Relationships between in vitro lymphoproliferative responses and levels of contaminants in blood of free-ranging adult harbour seals (Phoca vitulina) from the North Sea

(1)

ContentslistsavailableatScienceDirect

Aquatic

Toxicology

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

Relationships

between

in

vitro

lymphoproliferative

responses

and

levels

of

contaminants

in

blood

of

free-ranging

adult

harbour

seals

(Phoca

vitulina)

from

the

North

Sea

Aurélie

Dupont

a

_,

_Ursula

_Siebert

b

_,

_Adrian

_Covaci

c,d

_,

_Liesbeth

_Weijs

c,d

_,

_Gauthier

_Eppe

e

_,

Cathy

Debier

f

_,

_Marie-Claire

_De

_Pauw-Gillet

g

_,

_Krishna

_Das

a,∗

a_Laboratory_of_Oceanology_–_MARE_Center,_University_of_Liège,_B6c,_allée_de_la_chimie_3,_B-4000_Liège_{(Sart-Tilman),}_Belgium

b_Institute_for_Terrestrial_and_Aquatic_Wildlife_Research,_University_of_Veterinary_Medicine_Hannover,_Foundation,_Werftstrasse_6,₂₅₇₆₁_Buesum,_Germany c_{Toxicological}_Centre,_University_of_Antwerp,_{Universiteitsplein}_1,₂₆₁₀_Wilrijk,_Belgium

d_Systemic_{Physiological}_&_{Ecotoxicological}_Research_(SPHERE),_Department_of_Biology,_University_of_Antwerp,_{Groenenborgerlaan}_171,₂₀₂₀_Antwerp,

Belgium

e_Center_for_Analytical_Research_and_Technology_(CART),_Inorganic_Analytical_Chemistry,_University_of_Liège,_B6c,_allée_de_la_chimie_3,_B-4000_Liège

(Sart-Tilman),Belgium

f_Unit_of_Nutrition_{Biochemistry,}_Catholic_University_of_Leuven,_Place_Croix_du_Sud_2/8,₁₃₄₈_{Louvain-la-Neuve,}_Belgium g_Mammalian_Cell_Culture_Laboratory_(GIGA-R),_University_of_Liège,_B6c,_allée_de_la_chimie_3,_B-4000_Liège_{(Sart-Tilman),}_Belgium

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received22January2013

Receivedinrevisedform18August2013 Accepted24August2013 Keywords: Phocavitulina Blood Contaminant Invitro Lymphoproliferation Immuneparameters

a

b

s

t

r

a

c

t

Invitrocultureofperipheralbloodleucocytes(PBLs)iscurrentlyusedintoxicologicalstudiesofmarine mammals.However,bloodcellsofwildindividualsareexposedinvivotoenvironmentalcontaminants beforebeingisolatedandexposedtocontaminantsinvitro.Theaimofthis studywastohighlight potentialrelationshipsbetweenbloodcontaminantlevelsandinvitroperipheralbloodlymphocyte proliferationinfree-rangingadultharbourseals(Phocavitulina)fromtheNorthSea.Bloodsamples of18individualswereanalyzedfortraceelements(Fe,Zn,Se,Cu,Hg,Pb,Cd)andpersistentorganic contaminantsandmetabolites(PCBs,HO-PCBs,PBDEs,2-MeO-BDE68and6-MeO-BDE47,DDXs, hexachlorobenzene,oxychlordane,trans-nonachlor,pentachlorophenolandtribromoanisole).Thesame sampleswereusedtodeterminethehaematologyprofiles,cellnumbersandviability,aswellasthe invitroConA-inducedlymphocyteproliferationexpressedasastimulationindex(SI).Correlationtests (Bravais-Pearson)andPrincipalComponentAnalysiswithmultipleregressionrevealednostatistically significantrelationshipbetweenthelymphocyteSIandthecontaminantsstudied.However,thenumber oflymphocytespermillilitreofwholebloodappearedtobenegativelycorrelatedtopentachlorophenol (r=−0.63,p=0.005).Inadultharbourseals,theinterindividualvariationsofinvitrolymphocyte pro-liferationdidnotappeartobedirectlylinkedtopollutantlevelspresentintheblood,anditislikely thatotherfactorssuchasage,lifehistory,orphysiologicalparametershaveaninfluence.Inageneral manner,experimentswithinvitroimmunecellculturesofwildmarinemammalsshouldbedesignedso astominimizeconfoundingfactorsinwhichcasetheyremainavaluabletooltostudypollutanteffects invitro.

1. Introduction

During these last decades, large-scale mortalities of marine mammals have been increasingly documented (Greenland and Limpus,2007;Osingaetal.,2012).Epizooticsand mass strand-ingeventshaveimpactednumerouscetaceanandpinnipedspecies (Colegroveetal.,2005;deSwartetal.,1995a;Duignanetal.,1995;

∗ Correspondingauthor.Tel.:+32043663321;fax:+32043665147. E-mailaddresses:Krishna.das@ulg.ac.be,Aurelie.Dupont@gmail.com(K.Das).

GreenlandandLimpus,2007;Hannietal.,1997;Lipscombetal., 1996;Roganetal.,1997).In1988and2002,twomajorepizootics due tothephocinedistemper virus(PDV) affectedtheharbour seal(Phocavitulinavitulina)populationinhabitingtheNorthSea (Härkönenetal.,2006;Mülleretal.,2004).Thepopulationofthe WaddenSeasituatedalongtheeastandsouth-eastcoastsofthe North Sea,suffered lossesuptoapproximately57% and47% in 1988and 2002,respectively(Dietzetal.,1989;Härkönenetal., 2006).Asaconsequence,thesemassdie-offshaveinitiated numer-ous pathological and toxicological studies(Das et al.,2008; de Swartetal.,1996;Rossetal.,1996b;Siebert etal.,2007,2012)

(2)

thatnotablydiscussedtheimpactofenvironmentalpollutantson marine mammal health. Indeed, sealsare exposed to a variety ofanthropogeniccontaminantsmainlythroughtheirdiet.Astop predators,theyareparticularlyatriskforbioaccumulating contam-inantsthatbiomagnifyupthetrophicfoodchain(DeGuiseetal., 2003).Infact,thechemicalcompoundsknowntobioaccumulate in harbourseal tissuesare toxictrace elementsincluding mer-cury(Hg)(Dasetal.,2003)andpersistentorganicpollutants(POPs) suchasorganohalogenatedcompounds(polychlorinatedbiphenyls –PCBs),thedichlorodiphenyltrichloroethanes(DDTs), polybromi-nated diphenyl ethers (PBDEs), hexachlorobenzene (HCB) and chlordanes(O’SheaandTanabe,2003;Reijnders,1980).

Evidencesuggeststhatvariousimmunotoxiccontaminants(e.g. PCBsandmetabolites)mighthavecontributedatleastpartlytothe severityandextentoftheepizootics(Ross,2002).Pollutantsmay indeedhavetoxic effectsonperipheralblood leucocytes(PBLs), including lymphocytes, monocytes and granulocytes, thereby diminishingdefenses against invadingpathogens (Lahvis et al., 1995;Levinetal.,2007;Mosetal.,2006;Nakatsuruetal.,1985). Exposuretopollutantswashypothesizedtobenotablyrelatedto thehighlevelsofparasiticorbacterialinfectionsobservedinseals founddeadorkilledduetosevereillnessbeforeandafterthe2002 epizootic(Siebertetal.,2007).

Bloodsamplingofmarinemammalsisconsideredasa conve-nientand minimallyinvasivemethod,relevantfor theanalyses ofpollutantsandbiomarkers(Habranetal.,2011,2013;Vanden Bergheetal.,2012),andforinvitroassays(Rossetal.,2003a).Invitro culturesofPBLsareindeedroutinelyusedtoassesstherelationship betweenenvironmentalcontaminantsandimmunealterationsin free-rangingmarinemammals(Cámara Pellissóetal.,2008; De Guiseetal.,1998;Kakuschkeetal.,2008;Levinetal.,2005,2007). Organic and inorganic pollutants were analyzed recently in harboursealsfromtheNorth Searevealinghighconcentrations comparedtootherspeciesorlocations(Dasetal.,2008;Habran etal.,2013;VandenBergheetal.,2012;Weijsetal.,2009a,b,c). PCBsoftenexceedthethresholdvaluesof20ngPCBs/gwetweight forsealblood(Weijsetal.,2009a)andof17␮gPCBs/glipidweight for marinemammalblubber (Jepsonet al.,2005; Siebert etal., 2012;Weijsetal.,2009b)abovewhichdeleteriouseffectsmightbe observed(Kannanetal.,2000).Nothresholdhasbeendocumented forothercompoundssuchaspesticides,brominatedcompounds or toxic metals in this species, but deleterious effects of POPs and metallic traceelementson theimmune systemhave been documentedinbothinvivoandinvitroexposuresinlaboratory animalsandwildlife(CámaraPellissóetal.,2008;Dufresneetal., 2010;KakuschkeandPrange,2007;Kakuschkeetal.,2005;Luebke etal.,1997;Rossetal.,1996c).Associationshavebeenestablished betweenenvironmentalcontaminantsandimmunotoxicityin free-rangingsealpopulations(Dasetal.,2008;deSwartetal.,1996; Dufresneetal.,2010;Rossetal.,1996a,b,c;vanLoverenetal.,2000) asdemonstratedforinstancebysignificantimpairmentofnatural killercellactivityandspecificT-lymphocyteresponsesinvitro(Ross etal.,1996b).Nevertheless,thecurrentknowledgeofthosecellular effectsremainsfragmentary.Moreover,theinfluencethattheblood contaminantspresentinfree-rangingharboursealsmayhaveonin vitroimmuneresponsesoncePBLsareisolatedandculturedin con-trolledconditionsisunknown.Thispointishighlyrelevantwhen cellviabilityandproliferationareevaluatedintheframeworkof invitroexposuretocontaminants.

The present study aims to highlight potential relationships betweenbloodpollutantlevelsandinvitroimmuneresponsesof free-rangingadultharboursealfromtheNorthSea.Wedetermined theinvitromitogen-inducedlymphocyteproliferationandabroad rangeofblood parametersincludinghaematologyproﬁles, lym-phocytenumbersandviability,essentialelementconcentrationsas wellasnon-essentialelement,persistentorganicpollutant(POP)

and naturally-produced methoxylatedpolybrominated diphenyl ether(MeO-PBDE)levels.

