Use of combined microscopic and spectroscopic techniques to reveal interactions between uranium and Microbacterium sp. A9, a strain isolated from the Chernobyl exclusion zone

HAL Id: hal-01561096

https://hal-amu.archives-ouvertes.fr/hal-01561096

Submitted on 12 Jul 2017

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of

sci-entific research documents, whether they are

pub-lished or not. The documents may come from

teaching and research institutions in France or

abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est

destinée au dépôt et à la diffusion de documents

scientifiques de niveau recherche, publiés ou non,

émanant des établissements d’enseignement et de

recherche français ou étrangers, des laboratoires

publics ou privés.

Use of combined microscopic and spectroscopic

techniques to reveal interactions between uranium and

Microbacterium sp. A9, a strain isolated from the

Chernobyl exclusion zone

Nicolas Theodorakopoulos, Virginie Chapon, Frederic Coppin, Magali

Floriani, Thomas Vercouter, Claire Sergeant, Virginie Camilleri, Catherine

Berthomieu, Laureline Fevrier

To cite this version:

Use

of

combined

microscopic

and

spectroscopic

techniques

to

reveal

interactions

between

uranium

and

Microbacterium

sp.

A9,

a

strain

isolated

from

the

Chernobyl

exclusion

zone

Nicolas

Theodorakopoulos

_,

_Virginie

_Chapon

_,

_Fréderic

_Coppin

_,

_Magali

_Floriani

_,

Thomas

Vercouter

_,

_Claire

_Sergeant

_,

_Virginie

_Camilleri

_,

_Catherine

_Berthomieu

_,

Laureline

Février

a_{CEA,DSV,IBEB,SBVME,LIPM,F-13108Saint-Paul-lez-Durance,France} b_{CNRS,UMR7265,F-13108Saint-Paul-lez-Durance,France}

c_{Universitéd’Aix-Marseille,F-13108Saint-Paul-lez-Durance,France}

d_{IRSN/PRP-ENV/SERIS/L2BT,bat183,B.P.3,F-13115SaintPaul-lez-Durance,France} e_{CEA,DEN,DANS,DPCSEARS,LANIE,F-91191Gif-Sur-YvetteCedex,France} f_{UnivBordeaux,CENBG,UMR5797,F-33170Gradignan,France}

g_{CNRS,IN2P3,CENBG,UMR5797,F-33170Gradignan,France}

1. Introduction

Uranium(U)is a long-lived naturallyoccurringradionuclide whichisbothchemoandradio-toxic.ItsconcentrationinEuropean

∗ Correspondingauthor.Tel.:+33442199517;fax:+33442199151. E-mailaddress:laureline.fevrier@irsn.fr(L.Février).

radioactivewasterepositories,resultinginlocallyhighuranium concentrations[3].

Researchfocusingonevaluatingthemobility ofthiselement andonfindingbioremediationprocesseshasbeenconductedon aworldwide-scale[4].Importantly,somelaboratoryexperiments havealreadydemonstratedtheutilityofcertainbacteriafor reduc-inguraniummobility.Inoxicenvironments,bacteriamayinteract directlywithuraniumthroughbiosorptionatthecellsurfaceorby intracellularbioaccumulation,orindirectlythroughmodification ofsurroundinggeochemicalconditions,leadingtoprecipitationof U[5,6].However,theviabilityofbacteriaintheseprocessesisrarely considered,althoughthisfactorcouldbecriticaltothesuccessof long-termbioremediationapplications.

It has already been shown that highly contaminated envi-ronmentsactaspotentialreservoirssuitablefortheisolationof U-resistantbacteria[7,8].Inapreviouswork,weexaminedaset of50bacterialstrainsisolatedfroma radioactivewaste reposi-toryin theChernobylexclusion zone [3]. Fromthis screen, we selectedastrainaffiliatedtothegenusMicrobacterium,whichis abletosurviveU(VI)exposure.PreviousreportsonU(VI) interac-tionwithMicrobacteriumstrainsrevealedtheirabilitytointeract withupto500_␮MofUO2(NO3)2[9].Membersofthisgenushave beendetectedinradioactivewastecontaminatedareasorin natu-rallyU-richsoils[10–12],andarethusgoodcandidatestostudythe interactionsofcellswithU.Moreover,someMicrobacteriumspecies exhibitinterestingfeaturessuchaspolymetalresistance[13]and havebeenproposedforbioremediationapplications[14].

Thegoalofthisworkwastoidentifyinteractionmechanisms employedbylivingMicrobacteriumsp.A9straincellsexposedto arangeof U(VI)concentrations.Todistinguishactivefrom pas-sive mechanisms, we varied the experimental temperature, as it influences the activity of bacteria. The survival rate of bac-teria was assessed, in addition to the quantification of U(VI) removalandphosphaterelease. Finally,thelocalisationandthe speciationofUthat interactedwiththecells wereanalysed by microscopic(TEM–EDX)andspectrometricapproaches(ATR-FTIR andTRLFS).

2. Materialandmethods

2.1. Bacterialstrain

ThestrainMicrobacteriumsp.A93sp312(referredtohereas Microbacteriumsp.A9strain)wasisolatedfromChernobyltrench T22soil[3],acontaminatedwastestoragesitelocatednearthe Chernobylnuclearpowerplant[15].Thestrainwasroutinely cul-tivatedin0.1×TrypticSoyBroth(TSB,DifcoLaboratories)at32◦C withshaking.

