Artemiaparthenogenetica:Anexperimentalstudy

Aviancestodesaffectthebehaviouroftheirintermediatehost

Artemiaparthenogenetica:Anexperimentalstudy

M.I.Sa´ncheza,∗, B.B.Georgievb,c,A.J.Greena

aWetlandEcologyGroup,Estacio´nBiolo´gicadeDon˜ana–CSIC,Avda.Mar´ıaLuisas/n,41013Seville,Spain

bDepartmentofZoology,NaturalHistoryMuseum,CromwellRoad,LondonSW75BD,UK

cCentralLaboratoryofGeneralEcology,BulgarianAcademyofSciences,2GagarinStreet,1113Sofia,Bulgaria

Received23May2006;receivedinrevisedform3November2006;accepted3November2006

Abstract

ThebrineshrimpArtemiaparthenogenetica(Crustacea,Branchiopoda)isintermediate hostforseveralcestodespecieswhosefinalhostsare waterbirds.Previousfieldstudieshaveshownthatbrineshrimpsinfectedwithcestodeshaveabrightredcolourandarespatiallysegregatedin thewatercolumn.However,theethologicalmechanismsexplainingsuchfieldobservationsareunknown.Changesinappearanceandbehaviour inducedbytrophicallytransmittedparasiteshavebeenshowntoincreasetheriskofpredationbythefinalhost.Inthisexperimentalstudy,we comparedthebehaviourofuninfected Artemiaandthoseinfectedbyaviancestodes.Wefoundthatparasitised individuals behavedifferently fromunparasitised onesinseveralways.Incontrasttouninfectedindividuals,infectedbrineshrimpswerephotophilous andshowedincreased surface-swimmingbehaviour.Theseobservationssuggest thatthemodifiedbehaviour(inadditiontothebrightredcolour ofthemajorityofthe infectedindividuals)resultsininfectedbrine shrimpsbecomingmore vulnerabletoavian finalhosts, which facilitatesparasitetransmission.We discussourresultsintermsoftheadaptivenatureofbehaviouralchangesandtheirpotentialimplicationsforthehypersalineecosystem.

Keywords:Artemiaparthenogenetica;Cestodes;Parasitemanipulation;Hostbehaviour

1.Introduction

Theparasitemanipulation ofhostbehaviouriscurrentlya widely studiedphenomenonbecauseofitsecologicalandevo- lutionarysignificance(seeCombes,1991,2001;Poulin,1995,

1998;Moore,2002;Thomasetal.,2005;JogandWatve,2005). Recentstudiesemphasisetheimportanceofgatheringempirical datafromdiversehost–parasite systems(Thomasetal.,2005; Klein,2005;Hurd,2005;RigaudandHaine,2005;Poulinetal.,

2005;Pontonetal.,2006).Caseswhentwoormoretrophically transmittedparasitesco-occurinthesamehostpopulationand oneormoreofthemmanipulatehostbehaviourareofparticular interestbecauseofthecomplexityandthediversityofinterspe- cificinteractions(e.g.Thomasetal.,1998;Ce´zillyetal.,2000; Babiratetal.,2004;Haineetal.,2005;RigaudandHaine,2005; KostadinovaandMavrodieva,2005).Parasite-induced alter- ationsinthehostbehaviourmayhavemajorimplicationsatthe

∗Corresponding author.Presentaddress:GEMI/UMRCNRS-IRD2724, Equipe:“EvolutiondesSyste`mesSymbiotiques”,IRD,911AvenueAgropolis, B.P.64501,34394Montpelliercedex5,France.Tel.:+467416318.

E-mailaddress: (M.I.Sa´nchez).

ecosystem level(Combes, 1996;Thomasetal.,1999;Thomas andRenaud,2001;Mouritsen andPoulin,2005;Thompson et al.,2005)butthisremainsalargelyunexploredarea.

Amongthetrophically transmitted avianhelminths,most previousstudiesonparasite-inducedbehaviouralmodifications ininvertebrateintermediatehosts have focussedontrematodes (e.g.Helluy,1983,1984;Thomasetal.,1995;Babiratetal.,

2004;Kostadinova andMavrodieva, 2005)andacanthocepha- lans(e.g.Hindsbo, 1972;Moore,1983;Ce´zillyetal.,2000). Studies onparasite manipulationininvertebratesinfected with larvalcestodesarefew(Graham, 1963;HurdandFogo,1991; Yanetal.,1994).

BrineshrimpsofthegenusArtemia(Branchiopoda: Anos- traca)actasintermediatehostsfor13speciesofaviancestodes; theirparasitictransmissiondependsonpredationbybirds(see reviewinGeorgiev etal.,2005).Cestodes havebeenreported toinduce changesincolour andspatial distributionofinfected brineshrimpsinthefield;thesealterationsaresuspectedtobe associatedwithbehaviouralmodificationsfollowinginfection (Gabrionetal.,1982;Thie´ryetal.,1990;Sa´nchezetal.,2006a). However,theeffectoflarvalcestodesonArtemiabehaviourhas neverbeenassessed.Inthisstudy,weexplorepossiblecausesof

parasite-inducedspatialsegregationandcolourchangeobserved inthefield byexaminingthebehaviourofnaturallyinfected anduninfected brineshrimpsinthelaboratory. Sincemaking theupstream hostmorevisibletopredators isbelievedtobea majormechanism facilitating parasitetransmissionalongfood chains(Combes,2001),wequantifythelight/darkmicrohabitat preferenceandmeasuretimespentatthewatersurfacetotestthe predictionthatinfectedbrineshrimpsbecomemorevulnerable toaquaticbirds.Wealsotestthehypothesis thattheassocia- tionbetweenthehighcestodeburdenandtheredcolourofbrine shrimpsisduetotheincreasedtimespentatthesurface(resulting inaccumulationofphotoprotectivecarotenoids,seeSa´nchezet al.,2006a).Inaddition,cestodeshavebeensuggestedtoincrease thebuoyancyofinfectedArtemiabyincreasingtheirlipidcon- tent(Amatetal.,1991)andthereforesurfacingbehaviourcould easilybeconfoundedwithnegativegeotaxis.Thus,wedesigned anexperimenttoseparatetheeffectsofphototaxisandgeotaxis andtoevaluatewhetherphotophilyorchangesinbuoyancy areresponsible forsurfacingbehaviour;forthispurpose,we usedlightstimulitodrivethemovementofbrineshrimpin thewatercolumn.Finally,wediscussthepotentialeffectof cestodesonsaltpancommunitiesthroughtheirinfluenceon Artemiabehaviour.

