Marijn M. Kool - WUR

Enzymatic modification and characterization of xanthan

MarijnM.Kool

ThesiscommitteePromotor

Prof.Dr.Ir.H.GruppenProfessorofFoodChemistryWageningenUniversityCo‐promotor

Prof.DrH.A.ScholsPersonalchairattheLaboratoryofFoodChemistryWageningenUniversityOthermembers

DrT.J.Foster,UniversityofNottingham,UnitedKingdomProf.DrJ.G.MJanssen,UniversityofAmsterdam,TheNetherlandsProf.DrJ.vanderOost,WageningenUniversity,TheNetherlandsProf.DrC.Sandström,SwedishUniversityofAgriculturalSciences,Uppsala,SwedenThisresearchwasconductedundertheauspicesoftheGraduateSchoolVLAG(AdvancedstudiesinFoodTechnology,Agrobiotechnology,NutritionandHealthSciences).

Enzymatic modification and characterization of xanthan

MarijnM.Kool

Thesissubmittedinfulfillmentoftherequirementsforthedegreeofdoctor

atWageningenUniversitybytheauthorityoftheRectorMagnificus

Prof.DrM.J.Kropff.inthepresenceofthe

ThesisCommitteeappointedbytheAcademicBoardtobedefendedinpubliconFriday7February2014at04.00p.m.intheAula.

MarijnM.KoolEnzymaticmodificationandcharacterizationofxanthan144pagesPhDthesis,WageningenUniversity,Wageningen,NL(2014)Withreferences,withsummariesinEnglishandDutchISBN:978‐94‐6173‐865‐3

ABSTRACT

In this thesis an enzymatic approach for the modification and characterization ofxanthans was introduced. Complete backbone degradation of xanthan by cellulaseswas obtained independent on the molar composition of a xanthan sample. It wasshown that only xanthan segments that occurred in a disordered xanthanconformationwere susceptible to enzymatic backbone degradation. HILIC‐ELSD‐MSanalysisrevealedthepresenceofsixdifferentxanthanrepeatingunits(RUs).AllRUsconsistedof the samepentasaccharide structure,withdifferent acetyl andpyruvatesubstitutionpatterns.InterestinglythepresenceofanacetylgroupattheO‐6positionoftheoutermannoseunitwasshown.Analysisof5xanthansamplesshowedthat5–19%ofallacetylgroupspresentarepositionedontheoutermannose.Furthermore,the relative abundanceof theRUspresent in xanthan samples canvary, evenwhentheirmolarcompositionsarethesame.

AnalysisofthetransitionalbehaviorofxanthanbasedontheenzymaticreleaseofthesixtypesofRUsshowedthattheacetylgroupsontheoutermannose,andnotontheinnermannose, aswas previously reported, are responsible for the stabilization ofxanthans conformation. Itwas proposed that acetylation of the outermannose alsodetermines the functional properties of a xanthan solution. Furthermore, it waspostulatedthat1)TheRUsthatareeitheracetylatedontheoutermannoseunitsorsolely acetylated on the inner mannose units are block wise distributed over thexanthan molecule. 2) Pyruvylated RUs and unsubstituted RUs are randomlydistributed.

Screening for xanthanmodifying enzymes resulted in the discovery of the first twoacetylesterasesbeingactivetowardsxanthan.AXE3,axylanacetylesteraseproducedbyMyceliophthorathermophilaC1,showedtobespecificfortheremovaloftheacetylgroupsattheinnermannoseunitandwasonlyactivetowardsthedisorderedxanthanconformation.YesY,apectinacetylesteraseproducedbyBacillussubtilis strain168,specificallyremoved theacetylgroupsat theoutermannoseunitsand itsactivity isnotinfluencedbyxanthan’sconformation.

Table of contents

Abstract

Chapter1 Generalintroduction 1

Chapter2 Theinfluenceoftheprimaryandsecondaryxanthanstructureontheenzymatichydrolysisofthexanthanbackbone

19

Chapter3 Comparisonofxanthansbytherelativeabundanceofitssixconstituentrepeatingunits

37

Chapter4 The influence of the six constituent xanthanrepeatingunitsontheorder‐disorderedtransitionofxanthan, based on the cellulase degradation ofdisorderedxanthansegments

55

Chapter5 Characterization of an acetyl esterase fromMyceliophthora thermophila C1 able to deacetylatexanthan

71

Chapter6 Characterization of an acetyl esterase fromBacillussubtilis strain 168 able to deacetylate the outermannoseofxanthan

87

Chapter7 Generaldiscussion 101

Summary 119

Samenvatting 123

Acknowledgements 127

Abouttheauthor 131

Chapter 1

General introduction

Chapter 1

2

PROJECT OUTLINE

Polysaccharides are commonly used in food and medical application due to theirphysico‐chemicalpropertiesand/orbiologicalfunctionalities.Polysaccharidescanbeextracted from plants and aquatic sources or produced through fungal or bacterialfermentation.1 The properties of polysaccharides often depend on their chemicalstructure.2Alteringthechemicalstructurecould,therefore,resultintheoptimizationofthebiopolymersfunctionality.Industrialmodificationsofpolysaccharidestructuresare mainly made chemically.3 However, the use of enzymes instead of chemicalswouldreducetheenvironmentalimpactoftheprocessandpossiblytheenergyinput.3Furthermore,theuseofenzymeshastheadvantagethatmodificationsmaybehighlyspecific, and could thus yield in the production of tailored polymers with novelproperties. To date, however, there is a lack in suitable enzymes to enable theenzymaticmodificationandoptimizationofpolysaccharides.TheresearchdescribedinthisthesiswascarriedoutwithintheFP7EU‐project‘Novelpolysaccharidemodifyingenzymestooptimizethepotentialofhydrocolloidsforfoodand medical applications’ (PolyModE). The general aim of this EU‐project was todevelopnewtechniquesfortheanalysisofthechemicalstructureofpolysaccharides.Furthermore, theproject aimed for theproduction and characterizationof enzymesthat specifically alter the polysaccharide structure in order to optimize both theproductionaswellastheindustrialapplicationofthesepolysaccharides.Xanthanwasone of the hydrocolloids studiedwithin the EU‐project and is the focus in the PhDresearchpresented.

XANTHAN APPLICATIONS

XanthanisanegativelychargedexocellularheteropolysaccharidethatisproducedbyXanthomonasspp.,4,5withamolecularmassrangingfrom1–7*106Dalton.6,7Xanthanis produced by aerobic batch fermentation and is isolated through ethanolprecipitation.Removal of the bacterial cells prior to precipitation results in awhiteendproduct.5,8When dispersed in aqueous solvents, xanthan exhibit a weak gel‐like behavior,4resultinginhighlyviscoussolutions.Uponshearthegelisdisrupted,whichresultsinadecreaseinviscosity.Theinitialviscositycanberecoveredeasilybytheremovalofthe shear and is rather stable over awide range of ionic strength, pH andpolymerconcentration.5, 9 As xanthan is soluble in both cold and hot water and due to itspseudoplastic behavior already obtained at low xanthan concentrations, xanthan issuitableformanyindustrialapplications(Table1.1.).5,10


3

Table 1.1. Industrial application of xanthan

Industry Function Possible applications

Oil recovery Reducer of water mobility Improve pumpability

Personal care Stabilizer Lotions , tooth pastes

Food Stabilizer and rheology modifier

Dairy

Milkshakes, whipped cream; custards and puddings; yoghurts; water gels; ice creams; sorbets

Bakery Pastry fillings; gluten free baking

Meat and fish

Canned food; spreads; frozen and ready‐to‐eat foods

Condiments Mayonnaise; salad dressings; sauces; soups; jams

Beverages Fruit juices; powdered beverages

Asthefunctionalpropertiesofxanthanarestableoverawidetemperaturerange,10,11applications in food are mainly within the field of stabilization of the texturalcharacteristicsofproducts,whichareexposedtoeithercoldorheat.Furthermore,theenhanced gelling properties obtained through interactions of xanthan withgalactomannansareusedfor:thepreparationofgelledorthickenedfoods;thecontrolofparticlesedimentationinjuicesanddrinks;andthepreventionofcrystalformationinfrozenproductssuchasicecreams.5

THE XANTHAN STRUCTURE

Xanthanisanextracellularpolysaccharideanditsstructureisbelievedtoberepetitive(Figure 1.1). It consists of a cellulose‐like β‐1,4‐glucan backbone with an α‐D‐mannose‐(21)‐β‐D‐glucuronic acid‐(41)‐β‐D‐mannose side chain on the O‐3positionof every secondglucoseunit.12, 13Approximately90%ofall innermannoseunitsisacetylatedattheO‐6positionandabout50%oftheoutermannoseunitscarrya4,6linkedpyruvateketal.12,14‐16Theexactdegreeofsubstitutionofxanthansamplesisknowntovarydependingonthefermentationconditions8,17andtheXanthomonasstrain used for the xanthan production.14, 15 The biosynthesis of xanthan inXanthomonascampestrishasbeenextensivelystudied.18‐23Aclusterof12‘gum‐genes’hasbeenidentified,whichencodesfortheenzymesinvolvedinthebiosynthesis,thepolymerizationandthesecretionofxanthan.Anoverviewofthemetabolicpathwayisgiven in Figure 1.2. The pentasaccharide repeating unit is assembled by sequentialaddition of individual monosaccharide residues. Each monosaccharide addition iscatalyzedbyaspecificglycosyltransferase(GT).

Chapter 1

4

Figure 1.1. The xanthan repeating unit as identified by Jansson et al.12

The mannose residues of the repeating unit are substituted by distinct enzymes:Acetyltransferase‐I (AT‐I) catalyzes the acetylation of the innermannose and ketalpyruvate transferase (KPT) catalyzes the pyruvylation of the outer mannose.Unexpectedly, a gene was found that encodes for a third decoration enzyme,acetyltransferase‐II (AT‐II) that catalyzes the acetylation of the outer mannose.Finally, thesynthesizedsubstitutedrepeatingunit isadded to thegrowingpolymer,whichissubsequentlysecreted.The presence of a ‘gum‐gene’ that encodes for the production of an enzyme that isinvolvedintheacetylationoftheoutermannoseunitindicatesthatvariationsinthexanthanstructureasproposedbyJanssonetal.12are likelytooccur.SuchvariationscouldbecontrolledbyspecificmutationsintheXanthomonasgenome,asthereisonespecific‘gum‐gene’fortheexpressionofeachenzymeinvolvedinthebiosynthesis.18,20,23Severalmutantstrainshavebeenproducedinordertocontrolthebiosynthesisofxanthan.23Basedonthemolecularcompositionofthexanthansobtained,thexanthanswerehypothesizedtovaryinthelengthandsubstitutionpatternoftheirsidechains.However,detailedstructuralanalysisof thesexanthans toconfirmthehypothesizedstructureshasneverbeenconducted.Therefore, theexactmodifications inducedbythemutations are uncertain.Moreover, geneticmodificationsmay go hand in handwith unintentional effects, as the modification at one level can cause unfavorableeffects at otherparts of thebiosynthesis aswell. Thismay results in e.g. an alteredmolecular mass and changes on molecular level.24 Controlled modification of thexanthanstructureusingmutantstrainsis,therefore,notfullypossible.The biosynthesis of xanthan can also be influenced by changing the fermentationconditionsandtherebystimulateorsuppresscertainenzymesinvolvedinthebiosyn‐

OO

OH

OHO

HO

O

O

O

OH

CH2OH CH2OH

O

HOOH

O

COO-

O

HO

OH O

OH

n

CH3COOCH2

-OOCO

OH3C

CH2


5

Figure 1.2. The biosynthesis of xanthan in Xanthomonas campestris and the enzymes involved as proposed by Katzen et al.18 p = lipid; GT = glucosyltransferase; AT = acetyltransferase; KPT = ketal pyruvate transferase; Glc = glucose; Man = mannose; GlcA = glucuronic acid; triangle = acetyl groups; diamond: pyruvate group. Dashed arrows indicate that the substitution level may vary.

thesis. A deficiency of oxygen during the fermentation is known to decrease thepyruvate content.17, 25 In contrast, a deficiency of nitrogen during the fermentationincreases the pyruvate content.8 Furthermore, it is known that the presence orabsenceofcitricacidandmagnesiuminthegrowthmediarespectively increaseanddecreasethepyruvatelevelsinthexanthanproduced.26‐28Theeffectoffermentationconditionsonthelengthofthexanthansidechainoronthelevelofacetylationhavenotbeenstudied.Studies on the xanthan biosynthesis clearly show that variations in the xanthanstructurecanoccurandthatnotonlythelevelofacetylationandpyruvylation,butalsothepositionofthesesubstituentscanvarybasedontheproductionconditions.Apartfrom the methylation analysis performed by Jansson et al.12 only one other studyfocusedontheactualanalyticalcharacterizationofthexanthanstructure.Inthisstudythestructurewasdescribedusingglycosidiclinkageanalysis.29anditwasconfirmedthattheoutermannosecanbeacetylatedontheO‐6position.Theratiobetweentheacetyl groups on the inner and outer mannose units was also determined for theentirepolymer.However,structuralanalysisontherepeatingunit(RU)levelandtheexactdistributionofdifferentxanthansidechainsoverthexanthanbackbonehasnotbeenstudiedyet.

Chapter 1

6

CONFORMATION AND PROPERTIES OF XANTHAN IN SOLUTION

Due to the polyelectrolyte nature of xanthanmolecules, xanthan is very soluble inboth cold and hot water. Xanthan solutions exhibit a pseudoplastic, shear thinningbehavior, which depends on temperature, salt concentration, pH and xanthanconcentration.30‐32 This unique solution behavior is associated with theconformationalbehaviorofxanthaninsolution(Figure1.3.).Nativexanthan,asproducedbyXanthomonascampestris,existsinadouble‐strandedhelicalconformationwithanorder‐disordertransitionuponchangesintemperatureand/orionicstrength.33‐35The‘renaturingofxanthan’,aprocessinducedbycoolingorby an increase in salt concentration, results in the recovering of an orderedconformation. Renatured xanthan, however, does not exhibit exactly the sameconformation and rheological properties as the native xanthan.7, 35‐38 The order‐disorder transition of the native xanthan can thus be considered as non‐reversible.Renatured xanthan also exists in a helical conformation and has an order‐disordertransition, that is, in contrast tonative xanthan, reversible.30, 31, 39Awidevarietyofphysical and physicochemical techniques have been used to study the exactconformationofrenaturedxanthan.However,thenatureofthisconformationremainscontroversial and several models have been proposed (Figure 1.3c.). Some studiesproposed a double ormultiple stranded double helical structure.40‐42 Other studiesshowed evidence for a single stranded double helix, in which intramolecularinteractionsresultsintheformationofhairpinloops.7,32,43,44Morrisandco‐workers31argued that renatured xanthan has a single stranded helix that is stabilized by thealignmentofsidechainsalongthebackbone,aswasconfirmedbyothers.32,39Although the unique solution properties of xanthans are believed to be stronglyrelated to the order‐disorder transition, this transition alone is not sufficient toexplainthetotalrheologicalbehaviorofaxanthansolution.9,45Twomodels have been proposed to explain theweak gel‐like behavior of xanthan.Bothmodelsproposetheformationofanetworkofxanthanmolecules,whichiseasilydisrupteduponshear,causingxanthanspseudoplasticbehavior.35,45Thenatureoftheintermolecularinteractionsinvolvedinthenetworkformation,however,isuncertain.ThemodelproposedbyRoss‐Murphy45believesthatanetworkisformedthroughtheside‐by‐side alignment of ordered xanthan molecules. This network is stabilizedthroughnon‐covalentinteractions(Figure1.3h.).Consideringthataxanthanhelixcanpartlydissociateatmultiplepositionswithinahelix (Figure1.3f.), variations in thismodel areproposed inwhich theorderedpartsofonexanthanmoleculealignwithorderedpartsofmultipleotherxanthanmolecules(Figure1.3g.).32,46


7

Figure 1.3. Models proposed for the conformational behavior of xanthan in solution. a) native ordered double stranded double helix; b) disordered xanthan strains; c) proposed conformations for renatured xanthan; d) partially dissociated native strains;41 e) network formation by the association of disordered xanthan segments;35 f) partially dissociated renatured strains; g) network formation by stacking of ordered xanthan segments;32, 46 h) network formation by stacking of completely ordered xanthan molecules.45

TheothermodelelaboratesonthedissociationbehaviorasdescribedbyLiuetal.,41inwhich double stranded xanthan gradually unfolds from the end points of the helix(Figure 1.3d.). Based on this dissociation behavior, it has been argued35 that thedissociated parts of one helix interact with dissociated parts of one ormore otherdouble stranded helices upon renaturing, leading to the formation of a network ofdoublestrandedhelices(Figure1.3e.).

Chapter 1

8

Models describing xanthan‐galactomannan interactions

Oneofthemaincharacteristicofxanthanusedinindustryistheabilitytoformstronggelswithgluco‐andgalactomannans(GM).TheinteractionbetweenxanthanandGMsis believed to strongly depend on the conformation of xanthan. However, as theconformationofxanthanispoorlyunderstood,theexactnatureoftheseinteractionsalso remains controversial. Three different models for the interaction have beenproposed.AnoverviewofthesemodelsisgiveninFigure1.4.TheUnilevermodel,showsthatonlyorderedxanthanstructuresinteractwithGM.31,47Accordingtothismodel themannanbackbonehastobeavailable for interaction,asinteractionsonlyoccurwithsmoothmannanregionsorwithmannanregionswhichare substituted on every other mannose unit.48 In contrast, the Norwich modelindicates thatonlydisorderedxanthan fragments can interactwith thebackboneofGM.49Athirdmodel,theSilsoemodel,50suggeststhatthestrongxanthan‐GMgelsarearesult of homotypic (xanthan‐xanthan) and heterotypic interactions (xanthan‐galactomannan) with heterotypic junction zones formed by disordered xanthanfragments.

Figure 1.4. Proposed models for the xanthan‐galactomannan interactions. A) Unilever model, B) Norwich model, C) Silsoe model.

Influence of the primary xanthan structure on xanthans solution properties

Although multiple models for xanthans solution properties exist, studies areconsistentonthefactorsinfluencingtheseproperties.Theconformationalbehaviorisinfluencedbysolventconditions,wherethedisorderedconformationisfavoredwithincreasing temperature, decreasing ionic strength and/or increasing pH.30, 31, 51Furthermore, the primary xanthan structure influences the transitional behavior,where thepresenceof acetyl groups stabilizes theordered structure11, 52, 53 and thepresenceofpyruvategroupsresultsinamoreflexible,disorderedconformation.54,55


9

Theinfluenceoftheprimarystructureontherheologicalbehaviorofxanthanhasalsobeenstudiedextensively.Theremovalofacetylgroupgiverisetoxanthansolutionswith increasedviscosityanda lower sensitivity to changes inpH.11, 23 Furthermore,stronger interactions gels are observed using acetyl free xanthan insteadof normalxanthan.56 In contrast, the removal of pyruvate groups reduces the viscosity of axanthan solution, increases the sensitivity to changes in pH and decreases theinfluenceofsaltsontheviscosity.54,57Theremovalofpyruvategroupsdidnotshowacleareffectontheinteractionswithgalactomannans.56Studiesontheinfluenceofthelengthofthesidechainonxanthansolutionpropertiesshowedthattheremovaloftheouter mannose decreased the viscosity of a xanthan solution and that weakerinteraction gels are obtained.23, 56, 58 The additional removal of the glucuronic acidresultedinanincreasedviscosity.23,58As described previously, the xanthan conformation is believed to be important forxanthanssolutionproperties.Howevertheconformationissignificantlyinfluencedbyxanthansprimarystructure.Observedinfluencesoftheprimarystructureonsolutionproperties could, therefore, origin from the induced changes in the xanthanconformation.

ACCEPTED CHEMICAL XANTHAN STRUCTURE DOES NOT EXPLAIN

PHYSICAL OBSERVATIONS

Generally, it isassumedthatthestructureproposedbyJanssonetal. (Figure1.1.) isthe idealxanthanstructurewithonlyminorvariations in thedegreeofsubstitution.However, the variations observed in acetyl and pyruvate level, together withinformationonxanthansbiosynthesisclearlypointtopossiblevariationsinsidechaindecoration. If this indeed is the case, varieties in the distribution of xanthansubstituentsmaybeoneofthereasonsthatthewidevarietyinmodelsforxanthanssolution behavior still does not fully explain xanthans functionality. The hypothesisthat different xanthan repeating units exist is strengthened by the observation thatthe functionality of other carbohydrates strongly depends on the distribution ofsubstituentsand/orsidechains.2Unambiguousinformationonthexanthanstructureis thusneeded tohelpunderstanding the structure‐function relationshipof xanthanbetter,howeveragoodanalyticalapproachtostudythexanthanstructureislacking.Several methods are described in literature to study the chemical structure ofpolysaccharides. The total sugar composition can be determined after completechemicalhydrolysisofcarbohydrates.However,nostructuralinformationonpolymerlevel can be obtained from such analysis. Analysis on polymeric structures can beconductedusingNMR.Adrawback isthattheanalysisofviscoussamples isdifficult

Chapter 1

10

andusually leads tobroadnotwell resolvedsignals.59Anotherapproach is tostudydiagnosticoligosaccharidesreleaseduponhydrolysis,inordertorevealthechemicalstructure. Xanthan oligosaccharides can be obtained through chemicaldepolymerization,oxidative‐reductivedepolymerizationorsonication.6,60,61However,randommodificationsinthesidechainscanbeapparentusingsuchdepolymerizationtechniques.Enzymaticdepolymerizationof thexanthanbackboneas investigatedbyRinaudo&Milas62is,therefore,moresuitablefortheproductionofdiagnosticxanthanoligosaccharides.Anoverviewof xanthandegradingandmodifyingenzymeswill begivenbelow.