2.2. Uraniumexposure

TheMicrobacteriumsp.A9 strain wascultivated in0.1×TSB mediumuntiltheexponentialgrowthphase,andcellswere har-vestedbycentrifugationfor10minat5000g.Fromthisstageon, thesamplesweremaintainedeitherat4◦Corat25◦Cthroughout theexperiment.Thecellpelletswerewashedtwicein0.1MNaCl andwerere-suspendedatabout6×109_bacteriamL−1_in0.1MNaCl pH5with0,10or50␮MU(VI).U(VI)wasaddedasuranylnitrate UO2(NO3)2·6H2O(Sigma–Aldrich)froma7.51mMstocksolutionin 0.016MHNO3.Thenitrateconcentrationwasadjustedto0.416mM byaddingNaNO3whenneeded.Speciationof10and50␮MU(VI) intheexposuremediaat25◦Cor4◦CwassimulatedusingVisual MINTEQ(ver.3.0).

BacteriaexposedtoU(VI)wereincubatedwithshakingat25◦C or4◦C.Triplicatesweremadeforeachcondition.Aliquotswere

takenafter0.5,2,4,6,10and24h.Blankcontrolswithout bacte-riawereperformedtoexcludeabioticuraniumremovalfromthe exposuresolution.

2.3. Uraniumandphosphatequantification

The sampleswere centrifuged at8000g for 5min. Uin the supernatant was analysed by inductively coupled plasma-atomic emission spectrometry (ICP-AES Optima 4300DV, PerkinElmer).

Inorganicphosphate(Pi)inthesupernatantwasquantifiedby colorimetric measurements, using the molybdophosphoric acid bluemethod[16].Potassiumdihydrogenphosphatesolutionswere usedasstandards.Allmeasurementswereperformedat25◦Cina 96-wellmicroplateandrecordedat720nm.

2.4. Bacterialviability

Aliquotsofcellsuspensionstakenat0.5and24hwerediluted in0.1×TSBandspreadon0.1×TSBagarplates.ColonyForming Units(CFUs)werecountedafter24hat30◦C.

2.5. Microscopy(TEM–EDX)analysis

Bacterialcell pellets werefixedin sodiumcacodylate buffer (0.1M,pH7.4)supplementedwith2.5%glutaraldehyde.After24h at4◦C,thesampleswerewashedthreetimeswithsodium cacody-latebufferandpost-fixedinthesamebuffercontaining1%osmium tetroxide(OsO4)for1h.Thesamplesweredehydratedthrougha gradedethanolseries,andfinallyembeddedinanEpon812resin. Allchemicals used for histologicalpreparation were purchased fromElectronMicroscopySciences.Sampleswerecutinultra-thin sectionsusingaUCTultramicrotome(LeicaMicrosystemsGmbH). Sectionswerethenmountedoncoppergridsandexaminedwith a Scanning TransmissionElectronMicroscope (TEM/STEM; Tec-naiG2_{Biotwin, FEI)equipped witha CCD camera Megaview III} (OlympusSoftImagingSolutionsGmbH).Atleast200imageswere analyzedforeachcondition.ThelocalizationofUwasconducted using a Phoenix Energy Dispersive X-rayanalyzer (EDAX Inc.), equippedwithaSuperUltra-ThinWindowmodelsapphiredetector withacountingtimeof100sec.

2.6. Time-resolvedlaser-inducedfluorescencespectroscopy (TRLFS)

2.7. Attenuatedtotalreflectionfouriertransforminfra-red spectroscopy(ATR-FTIR)

Thecellpelletswerewashedwithultrapurewaterandwere deposited and dried on the ATRsurface (a 9 bounce diamond microprismwitha4.3mmsurfacediameterandZnSeoptics; Sen-sIrTechnologies)toavoidbackgroundabsorptionfromwater.All infraredspectrawere recordedwitha 4cm−1 resolution inthe 4000–400cm−1 range, using a Bruker IFS28 FTIR spectrometer equippedwithaDTGSdetector(SensIRATRSetup).Twospectra wereacquiredforeachcondition.Typically,300scanswere accu-mulatedforeachspectrum.

2.8. Statisticalanalyses

AllstatisticalanalyseswereperformedwithRsoftware[18]. ResultsofcellviabilityandPireleasewereanalyzedbyANOVA, aftercheckingassumptionsofnormalityandvariancehomogeneity ofresiduals.Alphalevelswere≤0.05(*)and≤0.001(**).Datawere presentedasmean±standarddeviationofthemean.

3. Results

3.1. U(VI)speciationintheexposuremedia

Asnoprecipitatewasobservedduringthepreparationofour exposuremedia,precipitationhasbeenexcludedfromthe simula-tionsofU(VI)speciation.U(VI)wasmainlypresentasbioavailable forms(UO22+andUO2OH+)(Table1)[19].At50␮M,bioavailable U(VI)formsconstituted60%(at25◦C)and80%(at4◦C)ofthe sol-ubleU(VI)forms,whereastheyconstituted86%(25◦C)and94% (4◦C)at10␮M.

3.2. BacterialviabilityuponU(VI)exposure

NomortalitywasmeasuredinthecontrolswithoutU(VI)after 24hat4◦Cand25◦C,indicatingthatbacteriaremainedviable dur-ingtheexperiment(Fig.1).Similarresultswereobtainedforthe samplesincubatedat4◦CwithbothU(VI)concentrationsandthe samplesincubatedat25◦Cwith10␮MU(VI).Incontrast,

incuba-Table1

SimulationofU(VI)speciationintheexposuresolutions.