2. Materialandmethods

2.1. Bloodsamplecollectionandconservation

Bloodsamples were collectedfrom18 free-ranging harbour sealscaughtonsandbanksoftheGermanpartoftheWaddenSea: 3 fromKolumbus Loch(geographical coordinates54◦23 N and 8◦35E)inApril(spring)2010;and15onLorenzenplate(54◦25N and8◦38E)inSeptember(autumn)2010andApril(spring)2011 (Fig.1).Sealswerecapturedinseinenetsatinter-tidalhaul-out sites.Theyweretransferredtoindividualtubenetsandrestrained manuallyfor clinicalexamination.Sex,lengthandweight were determined beforebloodsamplings (Table1).Collection proce-duresweredesignedtopreservesampleintegrityandminimize samplecontaminationduringprocessingandhandling.All sam-pledanimalswereconsideredolderthan3years(adults)basedon lengthandweight(Weijsetal.,2009b).Duringthesampling proce-dure,theanimalswerecontinuouslyundercareoftwoexperienced veterinarians.Allindividualswerereleasedatthesiteofcapture aftercompletingtheinvestigations.Themaximumdetentiontime, betweenthetransferofallanimalsintubenetsandthereleaseof thelastone,wasonehour.Allanimalsappearedhealthyby vet-erinarycheck.Theywereinnormalnutritionalstatusandshowed nosignofdisease.Theﬁeldstudieswerecarriedoutunderthe rel-evantpermitsoftheNationalParkOfﬁceSchleswig-Holsteinand theanimalexperimentpermit(AZ312-72241.121-19).

Blood was drawn from the extradural venous sinus into sterile evacuated blood collection tubes using a 20ml syringe anda1.2mm×100mmneedle(Carromco,Hamburg-Norderstedt, Germany).BloodwastransferredintoserumtubesS-Monovette®

forPOPanalysisandin7.5mlS-Monovette® _tubes _treated_with

Lithium-Heparin (LH) for trace metal analysis (Sarstedt, Essen, Belgium).SerumforPOPanalysiswasisolatedbycentrifugation at 1500×g for20minat 21◦C (Multifuge 3S-R, Kendro, Ham-burg,Germany).Wholeblood(2ml)andserumsamples(2ml)were storedat−20◦_C_until_analysis.

Bloodsamplingswerealsousedtotesttheinvitrolymphocyte proliferation.Blood(15–30ml)wascollectedinCellPreparation TubeswithSodiumHeparin(CPTTM_,_BD _Vacutainer®_,_Plymouth,

UK)thatweregentlyinverted10timesandstoreduprightatroom temperatureuntilfurtherprocessing,withinmaximum5hofblood collection.Additionalbloodsamplesweretakentodeterminethe clinicalbloodchemistryandhaematologyproﬁles,andotherhealth parametersaspartofanongoinghealthmonitoring.

2.2. Haematologicalproﬁles

Thehaematologyproﬁles(whitebloodcells[WBC],redblood cells[RBC],haemoglobin[HGB],hematocrit[HCT],platelets[PLT]) weredeterminedforallwholebloodsamples(beforePBLisolation) usingascilVetabchaematologyanalyzer(scilAnimalCare Com-panyGmbH,D-68519Viernheim,Germany)whichwascalibrated forsealblood(Hasselmeieretal.,2008).

2.3. PBLisolationandconservation

PBLswereisolatedaccordingtotheprincipleofdensitygradient centrifugationonFicollHypaqueTM_(density:_1.077_g/ml)_contained

inCPTTM_(BD_Vacutainer®_), _using_a_modiﬁed_protocol_described

bydeSwartetal.(1993).ThebloodsamplescontainedinCPTTM

(BDVacutainer®₎_were_centrifuged_at₁₈₀₀_×_g_for₃₀_min_at_room

(3)

Fig.1. ThesamplingsitesLorenzenplateandKolumbusLocharesandbanksoftheWaddenSeaoffshoreofGermany.Theyaremarkedbystars().

by 10 gentle inversions to improve cell viabilityduring trans-portationatroomtemperature,asadvisedbythemanufacturer (BDBeneluxN.V.,Erembodegem,Belgium).TheFicollHypaqueTM

densityﬂuidcontainedin CPTubesTM _(BD_Vacutainer®₎_is

origi-nallyusedtoseparateperipheralbloodmononuclearcells(PBMCs, includinglymphocytesandmonocytes)andplasmafrom granu-locytesanderythrocytes.However,ﬂowcytometryscatterplots of cell suspensions indicated that all leucocyte subpopulations (i.e.,lymphocytes,monocytesandgranulocytes)werepresent(see point2.4,Fig.2).Aftertransportationtothecellculture labora-torywithin18h,thePBLswerepipettedintoseparatedvesselsand washed3timesinphosphatebufferedsaline(PBS,sterileﬁltered, BioWhittaker®_,_Lonza,_Verviers,_Belgium)._Cells_were_suspended_in

RPMI1640medium(BioWhittaker®_,_Lonza,_Verviers,_Belgium)

sup-plementedwith10%foetalcalfserum(Gibco®_,_Invitrogen,_Paisley,

UK),0.33%l-alanyl-l-glutamine(GlutaMAXTM_,_Gibco®_,_Invitrogen,

Paisley,UK),1%non-essentialamino-acids(BioWhittaker,Lonza, Verviers,Belgium),1%Napyruvate(BioWhittaker,Lonza,Verviers, Belgium)and 1%penicillin-streptomycin (100IUand 100␮g/ml, BioWhittaker, Lonza, Verviers, Belgium). This complete culture mediumwasfurtherreferredtoasculturemedium.Theviability

andnumberofcellsweredeterminedwithaNucleocounter®

NC-100TM_(Chemometec,_Allerød,_Denmark)._The_{NucleoCounter}®_is_an

integratedﬂuorescencemicroscopedesignedtodetectsignalsfrom theﬂuorescentdye,propidiumiodide(PI)boundtoDNAofcells withcompromisedcellmembranes.Thecellviabilityofsamples wasdeterminedusingthetotalcellcount(followingatreatment renderingallthecellnucleisusceptibletostaining)andthecount ofnon-viablecells(ChemoMetecA/S,2013).Aftercounting,cells werecentrifuged,andthepelletwasresuspendedinadetermined volumeofculturemediumaddedwith10%DMSO(dimethyl sulfox-ide,Sigma–AldrichChemieGmbH,Steinheim,Germany)at4◦Cto adjusttheirconcentrationto10×106_viable_cells/ml._The_cell

sus-pensionwastransferredintocryovials(1ml/vial)andplacedinto afreezingcontainer(MrFrosty, Nalgene)at−80◦_C_during₂₄_h,

beforebeingtransferredinliquidnitrogen. 2.4. PBLinvitrocultures

PBLcryovialswerequicklythawedandtransferredinto50ml Falcontubescontaining40mlofcompleteculturemedium pre-warmedat37◦C.Theywerecentrifugedat300×gduring10min,

Table1

Harbourseal(Phocavitulina)information(n=18;7females(F)and11males(M)).

Samplingperiod Harboursealnumber Gender Length(cm) Weight(kg)

April2010 1 F 163 73.5 2 F 168 92 3 F 170 96 September2010 4 F 150 54.5 5 F 158 61 6 M 160 66.5 7 M 160 63.5 8 M 166 67.5 9 F 167 79 10 M 170 63.5 April2011 11 M 148 35 12 M 149 43 13 M 160 52.5 14 M 162 45 15 M 163 85.5 16 M 170 45 17 F 170 82.5 18 M 179 102.5

(4)

Fig.2.Flowcytometryscatterplotofcellpopulationsobtainedafterwholeblood centrifugationinCPTTM_._Four_{subpopulations}_were_gated_based_on_forward_scatter

(FSC–size)andsidescatter(SSC–internalcomplexity)characteristics.ThegateA mostlikelyincludeslymphocytes.GatesBandCprobablyincludemonocytesand neutrophils,respectively.GateDpredominantlyincludesearlyapoptoticanddead cells,andthesquareintheleftcornerincludescelldebris.

thesupernatantwasdiscardedandthecellpelletwasresuspended

inwarmculturemedium.Monocyteswerepartiallydepletedby

plastic adherenceintissue cultureﬂasks (VentedCap,Sarstedt,

Newton, NC, USA) incubated during 1h at 37◦C in a

humidi-ﬁedatmosphereenrichedwith5%CO2.Thecellsuspensionwas

harvested,thelivingcells werecountedwitha nucleocounter©

(Chemometec, Allerød, Denmark) and the concentration was

adjustedto1×106 _cells/ml_(ﬁnal_{concentration).}_Cells_were_also

co-stainedwithpropidiumiodide(PI)andAnnexinV-ﬂuorescein

isothiocyanate(FITC)conjugate accordingtothemanufacturer’s

instructions(AnnexinV:FITCApoptosisDetectionKitI,BD,

Erem-bodegem,Belgium)and analyzedbyﬂowcytometryinorderto

determinetheproportionsofviable,earlyapoptoticanddeadcells

(inlateapoptosisandnecrosis).TheAnnexinVassaywasusedto

detectphospatidylserinetranslocationonthemembranesurface

ofcellsundergoingapoptosis(Vermesetal.,1995).Cellswere

dis-playedasscatterplotsbasedontheirsizes,estimatedbyforward scatter(FSC),andinternalcomplexity(granulosity),estimatedby sidescatter(SSC).TheanalysiswasconductedwithaFACSCantoTM

II(BD,Erembodegem,Belgium)flowcytometer.Thelightsource usedwasanair-cooledlasertunedto488nm.FITC-fluorescence (FL1)wascollectedthrougha502nmlongpass(LP)filterandthen througha530/30nmbandpassfilter,andPIfluorescencethrough a585/42nmLP filter(FL2),andthrough655and670nmLP fil-ters(680nm, FL3). Five-parameter (FSC-A, SSC-A,FL1,FL2, and FL3)dataofminimum50,000eventswerecollectedandanalyzed foreach sampleusingFACSDivasoftware version6.1.2.Scatter plotsobtainedbyflowcytometryshowedthatallleucocyte sub-populations(i.e.,lymphocytes,monocytesandgranulocytes)were present,andthatcellsuspensionswerecomposedof lymphocyte-enrichedleucocytes(Fig.2).AllcellswereanalyzedforAV-FITC and PI labelling.In alllikelihood, the gate A included lympho-cytes (mostlybut not exclusively) astheyare characterizedby smallsizesandlowcomplexitycomparedtotheotherleucocyte types(Fig.2).However,theexactpositionandquantificationof leucocytesubpopulationsinscatterplotswasnotpossibleasno specificorsuitablemarkerwasavailableforharboursealatthetime

ofexperiment.Monocyteandneutrophil(mostabundanttypeof granulocyte)gateswerethushypotheticallydrawn(gatesBandCin Fig.2)basedonsizesandcomplexitycharacteristicsaswellason lit-eraturedata(Pilletetal.,2002;Zharinovetal.,2006).Thepresence ofmonocytes(despitedepletion)andgranulocyteswasconﬁrmed bymicroscopyanalysisofcellsstainedwithGiemsa(Modiﬁed Solu-tion,FlukaChemika,Sigma–Aldrich,Steinheim,Germany)(datanot shown).Lymphocyte-enrichedleucocyteswerefurtherreferredto asPBLs.

