Adsorption of DNA on biomimetic apatites: Toward the understanding of the role of bone and tooth mineral on the preservation of ancient DNA

Any correspondence concerning this service should be sent to the repository administrator:

staff-oatao@inp-toulouse.fr

Identification number: DOI : 10.1016/j.apsusc.2013.12.063

Official URL:

http://dx.doi.org/10.1016/j.apsusc.2013.12.063

This is an author-deposited version published in:

http://oatao.univ-toulouse.fr/

Eprints ID: 12054

To cite this version:

Grunenwald, Anne and Keyser, Christine and Sautereau, Anne-Marie and

Crubézy, Eric and Ludes, Bertrand and Drouet, Christophe Adsorption of DNA

on biomimetic apatites: Toward the understanding of the role of bone and tooth

mineral on the preservation of ancient DNA. (2014) Applied Surface Science,

vol. 292 . pp. 867-875. ISSN 0169-4332

O

pen

A

rchive

T

oulouse

A

rchive

O

uverte (

OATAO

)

OATAO is an open access repository that collects the work of Toulouse researchers and

makes it freely available over the web where possible.

Adsorption

of

DNA

on

biomimetic

apatites:

Toward

the

understanding

of

the

role

of

bone

and

tooth

mineral

on

the

preservation

of

ancient

DNA

A.

Grunenwald

a,b

_,

_C.

_Keyser

b

_,

_A.M.

_Sautereau

a

_,

_E.

_Crubézy

c

_,

_B.

_Ludes

b

_,

_C.

_Drouet

a,∗

a_{CIRIMATCarnotInstitute–Phosphates,Pharmacotechnics,Biomaterials,UniversityofToulouse,CNRS/INPT/UPS,ENSIACET,4alléeEmileMonso,31030}

ToulouseCedex4,France

b_{InstituteofLegalMedicine,AMISLaboratory,CNRSUMR5288,UniversityofStrasbourg,11rueHumann,67085StrasbourgCedex,France}

c_{MolecularAnthropologyandImageSynthesisLaboratory(AMIS),CNRSUMR5288,UniversityofToulouse,37alléeJulesGuesde,31000Toulouse,France}

Keywords: Nanocrystallineapatite Hydroxyapatite AncientDNA Polyelectrolyteadsorption Temkinisotherm Elovich

a

b

s

t

r

a

c

t

InordertoshedsomelightonDNApreservationovertimeinskeletalremainsfromaphysicochemical viewpoint,adsorptionanddesorptionofDNAonawellcharacterizedsyntheticapatitemimickingbone anddentinbiomineralswerestudied.Batchadsorptionexperimentshavebeencarriedouttodetermine theeffectofcontacttime(kinetics),DNAconcentration(isotherms)andenvironmentallyrelevantfactors suchastemperature,ionicstrengthandpHontheadsorptionbehavior.Theanalogyofthenanocrystalline carbonatedapatiteusedinthisworkwithbiologicalapatitewasfirstdemonstratedbyXRD,FTIR,and chemicalanalyses.Then,DNAadsorptionkineticswasfittedwiththepseudo-firstorder,pseudo-second order,Elovich,Ritchieanddoubleexponentialmodels.ThebestresultswereachievedwiththeElovich kineticmodel.TheadsorptionisothermsofpartiallyshearedcalfthymusDNAconformed satisfacto-rilytoTemkin’sequationwhichisoftenusedtodescribeheterogeneousadsorptionbehaviorinvolving polyelectrolytes.Forthefirsttime,theirreversibilityofDNAadsorptiontowarddilutionandsignificant phosphate-promotedDNAdesorptionwereevidenced,suggestingthataconcomitantionexchange pro-cessbetweenphosphateanionicgroupsofDNAbackboneandlabilenon-apatitichydrogenphosphate ionspotentiallyreleasedfromthehydratedlayerofapatitecrystals.Thisworkshouldprovehelpfulfor abetterunderstandingofdiageneticprocessesrelatedtoDNApreservationincalcifiedtissues.

1. Introduction

Boneand toothremains oftenrepresentthe only– butalso thebest–biologicalmaterialsavailablefordeoxyribonucleicacids (DNA)typinginanthropologyandmayadditionallybeexploited in forensicsciences [1,2]. Inancient samples,DNAarisingfrom degradedcellsisgenerallystronglyfragmented,withameansize around200basepairs, anditsamplification maybelimitedby thepresenceofsubstancespersistingafterpurification,and inhib-itingPCR[3].Althoughhardtissue samplesarecommonlyused asa sourceofancient DNAsequences,theprocessesof ancient DNApreservationinthesemineralizedtissueshavereceivedlittle systematicattention,bothregardingqualitativeandquantitative aspects.

∗_{Correspondingauthorat:CIRIMATCarnotInstitute,ENSIACET,4alléeEmile} Monso,31432ToulouseCedex4,France.Tel.:+33034323411;

fax:+33034323499.

E-mailaddress:christophe.drouet@ensiacet.fr(C.Drouet).

Calciumphosphateapatitesarethemaincomponentsofhuman boneandteeth.Thephysico-chemicalfeaturesofenamelhighly resemblethoseofstoichiometrichydroxyapatiteCa10(PO4)6(OH)2,

both in terms of crystallographic structure and composition. In contrast, bone and dentin minerals are composed of non-stoichiometricapatitenanocrystals.Severalstudieshave shown thepossibilitytoprepare,underclose-to-physiologicalconditions (temperature, pH), syntheticanalogs tobiological apatites that mimictheirphysico-chemicalcharacteristics[4–6].Forboth bio-logical specimens and biomimetic analogs, the surface of the nanocrystals was found toexhibit a structured but metastable non-apatitichydratedlayer[7–9],containinglabileions(e.g.Ca2+_,

HPO42−,CO32−...)leading to an exceptional surface reactivity,

either in terms of ionic exchanges or of molecular adsorption

[10–13]. Furthermore, the possibility to control, for synthetic biomimeticapatites,thematurationstateofthenanocrystalswas evidencedbymodifyingexperimentalconditionssuchas temper-ature,pHormaturationtimeinsolution[7,14,15].Thesefindings thenenablethepreparationofsamplesmimickingeithernewly formedormaturebonemineral.

The role of the mineral apatitic matrix, potentially acting asa physical and a chemical barrier against DNA deterioration (throughmicrobiologicalactivityandenvironmentalfactors)has beenwidelymentionedintheliteraturetoexplaintheexceptional preservationofDNAinhardtissues[16–18].Infact,eveninthe absenceofbacterialattack,thedecayof freeDNA moleculesin aqueoussolutionsiscompletelyachievedoverfewthousandyears, based onspontaneoushydrolysis [17].The presence of ancient DNAinsamplesagedover600000yearsappearsthenconsistent withthehypothesisofanadsorptionofDNAfragmentsonbone andtoothapatiticmatrix,thuslimitingthepossibilityof chemi-caldegradationandtheaccessibilityofbacteriaandtheirenzymes

[19].

