experimental批处理for

PrivateHeatforPublicWarmth:HowHuddlingShapesIndividualThermogenicResponsesofRabbitPupsCarolineGilbert1*.
,DominicJ.
McCafferty2.
,SylvainGiroud3,AndreAncel4,5,StephaneBlanc4,51UniversiteParis-Est,EcoleNationaleVeterinaired'Alfort,UMR7179,CNRS,MNHN,Maisons-Alfort,France,2InstituteofBiodiversity,AnimalHealthandComparativeMedicine,CollegeofMedical,Veterinary&LifeSciences,UniversityofGlasgow,Glasgow,UnitedKingdom,3ResearchInstituteofWildlifeEcology,UniversityofVeterinaryMedicine,Vienna,Austria,4UniversitedeStrasbourg,IPHC,Strasbourg,France,5CNRS,UMR7178,Strasbourg,FranceAbstractBackground:Withintheirlitter,youngaltricialmammalscompeteforenergy(constraininggrowthandsurvival)butcooperateforwarmth.
Theaimofthisstudywastoexaminethemechanismsbywhichhuddlinginaltricialinfantsinfluencesindividualheatproductionandloss,whileprovidingpublicwarmth.
Althoughconsideredasatextbookexample,itissurprisingtonotethatphysiologicalmechanismsunderlyinghuddlingarestillnotfullycharacterised.
Methodology/PrincipalFindings:Thebrownadiposetissue(BAT)contributiontoenergyoutputwasassessedasafunctionoftheabilityofrabbit(Oryctolaguscuniculus)pupstohuddle(placedingroupsof6and2,orisolated)andoftheirthermoregulatorycapacities(non-insulatedbefore5daysoldandinsulatedatca.
10daysold).
BATcontributionofpupsexposedtocoldwasexaminedbycombiningtechniquesofinfraredthermography(surfacetemperature),indirectcalorimetry(totalenergyexpenditure,TEE)andtelemetry(bodytemperature).
Throughlocalheating,thehuddleprovidedeachpupwhatevertheiragewithanambient''publicwarmth''inthecold,whichparticularlybenefitednon-insulatedpups.
Huddlingallowedpupsfacingaprogressivecoldchallengetobufferthedecreasingambienttemperaturebydelayingtheactivationoftheirthermogenicresponse,especiallywhenfur-insulated.
Inthisway,huddlingpermittedpupstoeffectivelyshiftfromanon-insulatedtoapseudo-insulatedthermalstatewhilecontinuouslyallocatingenergytogrowth.
ThehighcorrelationbetweenTEEandthedifferenceinsurfacetemperaturesbetweenBATandbackareasofthebodyrevealsthatenergylossfornon-shiveringthermogenesisisthemajorfactorconstrainingtheamountofenergyallocatedtogrowthinnon-insulatedaltricialpups.
Conclusions/Significance:Byprovidingpublicwarmthwithminimalindividualcostsatastageoflifewhenpupsarethemostvulnerable,huddlingbufferscoldchallengesandensuresaconstantallocationofenergytogrowthbyreducingBATactivation.
Citation:GilbertC,McCaffertyDJ,GiroudS,AncelA,BlancS(2012)PrivateHeatforPublicWarmth:HowHuddlingShapesIndividualThermogenicResponsesofRabbitPups.
PLoSONE7(3):e33553.
doi:10.
1371/journal.
pone.
0033553Editor:LewisGeorgeHalsey,RoehamptonUniversity,UnitedKingdomReceivedJanuary28,2011;AcceptedFebruary16,2012;PublishedMarch16,2012Copyright:2012Gilbertetal.
Thisisanopen-accessarticledistributedunderthetermsoftheCreativeCommonsAttributionLicense,whichpermitsunrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalauthorandsourcearecredited.
Funding:ThisresearchwasfundedbytheCentreNationaldelaRechercheScientifique(CNRS),France,http://www.
cnrs.
fr/.
CGwassupportedbyaBDI-CNRSfellowship.
Thefundershadnoroleinstudydesign,datacollectionandanalysis,decisiontopublish,orpreparationofthemanuscript.
CompetingInterests:Theauthorshavedeclaredthatnocompetinginterestsexist.
*E-mail:cgilbert@vet-alfort.
fr.
Theseauthorscontributedequallytothiswork.
IntroductionInshareddevelopmentalenvironments,competitionforlimitedresourcesisamajormechanismshapingphenotypicdifferencesamongsiblings[1].
However,atthesametimethatoffspringwithinanestcompeteforfoodsupply,theyjointlycreateanenvironmentthatpromotestheirgrowth[2].
Theadvantageofsuchasharedenvironmentisespeciallynoticeableinyoungaltricialmammalsforwhichthermalrequirementsintheirfirstdaysdeterminetheirgrowthtrajectoriesandsurvival(e.
g.
[3–7]).
Huddling,anactiveandcloseaggregationofanimals(reviewin[8]),isawidespreadcooperativegroupbehaviourthatallowsindividualstoreducetheirthermoregulatoryexpenses,thusenhancingsurvivalwhenenergybecomesalimitingfactor.
Byhuddlingtogetherinthenest,altricialpupsthenreducetheirthermoregulatorycostsandmaintainastablebodytemperaturepromotinggrowth[5–7,9–11].
However,withinahuddle,energyisallocatedbyeachindividualforthermogenesis,whilethebenefitsresultingofgroupheatingareshared.
AspointedoutbyHaig[12],''heatgeneratedbyhuddlinganimalsisapublicgoodwithaprivatecost''.
Asaresponsetocold,heatisfirstproducedinyoungmammalsbynon-shiveringthermogenesis(NST),dependingontheactivationofbrownadiposetissue(BAT)[13–16].
BecauseNSTrapidlygeneratesagreatamountofheatoverashortperiod,itisanessentialresponsetocoldthatpreventshypothermiainsmallmammalsornewborns[17].
NSTactivationhas,however,acost.
Asanaerobicprocessverydemandinginoxygenitsacuteuseinaltricialnewbornwillreducegrowthrateswhereasitschronicusewillultimatelyimpactsurvival.
Amongaltricialmammals,ratandrabbitnewbornsparticularlyfacethistrade-off.
TheyareabletogenerateheatviaNSTinBATintheirfirstdayoflifebutareunabletokeepitduetothelackofbodyinsulationandmustrelyonhuddlingtosharethecostofthermoregulation[4,10,11,18–PLoSONE|www.
plosone.
org1March2012|Volume7|Issue3|e3355322].
BATinnewbornrabbitsissituatedbetweenthescapulaeandaroundtheneck;itsmassismaximumatbirth,anddecreasesbyhalfduringthefirstdaysoflife[18].
Thissuggestthatrabbitpupsmaygainathermalbenefitfromtheirlittermates'presenceduringthefirstfivedaysoflifewhentheyarenon-insulated[5,6]butnotlateron.
Thermoregulatorycostsofrabbitpupsmaythereforedependbothontheirhuddlingbehaviourandtheirchangingdevelopmentalthermoregulatoryconstraints[6].
Sincealtricialmammalswithintheirlittercompeteforenergybutcooperateforwarmth,weexploredinthisstudytheproximalthermoregulatoryconstraintsthatultimatelyimpactontheirgrowthandsurvivalintheirlitter.
Wepredictedthattheefficiencyofhuddlingonthethermoregulatoryresponsesofaltricialrabbitpupsfacingacoldchallengewillrapidlydecreaseasafunctionofageandthermoregulatorycapacitiesi.
e.
insulatedvs.
non-insulated.
Inparticular,weinvestigatehowthehuddleinfluences:1)thethermalenvironmentofthepupand2)thermogenicresponsesofindividualsaswellas3)howindividualpupscontributetogroupthermoregulatoryprocessesinthelitter.
Inthiscontext,wetestedthephysiologicalresponsesofrabbitpupsinthecoldfrom23to11uC.
Bycombiningindependenttechniquestomeasureenergyexpenditure(indirectcalorimetry),bodytemperature(implantedsensors)andsurfacetemperature(infraredthermography[23–25]),wewereabletoevaluatetheenergeticcontributionofnon-shiveringthermogenesisthroughBATactiva-tion.
