Epigenetic memory of environmental organisms : a reflection of lifetime stressor exposures

(1)

ContentslistsavailableatScienceDirect

Mutation

Research/Genetic

Toxicology

and

Environmental

Mutagenesis

j ou rn a l h o m e pa ge :w w w . e l s e v i e r . c o m / l o c a t e / g e n t o x

C o m mun i ty a d dr e s s :w w w . e l s e v i e r . c o m / l o c a t e / m u t r e s

Epigenetic

memory

of

environmental

organisms:

A

reﬂection

of

lifetime

stressor

exposures

Leda

Mirbahai,

James

K. Chipman

∗

SchoolofBiosciences,UniversityofBirmingham,Edgbaston,BirminghamB152TT,UK

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received8October2013 Accepted9October2013 Availableonline17October2013

Keywords:

Epigenetic DNAmethylation Toxicity

Epigeneticmemory Environment Epigeneticfoot-printing

a

b

s

t

r

a

c

t

Both geneticand epigenetic responses of organisms toenvironmental factors, including chemical exposures,influenceadaptation,susceptibilitytotoxicityandbiodiversity.Inmodelorganisms,itis establishedthatepigeneticalterations,includingchangestothemethylome,cancreateamemoryof thereceivedsignal.Thisispartlyevidencedthroughtheanalysisofepigeneticdifferencesthatdevelop betweenidenticaltwinsthroughouttheir lifetime.Theepigenetic marks inducealterationstothe geneexpressionprofile,which,inadditiontomediatinghomeostaticresponses,havethepotentialto promoteanabnormalphysiologyeitherimmediatelyoratalaterstageofdevelopment,forexample leadingtoanadultonsetofdisease.Althoughthishasbeenwellestablished,epigeneticmechanisms arenot consideredinchemical riskassessmentorutilised inthe monitoringoftheexposure and effectsofchemicalsand environmentalchange. Inthis review,epigeneticfactors,specifically DNA methylation,arehighlightedasmechanismsofadaptationandresponsetoenvironmentalfactorsand which,ifpersistent,havethepotential,retrospectively,toreflectpreviousstressexposures.Thus,itis proposedthatepigenetic“foot-printing”oforganismscouldidentifyclassesofchemicalcontaminants towhichtheyhavebeenexposedthroughouttheirlifetime.Insomecases,thepotentialforpersistent transgenerationalmodificationoftheepigenomemayalsoinformonparentalgermcellexposures. Itisrecommendedthatepigeneticmechanisms,alongsidegeneticmechanisms,shouldeventuallybe consideredinenvironmentaltoxicitysafetyassessmentsandinbiomonitoringstudies.Thiswillassist indeterminingthemodeofactionoftoxicants,noobservedadverseeffectlevelandidentificationof biomarkersoftoxicityforearlydetectionandriskassessmentintoxicologybuttherearecriticalareas thatremaintobeexploredbeforethiscanbeachieved.

1. Introduction

Organisms have the ability to respond to environmental

stressorssuchastoxic chemicalsand adaptbeneﬁciallytonew environments. This is accomplished, in part, by altering their epigenomesand subsequently theirtranscription proﬁles. Thus, overlaidonthegenomeareepigeneticmarks,particularly methyl-ationandhydroxymethylationofCpGsitesthatdetermine,inpart,

how the genome responds through regulation of transcription

[1–3].Itiswellestablishedthatvariousenvironmentalstressors, includingdietary deﬁciencies and exposuretoa wide range of

chemical pollutants, can modulate the epigenome [4–7]. The

changesinresponsetoenvironmentalstressorsmaycontributeto anadaptivesurvivaladvantageoflocalpopulationsoforganisms

∗Correspondingauthor.Tel.:+44(0)1214145422;fax:+44(0)1214145925.

E-mailaddress:[email protected](J.K.Chipman).

throughadvantageous geneexpression[8,9].However,in some

cases these changes are associated with marked phenotypic

endpoints that can be detrimental [10,11]. Such accumulated modiﬁcationsofDNA,andtheconsequentchangesingene expres-sion, have important implications in diseases, including cancer

[12]andmay,insomesituations,bepersistentthroughsubsequent generations.Furthermore,itisbecomingmoreevidentthat epige-neticmechanismsareinvolvednotonlyinadaptiveresponsesin individuals;theyalsohaveasignificantroleinhost-pathogen inter-actionsasreviewedbyGomez-Diazetal.[13].Thisdemonstrates theroleofepigeneticmechanismsinmultiplespeciesinteractions. Inthisreview,weexploretheepigeneticresponsesoforganisms toenvironmentalstressorswithaparticularfocusonthe persis-tenceor“memory”ofsuchmodificationsandthewaysinwhichthis memorycanusefullyreflectthestatusoftheenvironmentinwhich humansand other organismsreside. Epigeneticfactors, specifi-callyDNA methylation,are introducedas aninterface between

the genome and the environment, providing partial

mechanis-ticexplanationsforthesensitivityoforganismstoenvironmental factors.WearguethatepigeneticmechanismssuchasDNA meth-ylationare essential indetermining how organismsrespondto

http://dx.doi.org/10.1016/j.mrgentox.2013.10.003

Open access under CC BY license.

(2)

environmentalagents and wepresent examplesof studiesin a rangeofspeciesshowinghowexposuretochemicalscanpromote persistentchangesintheepigenomewithphenotypicoutcomes. Furthermore,thesestudiesleadtotheconceptof“epigenetic foot-printing”forretrospectiveassessmentofchemicalexposures.The relevanceandsigniﬁcanceofepigeneticmechanismsin environ-mental risk assessment and the potential for establishment of suitablebiomarkersisdiscussedinthisreview.Theseinsightsmay shapethefutureofregulatorytoxicologyandenvironmental mon-itoring,especiallywheretherearechronicexposurestopollutants.

2. Epigenetics

Epigenetics is deﬁned as meiotically and mitotically

herita-ble changes in gene expression that cannot be explained by

changesinDNAsequence[14,15].SuchmodiﬁcationsincludeDNA methylation, post-transcriptional chemical modiﬁcations of the

N-tailof histones and amino acids within theglobular histone domains,binding ofnon-histonechromatinproteinstoDNA or histonemodiﬁcations(i.e.transcriptionfactors),non-codingRNA,

nucleosome positioning and higher order chromatin

organisa-tion[10,14,16–19].Moreover,thesemodiﬁcationsarenotisolated eventsandarecloselyinter-linkedbyinﬂuencingchromatin struc-tureatvariouslevelsandbyfurtherinteractionswiththegenome

[19].Undernormalconditions,cellsofanorganismdisplayafinely tunedepigeneticequilibrium[1].However,disruptionofthe activ-ityofenzymesregulatingtheepigenomeorchangesinthelevels ofthemetabolitesrequiredfortheactionoftheseenzymescan resultinalterationofepigeneticmarksandtheepigenomic equi-librium[10] leadingtoinappropriateregulationoftranscription andpotentialdisorders[14,19].MethylationofDNAatCpG dinu-cleotidesisthemoststudiedepigeneticmodification[1]anditisthe principalfocusofthisreview.DNAmethylationisthetransferofa methylgroup,byDNAmethyltransferases(DNMTs),fromthe uni-versalmethyldonorS-adenosylmethionine(SAM)tothe5thcarbon positionofacytosinepyridinering[20,21].Thecrosstalkbetween histonemodificationandDNAmethylationatthetranscriptionally activeandinactiveregionsispartlyaccomplishedbyacfp1(CXXC fingerprotein1)andmethyl-CpGbindingproteins(MeCPs), respec-tively[22].Theseproteinsselectivelybindtomethylation-freeand methylatedCpGdinucleotides,respectively[22–24],encouraging recruitmentofhistoneacetyltransferasesandde-acetyltransferase andotherepigeneticandnon-epigeneticfactorsleadingto regu-lationoftranscription[23,24].Althoughtheprecisemechanisms ofDNAde-methylationarenotknown,ithasbeensuggestedthat DNAcanalsoeitherpassivelyoractivelyundergode-methylation. Duringpassivede-methylation,5-methylcytosineisremovedin

a replication-dependent mannerby inhibition of DNMTs while

active de-methylation depends upon enzymatic removal of

5-methylcytosine.Recently,TET(ten-eleventranslocation)proteins havebeen foundtobeinvolved inregulation ofDNA methyla-tion.TET1-3proteinscatalysetheconversionof5-methylcytosine to5-hydroxymethylcytosine(5-hmC)whichispoorlyrecognised by DNMTs.This resultsin a passive replication-dependentloss ofDNA methylation.Alternatively,5-hmCcanberecognisedby therepair machineryand converted tocytosine [25–27]. Thus, althoughmethylationcan beremovedfromCpGislands,this is a highlyregulatedprocess and,as discussedbelow, many such modiﬁcationsarepersistentand“memorised”incells.

