ThealluringsimplicityandcomplexrealityofgeneticrescueDavidA.Tallmon,Go - PDF document

ThealluringsimplicityandcomplexrealityofgeneticrescueDavidA.Tallmon,GordonLuikartandRobinS.WaplesLaboratoired’EcologieAlpine,Ge´nomiquedesPopulationsetBiodiversite´,CNRSUMR5553,Universite´JosephFourier,BP53,38041Grenoble,Cedex09,FranceNationalMarineFisheriesService,NorthwestFisheriesScienceCenter,2725MontlakeBoulevardEast,Seattle,WA98112,USAAseriesofimportantnewtheoretical,experimentalandobservationalstudiesdemonstratethatjustafewimmigrantscanhavepositiveimmediateimpactsontheevolutionarytrajectoryoflocalpopulations.Inmany interactionsamonggenesatdifferentlocithatinuenceaphenotypicGeneticrescue:anincreaseinpopulationtnessowingtoimmigrationofnew Correspondingauthor:DavidA.Tallmon(dtallmon42@yahoo.com).Availableonline20July20040169-5347/$-seefrontmatterCrownCopyright2004PublishedbyElsevierLtd.Allrightsreserved.doi:10.1016/j.tree.2004.07.003 TRENDSinEcologyandEvolutionVol.19No.9September2004 residentstothenraisetheoverallmeanÞtnessofthelocalpopulation.Ideally,geneticrescueismeasuredbyanincreaseinpopulationgrowthrateovermultipleTheincreaseinÞtnessisthoughttobedueprimarilytoHETEROSISintheoffspringthatresultfrommatingsbetweenimmigrantsandlocalindividuals.Heterosisoccursviatwomechanisms.First,immigrantallelescanmaskdeleterious,recessiveallelesthathaveachievedahighlocalfrequencyviageneticdrift,whichcanoccurwhentheproductofpopulationsize()andtheselectioncoefÞcient(s)islessthanone(1).Second,matingsbetweenimmigrantsandlocalindividualsproducehighlyheterozygousoffspring,whichareoftenfavoredbynaturalselection.Therecentliteraturesuggestsmaskingofdeleteriousallelesisthemoreprevalentmechanismofof.TounderstandhowimmigrationcaneitherincreaseorreducepopulationÞtness,itisnecessarytoconsidermultiplegenelociovermultiplegenerations(Figure1TheF1generationwillhavehighheterozygosity(and,hence,heterosis)asaresultofallelefrequencydifferencesbetweenparentaltypes.Positiveepistasiswillalsobemaintainedbecauseonesetofchromosomesfromeachparentallineageremainsintact.However,immigrationwillalsodecreaseLOCALADAPTATIONbydilutinglocalallelesthatmakepositiveadditivecontributionstoÞtness.Therefore,thenetÞtnesseffectsofimmigrationintheF1generationcanbeeitherpositiveornegative,depend-ingonwhetherthepositiveeffectsofheterosismorethancompensateforthedilutionoflocaladaptation.Insubsequentgenerations(F2andbeyond),recombi-nationwilldisruptpositiveepistaticinteractionsamongparentalallelesatdifferentloci(Figure1),thusreducingINTRINSICCOADAPTATION.Iftheimmigrationrateishighandfromageneticallydivergentsource,populationÞtnesscanbereducedintheF2generationandbeyondasimmi-grantsdilutelocallyadaptiveallelesanddisruptco-adaptedgenecomplexes.Furthermore,heterozygosityandassociatedheterosispeaksintheF1generationanddeclinesthereafter.Asaconsequence,descendantsofimmigrantsoftenexhibitinitialheterosisfollowedbysubsequentoutbreedingdepressioninfollowinggener-ations(Box1),sowhetherapopulationisrescuedgeneticallybyimmigrantslargelydependsontherelativeimportanceoftheseopposingphenomena.ObservationalstudiesofgeneticrescueSeveralrecentstudiesreportincreasedÞtnessinresponsetolowlevelsofimmigration(geneßow)intopopulationsthathavesufferedrecentdeclines(Table1).Forexample,immigrantsapparentlyincreasedthehatchingrateofimperiledprairiechickensTympanuchuscupidocupidoandreducedtheproportionofstillbornbirthsinarelictpopulationofaddersViperaberus,whileincreasingmoleculargeneticvariation,recruitmentandpopulationgrowthraterate.Asingleimmigrantisthoughttohavespurredthegrowthofastagnant,recentlyre-foundedScandinavianwolfCanislupuslupus.Thesestudies,whichencompassawiderangeofvertebratetaxa,suggestthatimmigrationcanincreasetheÞtnessofsmall,inbredpopulationsbyrestoringadaptivegeneticvariation.However,becauseeachofthesestudieswaslimitedtoasingle,unreplicatedpopulationwithoutexperimentalcontrols,itisimpossibletoattributeunequivocallytheÞtnessincreasestogeneticrescueratherthantootherpotentiallyconfoundingfactors,suchasbenignchangesinthelocalenvironmentornon-geneticinßuencesofimmigrants.Forexample,itispossiblethattheScandinavianwolvesweresocloselyrelatedtheysimplyavoidedbreedingforbehavioralreasonsuntiltheunrelatedimmigrantarrivedarrived.ExperimentalstudiesofgeneticrescueOtherrecentstudieshaveisolatedthegeneticinßuencesofimmigrantsbycontrollingfortheirdemographicimpacts,andtheysupportthepremisethatimmigrationcanincreasepopulationÞtnessÞtness.ThemoststrikingexampleofgeneticrescueisfromawaterßeametapopulationinhabitingBalticseashorerock-rock-.Inreplicatepopulations,theresearchersmatedresidentswithresidents,residentswithimmigrantsand TRENDS in Ecology & Evolution AABBimmigrantIndividual levelGenerationF1F2ParentalAABBAAB