Theoverallaimofthisstudyistoprovideinformationonhowphysiologicalconstraintscontrolthermoregulationinearlylifeanditsconsequencesfortheirlifehistory.
MaterialsandMethodsAnimalsandhousingconditionsTheexperimentswerecarriedoutonrabbits(Oryctolaguscuniculus)ofacrossedstrain,''Hyplus''fromGrimaud(NewZealandWhite6Californianrabbits;GrimaudLaCorbie`re,Roussay,France;http://www.
grimaud.
fr/).
Fourdoesandtwomaleswerehousedindividuallyincages(506100cm,50cmhigh).
Parturitionoccurredspontaneouslyafter30–31daysofgestation.
Theslidingdoorofalitterbox(30650cm,40cmhigh)filledwithfreshstrawandhookedtothedoe'scagewasopened3daysbeforeparturition.
Roomtemperaturevariedfrom18to23uC,anda16:8-hlight-darkcyclewasmaintained.
TheanimalswerekeptandtreatedduringexperimentsinaccordancewiththeEuropeanGuidelinesforAnimalCarewithfullapprovalsfromtheFrenchGovernment,theCentreNationaldelaRechercheScientifiqueandtheDirectionoftheVeterinaryServices(no.
67-188).
Atbirth,pupswereleftafewhourswiththedoeinthenestingboxtoallowthemtosuckleoncewithoutdisturbance.
Thenpupswereseparatedfromthedoe,color-markedontheirbackwithananimalmarkingstick(Raidex,Dettingen/Erms,Germany)foridentification,andweighed(Sartoriusmodel1403,Germany,www.
sartorius.
com/,60.
1g).
Pupswereplacedinplasticboxeswithanopentop(28642cm,16cmhigh).
Exceptduringtheexperiment,allpupswerehousedingroupsfromtheiroriginallitterinaroomwithacontrolledambienttemperatureof23–24uCandcontinuouslighting.
Ambienttemperaturesetpointwaschosentolimitanymortalityofthepups,basedonapreviousstudybyourteam[6].
Sucklingwasallowedonceaday,around9:30–10:00AM,andlastedfor3–4minonaverage.
Pupswereweighedbeforeandaftersuckling(60.
1g)tomonitortheirgrowth.
Experimentswereundertakenduringtheafternoon,afewhoursaftersuckling(ca.
4hours),inordertominimizethermogeniceffectsduetopost-prandialdigestion[26–27].
CompositionofgroupsandcoldexposureExperiment1(n=24pupstakenfromfourlittersof7–8pups)wasdesignedtoexplorethermoregulatoryresponsesofpups.
Pupswereplacedinaroomadjacenttothebreedingroom,andexposedtodecreasingambienttemperaturesof23(for2hours),18,15,and11uC(for1hour).
Roomtemperatureafterthecoldexposurewasagainregulatedat23–24uC.
Non-insulatedpupsaged4daysoldwereplacedingroupsofsix(G6,4groupsof6pups),andthesubsequentday(at5daysold),thesamepupswereseparatedingroupsoftwo(G2,i.
e.
12groupsof2pups).
Thesameprocedurewasrepeatedwhenpupswere15daysold(G6insulatedpups;n=4G6groups)and16daysold(G2insulatedpups;n=12G2groups).
Aftertheexperimentsallpupswereplacedinthebreedingroom.
Independently,Experiment2wasdesignedtoinvestigatebrownadiposetissueactivationusingthermalimaging(n=12pupstakenfromtwolittersof7–8pups).
Non-insulatedpupsaged3and4daysoldandinsulatedpupsaged10and11daysoldwereused.
Theywereexposedtoambienttemperaturesof23uC(for2hours)and14uC(for1hour).
Onday1,agroupofsixpups(G6,3daysold)andtwoisolatedpups(G1,10daysold)weretested.
Onday2,agroupofsixpups(G6,aged11daysold,includingthetwoG1ofday1)andtwoisolatedpups(G1,4daysold,randomlyselectedfromtheG6ofday1)weretested.
Totalenergyexpenditure(TEE)Oxygen(O2)consumptionandcarbondioxide(CO2)produc-tionweremeasuredusinganopen-circuitrespirometrysystem(SableSystemsInternational,LasVegas,NV,USA).
TheconcentrationsofO2andCO2intheoutgoingairweremeasuredinoneexperimentalchamberforG6(at4and15daysold,25636616cm,V=14.
4L),andinthreechambersforthethreeG2(at5and16daysold,17624615cm,V=6.
1L).
Thetwodifferentsizesofchambersweredefinedaftertheoreticalcalculationstakingintoaccounttheairflowandtheenergyexpenditureofpups,toensurerobustrespirometrymeasurements.
Measurementswereperformedcontinuouslyoverthecoldchallenge(from23to11uC).
CalibrationofO2andCO2analyzerswereundertakenbeforeandaftereachexperiment.
Forthe''G6-procedure'',thecagewassampledfor120s(1samplepersecond)ataflowrateof1L.
min21every4min,whileforthe''G2-procedure'',thethreecagesweresuccessivelysampledfor120s(1samplepersecond)ataflowrateof1L.
min21every8min.
Forbothexperiments,finalvaluesofO2andCO2concentrationswerethemeanofvaluesrecordedduring60and90s,respectively.
Thesystemwasflushedwithairfor120sbetweeneachcycle.
Energyexpenditurewascalculatedusinganenergyequivalentof16.
47J.
ml21ofO2consumedand4.
62J.
ml21ofCO2produced,accordingtoWeir'sequations[28].
MetabolicratewasexpressedaskJ.
day21.
g21.
Themeanbodymassofrabbitswasdeterminedfromaweightbeforemeasurementsandaweightthedayaftertheexperiment,beforesubsequentsuckling.
ForExperiment2,theG6wasplacedinachamberof25636616cm,whilethetwoG1rabbitpupswereplacedincagesof17624615cm.
TheconcentrationsofO2andCO2intheoutgoingairweremeasuredsimultaneouslyinthethreechambers.
Measurementswereperformedcontinuouslyoverthecoldchallenge(at23uCfor2hoursand14uCfor1hour).
SamplingsandcalculationsweresimilartoExperiment1,accordingto''G2-procedure''.
MicroclimatewithinthecalorimetricchambersInordertoevaluateanylocalheatingeffect,inparticularinthechambersforG6pupscomparedwithG2pups,temperaturesPrivateHeatforPublicWarmthPLoSONE|www.
plosone.
org2March2012|Volume7|Issue3|e33553werealsomeasuredineachcalorimetricchamber.
Theambienttemperaturewithinchamberswascontinuouslymonitoredbyathermoresistivedevice(Pt-RTD100,Jumo,Jumo-Regulation,Metz,France)placedinthecentreofthechamber(seeaboveforboxdimensions).
ThethermalprobeswereconnectedtoaSmartA/Dcomputercard(modelno.
619),anddatawererecordedevery30susingSensorayQuicksensesoftware(version3.
3;SmartA/DandQuicksense,Sensoray,Tigard,OR;USA;http://www.
sensoray.
com).
Thermoresistivedeviceswerecalibratedinathermostaticbathfrom20to40uC,with5uCincrementsbeforeandaftertheexperiments.
BodytemperatureInExperiment1,11randomlyselectedpupswereimplantedwithtransmittersat2.
5daysofage(meanbodymassofca.
85g).
TA10TA-F20transmitters(DataSciencesInternational,StPaul,MN;USA;3.
5g,1.
75cm3)wereplacedintraperitoneally,undergaseousanesthesia(isoflurane,Forene)andstrictlyasepticsurgicalconditions.
Thesurgerytookplaceatleast5haftersuckling,andaheatingpadwasusedtopreventhypothermia.
Antibiotics(oxytetracycline,TerramycineLA)andanti-inflammatorymole-cules(ketoprofen,Ketofen)wereinjectedattheendofsurgeryandthepupwasreturnedtoitslittermateswithin1–2h.