XANTHANASES

Enzymatic backbone degradation

Xanthanmodifying or degrading enzymes can be obtained from xanthan degradingmicroorganisms,whencultivatedinagrowthmediawithxanthanascarbonsource.63‐66Severalstudiesusedmixedorpurifiedculturesfortheproductionofxanthanases.63‐65, 67‐70 Most of the cultures that could degrade xanthan were rich in Bacillus spp.Generally,theseculturesproducedamixtureofenzymesthatsynergisticallydegradethexanthanpolymer.Theanalysisofthedegradationproductsshowedthatsidechainmodifications are necessary prior to backbone degradation, as will be furtherdiscussedlater.Theoligosaccharidesproducedare,therefore,notfullyrepresentativeforthecompletexanthanpolymer.Other studies focused on the degradation of the xanthan backbone using purecellulasesandshowedthatfungalcellulasescanpartlydegradexanthan.Thiscanonlybedoneunderaqueous conditions inwhichxanthanappears inapartlydisorderedconformation.62, 71‐73Thexanthandegradation inmoststudieswasmonitoredbythedecreaseinviscosityonly,anddidnotfocusontheelucidationoftheprimaryxanthanstructure.62, 73 However, some studies followed the degradation by gel permeationchromatography.68, 71 It was shown that high molecular weight products, rich insubstituents,remainedattheendpointoftheenzymatichydrolysis.71,72Itwasarguedthatsubstituents in thexanthansidechainsmighthinder thecellulasesandpreventthe complete hydrolysis of the xanthan backbone into oligosaccharides. As xanthanwasonlypartiallyhydrolyzed,theoligosaccharidesproducedareconsideredtobenotrepresentativeforthecompletexanthanpolymer.Inordertoaccuratelycomparethechemical structure of different xanthans, an enzyme system is necessary which iscapableofcompletebackbonehydrolysisindependentonthesubstituentspresent.


11

Enzymatic modification of the xanthan side chains

Acombinationof specific sidechainmodifications followedbystructuralanalysisofthe modified xanthan could be a good approach for the further elucidation of thexanthan structure. Several techniques to modify the xanthan structure are known.Various chemical treatments can be used for the production of pyruvate and/oracetyl‐free xanthan or to produce the xanthan polytetramer in which the outermannoseunithasbeenremoved.53,74‐76Xanthanswithvaryingstructurescanalsobeobtainedbyalteringthexanthanbiosynthesis,asdescribedpreviously.However,noneofthesemethodsresultsinacontrolledmodificationofthexanthansidechain,whichis necessary for the further elucidation of the xanthan structure. Enzymaticmodificationofthexanthansidechainscouldbeasuperiormethod,as itprovidesacontrolledremovaloftargetgroups.

Xanthan lyases

Todatexanthan lyasesare theonlywell characterizedxanthanmodifyingenzymes,whichareactive towards thexanthanpolymer.Theyactexolyticallyonthexanthansidechainandliberatetheoutermannosethroughβ‐elimination.Inthatperspectivexanthan lyasesareratherpeculiarasmostotherpolysaccharide lyasesendolyticallycleave the glycosidic backbone of a polysaccharide.77, 78 Based on the amino acidsequence, xanthan lyases are classified intoCAZypolysaccharide lyase family 8,79‐81which contains lyases that are active towards hyaluronate and chondroitin aswell(www.cazy.org).Xanthanlyasescanbedividedintotwoclasses:1)pyruvatespecificlyases,whichneedthepresenceofpyruvategroupstobeactiveand2)xanthanlyaseswhich are active independent on the degree of acetylation and pyruvylation.24, 82Several studieshavedescribed lyasesbelonging to the first class,64, 77, 83, 84 ofwhichtwohavebeenpurifiedandcharacterized.79,85Notype2lyasehasbeenpurifiedsofar.Theexactsubstratespecificityoftype2lyases,therefore,remainsuncertain.Asthecharacterizedxanthanlyasesarespecificfortheremovalofpyruvylatedoutermannose units, these enzymes could be used for the elucidation of the distributionpatternofthepyruvategroupsalongthebackbone.

Putative xanthan side chain modifying enzymes

Althoughnoxanthanmodifyingenzymes,other thanxanthan lyases,havebeen fullycharacterized, some studies reported on the expression of other putative xanthanmodifying enzymes by micro‐organisms, when cultivated in a growth media withxanthanascarbonsource.

Chapter 1

12

Figure 1.5. Complete enzymatic degradation of xanthan by Bacillus sp. strain GL1 as described by Nankai et al.63 Glc = glucose; Man = mannose; GlcA = glucuronic acid; U = unsaturated glucuronic acid triangle = acetyl groups; diamond: pyruvate group.

The complete degradation of xanthan by Bacillus sp. strain GL1 and the enzymesinvolved have been described by Nankai et al.63 (Figure 1.5.). A pyruvate specificxanthan lyase and a β‐D‐glucanase are produced extracellular. Removal of theterminalmannosethroughβ‐eliminationshowedtobenecessarybeforethebackbonecan be hydrolyzed by β‐D‐glucanase.63, 66 The tetramers obtained are broughtintracellulartobefurtherdegradedintomonosaccharides.Incubationofthexanthanpolymerwiththeintracellularenzymefractionshowedthatnoneofthesesidechainmodifyingenzymesareactivetowardsthexanthanpolymer.66,86Anotherstudyreportedonthesecretionofseveralxanthanmodifyingenzymesduringthe growth of Paenibacillus alginolyticus XL1 on xanthan media (Figure 1.6.).24Analysis of the molecular weight distribution showed that high molecular weight(HMW)materialremainedinthexanthanmediaaftergrowth.ThisHMWmaterialhadlimited solubility and showed interaction with Congo red, which interacts with β‐glucans, indicating that the cellulosic backbonewas still intact. The enzymeswere,therefore,concludedtobeactivetowardstheintactxanthanpolymer.Asthecompleteculturebrothshowedlyaseactivitytowardsbothintactxanthanaswellaspyruvate‐freexanthan,itwashypothesizedthatnexttothepurifiedpyruvate‐specificlyase84,asecond (type 2) lyase was present. Furthermore, a deacetylase and an uronic acidreleasing enzymewere secreted by P. alginolyticusXL1. Both enzymeswere partlypurified. The uronic acid releasing enzyme was only active towards lyase‐treatedxanthanand is, therefore, similar to theunsaturatedglucuronylhydrolaseproducedbyBacillussp.strainGL1.86However,thisenzymeisproducedextracellularanddoesnot require backbone degradation to be active. The deacetylase was not furthercharacterized.Hence,noenzymesarereadilyavailableforthestructuralanalysisandmodificationofxanthan.

Xanthan lyase

n n

β-D-glucanase β-D-glucosidase

Unsatutatedglucuronylhydrolase

α-D-mannosidase

Glc Glc

Man

GlcA

Man

U


13

Figure 1.6. Xanthan modifying enzymes produced by Paenibacillus alginolyticus XL1 as proposed by Ruijssenaars.24 Glc = glucose; Man = mannose; GlcA = glucuronic acid; U = unsaturated glucuronic acid triangle = acetyl groups; diamond: pyruvate group.

AIM AND OUTLINE OF THE THESIS

Thesubstitutionpatternofseveralpolysaccharidesshowedtobeimportantfortheirfunctionalproperties.Thewidevarietyinmodelstodescribethefunctionalpropertiesofxanthanwastherefore,hypothesizedtobetheresultofamorecomplexchemicalxanthan structure than generally assumed, especially regarding the substitutionpattern of the substituents. The aim of the research described in this thesis was,therefore,todevelopanenzymaticmethodtocharacterizeandcomparetheprimarystructureofdifferentxanthans.Furthermore,enzymesthatsolelymodifythexanthansidechainweresearchedfortoenablethecontrolledmodificationofxanthan.Inchapter2theinfluenceofboththeprimaryaswellasthesecondarystructureofxanthanontheenzymaticdegradationofthebackbonebycellulaseswasinvestigated.The structure of oligosaccharides obtained from different xanthan samples afterenzymatichydrolysisofthebackbonewereelucidatedusingHPAECandLC‐ELSD‐MSnandarediscussedinchapter3.Inchapter4thecorrelationbetweenthetransitionalbehaviorofxanthanandthepositionofsubstituentswithinthexanthansidechainisdiscussed,basedonthechemicalstructureofdiagnosticoligosaccharidesreleasedbycellulases at different levels of disordered conformation. From the correlationsobserved,informationonthedistributionpatternofthesubstituentsoverthexanthanbackbonewasobtainedaswell.Chapters5and6reportthecharacterizationoftwodifferent acetyl esterases, fromMyceliophthora thermophila C1 and Bacillus subtilisstrain 168, which partly deacetylate xanthan in a different manner. Chapter 7discusses the most important findings of this study and their impact for futureresearchonthestructure‐functionrelationshipofxanthan.

Deacetylase

Xanthan lyase type 1Xanthan lyase type 2

Xanthan lyase type 2

Uronic acid releasing enzyme

Glc Glc

Man

GlcA

Man

U

Chapter 1

14

REFERENCES

[1] Williams PA, Phillips GO. Introduction to food hydrocolloids. In: Phillips GO, Williams PA, eds.Handbookofhydrocolloids.BocaRaton(FL),USA:CRCPress,2000;1‐19.

[2] BeMiller JN,WhistlerRL.Carbohydrates. In:FennemaOR,ed.FoodChem.3edn.NewYork(NY),USA:MarcelDekker,Inc.,1996;205‐24.

[3] Cheng HN, Gu Q‐M. Enzyme‐catalyzedmodifications of polysaccharides and polyethylene glycol.Polymers.2012;4:1311‐30.

[4] JeanesA,PittsleyJE,SentiFR.PolysaccharideB‐1459:Anewhydrocolloidpolyelectrolyteproducedfromglucosebybacterialfermentation.JApplPolymSci.1961;V:519‐26.

[5] Garcı́a‐OchoaF,SantosVE,CasasJA,GómezE.Xanthangum:production,recovery,andproperties.BiotechnolAdv.2000;18:549‐79.

[6] MilasM,RinaudoM,TinlandB.Comparativedepolymerisationof xanthangumbyultrasonicandenzymatictreatments‐Rheologicalandstructuralproperties.CarbohydrPolym.1986;6:95‐107.

[7] Milas M, Reed WF, Printz S. Conformations and flexibility of native and re‐natured xanthan inaqueoussolutions.IntJBiolMacromol.1996;18:211‐21.

[8] FloresCandia JL,DeckwerWD.Effectof thenitrogensourceonpyruvatecontentandrheologicalpropertiesofxanthan.BiotechnolProgr.1999;15:446‐52.

[9] Frangou SA, Morris ER, Rees DA, Richardson RK, Ross‐Murphy SB. Molecular origin of xanthansolutionrheology:Effectofureaonchainconformationandinteraction.JPolymSci:PolymLettEd.1982;20:531‐8.

[10] SwornG.XanthanGum.In:ImesonA.eds.FoodStabilisers,ThickenersandGellingAgents.Oxford,UK:Wiley‐Blackwellpublishing,2009;325‐42.

[11] MorrisonNA,ClarkR,TalashekT,YuanCR.Newformsofxanthangumwithenhancedproperties.In:WilliamsPA,PhillipsGO,eds.GumsandStabilisers forFood Industry12.Cambridge,UK:TheRoyalSocietyofChemistry,2004;124‐30.

[12] Jansson PE, Kenne L, Lindberg B. Structure of extracellular polysaccharide from Xanthomonascampestris.CarbohydrRes.1975;45:275‐82.

[13] Melton LD,Mindt L, Rees DA, Sanderson GR. Covalent structures of extracellular polysaccharidefromXanthomonascampestris‐Evidencefrompartialhydrolysisstudies.CarbohydrRes.1976;46:245‐57.

[14] Cadmus MC, Rogovin SP, Burton KA, Pittsley JE, Knutson CA, Jeanes A. Colonial variation inXanthomonas campestrisNRRL B‐1459 and charaterization of the polysaccharide from a variantstrain.CanJMicrobiol.1976;22:942‐8.

[15] SutherlandIW.Xanthomonaspolysaccharides‐Improvedmethodsfortheircomparison.CarbohydrPolym.1981;1:107‐15.

[16] Orentas DG, Sloneker JH, Jeanes A. Pyruvic acid content and constituent sugar of exocellularpolysaccharides fromdifferentspeciesof thegenusXanthomonas.Can JMicrobiol.1963;9:427‐30.

[17] Peters HU, Suh IS, Schumpe A, Deckwer WD. The pyruvate content of xanthan polysaccharideproducedunderoxygenlimitation.BiotechnolLett.1993;15:565‐6.

[18] KatzenF,FerreiroDU,OddoCG,etal.Xanthomonascampestrispv.campestrisgummutants;Effectonxanthanbiosynthesisandplantvirulence.JBacteriol.1998;180:1607‐17.

[19] Vojnov AA, Slater H, M.J. D, Maxwell Dow J. Expression of the gum operon directing xanthanbiosynthesis inXanthomonascampestrisand its regulation inplants.MolPlantMicrobe In.2001;14:768‐74.

[20] Vorhölter F‐J, Schneiker S, Goesmann A, et al. The genome of Xanthomonas campestris pv.campestris B100 and its use for the reconstruction of metabolic pathways involved in xanthanbiosynthesis.JBiotechnol.2008;134:33‐45.


15

[21] HardingNE,ClearyJM,CabañasDK,RosenDK,KangKS.Geneticandphysicalanalysisofaclusterofgenes essential for xanthan gumbiosynthesis inXanthomonas campestris. J Bacteriol. 1987; 169:2854‐61.

[22] KimS‐Y,Kim J‐G, LeeB‐M, Cho J‐Y.Mutational analysis of the gumgene cluster required for thexanthanbiosynthesisinXanthomonasoryzaepvoryzae.BiotechnolLett.2009;31:265‐70.

[23] HasslerRA,DohertyDH.Geneticengineeringofpolysaccharidestructure:ProductionofvariantsofxanthanguminXanthomonascampestris.BiotechnolProgr.1990;6:182‐7.

[24] RuijssenaarsHJ.Enzymaticmodificationofbacterialexopolysaccharides ‐Xanthan lyaseasa toolfor structural and functional modification of xanthan. PhD dissertation, Wageningen Univeristy,Wageningen,TheNetherlands,2001.

[25] Flores F, Torres LG, Galindo E. Effect of the dissolved oxygen tension during cultivation of X.campestrisontheproductionandqualityofxanthangum.JBiotechnol.1994;34:165‐73.

[26] Davidson IW. Production of polysaccharide by Xanthomonas campestris in continuous culture.FEMSMicrobiolLett.1978;3:347‐9.

[27] Cadmus MC, Knutson CA, Lagoda AA, Pittsley JE, Burton KA. Synthetic media for production ofqualityxanthangumin20literfermentors.BiotechnolBioeng.1978;20:1003‐14.

[28] JanaAK,GhoshP.Effectofcitricacidonthebiosynthesisandcompositionofxanthan. JGenApplMicrobiol.1999;45:115‐20.

[29] Stankowski JD, Mueller BE, Zeller SG. Location of a second O‐acetyl group in xanthan gum byreductive‐cleavagemethod.CarbohydrRes.1993;241:321‐6.

[30] Holzwarth G. Conformation of the extracellular polysaccharide of Xanthomonas campestris.Biochemistry.1976;15:4333‐9.

[31] MorrisER,ReesDA,YoungG.Orderdisordertransitionforabacterialpolysaccharideinsolution.AroleforpolysaccharideconformationinrecognitionbetweenXanthomonaspathogenanditsplanthost.JMolBiol.1977;110:1‐16.

[32] Norton IT, Goodall DM, Frangou SA, Morris ER, Rees DA. Mechanism and dynamics ofconformationalorderinginxanthanpolysaccharide.JMolBiol.1984;175:371‐94.

[33] BezemerL,UbbinkJB,KookerdeJA,KuilME,LeyteJC.Ontheconformationaltransitionsofnativexanthan.Macromolecules.1993;26:6436‐46.

[34] LiuW, Norisuye T. Order‐disorder conformation change of xanthan in 0.01 M aqueous sodium‐chloride‐Dimensionalbehaviour.Biopolymers.1988;27:1641‐54.

[35] MatsudaY,BiyajimaY, SatoT. Thermal denaturation, renaturation, and aggregationof a double‐helicalpolysaccharidexanthaninaqueoussolution.PolymJ.2009;41:526‐32.

[36] Capron I,BrigandG,MullerG.Thermaldenaturationandrenaturationofa fermentationbrothofxanthan:Rheologicalconsequences.IntJBiolMacromol.1998;23:215‐25.

[37] CalletF,MilasM,RinaudoM.Influenceofacetylandpyruvatecontentonrheologyicalpropertiesofxanthanindilutesolution.IntJBiolMacromol.1987;9:291‐3.

[38] OviattJrHW,BrantDA.Viscoelasticbehaviorofthermallytreatedaqueousxanthansolutionsinthesemidiluteconcentrationregime.Macromolecules.1994;27:2402‐8.

[39] Milas M, Rinaudo M. Conformational investigation on the bacterial polysaccharide xanthan.CarbohydrRes.1979;76:189‐96.

[40] Holzwarth G, Prestridge EB. Multistranded helix in xanthan polysaccharide. Science. 1977; 197:757‐9.

[41] LiuW,NorisuyeT.Thermallyinducedconformationchangeofxanthan‐Interpretationofviscositybehaviourin0.01Maqueoussodium‐chloride.IntJBiolMacromol.1988;10:44‐50.

[42] SatoT,KojimaS,NorisuyeT,FujitaH.Double‐strandedhelixofxanthanindilutesolution:furtherevidence.PolymJ.1984;16:423‐9.

[43] Kawakami K, Okabe Y, Norisuye T. Dissociation of dimerized xanthan in aqueous solution.CarbohydrPolym.1991;14:189‐203.

Chapter 1

16

[44] Capron I, Brigand G, Muller G. About the native and renatured conformation of xanthanexopolysaccharide.PolymJ.1997;38:5289‐95.

[45] Ross‐MurphySB,MorrisVJ,MorrisER.Molecularviscoelasticityofxanthanpolysaccharide.FaradaySymposiaoftheChemicalSociety.1983;18:115‐29.

[46] Morris VJ. Biotechnically produced carbohydrates with functional properties for use in foodsystems.FoodBiotechnol.1990;4:45‐57.

[47] Dea ICM, Morris ER, Rees DA, Welsh EJ, Barnes HA, Price J. Associations of like and unlikepolysaccharides: Mechanism and specificity in galactomannans, interacting bacterialpolysaccharides,andrelatedsystems.CarbohydrRes.1977;57:249‐72.

[48] McClearyBB.Enzymichydrolysis,finestructure,andgellinginteractionoflegume‐seedD‐galacto‐D‐mannans.CarbohydrRes.1979;71:205‐30.

[49] Cairns P, Miles MJ, Morris VJ. Intermolecular binding of xanthan gum and carob gum. NatBiotechnol.1986;322:89‐90.

[50] Foster TJ, Wolf B. Hydrocolloid gums ‐ Their role and interactions in foods. In: Norton IT,Spyropoulos F, Cox P, eds. Practical food rheology: An interpretitive approach. Oxford, UK:BlackwellPublishing,2010;61‐84.

[51] Bejenariu A, PopaM, Picton L, Cerf DL. Effect of concentration, pH and temperature on xanthanconformation:Apreliminarystudybeforecrosslinking.RevRoumChim.2010;55:147‐52.

[52] Pelletier E, Viebke C, Meadows J, Williams PA. A rheological study of the order‐disorderconformationaltransitionofxanthangum.Biopolymers.2001;59:339‐46.

[53] Tako M, Nakamura S. Rheology properties of deacetylated xanthan in aqueous‐media. Agr BiolChem.1984;48:2987‐93.

[54] SandfordPA,PittsleyJE,KnutsonCA,CadmusMC,WatsonPR,JeanesA.VariationinXanthomonascampestrisNRRLB‐1459;Characterisationofxanthansamplesofdifferentpyruvicacidcontent.In:Sandford PA, Laskin A, eds. ExtracellularMicrobial Polysaccharides.Washington (DC), USA: ACS,1977;192‐210.

[55] Shatwell KP, Sutherland IW, Dea ICM, Ross‐Murphy SB. The influence of acetyl and pyruvatesubstituentsonthehelix‐coiltransitionbehaviourofxanthan.CarbohydrRes.1990;206:87‐103.

[56] ShatwellKP,Sutherland IW,Ross‐MurphySB,Dea ICM. Influenceof theacetyl substituenton theinteractionofxanthanwithplantpolysaccharides‐I.Xanthan‐locustbeangumsystems.CarbohydrPolym.1990;14:29‐51.

[57] SmithCJH,SymesKC,LawsonCJ,MorrisER.Theeffectofpyruvateonxanthansolutionproperties.CarbohydrPolym.1984;4:153‐7.

[58] TaitMI, Sutherland IW. Synthesis and properties of amutant type of xanthan. J of ApplMicrob.1989;66:457‐60.

[59] LeesAJ, StraughanBP,GardinerDJ. AnNMRand IR studyof ion—formamide interactions. JMolStruct.1979;54:37‐47.

[60] ChristensenBE,SmidsrødO.Hydrolysisofxanthanindiluteacid:Effectsonchemicalcomposition,conformation,andintrinsicviscosity.CarbohydrRes.1991;214:55‐69.

[61] Christensen BE, Myhr MH, Smidsrød O. Degradation of double‐stranded xanthan by hydrogenperoxideinthepresenceofferrousions:Comparisontoacidhydrolysis.CarbohydrRes.1996;280:85‐99.

[62] RinaudoM,MilasM.Enzymic‐hydrolysisofthebacterialpolysaccharidexanthanbycellulase.IntJBiolMacromol.1980;2:45‐8.

[63] Nankai H, Hashimoto W, Miki H, Kawai S, Murata K. Microbial system for polysaccharidedepolymerization: enzymatic route for xanthan depolymerization byBacillus sp strainGL1. ApplEnvironMicrob.1999;65:2520‐6.

[64] CadmusMC, SlodkiME,Nicholson JJ.High‐temperature, salt‐tolerant xanthanase. J IndMicrobiol.1989;4:127‐33.