2.Materialandmethods

2.1.Thehost–parasitesystem

BrineshrimpsArtemiaarewell-studiedcrustaceanswith anearworldwidedistributioninhabitingextremehypersaline environments suchassaltlakes,coastallagoonsandsaltpans (Persooneetal.,1980;Abatzopoulos etal.,2002).Artemia tendstobethedominantinvertebrateinnumberandbiomass intheplaceswhereitoccurs(e.g.Sa´nchezetal.,2006b).In ourstudyarea,Artemiaisintermediatehostfor8speciesof cyclophyllidean tapeworms;theiradultsinfectseveralspecies ofwaterbirds,includingshorebirds,flamingos,gullsandgrebes (Georgievetal.,2005).Theprevalenceofthecestodeinfectionin brineshrimpsisav.27%(Georgievetal.,2005),locallyreaching upto90%(unpublished data).Brineshrimpsbecomeinfected byconsumingcestodeeggs(containingalarvatermedanonco- sphere)releasedintothewater with thefaecesofthedefinitive hosts.Brineshrimpsare non-selectivefilterfeedersconsuming particlesinsuspension(Reeve,1963a,b)ranging6.8–27.5pm (Ferna´ndez,2001).Theoncosphere(20pm indiameterfor Flamingolepisliguloides,seeRobertandGabrion,1991)pen- etratesthroughtheintestinewallintothehemocoel whereit developsintoacysticercoid (cestodelarvalstagecontaining scolex).Thelifecyclecontinues whenaninfectedArtemiais predatedbythefinalhost.

2.2.Sampling,rearinganddeterminationofinfectionstatus

Basedontheobservationthat red colourinArtemiaisasso- ciatedwithcestode parasitism(Thie´ryetal.,1990;Robertand Gabrion, 1991;Amatetal.,1991;Sa´nchez etal.,2006a),we selected50redand50transparentArtemiaparthenogenetica

inthefieldinordertocomparethebehaviourofinfectedver- susnon-infectedindividuals.A.parthenogeneticaarefrequently rearedinthelaboratoryinconditionsfreeofparasites(Hontoria andAmat,1992),when theintenseredcolourofinfectedindi- vidualsfoundinthefield hasneverbeenobserved(F.Amat, personal communication).Weselectedadultindividualsofthe samesizeinordertocontrol forageasthesetwovariablesare correlated(F.Amat,personalcommunication).Wecollectedthe sampleswithanetof0.1mmmeshfromanevaporationpond ofintermediatesalinityatanindustrialsaltpanintheOdiel Marshes(SWSpain,37◦17tN, 06◦55tW) inNovember2003 (seeSa´nchezetal.,2006b,fordetailsofthestudyarea).

Inthelaboratory,onthedayofcollection,brineshrimpswere introducedintoatank,50(length)×10(height)×30(width) cm,filledtowithin1cmofthetopwithwaterfromthepond andunderanaturalphotoperiod.After24h,weconducteda seriesofthreedifferentbehaviouraltests(seebelow)oneach ofthe100individuals,onebyone,alternatinggroupsof10red

and10transparentindividuals.Onceallthetestswerefinished, brineshrimpswerekilledbyheatingto80◦Cand preservedin

70%ethanolforidentificationofparasites(seeGeorgievetal.,

2005formethodsofcestodespeciesidentification).

2.3.Behaviouraltests

2.3.1.Experiment1:light/darknesschoice

Thisexperimentwascarriedoutwithapetridishdividedinto darkandlighthalves.Weusedblackplastictocoveronehalfof thedishandilluminatedtheotherhalfwithacoldlightsource topreventathermalgradient. EachArtemiawasindividually placedinthemiddleofthedishandafter1minwerecordedits position.Wethenrepeatedtheexperimentcoveringtheopposite sideofthepetridishinordertoconfirmthelight/darkprefer- ence.Incasesinwhichtheresponse ofbrineshrimpchanged between thetwotrials(only4of100cases),werepeated the experiment.Thelight/darkpreferencewasscoredasa dichoto- mousvariable.Weusedχ2 teststocomparetheresponseto lightbetweentransparent uninfected, transparent infected,and redinfectedindividuals.Whenover20%oftheexpectedvalues inthecontingencytablewereless than 5,weperformedFisher exacttests.

2.3.2.Experiment2:timeatwatersurface

Thisexperiment wascarriedoutinawhiteshallow(50 (length)×10(height)×30(width)cm)tankthatallowedlightto penetratetheentirewatercolumnfromabove.EachArtemiawas introducedindividuallyinto thetank.After1min,werecorded thetime(inseconds,s)spentwithin1cmofwatersurfaceduring thefollowing30s.WeusedKruskal–Wallistestforcomparisons

ofmediantimeatthewatersurfacebetweentransparent unin- fected,transparentinfectedandredinfectedbrineshrimps.For determiningwhichpairsofgroupsweredifferent,weperformed multiple-comparisonposthoctests(SiegelandCastellan,1988).