ProliferationtoConcanavalinA(ConA,Sigma–AldrichChemie GmbH,Steinheim,Germany),aT-cell-speciﬁcmitogeninharbour seals(deSwartetal.,1993),wasassessedbyculturingPBLsina ﬂatbottom96-wellplate(transparent,Sarstedt,Newton,NC,USA) at2×105_cells_in₂₀₀_␮l_culture_medium_per_well_with_an

opti-malconcentrationof5␮g/mlConAinanincubatorat37◦Cunder humidiﬁedatmosphereenrichedwith5%CO2.ControlPBLswere

culturedinmediumaloneandeachcultureconditionwasin trip-licate.Theactivelyproliferatinglymphocyteswereassayedwith bromodeoxyuridine(BrdU)incorporationintonewlysynthesized DNA(BrdUProliferationAssay,Calbiochem®_,_Overijse,_Belgium).

TwotypesofcontrolswereusedfortheBrdUtests:blanks (cul-turemediumalone)andbackgroundcontrols(cellswithoutBrdU). Eachcultureconditionwasintriplicate.BrdUlabelwasaddedto wellsofthemicrotiterplateforthelast24hofculture.After72h ofincubation,cellswerefixed,permeabilisedandtheDNA dena-turedtoenableantibody(Ab)bindingtotheincorporatedBrdU. Detectoranti-BrdUmonoclonalAbwaspipettedintothewellsand allowedtoincubatefor1h,duringwhichtimeitbindstoany incor-poratedBrdU.UnboundAbwasthenwashedawayandhorseradish peroxidase-conjugatedgoatanti-mousewasaddedtobindtothe detectorAb.Thehorseradishperoxidasecatalyzedtheconversion ofthesubstratetoablueproduct,whichisthenconvertedto yel-lowaftertheadditionofthesolutionwhichstopsthereaction.The intensityofthecolouredreactionproduct,whichisproportional totheamountofincorporatedBrdUinthecells,wasquantifiedby measuringtheabsorbance(opticaldensity,OD)ineachwellusing aspectrophotometricplatereader(PowerWaveX,Bio-Tek Instru-ments,Inc.,Winooski,VT,USA)atdualwavelengthof450–540nm. Thesignalsofproliferatingcellsweresignificantlyhigherthanthe correspondingblanksandbackgroundcontrols.ForeachwellOD, theaverageblankabsorbancewassubtracted.Lymphocyte prolif-erationwasexpressedasastimulationindex(SI),calculatedasthe ratiooftheproliferationofmitogen-inducedstimulatedcells to thatofcellsculturedwithoutmitogen(Levinetal.,2005). SI=_ODOD_of_{corresponding}ofstimulatedcells_control

In addition, toconﬁrm thepresence of lymphoblastsin stimu-latedconditionsandthelackofcytotoxicityincontrolwells,the correspondingcellsuspensionswerestainedwithPIandAnnexin V(AnnexinV:FITCApoptosisDetectionKitI,BD,Erembodegem, Belgium)andanalyzedbyﬂowcytometryasdescribedabove. 2.5. Chemicalanalysis

2.5.1. Traceelements

Afterbeinglyophilized,approximately200mgofhomogenized wholebloodpowderweredigestedinTeﬂonjarwithconcentrated HNO3,H2O2andultrapurewater(Milli-Q®ultrapurewaterfortrace

elementanalysis,Millipore,Billerica,MA,USA)inamicrowaveoven (20minbetween0and600W,ETHOS,Milestones).Aftercooling, samplesweredilutedto50mlwithMilli-Qwaterinavolumetric ﬂask.ThetraceelementsCd,Fe,Ni,Cu,Se,Pb,andZnwereanalyzed byInductively CoupledPlasma-MassSpectrometry(ICPMS,Elan DRCII,PerkinElmer,Shelton,USA).Totalmercury(T-Hg)was deter-minedbyatomicabsorptionspectrophotometrywiththeDirect

(5)

MercuryAnalyzer(DMA80,Milestones,Sorisole,Italy). Concentra-tionsareexpressedin␮g/gdryweight(dw).Concurrentanalyses ofasetofcertiﬁedreferencematerials(CRM)andofprocedural blankswererepeatedthroughouteachsettoensuremethod’s accu-racyandprecision.ThecontrolmaterialsusedwereSeronormTM

Human,TraceElementWholeBloodLevel-3(SEROAS,Billingstad, Norway)and DOLT-3,DogﬁshLiverCertiﬁedReferenceMaterial forTrace Elements(National Research Council Canada,Ontario, Canada).Thehandlingandpreparationprocedureswereidentical forallsamplesandreferencematerialsanalyzed.

2.5.2. Organiccontaminants

Inallserumsamples,29PCBs(PCBs),18HO-PCBs(HO-PCBs), 7 PBDEs (PBDEs), 2 MeO-PBDEs (2-MeO-BDE68 and 6-MeO-BDE47), DDXs (o,p-DDT, p,p-DDT and metabolites o,p-DDD, p,p-DDDandp,p-DDE),hexachlorobenzene(HCB),oxychlordane (OxC),trans-nonachlor(TN),pentachlorophenol(PCP)and 2,4,6-tribromoanisole(TBA)weretargeted.Moreprecisely,investigated PCBswereIUPACnumbers(#)28,52,74,87,95,99,101,105,110, 118,128,138,146,149,153,156,170,171,172,174,177,180,183, 187,194,199,196/203,206,209.TargetedHO-PCBcongenerswere 3HO-CBs(#118,138,153and180),4HO-CBs(#79,107,120,127, 130,146,162,163,172,177,187,193and208)anddi-4HO-CB202. AnalyzedPBDEswereIUPAC#28,47,99,100,153,154and183.

Themethodforserumanalysiswasslightlyadaptedfromthe methodsdescribedpreviously(Weijsetal.,2009a).Avolumeof serum(1.5ml)wasspikedwithinternalstandards(PCB143,BDE 77and4-HO-CB159),diluted1:1withMilli-Qwater,mixedwith formicacid,sonicatedfor20minandextractedusingsolid-phase extraction(SPE)cartridges(6ml/500mgOasisHLB,Waters). Elu-tionwasdoneby10mlofMeOH:dichloromethane(DCM,1:1).After evaporationtoneardryness,theanalyteswerereconstitutedin 500␮lhexane:DCM(1:1)andfractionatedonsilicaSPEcartridges (3ml/500mg,Varian)toppedwith200mgacidsilica(44%,w/w).A ﬁrstfractioncontainingPCBsandPBDEswaselutedwith5ml hex-ane,whilethephenoliccompoundswereelutedinasecondfraction with6mlDCM(1:1,v/v).Bothfractionswereevaporatedto dry-ness.Theﬁrstfractionwasreconstitutedin100␮liso-octane,while thesecondfractionwasderivatizedfor30minwithdiazomethane toformMeO-PCBsandMeO-PBDEsafterwhichitwasevaporated todrynessandreconstitutedin75␮liso-octane.

FortheanalysisofPBDEsandMeO-derivatives,aGC-MS oper-ating in electron capture negativeionization (ECNI) modewas equippedwitha30m×0.25mm×0.25␮mDB-5capillarycolumn (J&WScientific,Folsom, CA,USA).TheMS wasused intheSIM modewithtwo ionsmonitoredfor each MeO-PCB congener in specificwindows,whileionsm/z=79and81weremonitoredfor MeO-PBDEs and for PBDEs during the entire run. For the PCB and organochlorine pesticides(OCP) analysis,a GC–MS operat-inginelectronimpactionization(EI)modewasequippedwitha 25m×0.22mm×0.25␮m HT-8capillary column(SGE). TheMS wasusedintheSIMmodewithtwoionsmonitoredforeachPCB homologuegrouporindividualOCPinspecificwindows. Multi-levelcalibrationcurves(r2_>_0.999)_in_the_linear_response_interval

ofthedetectorwerecreatedforthequantiﬁcation.Quality con-trol(QC)wasperformedbyregularanalysesofproceduralblanks, by random injection of standards, spiked samples and solvent blanks.TheQCschemeisalsoassessedthroughregular participa-tiontointerlaboratorycomparisonexercisesorganizedbyAMAP (POPsinhumanserum).Obtainedvaluesweredeviatingwith<20% fromconsensusvalues.Foranalytesdetectedinproceduralblanks, themeanproceduralblankvaluewasusedforsubtraction.After blanksubtraction,thelimitofquantiﬁcation(LOQ) wassetat3 timesthe standard deviation ofthe procedural blank.For ana-lytesnotdetectedinproceduralblanks(HO-PCBsandMeO-PBDEs), LOQswerecalculatedforasignal-to-noiseratio=10.Resultswere

expressedona perserum volumebasis asthe lipidfractionof serumiscomposedofmorepolarlipidswhencomparedto subcu-taneousfatorotherorgans(Hendersonetal.,1994),andcontains non-polarlipids(i.e.hexane-extractablefat)atlowerlevels(Debier etal.,2003).ExpressingPOPconcentrationsinserumonalipid weightbasis (i.e.hexane-extractablefat)wouldthus leadtoan overestimation(Debieretal.,2003).

2.6. Dataanalysis

StatisticalanalysiswasperformedusingStatistica9.1.(StatSoft, Maisons-Alfort,France).Eachcultureconditionwasintriplicatefor eachindividual.Opticaldensityvalues(BrdUtest)andcellcounts wereaveragedforeachindividual,givingonemeanvalueper cul-turecondition.Coefﬁcientsofvariations(CV)werecalculatedasthe ratioofthestandarddeviation(SD)tothemean.Comparisonsof theODmeansofcontrolandstimulatedPBLsafterthe72hof incu-bationwasassessedbytheWilcoxontestforpairedsamples.The normalityofalldatadistributionwasassessedbyvisual inspec-tionand using theShapiro–Wilk test. All dataincluded length, weight,haematologyvalues,cellnumbersandviability,SI,essential and non-essential element levels,and persistent organic pollu-tant(POP)andnaturallyproducedmethoxylatedpolybrominated diphenylether(MeO-PBDE)levels.Allvalueswerelog-transformed (log10)toachievenormaldistributions.Comparisonsofthemeans

oflymphocytestimulationindexesbetweensexes,locations, sea-sonsand yearsofsamplingwereperformedusingtheStudent’s t test (two-tailed),respecting thenormalityof distributionand equality of variances (Bartlett, p>0.05). When variances were unequal,theSatterthwaite’sapproximatettestwasused.Thelevel ofstatisticalsigniﬁcancewasdeﬁnedatp<0.05.Whennormalityof distributioncouldnotbeachieveddespitethelog-transformation, thenon-parametricMann–WhitneyUtestwasused.Assample sizesweresmall (n<20 for each group), theobtainedUvalues werecomparedtorespectivecriticalUvalues(AppalachianState University,2013).