17

[65] HouCT,BarnabeN,GreaneyK.Biodegradationofxanthanbysalt‐tolerantaerobicmicroorganisms.JIndMicrobiol.1986;1:31‐7.

[66] HashimotoW,MommaK,MikiH,MishimaY,KobayashiE,MiyakeO,KawaiS.NankaiH,MikamiB.Murata K. Enzymatic and genetic bases on assimilation, depolymerization, and transport ofheteropolysaccharidesinbacteria.JBiosciBioeng.1999;87:123‐36.

[67] Lesley SM. Degradation of polysaccharide of Xanthomonas phaseoli by an extracellular bacterialenzyme.CanJMicrobiol.1961;7:815‐&.

[68] SutherlandIW.AnenzymesystemhydrolysingthepolysaccharidesofXanthomonasspecies.JApplBacteriol.1982;53:385‐93.

[69] CadmusMC, Jackson LK, Burton KA, Plattner RD, SlodkiME. Biodegradation of Xanthan gum byBacillusspp.ApplEnvironMicrob.1982;44:5‐11.

[70] AhlgrenJA.Purificationandpropertiesofaxanthandepolymerasefromaheat‐stablesalt‐tolerantbacterialconsortium.JIndMicrobiol.1993;12:87‐92.

[71] Cheetham NWH, Mashimba ENM. Characterisation of some enzymatic‐hydrolysis products ofxanthan.CarbohydrPolym.1991;15:195‐206.

[72] Sutherland IW.Hydrolysis of unordered xanthan in solutionby fungal cellulases. CarbohydrRes.1984;131:93‐104.

[73] Christensen BE, Smidsrød O. Dependence of the content of substituted (cellulosic) regions inprehydrolysedxanthansontherateofhydrolysisbyTrichodermareeseiendoglucanase. Int JBiolMacromol.1996;18:93‐9.

[74] Pinto EP, Furlan L, Vendruscolo CT. Chemical deacetylation natural xanthan (Jungbunzlauer®).Polimeros.2011;21:47‐52.

[75] Bradshaw IJ,NisbetBA,KerrMH,Sutherland IW.Modifiedxanthan‐itspreparationandviscosity.CarbohydrPolym.1983;3:23‐38.

[76] Tait MI, Sutherland IW, Clarke‐Sturman AJ. Acid hydrolysis and high‐performance liquidchromatographyofxanthan.CarbohydrPolym.1990;13:133‐48.

[77] HashimotoW,Miki H, Tsuchiya N, Nankai H,Murata K. Xanthan lyase ofBacillus sp. strain GL1liberates pyruvylatedmannose from xanthan side chains. Appl and EnvironmMicrob. 1998; 64:3765‐8.

[78] SutherlandIW.Polysaccharidelyases.FEMSMicrobReviews.1995;16:323‐47.

[79] Ruijssenaars HJ, Hartmans S, Verdoes JC. Novel gene encoding xanthan lyase of PaenibacillusalginolyticusstrainXL‐1.ApplandEnvironmMicrob.2000;66:3945‐50.

[80] HashimotoW,NankaiH,MikamiB,MurataK.Crystal structureofBacillus sp.GL1xanthan lyase,whichactsonthesidechainsofxanthan.JBiolChem.2003;278:7663‐73.

[81] Bourne Y, Henrissat B. Glycoside hydrolases and glycosyltransferases: families and functionalmodules.CurrOpinStrucBiol.2001;11:593‐600.

[82] SutherlandIW.Xanthanlyases‐Novelenzymesfoundinvariousbacterialspecies.JGenMicrobiol.1987;133:3129‐34.

[83] AhlgrenJA.Purificationandcharacterizationofapyruvated‐mannose‐specificxanthanlyase fromheat‐stable,salt‐tolerantbacteria.ApplEnvironMicrob.1991;57:2523‐8.

[84] RuijssenaarsHJ,deBontJAM,HartmansS.Apyruvatedmannose‐specificxanthanlyaseinvolvedinxanthandegradationbyPaenibacillusalginolyticusXL‐1.ApplEnvironMicrob.1999;65:2446‐52.

[85] HashimotoW,Miki H, Tsuchiya N, Nankai H,Murata K. Polysaccharide lyase:Molecular cloning,sequencing,andoverexpressionofthexanthanlyasegeneofBacillussp.strainGL1.ApplBiochemBiotechn.2001;67:713‐20.

[86] HashimotoW,KobayashiE,NankaiH,SatoN,MiyaT,KawaiS,MurataK.Unsaturatedglucuronylhydrolase of Bacillus sp. GL1: Novel enzyme prerequisite for metabolism of unsaturatedoligosaccharidesproducedbypolysaccharidelyases.ArchBiochemBiophys.1999;368:367‐74.

Chapter 1

18

Published as: Marijn M. Kool, Henk A. Schols, Roy J. B. M. Delahaije, Graham Sworn, Peter A. Wierenga and Harry Gruppen. Carbohydrate Polymers. 2013, 97(2), 368‐375.

Chapter 2

The influence of the primary and secondary xanthan structure on the enzymatic hydrolysis of the xanthan backbone

ABSTRACT

Differentlymodified xanthans, varying in degree of acetylation and/or pyruvylationwere incubatedwith theexperimental cellulasemixtureC1‐G1 fromMyceliophthorathermophilaC1.Theionicstrengthand/ortemperatureofthexanthansolutionswerevaried, to obtain different xanthan conformations. The exact conformation at theselected incubation conditions was determined by circular dichroism. The xanthandegradationwasanalyzedbysizeexclusionchromatography. Itwasshownthatatafixedxanthanconformation,thebackbonedegradationbycellulasesisequalforeachtypeofxanthan.Completebackbonedegradationisonlyobtainedatafullydisorderedconformation, indicating that only the secondary xanthan structure influences thefinaldegreeofhydrolysisbycellulases.Itistherebyshownthat,independentlyonthedegree of substitution, xanthan can be completely hydrolyzed to oligosaccharides.These oligosaccharides can be used to further investigate the primary structure ofdifferent xanthans and to correlate the molecular structure to the xanthanfunctionalities.

Chapter 2

20

INTRODUCTION

Xanthan is an extracellular polysaccharide that is secreted by the microorganismXanthomonas campestris,1with a molecular mass ranging from 1–7*106 Dalton.2, 3Xanthaniswater‐solubleandxanthansolutionsexhibithighpseudoplasticflowevenat low concentrations. Xanthan has a β‐1,4‐glucan backbone with on every secondglucose unit a (31) linked α‐D‐mannose‐(21)‐β‐D‐glucuronic acid‐(41)‐β‐D‐mannose side chain (Figure 2.1.).4 Depending on the Xanthomonas strain and thefermentationconditionsusedforxanthanproduction,approximately90%oftheinnermannoseunitsareO‐6acetylated,and30‐50%oftheterminalmannosegroupscarrya 4,6‐linked pyruvic acid acetal group.5‐7 No information is available regarding thedistributionpatternofthesesubstituents.

In solution, native xanthan as produced by Xanthomonas campestris, exists in a

double‐strandedhelicalconformationwithanorder‐disordertransitionuponchanges

intemperatureand/orionicstrength.8‐10Anorderedconformationofxanthancanbe

recovered by cooling or by increasing the ionic strength in a process called

renaturationofxanthan.Renaturedxanthanhowever,doesnotexhibittheexactsame

conformation and rheological properties as thenative xanthan.3, 10‐13 In this respect

theorder‐disordertransitionofthenativexanthancanbeconsiderednon‐reversible.

Renatured xanthan also exists in a helical conformation, and has an order‐disorder

transition,whichis,incontrasttonativexanthan,reversible.14‐16

Figure 2.1. The ideal repeating unit of xanthan as reported by Jansson et al.4

OO

OH

OHO

HO

O

O

O

OH

CH2OH CH2OH

O

HOOH

O

COO-

O

HO

OH O

OH

n

CH3COOCH2

-OOCO

OH3C

CH2

Enzymatic hydrolysis of xanthan

21

Theorder‐disordertransitionsofbothnativeandrenaturedxanthan,andtherebytheviscosity of a xanthan solution, depends strongly on the molecular composition ofxanthan,particularlywithrespecttothepresenceofacetyland/orpyruvategroups.Amore ordered, helical conformation is obtained by the removal of pyruvate groups,whereas a more disordered conformation is obtained by the removal of acetylgroups.12, 17‐19BecauseboththefermentationconditionsandtheXanthomonasstrainused during the production of xanthan influence the primary structure of xanthan,differences inconformationalbehaviormaybepresentbetweendifferentbatchesofxanthan. Therefore, the primary structure of xanthan is of importance when thetransitionalbehaviorisstudied.However,untiltodatenosuitablemethodisavailablefor the comparison of the primary structure of different batches of xanthan.Characterization of diagnostic xanthan oligosaccharides could help reveal the exactxanthanstructure.Productionofsucholigosaccharidesbychemicaldegradationofthexanthan backbone will also cause degradation of the xanthan side chains and istherefore not suitable. An enzyme‐based method for the production of xanthanoligosaccharides,thatwillspecificallydegradethexanthanbackbone,leavingthesidechainsintact,wouldbemoreuseful.Previous studies have shown that xanthan can be degraded by cellulases underaqueous conditions in which xanthan appears in a disordered conformation.20‐23

However,analysisofthedegradationproductsusinggelpermeationchromatographyshowed that high molecular weight products remain. Such high molecular weightfractions have a higher content of pyruvate and acetyl groups than the parentalxanthan. Therefore, it has been discussed that the accessibility of the backbonetowards enzymatic degradation might be reduced by the presence of thesesubstituentsinthesidechains.20,22Becausedifferencesintheprimarystructurealsocause differences in the order‐disorder behavior, it was also hypothesized thatdifferences in conformation, due to differences in the primary structure, result inenzyme resistant xanthan strands.21, 22 However, no conclusive studies wereperformedtoconfirmtheseassumptions.Toourknowledge,inliteraturenoclearoverviewispresentfortheeffectofboththeprimary xanthan structure as well as the secondary xanthan structure on theenzymatichydrolysisofthexanthanbackbone.Inthisstudywethereforeanalyzedthecombinedinfluenceofthedegreeofsubstitutionaswellasthexanthanconformation,ontheenzymaticdegradationofxanthan.

Chapter 2

22

MATERIALS AND METHODS

Xanthan samples

Unmodified, renatured xanthan in Na+ salt form (RX) was obtained from DuPont(Melle,France).Acetylfreexanthan(AFX)wasproducedbyasaponificationtreatmentof RX with 1 M NaOH (18 h; 4°C). Pyruvate free xanthan (PFX) was produced byheatingRXto100°Cina5mMtrifluoroaceticacid(TFA)solutionfor90min.24Acetyl‐andpyruvate‐freexanthan(APFX)wasproducedbya5mMTFAtreatmentfollowedbysaponification.Allmodifiedpolymersweredialyzedagainstdemineralizedwaterfor24handthenlyophilized.In order to analyze the influence of the sodium counter ions on the xanthanconformation, the chemically modified xanthans and the normal xanthan wereconvertedtotheirH+‐form.A2mg·mL‐1xanthansolutionwasmixedwithAmberliteIR‐120‐H+ ion‐exchange material (BDH, Poole Dorset, UK) for 30 minutes at roomtemperature.The ionexchangematerialwas removed fromthexanthansolutionbycentrifugation(5000xg,15min.,20°C).25Thesupernatantwasneutralizedusing100mMNaOH,dialyzedagainstdemineralizedwater for24h,andthen lyophilized.Thegeneratedxanthanswillhereafterbereferredtoas“H+”for“proton‐form”.

Xanthan composition

The constituent monosaccharide compositions of the unmodified xanthan and thechemicallymodifiedxanthans,inthesodium‐formaswellasintheproton‐form,weredetermined by methanolysis.26 The degree of acetylation of xanthan samples wasmeasured using a Megazyme acetic acid kit (Megazyme, Wicklow, Ireland) after asaponification step with 1 M NaOH (18 h; 4°C). The degree of pyruvylation wasmeasuredusingaMegazymepyruvicacidkit(Megazyme)afteracidhydrolysiswith1MTFA(100min.;90°C).24

Circular Dichroism

Theellipticities(inmDeg)of2mg·mL‐1xanthansolutionsweremonitoredat219nmfrom10‐85°CusingaJascoJ‐715Spectropolarimeter(JascoCorp.,Tokyo,Japan)withaheatingrateof30°C·h‐1,adatapitchof0.5°C,aresponsetimeof1sec,asensitivityof100mDegandat abandwidthof2nm.The transitionprofilesofRX,AFX,PFXandAPFXwere determined in 0, 2 and 10mMNaCl solutions, the transition profiles ofRX‐H+, AFX‐H+, PFX‐H+ and APFX‐H+ were determined in demineralized water. The


23

temperaturewascontrolledusingaJascoPTC‐348WIcontroller.Quartzcuvetteswithanopticalpathof1mmwereused.The transition profiles obtainedwere used to determine the fraction of disorderedconformation(α)atacertainionicstrengthandtemperatureusingEquation1with:θt = ellipticity at a given temperature; θU = ellipticity of a completely disorderedstructureandθF=ellipticityofacompletelyorderedstructure.27

α=1–(θt–θU)/(θF–θU) (1)

Theminimumandmaximumellipticitiesweredeterminedforeachtypeofxanthan,θFwas determined in 10 mM NaCl solutions at 15°C and θU was determined in theH+‐format85°C.Theobtainedcurveswerenormalizedbythebest‐fitparameters.

Enzymatic hydrolysis

Cellulases in the experimental enzyme preparation C1‐G1 from MyceliophthorathermophilaC1 (DyadicNetherlands,Wageningen, TheNetherlands)28were used tohydrolyzethexanthanbackboneatseveraltemperatureandsaltconditions.Solutionscontaining2mg·mL‐1xanthanwerepreparedin0,2and10mMNaClforRX,AFX,PFXand APFX, and solutions containing 2 mg·mL‐1 xanthan were prepared indemineralized water for RX‐H+, AFX‐H+, PFX‐H+ and APFX‐H+. The hydrolysis wasperformedbyincubating1mLofaxanthansolutionwith60µgproteinfor0,3,24or48 h at 40, 45, 50, 55 or 60°C. The hydrolysis was stopped by rapidly cooling thedigests to6°C.Sampleswerekeptat6°Cuntil analysis. Inorder toexcludepossibleinfluencesofthechosenincubationconditionsontheenzymeactivity,carboxymethylcellulose(Sigma‐Aldrich,Tseinheim,Germany)wasusedasreferencesubstrate.TheamountofreducingendsugarswasdeterminedusingthePAHBAHassay.29Thedegreeofhydrolysis(DH)isdeterminedbytheincreaseinreducingendsugarsafterenzymatichydrolysis.Ifeveryglucoselinkageinthebackboneissplit,aDHof100%isobtained. Complete hydrolysis of xanthan to its repeating units, would, therefore,correspond to a maximal DH of 50%. Assuming that the maximum degradation ofxanthan is obtainedwhenxanthan ishydrolyzed to its repeatingunit, aDHof50%wouldcorrespondtoadegreeofdegradationof100%.

High performance size exclusion chromatography (HPSEC)

HPSECwasperformedonanUltimate3000system(Dionex,Sunnyvale,CA,USA).AsetofthreeTSK‐Gelcolumns(TosohBioscience,Tokyo,Japan)wasusedinserieswithseparationcolumnsG‐6000PW,G‐3000PWandG‐2500PW(7.8mm×300mm).The

Chapter 2

24

columntemperaturewassetat55°C.Thesamples(20μl,2mg·mL‐1)wereelutedwith1%(v/v)ethyleneglycolin0.2–μm‐filtered0.2MNaNO3ataflowrateof0.8ml·min‐1.2Theeluatewasmonitoredusingrefractiveindexdetection(ShodexRI101;ShowaDenko K.K., Kawasaki, Japan). Molecular masses were estimated with the help ofpullulanmolecular‐mass standards (Polymer Laboratories, Palo Alto, CA, USA). Thexanthan degradation products were divided into three fractions: 1) non‐degradedxanthan(Rt=18‐25min);2)intermediatedegradationproducts(Rt=25‐31min)and3) completely degraded xanthan (Rt = 31‐34 min). The relative molecular weightdistributionwascalculatedbydividingtheintegratedRIpeakareaofeachfractionbythetotalRIpeakareameasuredfrom18‐34minutes.

RESULTS AND DISCUSSION

Xanthan composition

Toensurethatnochangesweremadeduringthechemicaltreatmentsotherthanthetargetedmodifications,unmodifiedxanthan(RX),acetylfreexanthan(AFX),pyruvatefree xanthan (PFX) and acetyl‐ and pyruvate‐free xanthan (APFX), both in theNa+‐formandintheH+‐form,wereanalyzedfortheirmolecularcompositions(Table2.1.)andfortheirmolecularweightdistributions.The ratio glucose:mannose:glucuronic acid is approximately 1.00:0.79:0.47 for allxanthans.Thisindicatesthatnochangesinthesugarcompositionoccurredduringthechemical treatments.Assuming an ideal repeating xanthan structure (Figure2.1.), aratio of 1:1:0.5 is expected. Hence, our results indicate that there are someirregularities in the repeating structure of xanthan. These irregularities could becausedbydownstreamprocessingorbyirregularitiesinthebiosynthesis.22,30The conversion of xanthan to theH+‐form, removes the remaining acetyl groups ofAFX and APFX, as well as some residual pyruvate groups from PFX. Because theremoval of acetyl groups andpyruvate groupswas intended inAFX/APFXandPFX,respectively, these deviations in substitution are acceptable within our study. Nosignificantchangesareobservedinthemonosaccharideratios.Toconfirmthatthechemicaltreatmentsdidnotresultinbackbonedegradation,themolecular weight distributions (Mw) of the generated xanthans were determinedusing HPSEC (results not shown). No changes in Mw were observed in any of themodifiedxanthans.Therefore,itwasconcludedthatthetreatmentsusedtogeneratethemodified xanthansdidnot cause relevant changes in the xanthan compositions,exceptfortheremovalofpyruvateand/oracetylgroupsasintended.


25

Table 2.1. Molecular composition of normal xanthan and chemically modified xanthan in the Na+‐form and in the H+‐form

Xanthan

Glc:Man:GlcA Molar ratio

Acetyl content w/w%

Pyruvate content w/w%

RX 1.00 : 0.80 : 0.46 5.79 3.70

AFX 1.00 : 0.81 : 0.48 1.31 3.88

PFX 1.00 : 0.79 : 0.49 6.24 0.87

APFX 1.00 : 0.81 : 0.47 1.06 0.60

RX ‐H+ 1.00 : 0.75 : 0.46 5.87 3.77

AFX ‐H+ 1.00 : 0.79 : 0.46 0.00 3.98

PFX‐H+ 1.00 : 0.76 : 0.48 6.30 0.29

APFX‐H+ 1.00 : 0.81 : 0.45 0.00 0.64

Order‐disorder transitions

Thetransitionalbehaviorofthefourxanthans,inducedbyincreasingthetemperature,wasanalyzedbycirculardichroismatdifferentsaltconcentrations(0,2and10mMNaCl). The temperature profiles (Figure 2.2.) show that the mid‐point transitiontemperature(Tm)ofxanthanincreaseswithincreasingionicstrength;itdecreasesdueto the removal of acetyl groups; and it increases due to the removal of pyruvategroups, aswas previously reported.16, 18 An overview of all Tm observed is given inTable 2.2. When dissolved in demineralized water, removal of both the acetyl andpyruvategroupsresultisalowerTm.However,atincreasingsaltconcentrations,thiseffectontheTmisnotobserved.Furthermore,thetemperaturerangeofthetransitionof themodified xanthans is significantly smaller than for RX, probably because theremovalofthesubstituentsclearlyresultsinlessmolecularvariability. Table 2.2. Midpoint‐transition temperatures of normal xanthan and chemically modified xanthan at different salt concentrations

Xanthan Midpoint‐transition Temperature Demineralized water 2 mM NaCl 10 mM NaCl

RX 44 49 61

AFX 27 32 47

PFX 80 81 ≥ 85

APFX 33 43 59

RX ‐H+ 40 n.a.

a n.a.

AFX ‐H+ 25 n.a. n.a.

PFX‐H+ 37 n.a. n.a.

APFX‐H+ 28 n.a. n.a.

a: n.a. = not analyzed

Chapter 2

26

In contrast to the other xanthans studied, the transitional behavior of PFX is notsignificantly affected by an increasing ionic strength in the temperature rangemeasured.Inthepresenceofpyruvategroups,anincreaseinionicstrengthwilllowerthe repulsive forces by shielding the negative charges of the pyruvate groups. Thisresults in a more rigid, helical structure, as is observed for RX. The removal ofpyruvate groups, and thereby removal of negative charges, already reduced therepulsiveforcesinPFX.AnincreaseinionicstrengththereforedoesnotinfluencetheconformationalbehaviorofPFXinthemeasuredtemperaturerange.14,16,31AsshowninFigure2.2c.,loweringtheionicstrength,bytheconversionofPFXintoPFX‐H+,doesinfluencetheconformationalbehavior.Duetotheremovalofthecounterionsoftheglucuronic acid units, negative charges are induced in the xanthan side chains.Thereby the electrostatic repulsion between side chains is increased, resulting in alowerTm.Fromthetemperatureprofilesobtained,thefractiondisorderedconformation(α)atagiven temperature and ionic strength was determined using Equation 1. ThenormalizedfittedgraphsaredepictedinFigure2.3.Thesefittedcurvescanbeusedtocalculate the fraction of disordered conformation at a given temperature and ionicstrength.Furthermore, it ispossible todeterminethesolutionconditionsneeded, inorder to obtain a certain fraction of disordered conformation. The xanthanconformation during an enzyme incubation can now be controlled by selecting aspecific incubation condition. Hence it is now possible to independently study theinfluence of the xanthan conformation on the enzymatic hydrolysis of xanthan forxanthans having different levels of substitution, but originating from the samexanthanbatch.