Inanothercontext,theabilityofhydroxyapatitetoadsorbDNA hasbeenshownsince1957forchromatographicpurposes,using phosphatesolutionswithaconcentrationgradientasmobilephase

[20–22]. More recently,apatites and other calciumphosphates nanoparticleshavebeeninvestigatedasdrugcarrierstodeliver geneticmaterialstoeukaryoticcells[23,24].Okasakietal.[25]have alsoexaminedthecrystalgrowthofhydroxyapatiteinthe pres-enceofDNAandsuggestedanaffinitybindingphenomenonbased uponelectrostaticinteractionbetweennegativelycharged phos-phategroupsofthebackboneofDNAandcalciumionsfromthe apatitesurface.

However,inspiteofthesevariousapplications,thechemical interactionbetweenapatitesandDNAisgenerallyassumedapriori, withoutanin-depthunderstandingofthenatureofthebinding mechanisminvolved;moreoververyfewquantitativedatawere reportedto-dateonapatite/DNAinteractions.

The adsorptionof several biological (macro)molecules other than nucleic acids (e.g. bovine serum albumin and other pro-teins,antibiotics,aminoacids...)onsyntheticbiomimeticapatites has on the contraryalready been the object of various works

[11,12,26–28].Interestingly,adsorptionprocessesonbiomimetic apatiteswasfoundtoinvolveinsomecasesmorecomplex mech-anismsthansolelybasedonthedepositionofthemoleculeonthe surfaceofthesolidphase:insomeinstancesindeed,thereleaseof surfaceionswasfoundtooccursimultaneouslytotheadsorption process[11,29,30].Moreover,considerablevariationsofadsorption parametersareboundtobeobserveddependingonthe physico-chemicalcharacteristicsoftheapatitecrystals.

Basedontheseconsiderations,wereportinthiscontributiona firstphysico-chemicalinsightontheinteractionbetweenDNAand abiomimeticapatitemimickingboneanddentinbiominerals,with thegoaltoexplainthediageneticpersistenceofDNAextractedfrom skeletalremains.Thisexperimentalmodelwasaimedat determin-ingadsorptionand desorptionfeatures (usingcalf thymusDNA asmodelDNA),andatcomparingtheobtaineddatawithreports dealingwithotherbiologicalmolecules.

2. Materialsandmethods

2.1. Synthesisofapatiteanalogoustodentinandbonemineral

Abiomimeticcarbonated apatitesample,referred toas hac-1w,wassynthesizedbydoubledecompositionaccordingtothe proceduredescribed byRey etal. [8].Briefly, a calciumnitrate Ca(NO3)2·4H₂Osolution(52.2gin750mlofdeionizedwater)was poured at room temperature (20◦_{C) into a solution of}

ammo-niumhydrogenophosphate(NH4)2HPO4andsodiumbicarbonate

NaHCO3(90gofeachin1500mlofdeionizedwater).Theexcess

ofphosphateinthesecondsolutionwasusedtobufferthepHof thesolutionataclose-to-physiologicalvalue,namely7.2. Matura-tioninsolutionhasthenbeencarriedoutfor7days(1week).The pHofthemediumwasfoundtoremainconstantduringthewhole

maturationprocess.Thesuspensionwasthenfilteredona Büch-nerfunnel,thoroughlywashedwithdeionizedwater,freeze-dried andstoredinafreezer(−18◦_{C)soastoavoidanyalterationofthe}

apatitenanocrystals.

Theapatitepowderwasthensieved,andthesizefractioninthe rangeof100–200mmwasusedforallfurtherexperiments (adsorp-tion/desorption).

2.2. DNA

Adsorptionanddesorptionexperimentswereperformedwith DNAsolutionspreparedfromshearedcalfthymusDNAsolutions (Trevigen).Accordingtothemanufacturer,theconcentrationofthe mothercalfthymusDNAsolutionisprovidedat10mg/mlinTE buffer(10mMTris,1mMEDTA,pH8)withafragmentsizerange of500–1000bp.

2.3. Physico-chemicalcharacterization

Thechemicalcompositionoftheapatitecompoundsynthesized wasdeterminedfromtitrationscarriedoutbycomplexometryfor thedeterminationofthecalciumcontent,byspectrophotometry forthetotalphosphatecontent(determinationofthesumofPO43−

andHPO42−ions,usingthephospho-vanado-molybdenicmethod

[31])andbycoulometricmethod(UIC,Inc.CM5014coulometer withCM5130acidificationunit)forthecarbonatecontent.

Thecrystalstructureofthesampleswascheckedbypowder X-ray diffractionusing an Inel diffractometer CPS120 and the monochromaticCoKaradiation(Co=1.78892 ˚A).

Thespecificsurfacearea,Sw,wasdeterminedusingafivepoints

BET method(nitrogen adsorption) ona Tristar IIMicromeritics apparatus.

Fouriertransforminfrared(FTIR)analyseswereperformedona PerkinElmer1700spectrometerwitharesolutionof4cm−1_,using

theKBrpelletmethod.

TheamountofadsorbedDNA,atequilibrium,afteradsorption experimentswasdrawnbycomparisonoftheconcentrationsin DNAbeforeandafteradsorption,usingtheUVabsorptionofthe bandatca.260nmfollowedbyspectrophotometry(Thermo Scien-tificNanoDrop2000cspectrophotometer).

Theamountsofcalciumand(inorganic)phosphateionspresent inthesupernatantsafteradsorptionordesorptionweretitrated respectively byatomic absorption(ContrAA 300, AnalytikJena) andbyspectrophotometryusingthephospho-vanado-molybdenic methodasmentioned above.Priortotheseanalyses, the resid-ualDNAmoleculespresentinthesupernatantswereremovedby pretreatmentinacidicmediumandcentrifugation.

2.4. Adsorptionanddesorptionexperiments

Forallexperiments,theamountofapatitepowderandvolume ofDNAsolution(inglassflasks)werekeptconstantatrespectively 10mgand7.5ml.DNAsolutionswerepreparedatincreasing con-centrationsrangingfrom0to500mgDNA/ml,indeionizedwater. Thisconcentrationrangehasbeenchosentoremaininthelinear zoneofBeerLambertLawandavoidhighsolutionviscosity,which couldinterferewithadsorptionprocesses.

Adsorptionkineticswasfollowed,atroomtemperatureandpH 7,overaperiodof4weekswithaDNAsolutionofintermediate concentration(160–195mg/ml).

For the establishment of adsorption isotherms, experiments werecarried out withvariouspHvalues (byadding eitherHCl 0.1MorNaOH0.1M),ionicstrengths(byaddingKCl)and tem-peratures, under constant mild stirring. Thenthe mixture was centrifuged(20minat1850gandthesupernatantwasretrieved byfiltrationwithAcrodisc®_{syringefilterswithnylonmembrane}

Fig.1.XRDpatterns(2between20◦_and70◦_{)of(a)biomimeticcarbonatedapatite} sample,hac-1w(maturationof1week),ascomparedtobonespecimens:(b)rat(12 months),(c)human(man57y.o.).