Transmittersrecordedbodytemperatureat30-sintervalsduringtheexperi-ments.
Becauseofradiofrequencyinterference,onlyonepuppergroupcouldbemonitored(at4and15daysold:4pupsplacedinG6;at5and16daysold:11pupsplacedinG2).
InExperiment2,4randomlyselectedpupswereimplantedwithtransmittersfollowingasimilarprotocol.
ThermalimagesInExperiment2,thesurfacetemperatureofrabbitpupswasrecordedatthestartandendofrespirometryperiods(at23and14uC)usingathermalimagingcameraPM595(FLIR,USA).
Thelidoftherespirometrychamberwasremovedandimageswererecordedataheightof1mabovethechamber.
ThermalimageswereanalysedusingThermacamReporter7.
0(FLIR,USA)usinganemissivityforfurof0.
98[29].
Surfacetemperatureswereaveragedfromimagestakenatstartandendoftherespirometryperiods.
ThemeansurfacetemperatureofpupswasdeterminedbyfittingapolygonaroundtheindividualanimalinthecaseofisolatedpupsandaroundtheentirehuddleforG6.
Wechosetomeasurethreesitesofsurfacetemperatures:back,earandbrownadiposetissue(BAT).
Asaresponsetocold,vasoconstrictionwasinvestigatedbymeasuringeartemperature(Tear),thepinnabeingtheprincipalsiteofheatdissipationforrabbits[30].
Heatproductionthroughnon-shiveringthermogenesiswasestimatedmeasuringBATsurfacetemperature(TBAT).
Thebacktempera-ture(Tback)thereforerepresentedareferencesurfacetemperature,fromaneutralareawithrespecttoheatflow.
Thegradient(TBAT-Tback)allowedthecomparisonofBATactivationindependentlyofambienttemperatures(23or14uC).
ForG1pups,surfacetemperatures(back,earandBAT)werecalculatedasthemean(6SD)oftwoindividualsforeachagegroup.
ForG6pups,themean(6SD)surfacetemperatures(back,earandBAT)weredeterminedfromthreeindividualswithinthegroupaspositionsofpupspreventedaclearviewofallpupswithinthehuddle.
Thebackregionwasdefinedusingacircle(diameter=20pixels)positionedonthedorsalsurfaceinthecentreofthehips.
Eartemperaturesweredeterminedbyfittingapolygonaroundtheouteredgeofbothleftandrightears.
AsitwasnotpossibletopreciselydefinethesurfaceoverlyingBAT,acircle(diameter=20pixels)waspositionedontheimagebetweenscapulae(following[31,32]).
StatisticalanalysesPriortoanalyses,weverifiedonrawdatatheabsenceofanydifferencebetweenlittersandcategoriesofpups.
Coldchallengesinthechambersweretestedindependentlyforthefourcategoriesofpups,betweenthefourambienttempera-tures,usingANOVAsandKruskal-WallisANOVAswhendistributionwasnotnormal.
InordertocomparelocalheatingbetweenG6andG2pups,insulatedandnon-insulatedpupswerepooledandMann-Whitneytestswereusedforeachroomtemperature.
Bodymassofpupswerecomparedwithpairedt-testsbetweenthetwosessions:G6pupsatday4,andG2pupsatday5;G6pupsatday15,andG2pupsatday16.
Toassesstheeffectsofcoldchallenge,huddling(G2vs.
G6),andage(non-insulatedvs.
insulatedpups)oneitherenergyexpenditureorbodytemperature,andtotakeintoaccountrepeatedmeasurementnatureofthedata,multivariatelinearmixed-effectsregressionmodelswereimplementedusingtheSASPROCMIXEDprocedure(SASInstitute,Cary,NC).
Independentvariablesincludedexternaltemperature,group(G2vs.
G6),andage(insulatedvs.
non-insulated).
Bodymasswasalsoincludedasapotentialconfounder.
Thesemodelswerefittedtothedatawiththeinterceptastherandomeffect.
Surfacetemperatureofthedifferenttissues(ear,backandBATforisolatedpups;n=4,andear,backandBATforpupsingroupsof6;n=6)werecomparedwithone-wayANOVAsonrepeatedmeasuresforeachcategory(post-hocTukeytests).
Resultsareexpressedasmeans6SD.
Statisticaltestsareconsideredsignificantatp,0.
05.
ResultsHowdoesthehuddledeterminethethermalenvironmentandphysiologicalresponsesofpupsComparisonofcoldchallengesexperiencedbyallcategoriesofpups.
Roomtemperatureduringthecoldchallengewasdecreasedsuccessivelyfrom22.
860.
8to18.
260.
8,15.
060.
8,and10.
661.
3uC(Table1).
Decreasingambienttemperaturesintheexperimentalroomdidnotdifferbetweengroups(non-insulatedandinsulated,G6andG2pups,p.
0.
08for23,18,15and11uC).
Coldchallengesinsidetheboxesweresimilarforthefourcategoriesofpups(G6andG2,non-insulatedandinsulatedpupscomparedat23,18,15and11uC:H=8.
938,p=0.
03,noposthocdifferencesfor23uC;p.
0.
12forothertemperatures;Table1).
However,bypoolingcoldchallengesofinsulatedandnon-insulatedpupsgroupedbysixortwo,localheatingforG6pupswassignificantat18uC(+1.
4uCcomparedwithG2pups,U=154,p=0.
05),andclosetosignificanceat15uC(+1.
8uC,U=149,p=0.
085)andat11uC(+2.
4uC,U=150,p=0.
076;Table1).
Localheatingwasnotsignificantat23uC(U=139.
5,p=0.
202).
Hence,duetotechnicalconstraintsimposedbyrespirometrymeasurements,G6pups(insulatedandnon-insulated)wereexposedlocallytoslightlylessofacoldchallengeincomparisontoG2pups.
Betweenthetwosessions(atday4andday5),pupsgainedonaverage13.
465.
3g,correspondingtoa14%increaseinbodymass(bodymassofG6pupsatday4:98.
9615.
9g,ofG2pupsatday5:112.
3618.
9g,t=212.
3,p,0.
001).
Inthesameway,betweenday15andday16,pupsgained15.
267.
0g,i.
e.
6%oftheirtotalbodymass(bodymassofG6pupsatday15:267.
6641.
7g,ofG2pupsatday16:282.
8643.
5g,t=210.
6,p,0.
001).
Hence,G2pups,onedayolderthanwhenplacedingroupsofsix,possessedahigherbodymass,i.
e.
alowersurfaceareatovolumeratiothanG6pups.
Thismayhaveminimisedanyphysiologicalvariationsandeffectslinkedtohuddling.
PrivateHeatforPublicWarmthPLoSONE|www.
plosone.
org3March2012|Volume7|Issue3|e33553Physiologicalresponsestocoldfornon-huddling(G2pups)vs.
huddling(G6pups)andnon-insulatedvs.
insulatedpups(Figures1A&1B).
Forallambienttemperatures,non-insulatedpupsshowedatotalenergyexpenditure(TEE)morethantwofoldhigherthaninsulatedpups(0.
9960.
17vs.
0.
4960.
06kJ.
g21.
day21;p,0.
001afteradjustmentforcold,groupandbodymass).
Inaddition,TEEofG6pupswas14%lowerthanG2pups(0.
6660.
24vs.
0.
7760.
29kJ.
g21.
day21;p=0.
02afteradjustmentforcold,ageandbodymass).
TEEofnon-insulatedpups(Figures1A&1B),bothinG6andG2,startedtoincreasesignificantlyat18uC,andfurtherat15and11uC(p,0.
004foralltemperaturesafteradjustmentforbodymass).
WhilebodytemperatureofG2pupsstartedtodecreasesignificantlyat15uC(p=0.
04afteradjustmentforbodymassbetween18uCand15uC),bodytemperatureofG6pupsstartedtodecreaseatthelowertemperatureof11uC(p,0.
001afteradjustmentforbodymassbetween18uCand23uCand11uC).