27

Figure 2.2. Transition profiles of xanthan in: the H+‐form in deionized water ( ); the Na+‐form in deionized water ( ); the Na+‐form in 2 mM NaCl ( ); and the Na+‐form in 10 mM NaCl ( ). A) unmodified xanthan; B) acetyl free xanthan; C) pyruvate free xanthan; D) acetyl and pyruvate free xanthan.

Figure 2.3. Fraction of disordered xanthan as function of temperature (°C) in: the H+‐form in deionized water ( ); the Na+‐form in deionized water ( ); the Na+‐form in 2 mM NaCl ( ); and the Na+‐form in 10 mM NaCl ( ). A) unmodified xanthan; B) acetyl free xanthan; C) pyruvate free xanthan; D) acetyl and pyruvate free xanthan.

-3

-2

-1

0

1

Ell

ipti

city

(m

De

g)

Temperature (˚C)

-3

-2

-1

0

1

Ell

ipti

city

(m

De

g)

Temperature (˚C)

-3

-2

-1

0

1

Ell

ipti

cit

y (m

Deg

)

Temperature (˚C)

-3

-2

-1

0

1

Elli

pti

cit

y (m

Deg

)

Temperature (˚C)

A B

C D

0 20 40 60 80 100

0 20 40 60 80 100 0 20 40 60 80 100

0 20 40 60 80 100

0

0.2

0.4

0.6

0.8

1

0 20 40 60 80 100

Fra

ctio

n d

iso

rder

ed x

anth

an (α

)

Temperature (˚C)

A

0

0.2

0.4

0.6

0.8

1

0 20 40 60 80 100

Fra

ctio

n d

iso

rder

ed x

anth

an (α

)

Temperature (˚C)

B

0

0.2

0.4

0.6

0.8

1

0 20 40 60 80 100

Fra

ctio

n d

iso

rder

ed x

anth

an (α

)

Temperature (˚C)

C

0

0.2

0.4

0.6

0.8

1

0 20 40 60 80 100

Fra

ctio

n d

iso

rder

ed x

anth

an (α

)

Temperature (˚C)

D

Chapter 2

28

Enzymatic degradation

Todeterminetheinfluenceofthexanthanconformationontheenzymatichydrolysisof xanthan, the fraction of disordered conformation (α) was varied for differentenzyme incubations. Based on Figure 2.3. several temperature and salt conditionswere selected in which the conformation of xanthan ranges from α=0 to α=1. AnoverviewofallconditionsisgiveninTable2.3.

Influence of the incubation conditions on the cellulase activity

Theactivityofenzymesmayalsobeaffectedbytheionicstrengthandtemperatureofasolution.Theinfluenceoftheselectedinubationconditionsonthecellulaseactivitywas, therefore,determinedusingcarboxymethyl celluloseasamodel substrate.Thecellulaseactivitywasnotsignificantlyinfluencedbythechangesinionicstrength;thetemperature, however, does affect the cellulase activity. The highest activity wasobserved at 55°C. This activity is reduced to 72%, by cellulase activity, must beminimized. Therefore, the end point of the enzymatic hydrolysis has been used todeterminetheinfluenceoftheconformationontheenzymatichydrolysisofxanthan.Theendpointof thereactionwasdeterminedbymonitoring theRXdegradationbycellulases in time. Because the lowest cellulase activity was detected at 40°C, thistemperature was used to verify the end point of the degradation. The molecularweight distributions of RX digests in time are shown in Figure 2.4a. After 3 h ofincubation the enzyme digest shows non‐degraded highmolecularweightmaterial,some intermediate degradation products and completely degraded low molecularweightmaterial.After24hofincubation,theintermediatedegradationproductsarefurtherdegradedintocompletelydegradedlowmolecularweightmaterial.Howeverthenon‐degraded,highmolecularweightmaterialremains.Thisindicatesthatunderthese conditions part of the xanthan is resistant to enzymatic hydrolysis. Nosignificant changes in the molecular weight distribution are observed when theincubationwasextendedforanother24h.We,therefore,concludethatafter48hofincubation, the maximal degradation will surely be reached at every incubationconditiontested.


29

Figure 2.4. HPSEC elution patterns of normal xanthan digests. A: xanthan degradation followed in time at 40˚C in demineralized water. B: 48 h digests of normal xanthan differing in conformation (α = fraction of disordered conformation; blank = untreated xanthan).

16 18 20 22 24 26 28 30 32 34 360

1

2

3

4

5

6

7

8

0 hours

3 hours

24 hours

48 hours

Retention time (min)

RI r

esp

on

se (μ

RIU

) 3621 732 50.9 10.3 2.7 0.9 MW (kDa)

HMW Intermediate LMW

16 18 20 22 24 26 28 30 32 34 36

0

4

8

12

16

20

24


RI r

esp

on

se

(μR

IU)

3621 732 50.9 10.3 2.7 0.9 MW (kDa)

HMW Intermediate LMW NaCl

α = 1.00

α = 0.81

α = 0.67

α = 0.57

α = 0.23

α = 0.02

blank

B

A

Chapter 2

30

Table 2.3. Degree of hydrolysis obtained after incubation of xanthan with cellulases for 48 hours at different fractions of disordered conformation. The xanthan conformation was controlled by varying the temperature and ionic strength

Sample Temperature NaCl added Disordered fraction DHa

(˚C) (mM) (α) (%)

RX‐ H+ 40 0 0.81 41

RX‐ H+ 50 0 0.95 51

RX‐ H+ 60 0 0.99 50

RX 40 0 0.81 42

RX 50 0 0.95 50

RX 60 0 1 57

RX 40 2 0.47 9

RX 45 2 0.57 18

RX 50 2 0.67 35

RX 55 2 0.78 43

RX 60 2 0.86 54

RX 40 10 0 5

RX 45 10 0.02 6

RX 50 10 0.08 8

RX 55 10 0.23 10

RX 60 10 0.48 31

AFX 40 0 0.92 53

AFX 60 0 1 43

AFX 40 2 0.93 32

AFX 50 2 1.0 46

AFX 60 2 1.0 46

AFX 40 10 0.18 10

AFX 45 10 0.46 27

AFX 50 10 0.77 45

AFX 60 10 0.99 54

PFX‐ H+ 40 0 0.78 44

PFX‐ H+ 45 0 0.97 55

PFX‐ H+ 50 0 0.99 50

PFX‐ H+ 60 0 1.0 53

PFX 40 0 0.02 5

PFX 60 0 0.14 8

PFX 40 10 0.05 8

PFX 60 10 0.17 6

APFX‐ H+ 40 0 0.99 50

APFX‐ H+ 60 0 1 64

APFX 40 2 0.47 14

APFX 45 2 0.76 40

APFX 50 2 0.95 57

APFX 40 10 0.01 2

APFX 50 10 0.14 6

APFX 55 10 0.32 10 a: DH = degree of hydrolysis. DH=100%: all backbone linkages are degraded, based on the increase in reducing end sugars as measured by PAHBAH assay.


31

Influence of the xanthan conformation on the enzymatic hydrolysis of xanthan

The xanthan conformation during the enzymatic hydrolysis was controlled byselectingdifferenttemperaturesandionicstrengthsforxanthansolutions.Table2.3.givesanoverviewof:1)thetypeofxanthan;2)theionicstrengthandtemperatureofthe xanthan solution during enzyme hydrolysis; 3) the corresponding fraction ofdisordered conformation (α); and 4) the degree of hydrolysis (DH) after 48 h ofincubation.Figure2.5a.showsthecorrelationbetweenthexanthanconformationandthefinaldegreeofdegradationbasedontheDH,where,duetoarepeatingbackboneunitwith2glucoseunits,aDHof50%correspondtoadegreeofdegradationof100%.Itisclearlyshownthatanincreaseinαleadstoahigherdegreeofdegradationattheendpointofthereaction.Whenxanthanexistsinacompletelyorderedconformation,no enzymatic hydrolysis is observed. It is, therefore, concluded that a disorderedconformation is necessary for enzymatic degradation. A previous study showed acorrelationbetweenthespeedofhydrolysisandthexanthanconformation.21Inthatstudy it was also observed that no enzymatic degradation occurs at a completelyordered conformation. However, the influence of substituents on the enzymatichydrolysiswas not analyzed. From the results shown in Figure 2.5a. it can nowbeconcluded that the degree of substitution does not significantly influence the finaldegree of degradation, as long as xanthan exists in the same conformation.Furthermoreourresultsshowthatnotonlythespeedofhydrolysis,butalsothefinaldegreeofhydrolysisisinfluencedbythexanthanconformation.WhenxanthanexistsinacompletelydisorderedconformationamaximumDHof~60%isobserved(Table2.3.).ThisDHexceedsthemaximumtheoreticalvalueof50%assumingthatcellulasescan hydrolyze xanthan to the repeating units. The assay used to determine theincrease in reducing end sugars could give a different response to the xanthanrepeating units than towards glucose, as is also the case with differentmonosaccharides.29Therefore,anoverestimationintheDHcouldexist.Based on the DH, cellulases seem to be able to completely degrade xanthan to itsrepeatingunitwhenxanthanispresentinacompletelydisorderedconformation.Toconfirmthesefindingsthemolecularweightdistributionsofthexanthandigestsweredetermined by HPSEC. The HPSEC elution patterns of normal unmodified xanthandigests,obtainedatdifferentxanthanconformations(α),areshowninFigure2.4b.Tobe able to compare the cellulase degradability of all xanthans tested, the relativemolecularweightdistributionsofalldigests,obtainedatdifferentα,weredetermined.TheresultsareshowninFigure2.5b.

Chapter 2

32

Figure 2.5. Correlation between the fraction disordered xanthan (α) and the enzymatic hydrolysis after a 48 h incubation with the experimental enzyme preparation C1‐G1 based on: A) The degree of hydrolysis measured by the increase in reducing end sugars B) The molecular weight distribution measured by HPSEC. Normal xanthan (); AFX (); PFX (●); APFX (). Open symbols: xanthan oligosaccharides; grey symbols: intermediate degradation products; closed symbols: non‐degraded xanthan.

The elutionprofiles show that as long as xanthan exists in a completely disorderedconformation,xanthaniscompletelydegradedtolowmolecularweightmaterial.TheHPSECresults therebyconfirmthatthecellulasescancompletelyhydrolyzexanthanto the xanthan repeating units, independently of the degree of substitution. Whenxanthan is not completely in the disordered conformation, high molecular weightmaterial remains in the enzymedigests. Thedegreeof degradation asmeasuredbythereducingendassay(Figure2.5a.),andtherelativeabundanceofsmallerfragments(Figure2.5b.)therebyfullymatchandshowaclearcorrelationbetweenthexanthandegradationandthexanthanconformation.Thiscorrelationissimilarforeachtypeofxanthan.Thefinalxanthandegradationbycellulasesatagivenαis,therefore,solelycontrolledbythexanthanconformation.Inearlierstudies itwashypothesizedthat theaccessibilityof thebackbonetowardsenzymaticdegradationmightbereducedbythepresenceofsubstituents in thesidechains.20,22 Because the precise conformation of xanthan under the chosen enzymeconditionswasnotmonitoredinthesestudies,weassumethattheobservedenzymeresistancyinthesestudiesisduetothepresenceof(partly)orderedxanthanstrains,aswasalsopostedasoneofthehypothesisbySutherland.22

Considerations on the transitional behavior of renatured xanthan

Figures2.4b.and2.5a.showthatallenzymedigestsobtainedwithanα≤0.95,containenzymeresistantxanthanwiththesamemolecularmassasthatofuntreatedxanthan.This indicates that at a given condition xanthan molecules are either completelydegradedto lowmolecularweightmaterialorcompletelyenzymeresistant.Because

0

20

40

60

80

100

120

0 0.2 0.4 0.6 0.8 1

Rel

ativ

e R

I pea

k ar

ea (

%)

Fraction disordered conformation (α)

0

20

40

60

80

100

120

0 0.2 0.4 0.6 0.8 1

Deg

ree

of

deg

rad

atio

n (

%)

Fraction of disordered conformation (α)

A B


33

only xanthan molecules in the disordered conformation will be hydrolyzed bycellulases, this indicates that during the order‐disorder transition two populationsexists:1)completelyorderedxanthanmoleculesand2)completelydisorderedxanthanmolecules. However, studies on the transitional behavior of xanthan report thatxanthan gradually dissociates from the outsides of the helices or that due tointramoleculardifferences,sequencesoforderedanddisorderedconformationsexistswithinamolecule.32,33Inthatcase,enzymatichydrolysisofthesepartiallydissociatedxanthanheliceswouldresult inenzymeresistantdegradationproductswitha lowermolecular weight than the untreated xanthan. Because this is not observed withHPSEC,itismostlikelythattherenaturedxanthanusedinthisstudydoesnotfollowthesametransitionalbehavioraspreviouslydescribed.Different populations of ordered and disordered conformation, however, might beobtained when high intermolecular variations exist within a xanthan batch.12, 18Differences inprimarystructureswithinonebatchwould results indifferentTm foreach xanthan molecule. At a given condition a certain xanthan molecule could,therefore, completely exist in an ordered conformation whereas another type ofxanthanmoleculecompletelyexistsinadisorderedconformation.Althoughpossible,it is unlikely that at all conditions chosen in this study, such a sharp division isobtainedinthexanthanconformations,especiallywhenit isassumedthateachtypeofxanthangraduallydissociatesasdescribedabove.High intermolecularvariations,therefore,donotseemtoexplaintheHPSECresultsobtainedinthisstudy.Anexplanationforthehighmolecularweightmaterialintheenzymedigestscouldbethatrenaturedxanthandoesnotexistsassingleordoublestrandedhelices,butasamultiplestrandednetworkofxanthanhelicesaswasrecentlyreported.34Becausethesizeexclusionmethodusedinthisstudyisnotabletodistinguishbetweenmolecularmassvaluesbeyond5.0*106Dalton.Partialdissociationofxanthanmoleculesfromamultiple stranded network would explain our finding as long as the remainingnetworkhas aMw≥ 5.0*106Dalton. Intermoleculardifferencesmight controlwhichpartsofthenetworkdissociatefirst,explainingtheobserveddifferencesinthedegreeof substitution between the degraded and non‐degraded xanthan in previousstudies.22Anotherstudyreportedontheside‐by‐sideassociationoforderedxanthanstructures.32 Alignment of all enzyme resistant ordered xanthan structures into anetwork, which is larger than the HPSEC detection limit of our method, mighttherefore also explain the high molecular weight observed in the enzyme digests.Basedonourfindingswewouldthereforeconcludethatorderedxanthanstructuresdonotexistassingleordoublehelices,butasanetworkofmultiplehelices.

Chapter 2

34

CONCLUSIONS

Wehaveinvestigatedtheinfluenceoftheprimaryandsecondarystructureofxanthanontheenzymatichydrolysisofthexanthanbackbone.Aclearcorrelationbetweenthesecondarystructureandtheextentofenzymaticdegradationofxanthanisobserved,whereonlydisorderedxanthanstructuresarehydrolyzedbycellulases.Wheninthedisorderedform,nocorrelationexistsbetweentheprimarystructureofxanthanandthefinalxanthandegradationbycellulases.Bycontrollingthexanthanconformationitis, therefore,possible tocompletelydegradedifferent typesofxanthan intoxanthanoligosaccharides. Further characterization of the oligosaccharides produced fromdifferenttypesofxanthan,enablesthecomparisonoftheprimaryxanthanstructures,especiallyregardingtherepeatingunitspresent.Consequently, furtherresearchintotheinfluenceofthedistributionoftherepeatingunitsonxanthansfunctionalitywillbepossible.The presence of non‐degraded, high molecular weight xanthan after enzymatichydrolysis of xanthan which was partly present in a disordered conformation washypothesizedtobecausedbythepresenceofaxanthannetwork.

ACKNOWLEDGMENTS

This research was supported by the European Community within a consortiumPolyModEKBBE‐2007‐3‐3‐07andisgratefullyacknowledged.


35

REFERENCES[1] JeanesA,PittsleyJE,SentiFR.PolysaccharideB‐1459:Anewhydrocolloidpolyelectrolyteproduced

fromglucosebybacterialfermentation.JApplPolymSci.1961;V:519‐26.

[2] MilasM,RinaudoM,TinlandB.Comparativedepolymerisationof xanthangumbyultrasonicandenzymatictreatments‐Rheologicalandstructuralproperties.CarbohydrPolym.1986;6:95‐107.












[14] Holzwarth G. Conformation of the extracellular polysaccharide of Xanthomonas campestris.Biochem.1976;15:4333‐9.



[17] RinaudoM. Role of substituents on the properties of some polysaccharides. Biomacromolecules.2004;5:1155‐65.


[19] DentiniM,CrescenziV,BlasiD.Conformationalpropertiesofxanthanderivativesindiluteaqueoussolution.IntJBiolMacromol.1984;6:93‐8.

Chapter 2

36




[23] Christensen BE, Smidsrød O. Dependence of the content of substituted (cellulosic) regions inprehydrolysedxanthansontherateofhydrolysisbyTrichodermareeseiendoglucanase. Int JBiolMacromol.1996;18:93‐9.

[24] Bradshaw IJ,NisbetBA,KerrMH,Sutherland IW.Modifiedxanthan‐itspreparationandviscosity.CarbohydrPolym.1983;3:23‐38.

[25] Rinaudo M, Milas M. Polyelectrolyte behaviour of a bacterial polysaccharide from Xanthomonascampestris:Comparisonwithcarboxymethylcellulose.Biopolymers.1978;17:2663‐78.

[26] RuiterGAde,ScholsHA,VoragenAGJ,RomboutsFF.Carbohydrateanalysisofwatersolubleuronicacid‐containingpolysaccharideswiththehigh‐performanceanion‐exchangechromatographyusingmethanolysis combined with TFA hydrolysis is superior to four other methods. Anal Biochem.1992;207:176‐85.

[27] Greenfield NJ. Using circular dichroism collected as a function of temperature to determine thethermodynamicsofproteinunfoldingandbindinginteractions.NatProtoc.2006;1:2527‐35.

[28] Kühnel S, Schols HA, Gruppen H. Aiming for the complete utilization of sugar‐beet pulp:Examination of the effects of mild acid and hydrothermal pretreatment followed by enzymaticdigestion.BiotechnBiofuels.2011;4:1‐14.

[29] LeverM.Anewreactionforcolorimetricceterminationofcarbohydrates.AnalBiochem.1972;47:273‐9.


[31] ShatwellKP,SutherlandIW,Ross‐MurphySB.Influenceofacetylandpyruvatesubstituentsonthesolutionpropertiesofxanthanpolysaccharide.IntJBiolMacromol.1990;12:71‐8.

[32] Norton IT, Goodall DM, Frangou SA, Morris ER, Rees DA. Mechanism and dynamics ofconformationalorderinginxanthanpolysaccharide.JMolBiol.1984;175:371‐94.

[33] LiuW,NorisuyeT.Thermallyinducedconformationchangeofxanthan‐Interpretationofviscositybehaviourin0.01Maqueoussodium‐chloride.IntJBiolMacromol.1988;10:44‐50.

[34] GulrezSKH,Al‐AssafS,FangY,PhillipsGO,GunningAP.Revisitingtheconformationofxanthanandtheeffectofindustriallyrelevanttreatments.CarbohydrPolym.2012;90:1235‐43.

Published as: Marijn M. Kool, Harry Gruppen, Graham Sworn and Henk A. Schols. Carbohydrate Polymers. 2013, 98(1), 914‐921.

Chapter 3

Comparison of xanthans by the relative abundance of its six constituent repeating units

ABSTRACT

Fivexanthanswerehydrolyzedto theirrepeatingunitsusingcellulases.Hydrophilicinteraction chromatography with online electrospray ionization ion trap massspectrometry and evaporative light scattering detection was used to analyze theoligomersreleased.Itwasconcludedthatsixdifferentpentamerrepeatingunits(RUs)existswithinaxanthansample.ThemostabundantRUshowsacetylationontheinnermannose and pyruvylation on the outer mannose. The second most abundant RUshows acetylation on both the inner and the outermannose. It becomes clear thatmorevariationsinthexanthanstructureexistthangenerallyrecognized.Comparisonoffivedifferentxanthansamplesrevealedthat,althoughthemolecularcompositionofxanthansamplescanbeexactlythesame,theratioinwhichtheRUsoccurcandiffersignificantly.Itis,therefore,concludedthatxanthansamplesshouldbecharacterizedforboth,theirmolecularcompositionandtherelativeabundanceoftheRUspresent.

Chapter 3

38

INTRODUCTION

Xanthan gum is an exopolysaccharide secreted by Xanthomonas spp. The generallyaccepted xanthan structure consists of a cellulosicbackbonewith trisaccharide sidechains linkedtoeveryalternateglucoseunit.Thesesidechainsconsistofmannose–glucuronic acid–mannose units, which are substituted with an acetyl group on theinnermannoseandapyruvicacidketalon theoutermannose.1Theexactdegreeofsubstitution,however,isknowntovarydependingonthefermentationconditions2,3andtheXanthomonasstrainusedforxanthanproduction.4,5Theprimarystructureofthe produced xanthan can be controlled by specific mutations in the Xanthomonasgenome.6SixdifferentpentamerrepeatingunitsareproposedbasedonthegenotypeoftheXanthomonasstrainused.Itwasshownthatsuppressionofthegeneinvolvedinthepyruvylationof theoutermannose resulted in a higherdegree of acetylation. Itwas,therefore,suggestedthatxanthansidechainscanalsobeacetylatedontheoutermannose.However, theexactpositionof theacetylgroupswasnotdetermined.Thesix repeating units proposed by Hassler & Doherty6 were, therefore, onlyhypothesized and not experimentally determined. Another study also report ondouble acetylated side chains.7 In this study sugar linkage analysis was used todetermine thepositionof thesecondacetylgroup. ItwasshownthatacetylationontheO‐6 position of the outermannose, as proposed by Hassler &Doherty,6 indeedoccurs in xanthanmolecules. About 24% of all outermannose units are acetylatedaccording to this study. However, the precise structure has not been linked to theindividual repeating units and the ratio in which they coexist has not beendetermined.Todate,noconclusivestudyhasbeenperformedon theexactpositionof theacetylgroups in the xanthan side chains. The exact structure of the different xanthanrepeatingunitspresentinonexanthansamplehas,therefore,neverbeendetermined.Inapreviousstudyitisshownthatcellulasescancompletelydegradexanthantothexanthan repeatingunits.8 In thepresent study, these repeatingunitswill be furthercharacterized and used for quantification of the xanthan repeating units present indifferent xanthan samples. We thereby introduce a method that enables theunambiguouscomparisonofdifferentxanthans.