Speciesname 50␮M25◦C 50␮M4◦C 10␮M25◦C 10␮M4◦C (%)oftotalconcentration UO22+ 46.96 63.17 67.53 73.83 UO2OH+ 12.66 17.00 18.19 19.84 (UO2)2(OH)22+ 19.98 8.39 8.25 2.29 (UO2)3(OH)5+ 14.98 0.93 1.77 0.06 UO2Cl+ 2.51 2.72 3.63 3.19 (UO2)3(OH)42+ 1.40 3.63 0.17 0.23 (UO2)4(OH)7+ 0.79 2.94 0.03 0.04 (UO2)2OH3+ 0.57 1.02 0.23 0.28 UO2(OH)2 0.12 0.17 0.18 0.20 UO2NO3+ 0.01 0.03 0.02 0.03

tionat25◦Cinthepresenceof50␮MU(VI)ledto61%mortality after24h.

3.3. SequestrationbyMicrobacteriumsp.A9strain

ThekineticsofU(VI)removalfromtheexposuresolutionby bacterial cells wasdetermined by measuringU content in the supernatant(Fig.2).AbioticcontrolsshowedthatU(VI)remains solublewithinthe24hexposureinallconditionsandconfirmed that U(VI) removal resulted exclusivelyfrom biotic interaction. Withinthefirst30minofexposureat50␮MU(VI),78%ofthe ini-tialamountofU(VI)wasremovedfromthesupernatantatboth temperatures(Fig.2B).Then,U(VI)removalreachedanapparent equilibriumat4◦C,whereasitcontinuedat25◦C.Theproportions ofU(VI)removedfromthesupernatantafter24hwere94%and86% at25◦Cand4◦C,respectively.

At10␮MU(VI),arapidmetalremovalcorrespondingtomore than90%ofthetotalU(VI)tookplacewithinthefirst30minat bothtemperatures(Fig.2A).Afterthisstep,greatdifferenceswere observedbetweenthetwotemperatures. At4◦C,U(VI)removal ceasedandthesystemremainedstableuntiltheendofthe exper-iment.Bycontrast,U(VI)releaseintheexposuremediumat25◦C was observed between 0.5 and 4h, followed by a slow U(VI) removalthatfinallyresultedinatotalaccumulationof86%U(VI) after24h. Control 10 µM U 50 µM U Control 10 µM U 50 µM U 104 105 106 107 108 109 1010 25°C 4°C * B ac te ri al co u nt (log C F U/ ml )

0 2 8 10 12

U

rem

ai

nin

g

in th

e

ce

ll

-fr

ee

sup

er

n

ata

nt

(µM

)

U10 abioticcontrol 25°C

U10 abioticcontrol 4°C

U10 25°C

U10 4°C

Expo

sure ti

me (hours)

0 5 10 15 20 25

0 10 40 50

U50 abioticcontrol 25°C

U50 abioticcontrol 4°C

U50 25°C

U50 4°C

B

Fig.2. Uconcentrationovertimeinthecell-freesupernatantduringcellexposure to10␮M(A)and50␮M(B)U(VI)at4◦C(opensymbols)and25◦C(filledsymbols). Abioticcontrolsarerepresentedbysquares;bioticconditionsarerepresentedby circles.

3.4. PireleaseduringU(VI)exposure

WemeasuredPicontentinallsupernatants,sinceitsrelease bycellscanstronglyinfluenceUspeciationandsolubility.At4◦C, aconstantandlowquantityofPi(7.4±0.76␮Minaverage)was measuredinthesupernatantsandnostatisticallysignificant dif-ferenceswerefoundbetweenthecontrolsampleswithoutU(VI) andsamplesexposedto10␮M U(VI)(Fig.3A).However,foran unexplainedreason,inthecontrolsamples,thisamountwashigh comparedtothatobservedat25◦Catthebeginningofthe incu-bation.Nevertheless,strongdifferencesappearedatlongertimeof incubation.Forcellsexposedto50␮MU(VI),thePiconcentration valueswerealwayssignificantlylowerthanintheotherconditions andshowedastrongdeclineduringtheincubation.Attheendof theexperiment,thePiconcentrationwas7-foldlower(1␮M)than inthecontrolsamples.

At25◦C(Fig.3B),anincreaseinPicontentwasobservedover thetime-courseoftheexperimentforallconditions,yetto differ-entextents.ThemaximumPireleaseoccurredforcellsexposedto 10␮MU(VI),withtheexceptionofthe24htimepoint,wherethe highestvaluewasmeasuredinthecontrolsample(43␮M).Inthe 50_␮MU(VI)condition,Piconcentrationsexhibitedvalues2.5to 4-foldlowerthanintheotherconditionsafter6hofexposure. 3.5. MicroscopicobservationsofU(VI)-exposedcells

TEM-EDXanalysisrevealedneedle-likestructuresinsidethe bacteriathat containU(Fig.4).Thesestructureswere detected exclusivelyinthesamplesincubatedat25◦CandexposedtoU(VI) foratleast6h,andtheirsizeincreasedasafunctionoftime.EDX analysisrevealedthat Uin thesestructuresco-localisedmainly withphosphorus(P)andcalcium(Ca).

