

mentionedbeforeonlythepresenceoftheproteinaceousfluidwas foundinfishexposedtothemixture(3outof6).Similarly,Schäfers etal.(2007)inafulllife-cyclezebrafishexposuretoEE2saw fibro-sisandinfiltrationwithmacrophagesandotherinflammatorycells intheovary.Althoughtheexactcauseofthepathologicalchanges observedinfishexposedtotheMIXremainsunknown,itcouldbe logicallyarguedthattheyarelinkedtotheestrogenexposureand canadverselyaffectbreeding.

5. Conclusions

Exposure of adult zebraﬁsh to EE2 for 21 days caused dis-ruptionof gametogenesisin both sexes,witha decreaseofthe volumesofthemoreadvancedmaturationcellstages.We demon-stratedthatgametogenesisrespondedsensitivelytoexposureto xenoestrogensandthereforeunbiasedstereologicalparametersare valuableadditionalendpointsinexperimentaland ecotoxicologi-calstudies.Onemajoroutcomeofthisstudywasthattheexposure totheMIXfoundintheDouroRiverestuary(aproxyforother similarlypolluted ecosystems)disrupted gametogenesisonly in males,bydecreasingthevolumeofspermatozoa.Howthe mix-turepromptedthesechangesinmalesisnotyetknown,butthe datafromEE2exposureessentiallypointtoestrogenicityasthe cul-prit,withincreasingeffectsobservedwithicreasingestrogenlevel. However,other underlyingmechanisms, suchasanti-androgen activity,mayalsoplayarole.Itshouldbeemphasizedthatprior tothisinvestigationveryfewstudieshaveunbiasedlyevaluated quantitativeeffectsof mixturesof toxicantsonthehistologyof gonads.Thepresence of estrogenic chemicals,theirpersistence andaccumulationintheDouroRiverandotheraquaticsystems, togetherwiththehereinreportedeffectsonﬁshgametogenesis, indicatethatsuchlocal“estrogenicmixtures”canhavesevere con-sequences.

Acknowledgments

ThisworkwasﬁnanciallysupportedbyFEDERfundsthroughthe CompetitivenessandTradeExpansionProgram– COMPETEandby NationalFundsprovidedbyFundac¸ãoparaaCiênciaeaTecnologia – FCT,viatheresearchprojectPTDC/MAR/70436/2006.

References

Andersen,L.,Bengtsson,B.E.,Björk,M.,Gessbo,A.,Holbech,H.,Hylland,K.,Norrgren, L.,Pedersen,K.L.,Lundgren,A.,Petersen,G.I.,Steinholz,A.,Örn,S.,2000.Zebraﬁsh fortestingendocrinedisruptingchemicals.TemaNord555,Copenhagen. Andersen,L.,Holbech,H.,Gessbo,A.,Norrgren,L.,Petersen,G.I.,2003.Effectsof

exposureto17␣-ethinylestradiolduringearlydevelopmentonsexual differen-tiationandinductionofvitellogenininzebraﬁsh(Daniorerio).Comp.Biochem. Physiol.CToxicol.Pharmacol.134,365–374.

Andreassen,T.K.,Skjoedt,K.,Anglade,I.,Kah,O.,Korsgaard,B.,2003.Molecular cloning,characterisation,andtissuedistributionofoestrogenreceptoralpha ineelpout(Zoarcesviviparus).Gen.Comp.Endocrinol.132,356–368. Ashﬁeld,L.A.,Pottinger,T.G.,Sumpter,J.P.,1998.Exposureoffemalejuvenile

rain-bowtrouttoalkylphenoliccompoundsresultsinmodiﬁcationstogrowthand ovosomaticindex.Environ.Toxicol.Chem.17,679–686.