B
A
A
B
B
A
A
BB
AA
B
B
AA
BB
AABB
A
A
BB A
ABB
AA
BBOnly A
A
B
B
genotype Population level A AB B
A
B
B
andAA
BB
genotypes genotypes Genotypes ofimmigrant descendants B xImmigrant A AB B A Figure1.Geneticrescue.Geneticrescuedependsuponwhetherselectionfavorsthenewgenotypescreatedbyimmigration.Thematingofageneticallydivergentimmigrant(homozygousforonesetofallelesAandBatloci1and2,respectively)withanindividualfromalocalpopulation(homozygousfordifferentallelesA)producesF1offspringthatareheterozygousandaddsgeneticvariationtothepopulation.BytheF2generation,thismatinggeneratestenpossiblegenotypesattwoloci.Inrealpopulations,therearemanymorelociandnovelgenotypesgeneratedbyimmigrationevents. TRENDSinEcologyandEvolutionVol.19No.9September2004 immigrantswithimmigrants.RepeatedgenerationsofasexualreproductionlockedinplaceF1genotypes.Becauseasexualreproductionpreventedsegregationofalleles,selectioncouldactovermanygenerationsonintactresident,F1hybrid,andpureimmigrantgenomesandamplifyÞtnessdifferencesamonglineagesovertime.Ultimately,thehybridF1clonallineageshadanaverageÞtnessthatwasover35timesthatofresidentlineages.ThisstudydemonstratesthatinbreedingdepressionisabiologicallyimportantphenomenoninthiswaterßeametapopulationandthatimmigrationcanincreaseÞtnessgreatly.Becauseentirelifecycleswereobserved,ÞtnessimpactswereintegratedoverthefulllifehistoryandampliÞedovertime.Bycontrast,moststudiesinferÞtnesseffectsofexperimentaltreatmentsononlyoneortwotraitsmeasuredinasinglegeneration.Selectionintensityonindividuallifehistorytraitscanvarygreatlyovertimeandspace,andthecumulativeeffectsofselectiononmultipletraitswillinteracttoproduceoverallÞtnesseffects.Thisimpliesthatshort-termstudiesofafewtraitscanresultinmisleadingconclusions.Forexample,Bryantetal.foundastronggeneticrescueeffectonlarvalemergenceinexperimentalhouseßyMuscadomesticapopulations,butonlyafteraninitialÞve-generationperiodinwhichlowlevelsofimmigrationprovidednodetectablebeneÞtstorecipientpopulationspopulations.TworecentplantstudiesprovidefurtherevidenceoftheÞtnessbeneÞtsimmigrantscanprovidetoinbredpopu-lations.InexperimentallyinbredpopulationsofthemustardBrassicacampestris,oneimmigrantpergenerationsigniÞcantlyincreasedtheÞtnessoffouroutofsixÞtnesstraitsintreatmentpopulationscomparedwith(noimmigrant)controlscontrols.NoÞtnessdifferencewasfoundbetweentheone-immigrantand2.5-immigranttreatments,whichsuggeststhatonlyverylowlevelsofimmigrationarenecessaryforaÞtnessboost.However,greaterphenotypicdivergenceamongpopulationswasfoundintheone-immigranttreatmentcomparedwith2.5-immigranttreatment.Thisisinterestingbecauseitshowsthatlowerimmigrationcanfacilitatelocaladaptationinspatiallystructuredpopulationsthataresubjecttovariableordivergentselectionpressures.InsmallwhitecampionSilenealbapopulations,geneßowincreasedgerminationsuccessandthesuccessofimmigrantpollencorrelatedpositivelywiththeamountofinbreedinginrecipientpopulationspopulations.Thisworkdemonstratesinbreedingdepressioninthesepopulationsandsuggeststhat,allelsebeingequal,populationsthataremoreinbredaremorelikelytobeneÞtfromageneticrescueeffect.Togetherwiththestudyandothersothers,thisimpliesthatverylowlevelsofimmigra-tioncouldcausegeneticrescueininbredpopulationswithoutgreatlyconstraininglocaladaptation.Fortu-nately,itisincreasinglypossibletoidentifyrecentlyinbredpopulationsusingnewmolecular-basedstatisticalstatistical,andsocandidatepopulationsforgeneticrescueshouldbeeasiertoÞnd.RescueviaothergeneticpathwaysImmigrantscanalsoincreaseÞtnessviamechanismsotherthanheterosisheterosis.Onesuchmechanismisfrequency-dependentselectionforrarealleles.Insmallpopulationsofself-incompatibleplantswithfewallelesattheS-locus(whichinhibitspollinationbydonorswiththesameS-alleletypeasthematernalplant),immigrantpollenwithnovelS-allelescansubstantiallyincreasereproductivesuccesssuccess.ThereisalsoevidenceinHymenopterathatfrequency-dependentselectionfornovelsex-determiningallelesintroducedintobottle-neckedpopulationsbyimmigrantswouldraisecolonyÞtnessbyincreasingthefrequencyofdiploidheterozy-gotes(females,whichcontributetocolonysuccess),attheexpenseofdiploidhomozygotes(males,whichdrainrain.GemmellandcolleagueshypothesizethatfemaleimmigrantsmightalsobeneÞtsmallpopulationsbyintroducingnewmitochondrialDNA(mtDNA)alleles,whicharematernallyinheritedinmostspeciesspecies.TheysuggestthatmtDNAalleleswithnegativeÞtnesseffects(e.g.onmalefertility)wouldbecomeÞxedinsmallpopulations,becauseselectionactslessefÞciently(anddifferently)onmtDNAalleles,owingtotheirmaternalinheritance,thanitdoesonnuclearallelesalleles.Inthisscenario,immigrantfemalesthatpassonsuperiormtDNAalleleswouldboostpopulationÞtnesswiththeirextra-nucleargeneticcontributions.InferencesformetapopulationconservationTheupshotoftheserecentstudiesisthatgeneticrescuecouldbecruciallyimportanttoentiremetapopulationsbyreducinglocalinbreedingdepressionoraddingadaptive