The effects of divalent cations in the presence of phosphate, citrate and chloride on the aggregatio

www.elsevier.com/locate/foodres

Thee󰂀ectsofdivalentcationsinthepresenceofphosphate,citrate

andchlorideontheaggregationofsoyproteinisolate

MarõaIsabelMolina,JorgeRicardoWagner*

ÂnyDesarrolloenCriotecnologõÂadeAlimentos(CIDCA)CONICET,FacultaddeCienciasExactas,UniversidadNacionaldeLaPlata,CentrodeInvestigacio

Calle47y116(1900)LaPlata,ProvinciadeBuenosAires,Argentina

Abstract

Theproteinaggregation±dissociationprocessinducedbydivalentcationswasstudiedusingdilutedaqueousdispersionsofanativesoyisolate.Turbidity,asaproteinaggregationindicator,wasquanti®edata®xedtimeafteradditionofincreasingamountsofCaCl2,MgCl2oranequimolarmixtureofthem.InallcasesturbidityvaluesattainedwithCa2+werehigherthantheseobtainedwithMg2+.Thee󰂀ectsofphosphate,citrateandchloridepresencewerealsodetermined.Phosphate,becomesincorporatedintoCa2+-inducedaggregates,butnotinMg2+ones.Bothcitrateandchlorideionsinhibitproteinaggregationandcontributetothedissociationofpreviouslyformedaggregates.Lowcitrateconcentrationsallowedaggregateformation,evenathighCaCl2con-centrations;thisbehaviorwasnotobservedinthepresenceofMgCl2.Asedimentationbasedtestwasdevelopedasastabilityparameterofdispersionsofsoyproteinaggregates.Resultsshowedthateventhoughaggregationwasinducedbydivalentcations,competitionbetweenphosphate,citrateandsoyproteincouldmodifytheaggregates'properties.Optimumaggregationconditionswereestablishedfromcombinationsofthesevariables.#1999CanadianInstituteofFoodScienceandTechnology.PublishedbyElsevierScienceLtd.Allrightsreserved.

1.Introduction

Thecomplexbehaviorofmilkcaseinisbasedonadelicateequilibriuminwhichtheseproteinstogetherwithdi󰂀erentsaltsformhighlyhydratedmicelles.Caseinmicellesconstituteahighlystabledispersionthatprovidesacharacteristicwhiteopalescencetothemilk.AlthoughmicelleformationdependsonCa2+concentration,thee󰂀ectofphosphateisimportantaswell.Themaininor-ganiccomponentofcaseinmicellesismicellarcalciumphosphate,alsocalledcolloidalcalciumphosphate(Holtetal.,19;vanDijk,1990).Theformationofarti®cialcaseinmicellesystemsbyaddingCa2+,phosphateandcitratetosodiumcaseinatesolutionswerealreadystudied(Aoki,Uehara,Yonemasu&ZinEl-Din,1996).Althoughthestructureandphysicochemicalcharacter-isticsofsoybeanproteinsarehighlydi󰂀erentfromthoseofbovinemilkcasein,milkisahelpfulmodeltoimitate.Thus,themainsaltsofbovinemilkarevariablesto

*Correspondingauthor.Tel.:+-21-249287/2853.

E-mailaddress:jrwagner@nahuel.biol.unlp.edu.ar(J.R.Wagner)

consideronprocessofformationofsoybeanproteinaggregates.

Bycombiningvegetableoils,sugarsotherthanlac-tose,proteins(forexamplesoyproteins),stabilizers,emulsi®ersanddrinkingwater,researchershavedesignedproductssimilarinbothappearanceandnutritionalvaluetoadairyproduct.Theimitationdairyproductsutilizephosphatesasbu󰂀erstomaintaintheoptimumpHforproteindispersion;thus,theyaidinstabilizingthefatparticlesintheproduct(Ellinger,1972).Themostcommonphosphatesinusearedis-odiumphosphateand;thelaterisusuallypreferred,becauseitdoesnotcontributeasmuchastringentasdisodiumphosphate(Holland&Bandler,1967;Moede,1968).Itwasreported(Sabharwal&Vakaleris,1969)thattheadditionofCa2+,phosphateandcitrateattheproperlevelsinthepropersequencewasnecessarytoproducestabledispersionssimilartomilk.