(0.45mm,25mm).Eachpointwasdoneintriplicatetocheckthe reproducibilityofourexperimentalmodel.

Whenmentionedinthetext,theeventualdesorptionof pre-adsorbedDNAmoleculeswasexamined–inthesameconditionsof temperatureandcontacttimeasfortheadsorptionexperiments– byinvestigatingtheeffectofdilutionbyafactor5,where80vol.%of thesupernatantwerereplacedbydeionizedwater,withorwithout additionofKH2PO4atafinalconcentrationof18mM.

3. Resultsanddiscussion

3.1. Physico-chemicalcharacterizationofapatitesample

The apatite sample hac-1w synthesized in this work was characterized by way of X-ray diffraction (XRD) analysis and Fourier-transforminfrared(FTIR)spectroscopy.

TheXRDpatternobtained(Fig.1)wasfoundtobe characteris-ticofsingle-phasedapatite,asnosecondarycrystallizedphasewas detected.Acomparisonwiththepatternsobtainedonbone speci-mens,namelyhumanfemur(man,57y.o.)andratbone(12months old),asillustratedinFig.1,confirmedthebiomimeticcharacter ofthehac-1wsample.Ascanbenotedinallcases,thepatterns exhibitthefeaturesofapatitewitharatherlowcrystallinitystate characteristicofbiologicalapatites(exceptfortoothenamel).The application of Scherrer’sformula todiffraction lines(002) and (310),leadingrespectivelytoanestimateofthemeanlengthand widthoftheapatitecrystals,gavemeancrystallitedimensionsof 15.7nmlengthand4.9nmwidth.Thesevaluesareclearlyinthe nanometerscalethusconferringanadditionalelementinfavorof thebone-ordentin-likecharacterofthisapatitecompound.

TheapatitenatureofthesamplewasalsoconfirmedbyFTIR analysis(Fig.2).Severaldifferencescouldhowever,asexpected,be evidencedbycomparisontostoichiometrichydroxyapatite(HA).In particular,thepresenceofcarbonateabsorptionbandswaspointed outinthecaseofthesamplehac-1wintheranges1350–1550cm−1

(assignable to the 3(CO3) vibration mode) and 840–910cm−1

(attributableto2(CO3)).Incontrast,absorptionbandsat3572and

632cm−1 _{typicalofapatiticOH}− _{ionsarenotclearly detectable}

onthespectrumofhac-1w(seeforexampleinletofFig.2).This observationindicatesthatthisapatitesampleislargelydepleted inOH−_{ions.Thistypeofnonstoichiometryiscommonfor}

biolog-icalapatites(e.g.[32]), andistypicallyaccompaniedbycationic vacancies(lackofcalciumions)andthesubstitutionofPO43−ions

bydivalentHPO42− orCO32−.Beside thealreadyassessed

pres-enceofcarbonatespecies intheapatitelattice,theexistenceof

Fig.2. FTIRspectraforbiomimeticapatite(hac-1w)andforreferencestoichiometric hydroxyapatite(HA)showinglocalizationofapatitichydroxylions,phosphatesions (1–3)andcarbonatesions(2,3,hac-1wonly).Inset:spectraldecompositionof the4(PO4)vibrationdomainforhac-1wintherange400–800cm−1,revealingin particularthevibrationofnon-apatitichydrogenphosphatelabileions.

HPO42− ionscan indeed beestablishedhere onthebasis of IR

absorptionappearingasashouldertothe4(PO4)bands,around

535–550cm−1_{(seeinset,Fig.2).}

Thechemicalcompositionofthisapatitesamplewasthen inves-tigated.In particular,theCa/(P+C) molarratioofthesolid was determinedfromthechemicaltitrationofcalcium,carbonateand orthophosphateions,leadingtothevalue1.36.ThisCa/(P+C)ratio isnoticeablylowerthan1.67,whichischaracteristicof stoichio-metricHA,andthisfindingconfirmsthenonstoichiometryofthis compoundalreadysuggestedbyFTIRanalyses.

Alltheresultsreportedabovethereforeassessthebiomimetic characteroftheapatitesamplehac-1w,withbone-ordentin-like characteristics,thusmakingofitanadequatesubstrateforDNA adsorptionanalysesinourcontext.

Takingintoaccount theimportance ofthesurfaceofapatite exposedtotheDNAsolutioninadsorptionexperiments,thesample wassieved(between100and200mm)andthespecificsurfacearea oftheresultingpowderfractionwasfoundtobeofSw=180m2/g

asdeterminedbynitrogenadsorption(BETmeasurement).

3.2. Adsorptionkinetics

ThekineticsofadsorptionofcalfthymusDNAonthehac-1w biomimeticapatitesamplepreparedabovewasfirstfollowed,with theobjectivetodetermineexperimentalconditionsallowingus toreachthethermodynamicequilibrium.Tothisaim,adsorption experimentswerecarriedout(atroomtemperatureandpH7–7.5) starting from DNA solutions witha concentrationin the range 160–195mg/ml.

Fig.3a reports the experimentaldatapoints byplotting the amountof DNAadsorbed,denoted N_ads(given inmg DNA/g of apatite),asafunctionofcontacttimeandoveraperiodof4weeks. ThiskineticcurveshowsasteepincreaseofNadsduringthefirst

dayofcontactbetweenDNAmoleculesandtheapatitesubstrate, followedbyaprogressivestabilization(uptoabout110–120mg DNA/gapatiteforthisDNAconcentrationrange).Theanalysisofthe firstderivative(seeinsetonFig.3a)allowsonetoexamineinmore detailsthevariationofN_adsduringthefirst24hofapatite/DNA con-tact.Ascanbeseen,thevastmajorityofDNAmoleculesappeartobe adsorbedwithinthefirst2–3hofcontact–reachingcoverageofthe orderof70–80%(relativelytothemaximumequilibratedadsorbed amount).Thesefindingssuggestthattheadsorptionprocessisnot

Fig.3.KineticsofDNAadsorptionfollowedover3weeksatroomtemperature, neutralpH,onbiomimeticapatitehac-1w.Adsorptioniscompletewithin3days, withatrendfollowingElovich’sequation(dottedline):(a)Nads=f(t),inset:first derivativeoffirstdaydatapointsshowingthattheadsorptionprocessisalmost completewithin3h,(b)linearregressionusingElovich’sequation,Nads=f(Ln(t)). particularlyhindered,duringthisfirststageofadsorption,bysteric hindranceoftheDNA macromoleculesatthesurface ofapatite nanocrystals.However,beyondthisstage,such phenomenacan probablycontributetothemodificationofslopeobserved(upto reachingthermodynamicequilibrium)andtheratherslow evolu-tionbeforefinalstabilization.