2.3.3.Experiment3:abilitytodescendinthewatercolumn

Thiswasdesignedtoseparatetheeffectsofphototaxis and geotaxis(andalsotheeffectsofthenegativegeotaxisandthe

Table1

Presenceofcestodespecies(i.e.,numberofinfectedArtemiaindividuals)andnumberofthemparticipatinginsimpleandmultipleinfections

CestodespeciesPresenceSimpleinfectionMultipleinfection

2species3species4species

Thepositionofcrossesineachcolumnindicatestheprecisenatureofmultiplespeciesinfectionsobserved,withninthebottomrow.

possibleincreasedbuoyancy), usinglightstimulitodrivethe movementofbrineshrimpinthewatercolumn. Weusedagrad- uatedcylinderof5cmlengthfilledtothetopwithwaterand placedthebrineshrimpgentlyintothetopofthecolumn.After

1min,welaterallyilluminatedthetopofthetubewithacold light.ThepositionoftheArtemiaindividualinthecolumn(top orbottomhalf)wasscoredafterafurther30s(timeestimatedfor stabilization).Thiswayweevaluatedtheabilityofphotophobic individualsto go to the bottom.Thenwe placedthe tubein the darkand,after1min,werepeatedtheprocedurebutilluminated thebottomofthetube.Inthisway,weevaluatedtheabilityof photophilous individualstogotothebottom.Theresponseto thelightstimuluswasscoredasadichotomousvariable.

Werevealedtheeffectoflightongeotacticbehaviourby combiningtheresponseundertheprevioustests.Thus,weiden- tifiedindividualswhoseresponsewasmainlydeterminedby light(whentheirdepthwasdetermined bythepositionofthe lightinrelationtothewatercolumn)andindividualswhoseposi- tionwasmainlydeterminedbygravity(whentheirdepthwasthe sameregardless ofwhetherthewatercolumnwasilluminated fromaboveorbelow).

Chi-square testswereperformed fortwo-waycomparisons betweentransparent uninfected, transparent infectedandred infected individuals,comparingboththeabilitytoswimtothe bottomandtheeffectoflightonthegeotacticresponse.Forthe analyses,weexcluded16individuals thatshowedevidenceof notbeinginagoodcondition atthetimeofcarryingoutthe experiment. Fortheanalysisoftheeffectoflightonthegeo- tacticresponse,wealsoremovedafurther3individualsthat didnotrespondclearly(theystoppedinthemiddleofthewater column).

Inordertoexplorepossiblecausesforthedifferencesbetween transparentinfectedandredinfectedArtemia,wecomparedthe meanintensityofinfection(numberofcysticercoids perindi- vidual)betweenthesetwogroupsusingMann–WhitneyU-tests. Inthesameway,wecomparedtheprevalence ofthedifferent cestodespeciesbetweenredandtransparentinfectedArtemia.

Non-parametricstatisticaltestswerecarriedoutusingthe Statistica 6.0package(StatSoft Inc.,2001).Sincerepeated experiments andtestswerecarriedoutonthesameindividu- als,correctionformultipletestingisappropriate(Garc´ıa,2004). Allsignificantresultsforbehaviouralexperimentsremained

significantafterstrictBonferronicorrectionunlessotherwise stated.

3.Results

Weidentifiedsevenspeciesoftapewormsbelongingtothree differentfamilies(Hymenolepididae,DilepididaeandProgyno- taeniidae)inthesamplestudied(Table1).Cysticercoidsofupto

4differentspecieswereregisteredinasingleArtemiaandsimple infectionwasrecordedinonly26%oftheinfected individuals (Table1).Asmostparasitisedindividualsshowedmultipleinfec- tions,wewereunabletoisolatetheeffectofthedifferentspecies ofcestodesonbehaviourandcolourinasatisfactorymanner.

Theexperimentswereconductedon72infectedand28unin- fectedindividuals,assome ofthe50transparentbrine shrimps sampledprovedtobeparasitised.Therefore,weperformedour analysison3differentgroups:transparent uninfected(n=28), transparentinfected(n=22)andredinfectedbrineshrimps (n=50).

Amongthe72infectedbrineshrimps,69%werered(the prevalenceofinfectioninredindividualswas100%),whereasall

28(100%)uninfectedindividualsweretransparent(χ2=36.16,

d.f.=1,P0.001).Themeanintensityofinfectionwastwice ashighinred(4.28±0.33,mean±S.E.)thanininfectedtrans- parentArtemia(2.45±0.48)(Mann–Whitneytest,U=241.5, P=0.00013).Themaximumnumberofcysticercoidswas11in

transparent and13inredbrineshrimps.Therewerenoclear differencesincestodespeciescomposition betweeninfected individualsofdifferentcolour.However,theprevalencefortwo specieswassignificantlyhigherforredArtemia(Table2).

Ingeneral,parasitised individuals lookedmorevigorous thannon-parasitisedonesandresistedtheexperimentalmanip- ulationsbetter.Among14individuals thatdiedduringthe experiments, 13 were transparent and only one was red (χ2=10.05,d.f.=1,P=0.002).

Theproportion ofindividuals thatactivelymovedtowards lightinExperiment 1washighestforredinfected(0.92),fol- lowedbytransparentinfected (0.73)andfinallybytransparent uninfectedindividuals(0.54). Wefound significantdifferences betweentransparentuninfectedandredinfectedindividuals (χ2=13.378,P0.001).However,we foundno significant differencesbetweentransparentArtemiaofdifferentparasitic

Table2

Prevalenceofcestodespeciesintransparentinfectedandredinfectedindividualsandcomparisonbetweenthem

* ThesetestsremainsignificantwhencorrectedformultipletestingwithasequentialBonferroni-typeprocedure(BenjaminiandHochberg,1995).

status (χ2=1.192, d.f.=1, P=0.275) or between infected Artemia of different colours (Fisher exact test, P=0.058). When comparing all infected with all uninfected Artemia, wefoundthatasignificantly higherproportionofinfected individualsmovedtowardslight(χ2=10.29,d.f.=1,P0.001).

InExperiment2,werecordedstatisticallysignificantdif- ferences inthetimespentatthesurfacebetween groups (Kruskal–Wallistest,H(2,100)=23.39,P0.001).Redindi- vidualsspentmoretimeatthewatersurface(26.30±1.29s,

mean±S.E.)thantransparentones(infected12.21±2.64s,

uninfected12.21±2.63s)andthesedifferenceswerestatisti-

callysignificant.However,therewasnosignificantdifference

betweentransparent individuals ofdifferentparasiticstatus. Whencomparing allinfectedwithalluninfected Artemia,we foundthat infectedindividualsspentsignificantlymoretimeat thesurface(Mann–Whitneytest,U=627.0,P=0.001).