Previousworkshowedthatsoybeanglobulins,mainly11Sfraction,bindCa2+andMg2+inducingaggregateformation(AppuRao&NarasingaRao,1975a,1975b).

n(1995),havereportedSorgentini,WagnerandAno

thatdilutedaqueousdilutionsofnativesoyisolatereach

0963-9969/99/$20.00#1999CanadianInstituteofFoodScienceandTechnology.PublishedbyElsevierScienceLtd.Allrightsreserved.PII:S0963-9969(99)00086-1

136M.I.Molina,J.R.Wagner/FoodResearchInternational32(1999)135±143

ationturbidityMgClof40maximummMMgClduetoaggregationataconcentra-2.Athigherconcentrationsoftion2,thesequesteringofproteinturbidityaggregates.decreased,Thethuse󰂀ectindicatingofdissocia-tion±disaggregationagents(citrate,Recently,processphospate)chlorideandhasnotontheaggrega-havephate,studiedYazici,However,separately,thesequesteringAlvarez,Mangino,andbeenHansenstudied.(1997)onsoymilke󰂀ectstabilityofcitrateandandviscosity.phos-aggregatesitcitrateonascaseinincludeisstillmicellesCaquestioned2+whethersoyproteindo,and/orandhowMg2+,phosphateandwetheaggregates'characteristics.Inthosethepresentionsin¯uencework,inducingstudiedofagentsthee󰂀ectofCaCl2andMgCl2asaggregationtionphosphate,anddissociationcitrateandtheofande󰂀ectsoychlorideofvariousproteinionsconcentrationsaggregates.onbothforma-2.Materialsandmethods

trolledDefatted,proteinconditionssolvent(notfreesoy¯our,preparedundercon-(Brazil).

denaturation),wasthermallyprovidedinactivatedbySanbratoavoidS.A.2.1.Preparationofnativesoyisolates

soyNativeSoyIsolate(NSI)wasobtainedfrompH¯ourasfollows:soy¯ourwasextractedwithdefattedwateratperature,8.0adjustedwith2NNaOHfor2hatroomtem-adjustedcentrifugedtousingpHwashedat6900󰀀4.5awater/¯ourwithratio10:1.Theextractwasgfor1N20HCl,minkeptat4forC.2Thehatpellet4CandwasNaOH.withwaterandneutralizedtopH8.0with2Nand®nepassedThethroughobtainedaisolate0.25mm-meshwasfreeze-dried,sievegroundbothpowder.DSCthermogramofNSIsampletopresentedobtaina(peaktransitionstemperaturescharacteristicsof77.6󰂋1.2of7Sandand11S90.5󰂋1.6globulinsrespectively)J/g,nativewhichandatotaldenaturationenthalphyof17.6C,proteinscon®rm(Wagner,thatSorgentini,thissampleandisAncomposedon,1996).of

2.2.Preparationofnativesoyisolatesolutions

tionsNSIbu󰂀er(0.5,solutionsand10weremg/ml)preparedinatdi󰂀erentconcentra-7.0withsolutions1NNaOH.

(10,30and50distilledmM)waterandadjustedandKHto2POpH42.3.turbidityStudymeasurements

ofproteinaggregation±dissociationby(CaAggregation2+andMg2+was)toobtainedNSIsolutionsbyaddingatroomdivalenttemperature

cations(approximatelyMgCl20C).Solutionsof1MCaCl2,1Mused.2uatedTurbidityoranequimolarduetomixtureoftheformeronesweremeasuringwithofTurbidity0.5mg/mltheaBeckmam-DU650aggregateformationwaseval-proteinabsorbancedispersionat600spectrophotometer,by(Sorgentininm(A600nm)of1mlsuccessivewassolutionsperformed(toadditionsmeasuredobtainofwithoutthecationsetal.,and1995).after0±12010mlmMaliquotsrange).ofdivalentReadingscationtationafunctionbyinversion.15saftereachaddition,withapreviouswereagi-ofCa2+Turbidity,Mg2+andpro®lesCa2+were+recordedMg2+ascentrations,centrationsatwater,intheselectedphosphatecon-con-citrateproteinsolutions.andinSimilarthepresencedeterminationsandabsencewere10runmMturbidity.blanksproteinToanalyzetoevaluatethee󰂀ectssaltcontributiontototalonsolutionsdissociation,Citratewereaddedincreasingamountsofionicofstrength1MNaClonamountse󰂀ectswerefollowedoverabyrangeadditionof0to120mM.and70mMofcitrate.