3.6. AnalysisofUspeciationwithTRLFS

UspeciationwasanalyzedbyTRLFSonbacterialcellsexposed tobothU(VI)concentrationsat25◦Cfor0.5,6and24h. Informa-tionregardingpossibleUspeciescanbedeterminedbyanalyzing boththestaticemissionfluorescencespectrumbandpositionand thetime-resolvedfluorescencedecay.Theemittedsignalrecorded forthecellsexposedto10␮MU(VI)wastoolow.Thetwospectra recordedforcellsexposedto50␮MUafter6and24hofexposure werecharacterizedbyemissionmaximaat494,516and539nm (Fig.5).Twoadditionalimportantshoulders,characterizedby

emis-0 5 10 15 20 25 30 35 40 45 50 0.5 2 4 6 10 24 0.5 2 4 6 10 24 T ot al Pi in th e cell -free sup ern at ant (µM) Time (hours) Control 10 µM U 50µM U ** ** ** ** ** ** ** ** ** _{** ** **} B: 25°C A:4°C

Fig.3. Piconcentrationovertimeinthecell-freesupernatantduringcellexposureto0␮M(blackcolumn),10␮M(greycolumn)and50␮M(whitecolumn)U(VI)at4◦_C

Fig.4. TEMmicrographandEDXspectraofcellsexposedto50␮MU(VI)for24hat25◦_{C.Theintracellularbackground(A),Caandphosphategranules(B)andneedle-like}

structurescontainingU,CaKandP(C)areindicatedbyarrows.

sionsat503and523nm,wererecordedforcellsexposedfor24h. Spectraofcellsexposedfor0.5hexhibitedalowsignal-to-noise ratiowithpoorlyresolvedemissionpeaks,althoughtheyfollowed thesametrendobservedintheotherspectra.

Theemissionmaximaofexposedcellscoincidewiththoseof ourU-phosphatereferencecomplex(Fig.5:spectrum(A),494,516, 539,and565nm).However,theconstructionofaconvoluted spec-trumbasedonlyonthepresenceoftheU-phosphatecomplexand backgroundnoiseindicatesthatasecondcompoundwith emis-sionmaximacorrespondingtothoseofautunite(Fig.5:spectrum (B),503,523,547and573nm)mustbeconsidered.The percent-ageofautuniteneededtofitthespectraofexposedcellsincreased withexposuretimetoU(VI).After6hofexposuretoU,the spec-tracouldbeexplainedby6%autunite,34%U-phosphatecomplexes and61%backgroundnoise,whereasafter24hofexposure,these valueswere15,45and40%,respectively.

Luminescencelifetime analysesofthe exposed cells didnot revealanydifferencesaccordingtoexposuretime.Alldatafitwell witha bi-exponential decay curve, confirming the presence of twoluminescentspecies,withfirstlifetimevalues of29.9±5.4, 34.4±4.3and32.6±3.1␮sandsecondlifetimevaluesof2.0±0.7, 3.5±1.8 and 2.3±0.4␮s after 0.5, 6 and 24h of exposure to U(VI),respectively.Theselifetimevalueswereslightlyshorterthan thoseforautunite(105and18_␮s)andtheU-phosphatecomplex (94.6±3.6and3.1±0.9␮s),probablyduetothepresenceof fluo-rescencequenchersinbacteria[20].

3.7. AnalysisofUcoordinationbyATR-FTIR

WefurtherinvestigatedthefunctionalgroupsinvolvedinU(VI) complexationbyrecordingATR-FTIRspectraofnon-exposed bac-teria(control)aswellasbacteriaexposedto10and50␮MUfor 0.5,6and24hatbothtemperatures.Forthe10␮Mcondition,the signalwastoolow.TheresultsinFig.6arepresentedasdifference spectra,i.e.absorptionspectraofthebacteriaexposedto50␮M minusabsorptionspectraofthenon-exposedbacteria.Thespectra ofcontrolandexposedcellswerecalibratedusingthemain absorp-tionbandsofproteins(referredtoasAmideIandAmideIIbands)to calculatethedifferencespectra.Inthesespectra,positiveand neg-ativepeaksrepresentvibrationalmodesoffunctionalgroupsthat undergochanges duetoUcomplexation.Thisapproach primar-ilyrevealedatemperaturedependenceofthedifferencespectra, aswellasspectralchangesinresponsetoexposuretimeat25◦C. At4◦C (Fig.6A),spectraexhibited onlya slight evolution with exposuretime.Theseresultsarecharacterizedbyaclearpositive bandat916cm−1,andbynegativebandsat1080,1215,1400cm−1 and 1570cm−1. The positive band at 916cm−1 is assigned to the asymmetricstretching vibration of the uranyl ion and has beenobservedforuranylcomplexesinvolvingfunctionalgroups

450 500 550 600

Re

lativ

e i

nte

n

sity

(a.

u

.)

Wav

elength (n

m)

Fig.5.TRLFSspectraofcellsexposedto50␮MU(VI)for0.5,6and24hat25◦_C,as

wellasuranylphosphatereferencecompounds(A),autunite(B),andtheconvoluted spectraofA+B(C).Thepeakmarkedwithastar(*_{)representstheharmonicsignal}

at532nm.

belongingtovarious cellularcomponents[21,22].The bandsat 1570and1400cm−1areassignedtotheasymmetricand symmet-ricvibrationsofcarboxylategroups(asands(COO−)),suggesting that uranyliscoordinatedby carboxylategroupsin thecellsat 4◦C. In addition, IR bands between 1300 and 1000cm−1 cor-respond to bands of phosphoryl or phosphate groups [22,23], showingthatthesegroupsmayalsobeinvolvedinuraniumbinding at4◦C.

290 N.Theodorakopoulosetal./JournalofHazardousMaterials285(2015)285–293 B (25°C) A(4°C) Wavenumber cm-1 -0.012 -0.007 -0.002 0.003 0.008 0.013 800 950 1100 1250 1400 1550 1700 U50 T24-25°C U50 T6-25°C U50 T0.5-25°C -0.012 -0.007 -0.002 0.003 0.008 0.013 800 950 1100 1250 1400 1550 1700 U50 T24-4°C U50 T6-4°C U50 T0.5-4°C υasPO2 -υsCOO -υasUO₂2+ υasCOO -υasPO A bso rb an ce (a . u.)