Forchemicalelementmeasurements,eachvaluebelowLOQwas replacedwithavalueofLOQ/2andusedinthestatisticaltests.

Intheothertestsperformed(correlationtests,Principal Com-ponentAnalysisandregressions),pollutantswithmorethan50%of values<LOQweresuppressedasmostoftheirvalueswereidentical (LOQ/2),thuspresentingalowvariability.Afterlog-transformation, noneof theconsidered distributions differed signiﬁcantlyfrom normalityasshownbytheShapiro–Wilktests(allW>0.897and p>0.05).Correlationtests(Bravais-Pearson)wereusedto investi-gatepotentialrelationshipsbetweeninvitrolymphoproliferative responsesand peripheralbloodcontaminant andtraceelement concentrations two-by-two. Then, a more global approach was foreseen,consideringpollutantsascombinedcompoundsasthey canactasamixturewithsynergicandantagonisticbehaviours.A PrincipalComponentAnalysis(PCA)wasrealizedinordertoreduce thenumberoforiginalpollutantvariables.Then,thetwoﬁrstfactor scoresrevealedbythePCAwereusedinamultipleregressiontest astheindependentvariables,withSIandlymphocytenumbersas dependantvariables.

3. Results

3.1. Levelsoftraceelementsinwholeblood

Traceelementlevelswereallabovethedetectionlimitexcept thatforCd,forwhichsevensealssampledin2011exhibitednon detectableconcentrations(Table2).AgeneralLOQmeanvaluewas calculatedforallindividuals,andwhensomeindividualspresented concentrations below theirown LOQ, theirnumber wasadded

(6)

Table2

Medians,ranges(minima–maxima),andmeans±standarderrors(SE)oftraceelementsexpressedin␮g/gdwandin␮g/l(*exceptFe,inmg/l)ofwholeblood(n=18).

Concentrations(␮g/gdw) Concentrations(␮g/l)

Median Range Mean±SE Median Range Mean±SE

Fe 2682 2500–2822 2658±20 744* 647–1139* 772±33* Zn 11.5 9.8–13.2 11.4±0.2 3167 2956–4318 3276±88 Se 3.4 2.1–6.8 3.6±0.2 976 544–1884 1020±69 Cu 2.7 2.3–3.4 2.8±0.1 757 595–1088 801±36 Hg 1.1 0.7–1.5 1.1±0.1 305 167–661 322±26 Pb 0.032 <0.028–0.043 0029±0.003(9) 8.9 <8.3–16.9 8.6±1.0 Cd <0.006 <0.002–0.026 <0.006±0.001(12) <1.8 <0.6–8.5 <1.8±0.5

betweenparentheses.Forexample,Cdpresented12valuesbelow

itsLOQ,ofwhich7belowitsLOD.Theessentialtraceelements

exhibitedminorvariationsinconcentrationlevels,withrelative

standarddeviation(RSD)valueslowerthan20%forFe,Cu,andZn,

and28%forSe.TheirRSDexpressedgreaterinterindividual

varia-tionswith32%forT-Hg,55%forPband142%forCd.Themedian

Se/T-Hgmolarratiowas7.9(min.–max.=4.7–17.6).

Nosigniﬁcantdifferenceofmedians(Mann–WhitneyUtest)

was observed between genders, seasons of sampling

(spring-autumn) except for Cd and Fe (p<0.05), years of sampling

(2010–2011)exceptforCd(p<0.01),andplacesofsampling(KL-LP)

exceptforZn(p<0.01).

3.2. LevelsofPOPsinserum

Table3summarizesPCBs,HO-PCBs,DDXs,PCP,PBDEs, TBA,TN,OxCandHCBserumconcentrations(pg/ml).The contam-inantsfoundatthehighestconcentrationswerePCBs,constituting 86%ofthemeantotalPOPs,followedbytheirhydroxylated metabo-litesHO-CBs,andtotalDDT(DDXs)ofwhichthecontributionof p,p-DDEwasupto93%.ThePCBproﬁleof the18 animalswas CB153>CB138>CB187>CB146>CB99.TheHO-PCBproﬁlewas 4-HO-CB107>162>146>187.The4-HO-CB107,oneofthe low-estchlorinatedcompounds,indeedaccountedfor49%ofthesum inaverage(min.24–max.58%).Itsmedianconcentrationwastwo timeshigherthanthatof4-HO-CB162andtentimesthatof 4-HO-CB187.SomeHO-PCBs(3-HO-CB118,4-HO-CB127and3-HO-CB 180)werenotfoundinanysample,while4-HO-CB79and4-HO-CB 193weredetectedinlessthan25%ofallsamples.Themeanratio HO-PCBs/PCBsequalled0.10andtherewasagoodcorrelation betweenthetwogroupsofpollutants(r2₌_0.76).

For PBDEs,BDE 47 was thepredominantcongener followed byBDE 100.Trans-nonachlor and oxychlordanelevelsappeared variableamong individuals.Among extremevalues, 6 individu-alspresentedconcentrationsbeloworjustabovetheLOQ,and3 animalssampledinSeptember2010presentedconcentrations10 timeshigherthantheLOQ.HCBwas<LOQinmorethan65%ofthe samplesanalyzed.o,p-DDT,o,p-DDD,4-diHO-CB202,6-MeO-BDE 47and2-MeO-BDE68werebelowLOQforallindividuals.One individual(harboursealno.7–Table1)displayedhigherpollutant

Table3

Medians,ranges(minima–maxima),andmeans±standarderrors(SE)oforganic pollutantsmeasuredinserumofharbourseals,expressedinpg/ml(n=18).

Median(pg/ml) Range(pg/ml) Mean±SE(pg/ml)

PCBs 28,863 7498–102,819 34,315±6051 HO-PCBs 2326 1269–4129 2398±212 DDXs 2234 649–6377 2739±430 PCP 134 62–293 155±16 PBDEs 94 46–249 125±18 TBA 79 33–166 84±9 TN 65 13–245 92±18 OxC 48 <10–262 79±19 HCB <20 <20–35 <20±3

loadsthantheotherindividuals.ItsPCBsconcentrationwason

average5.5timeshigherthantheoneoftheotheranimals,while

itsPBDEslevelwasonaverage3timeshigherthanintheothers.

NostatisticaldifferenceofPOPmedians(Mann–WhitneyUtest)

wasobservedbetweenmalesandfemales forPCB28

(trichloro-biphenyl), tetraCBs, pentaCBs, octaCBs, PCB 206 and 209

(nona-anddeca-CB,respectively),HCB,HO-CBandPCP(p>0.05).

On the contrary, hexaCBs (p=0.030), heptaCBs (p=0.046),

PCBs (p=0.046), OxC, TN, DDXs, BDEs and TBA had

p-values<0.05.

Nostatisticallysigniﬁcantdifferencewasobservedbetweenthe

seasonand yearofsampling,exceptfor PCP(p<0.01)and HCB

(p<0.05).

3.3. Haematologicalproﬁles

Medianvalueswerecalculatedtakingtheseasonofsampling

intoaccountassomehaematologicalvaluesvarywithit(Table4;

Hasselmeieretal.,2008).Indeed,theMann–WhitneyUtestshowed thatthenumbersoflymphocytes,monocytes,RBCandHCTwere signiﬁcantlydifferentbetweenthetwoseasonsconsidered(U<16, p<0.05).Themedianvaluesforspringwere7900neutrophilsper blood millilitre(min. 6400–max.10,500), 1600lymphocytes/ml (min.900–max.2300)and500monocytes/ml(min.300–max.900). Inautumn,themedianvalueswere6800neutrophilsperblood ml(min.4400–max.9400),3000lymphocytes/ml(min.2500–max. 5400)and200monocytes/ml(min.100–max.200).

This distinction between seasons was helpfulto adequately evaluatethehealthparameterslinkedtohaematologicalproﬁle, ascomparedtoliteraturedata(Hasselmeieretal.,2008).

3.4. PBLinvitrocultures

ThelymphocyteproliferationresponsestoinvitroConA stim-ulationwerecharacterized for18 harbour sealsfromthesame populationafter72hofincubation.Theviabilityofcellsisolatedby densitygradientcentrifugationwasalwaysabove96.7%(median 98.9%)afterthawing.After72hofincubation,lymphocyte prolif-erationinpresenceofConAwasdemonstratedbytheBrdUtest analysis.TheﬂowcytometryAV-PIlabellingtestsshowedthatthe majorityofthelymphocyteswereoutoftheapoptoticandnecrotic pathwaysbeforeandaftertheincubationperiods.

Asregardsthestimulationindex,nocorrelationbetweenthe latterandbodylengthorweightwasdetected(p>0.05).Alldata werepooledastherewasnosigniﬁcantdifferenceinstimulation indexesbetweengenders,seasons,yearsandplacesofsamplings. Despiteastronginterindividualvariability,lymphocytesofall indi-vidualsshowedstrongstimulationresponses,withmeanvalues reachingatleast3timestheoneobtainedforthecorresponding control(Fig.3).Opticaldensityvalueswerefoundtobe0.09±0.01 (mean_±SE;median:0.07,min.–max.:0.01–0.24)inunstimulated control,comparedtoConA-stimulatedconditionswhichaverage valuewas0.9±0.08(mean±SE;median:0.9,min.–max.:0.3–1.5).

(7)

Table4

Medians,ranges(minima–maxima),andmeans±standarderrors(SE)ofbloodcellparametersofharbourseals(n=18)basedontheseasonofsampling(spring=April, autumn=September).RBC=redbloodcells;HGB=haemoglobin;HCT=haematocrit;PLT=platelets.