1Mtrisodiumcitrate,rangingofbetweenincreasing02.4.DeterminationofaggregatedproteinamountNSIAggregatedispersionswereobtainedfrom0.5mg/ml7),mumwithsolutionsconcentrationandwithoutinbothof10waterCamMand30mMphosphate(pH2+,citrateMg2+andorwithMg2+the+Caopti-2+thatofProtein3mlgavewerethecentrifugedmaximumatturbidity.12100󰀀gDispersionfor30minaliquotsat20C.byconcentrationwasdeterminedinthesupernatantRandall,theLowrymethod(Lowry,Rosebroug,Lewis&wasinitialcalculated1951)andconsideringthepercentageproteinofconcentrationaggregatedproteinofthe(Asolutionas100%.Relationshipbetweenturbiditydegree600nmbeforecentrifugation)andprotein100mM(%CaClofaggregated.

protein)wasanalyzedaggregationat10±22.5.cationsDeterminationamountsintheofphosphateaggregates

(Pi)anddivalentgationOntheextentsamewassupernatantmeasured,fromPi,whichCa2+proteinand/oraggre-tionconsideringintowereproteinalsoaggregatesdetermined.Thedegreeofincorpora-Mg2+contentsContentChen,ofPitheofeachionwascalculatedwascorrespondinginitialconcentration.sition2.5%wasToribara,determinedfollowingthemethodofoneandWarner(1956).Reactantcompo-acidammoniumvolumemolybdate,10%ascorbiconevolumeacid,6oneNvolumesulfuricthebathsampleandtwoinvolumesa7:3(v/v)distilledwater.Thereactantand820nm.at45DeterminationsCfor20min.ratiowereincubatedinawaterofAbsorbanceCa2+andMgwas2+measuredwerecarried

atM.I.Molina,J.R.Wagner/FoodResearchInternational32(1999)135±143137

outWhitneybytheend(19),indirectusingtitrationcalceinmethodasindicator.ofNtailianasToimproveandmodi®edpointdetermination,thetitrationmethodwascalcein,0.12usinggtimolftaleinethefollowingandmixture20gKNOindicator:0.2g3.2.6.Stabilityofproteinaggregates

pensedAliquotsof2mlNSIAtsolutionsin5mlNSIaggregatesdispersionsweredis-attubes10with0.1ml-intervalgraduation.necessaryzerotime,theamountmg/mlinweremltestedof1Matallconditions.imumtoget®naloptimumconcentrationcationsolutiongatesturbidity)wasadded.Sedimentationofaggre-(max-measuringwaslesstherecordedheightofforthe30zoneminbetweenat3min-interval,themoreAllclearassayssolutionwerecarriedandtheoutsettledatleastaggregates.

orinduplicate.3.Resultsanddiscussion

3.1.cations

InductionofproteinaggregationwithdivalentwasTurbidityinwater,asaproteinmaximumsimilar(Fig.Aforbothcationsaloneaggregationormixed,reachingindicator,a600nmofdueto1).CaTurbidity2+iscausedapprox.byformation0.6±0.7forof20±30aggregatesmMRaoexcess,etal.,1975a,/Mg2+1975b).and11STurbiditysoyfractiondecreasebindingbyMgCl(AppuviouslythishypothesisreportedattributedandbytoextendSorgentinidissociationittoetCaal.process,2+(1995).,thepercentageTowasvalidatepre-2of

Fig.ml)1.Turbidity(A600nmmeanasafunctionofdivalent)ofancationaqueousconcentration.dispersionofNSI(0.5mg/mean).

ofatleasttwoassays(SE=5%,expressedasapercentageEachvalueofistheaaggregatedwasimumdetermined.proteinAggregatedintheassayedproteinCaCl2concentrationsincreasedandMgCl(Fig.thendecreasedasCaClreachedamax-2concentrationturbidity(resultsnot2);shown).asimilartrendwasobservedfor2obtainedconsidered(Fig.andpercentageofAgoodaggregatedcorrelationproteinsbetweenwastionalreadydegree.asThea2b).parameterHence,turbidity(A600nm)couldbee󰂀ectsofsaltstoestimateonproteinproteinaggrega-dissociationknownMgClofaggregates(Damodaranobserved&Kinsella,solubilityathigh1982).CaClTheare2orion2concentrationscouldbeattributedtoofIntheexcess.Todemonstratethatthisanionisresponsiblechloridethedissociation®rstone(assayprocess,1)aggregationtwoassayswerewasinducedperformed.by