Theshape ofthe adsorptionkinetic curveNads=f(t)appears

roughlywithalogarithmicshape,andthistypeofvariationwas indeed confirmed by the rather good linearity (R2_{=0.9463) of}

theNads=f(Ln(t))plotasshowninFig.3b.Suchabehavior isin

particularrepresentativeofkineticprocessesfollowingElovich’s mathematicalmodel[33]:

Nads= 1

bLn(t+t0)+ 1

bLn(ab) (1)

where“t0”(pre-Elovichperiod)isoftenfoundtobeclosetozero

(whichmayreasonablybeconsideredheretakingintoaccountthe linearityfoundinFig.3b).Thisequationderivesinfactfromthe descriptionoftheadsorptionratera=dNads/dt:

ra=dNads

dt =a·exp (−b·Nads(t)) (2) Interestingly,theElovichmodelwasfrequentlyconsideredto describethekineticsofadsorptionof(bio)moleculeson hetero-geneoussurfaces[34,35],especiallywhenthefinalstabilizationis ratherslowtooccur(suchasforexamplethecaseofthe adsorp-tionofhydrogenonmixedoxides[36]).Basedonourfindings,this

modelthusappearstosimulateratherwelltheadsorptionofDNA onbiomimeticapatite.Aratherslowstabilizationisindeedlikelyto happenheretakingintoaccountthelengthofthemacromolecules ofDNAandpotentialmodificationsofconformationupon adsorp-tion.

Otherkineticmodelshavehoweveralsobeentestedinthiswork inviewofevaluatingtheirpotentialpertinence.Thegeneralmodel proposedbyRitchie[37,38]wasinparticulartested,beingbased onanequationofthetype:

d

dt =kn(1−)

n ₍₃₎

where“”representsthecoverageofthesurfaceattime“t”,and “n”istheorderofthereaction.Outofthisgeneralequation,two modelsareindeedoftenencounteredinliteraturestudiesrelatedto kineticsofadsorption,namelythepseudo-firstorderand pseudo-secondorderkineticmodels:

Nads=_Ne·[1−exp(−k₁·_t)] (pseudo-firstorder) (4)

Nads=_Ne·

h

1−



1 1+k₂·t

i

(pseudo-secondorder) (5) Theapplicationofthesetwotypesofequationtoour experi-mentaldatahoweverdidnotallowustoreachasatisfactoryfit overthewholetimerange.Inallcases,thefitsobtainedpredicted astabilizationatsignificantlyshortercontacttimesascompared totheexperimentaldata.Theuseofadouble-exponentialmodel (DEM):

Nads=Ne−a₁·exp (−k₁·t)−a₂·exp (−k₂·t) (6) proposedintheliterature[39]fordescribingtherateof adsorp-tionoftwovariantsofnucleotidesonmontmorillonitealsofailed tosimultaneouslydescribeadequatelythedifferentstagesofthe kineticadsorptioncurveobserved inthepresent case(both for shortandlongercontacttimes).

TheabovefindingsthussuggestthattheElovichmodelremains thebestmodeltestedhere,fordescribingsatisfactorilythekinetics ofadsorptionofDNAmacromoleculesonbiomimeticapatite.

3.3. Adsorptionisothermanddesorptionstudy

Takingintoaccounttheprecedingkineticdata,thefollowing sub-sectionswillaimatestablishingDNAadsorptionisotherms. Onthebasisoftheabovekineticsstudy,anapatite/DNAcontact timeof3dayshasbeenselectedforallexperiments.Inafirststage, theisothermrelativetoso-called“reference”conditions–namely roomtemperature,neutralpHanddeionizedwatermedium–was determined.Insubsequentsteps,thepotentialeffectsofpH,ionic strengthandtemperaturewereexaminedforderivinggeneral ten-denciesontheadsorptionprocess.

3.3.1. Referenceadsorptionisotherm(roomtemperature,neutral pH,deionizedwatermedium)

Areferenceadsorptionisothermwasbuiltby measuringthe amountNadsofDNAadsorbed(eithergiveninmg/mgorinmg/m2)

onthehac-1wcarbonatednanocrystallineapatitesample(Fig.4) asafunctionoftheequilibriumconcentrationofDNAinthe super-natantCeq(mg/ml).Theseexperimentswerecarriedoutatroom

temperature(∼22◦_{C),neutralpH(initialvaluecloseto7.4)andin}

deionizedwater.

TheprofileofthisN_ads=f(Ceq)curveistypicalofanadsorption

isothermplottendingtowardamonolayer-likecoverage,witha steepincreaseoftheadsorbedamountoveranarrowrangeinCeq

followedbytheobtainmentofaplateau(aroundNm≈160mg/g). It maybenoted howeverthat experimentaldatapointsshow a

Fig.4.ThereferenceadsorptionisothermofcalfthymusDNA(0.75–4.5mg, ini-tialconcentrationrange40–600mg/ml)onbiomimeticcarbonatedapatitehac-1w (10mg)instandardconditions(roomtemperature,3daysofincubation,neutralpH) fitswithTemkinlogarithmiccurve(straightline)

rather highdispersion, which canberelatedtorelative hetero-geneityinDNAfragmentlengthsandapatiteparticlesize(despite preliminarysieving).Adispersionofphysico-chemical character-isticsamongbiologicalapatitesishoweveralsonaturallyobserved invivoduetovaryingstatusinmineralremodeling/agingstates. Inaddition,somevariabilityinDNAfragmentlengthalso presum-ablyoccursinpostmortemevents.Therefore,suchexperimental adsorptionisothermsareexpectedtoapproximatereasonablywell thevariabilityofactualdiageneticphenomena.

In ordertoshed somemorelight onthetype ofadsorption behaviorobservedherebetweenDNAmoleculesandbiomimetic apatitecrystals,severaladsorptionmodelsweretestedonthebasis ofmathematicallinearizationofthedata.TheLangmuirmodelas wellastwoderivatives,namelyFreundlichandTemkin,were suc-cessivelytestedbyplotting1/Nads=f(1/Ceq)forLangmuir(Eq.(7)),

Ln(Nads)=f(Ln(Ceq))forFreundlich(Eq.(8))andNads=f(Ln(Ceq))for

Temkin(Eq.(9)),referringtotheisothermequations: Nads=_Nm



KL·Ceq 1+K_L·C_eq



(7) Nads=K_F·C1/n_eq (8) Nads=_B·_Ln(A)+_B·_Ln(Ceq) (9) whereKL,Nm,KF,n,AandBarethecorrespondingconstants(at

fixedtemperature).

After fitting our data toeach of these three equations (see

Table1),onlya poorfit wasfoundwiththeLangmuirequation (correlationR2coefficient∼0.20),thusstronglysuggestingthatthe associatedhypothesesofthismodel(uniqueheatofadsorptionfor alladsorptionsites,nointeractionbetweenadsorbatemolecules) do not strictly apply here.In contrast,a significantly better fit (R2_{>0.5)wasfoundfortheothertwomodels,i.e.Freundlichand}

Table1

Adsorptionparametersandcorrelationcoefficientscalculatedaccordingtothe the-oreticalmodelstestedinthiswork.