3. Environmentalsensitivityoftheepigenomethroughout lifetimeandretention(memory)of

environmentally-inducedchanges

Adaptiveresponsesandsensitivityofanorganismto environ-mentalstimuli (e.g.chemicalcontaminants, dietandstress)are

observedthroughoutthelifetimeofanorganism.However,akey questioniswhatarethemolecularmechanismsleadingtochanges intheexpressionofthegenome?Forexample,whydo homozy-goustwinshavedifferentdiseasesusceptibilitiesintheabsenceof geneticvariation?Asintroducedabove,itisbecomingmoreevident that,althoughepigeneticmarksarestableenoughtoregulategene expression,theyarealsosusceptibletochangebyenvironmental signals.Thismeansthattheepigenomecanchangeasaresponseto environmentalstimuli,whichthencanleadtoalterationinthe phe-notype[15].Inaway,theepigenomecanactasthelinkbetween environmentalcues(externalandinternal)totheorganismand phenotypebytranslatingtheenvironmentalsignalstophenotypic responsesthroughalteredgeneexpressionprofiles.Forexample, inresponsetotheirimmediateenvironment,afractionof pluripo-tentstemcellswilldifferentiateandformdistinctcelltypeswitha characteristicgeneexpressionprofile.Thetissue-specific expres-sion patternsare generally maintainedthroughout the lifetime oftheindividual.Thedifferencesintranscription profilesofthe differentiated cells arethen attributedtotheirdifferent herita-ble epigeneticprofiles. Hence,tissue-specific epigeneticprofiles provideamethodofsustainingthememoryofthedifferentiation processintheabsenceoftheinitiatingsignal[12,28–30].Oneofthe bestexamplesoftheinfluenceofenvironmentontheepigenome andsubsequentchangesingeneexpressionistheresponseof Ara-bidopsis to prolongedexposureto coldweather (vernalisation). Followingprolongedexposuretocoldweather(anenvironmental factor),floweringlocusC,arepressorofflowering,becomes epi-geneticallysilenced.Thisresultsincoordinationoffloweringtime (phenotype)withspringandsummer[31–33].

Although the epigenome is sensitive to the environmental

stimulithroughoutanindividual’slifetime,therearecritical

win-dows during development that the epigenome is at its most

sensitivewithlastingtranscriptionaleffects.For example,genes suchasoestrogenreceptor(ER)andglucocorticoidreceptor(GR) areregulated,inpart,throughDNAmethylationoftheirpromoter regions.Themethylationandsubsequentlythetranscription lev-elsofthesegenesaregender-andregion-specific.Furthermore, DNA methylation of these genes is substantially influenced by environmental factors, such as maternal care and exposure to chemicals,encounteredduringembryogenesisandearlypostnatal stages(reviewedin[34–36]).Anothergoodexampleof develop-mentalsensitivewindowsofalterationsintheepigenomecomes fromthestudiesconductedin fishspecieslookingat theeffect

of temperatureongender.Sex determinationin many ﬁshand

reptilespeciesisinfluencedbymanyfactorsincludingthe tem-peratureofthewaterduringearlystagesoflarvaldevelopment. InEuropeanseabass,hightemperatures(21◦C)andlow tempera-tures(15–16◦C)increasethenumberofmalefishandfemalefish, respectively.Itwasdemonstratedthatexposuretohigh temper-atureatcriticalstagesoflarvaldevelopmentincreasedtheDNA methylationlevelofthearomatase(cyp19a1)promoterinfemale gonadspriortoformationofgonadalridgeanddifferentiationof thegonads.Aromataseconvertsandrogentooestrogen.Adecrease intheexpressionofthisgeneasaresultofhightemperaturesand subsequentmethylationandsuppressionofthepromoterregion ofthis gene resultsin increasedlevelsofandrogen, differentia-tionofthegonads,formationoftestisandamale-biasedsexratio

(3)

DNA _Exposure_{s can}_{cause uniqu}_e

and persistent change to the

epigenome

( , )

Epigenec changes act as a

“foot-print” of prior exposures

Regulatory agencies can use

these markers for

idenﬁcaon of prior

exposure to hazards. This aids

risk assessment and can help

to priorise remediaon

Series of exposure to environmental

stressors (e.g. chemical exposures)

Some epigenec modiﬁcaons can lead

to an adverse outcome in life-me

Adapve response

Toxicity

Toxicity in subsequent generaons

Fig.1.“Epigeneticmemory”asanindicatorofpriorexposure.Epigeneticalterationsinducedbyenvironmentalstressors,includingchangestothenormalDNAmethylation pattern,cancreateapersistentmemoryofthereceivedsignal.Mostinterestingly,itisproposedthateachclassofchemicalscaninduceclass-specificalterationstothenormal patternofDNAmethylation(epigeneticfoot-print).Thesechangeswillfurtherinducealterationstothegeneexpressionprofile,whichcanpromotechangeinorganism’s traitseitherimmediatelyoratalaterstageofdevelopment.Suchpersistentmodulationsoftheepigenomeofferauniqueopportunitytoprovidealife-timehistoryofan organism’spriorexposuretofactorsinfluencingtheepigenome.Thereisalsopotentialforgerm-linemodificationstopersistintosubsequentgenerationsdependingonthe natureandcriticaltimingofexposure.

Furthermore,severalstudieshave shownphenotypic plastic-itydrivenbyepigeneticchanges.Thedifferencesinmorphology, behaviourandreproductiveabilityofgeneticallyidenticalfemale workerandqueenhoneybees(Apismellifera)havebeenexplained by distinct methylation proﬁles of theirbrain DNA and subse-quentimpactsontranscription.Thevariationintheirmethylation proﬁles has been linked, in part, to their different diets dur-inglarvaldevelopment.[40].Likewise,earlymaturationofmale Atlanticsalmon(Salmo salar)hasevolvedin response tolower

population densities. Moran and Perez-Figueroa [41]

demon-strated that transcription levels vary betweenbrain and testes ofmatureandimmaturesalmonintheabsenceofgenetic varia-tions.Furthermore,maturationstagecorrelatedwithdifferences

inDNA methylationproﬁlesof matureandimmatureﬁsh,

pro-vidingevidenceofepigeneticallydrivenphenotypicdifferencesas aresponsetoanenvironmentalfactorintheabsenceofgenetic variations.

Of particular relevance to evidence of epigenetic “memory” areseveralstudiesingeneticallyidenticalhumantwinsthathave clearlydemonstratedalinkbetweenenvironmentalfactors,change intheepigenomeand differentsusceptibilitytodisease.Oneof theﬁrst studiesthat demonstrated this link wasconducted by Fragaetal.[42].Inthisstudy,itwasshownthatdifferencesinthe epigenomicproﬁlesofmonozygotic(MZ)twinsaccountsfortheir differentphenotype(i.e.disease)inresponsetoenvironmental fac-torsovertime.Indeed,theseepigeneticdifferencesappearedmore prevalentwithincreasedageoftwinswithdifferentlifestyles(i.e. diet,smokingandphysicalactivity).Thisdemonstratesastronglink betweenaccumulationofepigeneticchangesovertimeandaltered phenotype.

4. Examplesoftherangeofenvironmentalstressorsthat canaltertheepigenome

Direct exposure to chemicals, such as chronic exposure to persistent lipophilic compounds or metals in the environment, cancauseadverseeffects byinducingchangeintheepigenome ([7], reviewed in [43,44]). For example, the carcinogenicity of someenvironmentalcontaminantssuchasendocrinedisrupters,

nickel, cadmium, chromium and arsenic cannot entirely be

explainedthroughgeneticbasedmechanisms[45].Alterationsto

theepigenome through exposureto endocrine disruptors have

beenlinked,forexample,tonegativeimpactsonneuroendocrine systems[34,46],alteredbehaviouralneuroendocrinology[21,47]

andhigher ratesof tumourigenesisat environmentallyrelevant concentrationsinmice[48].Metalscaninterferewiththeactivity ofDNAmethyltransferaseseitherdirectlyorthroughproduction ofreactiveoxygenspecies. Thisleadstoanaltered DNA meth-ylation proﬁle and subsequent alterations in gene expression. For instance, it hasbeen proposedthat cadmium (Cd)-induced globalDNAhypomethylationmaybeduetoCdinteractionwith

DNA methyltransferases (DNMTs) and subsequent interference

(4)

epigeneticallymodifyingandde-modifyingenzymesdependupon levelsofcofactorsandmetabolitespresentintheintra-and extra-cellularenvironment[10,44].Severalstudieshavedemonstrated thatalterationsin thelevelsofnutrientssuchasfolate,choline andmethionineindietoralterationstoothercomponentsofthe one-carboncyclesuchasthemethyl-donorSAMcanalsohave sig-niﬁcanthealtheffectsbyinducingaberrantDNAmethylation.In thesestudies,alinkbetweendietsdeﬁcientincholineorother

primarymethyldonors,DNAhypomethylationanddevelopment

of tumours in rodents, humans and ﬁsh has been established

[4–6,49–54].