TheresultsofExperiment 3didnotsupportthehypothesis thatincreasedsurface-swimmingbehaviourwasduetophysi- calimpediment(highfloatability)ofinfectedindividuals.Most oftheArtemiaindividuals(morethan83%inthethreedifferent groups)swamintothebottomhalfofthewatercolumnduringthe experiment,withnosignificantdifferencesbetweengroupsin theproportionsdoingso(P0.6inallcases).Brineshrimpposi- tioninthewatercolumn(toporbottomhalf)wasmodifiedby lightpositionin61%ofuninfectedtransparent,60%ofinfected transparentand78%ofredArtemia,therebeingnosignificant differencesbetweengroupsintheseproportions(P0.22forall comparisons).Therefore,light(phototaxis)wasthemostimpor- tantstimulusdeterminingthefinalpositionofArtemia(both infectedanduninfected)inthewatercolumn,gravity(geotaxis) havingasecondaryrole.Amongindividualswhosepositionwas notaffectedbylight,wefound60%ofredinfected(n=10),43% ofinfectedtransparent(n=7)andonly14%ofuninfectedindi- viduals(n=7)tousethetophalfofthewatercolumn(i.e.,to benegativelygeotactic).Howeverthesmallsamplesizewasnot enoughtodetectstatisticallysignificantdifferences(P0.16in allthecases).Forthesamereason,whencomparingallinfected withalluninfectedArtemia,wedidnotfoundasignificantdiffer- enceintheproportionofinfectedindividualsshowingnegative geotaxis(Fisherexacttest,P=0.172).

AsinExperiment1,theproportionofbrineshrimpsshowing positivephototaxis(excludingthoseindividualswhoseresponse wasmainly determinedbygravity) inExperiment3wasmuch

higheramongstredinfectedArtemia(75%,n=36)thantrans- parentones(36%,n=11,foruninfectedtransparentand40%, n=10,forinfectedtransparent). Thedifferencebetweenunin- fectedandredinfectedshrimpswassignificant(Fisherexacttest, P=0.029)butnotafterBonferronicorrection.Whencomparing allinfectedwithalluninfectedArtemia,wefoundnosignificant difference(Fisherexacttest,P=0.086).

4.Discussion

4.1.Generalcomments

Ourresultsshowpronounced differencesinbehaviourand colourofbrineshrimpswheninfectedbycestodes.Infected brineshrimpsexhibitedpositivephototaxis, spentmoretime swimmingatthesurfaceofthewaterandusuallydisplayed abrightredcolouration.Theseresultsexplainpreviousfield observations ofdifferentialdistributionofinfectedindividuals ofArtemiainthe watercolumn(Gabrionetal.,1982).

Weobservedthestrongestdifferences betweenthered infectedandthetransparentuninfectedArtemia.Inexperiments

1and3,thebehaviouroftransparent infectedArtemiawas intermediatebetweenthatoftheothertwogroups.Infection intensitymay explainour observations,asthemeannumberof cysticercoidswasnearlytwiceashighinredthanintransparent infectedArtemia.Manipulation ofhostsbyhelminthparasites islikelytoberelatedtoparasite loadasshownforacantho- cephalans (Maynard et al., 1998) and trematodes(Webber etal.,1987).Althoughoneparasiteindividual seemstobe enoughtoinducehostmodificationsinsomesystems,apositive correlation existsbetweenthenumberofparasitesandthe extentofthemodification(Trabalonetal.,2000).Incontrast, theeffectoflarvalacanthocephalansonamphipodhostsappear tobedependantontheirpresenceonly,notontheintensity (Ce´zillyetal.,2000;Baueretal.,2000).Otherstudies suggest thatthedegreeofmaturationoflarvaedeterminestheextentof behaviouralmodification(Valkounova,1983;Williamsetal.,

2004).

We wereunabletoisolatetheeffectofeachcestodespecies onArtemiaphenotype,asmultipleinfectionsareextremelyfre- quentinthefield.Asourexperimentsrelyonnaturallyinfected Artemia,theydonotinthemselvesruleoutthepossibilitythata givenbehaviourwasacauseof increasedexposureto parasites

ratherthanaconsequenceofit.However,existingknowledgeof normal,adaptivehostbehaviour supportsacausalrelationship betweeninfectionandbehaviour.

4.2.ThenormalbehaviourofArtemia

Artemiahasnosophisticated antipredatorstructuresandis aneasyandnutritionalpreyforbirds, fishesandaquaticinver- tebrates.Itsdefenceagainstpredators istoavoidthem,(1)by livinginhypersaline habitats,whicharetooextremeformost predatoryinvertebratesandfishand(2) undergoingdailyverti- calmigration(DVM)(Lenz,1980;ForwardandHettler,1992). UninfectedArtemiaascendinthewatercolumnatsunsetto feednearthesurfaceanddescendatsunrisetoavoidvisual avianpredators(ForwardandHettler,1992).Indoingso,unin- fectedbrineshrimpshowsastrongnegativephototaxis(Lenz,

1980;BradleyandForward,1984)andpositivediurnalgeotaxis. Additionally,DVMconstitutesanefficientecologicalprotection fromharmfulultravioletlight(Rhodeetal.,2001).InArtemia, carotenoid pigmentsarediverseandreachhighconcentrations (Nelisetal.,1984,1988);theyarebelieved tohave,interalia, photoprotectivefunctions(Amatetal.,1991)andtocausethe redcolourationofbrineshrimps(seealsoSa´nchezetal.,2006a). However,similarcolourationwasreportedtoincreasepredation riskforcopepods(Hairston,1979).