Bloodcellparameters Spring(n=11) Autumn(n=7)

Median Range Mean±SE Median Range Mean±SE

Lymphocytes(G/l)** _1.6 _0.9–2.3 _1.6_±_0.04 _3.0 _2.5–5.4 _3.5_±_0.2 Monocytes(G/l)** _0.5 _0.3–0.9 _0.6_±_0.02 _0.2 _0.1–0.2 _0.2_±_0.01 Neutrophils(G/l) 7.9 6.4–10.5 8.3±0.13 6.8 4.4–9.4 6.8±0.2 RBC(million/mm3₎* _5.0 _4.6–5.3 _5.0_±_0.02 _4.5 _4.3–5.1 _4.6_±_0.04 HGB(g/dl) 20.9 17.4–24.8 21.1±0.2 19.4 19.0–21.6 20.2±0.2 HCT(%)** _59.2 _55.6–64.3 _59.5_±_0.2 _51.9 _50.8–57.8 _53.9_±_0.5 PLT(thousand/mm3₎ ₃₅₈ _324–437 ₃₇₂_±₄ ₄₂₀ _334–476 ₄₁₆_±₆

* _Signiﬁcant_difference:_U_<_16,_and_p_<_0.05. ** _Signiﬁcant_difference:_U_<_10,_and_p_<_0.01.

Fig.3.Stimulationindex(SI)meanvaluesoftriplicateswithcoefﬁcientof varia-tionbarsfortheBrdUproliferationassayofharbourseallymphocytesafter72hof incubation(n=18).

3.5. Relationshipsbetweenlymphocyteparametersand contaminantlevels

Analyses ofin vitro lymphocyteSI suggested nostatistically signiﬁcantcorrelationwiththedifferenttraceelements,organic contaminantspeciesandsums studied(p>0.05,e.g.,correlation coefﬁcientofSIandPCBs=0.14,p=0.58).Ontheotherhand,the numberoflymphocytespermlofwholebloodappearednegatively correlatedtoPCB209(r=−0.51,p=0.03),4-HO-CB208(r=−0.53, p=0.02)andPCP(r=−0.63,p=0.005).

Tofurtherinvestigatethoserelationships,aPCAbasedonthe correlationmatrixwasrealizedwiththemetallicpollutantT-Hg andallsumsofPOP,exceptPbandHCBforwhichatleast50%of valueswere<LOQ,andwerenotnormallydistributedevenafter log-transformation.Thecorrelation-basedPCArevealedtwo prin-cipalcomponentsexplaining81.0%ofthetotalvariance.Thetwo ﬁrstcomponentloadsaresummarizedin Table5.Thevariables

Table5

PollutantloadsforthetwoﬁrstcomponentsofthePCA.

Component1 Component2 T-Hg(log) −0.49 −0.71 PCBs(log) 0.94 −0.13 OxC(log) 0.94 −0.04 TN(log) 0.97 0.07 DDXs(log) 0.97 0.03 PBDEs(log) 0.97 −0.02 HO-PCBs(log) 0.86 −0.46 PCP(log) 0.30 −0.50 TBA(log) 0.41 0.68

Varianceexplained(cumulative) 64.7% 81.0%

Fig.4.PC1andPC2loadingplot.Theninepollutantvectorsoflog-transformed concentrationsrepresenttheBravais-Pearsoncorrelationswiththeﬁrstandsecond principalcomponents.

PCBs, OxC,TN, DDXs,PBDE andHO-PCBs loaded

princi-pallyinPC1(allloadingspositiveandatleast0.86),whereasTBA

loadedmostlyonPC2(loading=0.68).T-HgloadedmostlyonPC2

butpresentedanegativevalue.Theloadingplot(Fig.4)highlighted

distinctivefeaturesofHg-TandTBAcomparedtotheotherPOPs ofwhichbloodlevelswerecorrelatedwithoneanother,indicating multipleexposure.

Multipleregressionsofthetwoprincipalcomponentswiththe lymphocyte stimulation indexand number of lymphocytesper mlofwholebloodrevealedasigniﬁcantpositivecorrelationonly betweenPC2andlymphocytenumbers(Table6,p-value<0.05).

Acorrelationtestwasalsoperformedgroupingthe3dioxin-like PCBcongeners(IUPAC#105,118,156)andthe26non-dioxin-like PCBcongeners(IUPAC#28,52,74,87,95,99,101,110,128,138, 146,149,153,170,171,172,174,177,180,183,187,194,196/203, 199,209)together.Nosigniﬁcantcorrelationwasfoundbetween theSIandoneofthosetwogroups(p=0.98and0.58,respectively), evenwhenconsideringmalesandfemalesseparately.

Table6

Multipleregressionresultsfortherelationshipbetweenthetwoﬁrstprincipal com-ponentsofthePCAandthelymphocytestimulationindex(SI),andnumberof lymphocytesperbloodmillilitre(×106_/ml).

t-Value p-Value SI(log) Component1 0.64 0.53 Component2 −0.84 0.41 Lymphocytenumber (log) Component1 0.69 0.50 Component2 4.43 0.00

(8)

3.6. Relationshipsbetweenbodyweightandcontaminantlevels

ThebodyweightwassigniﬁcantlycorrelatedtoPCB99

repre-senting65%ofpentaCB(r=−0.48,p=0.044),hexaCB,heptaCB,

octaCB, PCB 206 (nonaCB) and 209 (decaCB), and PCBs

(r=−0.57, p=0.015), DDXs (r=−0.53, p=0.025), BDE 100 and

HO-PCBs(r=0.5,p=0.04).Bodyweightwasnotsigniﬁcantly

cor-relatedtotheSI(p<0.05).

4. Discussion

During theselastdecades, theharboursealpopulationfrom

the North Sea hasbeen depleted by two epizootics caused by

phocinedistempervirus(Härkönenetal.,2006;Mülleretal.,2004).

Severalstudieshavethereforeinvestigatedthehealthand pres-enceofcontaminantsinharboursealsfromthisarea(Reijnders, 1980,1981;Siebertetal.,2012;Weijsetal.,2009b,c). Concomi-tantdevelopmentofhealthmonitoringprogrammesandinvitro assayscircumventedthelackofexperimentaldataabout contam-inantexposureand broughtinformationonpotentialeffects,as wellasonbiochemicalpathwaysoftoxiccompounds(deSwart etal.,1995b,c;Ross,2002;Rossetal.,2003a,b;Vosetal.,2003). Theuseofinvitroperipheralbloodleucocytesoffersawiderange ofapplicationsbutonedrawbackisthatthesecellshavealready beenexposedtoamixtureoforganicandinorganiccontaminants inthebloodofmarinemammals.Indeed,harboursealsofthe Wad-denSeadisplayedhighbloodlevelsofcontaminantsinthepresent study(PCBs,HO-PCBs,DDXsandT-Hg)(Tables2and3). Con-centrationsoforganiccontaminants(PCBs,HO-PCBsDDXs, PBDEs,MeO-PBDEs)andtraceelements(Hg,Cd,Pb,Zn,Cu,Fe) analyzedinthebloodofthese18harboursealsarefromthesame orderofmagnitudethanthosefoundinthelastfewyearsin individ-ualsfromthesamesealpopulation(Dasetal.,2008;Grieseletal., 2006,2008;Kakuschkeetal.,2008;Weijsetal.,2009a).Theproﬁle ofPCBs(CB153>CB138>CB187>CB146>CB99)analyzedinthe serumwassimilartothatdescribedpreviouslyforharbourseals collectedbetween2006and2008inthesamegeographicalarea (Weijsetal.,2009a),andforgreysealsfromtheNorthSeawhere PCB153and138werethemaincontributors(VandenBergheetal., 2012).

Bloodsamplesprovideawindowintoananimal’scomplex net-workofimmunefunction(DeGuise,2002).Haematologyprofiles ofthe18sealswerecomparabletoapreviousstudyonfree-ranging andcaptiveharboursealsfromtheWaddenSea(Hasselmeieretal., 2008).However,thelymphocytenumbersinautumnwerehigher comparedtothedatagivenbyHasselmeieretal.(2008),butstay intherangegivenforspring.Oneindividual(harbourseal num-ber4–Pv4)presentedamuchhigherautumnconcentrationwith 5400lymphocytes/ml.In allanimals ofthis study,the erythro-cytehaematologyprofilewasgenerallymorehomogeneousthan theleucocyteprofile,whichis moreinfluencedbyphysiological changes(e.g.,bacterialorviraldiseases,stress)(Hasselmeieretal., 2008).Thelifehistoriesoftheseindividualsareunknownasthey arelivinginthewildbutitisknownfrompathological investiga-tionsthatthosefree-ranginganimalshavemildinfectiousdiseases, whichexplainsthevariabilityobservedinbloodleucocyteprofiles (Siebertetal.,2007).

Lymphoproliferativeresponse,asaspecificT-cellresponse,can alsogivevaluableinformationontheadaptiveimmuneresponse. Despite the variability related to samples obtained from free-rangingharbourseals,aclearstimulationof theTlymphocytes associatedwithasignificantproliferationcouldbeobserved(Fig.3). Theinvitrostimulationindex(SI),currentlyusedincellculture, showed significantinterindividual variation. Nevertheless, high inter-andintra-individualvariationsoflymphocyteproliferation

alsooccurin humansin which thewholelifehistory is gener-allyknown,andinwhichhomogeneousgroupsweretested(e.g. Stoneetal.,2009).Ourresultsdidnothighlightanyclear corre-lationbetweenthestimulationindicesandthepollutantsorclass ofpollutantsstudied(component1,Table5).Incontrast,a pre-viousstudyshowedthatinvitroConA-inducedperipheralblood lymphocyteproliferationwasinverselycorrelatedtowholeblood levelsof tetra- toocta-CBs,p,p-DDT,o,p-DDEand p,p-DDEin a smallgroup(n=5)ofbottlenose dolphins (Tursiopstruncatus) (Lahvis et al., 1995). However, thevery small sample size and largeage range of those animals(3–32 years old, approximate age) precluded drawing extensive conclusions. Studies on har-boursealpupshighlightedmodulatingeffects ofPCBsfromthe blubber (Levinet al.,2005;Mos etal., 2006).Mos etal.(2006) showed that the proliferative responses of T lymphocytes, but not ofBlymphocytes(mimickedby stimulationwiththe poly-clonalactivatorsConAandlipopolysaccharide(LPS),respectively), weresignificantlydecreasedinharboursealpupspresentinghigher PCB concentrations in their blubber. On the other hand, Levin etal.(2005)reportedasignificantbodyweight-independent posi-tivecorrelationbetweenbothT-cellmitogen(phytohemagglutinin [PHA])-andB-cellmitogen(LPS)-inducedlymphocyteproliferation and the blubber concentrations of 11 PCBs, but no signifi-cantcorrelationswerefoundbetweenConA-inducedlymphocyte proliferationandtotalPCBconcentration(Levinetal.,2005). How-ever,thecorrelationsobservedbetweenlymphocyteproliferation indices and blubber or blood PCB levelsare different as these two tissuesdiffer in OC affinitiesand retention capacities. This isbecausetheirlipidcompositionis differentandbecause con-taminantshavedifferentlipophilicproperties(AddisonandBrodie, 1987;Sinjarietal.,1996),resultinginaselectiveretentionofthe morelipophiliccompoundsintheblubberandthemorepolar com-poundsintheblood(Lydersenetal.,2002).