5. Epigeneticmemoryanditslinkwithadult-onsetof disease

TherearetwoDNAmethylationre-programmingevents associ-atedwithdevelopmentinmammals.Duringembryogenesis,there issusceptibilitytomodificationsduetoenvironmentalstressors thatcaninfluencephenotypeandpotentialdiseaselater inlife. Asecondstagere-programmingin germcells oftheF1 genera-tion,atleastinmammals,isalsosusceptibletointerferencethat couldinfluencephenotypebutalsocaninvolvemodificationsto genes suchas those that are imprinted and that influence sex determination.Disruptionofepigeneticmechanismsduringthese twocriticalstagesofdevelopmenthasbeenrecognised,inpart, asafactoraffectingdevelopmentofcertainadulthoodphenotypes longafterthestimulatingfactorhasbeenremoved.Theseinclude linkageofvariouscancers,diabetes,obesityandbehaviouraland neurodegenerativedisorderswithenvironmentalfactors encoun-teredduringprenatalandearlypost-natalperiodsin mammals. Thisisknownas“foetalbasisorearly-lifeprogrammingof adult-onsetofdisease”[15,55–58].Theepigeneticmarksinflictedupon

the genome by environmentalfactors very earlyon during an

individual’slife actasan“epigeneticmemory”oftheexposure. Theseepigeneticmemoriescanmanifestinadultsasa pathologi-calphenotypeoftenfollowingasecondarytriggersuchasageingor changesinhormonelevels.Forexample,inthe1950s diethylstilbe-strol(DES)wasused duringpregnancytopreventspontaneous abortions.DNAmethylationchangesinducedbythisagent dur-ingembryogenesishavebeenidentiﬁed asthecauseofa range ofdisorderssuchasincreasedbreastandtesticularcancerinadult femaleandmaleoffspring[59].Exposuretooestrogensandarange ofendocrinedisruptorsinearlydevelopmenthasbeenshownto predisposetocancerinrodentsandhumans[60–62],alter hor-monalresponseslaterinlifeinthefrog(Xenopuslaevis)[63]and causearangeofreproductiveandbehaviouraleffectsinrodents (reviewedin [34–36]).Speciﬁcally,methylationchanges caused bydiethylstilbestroltoc-fosandlactoferringenesatthe neona-talstageinmicecontributetoanincreasedincidenceofuterine cancer[64].

Studiesusingtheagouti(Avy₎_mouse_model_have_clearly demon-stratedalinkbetweenadultonsetofdisease,DNA methylation-inducedchangesintheactivatorbindingprotein–intra-cisternalA particle(IAP)transposonlocatedintheAvyalleleduring embryo-genesisandenvironmentalexposureofthegestatingfemalemice

[56,65,66].For exampledietaryuptakeof genisteinin gestating agoutimice,atlevelscomparabletothedietofahumanwithhigh soyconsumption,resultsinhypermethylationoftheAvy_allele_and generationofpseudo-agoutioffspringwithalowerriskof devel-opmentofobesityinadulthood[66].

Furthermoreinourrecentpublications[54,67]weestablished alinkbetweenexposuretomarinepollutants,alterationinDNA methylationpatternsandlivertumoursdissectedfromthe flat-fishdab(Limandalimanda)capturedfromwatersaroundtheUK. Basedonthefindingofsignificantepigeneticmodifications and

disruptionofmetabolitesofthe1-carbonpathwayinnon-tumour hepatictissueofadenoma-bearing fish,ourstudylendssupport totheepigeneticprogenitormodelofcancer. Disruptionof epi-geneticmechanismscancausestable,heritablechangesingene expressionthatareindependentfrommutations.Thesechanges canoccurpriortomutations[68].Therefore, inthismodel,itis proposedthatepigeneticchangesinflictedasaresultofchemical exposurescaninitiatecarcinogenesis.Thusepigenetic“memory” maypredisposetocancerhighlightingthesignificantimpactsthat disruptionofepigenetic mechanismscanhave onthehealthof an exposed individual.In the light ofthis, the valueof assess-mentofepigeneticchangesinenvironmentalbiomonitoringand intheearlydetectionofadversehealtheffectsbecomesevident. Importantly,evenminorepigeneticchangesasaresponseto envi-ronmentalfactorscanaccumulateover time,leadingtogradual alterationofthephenotype[7].However,itisimportanttobear inmindthatestablishingacause-andeffect-relationshipbetween exposuretoenvironmentalfactors,changesintheepigenomeand diseaseischallenging.

6. Transgenerationalepigenetic“memory”

Perhapsthemostconcerningaspectofepigeneticmodulation by environmental toxicants is the potential for modulation of theprogramming of thegermline, causing a transgenerational “memory”(Fig.1)[58,69].Transmissionofaphenotypetoa fol-lowinggenerationisaccomplishedthroughgermlines.Therefore environmentallyinduced-epigeneticchangesinimprintedgenes during the re-programming at thecritical germ cell stage can inﬂuence bothsex determination and canpotentially be inher-itedleadingtotransgenerationalmodiﬁcations.Duringthegerm

cellre-programmingeventinmammals,mostepigenetic marks

are removedand reset in a gender-dependentmanner.Certain

epigenetic marksthataresex-speciﬁc, suchas imprintedgenes establishedduringgermcellreprogramming,escape thesecond waveofDNAmethylationchangesthatoccurinpre-implantation embryos[2,70].Transgenerationalepigenetic inheritanceor the epigeneticbasisforinheritanceofatrait[56,71]providesafurther aspectof“epigeneticmemory”relevanttothisreview.

Importantly, it is necessary to distinguish between epige-netic transgenerational effect and epigenetic transgenerational inheritance.Theformerisabroadtermincorporatingall pheno-typesinfollowinggenerationsthatarenotgeneticallydetermined

[69]. For example, stressed femalerats have a reduced mater-nallicking/groomingandarchedbacknursing(LG-ABN)behaviour towardstheirneonates.AlowerlevelofLG-ABNbehaviourresults inepigenetic silencingoftheglucocorticoidreceptor(GR)gene, resulting in fearfulbehaviourin the F2generation,and persis-tenceand transferofthephenotypetosubsequentgenerations. Theacquiredphenotypeisepigeneticallymaintainedinmultiple generations.However,crossnurturingoftheF3generationborn fromthelowLG-ABNF2groupresultsinhypomethylationofthe

GR gene and normal LG-ABN behaviour in thesemice [72–74].