Fig.6.FTIRdifferentialspectrumbetweencontrolsamplesanduranium-exposedbacteria(10and50␮MU)at0.5,6and24h,assayedat4◦_C(A)and25◦_C(B).

Fig.7.ProposedmechanismofinteractionsbetweenUandtheMicrobacteriumsp.A9strain.Grey,blackandwhitearrowsindicateapassive,active,andanactiveorpassive mechanism,respectively.Proposedscheme:(1)uraniumentersthecellbyadirectorindirectway;(2)uraniumreleaseisprocessedeitherundertheU(VI)formor(2_)the

U-Piform;(3)separately,enzymaticpathwaysareinduced,causingboundbreakageofpolyphosphategranules;(4)inorganicphosphateionsarereleased,complexing(5) extracellularlyand(6)intracellularlywithbioavailableuranium;(7)autuniteisformedintracellularywhenasuitablyhighU-Picomplexconcentrationisreached.

assignedtocarboxylategroupsat1400and1570cm−1decreased significantly for spectra recorded after 6 and 24h exposureto U(VI).Furthermore,bandsassignedtophosphateand/or phospho-rylgroups at1087,1220and 1274cm−1 significantlyincreased, especiallyafter24hexposure.Thesechangeshighlighttheroleof phosphategroupsinthebindingofuraniumat25◦Cduringlonger timescales.

4. Discussion

4.1. ResistanceofMicrobacteriumtoU(VI)

UspeciationaswellasUbioavailabilityisinfluencedbypH,the redox-potential,andthepresenceofvariouscomplexingagents. Inourstudy,cells wereexposedin amedium totallydevoidof complexingagentsinordertooptimizeU(VI)bioavailability.We demonstratedthatexposureto10␮MU(VI)inthismediumdid notimpactcellviability.Upto30%ofMicrobacteriumsp.A9cells remainedviabledespite24hexposureto50_␮MU(VI).Several envi-ronmentalisolatesaffiliatedwithMicrobacteriumarereportedto beresistanttonumerousmetal(loid)ssuchas,chromium,arsenic, cobalt,nickel,selenium,uranium,andzinc[12,14,24,25].

4.2. Phosphatereleaseanditsdetoxificationrole

Polyphosphategranulesaresynthesizedbymanybacteriafor bothenergyandphosphatestorage,andcanbeusedindefense mechanismsagainstmetalexposurewhenneeded(reviewedin

[26]).

Here,weobservedthatPireleasewastemperature-dependent. At4◦C,afteraninitialreleaseuponcellexposuretotheacidicsaline solution,thePiconcentrationremainedstable.Incontrast,Pi con-centrationincreasedsignificantlyat25◦C,demonstratingthatPi releaseisanactiveprocess.TheefficiencyofPireleasewasnot onlytemperature-dependent,but alsovariedaccordingtoU(VI) concentration.Pireleasewasslightlyinducedwithinthefirst10h whencellswereexposedto10␮MU(VI).Incontrast,Pireleasewas muchmorelimitedat50␮MU(VI).Thisfeaturecouldbea conse-quenceofthetoxiceffectofU(VI)at50␮M,whichhadanoticeable effectoncellviability.Interestingly,Pireleasebybacteriaexposed to10␮M U(VI) occurredwiththe effluxof U observed 30min afteruranylexposure,suggestingthaturanylmaybereleasedas uranyl-phosphatecomplexes(U-Pi).However,itisnotpossibleto determineifthesetworeleasesystemsareconnected.Itis there-foreverylikelythatPirelease representsanindirectprotection mechanismagainsturanium,eitherbypromotingU-Pieffluxorby limitingtheentranceofU(onceithasbeenextracellularychelated byPi).

Theseresultsareinagreementwithotherstudies demonstrat-ingthatseveralbacteria speciescanprecipitate uraniuminthe extracellular compartment viaa phosphate release mechanism mediatedbynon-specificacidphosphatases,thuspreventingor limitingitsentranceintothecytoplasm(reviewedin[5,27–30]). However,noUprecipitateswereobservedbyTEMintheexposure mediaoratthebacterialsurfaceunderourconditions.Therefore, itismoreprobablethattheU-Picomplexesremainsolubleinthe exposuremedia.

Finally,thereleaseofPiappearedtobeaffectedbythe intracel-lularaccumulationofU.Indeed,24-hoursexposuretoeither10or 50␮MU(VI),conditionsinwhichUaccumulationwasobserved byTEM(seebelow),inhibited therelease of Piascomparedto controls.Thismaybeduetothetotalorpartialinhibitionof phos-phataseactivitybytheuranylion[31,32].Itisalsopossiblethat phosphate was sequestered in these cells, as part of uranium-phosphateprecipitates[33].Rayetal.[34]alsoobservedthatU minimizesthefinalaverageconcentrationsofPiinsolution,ina bacterialcultureexposedto100␮MU(VI).Wecanalso hypothe-sizethatthetoxicityofUledtotheimpairmentofcellularfunction, whichwasconsequentlyoverwhelmedbythismassiveentrance ofU.