Betweenthetwoextremeambienttemperaturesof23uCand11uC,TEEofnon-insulatedG6pupssignificantlyincreasedby59%,whiletheirbodytemperaturedecreasedby0.
9uC(p,0.
0001andp=0.
0015respectively,afteradjustmentforbodymass).
Inthesameway,TEEofnon-insulatedG2pupsincreasedsignificantlyby38%inresponsetocold,whiletheirbodytemperaturedecreasedby1.
2uC(bothp,0.
0001afteradjustmentforbodymass).
Consideringinsulatedpups(Figures1A&1B),TEEofG6pupsdidnotincreaseinresponsetocold(p.
0.
13forallTable1.
Mean(6SD)ambienttemperaturesexperiencedbydifferentcategoriesofpupgroupsinthechambers(G6orG2pups,non-insulatedandinsulated)asafunctionofthecoldchallengeintheexperimentalroom.
G6non-insulatedG6insulatedG6G2non-insulatedG2insulatedG2Roomtemperature(uC)22.
860.
824.
760.
923.
461.
924.
0±1.
423.
860.
322.
960.
923.
4±0.
618.
260.
820.
861.
220.
262.
420.
4±1.
919.
261.
418.
861.
119.
0±1.
315.
060.
818.
462.
417.
163.
617.
6±3.
016.
062.
015.
561.
915.
8±1.
910.
661.
314.
462.
813.
164.
313.
7±3.
511.
662.
510.
962.
411.
3±2.
4doi:10.
1371/journal.
pone.
0033553.
t001Figure1.
Relationshipbetweentotalenergyexpenditure(TEE)andbodytemperatureofinsulatedandnon-insulatedpups,huddlingingroupsof6(G6)orplacedingroupsof2(G2),duringthecoldchallenge.
Differentlettersindicatesignificantdifferences.
***:significantdifferenceswithp,0.
0001between23uCand11uC;**:significantdifferenceswithp,0.
002between23uCand11uC.
NS:notsignificant.
doi:10.
1371/journal.
pone.
0033553.
g001PrivateHeatforPublicWarmthPLoSONE|www.
plosone.
org4March2012|Volume7|Issue3|e33553temperatureafteradjustmentforbodymass).
TEEofG2insulatedpupsincreasedwhenexposedtolowertemperaturesof15and11uC(p,0.
05afteradjustmentforbodymass).
Incontrast,bodytemperatureofbothG6andG2pupsstartedtodecreasesignificantlyat15uC(p=0.
0045forG6andp,0.
0001forG2pupsafteradjustmentforbodymass).
Betweenthetwoextremeambienttemperaturesof23uCand11uC,TEEofG6insulatedpupsincreasednonsignificantlyby6%,whiletheirbodytemperaturedecreasedby0.
5uC(p=0.
226andp,0.
0001respectively,afteradjustmentforbodymass).
Bycontrast,TEEofG2insulatedpupsincreasedby23%inresponsetocold,whiletheirbodytemperaturedecreasedby0.
3uC(bothp,0.
0001afteradjustmentforbodymass).
Brownadiposetissuethermogenesisofnon-insulatedandinsulatedpupsexposedtocoldwhilehuddling(G6)orisolated(G1):howdoindividualpupscontributetogroupthermoregulatoryprocessesinthehuddleBodytemperatureandenergyexpenditureresponsesofG6andG1pups,at23uC(23.
060.
1uC)and14uC(14.
060.
7uC)fornon-insulated(at3and4daysold)andinsulatedpups(at10and11daysold),wereconsistentwithExperiment1(Figure2).
Themean(6SD)massofG6pupsaged3dayswas98.
765.
9g,whileG1pupsaged4dayswas108.
964.
7g,representinga10%increaseinbodymass.
Forinsulatedpups,G1pupsaged10daysaveraged215.
5657.
2g,whilethemassofG6pupsaged11dayswas229.
4634.
3g,thusrepresentinga6.
5%increaseinbodymass.
Ambienttemperaturesweremonitoredineachexperimentalchamber.
Atwarmambienttemperature,localheatinginthechamberscontainingG6pupscomparedwithisolatedpupswas0.
8uCfornon-insulatedpups(24.
460.
1uCfor3-dayG6pupsvs.
23.
660.
3uCfor4-dayG1pups)and2.
1uCforinsulatedpups(25.
660.
6uCfor11-dayG6pupsvs.
23.
560.
1uCfor10-dayG1pups).
Inthesameway,atcoldambienttemperature,localheatingwas3.
7uCforG6non-insulatedpups(18.
060.
2uCfor3-dayG6pupsvs.
14.
360.
4uCfor4-dayG1pups)and2.
6uCforG6insulatedpups(18.
260.
3uCfor11-dayG6pupsvs.
15.
660.
3uCfor10-dayG1pups).
Coldchallengesweretherefore6.
4uCfor3-dayG6pups,9.
3uCfor4-dayG1pups,7.
4uCfor11-dayG6pups,and7.
9uCfor10-dayG1pups.
BATactivationandearvasoconstriction:comparisonofBAT,ear,andbacksurfacetemperatures.
Consideringnon-insulatedpups,atwarmambienttemperatures,themeansurfacetemperatureofthehuddle(i.
e.
groupof6pups)was1.
9uChigherthanthemeantemperatureofisolatedpups(36.
960.
9uCvs.
35.
060.
5,respectively).
Atcoldambienttemperatures,themeansurfacetemperatureofthehuddlewas4.
7uChigherthanthemeantemperatureofisolatedpups(35.
660.
4vs.
30.
960.
9uC).
BATsurfacetemperaturewassignificantlyhigherthanbackandearsurfacetemperaturesforeachcategoryofnon-insulatedpup(p,0.
001inallcases,post-hocp,0.
05;Figure3).
Backsurfacetemperatureswerealsosignificantlyhigherthanearsurfacetemperatures(post-hocp,0.
05),exceptforG6non-insulatedpupsat14uC(post-hoctestnonsignificant;Figure3).
Consideringinsulatedpups,at23uC,themeansurfacetemperatureofthehuddlewas1.
2uChigherthanthemeantemperatureofisolatedpups(32.
060.
8uCvs.
30.
860.
8,respec-tively).
At14uC,themeansurfacetemperatureofthehuddlewas0.
9uChigherthanthemeantemperatureofisolatedpups(26.
760.
4vs.
25.
861.
0uC).
ThemeanBATsurfacetemperaturewashigherthanthemeanbacksurfacetemperatureofG6insulatedpups(p,0.
001inbothcases;Figure3).
However,BATsurfacetemperatureofG1insulatedpupswasnotsignificantlydifferentfrombackandeartemperatures(p.
0.
05inbothcases;Figure3).
BATactivation,bodytemperatureandenergyexpenditure.
BAT,ear,andbacksurfacetemperatureswerecorrelatedwithbodytemperaturefornon-insulatedpups(R2$0.
95,p#0.
025inallcasesFigure4A).
Bycontrast,BAT,ear,andbacksurfacetemperaturesofinsulatedpupswerenotcorrelatedwithbodytemperature(R2#0.
73,p.
0.
14inallcasesFigure4B).
InordertodeterminewhetherBATsurfacetemperature,TBAT,couldbeusedasanindexofenergyexpenditure,therelationshipbetweentotalenergyexpenditure(TEE)andTBAT-Tback,atwarmandcoldtemperatureswasexamined(Figure5).
Fornon-insulatedpups,TEEwashighlycorrelatedwithTBAT-Tback(TEE=0.
534+0.
354.
(TBAT-Tback),r2=0.
991,F1,3=223.
9,p=0.
004;Figure5).
However,forinsulatedpups,TEEwasnotcorrelatedwithTBAT-Tback(r2=0.
500,p=0.
293;Figure5).
DiscussionColdchallenges(from23to11uC)weredesignedtobesignificantfordevelopingrabbitpups[4–6,18,19].
Duetoexperimentalconstraints,rabbitpupswerecoldchallengedat4Figure2.
Totalenergyexpenditure(TEE,kJg21day21;bars)andbodytemperature(6C;circles)foreachpupcategory(G6huddlingandG1isolatedpups,insulatedandnon-insulated,exposedat236Cand146C).
doi:10.