Adsorption model Fittedadsorption parameters Correlation coeff. Langmuir Nm=114.9mg/g KL=3.0ml/mg R2=0.201 Freundlich KF=74.66mg/g(ml/mg)n n=7.17 R2=0.561 Temkin A=450.078ml/mg B=13.416 R2=0.671

Temkin,whichtakeintoaccountheterogeneousheatsofadsorption (thelatterbeingrelatedtothelocalaffinityofsurfacesitesfor spe-cificmolecularfunctionalgroup).Inparticular,thebestagreement wasfoundwiththeTemkinmodel(R2 _{coefficientcloseto0.67),}

whichtheoreticallysupposesaproportionalvariationof adsorp-tionenthalpiesasa functionof thecoverage.Theseresultscan beparalleledtoliteratureresultsdealingwiththeadsorptionof ionicadsorbentsonheterogeneoussurfaces[40–43],whereTemkin isothermwasalsooftenlinkedtoanElovich-typekineticmodel.

Fig.4pointsout asituationwhere themaximalDNA cover-agereachedonourapatiticsubstratehac-1wiscloseto160mg/g. Althoughitappearsappealingtocomparethisadsorption parame-terwithreportedvaluesmeasuredonapatitesforothermolecules, itshouldberemindedherethatcareshouldbetakenforsuch com-parisonssincesuchsubstratesprobablydonotexhibitthesame surface features, due todifferent synthesis histories. Indeed, it waspreviouslysuggestedthattheapatitematurationstage (lead-ingtomodifiedsurfaceandbulkcharacteristics[15])mayhavea directsignificantinfluenceonadsorptionprocesses[11,12,26,29]. Also, when DNA macromolecules areinvolved, it appears diffi-culttoexpresstheadsorbedamountsintermsofmoles(rather thangrams)duetothepolydispersityofthemolecularweightof DNA fragments(typicallyintherange 10–1000bpinourcase). Inthehypothesisofasamplemadeofanaverageof500bp,an approximatedvalueforDNAmolecularweightwouldbe330kDa; thereforethemeasuredNmvalueof160mg/gwouldbeequivalent

toabout0.49mmol/g.Forinformationonly,themaximumadsorbed amountofanothermacromolecule–bovineserumalbumin–on a nanocrystalline apatite substrate (non-carbonated but with a rathersimilarmaturationstate)reached11.4mmol/g(764mg/g)

[12].

BasedonaTemkin-likebehavior,thelogarithmicfitof exper-imentaldatapointswasaddedinFig.4.Althoughcareshouldbe takenatthisstagefordrawingconclusivestatementsonthe distri-butionofsurfacesitesforDNAadsorption,thesefindingssuggestan evolutionoftheanchoringbehaviorofDNAmacromoleculesalong theadsorptionprocess(i.e.uponcoverageincrease),whichmayin turnberelatedtotheir3Dconformation(s).

3.3.2. Desorptionstudy

Inordertoshedsomemorelightontheadsorptionmechanism forsuchDNA/apatitesystems,thepotentialdisplacementofDNA macromoleculesoutoftheapatiticsurface(desorption)was inves-tigatedbyfollowingtheeffectofdilutionontheresidualadsorbed amount.

A first set of experiments was carried out by performing a dilution by a factor 5 in deionized water, and the results are reportedonFig.5a.Interestingly,asisshowninthisfigure,thedata essentiallyindicatetheabsenceofdesorptioninsuchconditions. Thesefindingsmayprobablybeparalleledwithpreviousworkson theadsorptionofbisphosphonateonapatites[11,29]wherethe absenceofdesorptionupondilutionwasalsoevidenced(pointing outanon-reversibleadsorptionprocessassociatedtoagenuine “anchoring”ofthemoleculesonthesurfaceofthesubstrate).In thecaseofbisphosphonates,thisnon-reversibilitywasrelatedby theauthorstothereleaseofphosphateionsfromthesolidsurface simultaneouslytothemoleculargraftingprocessviaaphosphate end-group.

In the present case, the lack of significant desorption upon dilution could possibly be related toa rather similarscenario, where anionic phosphate groups from the DNA backbonemay interactwithphosphateionsfromthesurface ofthe nanocrys-tals.Calcium andphosphateionsweretitratedfromadsorption supernatants(retrievedafteradsorptionexperimentsforvarious increasingvaluesofCeq)aswellasina“blank”experiment

Fig.5. Effectofdilutionin(a)deionizedwateror(b)inthepresenceofphosphate ions([P]=18mM),pH7,ontheadsorptionofDNAonbiomimeticapatite.(a)After3d ofadsorption,80%ofthesolutionwasreplacedwithdeionizedwater(dilutionstep) for3moredays:the“desorption”datapointsdonotfollowtheisothermdespite thedilutionofthemedium,Nadsremainingessentiallyunchangedforeachpoint, indicatingthequasi-absenceofDNAdesorption.(b)Sameexperimentwithdilution byneutralphosphatesolution.Partialdesorptionofca.30–50%ofDNAisobserved after3days.

cases,ourresultsindicatedthepresenceofbothtypesofionsin solution,andinincreasingamountsasafunctionoftheadsorbed amount,withaconcentrationrangeofphosphorusandcalcium inthesupernatants,respectivelybetweenca.0.3–0.8mmol/land 0.4–1.3mmol/l(datanotshown).Thesefindingscould hypothet-ically be attributed to the (partial) dissolution of the apatitic substrateand/or totherelease ofions fromthesurfacedue to theadsorptionprocessitself.Theconcomitantpresenceof both calcium cations and phosphate anions, however, strongly sup-portsatleastsomeextentofapatitedissolution.Atthispoint,it isdifficulttodifferentiatebetweenthetwotypesofcontributions (dissolution-relatedoradsorption-related);especiallytakinginto accountthenon-congruenceofdissolutionprocessesfor nonstoi-chiometricapatites[29,30].

Additional“desorption”testswerenonethelessruninthe pres-enceofphosphateionsinthemedium(byadditionof KH2PO4,

seeSection2),whilekeepingthepHvalueneutral.Remarkably, inthiscase,theresidualadsorbedamountsweresystematically foundtosharplydecrease(Fig.5b).Thistypeofbehaviorwas pre-viouslyobservedforexampleforbisphosphonates[44]wherethe adsorptionprocesswasaccompaniedbya releaseof phosphate ions.Takingallthesestatementsin consideration,andalthough additional investigation dedicated to these aspects (e.g. to the

Fig.6.EffectofpHontheadsorptionprocessinstandardconditionsofDNAonto biomimeticnanocrystallinecarbonatedapatite(6<pH<9.5).

incongruenceofapatitedissolution)willbeneeded,ourdata sug-gestinthepresentcasethatphosphateionscompetewithDNA moleculesforsurfacesitesofapatitecrystals.

3.3.3. EffectofpHonDNAadsorption

ThemaximumadsorptionamountNmofDNAontothepoorly

crystallinebiomimeticapatitesamplehac-1wwasthenfollowed (stillat22◦_{C)asafunctionoftheinitialpHofthemedium(inthe}

range6–9.5).Indeed,pHmaypresumablyplayanon-negligiblerole inadsorptionprocessesinvolvingioniccrystalssuchasapatites. Also,pHmayalsovaryuponpostmortemconditions.