4.3. Uadsorption,accumulationandreleasemechanisms

Inthisstudy,weevidencedtwodifferentmechanismsinvolved intheremovalofU(VI)bytheMicrobacteriumsp.A9strain.Afirst mechanismoccurredwithin30minatbothtemperatures,andwas

consideredtobemetabolism-independent.Suchamechanismhas alreadybeenreportedforalargenumberofbacterialstrains.Ithas beeninterpretedasadsorptionofU(VI)tothebacterialsurfacedue toelectrostaticinteractionsbetweenuranylandhighaffinitysites ofthecellenvelope[8,35].Thistypeofinteractionisinlinewiththe FTIRresultsfortheMicrobacteriumsp.A9strain,whichshowthat after30min,Uwascoordinatedmainlytothecarboxyland phos-phorylgroupsatbothtemperatures.Inadditiontothisadsorption atthecellsurface,arapidmetabolism-independentaccumulation ofUinsidethecells[5]resultingfrommembranepermeability[7]

cannotbeexcluded,althoughwecouldnotdetectUinthecellsby TEMatearlyexposuretime.

At25◦C,immediatelyafterUadsorption,thecellsexposedto 10␮MU(VI)releasedafractionofthemetal,possiblyasa protec-tivemechanism.Sincethisreleasewasobservedonlyat25◦C,we proposethatitinvolvesanactivemechanism,suchasanexport system.Metalexportsystemsarewidespreadandallowbacteria tosurviveinmetal-contaminatedenvironments[36].The involve-mentofaneffluxsysteminthedetoxificationofuraniumhasnot yetbeendescribed,althoughanup-regulationofgenesencoding metaleffluxpumpshasalreadybeenreportedfor Desulfotomacu-lumreducensexposedtoU(VI)inanaerobicconditions[37].Inour study,Ureleaseintothemediumwasonlyobservableatthe low-estexposureconcentration.Thissuggeststhatexposuretohigher U(VI)concentrationscouldeitherinhibitthismechanismorthat massiveentranceofUinthecellmasksthisefflux.

Finally, a temperature-dependent removal of U(VI) was observedatlongertimescalesforthetwoU(VI)exposure concen-trations,withahigherremovalefficiencyobservedat25◦Cvs.4◦C. Furthermore,theFTIRanalysisshowedacleardistinctioninU(VI) coordinationbetweenthetwotemperatures:at25◦Cphosphate groupsaremoreandmoreinvolvedasexposuretime increased whileat4◦Cthesystemremainedstable.DuringthisstepofU(VI) removal,theformationofintracellularneedle-likestructureswas observablebyTEM. Thesestructureswereidentifiedasautunite bytheTRLFSanalysis,whichwascoherentwiththeEDXanalysis showingthatUwasco-localisedwithPandCa.TheTRLFSdata demonstrated alsothepresenceofa secondU-specieidentified asU(VI)-Picomplex.Wecannotexcludethatthisspecies corre-spondstoU(VI)-ATP sincetheybothexhibit asimilarspectrum

promisingstrategyfortreatmentandclean-upofU-contaminated sub-surfaces.

5. Conclusionsandperspectives

OurdatarevealthattheMicrobacteriumsp.A9strain,isolated fromtrench T22 at the Chernobyl waste repository, exhibits a highcapacityofsurvivalandresistancetoU(VI)basedonvarious detoxificationmechanisms.Comparingdataobtainedattwo tem-peraturesandusingexperimentalconditionswhereupto100%cell viabilitywasmaintainedhaveenabledthefirst-ever discrimina-tionbetweenactiveandpassivemechanismsofU(VI)removaland theidentificationofanactivereleaseprocess.Threesuch mech-anismshavebeenidentified,allofwhichinvolvephosphate.The firstonemediatesphosphatereleaseintheexposuremedia,which maycomplexwithuraniumtopreventitsfurtherentranceintothe cells.ThesecondonemediatestheeffluxofUfromthebacteria andisonlyvisiblewhenbacteriaareexposedtolowU concentra-tion.TheeffluxofUaccompaniesthereleaseofphosphate,which maysuggestthatthesemechanismsarelinkedtogether.Thethird detoxificationmechanismisinvolvedinprecipitatingaccumulated uraniumintracellularlyasautunite.Basedontheseresults,we pro-poseamodelofU-bacteriainteractions(Fig.7).

ThisstudyhighlightsthepotentialuseoftheMicrobacterium genusforbioremediation.Tovalidatetheuseofthesebacteriainthe clean-upofU-contaminatedsub-surfaces,futureresearchshould bedirectedatinvestigatingthebehaviorofthesebacteriainsitu, anddeterminingtheimmobilizationcapacityofUinsoils.Although wehaveproposedaninteractionmechanismbetweenUand bacte-riathatincorporatesourcombinedresults,thesemechanismsare farfrombeingcompletelyunderstood.Therefore,elucidatingthese effluxanddetoxificationmechanismsshouldbemadearesearch priorityinthisfield.

Acknowledgements

ThisworkisfinanciallysupportedbytheCNRSandIRSNthrough theGNRTRASSE.WethankSachaPasquierforhistechnical assis-tance during TRLFS analysis. Nicolas Theodorakopoulos is the recipient ofa PhD Grant co-fundedby the IRSN and thePACA regional council. We thank BrandonLoveall of Improvence for Englishproofreadingofthemanuscript.

References

[1]W.DeVos,T.Tarvainen,GeochemicalAtlasofEurope.Part2-Interpretationof GeochemicalMaps,Additionaltables,FiguresMapsandRelatedPublications (php)(2006)http://www.gsf.fi/publ/foregsatlas/part2

[2]V.A.Kashparov,N.Ahamdach,S.I.Zvarich,V.I.Yoschenko,I.M.Maloshtan,L. Dewiere,KineticsofdissolutionofChernobylfuelparticlesinsoilinnatural conditions,J.Environ.Radioact.72(2004)335–353.