1371/journal.
pone.
0033553.
g002PrivateHeatforPublicWarmthPLoSONE|www.
plosone.
org5March2012|Volume7|Issue3|e33553Figure3.
Surfacetemperatureofback,earandbrownadiposetissue(BAT)ofisolated(G1)andhuddling(G6)pups,non-insulatedandinsulated,exposedat236Cand146C.
Theresultsofpost-hocTukeytestsareshownbythelettersatoc:samelettershowednosignificantdifference.
Leftpanel:Thermalimagesareshownforisolated(G1)andhuddling(G6)non-insulatedandinsulatedpups,exposedtoanambienttemperatureof14uC.
Imagesshowthelocationofcirclesusedtocomparesurfacetemperature(totalarea,back,ear,brownadiposetissue(BAT)).
doi:10.
1371/journal.
pone.
0033553.
g003PrivateHeatforPublicWarmthPLoSONE|www.
plosone.
org6March2012|Volume7|Issue3|e33553and15daysoldinG6(huddlingcondition)andadaylaterinG2(non-huddlingcondition).
Inasecondexperiment,G6pupswerecoldchallengedat3and11daysoldandisolated(G1)pupswerechallengedat4and10daysold.
Thisdifferenceinagemayhaveweakenedthecomparisonbetweenhuddlingandnon-huddlingpupssincetheydidnotpossessstrictlysimilarheatlossconstraintslinkedtotheirdifferentbodymass,andhencetheirdifferentsurfacetovolumeratio[33,34].
Asnon-insulatedpupswhenhuddlingpossessedlessfavourableheatlossconstraints,thisprocedurereinforcesourfindingswithrespecttoenergysavingsassociatedwithhuddling.
Thecomparisonbetweenhuddlingandnon-huddlingpupshoweverremainedvalidfornon-insulatedpups(upto5daysold,whentheyhavenoorlittlefurandahighsurfacetovolumeratio),andinsulatedpups(from5daysold,whentheirbodymasshasincreasedandtheyhavegrownfur;[6]).
AccordingtoAlberts'analysis[10],theexperimentaldesignmayhaveactuallystrengthenedthecomparisonbetweenhuddlingandnon-huddlinganimalsbyavoidingphysiologicaldifferencesbetweenindividuals.
HuddlingpromoteslocalheatingandreducesthecoldchallengeThechamberswereclosedforrespirometrymeasurements,andthereforelocalheatinginthechamberscontainingG6huddlingpupswasimportantincomparisontoisolatedorG2pups.
Inwarmconditions(23uC),localheatingwas0.
8uChigherforhuddlingnon-insulatedpups(aged3–4days),and2.
1uChigherforhuddlinginsulatedpups(aged10–11days)comparedwithisolatedlittermates.
Localheatingwasevenmorepronouncedinacoldenvironment(14uC):surroundingtemperaturesofhuddlingpupswere3.
7and2.
6uChighercomparedwithisolatedpups(non-insulatedandinsulatedpups,respectively).
Warmingofthelocalmicroclimateispartofthethermalbenefitsofhuddling[8].
Forexample,a5uCincreaseinambienttemperaturewithinthenestofhuddlingshort-tailedfieldvoles(Microtusagrestis)accountsfor55%oftheenergeticbenefitsofhuddling[35].
Bautistaetal.
[5]showedthattheambienttemperatureintheimmediatevicinityofgroupsoffourtosixhuddlingrabbitpupswas26uC,significantlyhigherthanforgroupsoftwo(24uC)orforisolatedpups(23uC)whenpupswereexposedtoasimilarambienttemperature.
Hencesinglepupsorpairseffectivelyexperiencedcolderconditionscomparedwithpupshuddlinginagroupofsix.
Localheatingmoreoverdirectlyimpactedonthemeansurfacebodytemperatureofthepups.
Thesurfacetemperatureofhuddlingnon-insulatedpupswas1.
9and4.
7uChigherthanisolatedpupsinwarmandcoldconditions,respectively.
Wheninsulated,thesurfacetemperatureofhuddlingpupswas1.
2and0.
9uChigherthanisolatedFigure4.
Surfacetemperatureofback,earandbrownadiposetissue(BAT)ofnon-insulated(Figure5A)andinsulated(Figure5B)pups(isolated,G1orhuddling,G6)asafunctionoftheirbodytemperature.
Regressionlinesarestatisticallysignificantfornon-insulatedpupsonly.
Non-insulatedpups:TBAT=217.
79+1.
46*Tb,r2=0.
992,p=0.
004;Tear=271.
46+2.
85*Tb,r2=0.
950,p=0.
025;Tback=242.
14+2.
08*Tb,r2=0.
988,p=0.
006.
Insulatedpups:TBAT,r2=0.
436,p=0.
340;Tear,r2=0.
732,p=0.
144;Tback,r2=0.
086,p=0.
707.
doi:10.
1371/journal.
pone.
0033553.
g004Figure5.
Totalenergyexpenditure(TEE)asafunctionofsurfacetemperaturedifferencebetweenbrownadiposetissue(TBAT)andback(Tback).
G1standsforisolatedpupsandG6forhuddlingpups,non-insulated(circle)orinsulated(triangle).
Regressionlineisstatisticallysignificantfornon-insulatedpupsonly:TEE=0.
534+0.
354*(TBAT-Tback),r2=0.
991,F1,3=223.
9,p=0.
004.
doi:10.
1371/journal.
pone.
0033553.
g005PrivateHeatforPublicWarmthPLoSONE|www.
plosone.
org7March2012|Volume7|Issue3|e33553littermateswhenpupswereexposedtowarmandcoldconditions,respectively.
Ourresultsthereforeshowthatlocalheatingiscrucialinreducingtheextentofthecoldchallengeespeciallyinthefirstdaysoflife,whenpupsexperiencethermalstress.
Thismaybefrequentforrabbitpupsasunlikemanyothermammals,femalerabbitsleavetheirpupssoonafterbirth,andonlynursethemfor3to5minonceaday[36–38].
Thethermalenvironmentofpupsdependsbothontheinsulativepropertiesofthenestandoftheheatproducedbyeachpupwithinthelitter.
Thehuddlethereforeprovidesforeachindividuala''public''warmththatisbeneficialtosurvival,especiallyinthecoldandwhenpupsarenon-insulated.
HuddlingactsasabufferfornewbornstodelaythermogenicresponsestothecoldWhenexposedtocold,rabbitpupsincreasetheirenergyexpenditurebyariseinmetabolicheatproductionbutwhenoverwhelmedbyheatlossbodytemperaturedecreases[4,6].
Inthisstudyhuddlingpupsshowedadelayedthermogenicresponsecomparedwiththeirnon-huddlinglittermateswhenfacedwithaprogressivecoldchallenge(from23downto11uC).
Whennon-insulated,theincreaseinenergyexpenditurestartedwhenbothhuddlingandnon-huddlingpupswereexposedat18uC.
At11uC,energyexpenditureincreasedby59%forhuddlingpupsand38%fornon-huddlingpups.
However,thedecreaseinbodytemper-aturewasdelayedandlesspronouncedforhuddlingpups(significantdropsof0.
5uCat15uCand1.
2uCat11uCfornon-huddlingpupscomparedwitha0.
9uCdropat11uCforhuddlingpups).
Wheninsulated,energyexpenditureincreasedfornon-huddlingpupsby23%at11uCbutthisdidnotoccurwhenpupswerehuddling.
Similarly,thebodytemperatureofinsulatednon-huddlingpupsdecreasedat18uC,whilethisonlyoccurredforhuddlingpupsat15uC.
Insulatedhuddlingpups,exposedtothesameambientconditionsdidnotappeartofaceanycoldchallenge.
Inendotherms,physiologicalresponsestocoldinvolvevaso-constrictiontoreduceskinsurfacetemperatureandthermoregu-latorythermogenesistomaintainhomeostasis[33].
Innewborns,thermogenesisprimarilyoccursthroughnon-shiveringthermo-genesis(NST)withBATactivation[4].