4.3.Parasite-inducedalterationsinArtemiabehaviour

Trophicallytransmittedparasiteshavebeenshowntoincrease theirtransmission successbyadaptivelymanipulating the behaviouroftheirintermediatehost(Moore,2002;Thomas etal.,2005).Theobservedphenotypic changes(surface- swimmingbehaviourandredappearance)seemtoincrease theexposureandvisibilityofArtemiatosurface-feedingavian predators.

Increasedsurfaceactivityhasbeenreported foravarietyof invertebratesparasitisedby nematodes(McCurdyetal., 1999), trematodes(CrowdenandBroom,1980;MouritsenandJensen,

1997),acanthocephalans(Hindsbo,1972)andcestodes(Lester,

1971;Yanetal.,1994).Inmostcases,theunderlying mech- anismisunknown. Problems ofbuoyancy havebeenreported forsticklebacksparasitisedbycestodes,causingthemtoremain closertothesurface(ArmeandOwen,1967;Giles,1983;LoBue andBell,1993;NessandFoster,1999).Ourresultsdonotsup- portthehypothesisthatextrabuoyancyderivedfromthehigher lipidcontentofparasitised Artemia(Amatetal.,1991)isthe proximatecauseforthesurfacingbehaviour,asparasitisedindi- vidualswereabletoswimtowardsthebottomwhenstimulated bylight.Likewise, alteredhostbehaviour doesnotseemtobe duetoinfection-inducedhostdebilitation.

Ingeneral,parasitisedindividualslookedmorevigorousthan non-parasitised onesandresistedtheexperimentalmanipula- tionsbetter.Sinceonefunctionofcarotenoids islipidstorage, carotenoidsmayberesponsibleforsuchdifferences.Lipidcon- tentisbelievedtobeagoodpredictorofcopepodsurvival(Franz andKurtz,2002).Increased lipidreservesininfectedArtemia (Amatetal.,1991)mayfunctionasaparasitestrategytoincrease

thelongevityofinfected individuals(thusenhancingtheprob- abilityofencounterwiththefinalhost)ormayincreasethe profitabilityofthepreyinordertoaugmentitsattractivenessfor thefinalhost.

We are confidentthat surface-swimmingbehaviour was notanincidentalconsequenceoftheinfectioncausedbyan increasedoxygen (Smith and Kramer,1987; Lester, 1971) orenergydemand(Milinski, 1985;GodinandSproul,1988) imposedbytheparasites,asweprovidednooxygenorfoodgra- dientintheshallowwaterusedinourexperiments. Moreover, atleastfor F. liguloides,infectionhasbeenshownnotto affect therespirationrateinArtemia(Varo´etal.,2000).Thewater columnwasuniformly lit,andneithercanincreased attraction ofparasitisedArtemiatolightexplainour results.One alterna- tiveexplanationisthatthepositioninthewatercolumnmay bealteredinArtemiabecausetheparasites reversethenormal positivegeotaxis.Weshowedlighttobethemostimportant factordrivingthemovements ofArtemiainthewatercolumn. However,amongtheindividualswhoseresponsewerenotinflu- encedbylight,86%ofuninfected Artemiashowedpositive geotaxis(bottombehaviour)whilst53%ofinfectedArtemia werenegativelygeotactic(surface-swimmingbehaviour).Neg- ativegeotaxisandpositivephototaxis havebeenreportedin amphipodsinfectedbytrematodes(Helluy,1983).

Theassociation observedbetweenthepresenceofcysticer- coidsandredcolourinArtemia agreeswithprevious studies (Thie´ryetal.,1990;RobertandGabrion,1991;Amatetal.,

1991;Sa´nchezetal.,2006a).Wesuggestthatcestodesinducered colourationinArtemiathroughanindirectmechanismmediated bythealterationinphotoresponse.Photophilousindividualsare moreexposedtoUVradiation, whichstimulates synthesisof photoprotective carotenoids.Accumulationofcarotenoidshas beenreportedforcopepods exposedtohighlevels ofUVradi- ation,resultinginintenselyredcrustaceans (Hairston,1979). Butthedefiniteprooffortheinductionofaredcolorationby surfacingbehaviourshouldcomefromanexperiment specifi- callytestingitbyforcinguninfected shrimpsandtransparent infected shrimps tostayatthesurface(cagedshrimps inthe field).Onelessparsimoniousexplanationisthatparasitesinduce redcolourinbrine shrimpsbyactingdirectlyonthecontrolof carotenoidsynthesis.Whateverthemechanism, redcolouris likelytoincreasethevisibilityofinfectedArtemiatovisual predators(Hairston,1979)asconfirmed inexperimentsusing wadersincaptivity(Sa´nchezetal.,inpreparation). Ifchanges inbehaviourareinitiatedpriortocolourchange,theparasite ismorelikelyto be matureenoughwhentransmissionto avian hostsoccurs.

4.4.Consequencesofparasite-inducedbehavioural changes

Ourstudydemonstratesthatparasite-inducedbehavioural alterationsresultinspatialsegregation inhostpopulations, whichsupportstheviewthatparasitesmayinfluence dynam- ics,distributionandgeneflowinhostpopulations(Wellnitz et al.,2003).However,therearealsoimmediateecologicalconse- quences forthehost(andfortheparasites)duetothedifferent

environmentalconditionsamongmicrohabitats (light,oxygen, temperature,nutrients,predators,etc.).Livinginalgae-richsur- facewaterincreasespredationrisk,butArtemiacanalsobenefit fromnutrientacquisition, whichmayexplainthelargerlipid reservesof infectedbrineshrimps(Amatetal., 1991).Artemia isakeystonespeciesinhypersaline habitatsandamajorfood resourceforwaterbirds(Cooperetal.,1984;Brittonetal.,1986; Verkuiletal.,2003;Sa´nchezetal.,2006b).Byincreasing the vulnerabilityandprofitabilityofbrineshrimpstoavianpreda- torsandbyreducingtheirfecundity(Amatetal.,1991),cestodes arelikelytobemajordeterminantsofpredator–preydynamics andpredatorhabitatselection.Attheecosystemlevel,cestodes are embeddedin the foodwebandmayhaveastrongeffecton energyflowandcommunitystructureinsaltpans.Moreresearch is requiredtoelucidatetheeffectsofparasitesonArtemiapop- ulationsandonhypersalineecosystems.