Box1.Heterosisandoutbreedingdepression Initialheterosis,followedbyoutbreedingdepression,hasbeenobservedrecentlyintaxaasdiverseasangiosperms[62,63]],insectsnsects,birdsirdsandmammalsammals.InthemarinecopepodTigriopuscalifornicus,Burton,EdmandsandcolleaguesfoundheterosisinF1hybridsfollowedbyF2outbreedingdepressionlevelsthatcorrelatedwiththegeneticdivergencebetweensourceandrecipientpopulations[64,68,69].InFensterandGalloway’sstudiesofthepartridgepeaChamaecristafascicu-,F1crossesofindividualsfrompopulationsatalldistancesstudied(0.1–2000km)showedheterosis,thusdemonstratinginbreedingdepressionwithinpopulationspulations.Subsequently,outbreedingdepressionwasfoundincrossesfrompopulationsseparatedby1000km,butnotuntiltheF3generation.BecauseunlinkedgenesbeginrecombiningintheF2generation,theappearanceofconsistentoutbreedingdepressiononlyintheF3generationdemonstratesthatdisruptionofepistasisasaresultoflinkedgenescancontinuetoreducetnessbeyondtheF2generation.Evidencealsosuggeststhatcytonuclearco-adaptations(i.e.thecoadaptationofnuclearandcytoplasmicorganellegenes)cancontributetooutbreedingdepressiondepression.Inaremarkableexampleofintrinsiccoadaptation,Gharrettetal.foundoutbreedingdepressionincrossesoftwopinksalmonOncorhynchusgorbuschapopulationsthatspawninthesamelocation(and,hence,experienceonaveragethesameenvironment)butareisolatedtemporallybecauseofstricteven-orodd-yearlifelife.Inter-yearcrossesshowedoutbreedingdepressionintheF2(butnotF1)generation,whichprovidesevidencefordisruptionofpositiveepistasisfollowingrecombination.Thesetheoreticalresultshelptoexplainthateventhesameselectionpressures(onaverage)overmanygenerationsdonotguaranteeasimilargeneticarchi-tecture,which,inturn,suggeststhatextensivemixingofpopu-lationsexperiencingsimilarselectiveregimesmightincreasetnessinrealpopulations. TRENDSinEcologyandEvolutionVol.19No.9September2004491 allelesthatincreasetheprobabilityoflocalpopulationpersistence.Inturn,thisbolsteringoflocalpopulationsmaintainsabroadgeographicaldistributionofpopu-lationsthatbuffersoverallmetapopulationextinctionandprovidesfutureimmigrantsforotherpopulationspopulations.Ageneticrescueeffectcanimprovethelikelihoodofsuccessofcolonizationeventsthatmightotherwisefailbecauseimmigrantscanreinforcecolonizingpopulationsthatoftenevolvefromafewfoundersandquicklybecomebecome.Fromaconservationperspective,thisalsomeansthatgeneticrescuemighthaveanimportantimpactinthespreadofunwantedinvasivespeciesalongtheleadingedgeofinvasionbysupplyingrecentlyestablished,smallpropaguleswithadequategeneticvariationtorespondtoselectionandadapttothenewnew.Therefore,ausefulstrategyforredu-cingthespreadofinvasivepopulationscouldbetominimizethechancesforgeneticrescuetooccur.AlthoughthetermÔgeneticrescueÕisnew,theunder-lyingconceptsarenot.Similartomanyotherusefulconceptsinevolution,thethreadofgeneticrescuecanbetracedbacktoCharlesDarwinandSewallWright(Box2RecenttheoryhasreÞnedourunderstandingofgeneticrescueandshownthatheterosisismaximizedwhenlocalpopulationsizeissmall,geneßowislowandselectionintensityisloworintermediateintermediate.Theseresultsaregermane,becausereductionsinlocalpopulationsize(and,hence,greaterdriftandlessefÞcientnaturalselection)andrestrictionsingeneßowaretwooftheprimaryconsequencesoftheongoingworldwidehabitatfragmen-tationcrisiscrisis.Thissuggeststhatanthropogenicinßuenceswillincreasethenumberofcaseswheregeneticrescuewillbeanimportantevolutionaryphenomenonandeffectiveconservationtool.ComplicatinggeneticfactorsInspiteoftheseexcitingresults,severalimportantcaveatsapplytothelaboratoryandÞeldstudiesreportinggeneticrescuethatshouldtempertheuseofgeneticrescueasamanagementtool.First,mostlaboratorystudieshaveusedhighlyinbredlinesorpopulations.Thelevelofinbreedingtypicallywasequivalenttofull-sibmatingfortwogenerationsormore,soitisnotsurprisingthatmatingwithanimmigrantwouldprovideÞtnessbeneÞts.ItisdifÞculttotranslatetheseexperimentalresultsdirectlytonaturalpopulationsinwhichinbreedinglevelsareusuallyunknownandunlikelytobeashigh.Second,theexperimentalpopulationsandimmigrantsareusuallyderivedfromasinglesourcepopulation.Giventhatimmigrantsshareacommongeneticarchitecturewiththerecipientpopulation,thechancesofoutbreedingdepressionarereduced.Finally,moststudieshave