AsshowninFig.6,thevalueofNmwasfoundtoincreaseupon

acidification (ca.+13%at pH6)of themedium, andconversely todecreaseuponalkalinization(ca.−23%atpH9.5).Several fac-torsmaypotentiallycomeintoplayforexplainingthisbehavior. In particular,partial apatite(surface) dissolutionis expectedto increaseuponacidification[45],thusmodifyingexposedsurface characteristics.Toalesserextent,acidificationcouldbeseenasa meanstofacilitatethereleaseofHPO42−ionsfromthesurfaceof

apatitenanocrystals,astheseionsareknowntobethepredominant formofphosphateionsonsuchsurfaces[46,47].Inturn,suchan increasedmobilityofHPO42−ionscouldthenfacilitatethe

anchor-ingofDNAphosphategroupsontothesurfaceofthecrystals(see discussioninSection3.3.2).Anotherpotentialexplanationforthis pHeffectonNm couldberelatedtoachangeinDNA

conforma-tionuponacidification/alkalinization,thusmodifyingtheprofileof thegraftedmolecules.IncontrastitappearsthatthesepHeffects maynotberelatedtothespeciationofthephosphategroups(main ionizationsiteofdouble-strandedDNA)fromtheDNAbackbone. Indeed,theirpKavalueofca.1.5[48]pointstoasituationwhere

thesephosphategroupsareionizedforanypH>1.5.

ItisinterestingtoremarkhoweverthattheexperimentalpH valuesofthemediaafteradsorption(datanotshown)showedthe generaltendencytoevolvetowardneutrality,independentlyofthe initialpHvalue(whetheracidicofalkaline).Thiseffectmaybeboth due(1)toasurfaceequilibrationphenomenonoftheapatitecrystal surfaceand(2)totheadsorptionprocessitselfleadingto phos-phateionrelease,asindicatedintheprevioussub-section.Indeed, thereleaseofphosphateions(whicharemostlyprotonatedonthe surfaceofbiomimeticapatites,namelyeitherHPO42−orH2PO4−)

inamediumexhibitingapHvaluebetweenca.5and10isexpected toevolvetowardneutralizationduetothebufferingeffectofthe H2PO4−/HPO42−acido-basiccouple(pKa=7.2)[49].

Fig.7. EffectofionicstrengthonDNAadsorptionwithincreasingamountofKCl (0–300mM)instandardconditions.

Inthediageneticcontext,ourobservationsoftheroleofpHon theadsorbedDNAamountsonbiomimeticapatitemayinciteone tohastilyconcludeona“positive”effectofacidicenvironmentson DNApreservation;howeveritshouldbekeptinmindthatapatitic substratesareboundtodegradefasterinsuchconditions,thusnot favoringlong-termDNApreservation[17,50].

3.3.4. Roleofionicstrength

The influence of the overall salinity of the solution (varied herebyincorporationofvariousamountsofKClintheadsorption medium,with[KCl]=1–300mM)wasalsocheckedat22◦_C.Indeed,

ionicstrengthisanotherparameterlikelytoinfluenceadsorption processes[51],aswellasDNAdecay[52].Theobtaineddata(see maintrendsonFig.7)indicateacleartendencytowardincreasing DNAadsorbedamountswhenincreasingtheionicstrengthofthe medium.

Thesefindingscouldbelinkedtoadecreaseininter-molecular interaction,thusfacilitatingtheadsorptionprocessontoapatite crystalssurfaces.Again,avariationinDNAconformationmayplay arole intheseobservations.Theimpactoftheconcentrationin DNA of a givensolution onitsconformation and onmolecular inter-penetration wasaddressed in past studies [53,54]. In the presentcase–aswellasinusualdiageneticprocesses–the den-sityofDNAmoleculesishoweverexpectedtoberatherlimited, thuscorrespondingto“dilute”systemsratherthanhighly concen-tratedones:intheseconditions,eachmoleculemaythushavea tendencytoactasaseparateentity,assuggestedbyTomicetal.

[54]. It should alsobeadded that the “criticalrole of counter-ionvalenceinmodulatinginter-polyionforces”hasbeenrecently underlined,especiallyinthecaseofDNAmacromolecules[54–56]. Thismaythuscomplicatefurthertheunderstandingandmodeling ofDNA moleculardynamicsundermodifiedsalinity conditions; and additional data are probably needed onthis field prior to drawmechanisticconclusionsontheeffectof ionicstrength on DNA conformation and consequently on adsorption. Neverthe-less,ourexperimentaldata,obtainedinthepresenceofKCl,point out a measurable impact onadsorption capabilities of DNA on biomimeticapatite,thussuggestingthatsalinityshouldbe consid-eredasanon-negligibleinfluentialparameterinsuchadsorption processes.

3.3.5. Roleoftemperature

TheeffectoftemperatureonDNAadsorptiononthehac-1w sample wasthenexamined. In viewofshedding somelight on

Fig.8. EffectoftemperatureonDNAequilibriumadsorptionofat4◦_C(5d incu-bation)andat37◦_{C(3dincubation).Forcomparison, theTemkinfitatroom} temperatureisalsoshown

the effect of temperature on DNA adsorption on apatite, dat-apoints were obtained at three temperatures covering a wide range betweenbody temperatureand near-permafrost temper-ature, namely 37, 22 and 4◦_{C and the results are reported in}

Fig.8. Asmaybenoticed, thevalues ofN_ads obtainedat22 or 37◦_{Cdonotdifferdrastically,in contrasttodatacorresponding}

to 4◦_{C. Such “cold”conditions (duringthe adsorptionprocess)}

indeedleadtonoticeablyloweramountsofadsorbedDNA,fora givenDNA concentrationat equilibrium.Theisothermobtained at 4◦_{C also appears to display a different profile compared}

to “warmer” conditions, and linearization tests corresponding to the Temkin, Freundlich or Langmuir models indicated in this case a better correlation with the latter model (correla-tion coefficient R2=0.8515 for Langmuir as opposed to 0.7832 and0.7583respectivelyforFreundlichandTemkin).This obser-vation then points out some modifications in the type of molecule/molecule and/or of molecule/substrate interactions in such a colder situation. It is however difficult, at this stage, to propose a more advanced modeling for this phenomenon as additional information would be needed for instance on the evolution of the surface reactivity (e.g. exchangeability of HPO42− surface ions) of apatitic substrates versus

tempera-ture.

4. Concludingremarks

The present contribution aimed at shedding some light, for thefirsttime ona physico-chemicalpointofview,onthetype of interaction existing between biomimetic apatites and DNA macromolecules,orrelevancebothinforensicandanthropologic contexts.