Acknowledgements

Theseniorauthorwassupported byaPhDgrantfromthe MinisteriodeCienciayTecnolog´ıaandanI3Ppostgraduate grantfromtheConsejoSuperiordeInvestigacionesCient´ıficas (CSIC).Consejer´ıadeMedioAmbiente,JuntadeAndaluc´ıaand Aragonesas IndustriasyEnerg´ıaS.A.providedpermissionto workinthesaltworks.Fre´de´ricThomasandtwoanonymousref- ereesprovideduseful commentsonthemanuscript.Thisstudy wascarriedoutintheframeworkofaco-operativeprogram betweentheBulgarianAcademyofSciencesandCSIC(project

2004BG0013).

References

Abatzopoulos,T.J.,Beardmore,J.A.,Clegg,J.S.,Sorgeloos,P.,2002.Artemia: BasicandAppliedBiology.KluwerAcademicPublishers,Dordrecht.

Amat,F.,Gozalbo,A.,Navarro,J.C.,Hontoria,F.,Varo´,I.,1991.Someaspectsof Artemiabiologyaffectedbycestodeparasitism.Hydrobiology212,39–44. Arme,C.,Owen,R.W.,1967.Infectionsofthethree-spinedstickleback,Gas- terosteusaculeatusL.,withplerocercoidlarvae ofSchistocephalussolidus (Muller, 1776),withspecialreferencetopathologicaleffects. Parasitology

57,301–314.

Babirat,C.,Mouritsen,K.N.,Poulin,R.,2004.Equalpartnership:twotrematode species,notone,manipulate theburrowing behaviouroftheNewZealand cockle,Austrovenusstutchburyi.J.Helminthol.78,195–199.

Bauer,A.,Trouve´,S.,Gre´goire,A.,Bollache,L.,Ce´zilly,F.,2000.Differential influenceofPomphorhynchuslaevis(Acanthocephala)onthebehaviourof nativeandinvadergammaridspecies.Int.J.Parasitol.30,1453–1457.

Benjamini,Y.,Hochberg,D.,1995.Controllingthefalsediscoveryrate:aprac- ticalandpowerfulapproachtomultipletesting.J.R.Stat.Soc.Ser.B57,

289.

Bradley,D.J.,Forward,R.B.J.,1984.PhototaxisofadultbrineshrimpArtemia salina.Can.J.Zool.62,2357–2359.

Britton,R.H.,deGroot, E.R.,Johnson,A.R.,1986. Thedailycycleoffeeding activityoftheGreaterFlamingo inrelationtothedispersion oftheprey Artemia. Wildfowl37,151–155.

Ce´zilly,F.,Gre´goire,A.,Bertin,A.,2000.Conflictbetweenco-occurringmanip- ulativeparasites?Anexperimentalstudyofthejointinfluence oftwo acanthocephalanparasitesonthebehaviourofGammaruspulex.Parasitol- ogy120,625–630.

Combes,C., 1991.Ethologicalaspectsofparasitetransmission.Am.Nat.138,

866–880.

Combes,C.,1996.Parasites,biodiversityandecosystemstability.Biodivers.

Conserv.5,953–962.

Combes,C.,2001.Parasitism:TheEcologyandEvolutionof IntimateInterac- tions.UniversityofChicagoPress,Chicago.

Cooper,S.D.,Winkler,D.W.,Lenz,P.H.,1984.Theeffectofgrebepredationon brineshrimppopulation.J.Anim.Ecol.53,51–64.

Crowden,A.E.,Broom,D.M.,1980.Effectsoftheeyefluke,Diplostomum spathaceum,onthebehaviourof dace(Leuciscusleuciscus).Anim.Behav.

28,287–294.

Ferna´ndez,R.G.,2001.Artemiabioencapsulation.I.Effectofparticlesizes onthefilteringbehaviorofArtemiafranciscana.J.Crust.Biol.21,435–

442.

Forward,R.B.,Hettler,W.F.,1992.Effectsoffeedingandpredatorexposure onphotoresponsesduringdialverticalmigrationofbrineshrimplarvae. Limnol.Oceanogr.37,1261–1270.

Franz, K., Kurtz, J., 2002. Altered host behaviour: manipulation or energydepletionintapeworm-infectedcopepods?Parasitology125,187–

196.

Gabrion,C.,MacDonald-Crivelli,G.,Boy,V.,1982.Dynamiquedespopulations larvairesducestodeFlamingolepisliguloidesdansunepopulationd’Artemia enCamargue.ActaOecol.3,273–293.

Garc´ıa,L.V.,2004.EscapingtheBonferroniironclawinecologicalstudies.

Oikos105,657–663.

Georgiev, B.B.,Sa´nchez, M.I.,Green,A.J.,Nikolov,P.N.,Vasileva, G.P., Mavrodieva,R.S.,2005.CestodesfromArtemiaparthenogenetica (Crus- tacea,Branchiopoda) intheOdielMarshes,Spain:asystematicsurveyof cysticercoids.ActaParasitol.50,105–117.

Giles,N.,1983.BehaviouraleffectsoftheparasiteSchistocephalussolidus(Ces- toda)onanintermediate host,thethree-spined stickleback, Gasterosteus aculeatusL.Anim.Behav.31,1192–1194.

Godin,J-G.J.,Sproul,C.D.,1988.Risktakinginparasitizedsticklebacksunder threatofpredation:effectsofenergeticneedandfoodavailability. Can.J. Zool.66,2360–2367.

Graham,G.L., 1963.The behaviorofbeetles,Triboliumconfusum,parasitized bythelarvalstageofachickentapeworm,Raillietinacesticillus.Trans.Am. Microscop.Soc.85,163.