Table1.RecentempiricalstudiesshowinggeneticrescueeffectsSpeciesStudyContextImmigrantsourcePrimaryresultsRefsWatereaExpFieldandlaboratoryNearbypopulationF1PopulationgrowthratemanytimesgreaterinoutbredthanininbredpopulationsnsHouseyExpLaboratorypopulationsSamepopulationF20PupalemergenceinimmigrantpopulationsgreaterthancontrolsntrolsFruitymelanogasterExpLaboratorypopulationsSamepopulationF3ReproductivetnessinimmigrantpopulationstwicethatofthecontrolsontrolsFlowerbeetleTriboliumExpLaboratorypopulationsSamepopulationF20Populationgrowthratehigherinpopulationsreceivingimmigrantsfromhightness(butnotlowtness)populationsnsAdderViperaberusObsIsolatedpopulationDistantpopulationF1Populationgrowthrateincreased,stillbornbirthratedecreaseddecreasedPrairiechickennuchuscupidoObsIsolatedpopulationDistantpopulationsF1EggviabilityincreasedeasedSongsparrowpizamelodiaObsIsolatedpopulationUnknownpopulationF2F1tnesselevated,butF2tnesstnessScandinavianwolfCanislupusObsIsolatedpopulationUnknownpopulationF2PopulationgrowthrateincreasedncreasedPlantMustardBrassicacam-ExpOutdoorpopulationsSamepopulationF6Immigrantpollenincreased4/6t-nesstraitstraitsWhitecampionExpGreenhousestudyofeld-collectedplantsNearbypopulationsF1Immigrantpollenincreasedgermina-tionsuccessinperipheralpopulationspopulationsScarletgiliaExpIsolatedpopulationsDistantpopulationF1Insmallpopulationonly,seedmassandgerminationincreasedbyimmi-grantpollenpollenYellowpitcherplantSarraceniaßavaExpIsolatedpopulationsDistantpopulationF1PlantheightincreasedbyimmigrantmmigrantScurvygrassExpOutdoorpopulationsNearbypopulationsF1SeedmassandplantsizeincreasedbyimmigrantpollenpollenaOnlystudiesthatshowpopulationtnessresponsestotreatmentsareincluded.Obs,observational;Exp,experimental.Sourcepopulationforimmigrantindividuals.Inmanyexperiments,immigrantsarerecentdescendantsfromsamesourceasrecipientpopulations.Inferredstudylengthingenerationsfromrstimmigrationevent,sometimesnotexplicitlystatedinstudies. TRENDSinEcologyandEvolutionVol.19No.9September2004 followedresultsonlythroughtheF1generation,whereheterosisismaximizedandoutbreedingdepressionisoftennotyetexpressed.Still,althoughthereisnodoubtthatimmigrationfromamaladaptedsourcecanreduceÞtnessandconstrainlocallocal,theideathatlimitedimmigrationcanplayamorepositiveroleinevolutionhasreceivedrenewedattentionattention.AkeytoresolvingthegeneralimportanceofgeneticrescueisunderstandingtherelativeimportanceofinteractionsamonggenesversustheadditiveeffectsofindividualgenesindeterminingtheÞtnessofpopulationsÐadebatethatdatesbacktoFisherandWrightandremainsanareaofactiveresearch.Somerecentexperimentalresultssuggestthat,iffavoredallelesareaddedtoapopulation,theywillspreadbecauseselectioncanefÞcientlychooseadvantageousallelesregardlessofthegeneticbackgroundbackground.Thisimpliesthatgeneinteractionsmightnotbeasimportantassinglelocus,additivecontributionstoÞtnessandthat,undersomecircumstances,atleast,selectioncantakecareofthingsifgivenenoughtimeandafewcopiesofadaptiveallelestoactupon.Thus,immigrantscanprovideusefulallelesthat,overlongtimescales,contributetoincreasedpopulationÞtnesseveniftheimmediatedescendantsofimmigrantsdonot.Complicatingnon-geneticfactorsGiventhatindividualsarenotboxesofgenes,demo-graphic,behavioralandenvironmentalfactorscanpro-foundlyinßuencetheÞtnesscontributionsofimmigrantsBox3).Demographictheoryhaspredicted,andempiricalstudieshavedemonstrated,thatindividualsofparticularagesorlifestagescanhavedisproportionateeffectsonthetrajectoryofapopulation.Usingstandarddemographicdataandanalysesanalyses,onecanpredicttherelativeimpactsofimmigrantsbyestimatingtheirfuturerepro-ductivepotential.Forexample,immigrantshavelittlechanceofprovidinggeneticrescueiftheyareimmatureandhavealowchanceofsurvivaltoreproductiveage,oraretoooldandhavelittlechanceoffuturereproduction.Jointconsiderationofdemographyandgeneticsinasingletheoreticalframeworkcansharpenpredictionsofoptimalconditionsforgeneticrescuerescue.However,subtletiesofhowdemographyisincorporatedintopopulationgeneticmodelscanalterconclusions,socarefulexaminationofassumptionsisnecessarynecessary.Evenarigorousinvestigationofgeneticanddemo-graphicfactorsmightbeinsufÞcienttopredictreliablywhethergeneticrescuewilloccur,becauseimportantbehavioralsubtletiessuchasmatechoice,dominancehierarchiesandinfanticide,canalsoinßuencetheevol-utionaryimpactofimmigrants.Forexample,femalemiceMusdomesticusprefergeneticallydissimilarmales,whichwouldfavorimmigrantmalesandfacilitategeneticrescue.However,theyalsoassessmalequality,inadditiontogeneticdissimilarity,inchoosingamongpotentiall,whichcomplicatespredictionsofhowwellimmigrantmalesmightreproduce.Furthermore,insomespecies,asuccessfulmaleimmigrantmightkillexistingjuveniles,asoccurs,forexample,withimmigrantadultmalebrownbearsfollowingremovaloflocaladultmalesbyhuntershunters.Ingeneral,immigrantscouldhavewidelyvaryingreproductivesuccessdependinguponinteractionswithlocalindividuals.Thus,theycouldprovideeitherno,oragreatdeal,ofgeneßow,whichmakestenuousstudiesuninformedbybehavioralormatingsystemdatadata.Immigrantsmightalsoserveasvectorsfordisease-causingparasitesandpathogens,thusleadingtotheoppositeofarescueeffectfect.LittleempiricalevidenceisavailabletoindicateunderwhatconditionsthegeneticrescuebeneÞtsoflowlevelsofimmigrationoutweightheincreasedriskofdiseasetransfer.Fromtheaboveconsiderations,itisclearthat,evenifoneconsidersonlybiologicalconsequencesforthespeciesofinterest,evaluatingandpredictingtheconsequencesofageneticrescueattemptisexceedinglycomplex.Inaconservationcontext,biologicaltheorycanprovideinsights,butbroaderconsiderationsarealsolikelytoberelevant.Forexample,geneticrescueeffortsforsomeat-riskspecieshaveeconomic,social,legalandpoliticalramiÞcations.Integratingthesetypesofconsiderationintoanoverallcost-beneÞtanalysiswillbechallengingbutbut.FuturestudiesAnagginglimitationishowlittleweknowaboutthejointimpactsofinbreedingandoutbreedingdepressionin Box2.Theconceptualfoundationsofgeneticrescue TheconceptofgeneticrescuewaspresagedbyCharlesDarwin,who,inhisextensivesurveysofempiricaldatatobetterunderstandevolutionaryprocesses,notedthat(i)isolatedpopulationsoftensufferfrominbreedingandinbreedingdepression,andthat(ii)immigration(‘newblood’)canhelprecoverapopulation’stnesstness.AlthoughDarwin’sobservationsareover100yearsoldandreportinsightsfromhusbandrythatprobablyextendbackmillenniatotheearlydaysofdomestications,theorytounderstandtheseobservationsisstillbeingTheseminaltheoreticalworkrelevanttogeneticrescuewasproposedbySewallWright,whowasinterestedintheeffectsofimmigrationamongpopulationslinkedbygeneow.Heshowedthatonemigrantpergenerationamongpopulationsequallylinkedbyimmigrationwasadequatetokeepthesameneutralallelessegregat-inginallpopulations,andthattherelativestrengthofgeneowandselectiondeterminedthefateofnon-neutralalleleslleles.Wright’scontributionsremainthefoundationformuchoftheongoingresearchrelatedtopopulationsubdivision,migrationandthedistributionofgeneticvariationinneutralandselectedlociandtraits.RecentmodelsbyWhitlockandcolleagueshaveexpandedWright’sworktoincludeheterosisinimmigrantoffspringfspring.Theseeffortsdemonstratethat,assmallisolatedpopulationsdiverge,theybecomexedfordifferentdeleteriousallelesatdifferentloci.Asaresult,immigrantsbringinallelesthatmasklocaldeleteriousalleles,immigrantoffspringshowhighertnessthanlocaloffspring,andtheeffectiveimmigrationrate(i.e.geneow)iselevatedoverthatexpectedfromneutraltheoryasimmigrantallelesincreaserapidlyinfrequencyowingtoselection.Theseresultshelptoexplainsomeoftherecentremarkableresultsfromthelaboratoryandeld,eventhoughtheyarebaseduponsimplesingle-locuspopulationgeneticmodelsanddonotincludeheterozygoteadvantageorothergeneticmechanismsthatmightincreasetness. TRENDSinEcologyandEvolutionVol.19No.9September2004493 naturalpopulations,exceptthattheyprobablyoccurfrequentlyandsimultaneouslysimultaneously.AkeyevolutionaryquestioniswhethertheheteroticeffectsofafewsuccessfulimmigrantswilloutweighreductionsinÞtnessowingtooutbreedingdepression.SusceptibilitytothesetwofactorsvariesconsiderablyamongtaxaandspeciÞccircum-stances,includingpopulationhistoryandrecentinbreed-inglevels,degreeoflocaladaptationandgeneticdivergencebetweenimmigrantsandrecipientpopu-lations.Giventhepaucityofdataaddressingthisissueforwildorexperimentalpopulations,itisdifÞculttopredictwithanycertaintywhetherthenetgeneticeffectswillbepositiveornegativeforanyunstudiedsystem.Thisisasoberingtruthforappliedconservationefforts.However,empiricaldatadodemonstratethatlowlevelsofimmigrationcanprovideconsiderableÞtnessbeneÞtstorecentlyisolated,inbredpopulations,andrecentmolecu-lar-basedstatisticaladvanceswillhelptoidentifythesepopulationsefÞcientlyÞciently.Adauntingchallengeistodevelopageneraltheoryofrescueeffectsthatincorporatesgenetics,inadditiontodemography,behavioranddiseaseecology;allfactorsthatcanbeshowntobeofpreeminentimportanceinspeciÞccases.Thecruxofthechallengeisformulatingtheoriesthataresimpleenoughtobetestedempiricallyandthatareusefulbeyondonlynarrowconditions.Thereisavitalroleformulti-generationexperimentsindevelopingthistheory,aswellasforcreativeusesofmeta-analysestodistillcollectiveinsightsfromnumerousimperfectlycontrolledlaboratoryandnaturalÔexperimentsÕ.WethankJeffHard,JackHogg,LaurentKeller,andespeciallyFredAllendorfforhelpfulcomments.ThisarticlewaswrittenwhileD.A.T.wassupportedbyNSF(INT0202707)andG.L.wassupportedbyEuropeanCommission(QLK5-CT-2001Ð02461).R.S.W.contributedwhileavisitingscientistatLaboratoiredÕEcologieAlpine.1Grant,P.R.andGrant,B.R.(2002)Unpredictableevolutionina30-yearstudyofDarwinÕsÞnches.296,707Ð7112Brown,J.andKodric-Brown,A.(1977)Turnoverratesininsularbiogeography:effectofimmigrationonextinction.58,445Ð4493Stephens,P.A.andSutherland,W.J.(1999)ConsequencesoftheAlleeeffectforbehaviour,ecology,andconsequences.TrendsEcol.Evol.4Courchamp,F.etal.(1999)InversedensitydependenceandtheAlleeeffect.TrendsEcol.Evol.14,405Ð4105Caughley,G.(1994)Directionsinconservationbiology.J.Anim.Ecol.63,215Ð2446Keller,L.F.etal.(1994)Selectionagainstinbredsongsparrowsduringanaturalpopulationbottleneck.372,356Ð3577Keller,L.F.(1998)InbreedinganditsÞtnesseffectsinaninsularpopulationofsongsparrows(Melospizamelodia52,240Ð8Newman,D.andPilson,D.(1997)Increasedprobabilityofextinctionduetodecreasedgeneticeffectivepopulationsize:experimentalpopulationsofClarkiapulchella51,354Ð3629Saccheri,I.J.etal.(1998)Inbreedingandextinctioninabutterßymetapopulation.392,491Ð49410Thrall,P.H.etal.(1998)Metapopulationcollapse:theconsequencesoflimitedgene-ßowinspatiallystructuredpopulations.InSpatiotemporalDynamicsinEcology(Bascompte,J.andSole,R.V.eds),pp.83Ð104,SpringerVerlag11Richards,C.M.(2000)Geneticanddemographicinßuencesonpopulationpersistence:geneßowandgeneticrescueinSilenealbaGenetics,DemographyandViabilityofFragmentedPopulations(Young,A.G.andClarke,G.M.eds),pp.271Ð291,CambridgeUniversityPress12Ingvarsson,P.K.(2001)RestorationofgeneticvariationlostÐthegeneticrescuehypothesis.TrendsEcol.Evol.16,62Ð6313Gaggiotti,O.E.andHanski,I.(2004)Mechanismsofpopulation Box3.Thebiologicalcontextofgeneticrescue Inapplyingtheconceptofgeneticrescuetowildpopulations,muchinformationisneeded.Ideally,detailedbehavioral,demographic,environmentalandgeneticinformationwouldbeavailabletoprovideacomprehensiveframeworkforunderstandingwhenrescuemightbeachieved(FigureI).First,itisimportanttoknowhowmuchapopulationsuffersfrominbreedingdepressionandhowsuscep-tibleitistooutbreedingdepression.Thiswilldeterminetheoptimallevelofimmigrationtoeffectatnessincrease.However,thebalancebetweenthesetwoconsiderationsinwildpopulationsisdifculttoascertainbecauseitdependsuponmanyusuallyunknownfactors.ConservationeffortsonbehalfoftheendangeredFloridapantherFelisconcolor,whichhasshownstrongputativeinbreedingdepression,haveattemptedtointegratesomeofthesefactors.Hedrickusedsimplepopulationgeneticsmodelstoshowthatabriefperiodofhighgeneowfollowedbysubsequentgenerationsoflowgeneowcouldeffectgeneticrescuebyreducingthefrequencyofdeleteriousalleleswithoutsubstantiallyreducingthefrequencyoflocallyadaptivealleles[76].Knowledgeofpantherbehaviorsuggestedthatonlyfemalesshouldbeusedasimmigrantsbecausemaleswerelikelytoprecipitatemortalcombatwithresidentmales,riskingnotonlylocalpopulationdecline,butalsothefailureofimmigrantstoreproduce.Itwasalsoknownthatfemalesreadilyestablishterritoriesandbeginbreedingaftertheirrstyear.Thesebehavioral,genetic,anddemographicconsiderationswereincorporatedintoFloridapantherrecoverymanagementactionsandpreliminaryevidencesuggeststhattheinitialtargetlevelofintrogressionhasbeenachievedandthatatleastoneimportanttnesstrait,thenumberofdescendedtesticlespermale,hasincreased[77].Severalgenerationsofmonitoringwillbenecessarytoseeifthisattemptedgeneticrescueissuccessfulinboostingpopulationgrowth. TRENDS in Ecology & Evolution ¥ Sex¥ Age¥ Density dependence¥ Population growth rate ¥ Flutuations ¥ Social or mating system¥ Sexual selection Population fitness Individual fitness Outbreeding depression¥ Coadapted gene complexes Other Genetic rescue? FigureI.Someoftheimportantbiologicalfactorsthatinteracttodeterminewhethergeneticrescuewilloccur. TRENDSinEcologyandEvolutionVol.19No.9September2004 extinction.InEcology,Genetics,andEvolutionofMetapopulations(Hanski,I.andGaggiotti,O.E.eds),pp.337Ð366,ElsevierAcademic14Remington,D.L.andOÕMalley,D.M.