Therefore, transfer ofthe acquiredepigenetic phenotypeis not throughgametesanditisdependentuponconsistencyofthe envi-ronmentalcondition[70].Incontrast,transgenerationalepigenetic inheritancereliesoninheritanceoftheepigeneticallyacquired phe-notypethroughgametes.Thisrequirespersistenceofepigenetic marksthroughthereprogrammingevents. [75].Alsoimportant is the realisation that theDNA re-methylationof germ cells is inﬂuencedbythemicroenvironmentaffordedbythesurrounding somaticcellsandbysignalsreceiveddirectlyfrom environmen-tal factors. As a consequence, environmental factors can both directlyorindirectly inﬂuencethere-methylationofgermcells

(5)

Inviviparousspecies,forabiologicaltraittobecategorisedas inherited,thephenotypemustbemaintainedtoat leasttheF3 generation.Thisisbecauseexposureofthegestatingfemale(F0) tochemicalsthat modifytheepigenomecould resultin simul-taneousdirectexposureofthedeveloping embryo(F1)andthe developinggermlineoftheembryo(F2)[56,76].Severalstudies havedemonstratedthepossibilityofepigeneticinheritance ofa phenotypeinmultiplegenerationsinmammals,plantsandﬂies (forreviewssee[58,69]).Ofparticularinterestisthedemonstration that,followingintra-peritonealexposureofgestatingoutbred Har-lanSprague-Dawleyfemaleratstotheanti-androgenicendocrine disruptervinclozolin(100mg/kgbodyweight(bw)/day), epigene-ticallymalegermcelltransmittedphenotypiccharacteristicsare inducedduringcriticalstagesofsexdetermination(E12–E15),up toatleasttheF3generationinmaleratoffspring.The characteris-ticsinclude,forinstance,increasedspermatogeniccellapoptosis, decreased spermmotility and numbers,prostate abnormalities, tumoursandrenal lesions.Thereproducibilityand frequencyof the vinclozolin-induced phenotypes (i.e. rate of tumours) and identiﬁcationofgenes withaltered methylationin theaffected individuals compared to controls indicated that mutations are not the most likely cause of this abnormality [60,75–80]. Fur-thermore,usingaratmodelexposedtoseveralchemicals,ithas beendemonstratedthattheovariandiseasecanbeepigenetically inheritedacrossseveralgenerations [81]. However,it hasbeen demonstratedthattheeffectandepigeneticinheritanceofchanges inducedbyvinclozolin arehighly dependentondose[58], ani-malstrain[82] androuteof exposure[83].Oral administration ofvinclozolin(100mg/kgbw/day)inoutbredWistarrats[83]and intra-peritoneal(IP)injectionofvinclozolin(100mg/kgbw/day)in inbredCD-SpragueDawleyrats[82]duringthesexdetermination stagefailedtoinduceinheritedphenotypiceffects.However,there isagreat dealofcontroversysurroundingthis areaofresearch. Forexample,arecentpublicationfailedtodemonstrateany trans-generationalinheritanceafterIPadministrationofvinclozolin(0, 4,100mg/kgbw/d)ongestationdays6–15inoutbredWistarrats

[84].

In contrast to various studies in rodents, transgenerational epigenetics remains a fairly unexplored ﬁeld in other species. Epigeneticchangeshavebeenstudiedintwogenerationsofthe water ﬂea (Daphnia magna) following exposure to a range of environmentallyrelevantcompounds([85–88],reviewedin[89]).

5-Azacytidine, a demethylating drug, induced signiﬁcant DNA

hypomethylationinnon-directlyexposedF1andF2D.magna off-springaswellastheexposedF0 generation[88].Althoughit is possiblethattheseeffectsareepigeneticallyinherited,D.magna

reproducesbothsexuallyandasexuallyandthereforeobservation ofsimilarchangesinthephenotypesinF3generationarerequired priortoconcludingthattheobservedeffectsastruly transgenera-tional[90].

Inthenon-eutherianfish,onlytheF0andthegamete/oocyte oftheF1generationaredirectlyexposed.Therefore, after elim-inationofthepotentialexposuredirectly totheeggsfollowing spawning,itwould bepossibletocategorise anyepigenetically inheritedphenotypeintheF2generationasatrue transgenera-tionalepigeneticeffect.Beingnon-eutherian,fishdonotrequire imprintingtopreventdirect“parentconflict”anditisnotknown ifthereprogrammingeventingermcells duringthesex deter-mination stage (a critical stage for transgenerational effects in mammals)occursinfish.Thisisnottosaythatmethylationdoes notplayaroleduringsexdeterminationinfishsincethe regula-tionofaromatase,involvedinsexdeterminationinfish,ispartly throughDNAmethylation[37,38].Furthermorethemethylation, andasaresulttheexpressionofthis gene,issensitiveto com-poundssuchas17␣-ethinylestradiol[38].Thus,theexistenceof theenvironmentally-sensitivedifferentiallymethylatedaromatase

geneinmaleandfemaleﬁshfurtherjustiﬁestheinvestigationof transgenerationalepigenetic mechanismsandtheoccurrenceof

DNA methylationreprogrammingduringthesex determination

stageinnon-mammalianspecies.

7. Implicationsof“epigeneticmemory”forchemicalsafety assessmentandenvironmentalbiomonitoringandfuture directions

Inthecontextoftoxicitytesting,whetherinlaboratoryanimals toassesstherisktohumansorinspeciesrelevanttothenatural environment,epigeneticchangesarenotcurrentlyastandard fea-tureofsafetyassessment.Partofthereasonforthisistheinability tointerpretthefindingsinrelationtopotentialadverseoutcome withoutamorecompleteknowledgeofthefundamental relation-shipsbetweenspecificchangesanddisease.However,theevidence forchangesinDNAmethylationinearlydevelopmentinfluencing disease,includingcancer,inlaterlifeorbeyondintosubsequent generationsshouldbeborneinmind.

Currently, standard carcinogenicity testing doesnot usually

include exposure to the test compound during early

develop-ment. The evidence accruing for the contributionof change in DNAmethylationtocancersproducedbynumerousnon-genotoxic andgenotoxiccarcinogens[91–93]and theestablished carcino-geniceffectofDNAmethylationchangesfollowingdeﬁcienciesin cholineandotherprimarymethyl-donors[49–52],indicatesthat suchpotentialmechanismsshouldnotbeignored.However,unlike theassumptionsmadeaboutgenotoxiccarcinogens,apragmatic thresholdlevelofexposuretonon-genotoxiccarcinogensordietary deﬁciencymayberequiredforaclearimpactoncancer develop-ment.

ThefactthatmethylationofCpGislandsincreasestherateof mutationsatthesesitesbyaround10-fold[94]raisesimplications notonlyforcancerbutalsoforreproductivedeficienciesthatcould impactonecosystems[95].Moreover,environmentally-induced alterationsinthemethylationofsexdetermininggenesmayhave asignificantimpactonthepopulationandhealthofspecies(e.g. impairedrateoffertilisation).Asnotedabove,althoughsofaronly demonstrated in laboratorymaintainedand infarmed fish,the ratiooffishembryosthatdevelopintofemaleormalefishcanbe influencedbythemethylationlevel ofthearomatasegene[37]. Thereforeitispossiblethatenvironmentally-inducedDNA meth-ylationchanges,aswellascontributingtowardsadaptiveresponses ofvariousspeciestotheireverchangingenvironmentandspecies biodiversity,canhaveasignificantcontributiontowardssomeof theadverseeffectsobservedinresponsetoenvironmental stress-ors immediately or at a later stage in life (a concept referred toas epigenetic predisposition and adult-onset of disease). For example,epigeneticmechanisms,aswellassinglenucleotide

poly-morphisms (SNPs), canprove tocontribute tothemechanisms

involvedinthesexdifferentiationandsexratioshiftsobservedin responsetochangesinthetemperatureofthewaterandthe plas-ticityofthisresponseinwildseabassfromdifferentgeographical locations.

Inadditiontotheconcernsaboutthepotentialof

pollutant-induced epigenetic modulation to impact on disease and

(6)

contaminantstowhichtheyhavebeenexposedthroughouttheir life-time. Such epigenetic variability has been shown to occur innaturalpopulationsoftheclonalﬁshChrosomuseos-neogaeus [39]. The adaptive changes observed can help identify popula-tionsvulnerabletoenvironmentalchange.Tostudythiseffectively, geneticallyidenticalorganismscanprovideastablebackground uponwhichaccumulationofepigeneticchangescanbemeasured andwearecurrentlyassessingthisinclonalvertebrateand inver-tebratespecies.