[3]V.Chapon,L.Piette,M.-H.Vesvres,F.Coppin,C.LeMarrec,R.Christen,N. Theodorakopoulos,L.Février,S.Levchuk,A.Martin-Garin,C.Berthomieu,C. Sergeant,MicrobialdiversityincontaminatedsoilsalongtheT22trenchof theChernobylexperimentalplatform,Appl.Geochem.27(2012)1375–1383.

[4]M.Gavrilescu,L.V.Pavel,I.Cretescu,Characterizationandremediationofsoils contaminatedwithuranium,J.Hazard.Mater.163(2009)475–510.

[5]M.L.Merroun,S.Selenska-Pobell,Bacterialinteractionswithuranium:an environmentalperspective,J.Contam.Hydrol.102(2008)285–295.

[6]L.Newsome,K.Morris,J.R.Lloyd,Thebiogeochemistryandbioremediationof uraniumandotherpriorityradionuclides,Chem.Geol.363(2014)164–184.

[7]Y.Suzuki,J.F.Banfield,Resistancetoandaccumulationofuraniumbybacteria fromauranium-contaminatedsite,Geomicrobiol.J.21(2004)113–121.

[8]S.Choudhary,P.Sar,Uraniumbiomineralizationbyametalresistant Pseudomonasaeruginosastrainisolatedfromcontaminatedminewaste,J. Hazard.Mater.186(2011)336–343.

[9]M.Merroun,M.Nedelkova,A.Rossberg,C.Hennig,S.Selenska-Pobell, Interactionmechanismsofbacterialstrainsisolatedfromextremehabitats withuranium,Radiochim.Acta94(2006)723–729.

[10]R.Kumar,M.Nongkhlaw,C.Acharya,S.R.Joshi,Uranium(U)-tolerant bacterialdiversityfromUoredepositofdomiasiatinnorth-eastindiaandits prospectiveutilisationinbioremediation,MicrobesEnviron.28(2013)33–41.

[11]L.Mondani,K.Benzerara,M.Carrière,R.Christen,Y.Mamindy-Pajany,L. Février,N.Marmier,W.Achouak,P.Nardoux,C.Berthomieu,V.Chapon, Influenceofuraniumonbacterialcommunities:acomparisonofnatural uranium-richsoilswithcontrols,PLoSONE6(2011)e25771.

[12]M.Nedelkova,M.L.Merroun,A.Rossberg,C.Hennig,S.Selenska-Pobell, Microbacteriumisolatesfromthevicinityofaradioactivewastedepository andtheirinteractionswithuranium,FEMSMicrobiol.Ecol.59(2007) 694–705.

[13]T.N.Nazina,E.A.Luk’yanova,E.V.Zakharova,L.I.Konstantinova,S.N. Kalmykov,A.B.Poltaraus,A.A.Zubkove,Microorganismsinadisposalsitefor liquidradioactivewastesandtheirinfluenceonradionuclides,Geomicrobiol. J.27(2010)473–486.

[14]P.Kaushik,N.Rawat,M.Mathur,P.Raghuvanshi,P.Bhatnagar,H.Swarnkar,S. Flora,Arsenichyper-toleranceinfourMicrobacteriumspeciesisolatedfrom soilcontaminatedwithtextileeffluent,Toxicol.Int.19(2012)

188–194.

[15]D.Bugai,V.Kashparov,L.Dewiére,Y.Khomutinin,S.Levchuk,V.Yoschenko, Characterizationofsubsurfacegeometryandradioactivitydistributioninthe trenchcontainingChernobylclean-upwastes,Environ.Geol.47(2005) 869–881.

[16]F.Osmond,Dosagecolorimétriqueduphosphore,Bull.Soc.chim.(Paris)47 (1887)745–748.

[17]T.Vercouter,P.Vitorge,B.Amekraz,C.Moulin,Stoichiometriesand thermodynamicstabilitiesforaqueoussulfatecomplexesofU(VI),Inorg. Chem.47(2008)2180–2189.

[18]RCoreTeam,R:ALanguageandEnvironmentforStatisticalComputing,R FoundationforStatisticalComputing,Vienna,Austria,2013

http://www.R-project.org

[19]S.J.Markich,Uraniumspeciationandbioavailabilityinaquaticsystems:an overview,Sci.WorldJ.2(2002)707–729.

[20]T.Reitz,M.L.Merroun,A.Rossberg,R.Steudtner,S.Selenska-Pobell, BioaccumulationofU(VI)bySulfolobusacidocaldariusundermoderateacidic conditions,Radiochim.Acta99(2011)543–553.

[21]K.Popa,A.Cecal,G.Drochioiu,A.Pui,D.Humelnicu,Saccharomycescerevisiae asuraniumbioaccumulatingmaterial:theinfluenceofcontacttime,pHand anionnature,Nukleonika48(2003)121–125.

[22]A.Barkleit,H.Foerstendorf,B.Li,A.Rossberg,H.Moll,G.Bernhard, Coordinationofuranium(VI)withfunctionalgroupsofbacterial

lipopolysaccharidestudiedbyEXAFSandFT-IRspectroscopy,DaltonTrans.40 (2011)9868–9876.

[23]R.Pardoux,S.Sauge-Merle,D.Lemaire,P.Delangle,L.Guilloreau,J.-M. Adriano,C.Berthomieu,Modulatinguraniumbindingaffinityinengineered calmodulinEF-handpeptides:effectofphosphorylation,PLoSONE7(2012) e41922.

[24]A.C.Humphries,K.P.Nott,L.D.Hall,L.E.Macaskie,ReductionofCr(VI)by immobilizedcellsofDesulfovibriovulgarisNCIMB8303andMicrobacterium sp.NCIMB13,776,Biotechnol.Bioeng.90(2005)589–596.