(2000)Whole-genomecharacter-izationofembryonicstageinbreedingdepressioninaselfedloblollypinefamily.155,337Ð34815Westermeier,R.L.etal.(1998)Trackingthelong-termdeclineandrecoveryofanisolatedpopulation.282,1695Ð169816Madsen,T.etal.(1999)Restorationofaninbredadderpopulation.402,34Ð3517Vila,C.etal.(2002)Rescueofaseverelybottleneckedwolf()populationbyasingleimmigrant.Proc.r.Soc.Lond.Ser.B.270,91Ð9718Ingvarsson,P.K.(2003)Lonewolftotherescue.420,47219Ebert,D.etal.(2002)AselectiveadvantagetoimmigrantgenesinaDaphniametapopulation.Science295,485Ð48820Bryant,E.H.etal.(1999)Experimentaltestsofcaptivebreedingforendangeredspecies.Conserv.Biol.13,1487Ð149621Newman,D.andTallmon,D.A.(2001)ExperimentalevidenceforthebeneÞcialÞtnesseffectsofgeneßowinrecentlyisolatedpopulations.Conserv.Biol.15,1054Ð106322Richards,C.M.(2000)Inbreedingdepressionandgeneticrescueinaplantmetapopulation.Am.Nat.155,383Ð39423Spielman,D.andFrankham,R.(1992)Modelingproblemsinconservationgeneticsusingpopulations:improvementofreproductiveÞtnessduetoimmigrationofoneindividualintosmallpartiallyinbredpopulations.ZooBiol.11,343Ð35124Richards,C.M.etal.(1999)TheinßuenceofpopulationsizeandisolationongeneßowbypolleninSilenealba53,63Ð7325Heschel,M.S.andPaige,K.N.(1995)Inbreedingdepression,environ-mentalstress,andpopulationsizevariationinscarletgilia,aggregataConserv.Biol.9,126Ð13326Sheridan,P.M.andKarowe,D.N.(2000)Inbreeding,outbreeding,andheterosisintheyellowpitcherplant,Sarraceniaßava,inVirginia.Am.J.Bot.87,1628Ð163327Saccheri,I.J.andBrakeÞeld,P.M.(2002)Rapidspreadofimmi-grantgenomesintoinbredpopulations.Proc.R.Soc.Lond.Ser.B269,1073Ð107828Paschke,M.etal.(2002)EffectsofpopulationsizeandpollendiversityonreproductivesuccessandoffspringsizeinthenarrowendemicCochleariabavaria(Brassicaceae).Am.J.Bot.89,1250Ð125929Beaumont,M.A.(2003)Conservationgenetics.InHandbookofStatisticalGenetics(Balding,D.J.etal.,eds),pp.751Ð792,JohnWiley&Sons30Ball,S.J.etal.(2000)Thegeneticcontributionofsinglemaleimmigrantstosmall,inbredpopulations:alaboratorystudyusingDrosophilamelanogaster84,677Ð68431Charlesworth,D.(1985)Distributionofdioecyandself-incompat-ibilityinangiosperms.InEvolutionÐEssaysinHonorofJohnMaynardSmith(Greenwood,J.J.andSlatkin,M.eds),pp.237Ð268,HarvardUniversityPress32DeMauro,M.M.(1993)RelationshipofbreedingsystemtorarityintheLakesideDaisy(Hymenoxysacaulisvar.Conserv.Biol.33Reinartz,J.A.andLes,D.H.(1994)Bottleneck-induceddissolutionofself-incompatibilityandbreedingsystemconsequencesinAsterfurcatusAm.J.Bot.81,446Ð45534Ross,K.G.etal.(1993)Effectofafoundereventonvariationinthegeneticsex-determiningsystemoftheÞreantSolenopsisinvictaGenetics135,843Ð85435Gemmell,N.J.etal.(2004)MotherÕscurse:theeffectofmtDNAonindividualÞtnessandpopulationviability.TrendsEcol.Evol.19,238Ð24436Gemmell,N.J.andAllendorf,F.W.(2001)Mitochondrialmutationsmaydecreasepopulationviability.TrendsEcol.Evol.16,115Ð11737Sakai,A.K.etal.(2001)Thepopulationbiologyofinvasivespecies.Annu.Rev.Ecol.Syst.32,305Ð33238Ingvarsson,P.K.andWhitlock,M.C.(2000)Heterosisincreasestheeffectivemigrationrate.Proc.R.Soc.Lond.B.Biol.Sci.267,1321Ð132639Whitlock,M.C.etal.(2000)Localdriftloadandtheheterosisofinterconnectedpopulations.Heredity84,452Ð45740Laurance,W.F.andCochrane,M.A.(2001)Specialsection:synergisticeffectsinfragmentedlandscapes.Conserv.Biol.15,1488Ð148941Wade,M.J.andGoodnight,C.J.(1991)WrightÕsshiftingbalancetheory:anexperimentalstudy.Science253,1015Ð101842Boulding,E.G.andHay,T.(2001)Geneticanddemographicparametersdeterminingpopulationpersistenceafteradiscretechangeintheenvironment.86,313Ð32443Edmands,S.andTimmerman,C.C.(2003)Modelingfactorsaffectingtheseverityofoutbreedingdepression.Conserv.Biol.17,883Ð89244Burke,J.M.andArnold,M.L.(2001)GeneticsandtheÞtnessofAnnu.Rev.Genet.35,31Ð5245Rieseberg,L.H.andBurke,J.M.(2001)Thebiologicalrealityofspecies:geneßow,selection,andcollectiveevolution.Taxon50,47Ð6746Grant,P.R.etal.(2003)InbreedingandinterbreedinginDarwinÕsEvolution57,2911Ð291647Ungerer,M.C.etal.(2003)EffectsofgeneticbackgroundonresponsetoselectioninexperimentalpopulationsofArabidopsisthaliana163,277Ð28648Caswell,H.(2001)MatrixPopulationModels,SinauerAssociates49Wisdom,M.J.etal.(2000)Life-stagesimulationanalysis:estimatingvitalrateeffectsonpopulationgrowthforconservation.Ecology50Robert,A.etal.Releasingadultsversusyounginreintroductions:interactionsbetweendemographyandgenetics.Conserv.Biol.51Linklater,W.L.