Theclassicalecologicalendpointsforecotoxicologymaynotbe sufficientinthat,asmentionedabove,earlylifestagechangesinthe epigenomefollowingacuteexposurehavethepotentialtocause diseaselaterinlifeandpotentiallythroughgenerations.Thisraises thepossibilityofbenefitsofinvestigatingepigeneticmarksinthe contextofenvironmentalmonitoringandecotoxicological assess-ments.Alimitationofthecurrent practicesfor ecotoxicological assessmentsandamajordifficultyforregulatorsinthecontextof waterqualityregulationssuchastheEUWaterFramework Direc-tive[96],istheinabilitytomake assessmentsofexposuresand theireffectsotherthanthrough“snap-shot”,expensiveanalysesof organismbiodiversity,individualorganismhealth,chemical mea-surementsandafewbiomarkersoflimiteddiagnosticvalue.We proposeanovelmechanismthatallowsresearchersto retrospec-tively interrogateexposurehistory and chemical effects onthe epigenome,thuspotentiallyhelpingtoprovideamechanism-based assessmentofthequalityandimpactoftheenvironment.The epi-genetic“memory”caninformonthelife-timeexposuretostressors thatmodifytheepigenome(Fig.1).Irrespectiveofwhethersuch changes maybeindicative of toxicity per se,thesignature has thepotentialtoactasasurrogateforassessmentoftoxic expo-suresand otherenvironmentalstressorsthat couldmanifestas diseasethroughalternativemechanismsforthesameagents.For example,oxidative stress is knowntobecaused by a range of organicchemicalsandmetalsandisfrequentlydetectedinanimals exposedtoaquaticpollutants[97].OxidativestresscanleadtoDNA hypomethylationthroughtheinhibitionofmethyltransferases[7]

andthismay,atleastinpart,explainthechangesinDNA methyl-ationintheliveroffish[98]andinDaphnia[85–88]treatedwith certainmetals.Agoodexampleofthepotentialuseofmethylation inenvironmentalmonitoringcomesfromtheeffectsofendocrine disruptors.The5flankingregionoftheun-transcribedvitellogenin (vtg1)gene inthebrainofthemale andfemaleadultzebrafish andtheliverofthemaleadultzebrafishishighlyhypermethylated comparedtothevtg1intheliveroffemalezebrafish,whereitis veryhighlyexpressedinresponsetoendogenousoestrogenduring oocytogenesis.[99].Vtg1isnotexpressednormallyintheliverof malefishbut,uponexposuretoexogenousestrogeniccompounds,

vtg1becomeshighlyexpressedandisassociatedwithfeminisation ofmalefish.Therefore,expressionofvitellogeninproteininmale fishisusedasastandardbiomarkerofexposuretooestrogens[100]. Ithasbeenshownthatthe5flankingregionofthevtg1genein malefishexposedtooestrogensissignificantlyhypomethylated, correspondingwithitsinduction[99].Therefore,itisreasonableto suggestthatDNAmethylationchangeshavebiomarkerpotential. Thereisalsothepossibilitythataspecificpatternofgene methyla-tionrepresentsanindicatorofthetypeofexposures.Althoughthe degreeandmechanismsofspecificityarenotestablished,thereis evidencetosuggestitsoccurrence.Forexample,phenobarbitalisa non-genotoxiccompoundthatcancausetumoursthroughthe con-stitutiveandrostanereceptorpathway(CAR)intheliveroftreated rodents.Lempiainenetal.,[101]demonstratedthattreatmentwith phenobarbitalcausesnon-randomandtissue-specificchangesin DNA methylationand transcription. While Cyp2b10,one ofthe

maingenesaffectedthroughtheCARpathway,was

hypomethy-latedandover-expressedintheliveroftreatedmice,itwasnot affectedinnon-tumour bearingkidney.Furthermore,therewas

nosigniﬁcantoverlapbetweenDNAmethylationor transcriptio-nalchangesobservedintheliverandkidneyofthetreatedmice. Severalotherstudies havealso indicatedthat atvarious stages

of phenobarbital-inducedtumourigenesis DNAmethylation and

transcription changes arenon-random (in theabsenceor pres-enceofmutageniccompounds)[102–104].Furtherevidenceonthe specificityoftheDNAmethylationprofileonthecausativeagent comesfroma studydemonstrating that hepatocellular carcino-masinducedbyhepatitisCvirus(HCV)orhepatitisBvirus(HBV) havedifferentmethylationprofiles[105].Thesestudiesallsupport theideathateachcategoryofcompoundscangiverisetoa spe-cificDNAmethylation“footprint”whichwillnotonlyinformon thepotentialforadverseeffectsduringlife-time,butmost impor-tantlywillalsoestablishandintroduceanovelretrospectiveand predictivemonitoringtooltoassessenvironmentalquality. There-fore,inthesamewaythatgeneticpolymorphismscaninfluence thesusceptibilityofindividualorganismstotoxicity,differences intheepigenomethatemergethroughoutthelife-timeofan indi-vidualmayalsohavethesameeffects[12].Indeed,modulationof methylationasaresultofenvironmentalstressessuchasmetals, could,throughselection,affordlocalpopulationsoforganismsa survivaladvantagethroughconsequentadvantageousgene expres-sionchanges[8].However,inidentificationofsuitableepigenetic biomarkers,itisimportanttodifferentiatebetweentheinitialDNA methylationchanges(referredtoasdrivermethylation)andthe DNAmethylationchangesthataretriggeredasaconsequenceof achange(e.g.formationoftumours).Thelatterisreferredtoas passenger methylation[25]and islessrelevant to environmen-talmonitoring.Furthermore,itisimportanttobearinmindthat someDNAmethylationchangesmaysimplybebiomarkers indica-tiveofexposureratherthannecessarilypredictiveofanadverse effect[91].

Furthermore,theﬁeld ofpaleoecology(resurrectionecology) mayprovide a unique opportunityfor investigating therole of epigeneticmechanismsinpopulationdiversityandadaptationto environmentalchanges.Daphniaspeciescanproducediapausing eggsviablefordecades[8].Samplingoftheseeggsfromsediment coresfrompondsandlakes,alongsidesediment chemistrydata, provideauniqueopportunityforunveilingecologicaland evolu-tionarychangesasaresultofenvironmentalchangesthroughout severaldecades[106,107].InareviewbyEadsetal.[108],several studieshavebeenhighlightedlookingatphenotypicplasticityand geneticdifferenceasaresultofenvironmentalchanges(e.g. chem-icals,introductionofanewpredator)usingDaphniarestingeggs datingbackseveraldecades.However,thefocusofthemajority of such studies is genomic variation (e.g. copy number varia-tion)andtherelationshiptophenotypicplasticityandadaptation. Consideringthewell-establishedroleofepigeneticmechanismsin regulatingtheresponsesoforganismstoenvironmentalfactors, adaptationandevolution[9,109–112],itseemshighlypossiblethat epigeneticmechanismshaveasubstantialroleintheevolutionand adaptationthatareobservedthroughoutdecades,forexamplein therestingeggsofDaphnia.Certainly,therestingeggsprovidean exceptionalopportunityforstudyingtheconceptof epigenetically-drivenphenotypicevolutionandadaptation.

Inconclusion,theadvancesinknowledgeofepigenetic mech-anismsandthepotentialforsuchchangestopersistthroughout life-time, and even beyond into subsequent generations, give

concernastowhethersuchchangesimposedbyenvironmental

(7)

Conﬂictofinterest

Theauthorsdeclarethattherearenoconﬂictsofinterest.

Acknowledgements

The authors thank the UK Natural Environmental Research

Council(NERC)for fundingtheirworkinthis area.Theauthors wouldalsoliketothankDr.TimWilliams(Universityof Birming-ham,UK)forhisinputintotheresearchrelatedtothisreview.

References

[1]M.Esteller,J.G.Heman,Cancerasanepigeneticdisease:DNAmethylation andchromatinalterationsinhumantumours,J.Pathol.196(2002)1–7.

[2]E.Kriukiene,Z.Liutkeviciute,S.Klimasauskas, 5-Hydroxymethylcytosine-theelusiveepigeneticmarkinmammalianDNA,Chem.Soc.Rev.41(2012) 6916–6930.

[3]T.Y.Xing,W.Ding,RecentadvanceinDNAepigeneticmodiﬁcations-thesixth baseinthegenome,ShengLiKeXueJinZhan43(2012)164–1670.

[4]J.-L.Gueant,F.Namour,R.-M.Gueant-Rodriguez,J.-L.Daval,Floateandfetal programming:aplayinepigenomics?TrendsEndocrinol.Metab.(2013),

http://dx.doi.org/10.1016/j.tem.2013.01.010.

[5]P.Dominguez-Salas,S.E.Cox,A.M.Prentice,B.J.Hennig,S.E.Moore,Maternal nutritionalstatus,C1metabolismandoffspringDNAmethylation:areviewof currentevidenceinhumansubjects,Proc.Natl.Acad.Sci.71(2012)154–165.

[6]J.A.Alegria-Torres,A.Baccarelli,V.Bollati,Epigeneticsandlifestyle, Epige-nomics3(2011)267–277.

[7]A.Baccarelli,V.Bollati,Epigeneticsandenvironmentalchemicals,Curr.Opin. Pediatr.21(2009)243–251.

[8]A.J.Morgan,P.Kille,S.R.Sturzenbaum,Microevolutionandecotoxicologyof metalsininvertebrates,Environ.Sci.Technol.41(2007)1085–1096.

[9]K.B.Flores,F.Wolschin,G.V.Amdam,TheroleofmethylationofDNAin envi-ronmentaladaptation,Integr.Comp.Biol.(2013),PMID:23620251.