Blázquez,M.,Zanuy,S.,Carrillo,M.,Piferrer,F.,1998.Structuralandfunctionaleffects ofearlyexposuretoestradiol-17␤and17␣-ethinylestradiolonthegonadsofthe gonochoristicteleostDicentrarchuslabrax.FishPhysiol.Biochem.18,37–47. Carrola,J.,Fontaínhas-Fernandes,A.,Pardal,M.,Castro,B.,Rocha,E.,2009.Intersex

ratioingreymulletsfromtheMondego,DouroandAveestuaries,Portugal– preliminarydata.In:15thInternationalSymposiumonPollutantResponsesin MarineOrganisms–PRIMO15,Bordeaux(France),17–20ofMay,p.326,Abstract book.

Christiansen,T.,Korsgaard,B.,Jespersen,Å.,1998.Effectsofnonylphenoland 17␤-oestradiolonvitellogeninsynthesis,testicularstructureandcytologyinmale eelpoutZoarcesviviparus.J.Exp.Biol.201,179–192.

Colombo,G.,Grandi,G.,1995.SexdifferentiationintheEuropeaneel:histological analysisoftheeffectsofsexsteroidsonthegonad.J.FishBiol.47,394–413. Depledge,M.H.,Billinghurst,Z.,1999.Ecologicalsigniﬁcanceofendocrinedisruption

inmarineinvertebrates.Mar.Pollut.Bull.39,32–38.

Drèze,V.,Monod,G.,Cravedi,J.P.,Biagianti-Risbourg,S.,LeGac,F.,2000.Effectsof 4-nonylphenolonsexdifferentiationandpubertyinmosquitoﬁsh(Gambusia holbrooki).Ecotoxicology9,9–103.

Falconer,I.R.,Chapman,H.F.,Moore,M.R.,Ranmuthugala,G.,2006.Endocrine dis-ruptingcompounds:areviewoftheirchallengetosustainableandsafewater supplyandwaterreuse.Environ.Toxicol.21,181–191.

Ferreira,M.,Antunes,P.,Gil,O.,Vale,C.,Reis-Henriques,M.A.,2004.Organochlorine contaminantsinXounder(PlatichthysXesus)andmullet(Mugilcephalus)from Douroestuary,andtheiruseassentinelspeciesforenvironmentmonitoring. Aquat.Toxicol.69,347–357.

Gimeno,S.,Komen,H.,Jobling,S.,Sumpter,J.,Bowmer,T.,1998.Demasculinization ofsexuallymaturemalecommoncarp,Cyprinuscarpio,exposedto 4-tert-pentylphenolduringspermatogenesis.Aquat.Toxicol.43,93–109.

Gray, M.A., Metcalfe, C.D., 1997. Induction of testis-ova in Japanese medaka (Oryzias latipes) exposed to p-nonylphenol. Environ. Toxicol. Chem. 16, 1082–1086.

Gray,M.A.,Niimi,A.J.,Metcalfe,C.D.,1999.Factorsaffectingthedevelopmentof testis-ovainmedaka.Oryziaslatipes,exposedtooctylphenol.Environ.Toxicol. Chem.18,1835–1842.

Gutendorf,B.,Westendorf,J.,2001.Comparisonofanarrayofinvitroassaysfor theassessmentoftheestrogenicpotentialofnaturalandsyntheticestrogens, phytoestrogensandxenoestrogens.Toxicology166,79–89.

Hill,R.L.,Janz,D.M.,2003.Developmentalestrogenicexposureinzebraﬁsh(Danio rerio):I.Effectsonsexratioandbreedingsuccess.Aquat.Toxicol.63,417–429. Howard,C.V.,Reed,M.G.,2005.UnbiasedStereology.Three-Dimensional Measure-mentsinMicroscopy,2ndedition.GarlandScience/BiosScientiﬁcPublishers, OxonandNewYork.

Hutchinson,T.H.,Shillabeer,N.,Winter,M.J.,Pickford,D.B.,2006.Acuteandchronic effectsofcarriersolventsinaquaticorganisms:acriticalreview.Aquat.Toxicol. 76,69–92.