(2003)AnovelapplicationoftheTriversÐWillardmodeltotheproblemofgeneticrescue.Conserv.Biol.17,906Ð90952Mills,L.S.andAllendorf,F.W.(1996)Theone-migrant-per-generationruleinconservationandmanagement.Conserv.Biol.10,1509Ð151853Wang,J.(2004)Applicationoftheone-migrant-per-generationruletoconservationandmanagement.Conserv.Biol.18,332Ð34354Roberts,S.C.andGosling,L.M.(2003)Geneticsimilarityandqualityinteractinmatechoicedecisionsbyfemalemice.Nat.Genet.35,103Ð10655Swenson,J.B.etal.(1997)Infanticidecausedbyhuntingofmale386,450Ð45156Hogg,J.T.(2000)Matingsystemsandconservationatlargespatialscales.InVertebrateMatingSystems(Appolonio,M.etal.,eds),pp.214Ð252,WorldScientiÞcPress57Hess,G.(1996)Diseaseinmetapopulationmodels:implicationsfor77,1617Ð163258McCallum,H.andDobson,A.(2002)Disease,habitatfragmentation,andconservation.Proc.R.Soc.Lond.Ser.B269,2041Ð204959Waples,R.S.andDrake,J.(2004)Risk/BeneÞtconsiderationformarinestockenhancement:aPaciÞcsalmonperspective.InEnhancementandSeaRanching:Development,PitfallsandOppor-(Leber,K.M.etal.,eds),pp.260Ð306,Blackwell60Lynch,M.(1991)Thegeneticinterpretationofinbreedingdepressionandoutbreedingdepression.45,622Ð62961Beaumont,M.A.andRannala,B.(2004)TheBayesianrevolutioninTrendsGenet.5,251Ð26162Fenster,C.B.andGalloway,L.F.(2000)InbreedingandoutbreedingdepressioninnaturalpopulationsofChamaecristafasciculataConserv.Biol.14,1406Ð141263Fenster,C.B.andGalloway,L.F.(2000)Populationdifferentiationinanannuallegume:geneticarchitecture.Evolution54,1157Ð117264Edmands,S.(1999)Heterosisandoutbreedingdepressionininter-populationcrossesspanningawiderangeofdivergence.65Armbruster,P.etal.(1999)Evolutionaryresponsestoenvironmentalstressbythepitcher-plantmosquito,Wyeomyiasmithii66Marr,A.B.etal.(2002)Heterosisandoutbreedingdepressionindescendantsofnaturalimmigrantstoaninbredpopulationofsongsparrows(Melospizamelodia).Evolution56,131Ð14267Marshall,T.C.andSpalton,J.A.(2000)SimultaneousinbreedingandoutbreedingdepressioninreintroducedArabianoryx.Anim.Conserv.3,241Ð24868Burton,R.S.(1990)HybridbreakdownindevelopmentaltimeintheTigriopuscalifornicus44,1806Ð181369Burton,R.S.(1990)Hybridbreakdowninphysiologicalresponse:amechanisticapproach.44,1814Ð1822 TRENDSinEcologyandEvolutionVol.19No.9September2004495 70Galloway,L.F.andFenster,C.B.(2001)NuclearandcytoplasmiccontributionstointraspeciÞcdivergenceinanannuallegume.Evolution55,488Ð49771Edmands,S.andBurton,R.S.(1999)Cytochromeactivityininterpopulationhybridsofamarinecopepod:atestfornuclear-nuclearornuclear-cytoplasmiccoadaptation.Evolution53,1972Ð197872Gharrett,A.J.etal.(1999)Outbreedingdepressioninhybridsbetweenodd-andeven-broodyearpinksalmon.173,117Ð12973Darwin,C.E.(1883)TheVariationofAnimalsandPlantsUnderDomestication,D.Appleton&Co74Wright,S.(1931)EvolutioninMendelianpopulations.Genetics75Wright,S.(1940)BreedingstructureofpopulationsinrelationtoAm.Nat.74,232Ð24876Hedrick,P.(1995)Geneßowandgeneticrestoration:theFloridapantherasacasestudy.Conserv.Biol.9,996Ð100777Hedrick,P.Neutral,detrimental,andadaptivegeneticvariationinconservationgenetics.InConservationGeneticsintheAgeof(Amato,G.etal.,eds),ColumbiaUniversityPress(inpress)78Keller,L.F.etal.(2001)Immigrationandtheephemeralityofanaturalpopulationbottleneck:evidencefrommolecularmarkers.Proc.R.Soc.Lond.Ser.B268,1387Ð1394 ImportantinformationforpersonalsubscribersDoyouholdapersonalsubscriptiontoajournal?Asyouknow,yourpersonalprintsubscriptionincludesfreeonlineaccess,previouslyaccessedviaBioMedNet.Fromnowon,accesstothefull-textofyourjournalwillbepoweredbyScienceDirectandwillprovideyouwithunparalleledreliabilityandfunctionality.Accesswillcontinuetobefree;thechangewillnotinanywayaffecttheoverallcostofyoursubscriptionoryourentitlements.Thenewonlineaccesssiteofferstheconvenienceandexibilityofmanagingyourjournalsubscriptiondirectlyfromoneplace.Youwillbeabletoaccessfull-textarticles,search,browse,setupanalertorrenewyoursubscriptionallfromonepage.Inordertoprotectyourprivacy,wewillnotbeautomatingthetransferofyourpersonaldatatothenewsite.Instead,wewillbeaskingyoutovisitthesiteandregisterdirectlytoclaimyouronlineaccess.Thisisone-timeonlyandwillonlytakeyouafewYournewfreeonlineaccessoffersyou:QuicksearchBasicandadvancedsearchformSearchwithinsearchresultsSavesearchArticlesinpressExportcitationsE-mailarticletoafriendFlexiblecitationdisplayMultimediacomponentsHelplesIssuealerts&searchalertsforyourjournalhttp://www.trends.com/claim_online_access.htm TRENDSinEcologyandEvolutionVol.19No.9September2004