[10]B.M.Turner,Epigeneticresponsestoenvironmentalchangeandtheir evo-lutionaryimplications,Philos.Trans.R.Soc.Lond.B:Biol.Sci.364(2009) 3403–3418.

[11]D.C.Dolinoy,R.Das,J.R.Weidman,R.L.Jirtle,Metastableepialleles,imprinting, andthefoetaloriginsofadultdiseases,Pediatr.Res.61(2007)30–37.

[12]B.C.Christensen,E.AndresHouseman,C.J.Marsit,S.Zheng,M.R.Wrensch,J.L. Wiemels,H.H.Nelson,M.R.Karagas,J.F.Padbury,R.Bueno,D.J.Sugarbaker, R.Yeh,J.K.Wiencke,K.T.Kelsey,Agingandenvironmentalexposuresalter tissue-speciﬁcDNAmethylationdependentuponCpGislandcontext,PLOS Genet.5(2009)1–13.

[13]E. Gomez-Diaz, M. Jorda, M.A. Peinado, A. Rivero, Epigenetics of host–pathogeninteractions:theroadaheadandtheroadbehind,PLoSPathog. 8(2012)e1003007.

[14]G.Egger,G.Liang,A.Apqricio,P.A.Jones,Epigeneticsinhumandiseaseand prospectsforepigenetictherapy,Nature429(2004)457–460.

[15]R.Jirtle,A.Bird,Epigeneticregulationofgeneexpression:howthegenome integrates intrinsic and environmental signals, Nat. Genet. 33 (2003) 245–254.

[16]B.M.Turner,Deﬁninganepigeneticcode,Nat.CellBiol.9(2007)2–6.

[17]T.A.Chan,S.Glockner,J.M.Yi,W.Chen,L.VanNeste,L.Cope,J.G.Herman, V.Velculescu,K.E.Schuebel,N.Ahuja,S.B.Baylin,Convergenceofmutation andepigeneticalterationsidentiﬁescommongenesincancerthatpredictfor poorprognosis,PLOSMed.5(2008)823–838.

[18]P.Cui,L.Zhang,Q.Lin,F.Ding,C.Xin,X.Fang,S.Hu,J.Yu,Anovelmechanismof epigeneticregulation:nucleosome-spaceoccupancy,Biochem.Biophys.Res. Commun.391(2010)884–889.

[19]A.V.Probst,E.Dunleavy,G.Almouzni,Epigeneticinheritanceduringthecell cycle,Nat.Rev.Mol.CellBiol.10(2009)192–206.

[20]I.P. Pogribny, Epigenetic events in tumourigenesis: putting the pieces together,Exp.Oncol.32(2010)132–136.

[21]K.Gronbaek,C.Hother,P.A.Jones,Epigeneticchangesincancer,Agric.Prod. MarketInform.Syst.115(2007)1039–1059.

[22]J.P.Thomson,P.J.Skene,J.Selfridge,T.Clouaire,J.Guy,S.Webb,A.R.W.Kerr,A. Deaton,R.Andrews,K.D.James,D.J.Turner,R.Illingworth,A.Bird,CpGislands inﬂuencechromatinstructureviatheCpG-bindingproteinCfp1,Nature464 (2010)1082–1087.

[23]J.S.Butler,J.H.Lee,D.G.Skalnik,CFP1interactswithDNMT1independentlyof associationwiththeSetd1histoneH3K4methyltransferasecomplexes,DNA CellBiol.27(2008)533–543.

[24]C.M.Tate,J.Lee,D.G.Skalnik,CXXCﬁngerprotein1restrictstheSetd1A his-toneH3K4methyltransferasecomplextoeuchromatin,FASEBJ.277(2010) 210–223.

[25]Z.Herceg, T.Vaissiere, Epigeneticmechanismsandcancer. Aninterface betweentheenvironmentandthegenome,Epigenetics6(2011)804–819.

[26]M.Tahiliani,K.PengKoh,Y.Shen,W.A.Pastor,H.Bandukwala,Y.Brudno,S. Agarwal,L.M.Iyer,D.R.Liu,L.Aravind,A.Rao,Conversionof5-methylcytosine to5-hydroxymethylcytosineinmammalianDNAbyMLLpartnerTET1, Sci-ence324(2009)930–935.

[27]K.Williams,J.Christensen,M.TerndrupPedersen,J.V.Johansen,P.A.C.Cloos, J.Rappsilber,K.Helin,TET1andhydroxymethylcytosineintranscriptionand DNAmethylationﬁdelity,Nature473(2011)343–349.

[28]A.Bird,DNAmethylationpatternsandepigeneticmemory,GenesDev.16 (2002)6–21.

[29]R.Holliday,DNAmethylationandepigenotypes,Biochemistry(Mosc).70 (2005)500–504.

[30]M.Godmann,R.Lambrot,S.Kimmins,Thedynamicepigeneticprogramin malegermcells:itsroleinspermatogenesis,testiscancer,anditsresponse totheenvironment,Microsc.Res.Tech.72(2009)603–619.

[31]Y.He,S.D.Michaels,R.M.Amasino,Regulationofﬂoweringtimebyhistone acetylationinArabidopsis,Science302(2003)1751–1754.

[32]R.Bastow,J.S.Mylne,C.Lister,Z.Lippman,R.A.Martienssen,C.Dean, Vernal-izationrequiresepigeneticsilencingofFLCbyhistonemethylation,Nature 427(2004)164–167.

[33]J.E.Burn,D.J.Bagnall,J.D.Metzger,E.S.Dennis,W.J.Peacock,DNAmethylation, vernalization,andtheinitiationofﬂowering,Proc.Natl.Acad.Sci.90(1993) 287–291.

[34]M.E.Wilson,T.Sengoku,DevelopmentalregulationofneuronalgenesbyDNA methylation:environmentalinﬂuences,Int.J.Dev.Neurosci.(2013),PMID: 23501000.

[35]D.Crews,Epigenetics,brain,behaviour,andtheenvironment,Hormones (Athens)9(2010)41–50.

[36]M.M.McCarthy,D.Crews,Epigenetics-newfrontiersinneuroendocrinology, Front.Neuroendocrinol.29(2008)341–343.

[37]L.Navarro-Martin,J.Vinas,L.Ribas,N.Diaz,A.Gutierrez,L.DiCroce,F.Piferrer, DNAmethylationofthegonadalaromatase(cyp19a)promoterisinvolvedin temperature-dependentsexratioshiftsintheEuropeanseabass,PLOSGenet. 7(2011)1–15.

[38]R.G.Contractor,C.M.Foran,S.Li,K.L.Willett,Evidenceofgender-and tissue-speciﬁcpromotermethylationandthepotentialforethinylestradiolinduced changesinJapanesemedaka(Oryziaslatipes)estrogenreceptorandaromatase genes,Environ.Health67(2004)1–22.

[39]R.Massicotte,E.Whitelaw,B.Angers,DNAmethylation.Asourceofrandom variationinnaturalpopulations,Epigenetics6(2011)421–427.

[40]F.Lyko,S.Foret,R.Kucharski,S.Wolf,C.Falckenhayn,R.Maleszka,Thehoney beeepigenomes:differentialmethylationofBbrainDNAinqueensand work-ers,PLoSBiol.8(2010)1–12.

[41]P.Moran,A.Perez-Figueroa,Methylationchangesassociatedwithearly mat-urationstagesintheAtlanticsalmon,BMCGenet.12(2011)1–8.

[42]M.F.Fraga,E. Ballester,M.F.Paz, S.Ropero, F.Setien,M.L.Ballestar, D. Heine-Suner,J.C.Cigudose,M.Urioste,J.Benitez,M.Boix-Chornet,A. Sanchez-Aguilera,C.Ling,E.Carlsson,P.Poulsen,A.Vaag,Z.Stephan,T.D.Spector,Y. Wu,C.Plass,M.Esteller,Epigeneticdifferencesariseduringthelifetimeof monozygotictwins,Proc.Natl.Acad.Sci.102(2005)10604–10609.

[43]I.P.Pogribny,F.A.Beland,DNAmethylomealterationsinchemical carcino-genesis,CancerLett.334(2013)39–45.

[44]V.K.Cortessis,D.C.Thomas,A.J.Levine,C.V.Breton,T.M.Mack,K.D.Siegmund, R.W.Haile,P.W.Laird,Environmentalepigenetics:prospectsforstudying epigeneticmediationofexposure–responserelationships,Hum.Genet.131 (2012)1565–1589.