Jobling,S.,Sheahan,D.,Osborne,J.A.,Matthiessen,P.,Sumpter,J.P.,1996.Inhibitionof testiculargrowthinrainbowtrout(Oncorhynchusmykiss)exposedtoestrogenic alkyl-phenolicchemicals.Environ.Toxicol.Chem.15,194–202.

Karels,A.A.,Manning,S.,Brouwer,T.H.,Brouwer,M.,2003.Reproductiveeffectsof estrogenicandantiestrogenicchemicalsonsheepsheadminnows(Cyprinodon variegatus).Environ.Toxicol.Chem.22,855–865.

Khan,I.A.,Thomas,P.,1998.Estradiol-17betaando,p_-DDT_stimulate_gonadotropin

releaseinAtlanticcroaker.Mar.Environ.Res.46,149–152.

Kolpin,D.W.,Furlong,E.T.,Meyer,M.T.,Thurman,E.M.,Zaugg,S.D.,Barber,L.B., Buxton,H.T.,2002.Pharmaceuticals,hormones,andotherorganicwastewater contaminantsinU.S.streams,1999–2000:anationalreconnaissance.Environ. Sci.Technol.36,1202–1211.

Komen,J.,Lodder,P.A.J.,Huskens,F.,Richter,C.J.J.,Huisman,E.A.,1989.Effectsof oraladministrationof17␣-methyltestosteroneand17␤-estradiolongonadal developmentincommoncarp,Cyprinuscarpio.Aquaculture78,349–363. Lange,A.,Katsu,Y.,Miyagawa,S.,Ogino,Y.,Urushitani,H.,Kobayashi,T.,Hirai,T.,

Shears,J.A.,Nagae,M.,Yamamoto,J.,Ohnishi,Y.,Oka,T.,Tatarazako,N.,Ohta,Y., Tyler,C.R.,Iguchi,T.,2012.Comparativeresponsivenesstonaturalandsynthetic estrogensofﬁshspeciescommonlyusedinthelaboratoryandﬁeldmonitoring. Aquat.Toxicol.109,250–258.

Länge,R.,Hutchinson,T.H.,Croudace,C.P.,Siegmund,F.,Schweinfurth,H.,Hampe, P.,Panter,G.H.,Sumpter,J.P.,2001.Effectsofthesyntheticoestrogen 17a-ethinylestradiolonthelife-cycleofthefatheadminnow(Pimephalespromelas). Environ.Contam.Toxicol.20,1216–1227.

Laws,S.C.,Carey,S.A.,Ferrell,J.M.,Bodman,G.J.,Cooper,R.L.,2000.Estrogenicactivity ofoctylphenol,nonylphenol,bisphenolAandmethoxychlorinrats.Toxicol.Sci. 54,154–167.

Legler,J.,Broekhof,J.L.M.,Brouwer,A.,Lanser,P.H.,Murk,A.J.,VanderSaag,P.T., Vethaak,A.D.,Wester,P.,Zivkovic,D.,VanderBurg,B.,2000.Anovelinvivo bioassayfor(xeno-)estrogensusingtransgeniczebraﬁsh.Environ.Sci.Technol. 34,4439–4444.

Lin,L.L.,Janz,D.M.,2006.Effectsofbinarymixturesofxenoestrogensongonadal developmentandreproductioninzebraﬁsh.Aquat.Toxicol.80,382–395. Madureira,T.V.,Rocha,M.J.,Cruzeiro,C.,Galante,M.H.,Monteiro,R.A.,Rocha,E.,

2011.ThetoxicitypotentialofpharmaceuticalsfoundintheDouroRiverestuary (Portugal):assessingimpactsongonadalmaturationwithahistopathological andstereologicalstudyofzebraﬁshovaryandtestisaftersub-acuteexposures. Aquat.Toxicol.105,292–299.