Fig.anconcentration.aqueous2.(a)Turbiditydispersion(A600ofnmNSI)and(0.5percentagemg/ml)asofaggregatedproteinincles=ASolidsquares=%aggregatedafunctionprotein,ofopencalciumcir-centage600nm;(b)relationshipbetweenturbidity(A600nm)andper-atleasttwoofaggregatedassays(SE=5%,proteinexpressedintheassayas(a).apercentageEachvalueofistheameanmean).

of138M.I.Molina,J.R.Wagner/FoodResearchInternational32(1999)135±143

Ca2+inanalyzedtheandsecondMg2+inthepresenceof200mMNaCl,andsionscurveswithbyone(assay2)proteindissociationwaspreviouslyincreasingformedNaClconcentrationindisper-respondingofAaggregates.Inassay1the600nmvs.divalentcationconcentrationcor-compared.excess,Fig.tosystems3bshowswithlowandturbiditywithoutvaluesNaClforwereClIndispersionassayeven2increasingwithhighamountsCaCl2andofMgCl2concentration.0.5tionmg/mlNSIofproteinand20aggregatesmMCaClformedNaClwereinwateraddedwithto2.AsNaClconcentra-showedincreasedaggregationaninhibitingturbiditye󰂀ectdecreasedofCl(Fig.ion3a).onBothsoyassaysconcentrationsInthecaseinducedofsinglebydivalentglobularcations.

proteinproteins,atlowsaltwouldhighberesponsiblethedecreaseofelectrostaticinteractionstowatertheconcentrationsforthesalting-ine󰂀ect,whileatcompetitionthesalting-oute󰂀ectwouldbedueglobulins(Damodaran(highlyordered&betweenKinsella,saltsoligomeric1982).andstructures),Inproteinsthecaseforsolubility

ofsoytheFig.mM3.(a)Turbidity(A600nm)ofadispersionturbidityNaClsolution,(Asolution,asadispersionfunctionof0.5divalentmg/mlcation0.5mg/mlofNSIconcentration;ofNSIin200in20mMCaCl(b)600nm)ofa(󰂋2atleast10-4)astwo%aggregatedafunctionofassays(r=correlationproteinNaCl(rconcentration:2=0.993913).A2600nm=0.018+0.008coe󰂁cient).

Eachvalueisameanofdecreasesincreasing(salting-out)CaClathighconcentrations.atlowionicInstrengththisregard,(40.2bothM),being2tributioninandtheNaCl®rstshowedthesamesalting-oute󰂀ects,bridges.toproteincaseaggregatemoremarkedformationduetobyCa2+calciumcon-trostaticThismolecules.andhydrophobicbehaviorcaninteractionsbeattributedbetweentobothproteinelec-chargedthegroupsAlthough,bysaltsthemayelectrostaticincreaseproteinshieldingsolubility,oftheteinsdecreasetheirmayenhanceinthetheelectrostatichydrophobicrepulsioninteractionbetweenbetweenpro-gatesnonpolarcentration(Damodaransurfaces&andKinsella).leadtoIncreasingformationsaltofaggre-hydrophobicmayincreaseordecreasethemagnitudecon-natureandoninteraction,proteinhydrophobicity.dependingonThus,lyotropictheCaionof2+e󰂀ectinteractiononsoyproteinwouldbeahydrophobicbetweenmoleculesinteractionspolypeptidesfavoredbymediatedstrongthebyelectrostaticCa2+andandtohightheClweakeningatlowsaltanionconcentrationsofconcentrationsstructuringwaterthoseelectrostatic(aggregatione󰂀ect)(dissociationinteractionse󰂀ect).due3.2.dissociationE󰂀ectofprocess