Ourexperimentalfindingsstronglysupportthegeneral empiri-calhypothesis,oftenemittedintheancientDNAcommunity,after whichDNAstrandscouldinteractwithapatitefoundinhardtissues thusconsiderablylimitingitsdegradationwithtime.Thestudyof thekineticsofDNAadsorptiononasyntheticbiomimeticapatite sample showed that the datacould be adequately fitted toan Elovichianequation,oftenfoundforratherslowadsorption pro-cesses.Theshapeoftheadsorptionisothermsobtainedinvarious conditionswasfoundtobesatisfactorilydescribedbytheTemkin model(exceptatlowtemperature).TheeffectofpH,ionicstrength and temperature onthe adsorbed amounts hasbeen explored, suggestingthenon-negligible roleofenvironmentalparameters

(atleastduringtheadsorptionstage).Althougha “simple” dilu-tionofthemediumdidnotprovokethedesorptionofDNA,thus suggestingastrongbindingaffinityofDNAforapatiticsurfaces, theadditionofphosphateionswasfoundtopromotetherelease of the macromolecules (pointingout a competition for surface sites betweenDNA phosphate groups and inorganic phosphate ions).

Thisstudyisintendedtohelpunderstandingdiagenetic pro-cessesundergonebyDNAinskeletalremains,andalsotobetter apprehendtheinteractionsthatDNAmayundertakewithapatitic substratesfromwhichitisretrieved.Moreadvancedknowledge onsuchinteractionsmayalsoallowonetooptimize,inturn,DNA extractionprocedures.

ThededicatedanalysisofthesolidsafterDNAadsorptionand theexplorationoftheinteractionofshorterDNAfragmentswith apatiticsubstratesrepresentupcomingperspectivesforthiswork, andwillbetheobjectoffuturecomplementarystudies.

Acknowledgements

ThisresearchwassupportedbytheInstituteofEcology and Environment(INEE) andtheInstituteof Chemistry(INC)ofthe FrenchNationalCentreforScientificResearch(CNRS).

References

[1]T.Delabarde,C.Keyser,A.Tracqui,D.Charabidze,B.Ludes,Thepotentialof forensicanalysisonhumanbonesfoundinriverineenvironment,ForensicSci. Int.228(2013)e1–e5.

[2]C.Keyser,C.Bouakaze,E.Crubézy,V.G.Nikolaev,D.Montagnon,T.Reis,etal., AncientDNAprovidesnewinsightsintothehistoryofsouthSiberianKurgan people,Hum.Genet.126(2009)395–410.

[3]C.Keyser-Tracqui,B.Ludes,MethodsforthestudyofancientDNA,Methods Mol.Biol.297(2005)253–264.

[4]N.Nassif,F.Martineau,O.Syzgantseva,F.Gobeaux,M.Willinger,T.Coradin, etal.,Invivoinspiredconditionstosynthesizebiomimetichydroxyapatite, Chem.Mater.22(2010)3653–3663.

[5]A.S.Posner,F.Betts,Syntheticamorphouscalciumphosphateanditsrelation tobonemineralstructure,Acc.Chem.Res.8(1975)273–281.

[6]C.Rey,Calciumphosphatebiomaterialsandbonemineral.Differencesin com-position,structuresandproperties,Biomaterials11(1990)13–15.

[7]E.D.Eanes,J.L.Meyer,Thematurationofcrystallinecalciumphosphatesin aqueoussuspensionsatphysiologicpH,Calcif.TissueRes.23(1977)259–269.

[8]C.Rey,A.Hina,A.Tofighi,M.J.Glimcher,Maturationofpoorlycrystalline apatites:chemicalandstructuralaspectsinvivoandinvitro,CellsMater.5 (1995)345–356.

[9]C. Rey,J.Lian, M.Grynpas, F.Shapiro,L. Zylberberg,M.J.Glimcher, Non-apatiticenvironmentsinbonemineral:FT-IRdetection,biologicalproperties andchangesinseveraldiseasestates,Connect.TissueRes.21(1989)267–273.

[10]S.Cazalbou,D.Eichert,X.Ranz,C.Drouet,C.Combes,M.F.Harmand,etal.,Ion exchangesinapatitesforbiomedicalapplication,J.Mater.Sci.Mater.Med.16 (2005)405–409.

[11]F.Errassifi,A.Menbaoui,H.Autefage,L.Benaziz,S.Ouizat,V.Santran,etal., Adsorptiononapatiticcalciumphosphates:applicationstodrugdelivery,in: R.Narayan,J.McKittrick(Eds.),AdvancesinBioceramicsandBiotechnologies, AmericanCeramicSociety,Westerville,2010,pp.159–174.

[12]S.Ouizat,A.Barroug,A.Legrouri,C.Rey,Adsorptionofbovineserumalbuminon poorlycrystallineapatite:influenceofmaturation,Mater.Res.Bull.34(1999) 2279–2289.

[13]A.S.Posner,Thestructureofboneapatitesurfaces,J.Biomed.Mater.Res.19 (1985)241–250.

[14]S.Cazalbou,C.Combes,D.Eichert,C.Rey,M.J.Glimcher,Poorlycrystalline apatites:evolutionandmaturationinvitroandinvivo,J.BoneMiner.Metab. 22(2004)310–317.

[15]N.Vandecandelaere,C.Rey,C.Drouet,Biomimeticapatite-basedbiomaterials: onthecriticalimpactofsynthesisandpost-synthesisparameters,J.Mater.Sci. Mater.Med.23(2012)2593–2606.

[16]M.J.Collins,C.M.Nielsen-Marsh,J.Hiller,C.I.Smith,J.P.Roberts,R.V.Prigodich, etal.,Thesurvivaloforganicmatterinbone:areview,Archaeometry44(2002) 383–394.

[17]T.Lindahl,InstabilityanddecayoftheprimarystructureofDNA,Nature362 (1993)709–715.

[18]N.Tuross,ThebiochemistryofancientDNAinbone,Experientia50(1994) 530–535.

[19]L.Orlando,A.Ginolhac,G.Zhang,D.Froese,A.Albrechtsen,M.Stiller,etal., RecalibratingEquusevolutionusingthegenomesequenceofanearlyMiddle Pleistocenehorse,Nature499(2013)74–78.

[20]G.Bernardi,Chromatographyofnucleicacidsonhydroxyapatite,Nature206 (1965)779–783.

[21]R.K.Main,M.J.Wilkins,L.J.Cole,Partialchromatographicseparationof pentose-anddeoxypentosenucleicacids,Science129(1959)331–332.

[22]T.Watanabe,K.Makitsuru,H.Nakazawa,S.Hara,T.Suehiro,A.Yamamoto, etal.,Separationofdouble-strandDNAfragmentsbyhigh-performanceliquid chromatographyusingaceramichydroxyapatitecolumn,Anal.Chim.Acta386 (1999)69–75.

[23]F.L.Graham,A.J.vanderEb,Anewtechniquefortheassayofinfectivityof humanadenovirus5DNA,Virology52(1973)456–467.

[24]M.Jordan,A.Schallhorn,F.M.Wurm,Transfectingmammaliancells: optimiza-tionofcriticalparametersaffectingcalcium-phosphateprecipitateformation, NucleicAcidsRes.24(1996)596–601.