Metcalfe,C.D.,Metcalfe,T.L.,Kiparissis,Y.,Koenig,B.G.,Khan,C.,Hughes,R.J.,Croley, T.R.,March,R.E.,Potter,T.,2001.Estrogenicpotencyofchemicalsdetectedin sewagetreatmentplantefﬂuentsasdeterminedbyinvivoassayswithJapanese medaka(Oryziaslatipes).Environ.Toxicol.Chem.20,297–308.

Nash, J.P.,Kime,D.E.,Vander Ven,L.T.M., Wester, P.W.,Brion,F.,Maack, G., Stahlschmidt-Allner,P.,Tyler,C.R.,2004.Long-termexposureto environmen-talconcentrationsofthepharmaceuticalethynylestradiolcausesreproductive failureinﬁsh.Environ.HealthPerspect.112,1725–1733.

Panter,G.H.,Thompson,R.S.,Sumpter,J.P.,1998.Adversereproductiveeffectsin malefatheadminnows(Pimephalespromelas)exposedtoenvironmentally rel-evantconcentrationsofthenaturaloestrogens,oestradiolandoestrone.Aquat. Toxicol.42,243–253.

Papoulias,D.M.,Noltie,D.B.,Tillitt,D.E.,1999.Aninvivomodelﬁshsystemtotest chemicaleffectsonsexualdifferentiationanddevelopment:exposuretoethinyl estradiol.Aquat.Toxicol.48,37–50.

Papoulias,D.M.,Noltie,D.B.,Tillitt,D.E.,2000.Aninvivomodelﬁshsystemtotest chemicaleffectsonsexualdifferentiationanddevelopment:exposuretoethinyl estradiol.Aquat.Toxicol.48,37–50.

(10)

Paris,F.,Balaguer,P.,Terouanne,B.,Servant,N.,Lacoste,C.,Cravedi,J.P.,Nicolas,J.C., Sultan,C.,2002.Phenylphenols,biphenols,bisphenol-Aand4-tert-octylphenol exhibitalphaandbetaestrogenactivitiesandantiandrogenactivityinreporter celllines.Mol.Cell.Endocrinol.193,43–49.

Piferrer,F.,Donaldson,E.M.,1992.Thecomparativeeffectivenessofthenaturalanda syntheticestrogenforthedirectfeminizationofchinooksalmon(Oncorhynchus tshawytscha).Aquaculture106,183–193.

Porte,C.,Janer,G.,Lorusso,L.C.,Ortiz-Zarragoitia,M.,Cajaraville,M.P.,Fossi,M.C., Canesi,L.,2006.Endocrinedisruptorsinmarineorganisms:approachesand perspectives.Comp.Biochem.Physiol.C143,303–315.

Preuss,T.G.,Gehrhardt,J.,Schirmer,K.,Coors,A.,Rubach,M.,Russ,A.,Jones,P.D., Giesy,J.P.,Ratte,H.T.,2006.Nonylphenolisomersdifferinestrogenicactivity. Environ.Sci.Technol.40,5147–5153.

Rasmussen,T.H.,Teh,S.J.,Bjerregaard,P.,Korsgaard,B.,2005.Anti-estrogenprevents xenoestrogen-inducedtesticularpathologyofeelpout(Zoarcesviviparus).Aquat. Toxicol.72,177–194.

Rocha,M.J.,Ribeiro,C.,Ribeiro,M.,2010.DevelopmentandoptimizationofaGC–MS methodtoevaluaterelevantendocrinedisruptersinenvironmentalmatrices– estrogensandpersistentpollutants.Int.J.Environ.Anal.Chem.91,1190–1204. Rocha,M.J.,Ribeiro,C.,Ribeiro,M.,2011.DevelopmentandoptimizationofaGC-MS methodtoevaluaterelevantendocrinedisruptersinenvironmentalmatrices– estrogensandpersistentpollutants.Int.J.Environ.Anal.Chem.19,1191–1205. Rocha,M.J.,Cruzeiro,C.,Ferreira,C.,Rocha,E.,2012.Occurrenceofendocrine dis-ruptorcompoundsintheestuaryoftheIberianDouroRiverandnearbyPorto Coast(NWPortugal).Toxicol.Environ.94,252–261.