Intheframeworkofinvitroexposuretocontaminants,Neale et al. (2002) showed no significantdepression in harbour seal lymphoproliferationwhenexposedto10␮MCB-156andCB-80. Other studies of in vitro exposure tosome OC mixtures (PCBs andTCDD)haverevealedsignificantpositivecorrelationswiththe ConA-inducedTcellproliferationintheharbourseal,Northernfur seal,Stellersealion,belugawhale,Commerson’sdolphinandsea otter(Levinetal.,2007;Morietal.,2006).Onthecontrary, signifi-cantdecreasesinlymphocyteproliferationwereseenaftermercury exposureinharbourseals(Dasetal.,2008;Dufresneetal.,2010) andafterOCmixtureexposureinmice(Morietal.,2006),for exam-ple.Theinvitroexposureofharboursealgranulocytesto12␮M ofBDE-47,-99and -153had alsobeenshowntoinducea thiol depletionandphagocytosisinhibition(Frouinetal.,2010).

In the present study, the number of activated lymphocytes appearedsigniﬁcantlynegatively correlatedtoonlya fewblood pollutants (PCB 209, 4-HO-CB208 and PCP) but this cannot be clearlydiscussedforPCB209and4-HO-CB208astheyareminor compounds.Incontrast,PCP,awidelyusedpesticideubiquitous intheenvironment,hadbeendemonstratedtohavesuppressive effects on human lymphocytes in vitro (Lang and Mueller-Ruchholtz, 1991) as well as on diverse cellular and humoral immuneparameters(Danieletal.,1995,2001).Serumlevelsare notavailableforcomparisoninmarinemammals,buthumans usu-allypresentmuchlowerconcentrations(ButteandHeinzow,1995). PCPshouldthusbemonitoredinmarinemammalsinordertoﬁnd outmoreaboutitsbiologicaleffects.

Thefactthatnoclearcorrelationcouldbefoundbetweenthein vitroSIandthecontaminantspresentinbloodoffree-rangingadult harboursealsatthattimeoftheirannualcycledoesnotmeanthat chemicalshavenotoxiceffectonPBLs.Inalllikelihood,thereare morecomplexinteractionsbetweenconfoundingfactors explain-ingimmunological variationsobservedinvitro.Inanycase, the

(9)

resultsobtainedininvitroassaysshouldbeinterpretedcautiously asPBLsareisolatedfromenvironmentallycontaminatedharbour seals.Generallyspeaking,factorssuchaspreciseage,nutritional status,geneticfactors,stress,disease,nursing,orfasting,mayhave additionalinﬂuencesonimmunefunctions,therebylimitingthe directabilitytodetecttheeffectsofcontaminantsontheimmune system(Levinetal.,2005).Thosefactorsshouldthusbetakeninto consideration,withintherealmsofpossibility.Inthisstudy,all indi-vidualswereadultsinagoodbodycondition,presentingnosign ofdisease.Yetminorinjuriesorinfections andstressmayhave inﬂuencedthelymphocyteresponses.

Furthermore,contaminantsweremeasuredinserum(organic pollutants)orwholeblood(metals)andnotinblubberorinliver, both tissues known to accumulate these contaminants over a lifetime(Weijset al.,2011)incontrast toblood(Habranet al., 2013). However,correlations couldbe made betweenlevels of organochlorinesfoundinbloodandblubber(Lydersenetal.,2002). Phocidsealsregularlygothroughfastingorreducedfoodintake periodsandmostoftheirmasslossduringtheseperiodsisderived fromcatabolismoflipidsfromtheirinnerblubberlayer(Nordoy andBlix,1985).Inharbourseals,themobilizationoflipidsfrom theblubbernotablyoccursinindividualshavinglowerfoodintake duringthebreedingandmoultingseasons(Burns,2008),inpups duringthepostweaningfast(MuelbertandBowen,1993)andin femalesproducingmilk(Bonessetal.,1994).Lactationisindeed particularlydemanding,leadingtoimportantlossofbodyfatand energy(Bowenetal.,1992).Bloodsamplesfromasealwhichis inastateofpositiveenergybalancewouldthushavePOPlevels reflectingthoseinthediet,whilebloodlevelsduringfastingand lactatingperiodsarereflectinginnerblubberlevelsof bioaccumu-latedlipophiliccontaminants.Moreover,sealsinaprimecondition generallyhavelowerconcentrationsthananimalsinapoor con-dition(Lydersenet al.,2002).Harboursealsof this studywere sampledwhileinagoodconditionbeforethebreedingseasonand afterthemoultingseason,andnotwhilein apoorconditionas duringmoultforexample.Theirbloodlevelscouldthusreflectlow contaminantconcentrationscomparedtothoseoffastingandmilk productionperiods.

Resultsobtainedinourstudyhighlightedthefactthatvariations inproliferationcouldnotbedirectlylinkedtothebloodpollutants investigatedinadultharboursealsinagoodbodycondition,but thatotherfactorscouldmostlikelyhaveaninﬂuence.Field stud-iesdothusallowtheopportunitytostudythedirectandindirect effectsof real-worldexposurestoPOPs ontheimmunesystem whenconfoundingfactorsareminimized(Levinetal.,2005). 5. Conclusions

Weprovideherearelevantapproachtogetabetter understand-ingoftherelationshipsbetweentheimmunefunctionandseveral organicandinorganiccontaminantspresentinthebloodof free-rangingadultharboursealsfromtheNorthSea. Thisisthefirst studyreportingthesimultaneousmeasurementofblood haematol-ogyparameters,traceelements,PCBs,PBDEs,naturallyproduced MeO-PBDEs,DDXs,OxC, TN,HCBand themetabolitesHO-PCBs, inbloodofhealthyharbourseals,concomitantlytothestudyof lymphocyteproliferation invitro.Significantlymphocyte prolif-erationwasobservedandresultshighlighteda clearcorrelation betweenthenumberoflymphocytesperbloodmillilitreandPCP, butnotbetweenthelymphocytestimulationindicesandblood con-taminantlevels.Thiscorroboratesthefactthatperipheralblood lymphocytesisolatedfromwildharboursealsandculturedin con-trolledconditionsconstituteavaluabletoolforinvitrostudies,as longasconfoundingfactorsarelimited.Nevertheless,astudy car-riedoutona biggersamplesizewouldbenecessarytoconfirm thoseresults.

Acknowledgements

AurélieDupontwassupportedbyaFRIAgrant(Fondspourla for-mationàlarecherchedansl’industrieetdansl’agriculture).Krishna DasisaF.R.S.–FNRSResearchAssociate(FondsdelaRecherche Sci-entifique).ThisstudywassupportedbytheFRFCGrant#2.4502.07 (F.R.S.-FNRS).Thesealcapturewasconductedunderthefundingof theMinistryofRenewableEnergies,Agriculture,Environmentand RuralAreasofSchleswig-Holstein,Germany.LiesbethWeijsand AdrianCovaciacknowledgefinancialsupportfromFWO-Flanders (FoundationforScientificResearch).Theauthorswishtothankall peopleinvolved,fortheirhelpinthefieldtrials,samplingand pro-cessingofbloodanalyses.TheauthorsarealsogratefultoRenzo Biondo(LaboratoryofOceanology,ULg)foritsvaluabletechnical assistanceintraceelementanalysis,andtoRaafatStephanforhis helpin flow cytometry(Imaging andFlowCytometryplatform, GIGA,ULg).ThisisaMAREpublication.

References

Addison,R.F.,Brodie,P.F.,1987.Transferoforganochlorineresiduesfromblubber throughthecirculatorysystemtomilkinthelactatinggreysealHalichoerus grypus.Can.J.Fish.Aquat.Sci.44,782–786.

AppalachianStateUniversity,2013.CriticalValuesoftheMann-WhitneyU (Two-Tailed Testing) [Data ﬁle], Retrieved from: http://www.lesn.appstate.edu/ olson/statdirectory/Statistical%20procedures/MannWhitney%20U%20Test/ Mann-Whitney%20Table.pdf

Boness,D.J.,Bowen,W.D.,Oftedal,O.T.,1994.Evidenceofamaternalforagingcycle resemblingthatofotariidsealsinasmallphocid,theharborseal.Behav.Ecol. Sociobiol.34,95–104.

Bowen,W.D.,Oftedal,O.T.,Boness,D.J.,1992.Massandenergytransferduring lactationinasmallphocid,theharborseal(Phocavitulina).Physiol.Zool.65, 844–866.

Burns,J.J.,2008.HarborsealandspottedsealPhocavitulinaandP.largha.In:Perrin, W.F.,Wursig,B.,Thewissen,J.G.M.(Eds.),EncyclopediaofMarineMammals., seconded.AcademicPress,SanDiego,CA,pp.533–542.

Butte,W.,Heinzow,B.,1995.ReferenzwertederKonzentrationanPentachlorphenol inSerumundUrin(deutsch).KlinischesLabor12,31–35.

CámaraPellissó,S.,Mu ˜noz,M.J.,Carballo,M.,Sánchez-Vizcaíno,J.M.,2008. Determi-nationoftheimmunotoxicpotentialofheavymetalsonthefunctionalactivity ofbottlenosedolphinleucocytesinvitro.Vet.Immunol.Immunopathol.121, 189–198.

ChemoMetec A/S, 2013. NucleoCounter NC-100 concept. Cell via-bility, Retrieved from: http://www.chemometec.com/en-GB/ Products/NucleoCounter-NC-100/NC-100-Concept/Cell-Viability

Colegrove,K.M.,Greig,D.J.,Gulland,F.M.D.,2005.Causesoflivestrandingsof north-ernelephantseals(Miroungaangustirostris)andpaciﬁcharborseals(Phoca vitulina)alongthecentralCaliforniaCoast,1992–2001.Aquat.Mamm.31,1–10.

Daniel,V.,Huber,W.,Bauer,K.,Opelz,G.,1995.Impairedin-vitrolymphocyte responsesinpatientswithelevatedpentachlorophenol(PCP)bloodlevels.Arch. Environ.Health50,287–292.