InterscapularandcervicalBATisfunctionalatbirthinrabbits[18,19,39].
Astemperatureinfluencescellularmetabolism[33]andhencegrowth,mainte-nanceofahighbodytemperatureisanespeciallyimportantfactorshapingdevelopment.
Anydelayinthermalresponsesduetohuddlingreducesenergyrequirementsandpromotespupgrowth.
Theseresultsalsohighlightthatahigherandlessvariablebodytemperature,favourabletogrowth,ismoreeasilymaintainedbyhuddlingindividuals(inrabbits[5,6];inrats[10];forreview[8]).
Indeed,Bautistaetal.
[11]foundapositiverelationshipbetweenhuddlingbehaviourofindividualsandtheirbodytemperaturewithinthelitter.
Withintheirlitter,altricialmammalpupscooperateforwarmth,butcompeteforenergy,obtainedfrommaternalmilkandallocatedtogrowthandthermogenesis.
Pupgrowthinrabbitsishenceprimarilydependentonmaternalfactorsandlittersize[7,40].
Foodprovisioningforanindividualrabbitpupinalitterisindeedlimitedbymaternalmilkproduction[7,40,41]anditscompetitivenesstoaccessthemother'steatsduringthebriefdailysucklingperiod[38,42,43,44].
Moreover,survivalisinfluencedbythepups'bodymass[7]andtheamountofenergyallocatedtogrowthisdependentontheamountofenergylostforthermoregulation,linkedtoheatloss[6].
Inwildrabbitsthethermalenvironmentduringdevelopmentisanimportantdeterminantofoptimallittersize[7],throughthetrade-offbetweenalimitedenergysupply(milk)andthecooperativewarmthgainedbypupshuddlinginthelitter.
Theresultsofourstudysuggestthataltricialrabbitpupsareproximatelyconstrainedbythecostofthermoregulationbutthatthisisalleviatedbythebenefitsofhuddling.
Huddlingindeeddelaystheonsetofthermogenesisbyproviding''publicwarmth''thatinfluencesthecostofthermoregulationforeachindividual.
HuddlingreducesindividualBATheatproductionWhennon-insulated,bothhuddling(G6)andisolatedpupsproducedextraheatwhenexposedtocold:theirenergyexpenditureincreased,whiletheirbodytemperaturedecreased(Figure2).
Isolatedpupswerehoweverfacedwithahighercoldchallengethanhuddlingpupsastheircapacitytothermoregulatewasexceeded.
FornoninsulatedpupstheincreaseinTEEbetweenwarmandcoldconditionswas49.
5%forhuddlingpups,and12.
6%forisolatedpups(Figure6).
Eveninwarmconditionstheseisolatedpupsincreasedmetabolicheatproductionforeffectivethermoregulation.
However,whenexposedtocoldisolatedpupswereonlyabletoincreasethermogenesisbyarelativelysmallamount,resultinginadropinbodytemperatureof1.
7uC.
Furthermore,thermalimagesshowedthatTBATwashigherthanTbackinallcases,indicatingthatheatproductionwaslargelyduetoBATactivation.
Pupsalsoreducedtheirheatlossbyvasoconstriction,asseenbyadropineartemperature(Figure6).
AsBAT,ear,andbacksurfacetemperatureswerecorrelatedwithbodytemperature,bothforhuddlingandisolatedpupsinwarmandcoldconditions,theheatproducedbynon-shiveringthermogenesisthroughBATisthereforerequiredforthemaintenanceofastableandhighbodytemperature.
Previously,infraredthermographyprovidedausefultooltorevealBATactivation[4,21,31,32,45,46].
Asrevealedbythermalimages,itisknownthatthethermogeniccapacityofBATisdeterminedbybehaviouralmodulationofhuddlingininfantrats[4,20,22,47].
Inaddition,Oyaetal.
[46]reportedthatnon-shiveringthermogenesis(NST)isenhancedthroughBATactivityafterbirthinnewbornhumans.
Otherauthors[31,45]showedthatultrasonicvocaliza-tionsandthermogenesisarelinkedwhenratpupsareisolatedandexposedtovariousambientconditions.
SokoloffandBlumberg[21]moreoverinvestigatedcompetitionandcooperationamonglittersofratpupsbyinhibitingBATthermogenesisfor0,2,or4ratsineachhuddle.
InhibitionofBATthermogenesis,revealedbyIRT,indeedcompromisedtheabilitytomaintainhuddletemperatureduringacoldexposure.
Jacksonetal.
[32]howeverattemptedtouseIRTtocorrelatechangesinBATtemperaturewithenergyexpenditure.
Moreprecisely,theseauthorsattemptedtoquantifyNSTinBATofshort-tailedfieldvoles(cold-acclimatizedornot),followinganoradrenalineinjection,aBATactivator.
Theycouldnotfindanysignificantcorrelationbetweenchangesinsurfacetemperatureandthemetabolicpeakassociatedwiththenoradrenalineinjection.
Howeverinourstudy,TEEwashighlycorrelatedwiththegradientTBAT-Tback,indicatingthattheextentofBATnon-shiveringthermogenesisdirectlyinfluencedtheenergyexpenditureofthepup.
Theambienttemperaturesselectedinourprocedurewerechallengingbothforhuddlingandnon-huddlingnon-insulatedpups(alsosee[18,19,4–6]).
Thecoldtreatmentwashowevertoochallengingforisolatedpupsasheatlossoverwhelmedheatproduction.
Tocomparethethermogenicresponsesofinsulatedpups(olderthan5days)withthoseofnon-insulatedpups(frombirthupto5daysold),wechosetoselectsimilarambienttemperaturesforolderpups.
However,aspupsdeveloptheyincreasetheirbodymass(decreasetheirsurfacetovolumeratio)andgrowfur(increasetheirinsulationcapacities),theirthermo-regulatoryconstraintsandtheirthermoneutralzonechange[4].
PrivateHeatforPublicWarmthPLoSONE|www.
plosone.
org8March2012|Volume7|Issue3|e33553Hence,thetestedcoldchallengewasnotsufficientforhuddlinginsulatedpups:theywereparadoxicallyplacedinasituationofwarmchallengeinsteadofcoldchallenge,notablyduetothehighefficiencyofhuddlingandlocalheating.
Indeed,theireartemperatureexceededtheirbacktemperature(Figure6).
Sinceonlyisolatedinsulatedpupswerecoldchallenged,whilehuddlinginsulatedpupswerewarmchallenged,noneofthecorrelationsfornon-insulatedpupswerefoundsignificantfortheoldercategories.
Bodytemperatureshowednocorrelationwithsurfacetempera-tures,andinthesamewayTEEwasnotcorrelatedwiththegradientTBAT-Tback.
However,whencomparingisolatedinsulatedpupsatwarmandcoldtemperatures,theyseemtopresentthermogenicresponsessimilartonon-insulatedpups(Figure5&6),eventhoughtheymaintainedahigherbodytemperatureandpossessedlowersurfacetemperatures,presum-ablyduetogreaterinsulation.
Asenergysupplyislimitedforpups,extraenergyrequiredforthermallychallengedpupswillreduceenergyavailableforgrowthandpossiblyinfluencetheirshortorlong-termsurvivalandtheiradultperformances[40,48–50].
Inthecold(14uC),non-insulatedhuddlingpupshadbodyandsurfacetemperaturesthatwereequivalenttotheirisolatedlittermatesexposedtoawarmambienttemperature(23uC,Figure4A).
Moreover,inawarmenviron-ment,non-insulatedhuddlingpupsmaintainedaTEEequivalenttoolderinsulatedpupsthatwereisolated(Figure5).
Thesurfacetemperaturegradients(TBAT-TbackandTear-Tback)ofnon-insulatedhuddlingpupswereequivalenttoinsulatedisolatedindividuals(23uC,Figure6).
Hence,evenwhenpupsareyoungerthan5daysold,andnon-insulated,huddlingprovidesthem''publicinsulation''.