Comparison of xanthans based on its six constituent repeating units

39


Xanthan samples

FivetypesofrenaturedxanthanwerekindlyprovidedbyDuPont(Melle,France).Themolecular compositions (Table 3.1.) are determined as previously described.8 Allsamples have a molar glucose:mannose:glucuronic acid ratio which is close to theexpected ratio 2:2:1. The xanthan samples differ in their degree of substitution.XanthansAandBshowsimilaracetylandpyruvylcontentsof~5.7%(w/w)and4.5%(w/w),respectively.Thiscorrespondsto1.15acetyland0.56pyruvylgroupspersidechain,whichindicatesthatthesidechainscanindeedbemultipleacetylated.XanthansC and D also have a similar composition to one another but have a slightly higherpyruvyl content and a slightly lower acetyl content compared to xanthansA andB.Xanthan E has a higher degree of pyruvylation, corresponding to almost fullypyruvylatedsidechains.


Solutions containing 2mg·ml‐1 xanthanwere prepared in demineralizedwater andhydrolyzed using cellulases from the experimental enzyme preparation C1‐G1 fromMyceliophtorathermophilaC1(DyadicNetherlands,Wageningen,TheNetherlands).8,9Thehydrolysiswasperformedby incubating1mlof a xanthan solutionwith60µgprotein for 48 h at 60°C. After incubation, the digests were boiled (10 min) andcentrifuged (10,000 g; 10 min; 25°C). The supernatants were analyzed by HPSEC,HPAECandUPLC‐ELSD‐MSn.

Table 3.1. Molecular composition of five xanthan samples

Xanthan type

Glc:Man:GlcA Molar ratio

Acetyl content w/w%

Pyruvate content w/w%

Xanthan A 1 : 0.88 : 0.41 5.6 4.4

Xanthan B 1 : 0.91 : 0.43 5.9 4.6

Xanthan C 1 : 0.92 : 0.44 4.9 5.1

Xanthan D 1 : 0.91: 0.43 4.9 5.2

Xanthan E 1 : 0.94 : 0.45 4.8 7.3

Chapter 3

40


HPSEC was performed as described previously.8 Molecular masses were estimatedusingpullulanmolecular‐massstandards(PolymerLaboratories,PaloAlto,CA,USA).

High performance anion exchange chromatography (HPAEC)

HPAEC was performed on an ICS5000 HPLC system (Dionex, Sunnyvale, CA, USA),equippedwithaCarboPacPA‐1column(2mmIDx250mm;Dionex)incombinationwith a CarboPac PA guard column (2 mm ID x 25 mm) and an ISC5000 ED PAD‐detector (Dionex). The digests were centrifuged (10,000 g; 10 min; 25°C) and 2xdilutedbeforeinjectionontothecolumn(10μl).Sampleswereelutedataflowrateof0.3ml·min‐1withthefollowingelutionprofileof0.1Msodiumhydroxide(NaOH)and1M sodium acetate (NaOAc) in 0.1MNaOH: 0–10min, 0 – 50mMNaOAc in 0.1MNaOH; 10–35 min, 50–400 mM NaOAc in 0.1 M NaOH; 35–40 min, 400–1000 mMNaOAcin0.1MNaOH;40–45minwashingstepwith1MNaOAcin0.1MNaOH;45–60min,equilibrationwith0.1MNaOH.TheglucosereleasedwasquantifiedbasedontheresponsefactorofstandardD‐glucose.

Hydrophilic interaction liquid chromatography with evaporative light

scattering and mass spectrometry detection (HILIC‐ELSD‐MS)

DigestswereanalyzedusingUPLC‐ELSD‐MSnonaHILICBEHamidecolumn(WatersCorporation, Milford, MA, USA) as described elsewhere with a modified gradient.10Thefollowingelutionprofilewasused,withA)1%(v/v/)acetonitrile(ACN)inwater;(B)100%ACN;and(C)2%(v/v)formicacidin200mMammoniumformatesolution:0–1min,isocratic15%A,80%Band5%C;1–25min,linearto45%A,50%Band5%C;25–30minlinearto55%A,40%Band5%C;30–35min,isocratic55%A,40%Band5%C;35–35.1minlinearto15%A,80%Band5%C;35.1–40min,isocratic15%A, 80%B and 5%. The xanthan digestswere centrifuged and diluted 1:1with ACNbeforeinjection(5μl)intothesystem.TheAcquityBEHAmidecolumnwascoupledtoa splitter (Accurate,Dionex Corporation) directing the eluent to an ELSD and to anESI‐MSn‐detector with a ratio 10:1 respectively. The ELSD micro flow nebulizer(Sedere,France)hadagaspressureof3.5barandagasflowof1.75L·min‐1.Thedrifttubetemperatureof theELSDwasset to50°Candthegainto12.MS‐detectionwasperformedinnegativemodeonaVelosProiontrapMS(ThermoScientific,SanJose,CA, USA) with the ion source voltage set to –4.5 kV, capillary temperature 250°C,sheathgas30(arbitraryunits),auxiliarygas12(arbitraryunits).Massspectrawereacquired over the scan range m/z 300–2000. MSn‐collection parameters included


41

normalized collision energy35 (arbitraryunits), activationQ0.25 (arbitraryunits),activationtime30msandisolationwidth2m/z.


Enzymatic hydrolysis of the xanthan backbone

Thedifferentxanthanswereincubatedwithcellulasesinordertocompletelydegradethexanthansintotheirrepeatingunits.Inapreviousstudyweshowedthatcellulasesfrom the C1‐G1 preparation from Myceliophthora thermophila C1 can completelyhydrolyzethexanthanbackbonewhenxanthanispresentinacompletelydisorderedconformation.8 To ensure that the different xanthans were indeed completelyhydrolyzed at the incubation conditions chosen, HPSECwas used to determine themolecular weight distribution of the xanthan digests (Figure 3.1). After a 48 h.incubation all xanthan samples are indeed completely hydrolyzed to xanthanoligosaccharides. The xanthan degradation products can, therefore, be used forquantificationandcharacterization.

Figure 3.1. HPSEC elution patterns of xanthan digests obtained after 48 h of incubation at 60˚C (solid line), and the corresponding xanthan blanks (dotted line).

Xanthan D

Xanthan C

Xanthan B

Xanthan E

Xanthan A

RI R

esp

on

se

3621 732 50.9 10.3 2.7 0.9 MW (kDa)

18 20 22 24 26 28 30 32 34 36


Pullulan

Chapter 3

42

Analysis of xanthan degradation products using HPAEC

AsHPAECisacommonmethodformono‐andoligosaccharideanalysis,11thexanthandigestswereanalyzedbyHPAEC(Figure3.2).Smallamountsofglucosewerereleased(≤5%ofall glucose) inalldigests,while releaseofmannoseand/orglucuronicacidwasnotobserved.Itis,therefore,concludedthatthecellulasesuseddonotshowanyside chain degrading activity. The free glucose could originate from parts of thebackbonewherethesidechainislacking,12enablingthecellulasestoreleaseglucose.Nexttothereleaseofglucose,twoadditionalpeaksareobservedintheHPAECelutionpatterns.BecauseHPAECisconductedunderalkaliconditions,acetylestersoriginallypresent will be saponified online. The two peaks will, therefore, represent: 1) theunsubstitutedpentamerRUand2)thepyruvylatedpentamerRU.Becausesidechainscarrying a pyruvic acid acetal aremore negatively charged than unsubstituted sidechains,thepyruvylatedRUeluteslater,aswasconfirmedbytheanalysisofapyruvatefreexanthan(resultsnotshown).NootherdegradationproductsareobservedintheHPAEC elution pattern of any of the xanthan digests. Therefore, we conclude thatunder the chosen enzyme conditions, all xanthans are completely degraded to thexanthanRUs,confirmingtheHPSECresults.

Figure 3.2. HPAEC elution pattern of the 48h xanthan A cellulase digest.

5 10 15 20 25 30 35 40 45 50 55 60


PA

D R

esp

on

se

Glucose Repeating unit

Pyruvylated repeating unit


43

Characterization of xanthan repeating units using LC‐MS

NoinformationonthepositionofacetylgroupscanbeobtainedbyHPAECduetotheonline saponification of the acetyl esters. The xanthan repeating units (RUs) were,therefore,furthercharacterizedusinghydrophilic liquidinteractionchromatography(HILIC)withonlinemassspectrometry(MS).TheHILICelutionpatternofthexanthanAdigest(Figure3.3.)showsthatthedifferentRUs present in the cellulase digest are well separated. Five different peaks arerecognized.Thebroadpeaksarearesultsofpartialα‐/β‐anomerseparation,aswasalsoobservedformaltodextrinsusingthesamecolumn.10Electrosprayionizationiontrapmass spectrometry (ESI‐IT‐MSn) was used to identify all five compounds. Thedominantpeak(peak3)hasam/z‐valueof953,correspondingtothenegativeionoftheRU that is bothpyruvylated andacetylated.Figure3.4 shows the fragmentationpatternof thisRU,according to thenomenclaturedescribedbyDomon&Costello.13Themostabundantfragmentsarem/z909,791and703,whichwereinterpretedasfollows:m/z909(953‐44, lossofacarboxylgroup),791(953‐162, lossofaglucoseunit)and703(953‐250,lossofpyruvylatedmannose).Thefragmentm/z541canbeannotatedastheRUwithout2hexosesandapyruvylgroup.Thismasslosscanonlybeexplainedwhenoneglucoseiscleavedsimultaneouslywiththepyruvylatedoutermannose. Such double cleavage, from both sides of an oligosaccharides, has beenreportedpreviouslyinthefragmentationpatternofxyloglucanoligomers.14

Figure 3.3. HILIC ELSD elution pattern of the 48h xanthan A cellulase digest.

8 10 12 14 16 18 20 22 24


EL

SD

res

po

ns

e

1 2

3

4 5

Chapter 3

44

Figure 3.4. MS2 fragmentation pattern of the acetylated and pyruvylated pentamer repeating unit, eluting at 16.5 min. in Figure 3 and the chemical structure of m/z 953 with the observed cleavages according to the nomenclature of Domon & Costello.13

Table3.2 shows them/z‐valuesand the fragmentationpatternsofall compounds inthexanthanAdigest.Consideringthedoublecleavageobservedinthefragmentationpattern of the acetylated and pyruvylatedRU, all compounds could be identified asxanthanpentamerRUsdifferinginsubstitutionpattern.Peak1(Figure3.3.)hasam/z

(909)X5α+1Ac

A1α(44)

(162)

Z1

2,5A2(875)

0,2A2(893)C2+1Pyr+1Ac

(791)

C1α+1Pyr (250)

(324)Z2α

C3α+1Pyr+1Ac(629)

(342)

Y2α

B3α+1Pyr+1 Ac(611)

(703)Z4α+1Ac

(546)

Y3α

B2α+1Pyr(407)

X5α+ 1Ac

B2α

+ 1

Pyr

[M-H

]- -H

2O

2,5 A

2

C3α

+ 1

Pyr

+ 1

Ac

[Z4α

+ 1

Ac]

–Y

2α

[Z4α+ 1Ac]–

Z1

B3α

+ 1

Pyr

+ 1

Ac

[X5α

+ 1

Ac]

–0,

2 X1

0,2 A

2

C2+

1Pyr

+ 1

Ac

[X5α+ 1Ac]–

Z1Z

4α+

1A

c

[Z4α

+ 1

Ac]–

0,2 X

1

[Z4α

+ 1

Ac]

–2,

5 X1

300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000

m/z

0

Rel

ativ

e A

bu

nd

ance

909

791

703

541

747

893625 849

875629407643611 935361

10

20

30

40

50

60

70

80

90

100

[M-H]-= 953


45

Table 3.2. Fragm

entation pattern and identification of the peaks in Figure 3.3 according to the nomen

clature of Domon & Costello

13

as dep

icted in Figure 3.4.

Peak number

RT (m

in)

[M

‐H]‐

(m/z)

Ion fragm

ents

(m/z)

Proposed

structure

Structure

code

1

11.7 – 12.3

925

907 ([M

‐H]‐ ‐H2O), 865 ([M

‐H]‐ ‐

0,2X1), 847 ([M

‐H]‐ ‐

2,5X1),

763 ([M

‐H]‐ ‐Glc), 703 ([M

‐H]‐ ‐AcM

an), 625 ([M

‐H]‐ ‐

2,5 X

1; ‐AcM

an ),

541 ([M

‐H]‐ ‐Glc; ‐AcM

an), 379 ([M

‐H]‐ ‐Glc‐Glc‐AcM

an),

361 ([M

‐H]‐ ‐Glc‐Glc; ‐AcM

an)

RU‐1

2

14.9 – 15.3

883

823 ([M

‐H]‐ ‐

0,2X1), 805

(M‐H]‐ ‐

2,5X1), 721[M

‐H]‐‐Glc),

703([M‐H]‐‐M

an), 643 ([M

‐H]‐ ‐0,2X1; ‐Man),

625 ([M

‐H]‐ ‐2,5X1; ‐Man), 541([M‐H]‐ ‐Glc; ‐Man),

361([M‐H]‐ ‐G

lc‐Glc;‐Man)

RU‐2

823 ([M

‐H]‐ ‐

0,2X1), 805

(M‐H]‐ ‐

2,5X1), 721[M

‐H]‐‐Glc),

661([M‐H]‐‐AcM

an), 601 ([M

‐H]‐ ‐0,2X1; ‐AcM

an),

583 ([M

‐H]‐ ‐2,5X1; ‐AcM

an), 499([M‐H]‐ ‐G

lc; ‐AcM

an),

379([M‐H]‐ ‐G

lc‐Glc‐M

an)

RU‐3

3

16.3 – 16.7

953

909 ([M

‐H]‐ ‐CO

2), 893 ([M

‐H]‐ ‐

0,2X1), 849 (M‐H]—CO

2; ‐ 0,2 X

1),

791 ([M

‐H]‐ ‐Glc), 747 ([M

‐H]—Glc; ‐CO

2), 703 ([M

‐H]‐ ‐PyrMan),

625 ([M

‐H]‐ ‐2,5X1; ‐PyrMan), 541 ([M

‐H]‐ ‐Glc; ‐PyrMan),

407 ([M

‐H]‐ ‐Glc‐Glc‐AcM

an), 361 ([M

‐H]‐ ‐G

lc‐Glc; ‐PyrMan)

RU‐4

4

18.0 – 18.3

841

781 ([M

‐H]‐ ‐

0,2X1), 763

(M‐H]‐ ‐

2,5X1), 679 ([M

‐H]‐ ‐Glc),

661 ([M

‐H]‐ ‐Man), 601 ([M

‐H]‐ ‐0,2X1; ‐Man), 517 ([M

‐H]‐ ‐G

lc‐Glc),

499 ([M

‐H]‐ ‐Glc; ‐Man)

RU‐5

5

19.1 – 19.4

911

867 ([M

‐H]‐ ‐CO

2), 851 ([M

‐H]‐ ‐

0,2X1), 807 (M‐H]‐ ‐CO

2; ‐ 0,2X1),

749 ([M

‐H]‐ ‐Glc), 705 ([M

‐H]‐ ‐Glc; ‐CO

2), 661 ([M

‐H]‐ ‐PyrMan),

601 ([M

‐H]‐ ‐0,2X1; ‐PyrMan), 583 ([M

‐H]‐ ‐2,5X1; ‐PyrMan),

569 ([M

‐H]‐ ‐Glc‐Glc), 499 ([M

‐H]‐ ‐Glc; ‐PyrMan),

407 ([M

‐H]‐ ‐Glc‐Glc‐M

an)

RU‐6

glucose; m

annose; glucuronic acid; acetyl groups; pyruvic acid ketal

Chapter 3

46

valueof925,whichcorrespondsto thenegative ionofaRU that is substitutedwithtwoacetylgroups.TheMS‐fragmentationshowsthatoneacetylgroupispositionedontheinnermannoseandtheotherontheoutermannose,aswaspreviouslyshownbyStankowski et al.7 using linkage analysis. Peak 2 has a m/z‐value of 883, whichcorresponds to the negative ion of a single acetylated RU. Figure 3.5. shows thefragmentation pattern of this RU and the structures of the two possible acetylatedRUs.Themostabundantfragmentsarem/z721,703and541,whichwereinterpretedasfollows:721(883‐162,lossofaglucose),703(883‐180,lossofamannoseunit)and541(883‐342,lossofaglucoseandmannoseunit).Thelossofunsubstitutedmannoseisindicativeforacetylationontheinnermannose.Fragmentsm/z661and499wereinterpreted as follows:m/z661 (883‐222, loss of an acetylatedmannose unit) and499(883‐384,lossofanacetylatedmannoseandaglucoseunit).Thesefragmentsaretherebyspecific foracetylationon theoutermannose.The fragmentationpatternofpeak 2 is, therefore, indicative for the presence of two different single acetylatedrepeatingunits.Basedontherelativeabundanceofthespecificm/zfragments,itcanbe concluded that most of the single acetylated RUs are acetylated on the innermannose(≤85%).Peaks 4 and 5 in the HILIC elution pattern (Figure 3.3.) can be identified as theunsubstitutedxanthanRUandthenon‐acetylated,butpyruvylatedRUrespectively.Intotal six different pentamer RUs are identified. Thereby we confirm that the RUsproposed by Hassler & Doherty6 are indeed present in xanthan. However, theseauthorssuggestedthatsomeRUsareonlysynthesizedafterspecificmutationsintheDNA ofXanthomonas and that nativeXanthomonas strainswould only express RUsthat are substituted on both the inner and outer mannose. In this study it is nowshownthatallsixrepeatingunitscancoexist inonexanthansample,evenwhentheXanthomonas strain is not geneticallymodified. Therefore, it is concluded that ‘the’xanthan repeatingunit doesnot exist and that the xanthan structure ismuchmorecomplexthangenerallydepictedinliterature.WhetherthepresenceandabundanceofthedifferentRUsareresultingfromirregularitiesinthebiosynthesisororiginatefromthedownstreamprocessingofxanthanduringproductionremainsuncertain.


47

Figure 3.5. MS2 fragmentation pattern of the single acetylated pentamer repeating unit, eluting at 15.1 min. in Figure 3.3. and the two isomeric chemical structures of m/z 883 with the observed cleavages according to the nomenclature of Domon & Costello.13

(162)

Z1

2,5A2(805)

0,2A2(823)C2+1Ac

(721)

C1α+1Ac(222)

(324)Z2α

C3α+1Ac(559)

(342)

Y2α

B3α+1 Ac(541)

(661)Z4α

(504)

Y3α

B2α+1Ac(379)

(162)

Z1

2,5A2(805)

0,2A2(823)C2+1Ac

(721)

C1α(180)

(324)Z2α

C3α+1Ac(559)

(342)

Y2α

B3α+1 Ac(541)

(703)Z4α+1Ac

(546)

Y3α+1Ac

B2α(337)

*= specific for acetylation on the inner mannose¤ =specific for acetylation on the outer mannose

C2+

1Ac

Z4α

+1A

c

[Z4α

+1A

c] –

Z1

300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000m/z

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bu

nd

ance

721

703*

541*823

805625* 661¤499¤

583¤ 643*559361*379¤ 601¤ 865

[M-H

]- -H

2O

2,5

A2

0,2

A2

Z4α

[Z4α] –

0,2 X

1

[Z4α] –

2,5 X

1

[Z4α

+1

Ac]

–0,

2X

1

[Z4α

+1A

c] –

2,5

X1

[Z4α

] –Z

1

C3α

+1A

c

B2α

+ 1

Ac

[Z4α

+1A

c] –

Y2α

[M-H]-= 883

Chapter 3

48

Relative abundance of the xanthan repeating units in different xanthan

samples based on UPLC‐ELSD response

Nowthatitisshownthatthepositionoftheacetylgroupcanvary,xanthansampleswith the same levels of acetylation and pyruvylationmay still differ in the relativeabundance of the RUs present. The five xanthan samples were, therefore, alsocomparedbasedontheRUspresentintheircellulasedigests.HILIC elution patterns of the different cellulase digests showed that every type ofxanthancontainedallsixRUs(resultsnotshown).Becausenosuitablestandardswereavailable, the ratio in which the RUs are present in the xanthan samples wasdeterminedbasedontheELSD‐response,asapreviousstudyshowedthat theELSDresponseofdifferenttypesofoligosaccharidesissimilar.15BecausethetwoisomericsingleacetylatedRUselutesimultaneously,theMSfragmentationpatternwasusedtodeterminetheratioinwhichtheseRUsarepresent.Table 3.3. gives an overview of the relative abundance of the RUs present in thexanthan samples. The most abundant RU in all xanthan types is RU‐4, which isacetylatedonthe innermannoseandpyruvylatedontheoutermannose.Dependingon the type of xanthan, the secondmost abundant repeating unit in xanthan is thedouble acetylated repeating unit (RU‐1), the repeating unit acetylated on the innermannose (RU‐2)or thepyruvylated repeatingunit (RU‐6).Together the twodoublesubstituted side chains typically represent about 79% of all xanthan side chains,independentofthexanthantype.Theresults,therefore,indicatethatindependentoftheconditionsusedduringthexanthanproduction,approximatelythesameamountofdoublesubstitutedsidechainsarepresent.The influence of the production conditions on the biosynthesis of xanthan and thedegreeofpyruvylationofthexanthanproducedhasbeendescribedinvariousstudies.A deficiency of nitrogen during the fermentation increases the pyruvate content,whereasadeficiencyofoxygendecreasesthepyruvatecontent.2,3,16Furthermore,itisknownthatthepresenceofcitricacidinthegrowthmediaresultsinaxanthanwithahighpyruvatelevel17andlimitationsoftheamountofmagnesiumorphosphateinthegrowthmediaresultsinaxanthanwithlowpyruvatelevels.18Inmostofthesestudies,however,theeffectofthefermentationconditionsonthelevelofacetylationwasnotstudied.Thisstudyshowsthatthereisnoclearcorrelationbetweenthefermentationconditions and the total amount of double substituted side chains synthesizedHowever,acorrelationbetweenfermentationconditionsandtheacetyl:pyruvyl‐ratioon the outer mannose is observed. We therefore propose that suppression of thebiosynthesis of a pyruvate group on the outer mannose, by selective fermentationconditions, results in higher levels of acetylation on the outer mannose. This is inalignmentwithDavidson18whoobservedanincreaseinthetotalacetylcontentwith


49

Table 3.3. R

elative abundance (mol%) of the repeating units in different xanthan

sam

ples based

on the ELSD

‐response after HILIC sep

aration of 48 h

cellulase digests

Xanthan

type

RU‐1

RU‐2

RU‐3

RU‐4

RU‐5

RU‐6

Substitution

outer mannose

(%)

Substitution

inner m

annose

(%)

% of all acetyl

groups on outer

mannose

Double

substituted RUs

(%)

Ac:Pyr‐ratio

on the outer

mannose

Xanthan

A

19

10

2

62

3

4

87

91

19

81

1:3.2

Xanthan

B

17

11

1

61

3

7

86

89

14

78

1:3.8

Xanthan

C

4

12

1

75

2

6

86

91

5

79

1:16.2

Xanthan

D

10

8

0

64

3

15

89

82

11

74

1:7.9

Xanthan

E

5

3

1

73

3

15

94

81

6

78

1:14.7

glucose; m

annose; glucuronic acid; acetyl groups; pyruvic acid ketal

Chapter 3

50

decreasingpyruvatecontent. Italsoalignswithresults foundbyHassler&Doherty6whichshowedthatblockingtheketalaseactivitythroughmutationsingeneL,resultsintheproductionofxanthanwithhighacetyllevels.Intotal5‐19%ofallacetylgroupsarepositionedontheoutermannose,whileStankowski7mentionedvaluesof20‐25%for their xanthan based on sugar linkage analysis. No clear correlation is observedbetween the acetylation of the inner mannose and the substitution on the outermannose.