phosphateandcitrateonaggregation±degreeWhenMg2+,alsophosphatebutincreasedamountsnotwithinMgthe2+presenceincreased,ofCa2+aggregationorCa2+/WhilemaximumfortheACa2+assayapproximatelywithalone30mM(Figs.Pi4a±cand5).2.0,solution,forMgthe600nmwas2+under5).InthethesamecaseconditionsofthemixturethevalueCa2+wasabout0.5(Fig.turbidityencewassimilartothatofCa2+/Mg2+alone.maximumThepres-evenbtoandatofc).cationMg2+ProteinconcentrationsandCa2+allowedblanksshowedhigheraggregateformationamarkedthan30turbiditymM(Fig.due4toCa2+insolublesalts,probablyCa3(PO4)2,incontrastverifyMg2+thewhethersaltsturbidity(Fig.6).Thus,itshouldbepossibletotopresenceofPiinNSIincreasesolutions,isinducedorbyCa2+inprotein±Catheformation2+salts,aggregates,ofprotein±Ca2+±Piwhetheraggregates,itisdueoroneortothesimultaneousoroccurrenceonlytoCa2+ofmoreinsolublethanbidityprocessiumishighertodi󰂀erentinthepresencedegrees.ofConsideringthattur-in(Fig.6),themostprobablesituationproteinswouldinthebemed-thatIfwhichcompetitiononlyprotein±Caprotein±Ca2+2+±PiandaggregatesPi±Ca2+willexisted,predominate.bidityaswouldfordecreasethecationorshouldbeexpected,aandstrongtur-®edIntheatleastwouldbemaintainedorderblank.

toexplainourresults,weneouslygroupcation(Cainof2+,systemsequilibriumpostulateasimpli-Mg2+,Caofthatmaycoexistsimulta-2+/Mgprotein±phosphate±divalent2+),asshownfollowing:

M.I.Molina,J.R.Wagner/FoodResearchInternational32(1999)135±143139

Fig.oftiondivalent4.Turbiditycation(Aconcentration.600nm)ofadispersionAssayswere0.5mg/mlperformedofNSIasafunctionleastof:two(a)assays10mM,(SE=5%,(b)30mMexpressedand(c)as50amM.percentageEachvalueatoftheisPimean).

aconcentra-meanofatFig.ml)atleastresulting5.Turbiditymaximum(A600nmmax)ofNSIdispersions(0.5mg/twoassaysfromassays(SE=5%,showedexpressedinFig.as4a±c.apercentageEachvalueofistheameanmean).

ofFig.NSIconcentration.(0.56.Turbiditymg/ml)in(A30600mMnm)Piofsolution,dispersionsasawithout(blank)andwithexpressedasapercentageEachvalueofistheameanmean).

ofatleastfunctiontwoofassaysdivalent(SE=5%,cation2󰀀

$Ca2󰀀Pa󰀀PMg2󰀀

$

Mg2󰀀Pb󰀀

Ca2󰀀PO43$Ca3PO42

PS1025

c󰀀

Mg2󰀀PO43$

Mg3PO42

PS2:51013

d󰀀PCa2󰀀PO43$PO43Ca2󰀀Pe󰀀PMg2󰀀PO43$

PO43Mg2󰀀P

f󰀀

wherevaluePissoyprotein,andPSistheproductsolubilityJimeno,at20Ca+2concentrationor19).C(Burriel,Marti,LucendaConde&ArribasMg+2OtherhighmayenoughbephosphatepresentcompoundsthatbindtoatpH7.0,butnotin(a)Themaximunturbididyofbothcompetedispersionsforthese[reactionscations.theandButabsorbance(b)above]equilibriaifproteinismeasuredisreachedat[cation]=20mM,anddispersedisabout0.6(Figs.8aand9a).petecoexist,andbothinproteinphosphateandphosphatebu󰂀er,severalMg2+forthedivalentcation[reactions(c)±(f)above].Whencom-in[reactionthemedium,istheinducingcation,increasesinPiconcentrationMg(d)],whichleadsistoconsistentmoreMg2+withintheequilibriumhighPSvaluewithPiofwith3(PO4)2.Thisfactdoesnotincreaseturbidity,asseengationproteintionaggregates(f)[reactionblanksisnotinducedimportant,(b)].(Fig.Asit6)willbutbedecreaseexplainedproteinlater,aggre-reac-byMgPO+243isnotincorporatedintoresults,tendwhenphosphateincreases,andPi.MgAccording2+toourtowardstodecreasetheright(Fig.side.5),indicatingInthepresenceashiftofofreactionaggregatesCa2+,the

(d)140M.I.Molina,J.R.Wagner/FoodResearchInternational32(1999)135±143

increaseinPiconcentrationallowstheadditionalfor-mationofprotein±Ca2+±Piaggregates[reaction(e)],andCa3(PO4)2insoluble[reaction(c)]andbothofthemincreaseturbidity,reachingabsorbancevalueshigherthanonesobtainedbyproteininwater(Fig.2a).Inthecaseofcationmixtures,thefactthatmaximumA600nmFig.7.Turbidity(A600nm)ofNSIdispersions(0.5mg/ml)withoutPiaddedandwithPi(30mM),asafunctionofcitrateconcentration.Eachvalueisameanofatleasttwoassays(SE=5%,expressedasapercentageofthemean).