[45]R.Martinez-Zamudio,H.C.Ha,Environmentalepigeneticsinmetalexposure, Epigenetics6(2011)820–827.

[46]A.C.Gore,Developmentalprogrammingandendocrinedisruptoreffectson reproductiveneuroendocrinesystems,Front.Neuroendocrinol.29(2008) 358–374.

[47]D.Crews,Epigeneticsanditsimplicationsforbehavioralneuroendocrinology, Front.Neuroendocrinol.29(2008)344–357.

[48]S.Ho,W.Tang,J.B.deFrausto,G.S.Prins,Developmentalexposureto estra-diolandbisphenolAincreasessusceptibilitytoprostatecarcinogenesisand epigeneticallyregulatesphosphodiesterasetype4variant4,CancerRes.66 (2006)5624–5632.

[49]V.Michel,Z.Yuan,S.Ramsubir,M.Bakovic,Cholinetransportforphospholipid synthesis,Exp.Biol.Med.231(2006)490–504.

[50]J.Locker,T.V.Reddy,B.Lombardi,DNAmethylationandhepatocarcinogenesis inratsfedacholine-devoiddiet,Carcinogenesis7(1986)1309–1312.

[51]M.J.Wilson,N.Shivapurkar,L.A.Poirier,Hypomethylationofhepaticnuclear DNAinratsfedwithacarcinogenicmethyl-deﬁcientdiet,Biochem.J.218 (1984)987–990.

[52]A.K.Ghoshal,E.Farber,Theinductionoflivercancerbydietarydeﬁciencyof cholineandmethioninewithoutaddedcarcinogens,Carcinogenesis5(1984) 1367–1370.

[53]S.Sibani,S.Melnyk,I.P.Pogribny,W.Wang,F.Hiou-Tim,L.Deng,J.Trasler, S.J.James,R.Rozen,Studiesofmethioninecycleintermediates(SAM,SAH), DNAmethylationandtheimpactoffolatedeﬁciencyontumournumbersin Minmice,Carcinogenesis23(2002)61–65.

[54]L. Mirbahai, A.D. Southam, U. Sommer, T.D. Williams, J.P. Bignell, B.P. Lyons,M.R. Viant,J.K. Chipman,Disruption of DNAmethylation viaS -adenosylhomocysteineisakeyprocessinhighincidencelivercarcinogenesis inﬁsh,J.ProteomeRes.(2013),PMID:23611792.

[55]S.P.Barros,S.Offenbacher,Epigenetics:connectingenvironmentand geno-typetophenotypeanddisease,J.Dent.Res.88(2009)400–408.

[56]R.L.Jirtle,M.K.Skinner,Environmentalepigenomicsanddisease susceptibil-ity,Nat.Rev.Genet.8(2007)253–262.

(8)

diethylstilbesterol-induced cancers, Ann. N. Y. Acad. Sci. 983 (2003) 161–169.

[58]M.K.Skinner,M.Manikkam,C.Guerrero-Bosagna,Epigenetic transgenera-tionalactionsofenvironmentalfactorsindiseaseetiology,TrendsEndocrinol. Metab.21(2010)214–222.

[59]G.S.Prins,Estrogenimprinting:whenyourepigeneticmemoriescomeback tohauntyou,Endocrinology149(2008)5919–5921.

[60]M.D.Anway,M.A.Memon,M.Uzumcu,M.K.Skinner,Transgenerationaleffect oftheendocrinedisruptorvinclozolinonmalespermatogenesis,J.Androl.27 (2006)868–879.

[61]R.R.Newbold,R.B.Hanson,W.N. Jefferson,B.C.Bullock,J. Haseman,J.A. McLachlan,Proliferativelesionsandreproductivetract tumoursinmale descendantsofmiceexposeddevelopmentallytodiethylstilbestrol, Carcino-genesis21(2000)1355–1363.

[62]G.S.Prins,L.Birch,W.-Y.Tang,S.-M.Ho,Developmentalestrogenexposures predisposetoprostatecarcinogenesiswithaging,Reprod.Toxicol.23(2007) 374–382.

[63]A.Andres,D.B.Muellener,G.U.Ryffel,Persistence,methylationand expres-sionofvitellogeningenederivativesafterinjectionintofertilizedeggsof Xenopuslaevis,NucleicAcidsRes.12(1984)2283–2302.

[64]S.Li,K.A.Washburn,R.Moore,T.Uno,C.Teng,R.R.Newbold,J.A.McLachian, M.Negishi,Developmentalexposuretodiethylstilbestrolelicits demethyla-tionofestrogen-responsivelactoferringeneinmouseuterus,CancerRes.57 (1997)4356–4359.

[65]D.C.Dolinoy,Theagoutimousemodel:anepigeneticbiosensorfornutritional andenvironmentalalterationsonthefoetalepigenome,Nutr.Rev.66(2008) 1–8.

[66]D.C.Dolinoy,J.Wiedman,R.Waterland,R.L.Jirtle,Maternalgenisteinalters coatcolourandprotectsAvymouseoffspringfromobesitybymodifyingthe foetalepigenome,Environ.HealthPerspect.114(2006)567–572.

[67]L.Mirbahai,G.Yin,J.P.Bignell,N.Li,T.D.Williams,J.K.Chipman,DNA meth-ylationinlivertumourigenesisinﬁshfromtheenvironment,Epigenetics6 (2011)1319–1333.

[68]A.P.Feinberg,R.Ohlsson,S.Henikoff,Theepigeneticprogenitororiginof humancancer,Nat.Rev.Genet.7(2006)21–33.

[69]N.A.Youngson,E.Whitelaw,Transgenerationalepigeneticeffects,Annu.Rev. GenomicsHum.Genet.9(2008)233–257.

[70]W.Reik,J.Walter,Evolutionofimprintingmechanisms:thebattleofthesexes beginsinthezygote,Nat.Genet.27(2001)255–256.

[71]N.C.Whitelaw,E.Whitelaw,Transgenerationalepigeneticinheritancein healthanddisease,Curr.Opin.Genet.Dev.18(2008)273–279.

[72]D.Francis,J.Diorio,D.Liu,M.J.Meaney,Nongenomictransmissionacross generationsofmaternalbehaviourandstressresponsesintherat,Science 286(1999)1155–1158.

[73]I.C.G.Weaver,N.Cervoni,F.A.Champagne,A.C.D’Alessio,S.Sharma,J.R. Seckl,S.Dymov,M.Szyf,M.J.Meaney,Epigeneticprogrammingbymaternal behaviour,Nat.Neurosci.7(2004)847–854.

[74]I.C.G.Weaver,A.C.D’Alessio,S.E.Brown,I.C.Hellstrom,S.Dymov,S.Sharma, M.Szyf,M.J.Meaney,ThetranscriptionfactornervegrowthfactoriproteinA mediatesepigeneticrogramming:alteringepigeneticmarksby immediate-earlygenes,J.Neurosci.27(2007)1756–1768.

[75]M.D.Anway,A.S.Cupp,M.Uzumcu,M.K.Skinner,Epigenetic transgenera-tionalactionsofendocrinedisruptorsandmalefertility,Science308(2005) 1466–1469.

[76]M.D.Anway,M.K.Skinner,Epigenetictransgenerationalactionsofendocrine disruptors,Endocrinology147(2006)S43–S49.

[77]M.D.Anway,M.K.Skinner,Transgenerationaleffectsoftheendocrine dis-ruptorvinclozolinontheprostatetranscriptomeandadultonsetdisease, Prostate68(2008)517–529.

[78]M.D.Anway,S.S.Rekow,M.K.Skinner,Transgenerationalepigenetic program-mingoftheembryonictestistranscriptome,Genomics91(2008)30–40.

[79]M.D.Anway,S.S.Rekow,M.K.Skinner,Comparativeanti-androgenicactions ofvinclozolinandﬂutamideontransgenerationaladultonsetdiseaseand spermatogenesis,Reprod.Toxicol.26(2008)100–106.

[80]C.Stouder,A.Paoloni-Giacobino,Transgenerationaleffectsoftheendocrine disruptorvinclozolinonthemethylationpatternofimprintedgenesinthe mousesperm,Reproduction139(2010)373–379.

[81]E.Nilson,G.Larsen,M.Manikkam,C.Guerrero-Bosagna,M.I.Savenkova,M.K. Skinner,Environmentallyinducedepigenetictransgenerationalinheritance ofovariandisease,PLoSONE7(2012)e36129.