Differentiation between ‘similar’ xanthans based on their repeating units

Comparison of the molecular composition of the different xanthans (Table 3.1.)showedthatxanthanCandDarealmostidentical.However,Table3.3.illustratestheexistenceofdifferencesbetweenthesexanthans.XanthanDhastwiceasmuchRU‐6,whichisonlysubstitutedwithapyruvategroupontheoutermannose,thanxanthanC,whereasxanthanChashigher levelsofRU‐4.Furthermore,xanthanDhashigherlevelsofthedoubleacetylatedRU,whereasxanthanChasmoresingleacetylatedRUs(RU‐2andRU‐3).Thereby,xanthanChasaslightlyhigherdegreeof substitutionontheinnermannoseandaslightlylowerdegreeofsubstitutionontheoutermannosecompared to Xanthan D. Cross analysis of the acetyl and pyruvate content of thexanthans (Table 3.1.) corresponds rather well with the acetyl and pyruvate levelscalculated from the type and relative abundance of the repeating units. It can,therefore, be concluded that two xanthans similar in molecular composition, maysignificantlydifferintheirstructure.Establishingthesedifferencesinthestructureofxanthans,mayhelptoexplaindifferencesinfunctionalitybetweentwoxanthanswiththesamemolecularcomposition.

Impact of new findings

The effect of acetyl and pyruvate groups on the functionality of xanthan is widelydiscussed in literature and conclusions are generally based on the quantitativeanalysisofthesubstituents.Theinfluenceoftheexactdistributionofthesubstituentson the functionality of xanthan, however, has been largely ignored in functionalitystudies due to the assumption that the xanthan substitution pattern is close to theidealizedrepeatingunit.Somestudiesshowthatacetyland/orpyruvylgroupsdonot influencethexanthansfunctionality.19,20However,moststudiesshowthatacetylgroupsstabilizethexanthanconformation, reduce the interactionwith galactomannans and reduce the xanthanviscosity,whilepyruvylgrouphavethereversedeffect.21‐23Theeffectofacetylgroupsis attributed to increased association of the side chains with the backbone, due to


51

hydrogenbondingbetweenthexanthanbackboneandtheacetylgroupsontheinnermannose.24‐26Our study shows that the substitution pattern within the xanthan side chains isdifferentthangenerallyassumedandvariesbetweenxanthansamples.ItisthereforelikelythattherelativeabundanceofthedifferentRUsneedstobetakenintoaccountwhen studying the effect of the substituents on xanthan’s functionality, especiallyregarding the substitutionof theoutermannoseunit.Because it is nowshown thatabout11%ofallacetylgroupsarepositionedontheoutermannose,conclusionsontheeffectofacetylgroupsonxanthansfunctionalityshouldbereconsidered.

CONCLUSIONS

In this research an analytical method is introduced to characterize and comparexanthan samples. Theunambiguous identification andquantificationof six differentRUs,demonstratethatsubstitutionontheoutermannoseunitismuchmoreabundantthangenerallyassumed.Inadditiontothe66‐88%oftheoutermannoseunitsthatarepyruvylated, 5‐21% of the outer mannose units are acetylated. Comparison ofdifferentxanthansshowedthattheratioinwhichthesixrepeatingunitsarepresent,differsbetweenxanthansamplesevenwhenthemolecularcomposition issimilar. Itis, therefore, concluded that thecharacterizationof xanthansamples should includetherelativeabundanceoftheRUspresent,aswellasthemolecularcompositionofaxanthansample.

ACKNOWLEDGMENTS This research was supported by the European Community within a consortiumPolyModEKBBE‐2007‐3‐3‐07andisgratefullyacknowledged.

Chapter 3

52

REFERENCES



[3] Peters HU, Suh IS, Schumpe A, Deckwer WD. The pyruvate content of xanthan polysaccharideproducedunderoxygenlimitation.BiotechnolLett.1993;15:565‐6.





[8] Kool MM, Schols HA, Delahaije RJBM, Sworn G, Wierenga PA, Gruppen H. The influence of theprimary and secondary xanthan structure on the enzymatic hydrolysis of the xanthanbackbone.CarbohydrPolym.2013;97:368‐75.

[9] Kühnel S, Schols HA, Gruppen H. Aiming for the complete utilization of sugar‐beet pulp:Examination of the effects of mild acid and hydrothermal pretreatment followed by enzymaticdigestion.BiotechnolBiofuels.2011;4:1‐14.

[10] Leijdekkers AGM, Sanders MG, Schols HA, Gruppen H. Characterizing plant cell wall derivedoligosaccharidesusinghydrophilicinteractionchromatographywithmassspectrometrydetection.JChromatogrA.2011;1218:9227‐35.

[11] Hardy MR, Rohrer JS. High‐pH Anion‐Exchange Chromatography (HPAEC) and PulsedAmperometricDetection(PAD)forcarohydrateanalysis.In:KamerlingJP,BoonsGJ,LeeYC,SuzukiA,TaniguchiN,VoragenF,eds.Comprehensiveglycoscience‐Fromchemistrytosystemsbiology.Oxford,UK:ElsevierLtd.,2007;303‐24.


[13] DomonB,CostelloCE.AsystematicnomenclatureforcarbohydratefragmentationsinFAB‐MS/MSspectraofglycoconjugates.GlycoconjugateJ.1988;5:397‐409.

[14] Hilz H, de Jong LE, KabelMA, Schols HA, Voragen AGJ. A comparison of liquid chromatography,capillary electrophoresis, andmass spectrometrymethods to determine xyloglucan structures inblackcurrants.JChromatogrA.2006;1133:275‐86.

[15] Remoroza C, Cord‐Landwehr S, Leijdekkers AGM, Moerschbacher BM, Schols HA, Gruppen H.CombinedHILIC‐ELSD/ESI‐MSnenables the separation, identificationandquantificationof sugarbeetpectinderivedoligomers.CarbohydrPolym.2012;90:41‐8.

[16] Flores F, Torres LG, Galindo E. Effect of the dissolved oxygen tension during cultivation of X.campestrisontheproductionandqualityofxanthangum.JBiotechnol.1994;34:165‐73.

[17] JanaAK,GhoshP.Effectofcitricacidonthebiosynthesisandcompositionofxanthan. JGenApplMicrobiol.1999;45:115‐20.

[18] Davidson IW. Production of polysaccharide by Xanthomonas campestris in continuous culture.FEMSMicrobiolLett.1978;3:347‐9.



53



[22] RenouF,PetibonO,MalhiacC,GriselM.Effectofxanthanstructureon its interactionwith locustbeangum:Towardpredictionofrheologicalproperties.FoodHydrocolloid.2013;32:331‐40.





Chapter 3

54

Marijn M. Kool; Harry Gruppen; Graham Sworn and Henk A. Schols. Accepted for publication in Carbohydrate Polymers

Chapter 4

The influence of the six constituent xanthan repeating units on the order‐disorder transition of xanthan, based on the cellulase degradation of disordered xanthan segments

ABSTRACT

Xanthansoccurringindifferentlevelsofdisorderedconformationwereenzymaticallyhydrolyzed to their six pentamer repeating units (RUs). The RUs present in theenzymedigestswereanalyzedusingLC‐MS.Asonlydisorderedxanthansegmentsaredegraded by cellulases, the influence of the six different RUs on the transitionalbehavior of xanthan could be studied. The results indicate that especially xanthansegments rich in RUs that are acetylated on the outer mannose unit stabilize thexanthan conformation. Acetylation of the inner mannose did not show to have astabilizing effect on the xanthan conformation. As the enzymatic release ofpyruvylated RUs gradually increased with increasing levels of disordered xanthansegments, it is concluded that the distribution of these RUs is random. On thecontrary, xanthan segments rich in single or double acetylated RUs were instantlyhydrolyzedindicatingablockwisedistributionoftheseRUs.

Chapter 4

56

INTRODUCTION

Xanthanisapolysaccharidethatgivessolutionswithhighpseudoplasticflows,whendissolved inwater.1 This solution property holds over a wide pH and temperaturerange, making xanthan very suitable as viscosifier and/or stabilizer for the foodindustry.2, 3 The stability of a xanthan solution is mainly addressed to the helicalconformation of xanthan.4 The stability of this conformation depends on thetemperatureandsaltconcentrationofthesolventaswellasontheprimaryxanthanstructure.5‐7 Xanthan is a bacterial exo‐polysaccharide that consists of a β‐(14)linked glucose backbone, with a (31) linked α‐D‐mannose‐(21)‐β‐D‐glucuronicacid‐(41)‐β‐D‐mannose side chain attached to every other glucose unit (Figure4.1.).8Variationsintheprimaryxanthanstructurearemainlyduetothesubstituentspresentinthesidechains.Onaverage85%ofallinnermannoseunitsareacetylatedattheO‐6positionand50%ofalloutermannoseunitscarryapyruvategroup.9,10Thepresence of acetyl groups is reported to stabilize the helical conformation,2 whilepyruvategroupsdestabilizethehelicalconformation.5,11Itisgenerallyassumedthatonlytheinnermannoseunitcanbeacetylatedandthestabilizingeffectoftheacetylgroups is directed to hydrogen bonds formed with the xanthan backbone.12, 13However,inarecentstudyweshowedthatxanthanconsistsof6differentrepeatingunits and that depending on the production conditions, 5‐20% of all xanthan sidechainsareacetylatedontheoutermannoseunits.14

Figure 4.1. The xanthan repeating unit.

O

O

OH

OHO

HO

O

O

O

OH

CH2OH CH2OH

CH2OR1

O

HOOH

O

COO-

OR2O

HO

OHO

CH2OR3

OH

n

R1= H or COCH3

R2= H; R3 = COCH3

R2R3 = H

R2R3 =

or

or

Influence of the six repeating units on the transitional behavior of xanthan

57

Thereby,thequestionariseswhetherthesubstitutionoftheacetylgrouponeithertheinner or outer mannose is important for its stabilizing effect on the xanthanconformation. It was also shown that the total degree of substitution on the outermannose is rather constant,while the pyruvate:acetate ratio on the outermannosedependsonthefermentationconditions14.Itis,therefore,plausibletopostulatethatthe adverse effect of acetyl and pyruvate groups on the stability of the xanthanconformationisduetothetypeofsubstitutionattheoutermannoseunit:acetateorpyruvate.Thiscouldindicatethatthelocationoftheacetylgroupwithinthesidechainindeedisimportantforthestabilityofthexanthanconformation.Ifso,itcouldbethattheresultsobtainedinpreviousstudies,thatassumethatonlytheinnermannoseunitcan be acetylated, were misinterpreted and have not always led to the properexplanation.Hence,theaimofthepresentstudyistobetterunderstandtheinfluenceof theprimaryxanthan structureand theacetylationpattern,on the stabilityof thexanthanconformation.


Xanthan samples

Three types of renatured xanthan, differing in acetyl and pyruvyl contents, wereobtainedfromDuPont(Melle,France).Themolecularcompositionofthesamplesandthe relative abundance of the repeating units present weremeasured as describedpreviously.14AnoverviewofthexanthancompositionsisgiveninTable4.1.Themolarcompositions of xanthans A and B are rather similar. Variations between thesexanthansexistintheirsubstitutionpatterns,wherexanthanAhasmoreacetylgroupsontheoutermannosethanxanthanB.XanthanCisrichinpyruvategroupsandhasarelativelylowlevelofacetylationcomparedtotheothertwoxanthans.ThistranslatesintoalowdegreeofacetylationontheoutermannosecomparedtoxanthansAandB.

Table 4.1 Molar composition of different xanthan samples

Xanthan type

Glc:Man:GlcA Molar Ratio

Acetyl content (w/w%)

Pyruvate content (w/w%)

RU‐1 RU‐2

RU‐3

RU‐4

RU‐5

RU‐6

Ac:Pyr‐ratio on the outer mannose

Xanthan A 1:0.88:0.41 5.6 4.4 19 11 2 62 2 4 1:3.1

Xanthan B 1:0.91:0.43 5.9 4.6 14 12 1 67 2 4 1:4.7

Xanthan C 1:0.94:0.45 4.8 7.3 4 2 1 77 1 15 1:18.4

glucose; mannose; glucuronic acid; acetyl groups; pyruvic acid ketal

Chapter 4

58


Solutionscontaining2mg·ml‐1xanthanAwerepreparedin0,2and10mMNaClandsolutionscontaining2mg·ml‐1xanthanBorxanthanCwerepreparedin0and10mMNaCl. Xanthanwashydrolyzedby incubating1ml of a xanthan solutionwith60µgprotein from the experimental cellulase mixture C1‐G1 from Myceliophthorathermophila C1 (Dyadic Netherlands, Wageningen, The Netherlands). The cellulasemixture was desalted prior to incubation using Micro Bio‐Spin chromatographycolumns following the company’s description (Bio‐Rad Laboratories, Hercules, CA,USA).Incubationswereperformedattemperaturesintherange30–60°Candcontinuedfor48hinordertoobtaintheendpointoftheenzymaticdegradation.Afterhydrolysis,thedigestswerecooledto6°C.

Circular dichroism

Thexanthanconformationat theenzyme incubations chosenwasdeterminedusingcirculardichroism.The transitionprofilesof the threexanthans, in0, 2 and10mMNaClsolutions,weredeterminedasdescribedpreviously15andused toestimate thefractionofdisorderedxanthan(α)inthedifferentenzymeincubationsusingEquation1 with: θt = ellipticity at a given temperature; θU = ellipticity of a completelydisorderedstructureandθF=ellipticityofacompletelyorderedstructure.16

α=1–(θt–θU)/(θF–θU) (1)

Theminimumandmaximumellipticitiesweredeterminedforeachtypeofxanthan,θFwas determined in 10 mM NaCl solutions at 15°C and θU was determined indemineralized water at 85°C. The curves obtained were normalized by the best‐fitparameters.


HPSECwasperformedonanUltimate3000system(Dionex,Sunnyvale,CA,USA)asdescribedpreviously.15Apparentmolecularmassdistributionswereestimatedusingpullulan molecular‐mass standards (Polymer Laboratories, Palo Alto, CA, USA).Xanthandigestswerecentrifuged(10,000g;10min;25˚C)priortoinjection.


59

HILIC‐ELSD–ESI‐IT‐MSn

DigestswereanalyzedusingUPLC‐ELSD‐MSnonaHILICBEHamidecolumn(WatersCorporation, Milford, MA, USA), which was coupled to an ELSD and an ESI‐MSn‐detector asdescribed elsewhere.14 The xanthandigestswere filteredusingAmicon‐0.5mL centrifugal filter deviceswith a 10 kDa cut off (EMDMillipore Corporation,Billerica, MA, USA) to remove any remaining high molecular weight xanthan. Thefiltratewasdiluted1:1 (v:v)withmilliQwater followedbya1:1 (v:v)dilutionwithacetonitrilebeforeinjectionintothesystem.TheELSDpeakareawasusedtodeterminetheratio inwhichthedifferentxanthanrepeating units (RUs)were present in the xanthan digests.14 The peak areas of thedifferentRUsoffullydegradedxanthan,obtainedafterincubationunderconditionsinwhichα=1,wereusedtodeterminethemaximumELSD‐peakareaforeachtypeofRU. As the two single acetylated RU elute simultaneously, the MS‐fragmentationpatternwasusedtodistinguishandsemi‐quantifybetweentheRUswhicharesolelyacetylatedattheinneroroutermannose.14TodeterminewhichpartofeachspecificRUwas released at a given xanthan conformation, the ELSDpeak areas of eachRUpresentindigestsobtainedatα<1,areexpressedas%ofthemaximumESLDpeakareaofthatRU.


Order‐disorder transitions

The normalized fitted graphs for the three xanthans, obtained through CD analysis,areshowninFigure4.2.ItcanbeconcludedthatxanthanCexhibitsthelowestmid‐point transition temperature (Tm). Thiswas expected, as xanthanC has the highestpyruvate content, which destabilizes the ordered conformation.5 Xanthans A and Bhave lower levels of pyruvylation and higher levels of acetylation compared toxanthanC.Thereby, theyhave ahigherTm compared toxanthanC.This correlationhasbeenreportedbeforeforotherxanthans.5,17

0

0.2

0.4

0.6

0.8

1

1.2

0 10 20 30 40 50 60 70 80 90

Fra

ctio

n d

iso

rder

ed x

anth

an (α

)

Temperature ( ̊C)

0

0.2

0.4

0.6

0.8

1

1.2

0 10 20 30 40 50 60 70 80 90

Fra

ctio

n d

iso

rder

ed x

anth

an (α

)

Temperature ( ̊C)

0

0.2

0.4

0.6

0.8

1

1.2

0 10 20 30 40 50 60 70 80 90

Fra

ctio

n d

iso

rder

ed x

anth

an (α

)

Temperature ( ̊C)

Figure 4.2. Fraction of disordered xanthan as function of temperature (˚C) in: deionized water ( ),2mM NaCl ( ) and 10 mM NaCl ( ). α = fraction of disordered conformation.

0 10 20 30 40 50 60 70 80 90

Temperature (°C)

α

1.2

1

0.8

0.6

0.4

0.2

0

α

Temperature (°C)

1.2

1

0.8

0.6

0.4

0.2

0

0 10 20 30 40 50 60 70 80 90

α

1.2

1

0.8

0.6

0.4

0.2

0

0 10 20 30 40 50 60 70 80 90

Temperature (°C)

Xanthan A Xanthan B Xanthan C

Chapter 4

60

AlthoughxanthansAandBhavearathersimilarmolecularcomposition(Table4.1.),their transitionalbehavior isdifferent:xanthanBhas lower transitiontemperaturesthan xanthanA, indicating a less stableordered conformation.5, 17This difference ismost likely the result of the differences observed in the substitution patterns ofxanthan A and B. As xanthan A has a higher acetyl:pyruvate‐ratio on the outermannosethanxanthanB(Table4.1.), it ishypothesizedthatacetylationoftheoutermannoseincreasesthestabilityofthehelicalconformation.The graphswere further used to determinewhich incubation conditions should bechosen to ensure a specific fraction of disordered conformation. An overview of allincubationconditionsandthecorrespondingxanthanconformationsisgiveninTable4.2. Table 4.2. Enzyme incubation conditions and the corresponding fraction of xanthan present in a disordered conformation (α)

Xanthan

Salt concentration (mM NaCl)

Temperature (°C)


Xanthan A 10 30 0 40 0

50 0,05

60 0,38

0 30 0,42

40 0,52

50 0,93

60 1

Xanthan B 10 30 0

40 0

50 0,11

60 0,55

0 30 0,58a

40 0,62

50 0,75

60 1

Xanthan C 10 30 0

40 0,03

50 0,32

60 0,82

0 30 0,87a

40 0,91

50 0,95a

60 1 a: HILIC‐ESLD results are not further shown, as results were equal to the results of other digests of the same xanthan obtained at a similar α.


61

Molecular weight distribution of xanthan degradation products

Forthepresentstudyitisassumedthat,whenxanthanappearsinapartlydisorderedconformation,cellulaseswillcompletelydegradethedisorderedpartstothexanthanrepeating units (RUs). The ordered parts remain as non‐degraded high molecularweight material.15 To determine whether this assumption indeed applies to the 3xanthans used in this study, all xanthan digests were analyzed for the molecularweightdistributionofthedegradationproductspresent.Nointermediatedegradationproductswereobservedintheelutionprofilesobtained(datanotshown)confirmingtheassumptionmade:disorderedpartsarecompletelydegradedtoRUsandorderedpartsremainashighmolecularweightmaterial.Thelowmolecularweightfractionsofthedigestsobtainedatvariousαcanthusbeusedtostudytheinfluenceofdifferentprimaryxanthanstructuresonthedissociationbehaviorofxanthan.