remainedconstantisnoteasytoexplain.OnemaythinkthatsincePia󰂁nityforCa2+ismuchhigherthanforMg2+,thePianionwouldbeconsumedintheforma-tionofinsolublecalciumphosphateandprotein±Ca2+±Piaggregates,leavingmoreMg2+freetointeractwithproteinsandtoformprotein±Mg2+aggregates[reac-tion(c)],allprocessesleadingtoturbidityincrease.

Thecitratee󰂀ectwasstudiedforanaggregatedisper-sionof0.5mg/mlNSIin30mMCaCl2inwaterandin30mMPi(Fig.7).Forthelatter,theadditionofincreasingamountsofsodiumcitrateledtoagradualdecreaseinturbidity,duetoitssequesteringe󰂀ectoncations,thus,makingthecationlessavailableforinter-actionwithproteins.Thecitrate-inhibitinge󰂀ectonaggregationofproteinsordissociationofaggregatesishigherintheabsenceofphosphate.Withorwithoutphosphate,aggregationwastotallyinhibitedatcitrateconcentrationsabove40mM.Thus,alowcitratecon-centrationof10mMwaschosentostudyaggregatestability.Itshouldbestressedthatthisorganicpoly-anionispresentinbovinemilkinaapproximatelysimilarconcentrationandhelptomaintainthecaseinmicellarstructure(Yazicietal.,1997).Thepresenceof10mMcitrateintheaggregationprocessofNSIinwaterledtoadecreaseofattainedturbidity[reaction(a)],withnodi󰂀erencesinthepositionofthemaximumA600nmbetweenCa2+andMg2+inducedprocesses(Figs.8aand9a).Citrateanionformssolublecomplexes

Fig.8.Turbidity(A600nm)ofdispersionswithout(blank)andwithNSI(0.5mg/ml),asafunctionofmagnesiumconcentration.Assayswereper-formedwithoutandwithsodiumcitrate(10mM),and(a)withoutPiaddedor(b)withPiat10mM,(c)30mMand(d)50mM.Eachvalueisameanofatleasttwoassays(SE=5%,expressedasapercentageofthemean).

M.I.Molina,J.R.Wagner/FoodResearchInternational32(1999)135±143141

Fig.9.Turbidity(A600nm)ofdispersionswithout(blank)andwithNSI(0.5mg/ml),asafunctionofcalciumconcentration.Assayswereperformedwithoutandwithsodiumcitrate(10mM),and(a)withoutPiaddedor(b)withPiat10mM,(c)30mMand(d)50mM.Eachvalueisameanofatleasttwoassays(SE=5%,expressedasapercentageofthemean).

withCa2+andMg2+decreasingcationavailabilitytoformproteinaggregates.

Inthepresenceof10mMcitrate,whenaggregationbyCa2+isinducedinNSIdispersionsinbu󰂀erwith10,30and50mMPi(Figs.9b±d),onlyashiftoftheposi-tionofmaximumturbiditytowardshighercalciumconcentrationwasobserved.SincebothcitrateandphosphatecompeteforCa2+,moreCa2+isrequiredtoreachmaximumaggregation.TurbidityisstabilizedathighA600nmvalueinthepresenceofcitrate,incontrasttowhatoccurredindispersionswithoutcitrate.Appar-ently,citratebeingapolyanionmightcompetewithchlorideanion,avoidingitsinhibitinge󰂀ectonaggre-gation.Citratewouldnotonlystabilizeproteinaggre-gatesbutalsoinsolublephosphates,asseeninassayswithoutprotein.Figs.8b±dshowadi󰂀erentsituationforMg2+.Nodisplacementofmaximumturbiditywasobservedandthemaximumneverreachedvaluessimilartothoseofdispersionswithoutcitrate.Inthissystem,bothcitrateandphosphatecompetewithsoyproteinforMg2+ion,leadingtoMg2+±citrateandMg2+±phos-phatesolublecomplexes.WhenMg2+concentrationishighenoughtoovercomethiscompetitionandtoincreasetheamountofprotein±Mg2+aggregates,theexcessofchloridefromMgCl2startsthedissociationprocess(Figs.8b±d).