Huddlingnon-insulatedpupsthenreducetheirenergyrequirements,switchingfroma''non-insulatedstate''toapseudo''insulatedstate'',thankstothereductionintheircold-exposedbodysurfaceandthelocalheatingprovidedbyhuddling.
Huddlingthereforedrasticallymodifiesthermalconstraintsofpups,reducingthecoldchallengeandeffectivelyincreasingtheirdevelopmentalstageofthermoregulation.
HuddlingreducesBATthermogenesis,allowingpupstolowertheamountof''privateheat''giventothe''publicgood'',essentialforthemaximumallocationofenergyforgrowth.
Whenambienttemperatureswerechallenging,thepatternofthermogenicresponseswasequivalentwhenpupswereexposedtowarmorcoldtemperaturesinourstudy,butdifferedconsideringtheextentoftheresponse(Figure5).
Moreover,forcoldchallengingconditions,huddlingpupsrespondedsimilarlytoisolatedpups(significantregressionline,Figure5).
Huddlingthereforedidnotseemtomodifytheindividuals'thermogenicresponse,buttheextentoftheirresponse,andhencetheextentoftheenergyallocatedtomaintenanceoftheirbodytemperature.
Consideringtheseresults,wemayassumethatallpupsinvestedtothesameextentinthepublicgood,andthatpresumablynoselectiveforcesdrivesomeselfishless-relatedpupswithinalittertoinvestless''privateheat''intothe''publicwarmth''.
Inratpups,SokoloffandBlumberg[21]showedthattheinhibitionofBATthermogenesiscompromisedtheabilityofpupstomaintainhuddletemperature,butthisdidnotresultinenhancedhuddling.
TheyconcludedthattheheatprovidedbyBATappearedtoshapebehaviouralinteractionsinthehuddleduringdevelopment,sinceeffectivehuddlingduringcoldexposurerequiresthethermalresourcesprovidedbyBATactivation.
Inaddition,theabilityofindividualstoobtainaccesstowarmthwithinahuddleisnotrelatedtobirthweight,survival,milkintake,ormetabolicefficiency[11].
Itappearsfromthesestudiesthatneonatesshareoutthermallyadvantageouspositionsratherthancompetewithinthehuddle,astheycontinuallymovethroughit.
However,arecentstudyshowsFigure6.
Gradientsbetweenbrownadiposetissue(BAT)andbacksurfacetemperatures(6C),andearandbacksurfacetemperatures(6C)foreachpupcategory(G6-huddling,G1-isolated,non-insulatedandinsulatedpups).
ThepercentageincreaseintotalEnergyExpenditure(TEE)isbasedonFigure2.
OrangeandredcoloursrepresentBATandearsurfacetemperatureshigherthanbacksurfacetemperature(i.
e.
heatproductionorvasodilatation).
Bluecoloursrepresentearvasoconstriction.
Densershadingrepresentshighersurfacetemperaturegradients.
doi:10.
1371/journal.
pone.
0033553.
g006PrivateHeatforPublicWarmthPLoSONE|www.
plosone.
org9March2012|Volume7|Issue3|e33553thatpositionsofpupswithinthehuddlecouldpossiblybelinkedtodifferentpersonalities[51],andhencewoulddependonindividualphysiologicalandbehaviouralcharacteristics.
GiventhatitmaybepossibletoexploretheselectionofmaternalandpaternalgenescontrollingBATthermogenesis[12],furtherstudiesexaminingthebehaviourandphysiologicalperformancesofdifferentindividualswithinalittermayrevealnewinsightsintotheextenttowhichhuddlingisaltruistic,andtoinvestigateproximalfactorsthatdrivethedevelopmentofindividualpersonality[52].
ConclusionsProximalthermoregulatoryconstraintsthatultimatelygoverndevelopmentandsurvivalofrabbitpupswereexamined.
Huddling,through''publicwarmth'',drasticallymodifiesthermalconstraintsofpups,reducingcoldchallengesandeffectivelyincreasingtheirdevelopmentalstageofthermoregulation.
Sinceenergyislimitedforanorganism,variationsinitsthermoregu-latoryrequirementsimpactontheamountofenergyallocatedtogrowth.
Byprovidingpublicwarmthwhenpupsarethemostvulnerable,huddlingbufferscoldchallengesandensuresaconstantallocationofenergytogrowthbydelayingnon-shiveringthermogenesis.
Thisstudymaystimulatefurtherresearchfocusingontheenergeticimplicationsofcoldchallengesinotheraltricialinfants,andmoreparticularlybyinvestigatingindividualthermo-genicconstraintswithinalitterontheearlydevelopmentofpersonalitydifferencesbetweensiblings.
AcknowledgmentsWewouldespeciallyliketothankJohnCurrie(ScottishEnergyCentre,EdinburghNapierUniversity)foruseofthethermalimagingcameraandanalysissoftware.
WewouldalsoliketoaddressgratefulthankstoDr.
Lo¨cDesquilbetforhishelponstatisticalanalyses,andtotwoanonymousreviewersfortheirhelpfulcommentsonthemanuscript.
AuthorContributionsConceivedanddesignedtheexperiments:CGSBAADJM.
Performedtheexperiments:CGDJMSG.
Analyzedthedata:CGDJM.
Contributedreagents/materials/analysistools:DJMSB.
Wrotethepaper:CGDJM.
Providedacriticalsecondreadingofthemanuscript:SBSGAA.
References1.
MockDW,ParkerGA(1997)Theevolutionofsiblingrivalry.
OxfordUniversityPress,Oxford.
2.
HudsonR,TrillmichF(2008)Siblingcompetitionandcooperationinmammals:challenges,developmentsandprospects.
BehavEcolSociobiol62:299–307.
3.
BernardE,HullD(1964)Theeffectoftheenvironmentaltemperatureonthegrowthofnew-bornrabbitsrearedinincubators.
BiolNeonate7:172–178.
4.
BlumbergMS,SokoloffG(1998)Thermoregulatorycompetenceandbehavioralexpressionintheyoungofaltricialspecies–Revisited.
DevPsychobiol33:107–123.
5.
BautistaA,DrummondH,Martinez-GomezM,HudsonR(2003)Thermalbenefitofsiblingpresenceinthenewbornrabbit.
DevPsychobiol43:208–215.
6.
GilbertC,BlancS,GiroudS,TrabalonM,LeMahoY,etal.
(2007)Roleofhuddlingontheenergeticofgrowthinanewbornaltricialmammal.
AmJPhysiol293:R867–876.
7.
Ro¨delHG,HudsonR,vonHolstD(2008a)OptimallittersizeforindividualgrowthofEuropeanrabbitpupsdependsontheirthermalenvironment.
Oecologia155:677–689.
8.
GilbertC,McCaffertyD,LeMahoY,GiroudS,BlancS,AncelA(2010)Oneforallandallforone:theenergeticbenefitsofhuddlinginendotherms.
BiolRev85:545–569.
9.
StanierMW(1975)Effectofbodyweight,ambienttemperatureandhuddlingonoxygenconsumptionandbodytemperatureonyoungmice.
CompBiochemPhysiol51A:79–82.
10.
AlbertsJR(1978)Huddlingbyratpups:groupbehavioralmechanismoftemperatureregulationandenergyconservation.
JCompPhysiolPsychol92:231–245.
11.
BautistaA,Garcia-TorresE,Martinez-GomezM,HudsonR(2008)DonewborndomesticrabbitsOryctolaguscuniculuscompeteforthermallyadvantageouspositionsinthelitterhuddleBehavEcolSociobiol62:331–339.
12.
HaigD(2008)Huddling:brownfat,genomicimprintingandthewarminnerglow.
CurrentBiol18:R172–R174.
13.
JanskyL(1973)Non-shiveringthermogenesisanditsthermoregulatorysignificance.
BiolRev48:85–132.
14.
NedergaardJ,CannonB(1986)Brownadipose-tissuethermogenesisinneonatalandcold-adaptedanimals.
BiochemSocTrans14:233–236.
15.
NedergaardJ,GolozoubovaV,MatthiasA,ShabalinaI,OhbaK,etal.