Influence of the constituent repeating units on the xanthan conformation

as monitored by cellulase fingerprinting of disordered xanthan segments

As only disordered xanthan segments are susceptible to enzymatic backbonedegradation,degradationproductspresentinaxanthandigestmusthavebeenpartofa disordered xanthan segment. Analysis of the degradation products in xanthandigests obtained at given α will, therefore, enable the correlation between thetransitional behavior of xanthan and its primary structure on RU level instead ofmolarlevel.

Contribution of the differently substituted repeating units to the stability of the

xanthan conformation

Figure 4.3. shows the appearance of the individual RUs from 3 different xanthansobtained at various levels of disordered conformation and the dependence of theirabundanceonthetypeofxanthan.CleardifferencesinrelativeabundanceoftheRUsat different conformations could be recognized for the 3 xanthans. To enable thecorrelationbetweenthexanthanconformationandtheprimaryxanthanstructure,thepresence of each individual RU in the xanthan digests obtained at α≤0.95 wasexpressed as percentage of the total amount of that RU present in xanthan (Figure4.4.).

Chapter 4

62

8 10 12 14 16 18 20 22 24

α = 0.00

α = 0.38

α = 0.52

α = 1.00

α = 0.42

α = 0.93

α = 0.05

Time (min)

EL

SD

res

po

ns

e

8 10 12 14 16 18 20 22 24

Time (min)

α = 0.11

α = 0.62

α = 0.75

α = 1.00

α = 0.00

α = 0.55

EL

SD

resp

on

se

8 10 12 14 16 18 20 22 24

α = 0.00

α = 0.03

α = 0.32

α = 0.82

α = 1.00

α = 0.91

Time (min)

EL

SD

res

po

nse

Figure 4.3. HILIC‐ELSD elution profiles of xanthan digests obtained after a 48 h incubation withcellulases at different fractions of disordered conformation. Glucose ; mannose ;glucuronic acid ; acetyl group ; pyruvic acid acetal .

Xanthan A

Xanthan B

Xanthan C

RU‐1 RU‐3

RU‐2

RU‐5 RU‐6

RU‐4


63

The results suggest 3 different possible correlations between the xanthanconformationandtheenzymaticreleaseofaRU.Anexponentialtrendisobservedforthe enzymatic release of double acetylated RU (RU‐1) and the RU which is solelyacetylated on the outer mannose (RU‐3) when α ≥0.8. Digests obtained afterincubationat60˚Careanexceptiontothistrend,whichwillbediscussedlateron.ASigmoidal trend is observed for the enzymatic release of the RU that is solelyacetylatedon the innermannose (RU‐2). The xanthan conformation range inwhichthistrendisobserveddifferswiththetypeofxanthan:xanthanA0.3≤α≤0.6;xanthanB0.5≤α≤0.65andxanthanC0.0≤α≤0.4.AlmostmaximalenzymaticreleaseofRU‐2isobtainedalreadyatα=0.7forall3xanthans.TheremainingthreeRUsshowaratherlinear correlation between the xanthan conformation and their enzymatic releasethroughouttherangeα=0‐1.BecausepyruvylatedRUs,with(RU‐4)andwithout(RU‐6) acetylation at the innermannose, show the same linear correlation between thexanthan conformation and their enzymatic release, the results indicate thatacetylation on the inner mannose does not have a significant impact on xanthan’stransitionalbehavior.Segmentsrichinacetylgroupsontheoutermannose(RU‐1andRU‐3),however,haveastrongstabilizingeffectontheorderedxanthanconformationastheyarenotpresentinchainsegmentsunfoldingeasily.Previousstudies,12,13thatassumedthatonlytheinnermannosecanbeacetylated,concludedthatacetylgroupson the inner mannose are responsible for xanthan’s conformation stabilization. Incontrast,wenowconcludethat theacetylgroupsat theoutermannosestabilize thexanthanconformation.Thesubstitutionsattheoutermannosecould,therefore,bekeyin the xanthan conformation and thus in xanthan’s physical properties. As the totaldegree of substitution on the outer mannose is rather constant between xanthansamples, this could indicate that the acetyl:pyruvate‐ratio on the outer mannosedeterminesthetransitionalbehaviorofaxanthansample.AsxanthanAcontainsmoreRU‐1andRU‐3thanxanthanB(21%and15%respectively),theseresultsareinlinewiththehighertransitiontemperaturesobservedforxanthanAcomparedtoxanthanB(Figure4.2.).Our results do not necessarily conflict with a previous study reporting that thepositionoftheacetylgroupsoneithertheinneroroutermannosedidnot influencexanthan’sfunctionality.18Thexanthansused inthatstudywereproducedbymutantstrainsandwereeithersolelyacetylatedontheinnermannoseorsolelyacetylatedontheoutermannose.Thedisadvantageoftheuseofsuchxanthansisthattheylackthepyruvate groups. Their solution behavior may, therefore, not be representative forstandardxanthan,asusedinourstudy.

Chapter 4

64

0

25

50

75

100

0 0.2 0.4 0.6 0.8 1

% degrad

ed of each RU


0

25

50

75

100

0 0.2 0.4 0.6 0.8 1

% degrad

ed of each RU


0

25

50

75

100

0 0.2 0.4 0.6 0.8 1

% degrad

ed of each RU

Fraction fo disordered conformation (α)

0

25

50

75

100

0 0.2 0.4 0.6 0.8 1

% degrad

ed of each RU


0

25

50

75

100

0 0.2 0.4 0.6 0.8 1

% degrad

ed of each RU


0

25

50

75

100

0 0.2 0.4 0.6 0.8 1

% degrad

ed of each RU


100% of this RU represents: : 19% of all RUs : 14% of all RUs: 4% of all RUs

100% of this RU represents: : 62% of all RUs : 67% of all RUs : 77% of all RUs





* = Data from digests which were obtained after incubation at 60°C and are excluded from the trend line

*

*

* *

*

*

*

*

*

*

*

*

*

*

*

**

RU‐1

RU‐2

RU‐3

RU‐4

RU‐5

RU‐6

*

Figure 4.4. Degradation pattern of the each xanthan repeating unit, expressed as % of the maximumrelease of that repeating unit, as function of the xanthan conformation based on HILIC ELSD peakarea. Different lines represent the degradation pattern of the RUs in: Xanthan A (); xanthan B () andxanthan C (). glucose ; mannose ; glucuronic acid ; acetyl group ; pyruvic acid acetal .


65

Proposed distribution pattern of the xanthan repeating units

Thetrendsobservedforthecorrelationbetweentheenzymaticreleaseofthesixtypesof RU and the xanthan conformation, provides tentative information on thedistributionofthedifferentRUsoverthexanthanbackbone.RU‐4 constitutesat least60%of allRUspresent in thexanthan samples.The lineartrend observed for the enzymatic release of this RU, indicating a rather randomdistribution of this RU, is not surprising. RU‐5 and RU‐6, each responsible for only~5%ofall sidechains, also showa linear trend for theirenzymatic releaseandarethusrandomlydistributedaswell,followingthedistributionofRU‐4.RU‐1 andRU‐3, both acetylated on the outermannose, have exponential trends fortheir enzymatic release andRU‐2,which is solely acetylatedon the innermannose,exhibitsasigmoidaltrend(Figure4.4.).BasedonthesuddenreleaseoftheseRUsatacertainxanthanconformation,itisproposedthattheseRUsaredistributedinasemiblockwisemanner.This could indicate that aminimal abundanceof theseRUs inaxanthan segment already controls the unfolding behavior of that specific segment.SegmentsenrichedinRU‐1andRU‐3onlyunfoldwhenα≥0.8andsegmentsenrichedinRU‐2allunfoldatα≤0.6.ThelinearcorrelationbetweenthereleaseofRUs4‐6andthexanthanconformationcouldthusbeexplainedbyarandomdistributionoftheseRUsoverxanthansegmentswhichareeitherenrichedwithRU‐1andRU‐3,unfoldingat high levels of disordered conformation or enriched with RU‐2, unfolding at lowlevels of disordered conformation.Amoreprecisedescriptionof thedistributionofthe different RUs cannot be given on the basis of the current research. Furtherresearchonlargerxanthanoligosaccharides,consistingofseveralRUs,shouldrevealtheexactdistributionofthedifferentRUsalongthexanthanbackbone.

Interactions stabilizing the xanthan conformation – Effect of the solvent

conditions on the release of individual repeating units.

Itisknownthatdifferenttypesofmolecularinteractionsaredifferentlyinfluencedbyionic strength and/or temperature: an increase in ionic strength especially reducesthe electrostatic repulsion forces through shielding of the negative charges; anincrease in temperature reduces electrostatic interactions and reduces hydrogenbonding up to a certain temperature.19, 20 By studying the influence of the solventconditions on the unfolding behavior of different xanthan segments it is, therefore,possible to determine which interactions are involved in the stabilization of thexanthanconformation.XanthanAwasused forthis.Basedonthe transitionprofiles(Figure4.2a.),solutionconditionswerechoseninwhichthexanthanAconformationwasconstant(atα~0.50and~0.80),butinwhichtheionicstrengthandtemperature

Chapter 4

66

of the xanthan solutions differed. The relative abundance of theRUs present in thecellulase digests obtained after incubation at the chosen conditionswas compared.TheresultsareshowninTable4.3.Thetwodigestsobtainedatelevatedtemperature(2mMNaCl 55˚C; 10mMNaCl 60˚C)were relatively rich in RUs carrying an acetylgroup on the outer mannose. The digests obtained at low temperatures indemineralizedwaterdidnotcontainsuchRUs,butwererelativelyrichinpyruvylatedRUs.Duetotheabsenceofsaltsinthesedigests,thenegativechargesofthepyruvategroupsarenot shielded.The relativelyhigh abundanceofpyruvylatedRUs in thesedigestscanthusbeexplainedbyincreasedelectrostaticrepulsion,makingtheseRUsavailableforcellulasedegradation.AssegmentsrichinoutermannoseacetylatedRUsonly unfolded at elevated solution temperatures, these type of RUs most likelystabilizethexanthanconformationthroughtheformationofhydrogenbonds.ChangesintemperaturedidnotinfluencethedissociationbehaviorofsegmentsthatarerichinRUs thatare solelyacetylatedon the innermannose. It is, therefore, concluded thatthe acetyl groups on the outer mannose units, and not on the inner mannose, areinvolvedinintra‐and/orintermolecularinteractionsthroughhydrogenbonding.

Table 4.3 Influence of the solvent conditions on the relative abundance of the repeating units released during the enzymatic xanthan degradation at fixed xanthan conformations

Incubation condition

Conformation (α)

Total degradation (%)

RU‐1 RU‐2

RU‐3

RU‐4

RU‐5

RU‐6

10 mM NaCl; 60°C 0.48 34 16 16 2 59 3 4

Millipore; 40°C 0.52 37 0 18 0 71 5 6

2 mM NaCl; 55°C 0.78 87 6 25 3 56 5 5

Millipore; 45°C 0.81 85 0 22 0 70 3 5



67

CONCLUSIONS

Thisstudyhasshown, incontrasttopreviousreports, thatespeciallythoseRUsthatare acetylated on the outer mannose unit are involved in the stabilization of theordered xanthan conformation by hydrogen bonding. Acetylation of the innermannosedoesnothaveasignificanteffectontheconformationstability.BasedonthecorrelationbetweentheαandtheenzymaticreleaseofindividualRUsitispostulatedthatpyruvylatedRUsare randomlydistributedover thexanthanbackboneand thatdoubleandsingleacetylatedRUsareorganizedinamoreorlessblockwisemanner.Furthermore, the resultspoint out that it is possible to control anddirect theexacttransitionalbehaviorofaxanthanmoleculebycontrollingthesolventconditions.

ACKNOWLEDGEMENTS

ThisresearchwassupportedbytheEuropeanCommunitywithinaconsortiumPolyModEKBBE‐2007‐3‐3‐07andisgratefullyacknowledged.

Chapter 4

68

REFERENCES

[1] JeanesA,PittsleyJE,SentiFR.PolysaccharideB‐1459:Anewhydrocolloidpolyelectrolyteproducedfromglucosebybacterialfermentation.JApplPolymSci.1961;V:519‐26.


[3] SwornG.XanthanGum.In:ImesonAed. FoodStabilisers,ThickenersandGellingAgents.Oxford,UK:Wiley‐Blackwellpublishing,2009;325‐42.

[4] Morris ER. Molecular origin of xanthan solution properties. In: Sandford PA, Laskin A, eds.ExtracellularMicrobioalPolysaccharides.Washington(DC),USA:AmericanChemicalSociety,1977;81‐9.










[14] KoolMM,GruppenH,SwornG,ScholsHA.Comparisonofxanthansbytherelativeabundanceofitssixconstituentrepeatingunits.CarbohydrPolym.2013;98:914‐21.


[16] Greenfield NJ. Using circular dichroism collected as a function of temperature to determine thethermodynamicsofproteinunfoldingandbindinginteractions.NatProtoc.2006;1:2527‐35.

[17] RinaudoM. Role of substituents on the properties of some polysaccharides. Biomacromolecules.2004;5:1155‐65.



69

[19] DamodaranS,ParkinKL, FennemaOR.Fennema'sFoodChemistry, 4edn.BocaRaton (FL),USA:CRCPressTaylorandFrancisGroup,2008

[20] CreightonTE.Proteins:StructuresandMolecularProperties.NewYork(NY),USA:W.H.FreemanandCompany,1993;139‐70.

Chapter 4

70

Submitted as: Marijn M. Kool; Henk A. Schols; Martin Wagenknecht; Sandra W.A. Hinz; Bruno M. Moerschbacher and Harry Gruppen. Submitted for publication

Chapter 5

Characterization of an acetyl esterase from Myceliophthora thermophila C1 able to deacetylate xanthan

ABSTRACT

Screening of eight carbohydrate acetyl esterases for their activity towards xanthanresulted in the recognition of one active esterase. AXE3, a CAZy family CE1 acetylxylan esteraseoriginating fromMyceliophthora thermophilaC1, removed31%of allacetyl groups present in xanthan after a 48h incubation. AXE3 activity towardsxanthan was only observed when xanthan molecules were in the disorderedconformation. Optimal performance towards xanthan was observed at 53°C in thecomplete absence of salt, a condition favoring the disordered conformation. AXE3‐deacetylatedxanthanwashydrolyzedusingcellulasesandanalyzedfor itsrepeatingunits using HILIC‐ELSD‐MSn. It showed that AXE3 specifically removes the acetylgroups positioned on the inner mannose and that acetyl groups positioned on theouter mannose are not removed at all. After a prolonged incubation at optimalconditions,60%ofalltheacetylgroups,representing75%ofallacetylgroupsontheinnermannoseunits,werehydrolyzed.

Chapter 5

72

INTRODUCTION

The bacterial polysaccharide xanthan is a polymer having a β‐1,4‐glucan backbone,carrying a glycosidically linkedα‐D‐mannose‐(21)‐β‐D‐glucuronic acid‐(41)‐β‐D‐mannose side chain on theO‐3‐position of every second glucose unit (Figure 5.1.).1Approximately85%ofallinnermannoseunitsareacetylatedattheO‐6positionand50‐70% of all terminal mannose units are substituted with a pyruvic acid ketal.Additionally,about5‐25%ofallterminalmannoseunitsaresubstitutedwithanacetylgroupattheO‐6position.2,3Theacetylandpyruvategroupsofxanthanareknowntohavealargeinfluenceontheviscosityofxanthansolutionsandtheirstabilitytowardstheadditionofsalts,changesin temperature, andvariations in solvent acidity.4‐6Lowering thedegreeof xanthanacetylationresultsinimprovedviscosityandstabilityofxanthansolutionsaswellasin improved interactions of xanthan with galactomannans.4, 7 How the position ofacetyl influences the xanthan functionality, however, remains unknown. Targetedremovalofspecificacetylgroupsfromthexanthansidechainswouldthusbeusefultofurtherexplorethefunctionalityofxanthan.Todate,acetylgroupsareremovedusinganalkalitreatment.8,9However,suchaprocessrandomlyremovesacetylgroupsandbackbonedegradationmightbeapparent.


O

O

OH

OHO

HO

O

O

O

OH

CH2OH CH2OH

CH2OR1

O

HOOH

O

COO-

OR2O

HO

OH

O

CH2OR3

OH

n

R1= H or COCH3

R2= H; R3 = COCH3

R2R3 = H

R2R3 =

or

or

An acetyl esterase from M. thermophila C1 able to deacetylated xanthan

73

Another method to control the acetyl levels in xanthan is the use of specificXanthomonasstrainsand/orfermentationconditionsforthexanthanproduction.10,11Altering the fermentation conditions of xanthan, however, also influences thepyruvate levels in the xanthan produced, which are also known to be of greatimportance for xanthans functionality.12 Therefore, bothmethods described arenotapplicable for the specific removal of acetyl groups. Targeted modification usingenzymesthatspecificallyremoveacetylgroups fromxanthanwouldbemoreuseful.However,suchenzymeshavenotbeendescribedtodate.Becausenoxanthanacetylesterases are known, targeted database mining for xanthan acetyl esterases is notpossible. Nevertheless, due to the similarity in the mode of action of differentcarbohydrateacetylesterases,theclassificationofcarbohydrateacetylesterasesisfarless specific compared to classification of the carbohydrate hydrolases andcarbohydrate lyases.13 The exact activities of carbohydrate acetyl esterases are,therefore, much less predictable compared to other carbohydrate‐active enzymes.Manycarbohydrateacetylesterases,especiallyfromCAZyfamiliesCE1,CE3andCE6,areknowntohaveana‐specificbindingsite.14,15Hence,searchingforxanthanacetylesterases could be done by testing known carbohydrate acetyl esterases for theiractivitytowardsxanthan.However,suchascreeningforxanthanmodifyingenzymesis not fully straight forward. Several studies on the enzymatic degradation of thexanthan backbone by cellulases have shown that the conformation of xanthan insolution is critical for enzymatic degradation.16‐18 In solution xanthan can adapt anorderedhelicalconformationorarandomdisorderedconformation.19,20Althoughtheexact natureof thehelical conformation is still underdebate, it is believed that thexanthansidechainsarealignedwiththexanthanbackbone.21,22Thisalignmentofsidechains when xanthan appears in an ordered conformation and/or the stacking oforder structures into a network is assumed to make xanthan resistant againstenzymaticmodifications.17,18Recently,itwasproventhatonlyxanthanmoleculesthatappear in a disordered conformation are susceptible to enzymatic backbonedegradationbycellulases.16Whenscreeningcarbohydrateacetylesteraseforpossiblesideactivitiestowardsxanthan, thexanthanconformationshouldthusbetaken intoaccount.

In this study several carbohydrate acetyl esteraseswere screened for their activitytowardsxanthanappearinginanorderedorinapartlydisorderedconformation.Thetemperatureoptimumoftheactiveenzymewasdeterminedandtheinfluenceofthexanthanconformationontheenzymeactivitywasstudiedindetail.Structuralanalysisoftheenzymaticallymodifiedxanthanwasperformedtodeterminethespecificityoftheactiveenzymewithrespecttoacetylationoftheinneroroutermannoseunit.

Chapter 5

74


Chemicals and substrates

Allchemicalsusedwere,ifnotmentionedotherwise,ofanalyticalgrade.Thexanthanused was obtained by DuPont (Melle, France) and characterized as describedpreviously.3AdetailedoverviewofthechemicalcharacterizationisgiveninTable5.1.Acetylated xylooligosaccharides were obtained and characterized as described byKoutaniemi et al.23 Sugar beet pectin (SBP6230)24 was obtained from DuPont(Brabrand,Denmark).Chitinandchitosanoligosaccharides(degreeofpolymerizationof2–6)werepurchasedfromSeikagakuCorp.(Tokyo,Japan).

Circular Dichroism

Far‐UVCDspectraof2mg·mL‐1xanthanin0,1,2,5and10mMNaClsolutionsweremeasuredat20,40and70°Cusinga Jasco‐J‐715spectropolarimeter (JASCO,Tokyo,Japan).Aquartzcuvettewithanopticalpathof1mmwasused.Thetemperaturewasregulated using a PTC‐348 WI controller (JASCO). In the 190‐300 nm wavelengthregion (0.2 nm resolution) 10 scans were accumulated with a scan rate of 100nm·min−1 anda time constantof 0.125 s.The final spectra are the averageof thesescans. Prior to the spectral analysis the wavelength scans were corrected for thebufferbackgroundsignal.Previousresearchhasshownthatthedecreaseinellipticity(θ)at219nmcorrelatesalmostlinearlywiththefractionofxanthanpresentinthedisorderedconformation.16,21 The fraction of disordered conformation (α) was, therefore, estimated usingEquation1with:θs=ellipticityofthesampleat219nm;θU=ellipticityofacompletelydisorderedstructureat219nmandθF=ellipticityofacompletelyorderedstructureat219nm.

α=(θs–θF)/(θU–θF) (1)

TheellipticityofθFwasdeterminedin10mMNaClsolutionsat20°CandtheellipticityofθUwasdeterminedinMilliporewater70°C.


75

Carbohydrate acetyl esterases

Anoverviewofthecarbohydrateacetylesterasestested,theiroriginandtheirknownsubstrate specificities is given in Table 5.2. AXE2 and AXE3,25, 26 both produced byMyceliophthora thermophila C1 (formerly termed Chrysosporium lucknowense C127),wereobtainedfromDyadic(Wageningen,TheNetherlands).AnAXEandRG‐04wereextractedfromAspergillusspp.preparationsasdescribedbyKormelink28andSearle‐vanLeeuwen,29respectively.ThecodingsequencesofCDA‐II,CDA‐III,PAE2andPAE4were cloned in pET22b‐StrepIIc, a pET‐22b(+) (Novagen, Merck KGaA, Darmstadt,Germany) derivative that additionally contains a sequence coding for the StrepIIaffinity tag, and heterologously expressed in E. coli Rosetta 2(DE3)(pLysSRARE2)(Novagen)usingauto‐inductionmedium.30Enzymepurificationwasdonebyaffinitychromatography using a 1ml Strep‐Tactin Superflow Plus Column (Qiagen, Hilden,Germany)asdescribedelsewhere.31

Enzyme assays

AllenzymesweredesaltedpriortotestingtheiractivitytowardsxanthanusingMicroBio‐Spin chromatography columns following the company’s description (Bio‐RadLaboratories,Hercules,CA,USA).The carbohydrateacetyl esteraseswerescreened for their activity towardsxanthanbyincubating1mLofa2mg·mL‐1xanthansolutionwith8–35μgenzymeat40°Cfor48h.TheenzymeactivitywastestedinMilliporewater,10mMNaClsolutionsand50mM sodium citrate buffer (pH6.0). The acetic acid releasewasdetermined using aMegazymeaceticacidkit(Megazyme,Wicklow,Ireland)andexpressedaspercentageof the total acetyl content present in the parental xanthan. The company’s protocolwasdownscaledtomicrotiterplatescaletoenablemediumthroughputanalysis.Thetotalacetylcontentwasdeterminedbytheanalysisoftheaceticacidreleasedafterasaponification step with 1M NaOH (18 h; 4˚C). All incubations were performed induplicate.