Daniel,V.,Huber,W.,Bauer,K.,Suesal,C.,Mytilineos,J.,Melk,A.,Conradt,C., Opelz,G.,2001. Associationofelevatedbloodlevelsofpentachlorophenol (PCP)withcellularandhumoralimmunodeﬁciencies.Arch.Environ.Health56, 77–83.

Das,K.,Debacker,V.,Pillet,S.,Bouquegneau,J.-M.,2003.Heavymetalsinmarine mammals.In:Vos,J.G.,Bossart,G.,Fournier,M.,O’Shea,T.(Eds.),Toxicologyof MarineMammals.TaylorandFrancisPublishers,Washington,DC,pp.135–167.

Das,K.,Siebert,U.,Gillet,A.,Dupont,A.,Di-Poï,C.,Fonfara,S.,Mazzucchelli,G.,De Pauw,E.,DePauw-Gillet,M.-C.,2008.Mercuryimmunetoxicityinharbourseals: linkstoinvitrotoxicity.Environ.Health,7.

DeGuise,S.,2002.Cellularimmunologyofcetaceans.In:Pfeiffer,C.J.(Ed.),Molecular andCellBiologyofMarineMammals.KriegerPublishingCompany,Melbourne, Florida,pp.235–244.

DeGuise,S.,Martineau,D.,Béland,P.,Fournier,M.,1998.Effectsofinvitro expo-sureofbelugawhaleleucocytestoselectedorganochlorines.J.Toxicol.Environ. HealthA55,479–493.

DeGuise,S.,Beckmen,K.B.,Holladay,S.D.,2003.Contaminantsandmarine mam-malimmunotoxicologyandpathology.In:Vos,J.G.,Bossart,G.D.,Fournier,M., O’Shea,T.J.(Eds.),ToxicologyofMarineMammals.NewPerspectives:Toxicology andtheEnvironment.TaylorandFrancis,London,pp.38–54.

deSwart,R.L.,Kluten,R.M.G.,Huizing,C.J.,Vedder,L.J.,Reijnders,P.J.H.,Visser,I.K.G., UytdeHaag,F.G.C.M.,Osterhaus,A.D.M.E.,1993.MitogenandantigeninducedB andTcellresponsesofperipheralbloodmononuclearcellsfromtheharbour seal(Phocavitulina).Vet.Immunol.Immunopathol.37,217–230.

deSwart,R.L.,Harder,T.C.,Ross,P.S.,Vos,H.W.,Osterhaus,A.D.M.E.,1995a. Morbil-livirusesandmorbillivirusdiseasesofmarinemammals.Infect.AgentsDis.4, 125–130.

deSwart,R.L.,Ross,P.S.,Timmerman,H.H.,Hijman,W.C.,deRuiter,E.M.,Liem,A.K.D., Brouwer,A.,vanLoveren,H.,Reijnders,P.J.H.,Vos,J.G.,Osterhaus,A.D.M.E.,

(10)

1995b.Shorttermfastingdoesnotaggravateimmunosuppressioninharbour seals(Phocavitulina)withhighbodyburdensoforganochlorines.Chemosphere 31,4289–4306.

deSwart,R.L.,Ross,P.S.,Vedder,L.J.,Boink,F.B.T.H.,Reijnders,P.J.H.,Mulder,P.G.H., Osterhaus,A.D.M.E.,1995c.Haematologyandclinicalchemistryvaluesfor har-bourseals(Phocavitulina)fedenvironmentallycontaminatedherringremain withinnormalranges.Can.J.Zool.73,2035–2043.

deSwart,R.L.,Ross,P.S.,Vos,J.G.,Osterhaus,A.D.M.E.,1996.Impairedimmunityin harbourseals(Phocavitulina)exposedtobioaccumulatedenvironmental con-taminants:reviewofalong-termfeedingstudy.Environ.HealthPerspect.104, 823–828.

Debier,C.,Pomeroy,P.P.,Dupont,C.,Joiris,C.,Comblin,V.,LeBoulengé,E.,Larondelle, Y.,Thomé,J.-P.,2003.QuantitativedynamicsofPCBtransferfrommothertopup duringlactationinUKgreysealsHalichoerusgrypus.Mar.Ecol.Prog.Ser.247, 237–248.

Dietz,R.,Heide-Jørgensen,M.P.,Harkonen,T.,1989.Massdeathsofharborseals (Phocavitulina)inEurope.Ambio18,258.

Dufresne,M.M.,Frouin,H.,Pillet,S.,Lesage,V.,DeGuise,S.,Fournier,M.,2010.

Comparativesensitivityofharbourandgreysealstoseveralenvironmental contaminantsusinginvitroexposure.Mar.Pollut.Bull.60,344–349.

Duignan,P.J.,Saliki,J.T.,StAubin,D.J.,Early,G.,Sadove,S.,House,J.A.,Kovacs,K., Geraci,J.R.,1995.EpizootiologyofmorbillivirusinfectioninNorthAmerican harborseals(Phocavitulina)andgrayseals(Halichoerusgrypus).J.Wildl.Dis.31, 491–501.

Frouin,H.,Lebeuf,M.,Hammill,M.,Masson,S.,Fournier,M.,2010.Effectsof individ-ualpolybrominateddiphenylether(PBDE)congenersonharboursealimmune cellsinvitro.Mar.Pollut.Bull.60,291–298.

Greenland,J.A.,Limpus,C.J.,2007.Marinewildlifestrandingandmortalitydatabase annualreport2007.II.CetaceanandPinniped.In:EnvironmentalProtection Agency,Conservationtechnicalanddatareport.,pp.30.

Griesel,S.,Mundry,R.,Kakuschke,A.,Fonfara,S.,Siebert,U.,Prange,A.,2006. Min-eralelementsandessentialtraceelementsinbloodofsealsoftheNorthSea measuredbytotal-reﬂectionX-rayﬂuorescenceanalysis.Spectrochim.ActaB: Atom.Spectrosc.61,1158–1165.

Griesel,S.,Kakuschke,A.,Siebert,U.,Prange,A.,2008.Traceelementconcentrations inbloodofharborseals(Phocavitulina)fromtheWaddenSea.Sci.TotalEnviron. 392,313–323.

Habran,S.,Debier,C.,Crocker,D.E.,Houser,D.S.,Das,K.,2011.Blooddynamicsof mercuryandseleniuminnorthernelephantsealsduringthelactationperiod. Environ.Pollut.159,2523–2529.

Habran,S.,Pomeroy,P.P.,Debier,C.,Das,K.,2013.Changesintraceelementsduring lactationinamarinetoppredator,thegreyseal.Aquat.Toxicol.126,455–466.

Hanni,K.D.,Long,D.J.,Jones,R.E.,Pyle,P.,Morgan,L.E.,1997.Sightingsandstrandings ofGuadalupefursealsincentralandnorthernCalifornia,1988–1995.J.Mammal. 78,684–690.

Härkönen,T.,Dietz,R.,Reijnders,P.,Teilmann,J.,Harding,K.,Hall,A.,Brasseur,S., Siebert,U.,Goodman,S.J.,Jepson,P.D.,Rasmussen,T.D.,Thompson,P.,2006.The 1988and2002phocinedistempervirusepidemicsinEuropeanharbourseals. Dis.Aquat.Organ.68,115–130.

Hasselmeier,I.,Fonfara,S.,Driver,J.,Siebert,U.,2008.Differentialhematology pro-ﬁlesoffree-ranging,rehabilitated,andcaptiveharborseals(Phocavitulina)of theGermanNorthSea.Aquat.Mamm.34,149–156.

Henderson,R.J.,Kalogeropoulos,N.,Alexis,M.,1994.Thelipidcompositionof selectedtissuesfromaMediterraneanmonkseal,Monachusmonachus.Lipids 29,577–582.

Jepson,P.D.,Bennett,P.M.,Deaville,R.,Allchin,C.R.,Baker,J.R.,Law,R.J.,2005.

Relationshipsbetweenpolychlorinatedbiphenylsandhealthstatusinharbor porpoises(Phocoenaphocoena)strandedintheUnitedKingdom.Environ. Toxi-col.Chem.24,238–248.

Kakuschke,A.,Prange,A.,2007.Theinﬂuenceofmetalpollutionontheimmune systemapotentialstressorformarinemammalsinthenorthsea.Int.J.Comp. Psychol.20,179–193.

Kakuschke,A.,Valentine-Thon,E.,Griesel,S.,Fonfara,S.,Siebert,U.,Prange,A.,2005.

Immunologicalimpactofmetalsinharborseals(Phocavitulina)oftheNorthSea. Environ.Sci.Technol.39,7568–7575.

Kakuschke,A.,Valentine-Thon,E.,Griesel,S.,Rosenberger,T.,Mundry,R.,Siebert, U.,Prange,A.,2008.Bloodmetallevelsandmetal-inﬂuencedimmunefunctions ofharboursealsincaptivity.Mar.Pollut.Bull.56,764–769.

Kannan,K.,Blankenship,A.L.,Jones,P.D.,Giesy,J.P.,2000.Toxicityreferencevalues fortoxiceffectsofpolychlorinatedbiphenylstoaquaticmammals.Hum.Ecol. RiskAssess.6,181–201.

Lahvis,G.P.,Wells,R.S.,Kuehl,D.W.,Stewart,J.L.,Rhinehart,H.L.,Via,C.S.,1995.

Decreasedlymphocyteresponsesinfree-rangingbottlenosedolphins(Tursiops truncatus)areassociatedwithincreasedconcentrationsofPCBsandDDTin peripheralblood.Environ.HealthPerspect.103,67–72.

Lang,D.,Mueller-Ruchholtz,W.,1991.Humanlymphocytereactivityafterinvitro exposuretotechnicalandanalyticalgradepentachlorophenol.Toxicology70, 271–282.

Levin,M.,DeGuise,S.,Ross,P.S.,2005.Associationbetweenlymphocyte pro-liferationandpolychlorinatedbiphenylsinfree-rangingharborseal(Phoca vitulina)pups fromBritish Columbia,Canada. Environ.Toxicol. Chem.24, 1247–1252.

Levin,M.,Leibrecht,H.,Mori,C.,Jessup,D.,DeGuise,S.,2007.Immunomodulatory effectsoforganochlorinemixturesuponinvitroexposureofperipheralblood leucocytesdifferbetweenfree-rangingandcaptivesouthernseaotters(Enhydra lutris).Vet.Immunol.Immunopathol.119,269–277.

Lipscomb,T.P.,Kennedy,S.,Moffett,D.,Krafft,A.,Klaunberg,B.A.,Lichy,J.H.,Regan, G.T.,Worthy,G.A.J.,Taubenberger,J.K.,1996.Morbilliviralepizooticin bot-tlenosedolphinsoftheGulfofMexico.J.Vet.Diagn.Invest.8,283–290.