Haine,E.R.,Boucansaud,K.,Rigaud,T.,2005.Conflictbetweenparasiteswith differenttransmissionstrategiesinfectinganamphipodhost.Proc.Biol.Sci.

272,2505–2510.

Hairston,N.G.,1979.The adaptivesignificanceofcolor polymorphismintwo speciesofDiaptomusCopepoda.Limnol.Oceanogr.24,15–37.

Helluy,S.,1983.Relationshoˆtes-parasitesdutre´matodeMicrophalluspapilloro- bustus(Rankin,1940).II.ModificationsducomportementdesGammarus hoˆtesinterme´diairesetlocalisationdesme´tacercaires.Ann.Parasitol.Hum. Comp.58,1–17.

Helluy,S.,1984.Relationshoˆtes-parasitesdutre´matodeMicrophalluspapil- lorobustus(Rankin,1940)III.Facteursimplique´sdanslesmodificationsdu componentdesGammarushoˆtesinterme´diairesattestsdepredation.Ann. Parasitol.Hum.Comp.59,41–56.

Hindsbo,O.,1972.EffectsofPolymorphus(Acanthocephala) oncolourand behaviourofGammaruslacustris.Nature238,333.

Hontoria,F.,Amat,F.,1992.MorphologicalcharacterizationofadultArtemia (Crustacea,Branchiopoda) fromdifferentgeographicalorigin.Mediter- raneanpopulations.J.Plankt.Res.14,949–959.

Hurd,H.,2005.Parasitemanipulation:stretchingtheconcepts.Behav.Process.

68,235–236.

Hurd,H.,Fogo,S.,1991.ChangesinducedbyHymenolepisdiminuta(Cestoda) inthebehaviouroftheintermediatehostTenebriomolitor(Coleoptera).Can. J.Zool.69,2291–2294.

Jog,M.,Watve,M.,2005.Roleofparasitesandcommensals inshapinghost behaviour.Curr.Sci.89,1184–1191.

Klein, S.L.,2005. Parasitemanipulationofhostbehaviour:mechanisms,ecol- ogy,andfuturedirections.Behav.Process68,219–221.

Kostadinova,A.,Mavrodieva,R.S.,2005.MicrophallidsinGammarusinsensi- bilisStock, 1966fromaBlack Sealagoon:manipulationhypothesisgoing East?Parasitology131,337–346.

Lenz,P.H.,1980.EcologyofanalkaliadaptedvarietyofArtemia fromMono Lake,California,U.S.A.In:Persoone,G.,Sorgeloos,P.,Roels,O.,Jaspers, E.(Eds.),TheBrineShrimpArtemia,vol.III.UnversaPress,Wetteren, Belgium.

Lester,R.J.G.,1971.TheinfluenceofSchistocephalusplerocercoidsonthe respirationofGasterosteusandapossibleeffectonthebehaviourofthefish. Can.J.Zool.49,361–366.

LoBue,C.P.,Bell,M.A.,1993.Phenotypicmanipulationbythecestodeparasite Schistocephalussolidusofitsintermediatehost,Gasterosteusaculeatus,the three-spinestickleback.Am.Nat.142,725–735.

Maynard,B.J.,Wellnitz,T.A.,Zanini,N.,Wright,W.G.,Dezfuli,B.S.,1998.

Parasite-alteredbehaviorinacrustacean intermediatehost:fieldandlabo- ratorystudies.J.Parasitol.84,1102–1106.

McCurdy, D.G.,Forbes,M.R.,Boates,J.S.,1999.Evidencethattheparasitic nematodeSkrjabinoclavamanipulateshostCorophiumbehaviortoincrease transmissiontothesandpiper,Calidrispusilla.Behav.Ecol.10,351–357.

Milinski,M.,1985.Riskofpredationofparasitizedsticklebacks(Gasterosteus aculeatusL.)undercompetitionforfood.Behaviour93,203–216.

Moore,J.,1983.Responses ofanavianpredatoranditsisopodpreytoan acanthocephalanparasite.Ecology64,1000–1015.

Moore,J.,2002.ParasitesandtheBehaviourofAnimals.OxfordUniversity

Press,Oxford.

Mouritsen,K.N.,Jensen,K.T.,1997.Parasitetransmission betweensoft- bottominvertebrates:temperaturemediatedinfectionratesandmortality inCorophiumvolutator.Mar.Ecol.Progr.Ser.151,123–134.

Mouritsen,K.N.,Poulin,R.,2005.Parasiteboostsbiodiversityandchanges animalcommunitystructurebytrait-mediatedindirecteffects.Oikos108,

344–350.

Nelis,H.J.C.F.,Lavens,P.,Moens,L.,Sorgeloos,P.,Jonckheere, J.A.,Criel, G.R.,DeLeenheer,A.P.,1984. cis-Canthaxanthins,unusualcarotenoidsin theeggsandthereproductive systemoffemalebrineshrimpsArtemia.J. Biol.Chem.259,6063–6066.

Nelis,H.J.C.F.,Lavens,P.,VanSteenberge,M.M.Z.,Sorgeloos,P.,Criel, G.R.,DeLeenheer,A.P.,1988.Quantitativeandqualitativechanges inthe carotenoidsduring developmentofthebrineshrimpArtemia.J.LipidRes.

29,491–499.

Ness,J.H.,Foster,S.A.,1999.Parasite-associatedphenotypemodificationsin threespinestickleback.Oikos85,127–134.

Persoone,G.,Sorgeloos,P.,Roels,O.,Jaspers,E.(Eds.),1980.TheBrineShrimp Artemia.vol.III.Ecology,Culturing,UseinAquaculture. UniversaPress, Wetteren,Belgium.

Ponton,F.,Biron,D.G.,Moore,J.,Møller,A.P., Thomas,F.,2006.Facultative virulence:astrategytomanipulatehostbehaviour?Behav.Process72,1–5. Poulin,R.,1995.‘Adaptive’changesin thebehaviourof parasitizedanimals:a

criticalreview.Int.J.Parasitol.25,1371–1383.