(2001)LifewithoutUCP1:mitochondrial,cellularandorganismalcharacteristicsoftheUCP1-ablatedmice.
BiochemSocTrans29:756–763.
16.
CannonB,NedergaardJ(2004)Brownadiposetissue:functionandphysiologicalsignificance.
PhysiolRev84:277–359.
17.
NichollsDG,LockeRM(1984)Thermogenicmechanismsinbrownfat.
PhysiolRev64:1–64.
18.
DawkinsMJR,HullD(1964)Brownadiposetissueandtheresponseofnew-bornrabbitstocold.
JPhysiol172:216–238.
19.
HullD,SegallMM(1965)Thecontributionofbrownadiposetissuetoheatproductioninthenew-bornrabbit.
JPhysiol181:449–457.
20.
SokoloffG,BlumbergMS,AdamsMM(2000)AcomparativeanalysisofhuddlingininfantNorwayratsandSyriangoldenhamsters:DoesendothermymodulatebehaviorBehavNeurosci114:585–593.
21.
SokoloffG,BlumbergMS(2001)Competitionandcooperationamonghuddlinginfantrats.
DevPsychobiol39:65–75.
22.
SokoloffG,BlumbergMS(2002)ContributionsofendothermytohuddlingbehaviorininfantNorwayrats(Rattusnorvegicus)andSyriangoldenhamsters(Mesocricetusauratus).
JCompPsychol116:240–246.
23.
SpeakmanJR,WardS(1998)Infraredthermography:principlesandapplications.
Zoology101:224–232.
24.
McCaffertyDJ(2007)Thevalueofinfraredthermographyforresearchonmammals:previousapplicationsandfuturedirections.
MammalReview37:207–223.
25.
McCaffertyDJ,GilbertC,PatersonW,PomeroyPP,ThompsonD,etal.
(2011)Estimatingmetabolicheatlossinbirdsandmammalsbycombininginfraredthermographywithbiophysicalmodelling.
CompBiochemPhysiolA158:337–345.
26.
JilgeB,KuhntB,LandererW,RestS(2000)Circadianthermoregulationinsucklingrabbitpups.
JBiolRhythms15:329–335.
27.
JilgeB,KuhntB,LandererW,RestS(2001)Circadiantemperaturerhythmsinrabbitpupsandintheirdoes.
LabAnim35:364–373.
28.
WeirJB(1949)Newmethodsforcalculatingmetabolicratewithspecialreferencetoproteinmetabolism.
JPhysiol109:1–9.
29.
HammelHT(1955)Thermalpropertiesoffur.
AmJPhysiol182:369–376.
30.
MohlerFS,HeathJE(1988)ComparisonofIRthermographyandthermocouplemeasurementofheat-lossfromrabbitpinna.
AmJPhysiol254:R389–R395.
31.
BlumbergMS,StolbaMA(1996)Thermogenesis,myoclonictwitching,andultrasonicvocalizationinneonatalratsduringmoderateandextremecoldexposure.
BehavNeuroscience110:305–314.
32.
JacksonDM,HamblyC,TrayhurnP,SpeakmanJR(2001)Cannon-shiveringthermogenesisinbrownadiposetissuefollowingNAinjectionbequantifiedbychangesinoverlyingsurfacetemperaturesusinginfraredthermographyJThermalBiol26:85–93.
33.
Schmidt-NielsenK(1997)AnimalPhysiology.
Adaptationandenvironment.
FifthEdition.
CambridgeUniversityPress.
pp241–297.
34.
WhiteCR,SeymourRS(2003)Mammalianbasalmetabolicrateisproportionaltobodymass2/3.
PNAS100:4046–4049.
35.
HayesJP,SpeakmanJR,RaceyP(1992)Thecontributionsoflocalheatingandreducingexposedsurface-areatotheenergeticbenefitsofhuddlingbyshort-tailedfieldvoles(Microtusagrestis).
PhysiolZool65:742–762.
36.
ZarrowMX,DenenbergVH,AndersonCO(1965)Rabbit:frequencyofsucklinginthepup.
Science150:1835–1836.
37.
LincolnDW(1974)Suckling:atime-constantinthenursingbehaviouroftherabbit.
PhysiolBehav13:711–714.
38.
HudsonR,DistelH(1982)Thepatternofbehaviourofrabbitpupsinthenest.
Behaviour79:255–272.
39.
HeimT,HullD(1966)Thebloodflowandoxygenconsumptionofbrownadiposetissueinthenewbornrabbit.
JPhysiol186:42–55.
40.
Ro¨delHG,PragerG,Stefanski,vonHolstD,HudsonR(2008b)Separatingmaternalandlitter-sizeeffectsonearlypostnatalgrowthintwospeciesofaltricialsmallmammals.
Physiology&Behavior93:826–834.
41.
BautistaA,Garcia-TorresE,PragerG,HudsonR,Ro¨delHG(2010)Developmentofbehaviorinthelitterhuddleinratpups:within-andbetween-litterdifferences.
DevPsychobiol52:35–43.
42.
DrummondH,VazquezE,Sanchez-ColonS,Martinez-GomezM,HudsonR(2000)Competitionformilkinthedomesticrabbits:survivorsbenefitfromlittermatedeaths.
Ethology106:511–526.
43.
BautistaA,Mendoza-DeganteM,CoureaudG,Martinez-GomezM,HudsonR(2005)Scramblecompetitioninnewborndomesticrabbitsforanunusuallyrestrictedmilksupply.
AnimBehav70:1011–1021.
PrivateHeatforPublicWarmthPLoSONE|www.
plosone.
org10March2012|Volume7|Issue3|e3355344.
CoureaudG,SchaalB,CoudertP,RideaudP,Fortun-LamotheL,etal.
(2000)Immediatepostnatalsuckingintherabbit:Itsinfluenceonpupsurvivalandgrowth.
ReprodNutrDev40:19–32.
45.
BlumbergMS,EfimovaIV,AlbertsJR(1992)Thermogenesisduringultrasonicvocalizationbyratpupsisolatedinawarmenvironment:Athermographicanalysis.
DevPsychobiol25:497–510.
46.
OyaA,AsakuraH,KoshinoT,ArakiT(1997)Thermographicdemonstrationofnonshiveringthermogenesisinhumannewbornsafterbirth:itsrelationtoumbilicalgases.
JPerinatMed25:447–454.
47.
SokoloffG,BlumbergMS,BolineEA,JohnsonED,StreeperNM(2002)ThermoregulatorybehaviorininfantNorwayrats(Rattusnorvegicus)andSyriangoldenhamsters(Mesocricetusauratus):Arousal,orientation,andlocomotion.
JCompPsychol116:228–239.
48.
Ro¨delHG,BautistaA,Garcia-TorresE,Martinez-GomezM,HudsonR(2008c)WhydoheavylittermatesgrowbetterthanlighteronesAstudyinwildanddomesticEuropeanrabbits.
Physiology&Behavior95:441–448.
49.
MucinoE,BautistaA,JimenezI,Martinez-GomezM,HudsonR(2009)Differentialdevelopmentofbodyequilibriumamonglittermatesinthenewbornrabbit.
DevPsychobiol51:24–33.
50.
Ro¨delHG,vonHolstD(2009)FeaturesoftheearlyjuveniledevelopmentpredictcompetitiveperformanceinmaleEuropeanrabbits.
Physiology&Behavior97:495–502.
51.
Reyes-MezaV,HudsonR,Martnez-GomezM,NicolasL,Ro¨delHG,BautistaA(2011)Possiblecontributionofpositioninthelitterhuddletolong-termdifferencesinbehavioralstyleinthedomesticrabbit.
Physiology&Behavior104:778–785.
52.
HudsonR,BautistaA,Reyes-MezaV,MontorJM,Ro¨delHG(2011)Theeffectofsiblingsonearlydevelopment:apotentialcontributortopersonalitydifferencesinmammals.
DevPsychobiol53:564–574.
PrivateHeatforPublicWarmthPLoSONE|www.
plosone.
org11March2012|Volume7|Issue3|e33553