Scholz,S.,Gutzeit,H.O.,2000.17-alpha-ethinylestradiolaffectsreproduction,sexual differentiationandaromatasegeneexpressionofthemedaka(Oryziaslatipes). Aquat.Toxicol.50,363–373.

Schäfers,C.,Teigeler,M., Wenzel,A.,Maack, G.,Fenske,M.,Segner,H., 2007. Concentration-andtime-dependenteffectsofthesyntheticestrogen.17alpha ethinylestradiol,onreproductivecapabilitiesofthezebraﬁsh,Daniorerio.J. Tox-icol.Environ.HealthA70,768–779.

Segner,H.,Caroll,K.,Fenske,M.,Janssen,C.R.,Maack,G.,Pascoe,D.,Shäfers,C., Vandenbergh,G.F.,Watts,M.,Wenzel,A.,2003.Identiﬁcationof endocrine-disruptingeffectsinaquaticvertebratesandinvertebrates:reportfromthe EuropeanIDEAproject.Ecotoxicol.Environ.Saf.54,302–314.

Sohoni,P.,Sumpter,J.P.,1998.Severalenvironmentaloestrogensarealso anti-androgens.J.Endocrinol.158,327–339.

Sun,H.,Xu,L.C., Chen,J.F.,Song,L.,Wang, X.R.,2006.Effect ofbisphenolA, tetrachlorobisphenolAandpentachlorophenolonthetranscriptional activi-tiesofandrogenreceptor-mediatedreportergene.FoodChem.Toxicol.44(11), 1916–1921.

Tanaka,J.,Grizzle,J.M.,2002. Effectsofnonylphenolonthegonadal differen-tiationofthe hermaphroditicﬁsh,Rivulusmarmoratus.Aquat.Toxicol.57, 117–125.

Urbatzka,R.,Rocha,E.,Reis,B.,Cruzeiro,C.,Monteiro,R.A.F.,Rocha,M.J.,2012.Effects ofethinylestradiolandofanenvironmentallyrelevantmixtureofxenoestrogens onsteroidogenicgeneexpressionandspeciﬁctranscriptionfactorsinzebraﬁsh. Environ.Pollut.164,28–35.

VanDen Belt,K.,Wester, P.W.,VanDerVen,L.T.M.,Verheyen,R.,Witters,H., 2002.Effectsofethinylestradiolonthereproductivephysiologyinzebraﬁsh (Daniorerio):timedependencyandreversibility.Environ.Toxicol.Chem.21, 767–775.

Weber,L.P.,HillJr.,R.L.,Janz,D.M.,2003.Developmentalestrogenicexposurein zebraﬁsh(Daniorerio):II.Histologicalevaluationofgametogenesisandorgan toxicity.Aquat.Toxicol.63,431–446.

Ying,G.G.,Kookana,R.S.,Ru,Y.J.,2002.Occurrenceandfateofhormonesteroidsin theenvironment.Environ.Int.28,545–551.

Zillioux,E.J.,Johnson,I.C.,Kiparissis,Y.,Metcalfe,C.D.,Wheat,J.V.,Ward,S.G.,Liu,H., 2001.Thesheepsheadminnowasaninvivomodelforendocrinedisruptionin marineteleosts:apartiallife-cycletestwith17alpha-ethinylestradiol.Environ. Toxicol.Chem.20,1968–1978.

Furtherreading(Webreferences)

OECD, 2009. OECD Guidance Document for the Diagnosis of Endocrine-Related Histopathology of Fish Gonads. Draft of January 7, 2009, http://www.oecd.org/dataoecd/33/27/42140701.pdf.