[82]K.Inawaka,M.Kawabe,S.Takahashi,Y.Doi,Y.Tomigahara,H.Tarui,J.Abe, S.Kawamura,T.Shirai,Maternalexposuretoanti-androgeniccompounds, vinclozolin,ﬂutamideandprocymidone,hasnoeffectsonspermatogenesis andDNAmethylationinmaleratsofsubsequentgenerations,Toxicol.Appl. Pharmacol.237(2009)178–187.

[83]S.Schneider,W.Kaufmann,R.Buesen,B.VanRavenzwaay,Vinclozolin—the lackofatransgenerationaleffectafteroralmaternalexposureduring organo-genesis,Reprod.Toxicol.25(2008)352–360.

[84]S. Schneider, H. Marxfeld, S. Groters, R. Buesen, B. van Ravenzwaay, Vinclozolin—Notransgenerational inheritance of anti-androgenic effects aftermaternalexposureduringorganogenesisviatheintraperitonealroute, Reprod.Toxicol.37(2013)6–14.

[85]M.B.Vandegehuchte,D.DeConinck,T.Vandenbrouck,W.M.DeCoen,C.R. Janssen,Genetranscriptionproﬁles,globalDNAmethylationandpotential transgenerationalepigeneticeffectsrelatedtoZnexposurehistoryinDaphnia magna,Environ.Pollut.158(2010)3323–3329.

[86]M.B.Vandegehuchte,T.Kyndt,B.Vanholme,A.Haegeman,G.Gheysen,C.R. Janssen,OccuranceofDNAmethylationinDaphniamagnaandinﬂuenceof multigenerationCdexposure,Environ.Int.35(2009)700–706.

[87]M.B.Vandegehuchte,F.Lemiere,C.R.Janssen,QuantitativeDNA methyla-tioninDaphniamagnaandeffectsofmultigenerationZnexposure,Comp. Biochem.Physiol.150(2009)343–348.

[88]M.B.Vandegehuchte,F. Lemiere, L. Vanhaecke,W.Vanden Berghe, C.R. Janssen,DirectandtransgenerationalimpactonDaphniamagnaof chemi-calswithaknowneffectonDNAmethylation,Comp.Biochem.Physiol.151 (2010)278–285.

[89]M.B.Vandegehuchte,C.R.Janssen,Epigeneticsanditsimplicationsfor eco-toxicology,Ecotoxicology20(2011)607–624.

[90]K.D.M.Harris,N.J.Bartlett,K.L.Vett,Daphniaasanemergingepigeneticmodel organism,Genet.Res.Int.2012(2012)1–8.

[91]J.I.Goodman,K.A.Augustine,M.L.Cunnningham,D. Dixon,Y.P.Dragan, J.G. Falls, R.J. Rasoulpour, R.C. Sills, R.D. Storer, D.C. Wolf, S.D. Pettit, What do we need to know prior to thinking about incorporating an epigeneticevaluation into safety assessments? Toxicol. Sci. 116 (2010) 375–381.

[92]J.G.Moggs,J.I.Goodman,J.E.Trosko,R.A.Robertsd,Epigeneticsandcancer. Implicationsfordrugdiscoveryandsafetyassessment,Toxicol.Appl. Pharma-col.196(2004)422–430.

[93]J.Legler,Epigenetics:anemergingﬁeldinenvironmentaltoxicology,Integr. Environ.AssessManag.6(2010)314–315.

[94]J.Sved,A.Bird,TheexpectedequilibriumoftheCpGdinucleotidein ver-tebrategenomesunderamutationmodel,Proc.Natl.Acad.Sci.87(1990) 4692–4696.

[95]C.Guerrero-Bosagna,P.Sabat,L.Valladares,Environmentalsignalingand evolutionarychange:canexposureofpregnantmammalstoenvironmental estrogensleadtoepigeneticallyinducedevolutionarychangesinembryos? Evol.Dev.7(2005)341–350.

[96]http://www.wfduk.org/faq/HowAreTheNewClassiﬁcationSystemsDifferent FromPreviousClassiﬁcationSystems

[97]I. Ahmad,M.Pacheco,M.A.Santos,Anguilla anguillaL.,oxidativestress biomarkers:aninsitustudyoffreshwaterwetlandecosystem(Pateirade Fermentelos,Portugal),Chemosphere65(2006)952–962.

[98]X.W.Zhou,G.N.Zhu,M.Jilisa,J.H.Sun,InﬂuenceofCu,Zn,Pb,Cdandtheir heavymetalionmixtureontheDNAmethylationleveloftheﬁsh(Carassius auratus),J.Environ.Sci.(China)21(2001)549–552.

[99]M.Stromqvist,N.Tooke,B.Brunstrom,DNAmethylationlevelsinthe5 ﬂank-ingregionofthevitellogeninIgeneinliverandbrainofadultzebraﬁsh(Danio rerio)-sexandtissuedifferencesandeffectsof17␣-ethinylestradiolexposure, Aquat.Toxicol.98(2010)275–281.

[100]A.P.Scott,M.Sanders,G.D.Stentiford,R.A.Reese,I.Katsiadaki,Evidencefor estrogenicendocrinedisruptioninanoffshoreﬂatﬁsh,thedab(Limanda limanda),Mar.Environ.Res.64(2007)128–148.

[101]H.Lempiainen, A. Muller,S. Brasa,S. Teo,T.C. Roloff, L. Morawiec, N. Zamurovic,A.Vicart,E.Funhoff,P.Couttet,D.Schubeler,O.Grenet,J. Mar-lowe,J.Moggs,R.Terranova,Phenobarbitalmediatesanepigeneticswitchat theconstitutiveandrostanereceptor(CAR)targetgeneCyp2b10intheliver ofB6C3F1mice,PLoSONE6(2011)1–14.

[102]J.M.Phillips,L.D.Burgoon,J.I.Goodman,Phenobarbitalelicitsunique,early changesintheexpressionofhepaticgenesthataffectcriticalpathwaysin tumour-proneB6C3F1mice,Toxicol.Sci.109(2009)193–205.

[103]J.M.Phillips,J.I.Goodman,Multiplegenesexhibitphenobarbital-induced constitutiveactive/androstanereceptor-mediatedDNAmethylationchanges duringlivertumourigenesisandinlivertumours,Toxicol.Sci.108(2009) 273–289.

[104]A.N.Bachman,J.M.Phillips,J.I.Goodman,Phenobarbitalinducesprogressive patternsofGC-richandgene-speciﬁcalteredDNAmethylationintheliverof tumour-proneB6C3F1mice,Toxicol.Sci.91(2006)393–405.

[105]Y.Pei,T.Zhang,V.Renault,X.Zhang,Anoverviewofhepatocellularcarcinoma studybyomics-basedmethods,ActaBiochim.Biophys.Sin.(Shanghai)41 (2009)1–15.

[106]N.Brede,C.Sandrock,D.Straile,P.Spaak,T.Jankowski,B.Streit,K.Schwenk, Theimpactofhuman-madeecologicalchangesonthegeneticarchitectureof Daphniaspecies,Proc.Natl.Acad.Sci.106(2009)4758–4763.

[107]W.C.Kerfoot,L.J.Weider,Experimentalpaleoecology(resurrectionecology): chasingVanValen’sRedQueenhypothesis,Limnol.Oceanogr.49(2004) 1300–1316.

[108]B.D. Eads, J. Andrews, J.K. Colbourne, Ecological genomics in Daphnia: stressresponsesandenvironmentalsexdetermination,Heredity100(2008) 184–190.

[109]C.Furusawa,K.Kaneko,Epigeneticfeedbackregulationaccelerates adapta-tionandevolution,PLoSONE8(2013)e61251.

[110]O.Paun,Bateman,R.M.Fay,M.Hedren,L.Civeyrel,M.W.Chase,Stable epi-geneticeffectsimpactadaptationinallopolyploidorchids(Dactylorhiza: Orchidaceae),Mol.Biol.Evol.27(2010)2465–2473.

[111]K.Brautigam,K.J.Vining,C.Lafon-Placette,C.G.Fossdal,M.Mirouze,J.G. Mar-cos,S.Fluch,M.F.Fraga,M.A.Guevara,D.Abarca,O.Johnsen,S.Maury,S.H. Strauss,M.M.Campbell,A.Rohde,C.Diaz-Sala,M.T.Cervera,Epigenetic reg-ulationofadaptiveresponsesofforesttreespeciestotheenvironment,Ecol. Evol.3(2013)399–415.