[25]A.Achour-Rokbani,A.Cordi,P.Poupin,P.Bauda,P.Billard,Characterizationof thearsgeneclusterfromextremelyarsenic-resistantMicrobacteriumsp. strainA33,Appl.Environ.Microbiol.76(2010)948–955.

[26]L.Achbergerová,J.Nahálka,Polyphosphate–anancientenergysourceand activemetabolicregulator,Microb.CellFact.4(2011)10–63.

[27]V.Sivaswamy,M.I.Boyanov,B.M.Peyton,S.Viamajala,R.Gerlach,W.A.Apel, R.K.Sani,A.Dohnalkova,K.M.Kemner,T.Borch,Multiplemechanismsof uraniumimmobilizationbyCellulomonassp.strainES6,Biotechnol.Bioeng. 108(2011)264–276.

[28]M.J.Beazley,R.J.Martinez,P.A.Sobecky,S.M.Webb,M.Taillefert,Uranium biomineralizationasaresultofbacterialphosphataseactivity:insightsfrom bacterialisolatesfromacontaminatedsubsurface,Environ.Sci.Technol.41 (2007)5701–5707.

[29]R.J.Martinez,M.J.Beazley,M.Taillefert,A.K.Arakaki,J.Skolnick,P.A.Sobecky, Aerobicuranium(VI)bioprecipitationbymetal-resistantbacteriaisolated fromradionuclide-andmetal-contaminatedsubsurfacesoils,Environ. Microbiol.9(2007)3122–3133.

[30]M.L.Merroun,M.Nedelkova,J.J.Ojeda,T.Reitz,M.L.Fernández,J.M.Arias,M. Romero-González,S.Selenska-Pobell,Bio-precipitationofuraniumbytwo bacterialisolatesrecoveredfromextremeenvironmentsasestimatedby potentiometrictitration,TEMandX-rayabsorptionspectroscopicanalyses,J. Hazard.Mater.197(2011)1–10.

[31]J.A.Finlay,V.J.Allan,A.Conner,M.E.Callow,G.Basnakova,L.E.Macaskie, Phosphatereleaseandheavymetalaccumulationbybiofilm-immobilizedand chemically-coupledcellsofaCitrobactersp.pre-grownincontinuousculture, Biotechnol.Bioeng.63(1999)87–97.

[32]L.Lütke,H.Moll,G.Bernhard,Insightsintotheuranium(VI)speciationwith Pseudomonasfluorescensonamolecularlevel,DaltonTrans.41(2012) 13370–13378.

[33]J.Misson,P.Henner,M.Morello,M.Floriani,T.D.Wu,J.-L.Guerquin-Kern,L. Février,UseofphosphatetoavoiduraniumtoxicityinArabidopsisthaliana leadstoalterationsofmorphologicalandphysiologicalresponsesregulated byphosphateavailability,Environ.Exp.Bot.67(2009)353–362.

[34]A.E.Ray,J.R.Bargar,V.Sivaswamy,A.C.Dohnalkova,Y.Fujita,B.M.Peytond, T.S.Magnusona,Evidenceformultiplemodesofuraniumimmobilizationby ananaerobicbacterium,Geochim.Cosmochim.Acta75(2011)

2684–2695.

[36]D.H.Nies,Efflux-mediatedheavymetalresistanceinprokaryotes,FEMS Microbiol.Rev.27(2003)313–339.

[37]P.Junier,E.D.Vecchia,R.Bernier-Latmani,TheresponseofDesulfotomaculum reducensMI-1toU(VI)exposure:atranscriptomicstudy,Geomicrobiol.J.28 (2011)483–496.

[38]M.L.Merroun,G.Geipel,R.Nicolai,K.H.Heise,S.Selenska-Pobell, Complexationofuranium(VI)bythreeeco-typesofAcidithiobacillus ferrooxidansstudiedusingtime-resolvedlaser-inducedfluorescence spectroscopyandinfraredspectroscopy,BioMetals16(2003)331–339.

[39]N.Renninger,R.Knopp,H.Nitsche,D.S.Clark,J.D.Keasling,Uranyl precipitationbyPseudomonasaeruginosaviacontrolledpolyphosphate metabolism,Appl.Environ.Microbiol.70(2004)7404–7412.

[40]F.Jroundi,M.L.Merroun,J.M.Arias,A.Rossberg,S.Selenska-Pobell,M.T. González-Mu ˜noza,Spectroscopicandmicroscopiccharacterizationof

uraniumbiomineralizationinMyxococcusxanthus,Geomicrobiol.J.24(2007) 441–449.

[41]E.Krawczyk-Bärsch,H.Lünsdorf,K.Pedersen,T.Arnold,F.Bok,R.Steudtner, A.Lehtinen,V.Brendler,Immobilizationofuraniuminbiofilm

microorganismsexposedtogroundwaterseepsovergraniticrocktunnel wallsinOlkiluoto,Finland,Geochim.Cosmochim.Acta96(2012)94–104.

[42]S.Frelon,S.Mounicou,R.Lobinski,R.Gilbin,O.Simon,Subcellular fractionationandchemicalspeciationofuraniumtoelucidateitsfateingills andhepatopancreasofcrayfishProcambarusclarkia,Chemosphere91(2013) 481–490.

[43]D.M.Wellman,J.P.Icenhower,A.P.Gamerdinger,S.W.Forrester,EffectsofpH, temperature,andaqueousorganicmaterialonthedissolutionkineticsof meta-autuniteminerals,(Na,Ca)2-1[(UO2)(PO4)]2·3H2O,Am.Mineral.91