3.3.Percentageofphosphate,divalentcationandproteinaggregatedunderdi󰂀erentconditions

Fig.10showsthatMg2+,Ca2+orCa2+/Mg2+mix-turesproduceacomparableamountofaggregatedpro-teininwater.Whentheaggregationwascarriedoutinthepresenceof30mMPi,thepercentageofaggregatedproteinwassimilar(78±84%).Ca2+isslightlymoree󰂀ectivethatMg2+toproducesoyproteinaggregationinpresenceofPiion.Althoughtheamountofaggre-gatedproteinisalmostthesame,theturbidityvaluesweregreaterwithCa2+orCa2+/Mg2+thatthosewithMg2+alone(Figs.4and5).TheseresultscouldbeexplainedbytheparticipationofPiioninprotein±Ca2+orprotein±Ca2+±Mg2+aggregates,butnotinprotein±Mg2+aggregates(Fig.10).

BasedonCa2+,Mg2+andPiquanti®cationinaggregatesobtainedunderdi󰂀erentconditions,wecal-culatedtheamountofdivalentcationandPiinmmolpergoftotalprotein.Thecompositionanalysisofaggregatesformedunderdi󰂀erentconditionsisshowninFig.11.IntheabsenceofPiinthemedium,forbothMg2+andCa2+,theincorporatedamountofcationinsoyproteinaggregateswaslow(0.04mmolCa/gpro-tein;0.022mmolMg/gprotein),whichisinagreementwithpreviousreporteddata(Sakakibara&Noguchi,

142

1977).WhenaggregationwasinducedbyCa2+presenceaggregatesofPi,therelativeinthefortheMg2+wassixfoldlargeramountthaninofwater.cationHowever,intheCa2+agreementandPinearlycombinations,doubledrelativetheamounttowater.ofcationBesides,inmorePiaggregatesthanthoseincorporateinducedbybetweenMg2+10to20timestheideathatprotein±Ca2+,whichsupportFig.in11alsoshowsthatPiand±Piaggregatesdivalentcationwereamountsformed.obtainedtheaggregatescouldaccountforturbidityvaluesturbidityinsoyunderwereproteindi󰂀erentreached.

aggregatesconditions.was1:2,WhenhigherthevaluesPi/Ca2+ratioof3.4.sedimentation

StabilityofdispersionsofaggregatesagainstshortThetimesstability(30ofmin),aggregateataproteindispersionsconcentrationwasstudiedof1%

atFig.sions10.aggregation(0.5Percentagemg/ml)inofdistilledbothaggregatedwaterproteinandPiinNSIdisper-valuewasinducedbyadditionandof3030mMmMdivalentPisolution.cation.Proteincentageisofameanthemean).

ofatleasttwoassays(SE=7%,expressedasaEachper-Fig.protein11.Content(mmol/gprotein)ofPianddivalentcationsinsoydivalentaggregates.mMcationsatAggregationNSIdispersionswas(0.5inducedmg/ml)byadditionof30mMmeanPi,andwithoutorwithsodiumcitrate(10mM).withoutEachorvaluewith30mean).

ofatleasttwoassays(SE=6%,expressedasapercentageofistheaforkineticsallanalyzedhydratedofments.However,aggregatesaggregatesconditions.forvs.as30mMtime,quanti®edInwater,sedimentationPiwassolutions,similarbytheMgforheight2+alltreat-ofgatesaggre-thanhighCawere2+moreonesunstable,andsettledatahigherratelargedensityPisizeand/orofprotein±Mg(Fig.12).2+Thismaybeattributedtolowdegreeofaggregates,hydration.Theduetotheiringasacomponentofproteinaggregates,besidese󰂀ecttend-ofdegreetoincreasesdi󰂀erenceofhydration.inturbidity,Whenwouldbetoincreasetheiumbetweendensitiesofhydrationaggregatesisincreased,andthemed-thetoo.getsvelocityAssayslower,withandthensedimentationvelocityisloweradditionofaggregatescitrateformedshowedwasthatnotsedimentationmodi®edbyresultsrequired.

areofthisnotpolyanionconclusive,(dataadditionalnotshown).studiesSincetheseareFig.mM12.mMdivalentSedimentationcationsatofNSIproteindispersionsaggregates(1%formedp/v)inbyadditionof30expressedPisolution.asapercentageEachvalueofistheameanwaterorin30mean).

ofatleasttwoassays(SE=6%,