Luebke,R.W.,Hodson,P.V.,Faisal,M.,Ross,P.S.,Grasman,K.A.,Zelikoff,J.,1997.

Aquaticpollution-inducedimmunotoxicityinwildlifespecies.Toxicol.Sci.37, 1–15.

Lydersen,C.,Wolkers,H.,Severinsen,T.,Kleivane,L.,Nordøy,E.S.,Skaare,J.U.,2002.

Bloodisapoorsubstrateformonitoringpollutionburdensinphocidseals.Sci. TotalEnviron.292,193–203.

Mori,C.,Morsey,B.,Levin,M.,Nambiar,P.R.,DeGuise,S.,2006. Immunomodu-latoryeffectsofinvitroexposuretoorganochlorinesonT-cellproliferation inmarinemammalsandmice.J.Toxicol. Environ.HealthA:Curr.Iss. 69, 283–302.

Mos,L.,Morsey,B.,Jeffries,S.J.,Yunker,M.B.,Raverty,S.,Guise,S.D.,Ross,P.S.,2006.

Chemicalandbiologicalpollutioncontributetotheimmunologicalproﬁlesof free-rangingharborseals.Environ.Toxicol.Chem.25,3110–3117.

Muelbert,M.M.C.,Bowen, W.D., 1993. Durationoflactation andpostweaning changesinmassandbodycompositionofharbourseal,Phocavitulina,pups. Can.J.Zool.71,1405–1414.

Müller,G.,Wohlsein,P.,Beineke,A.,Haas,L.,Greiser-Wilke,I.,Siebert,U.,Fonfara, S.,Harder,T.,Stede,M.,Gruber,A.D.,Baumgärtner,W.,2004.Phocinedistemper inGermanSeals,2002.Emerg.Infect.Dis.10,723–725.

Nakatsuru,S.,Oohashi,J.,Nozaki,H.,Nakada,S.,Imura,N.,1985.Effectofmercurials onlymphocytefunctionsinvitro.Toxicology36,297–306.

Neale,J.C.C.,VandeWater,J.A.,Harvey,J.T.,Tjeerdema,R.S.,Gershwin,M.E.,2002.

Proliferativeresponsesofharborseal(Phocavitulina)Tlymphocytestomodel marinepollutants.Dev.Immunol.9,215–221.

Nordoy,E.S.,Blix,A.S.,1985.Energysourcesinfastinggreysealpupsevaluated withcomputedtomography.Am.J.Physiol.–Regul.Integr.Comp.Physiol.249, R471–R476.

O’Shea,T.J.,Tanabe,S.,2003.Persistentoceancontaminantsandmarinemammals: aretrospectiveoverview.In:Vos,J.G.,Bossart,G.D.,Fournier,M.,O’Shea,T.J. (Eds.),ToxicologyofMarineMammals.Taylor&Francis,London,NewYork,pp. 99–134.

Osinga,N.,ShahiFerdous,M.M.,Morick,D.,GarcíaHartmann,M.,Ulloa,J.A., Ved-der,L.,UdodeHaes,H.A.,Brakeﬁeld,P.M.,Osterhaus,A.D.M.E.,Kuiken,T.,2012.

Patternsofstrandingandmortalityincommonseals(Phocavitulina)andgrey seals(Halichoerusgrypus)inTheNetherlandsbetween1979and2008.J.Comp. Pathol.147,550–565.

Pillet,S.,Fournier,M.,Measures,L.N.,Bouquegneau,J.-M.,Cyr,D.G.,2002.Presence andregulationofmetallothioneinsinperipheralbloodleucocytesofgreyseals. Toxicol.Appl.Pharmacol.185,207–217.

Reijnders,P.J.H.,1980.Organochlorineandheavymetalresiduesinharbourseals fromthewaddenseaandtheirpossibleeffectsonreproduction.Neth.J.SeaRes 14,30–65.

Reijnders,P.J.H.,1981.Managementandconservationoftheharbourseal,Phoca vitulina,populationintheinternationalwaddenseaarea.Biol.Conserv.19, 213–221.

Rogan,E.,Baker,J.R.,Jepson,P.D.,Berrow,S.,Kiely,O.,1997.Amassstrandingof white-sideddolphins(Lagenorhynchusacutus)inIreland:biologicaland patho-logicalstudies.J.Zool.242,217–227.

Ross,P.S.,2002.Theroleofimmunotoxicenvironmentalcontaminantsin facilitat-ingtheemergenceofinfectiousdiseasesinmarinemammals.Hum.Ecol.Risk Assess.8,277–292.

Ross,P.S.,DeSwart,R.,Addison,R.,VanLoveren,H.,Vos,J.,Osterhaus,A.,1996a.

Contaminant-inducedimmunotoxicityinharbourseals:wildlifeatrisk? Toxi-cology112,157–169.

Ross,P.S.,DeSwart,R.L.,Timmerman,H.H.,Reijnders,P.J.H.,Vos,J.G.,VanLoveren,H., Osterhaus,A.D.M.E.,1996b.Suppressionofnaturalkillercellactivityinharbour seals(Phocavitulina)fedBalticSeaherring.Aquat.Toxicol.34,71–84.

Ross,P.S.,DeSwart,R.L.,VanLoveren,H.,Osterhaus,A.D.M.E.,Vos,J.G.,1996c.The immunotoxicityofenvironmentalcontaminantstomarinewildlife:areview. Annu.Rev.FishDis.6,151–165.

Ross,P.S.,Beckmen,K.B.,Pillet,S.,2003a.Immunotoxicologyoffree-ranging pin-nipeds:approachestostudydesign.In:Vos,J.G.,Bossart,G.,Fournier,M.,O’Shea, T.(Eds.),ToxicologyofMarineMammals.TaylorandFrancisPublishers, Wash-ington,DC,pp.570–591.

Ross,P.S.,Vos,J.G.,Osterhaus,A.D.M.E.,2003b.Theimmunesystem,environmental contaminantsandvirus-associatedmassmortalitiesamongpinnipeds.In:Vos, J.G.,Bossart,G.,Fournier,M.,O’Shea,T.(Eds.),ToxicologyofMarineMammals. TaylorandFrancisPublishers,Washington,DC,pp.534–557.

Siebert,U.,Wohlsein,P.,Lehnert,K.,Baumgartner,W.,2007.Pathologicalﬁndings inharbourseals(Phocavitulina):1996–2005.J.Comp.Pathol.137,47–58.

Siebert,U.,Heidmann,A.,Friedhoff,N.,Kruse,H.,Rigét,F.,Adler,S.,Maser,E.,2012.

Organochlorineburdensinharboursealsfromthegermanwaddenseacollected duringtwophocinedistemperepizooticsandringedsealsfromwestgreenland waters.J.Environ.Anal.Toxicol.2,2–8.

Sinjari,T.,Klasson-Wehler,E.,Oskarsson,A.,Darnerud,P.O.,1996.Milktransferand neonataluptakeofcoplanarpolychlorinatedbiphenyl(PCB)congenersinmice. Pharmacol.Toxicol.78,181–186.

Stone,K.D.,Feldman,H.A.,Huisman,C.,Howlett,C.,Jabara,H.H.,Bonilla,F.A.,2009.

Analysisofinvitrolymphocyteproliferationasascreeningtoolforcellular immunodeﬁciency.Clin.Immunol.131,41–49.

vanLoveren,H.,Ross,P.S.,Osterhaus,A.D.M.E.,Vos,J.G.,2000.Contaminant-induced immunosuppressionandmassmortalitiesamongharbourseals.Toxicol.Lett. 112–113,319–324.

(11)

VandenBerghe,M.,Weijs,L.,Habran,S.,Das,K.,Bugli,C.,Rees,J.-F.,Pomeroy, P.,Covaci,A.,Debier,C.,2012.Selectivetransferofpersistentorganic pollu-tantsandtheirmetabolitesingreysealsduringlactation.Environ.Int.46,6– 15.

Vermes,I.,Haanen,C.,Steffens-Nakken,H.,Reutellingsperger,C.,1995.Anovelassay forapoptosisﬂowcytometricdetectionofphosphatidylserineexpressionon earlyapoptoticcellsusingﬂuoresceinlabelledAnnexinV.J.Immunol.Methods 184,39–51.

Vos,J.G.,Ross,P.S.,deSwart,R.L.,vanLoveren,H.,Osterhaus,A.D.M.E.,2003.The effectsofchemicalcontaminantsonimmunefunctioninharborseals:resultsof asemi-ﬁeldstudy.In:Vos,J.G.,Bossart,G.D.,Fournier,M.,O’Shea,T.J.(Eds.), Toxi-cologyofMarineMammals,NewPerspectives:ToxicologyandtheEnvironment. TaylorandFrancis,pp.558–591.

Weijs,L.,Das,K.,Siebert,U.,vanElk,N.,Jauniaux,T.,Neels,H.,Blust,R.,Covaci,A., 2009a.Concentrationsofchlorinatedandbrominatedcontaminantsandtheir metabolitesinserumofharboursealsandharbourporpoises.Environ.Int.35, 842–850.

Weijs,L.,Dirtu,A.C.,Das,K.,Gheorghe,A.,Reijnders,P.J.H.,Neels,H.,Blust,R., Covaci,A.,2009b.Inter-speciesdifferencesforpolychlorinatedbiphenylsand polybrominateddiphenylethersinmarinetoppredatorsfromtheSouthern NorthSea:Part1.Accumulationpatternsinharboursealsandharbourporpoises. Environ.Pollut.157,437–444.

Weijs,L.,Dirtu,A.C.,Das,K.,Gheorghe,A.,Reijnders,P.J.H.,Neels,H.,Blust,R.,Covaci, A.,2009c.Inter-speciesdifferencesforpolychlorinatedbiphenylsand poly-brominateddiphenylethersinmarinetoppredatorsfromtheSouthernNorth Sea:Part2.Biomagniﬁcationinharboursealsandharbourporpoises.Environ. Pollut.157,445–451.

Weijs,L.,Covaci,A.,Yang,R.S.H.,Das,K.,Blust,R.,2011.Anon-invasiveapproach tostudylifetimeexposureandbioaccumulationofPCBsinprotectedmarine mammals:PBPKmodelinginharborporpoises.Toxicol.Appl.Pharmacol.256, 136–145.

Zharinov,A.,Tarasov,P.,Shvalov,A.,Semyanov,K.,vanBockstaele,D.R.,Maltsev,V., 2006.Astudyoflightscatteringofmononuclearbloodcellswithscanningﬂow cytometry.J.Quant.Spectrosc.Radiat.Transfer102,121–128.