[25]M.Okazaki,Y.Yoshida,S.Yamaguchi,M.Kaneno,J.C.Elliott,Affinitybinding phenomenaofDNAontoapatitecrystals,Biomaterials22(2001)2459–2464.

[26]H.Autefage,F.Briand-Mésange,S.Cazalbou,C.Drouet,D.Fourmy,S.Gonc¸alvès, etal.,AdsorptionandreleaseofBMP-2onnanocrystallineapatite-coatedand uncoatedhydroxyapatite/b-tricalciumphosphateporousceramics,J.Biomed. Mater.Res.B:Appl.Biomater.91B(2009)706–715.

[27]L.Benaziz,A.Barroug,A.Legrouri,C.Rey,A.Lebugle,Adsorptionof O-phospho-l-serineandl-serineontopoorlycrystallineapatite,J.ColloidInterfaceSci.238 (2001)48–53.

[28]D.N.Misra,Adsorptionandorientationoftetracyclineonhydroxyapatite,Calcif. TissueInt.48(1991)362–367.

[29]P.Pascaud,P.Gras,Y.Coppel,C.Rey,S.Sarda,Interactionbetweena bisphos-phonate,tiludronate,andbiomimeticnanocrystallineapatites,Langmuir29 (2013)2224–2232.

[30]H.Tanaka,K.Miyajima,M.Nakagaki,S.Shimabayashi,Incongruentdissolution ofhydroxyapatiteinthepresenceofphosphoserine,ColloidPolym.Sci.269 (1991)161–165.

[31]A.Gee,V.R.Deitz,Determinationofphosphatebydifferential spectrophoto-metry,Anal.Chem.25(1953)1320–1324.

[32]R.Legros,Apportdelaphysico-chimieàl’étudedelaphaseminéraledestissus calcifiés(Thèsed’État),Institutnationalpolytechnique,1984.

[33]M.J.D.Low,Kineticsofchemisorptionofgasesonsolids,Chem.Rev.60(1960) 267–312.

[34]C.Aharoni,F.C.Tompkins,KineticsofadsorptionanddesorptionandtheElovich equation,in:H.Pines,P.B.Weisz,D.D.Eley(Eds.),AdvancesinCatalysis, Aca-demicPress,1970,pp.1–49.

[35]F.-C.Wu,R.-L.Tseng,R.-S.Juang,CharacteristicsofElovichequationusedfor theanalysisofadsorptionkineticsindye–chitosansystems,Chem.Eng.J.150 (2009)366–373.

[36]J.M.Thomas,W.J.Thomas,PrinciplesandPracticeofHeterogeneousCatalysis, VCH,NewYork,USA,1996.

[37]C. Cheung,J. Porter,G. Mckay, Sorptionkinetic analysisfortheremoval of cadmium ions from effluentsusing bonechar, Water Res.35 (2001) 605–612.

[38]A.G.Ritchie,AlternativetotheElovichequationforkineticsofadsorptionof gasesonsolids,J.Chem.Soc.FaradayTrans.73(1997)1650–1653.

[39]L.Sciascia,M.L.TurcoLiveri,M.Merli,Kineticandequilibriumstudiesforthe adsorptionofacidnucleicbasesontoK10montmorillonite,Appl.ClaySci.53 (2011)657–668.

[40]J.O.Bockris,K.T.Jeng,In-situstudiesofadsorptionoforganiccompoundson platinumelectrodes,J.Electroanal.Chem.330(1992)541–581.

[41]G.Skodras,I.Diamantopoulou,G.Pantoleontos,G.P.Sakellaropoulos,Kinetic studiesofelementalmercuryadsorptioninactivatedcarbonfixedbedreactor, J.Hazard.Mater.158(2008)1–13.

[42]S.Trasatti,L.Formaro,Kineticsandmechanismoftheadsorptionof glyco-laldehydeonasmoothplatinumelectrode,J.Electroanal.Chem.Interfacial Electrochem.17(1968)343–364.

[43]M.A.Vannice,KineticsofCatalyticReactions,Springer,NewYork,2005.

[44]F.Errassifi,Mécanismesd’adsorptiondurisédronatepardesphosphatesde calcium biologiques: applicationsauxbiomatériaux, FacultédesSciences Semlalia-Marrakech,2011.

[45]J.C. Elliott, Structureand Chemistry of the Apatites and Other Calcium Orthophosphates,ElsevierScience&Technology,Amsterdam,1994.

[46]C.Rey,M.Shimizu,B.Collins,M.J.Glimcher,Resolution-enhancedFourier trans-forminfraredspectroscopystudyoftheenvironmentofphosphateioninthe earlydepositsofasolidphaseofcalciumphosphateinboneandenameland theirevolutionwithage:2.Investigationsinthe3PO4domain,Calcif.Tissue

Int.49(1991)383–388.

[47]C.Rey,M.Shimizu,B.Collins,M.J.Glimcher,Resolution-enhancedFourier trans-forminfraredspectroscopystudyoftheenvironmentofphosphateionsinthe earlydepositsofasolidphaseofcalcium-phosphateinboneandenamel,and theirevolutionwithage.I.Investigationsinthe4PO4domain,Calcif.Tissue

Int.46(1990)384–394.

[48]J.A.V.Butler,ProgressinBiophysicsandBiophysicalChemistry,PergamonPress, London,1951.

[49]G.Charlot,L’analysequalitativeetlesréactionsensolution,Masson,Paris, France,1963.

[50]R. Bollongino,A.Tresset,J.-D.Vigne,Environmentandexcavation.Pre-lab impactsonancientDNAanalyses,C.R.Palevol.7(2008)91–98.

[51]D.N.Misra,AdsorptiononandSurfaceChemistryofHydroxyapatite,Springer, NewYork,1984.

[52]T.Lindahl,B.Nyberg,Rateofdepurinationofnativedeoxyribonucleicacid, Biochemistry(Mosc.)11(1972)3610–3618.

[53]A.V.Dobrynin,M.Rubinstein,Theoryofpolyelectrolytesinsolutionsandat surfaces,Prog.Polym.Sci.30(2005)1049–1118.

[54]S.Tomi´c, D. Grgiˇcin,T. Ivek,T.Vuleti´c, S.DolanskiBabi´c, R. Podgornik, Dynamicsand structure ofbiopolyelectrolytes inrepulsion regime char-acterizedbydielectricspectroscopy,Phys.B:Condens.Matter407(2012) 1958–1963.

[55]A.Naji,M.Kanduc,R.R.Netz,R.Podgornik,Exoticelectrostatics: unusual featuresofelectrostaticinteractionsbetweenmacroions,ArXivPrepr,2010 arXiv:1008.0357.

[56]R.W.Wilson,V.A.Bloomfield,Counterion-inducedcondensationof deoxyri-bonucleic acid. Alight-scattering study, Biochemistry (Mosc.) 18 (1979) 2192–2196.