Poulin,R.,1998.EvolutionaryEcologyofParasites:FromIndividualstoCom- munities.ChapmanandHall,London.

Poulin,R.,Fredensborg,B.L.,Hansen,E.,Leung,T.L.F.,2005.Thetruecostof hostmanipulationbyparasites.Behav.Process68,241–244.

Reeve,M.R.,1963a.The filterfeedingArtemia.I.Inprecultureofplant cells.

J.Exp.Biol.40,195–205.

Reeve,M.R.,1963b.Thefilter feedingArtemia.II.Insuspensionofvarious particles.J.Exp.Biol.40,207–214.

Rhode,S.C.,Pawlowski,M.,Tollrian,R.,2001.Theimpactofultravioletradia- tionontheverticaldistributionofzooplanktonofthegenusDaphnia.Nature

412,69–72.

Rigaud,T.,Haine,E.R.,2005.Conflictbetweenco-occurringparasitesasa confoundingfactorinmanipulationstudies?Behav.Process68,259–262.

Robert,F.,Gabrion,C.,1991.Cestodosesdel’avifauneCamarguaise.Roˆle d’Artemia(Crustacea,Anostraca)etstrate´giesderecontrehoˆte-parasite. Ann.Parasitol.Hum.Comp.66,226–235.

Sa´nchez,M.I.,Georgiev,B.B.,Nikolov,P.N.,Vasileva,G.P.,Green,A.J.,2006a.

Redandtransparentbrineshrimps(Artemiaparthenogenetica):comparative studyoftheircestodeinfections.Parasitol.Res.,doi:10.1007/s00436-006-

0248-2.

Sa´nchez,M.I.,Green,A.J.,Castellanos,E.M.,2006b.Temporalandspatialvari- ationofanaquaticinvertebratecommunitysubjected toavianpredation at theOdielsaltpans(SWSpain).Arch.Hydrobiol.166,199–223.

Siegel,S.,Castellan,N.J.,1988.NonparametricStatisticsfortheBehavioral

Sciences,seconded.McGraw-Hill,NewYork.

Smith,R.S.,Kramer,D.L.,1987.Effectsofacestode(Schistocephalussp)onthe responseofninespinesticklebacks(Pungitiuspungitius)toaquatichypoxia. Can.J.Zool.65,1862–1865.

Thie´ry,A.,Robert,F.,Gabrion,C.,1990.Distributiondespopulationsd’Artemia etdeleurparasiteFlamingolepis liguloides(Cestoda,Cyclophyllidea), danslessalinsdulittoralme´diterrane´enfranc¸ais.Can.J.Zool.68,2199–

2204.

Thomas,F.,Renaud,F.,2001.Microphalluspapillorobustus(Trematoda):a reviewofitseffectsinlagoonecosystems.Rev.Ecol.(TerreVie)56,

147–156.

Thomas,F.,Lambert,A.,DeMeeuˆs,T.,Ce´zilly,F.,Renaud,F.,1995.Influence ofMicrophallushoffmanni(Trematoda,Microphallidae)onthesurvival, sexualselection,andfecundityofGammarusaequicauda(Amphipoda). Can.J.Zool.73,1634–1639.

Thomas,F.,Renaud,F.,Poulin,R.,1998.Exploitationof manipulators:’hitch- hiking’asaparasitetransmissionstrategy.Anim.Behav.56,199–206.

Thomas,F.,Poulin,R.,deMeeu¨s, T.,Gue´gan,J.F.,Renaud,F.,1999.Parasites andecosystemengineering:whatrolescouldtheyplay?Oikos84,167–171. Thomas,F., Adamo,S., Moore,J.,2005. Parasiticmanipulation:whereare we

andwhereshouldwego?Behav.Process68,185–199.

Thompson, R.M.,Mouritsen, K.N.,Poulin,R.,2005.Importance ofparasites andtheirlifecyclecharacteristics indeterminingthestructureofalarge marinefoodweb.J.Anim.Ecol.74,77–85.

Trabalon, M.,Plateaux, L.,Peru,L.,Bagneres, A.,Hartmann,N.,2000.Mod- ification ofmorphologicalcharactersandcuticularcompoundsinworker antsLeptothorax nylanderiinducedbyendoparasitesAnomotaeniabrevis. J.Insect.Physiol.46,169–178.

Valkounova, J.,1983.Biologyofcestodesofdomesticducksandwildwater birds.Vet.Med.(Praha)28,549–563.

Varo´,I.,Taylor,A.C.,Navarro,J.C.,Amat,F.,2000.Effectofparasitismon respirationratesofadultsofdifferentArtemiastrainsfromSpain.Parasitol. Res.86,772–774.

Verkuil,Y.,VanderHave,T.M.,VanderWinden,J.,Chernichko,I.I.,2003.

HabitatuseanddietselectionofnorthwardmigratingwadersintheSivash (Ukraine):theuseofbrineshrimpArtemiasalinainavariablysalinelagoon complex.Ardea91,71–83.

Webber,R.A.,Rau,M.E., Lewis, D.J.,1987. Theeffects ofPlagiorchisnoblei (Trematoda:Plagiorchiidae)metacercariaeonthebehaviorofAedesaegypti larvae.Can.J.Zool.65,1340–1342.

Wellnitz, T.,Giari,L.,Maynard, B.,Dezfuli,B.S.,2003.Aparasitespatially structuresitshostpopulation.Oikos100,263–268.

Williams,C.M.,Poulin,R., Sinclair,B.J.,2004.Increasedhaemolymphosmo- laritysuggestsanewrouteforbehaviouralmanipulationofTalorchestia quoyana(Amphipoda:Talitridae)byitsmermithidparasite.Funct.Ecol.

18,685–691.

Yan,G.,Stevens,L.,Schall,J.J.,1994.BehavioralchangesinTriboliumbeeteles infectedwithatapeworm: variationineffectsbetweenbeetlespeciesand amonggeneticstrains.Am.Nat.143,830–847.