Table 5.1 Chemical characterization of xanthan




RU‐1 RU‐2

RU‐3

RU‐4

RU‐5

RU‐6

Xanthan 1 : 0.88 : 0.41 5.6 4.4 19 11 2 62 2 4


Chapter 5

76

Table 5.2. Characterization of the carbohydrate acetyl esterases used in this study, including abbreviations, origin, Gen

e bank

accession number, CAZy fam

ily and references

Known activity

Abbreviation

Origin

Gen

eBank

Accession number

CAZy

family

Referen

ce

Acetyl xylan

esterase

AXE2

Myceliophthora thermophila

C1

HQ324256

CE‐5

Pouvreau (2011); Hinz (2009)

AXE3

Myceliophthora thermophila

C1

HQ324257

CE‐1

Pouvreau (2011); Hinz (2009)

AnAXE

Aspergillus nige r

unknown

CE‐1

Korm

elink (1993)

Chitin deacetylase

CDA‐II

Bacillus licheniform

is DSM

13

AAU39149

CE‐4

Not available

CDA‐III


is DSM

13

AAU39762

CE‐4

Not available

Pectin acetyl esterase

PAE2

Pectobacterium atrosepticum SCRI1043

CAG75311

Unknown

Not available

PAE4


is DSM

13

AAU42913

CE‐12

Rem

oroza (2013)

RG‐04

Aspergillus aculeatus

unknown

CE‐12

Searle‐van

Leeu

wen (1992)


77

AcetylesteraseAXE3wasfurthercharacterizedforitsactiononxanthan.Thepreciseinfluence of the xanthan conformation on the enzyme activity was determined byincubating2mg·mL‐1xanthaninMilliporewater,1mM,2mM,5mMand10mMNaClsolution with 25 μg of protein at 40°C for 48 h. The temperature optimum wasdeterminedinasaltfreeenvironmentwithinthetemperaturerange35–60°Caftera24h incubation.The specific activityofAXE3 towards the xanthanwasdeterminedafter a 24 h incubation at 55˚C in a salt free environment. A 24 h incubation forxanthanwaschosenasthereactionstillisinthelinearrangeandshorterincubationtimeswouldnotreleasesufficientamountsofaceticacidtoaccuratelydeterminethespecificactivity.Allincubationswereperformedinfourreplicates.ThemodeofactionofAXE3towardsxanthanwasdeterminedbystructuralanalysisofunmodified and enzymatically deacetylated xanthan. Solutions of the latter weredialyzed against Millipore water and lyophilized. The obtained xanthan wasredissolved in Millipore water (2 mg·mL‐1) and incubated with the cellulasepreparation C1‐G1 from Myceliophthora thermophila C1 (Dyadic Netherlands,Wageningen,TheNetherlands)at60°Cfor48h.16Thexanthandigestsobtainedwereanalyzed for their repeating units using HILIC‐ELSD‐ESI‐IT‐MSn as describedpreviously.3


Screening for xanthan acetyl esterase activity

All acetyl esterases (AEs)available inour laboratorieswere tested for their activitytowards xanthan. The enzyme activities of the AEswere tested at different solventconditionsinordertoanalyzetheactivitytowardsdifferentxanthanconformations.None of the pectin and chitin AEs (Table 5.2.)was able to release acetic acid fromxanthanatallincubationconditionstested(datanotshown).Inthepresenceofsalts,when xanthan appears in a completely ordered conformation,16 none of xylan AEsreleasedsignificantamountsofaceticacidfromxanthaneither.Intheabsenceofsalts,AXE3 (CE1 family) was able to release approximately 30% of all acetyl groups.Previousexperiments16showedthatundertheseconditionsapproximately80%ofallxanthan molecules appears in a disordered conformation. It is, therefore, likely toassumethatAXE3canonlydeacetylatexanthanthatisinadisorderedconformation,probably because the acetyl groups are only accessible in that conformation.Consequently,enzymaticdeacetylationofxanthanmightonlybepossiblebythoseAEsthat are activeunder conditions that favor thedisordered conformation.Hence, theactivity of all AEs towards their model substrate (acetylated sugar beet pectin,acetylatedxylanoligosaccharidesorchitin/chitosanoligosaccharides)wastestedina

Chapter 5

78

saltfreesolutionat40˚C.Noneoftheenzymes,exceptAXE3,wasactiveagainsttheirknown specific substrate at this condition, while all enzymes showed the expectedactivityinthepresenceofsalt(datanotshown).ExceptforAXE3,noneoftheenzymesis, therefore, active under conditions favoring the disordered conformation, whichcouldexplainwhyonlyAXE3wasfoundtobeactivetowardsxanthan.Previous research towardsAXE2 (CE5 family) andAXE3 (CE1 family) both fromM.thermophila C1, showed that both enzymes, although from different CE families,showed similar activity towards acetylated xylan oligosaccharides.25, 26 However,substratespecificitytestsshowedthatAXE2hasamorespecificmodeofaction.Moregeneral studies including acetyl xylan esterases from different CE families showedthat enzymes fromCE family1, towhichAXE3belongs, havea lower specificity forxylans than acetyl esterases from other CE families.14 The difference observedbetween the activities towards xanthanofAXE2andAXE3, even though theyoriginfrom the same bacterial host, can probably be explained by differences in theirsubstratespecificityand/orsubstratebinding.

Influence of the xanthan conformation on the acetic acid release by AXE3

To be able to determine the influence of the xanthan conformation on the AXE3activityinmoredetail,theconformationofxanthanat40˚CinsolutionswithdifferentNaCl concentrations was analyzed using circular dichroism (CD). The measuredellipticity (θ)at219nmand thecorresponding fractionofdisorderedconformation(α)aregiven inTable5.3.As theexactxanthanconformation isnowknown for thedifferent incubationconditions, it ispossible to correlateα to theabilityofAXE3 toreleaseaceticacidfromxanthan(Figure5.2).Withanincreasingα,anincreaseintheacetic acid release is observed, indicating that AXE3 is indeed only active towardsdisorderedxanthanfragments.Severalstudiesonthexanthanconformationproposedthattheacetylgroupspositionedontheinnermannoseunitshowinteractionwiththexanthanbackbone.Table 5.3. The ellipticity at 219 nm (θ219) and the corresponding fraction of disordered conformation (α) of xanthan (2 mg∙mL‐1) under different solvent conditions derived from far‐UV CD spectra

Sample condition θ219 α

Millipore water 70˚C ‐3.89 1.00 Millipore water 40˚C ‐3.02 0.78

1 mM NaCl 40˚C ‐1.37 0.35

2 mM NaCl 40˚C ‐0.93 0.24

5 mM NaCl 40˚C ‐0.55 0.14

10 mM NaCl 40˚C ‐0.20 0.05

10 mM NaCl 20˚C 0.00 0.00


79

Figure 5.2. Influence of the xanthan conformation (α) on the acetic acid release from xanthan after a 48 h incubation with AXE3 at 40˚C.

These acetyl groups would, therefore, be positioned at the inside of the xanthanhelix21,32,33andwouldthusnotbeaccessibleforacetylesteraseswhenxanthanisintheorderedconformation.In a recent study we showed that, depending on the xanthan production process,approximately 20% of all acetyl groups can be positioned on the outer mannose.3Whether these acetyl groups also fold to the inside of the helical conformation isunknown. However, it is known that xanthan lyases are able to remove the outermannoseinthepresenceofsalts,34,35indicatingthattheoutermannoseisaccessibleforenzymeswhenxanthanappearsinanorderedconformation.BecauseAXE3isonlyactivetowardsdisorderedxanthanfragments, it isexpectedthatAXE3isspecificfortheremovaloftheacetylgroupsfromtheinnermannoseunit.

Characterization of the AXE3 activity towards xanthan

Temperature optimum

Figure 5.3. shows the temperature optima of AXE3 towards xanthan in a salt freeenvironmentandtowardsacetylatedxylanoligosaccharidesatitsoptimalpHof7.0.26AXE3hasa clear temperatureoptimum forxanthandeacetylationbetween50‐55˚C,whereasabroaderand lowertemperatureoptimum(35‐45˚C)wasobservedforthedeacetylation of xylan oligosaccharides at pH 7.0.With increasing temperature, thexanthanconformationchangestoamoredisorderedstructure,whichisnecessaryforAXE3tobeactive.

0

10

20

30

40

50

0 0.2 0.4 0.6 0.8 1

Acetic acid release (% of total)


Chapter 5

80

Figure 5.3. Temperature optimum of AXE3 towards xanthan () and acetylated xylan oligosaccharides () at their optimal conditions. The data for acetylated xylan are based on Pouvreau et al.26

The higher temperature optimum observed for the AXE3 deacetylation of xanthan,compared to the deacetylation of xylan might, therefore, be the result of bothincreased substrate accessibility and the enzyme inactivation at elevatedtemperatures.

Specific activity

The specific activity of AXE3 towards xanthan was determined at its optimalconditions and compared to the specific activity towards xylan as determined byPouvreau et al.26 The specific activity towards acetylated xylooligosaccharides is8.3 U∙mg protein‐1. The specific activity towards xanthan is 600x lower: 13mU∙mgprotein‐1. It is, therefore, concluded that although the enzyme can remove acetylgroupsfromxanthan,theannotationoftheenzymebeinganacetylxylanesteraseisfullycorrect.

Characterization of the enzymatically modified xanthan

Xanthanwaspartlydeacetylatedbyincubatingxanthanat55˚CwithAXE3for1,2and3days,resultinginthereleaseof17%,47%and58%ofallacetylgroups,respectively.Subsequently,therelativeabundanceoftherepeatingunits(RUs)presentincellulasedigestsof thepartlydeacetylatedxanthanswasanalyzedusingHILIC‐ELSD‐MS2andcompared to the relative abundance of the RUs present in the cellulase digest ofunmodifiedxanthan(Figure5.4.).Quantificationoftheexactmodificationsintroducedby AXE3, based on the ELSD response, remains difficult due to a lack of properstandards.

0

20

40

60

80

100

120

30 35 40 45 50 55 60 65

Re

lati

ve a

ctiv

ity

(%)

Temperature (˚C)


81

8 10 12 14 16 18 20 22 24

Time (min)

ELS

Dre

spon

se

3 days modified xanthan

2 days modified xanthan

1 day modified xanthan

unmodified xanthan

1 3

2

4

5 6

A

0

10

20

30

40

50

60

70

0 1 2 3

Rel

ativ

e ab

unda

nce

of e

ach

RU

(%

)

Modification time (days)

B

Figure 5.4. AXE3 modification of xanthan followed in time. A) HILIC‐ELSD elution profiles of cellulasedigests of unmodified xanthan and xanthan modified with AXE3 for 1 day, 2 days or 3 days. Glucose ;mannose ; glucuronic acid ; acetyl groups ; pyruvic acid ketal . B) Relative abundance of the 6xanthan repeating units present in cellulase digests of xanthan at different levels of AXE3 modification.RU‐1 (); RU‐2 (); RU‐3 (*); RU‐4 (); RU‐5 (); RU‐6 (●)

Chapter 5

82

The observed trend of modification, however, is absolutely clear: After a one daytreatmentwithAXE3,thedoubleacetylatedRUsandtheacetylated+pyruvylatedRUsdecreased slightly in their abundance. The amount of single acetylated RUs,unsubstituted RUs and pyruvylated RUs slightly increased. Extending the AXE3treatment resulted in a further increase of the unsubstituted RUs and pyruvylatedRUs,whilealltheotherRUsdecreaseintheirabundance.Aftera3daytreatmentwithAXE3almostnodoubleacetylatedRUsoracetylated+pyruvylatedRUsare left intheAXE3deacetylatedxanthan.However,singleacetylatedRUsremainpresent.AsAXE3ishypothesizedtobespecificforthedeacetylationoftheinnermannose,thedoubleacetylated repeating unit would be converted into single acetylated RUs that areacetylated on the outer mannose. Simultaneously, single acetylated RUs that areacetylated on the innermannose are converted into unsubstituted RUs. As the twosingleacetylatedRUselutesimultaneouslyfromtheHILICcolumn,3nocleardecreasein the total amount of single acetylated RUswould be observed in the HILIC‐ELSDprofile. Nevertheless, changes in the ratio between the two single acetylated RUs,induced by the AXE3 modification, can be determined using the MS‐fragmentationpattern,as innermannoseacetylatedRUsandoutermannoseacetylatedRUshaveadifferentsetofdiagnostic fragment ions.3The intensityof thesedifferent fragments,was used to determine the ratio in which the two single acetylated were presentbefore and after AXE3modification (Figure 5.5.). Before modification, ~10% of allsingleacetylatedRUsareacetylatedontheoutermannose,whilethisvalueincreasedto~50%aftera3daymodification.ThisindicatesthattheRUsthatareacetylatedontheoutermannoseaccumulateduringAXE3deacetylation.It is,therefore,concludedthatAXE3ishighlyspecificfortheremovalofacetylgroupsontheinnermannoseunitandthattheacetylgroupsontheoutermannosearenothydrolyzed.Characterizationof the unmodified xanthan structure showed that ~80% of all acetyl groups aresubstitutedtotheinnermannose.Aftera3dayincubation~60%ofallacetylgroupsareremovedbyAXE3,whichcorrespondstotheremovalof~75%ofallacetylgroupspositionedontheinnermannoseunit.Whethercompleteremovaloftheacetylgroupsonthe innermannose ispossibleremainsuncertain,as the incubationtimewasnotfurtherextendedinthisresearch.


83

¤ Fragments specific for the acetylation on the outer mannose

* Fragments specific for the acetylation of the inner mannose

Figure 5.5. MS2‐fragmentation pattern of the single acetylated repeating unit (eluting at 12.5 min in Figure 5.4a.) present in the cellulases digests of A) unmodified xanthan and B) 3 days AXE3 modified xanthan.

CONCLUSIONS

Acetylxylanesterase3(AXE3)aCEfamily1esteraseoriginatingfromMyceliophthorathermophilaC1, iscapableof removing60%ofallacetylgroups isxanthan.Enzymeactivitywasonlyobservedwhenxanthanispresent inthedisorderedconformation.AlthoughthespecificactivityofAXE3towardsxanthanisverylow,itisthefirstacetylesterase reported that is active towards xanthan. Structural characterization of theAXE3modified xanthan showed that AXE3 is specific for the removal of the acetylgroupspositionedontheinnermannoseunit,althoughaftera3dayincubationnotallacetylgroupswereremoved.Asthexanthanconformationshowedtobeimportantforthe enzymatic deacetylation of xanthan, it is concluded that screening for potentialxanthanAEsshouldincludeincubationconditionsthatsupportboththedisorderedaswellastheorderedxanthanconformation.

ACKNOWLEDGEMENTS

ThisresearchwassupportedbytheEuropeanCommunitywithinaconsortiumPolyModEKBBE‐2007‐3‐3‐07andisgratefullyacknowledged.

300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000

m/z

Rel

ativ

e A

bu

nd

an

ce

721

703

541

823

625805661499 643583559 601

300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000

m/z

Re

lati

ve A

bu

nd

an

ce

721

703823661541

499

805625583

643559 601

A

B

*

*

*

*

*

*

*

*

¤¤ ¤ ¤

¤¤

¤¤

Chapter 5

84

REFERENCES





[5] ShatwellKP,Sutherland IW,Ross‐MurphySB,Dea ICM. Influenceof theacetyl substituenton theinteraction of xanthan with plant polysaccharides ‐ II. Xanthan‐guar gum systems. CarbohydrPolym.1991;14:115‐30.



[8] BradshawIJ,NisbetBA,KerrMH,SutherlandIW.Modifiedxanthan‐itspreparationandviscosity.CarbohydrPolym.1983;3:23‐38.




[12] SandfordPA,PittsleyJE,KnutsonCA,CadmusMC,WatsonPR,JeanesA.VariationinXanthomonascampestrisNRRLB‐1459;Characterisationofxanthansamplesofdifferentpyruvicacidcontent.In:Sandford PA, Laskin A, eds. Extracellular Microbial Polysaccharides.Washington (DC) USA: ACS,1977;192‐210.

[13] Davies GJ, Gloster TM, Henrissat B. Recent structural insights into the expanding world ofcarbohydrate‐activeenzymes.CurrOpininStrucBiol.2005;15:637‐45.

[14] Biely P. Microbial carbohydrate esterases deacetylating plant polysaccharides. Biotechnol Adv.2012;30:1575‐88.

[15] TenkanenM, Eyzaguirre J, Isoniemi R, Faulds CB, Biely P. Comparison of catalytic properties ofacetylxylanesterasesfromthreecarbohydrateesterasefamilies.In:SaddlerM,ed.ApplicationofEnzymestoLignocellulosics.Washington(DC),USA:ACS,2003;211‐29.







85



[23] Koutamiemi S, Gool MP, JuvonenM, Jokela J, Hinz SWA, Schols HA, TenkanenM. Importance ofcarbohydrate esterase familie 16 acetyl esterases in xylan deactylation and total hydrolysis. JBiotechnol.2013;acceptedforpublication

[24] BuchholtHC,ChristensenTMIE,FallesenB,RaletM‐C,Thibault J‐F.Preparationandpropertiesofenzymaticallyandchemicallymodifiedsugarbeetpectins.CarbohydrPolym.2004;58:149‐61.

[25] Hinz SWA, Pouvreau L, Joosten R, Bartels J, Jonathan MC, Wery J, Schols, HA. HemicellulaseproductioninChrysosporiumlucknowenseC1.JCerealSci.2009;50:318‐23.

[26] PouvreauL,JonathanMC,KabelMA,HinzSWA,GruppenH,ScholsHA.Characterizationandmodeof action of two acetyl xylan esterases from Chrysosporium lucknowense C1 active towardsacetylatedxylans.EnzymeMicrobTechnol.2011;49:312‐20.

[27] Visser H, Joosten V, Punt PJ, et al. Development of a mature fungal technology and productionplatformforindustrialenzymesbasedonaMyceliophthorathermophila isolate,previouslyknownasChrysosporiumlucknowenseC1.IndBiotechn.2011;7:214‐23.

[28] KormelinkFJM,LefebvreB,StrozykF,VoragenAGJ.PurificationandcharacterizationofanacetylxylanesterasefromAspergillusniger.JBiotechnol.1993;27:267‐82.

[29] Searle‐van LeeuwenMJF, Broek LAM, Schols HA, Beldman G, Voragen AGJ. Rhamnogalacturonanacetylesterase: a novel enzyme fromAspergillusaculeatus, specific for the deacetylation of hairy(ramified)regionsofpectins.ApplMicrobiolBiotechnol.1992;38:347‐9.

[30] StudierFW.Proteinproductionbyauto‐inductioninhigh‐densityshakingcultures.ProteinExpresPurif.2005;41:207‐34.

[31] Remoroza C,Wagenknecht M, Gu F, Buchholt HC, Moerschbacher BM, Gruppen H, Schols HA. ABacilluspectinacetylesteraseisspecificfordeacetylationofhomogalacturonan’sacetylatedatO‐3.AcceptedforpublicationinCarbohydratePolymers.



[34] HashimotoW,Miki H, Tsuchiya N, Nankai H,Murata K. Xanthan lyase ofBacillus sp. strain GL1liberatespyruvylatedmannosefromxanthansidechains.ApplEnvironMicrob.1998;64:3765‐8.


Chapter 5

86

Marijn M. Kool; Martin Wagenknecht; Bruno M. Moerschbacher; Harry Gruppen and Henk A. Schols. To be submitted

Chapter 6

Characterization of an acetyl esterase from Bacillus subtilis strain 168 able to deacetylate the outer mannose of xanthan

ABSTRACT

YesY, a pectin acetyl esterase originating fromBacillus subtilis strain 168, removed22% of all acetyl groups present in xanthan after a 24h incubation. YesY activitytowards xanthan was mainly observed in the presence of salts, when xanthanmolecules appear in the ordered conformation. Optimal performance towardsxanthanwasobservedatpH6.3and55°C.YesYdeacetylatedxanthanwashydrolyzedusing cellulases and the digest was analyzed for its repeating units present. It wasshown that YesY is specific for the removal of the acetyl groups positioned on theouter mannose and that acetyl groups positioned on the inner mannose are notremoved at all. After extended incubation, all acetyl groups on the outer mannoseresidueswerehydrolyzed.

Chapter 6

88

INTRODUCTION

Xanthan, the exopolysaccharide produced by Xanthomonas spp., is widely used inindustry as rheology modifier and food stabilizer.1, 2 In solution, xanthan adapts asecondaryorderedconformation,3,4makingtheviscosityofxanthansolutionsratherstable over awidepH and temperature range and towards addition of salts.2, 5 Thestability of this secondary conformation towards the addition of salt and towardschanges in temperature is strongly affected by the primary xanthan structure.6‐8Xanthanhasaβ‐1,4‐linkedglucanbackbonewith trisaccharidesidechains linkedtoevery other glucose unit (Figure 6.1.). The side chains consist of (31) linkedα‐D‐mannose‐(21)‐β‐D‐glucuronicacid‐(41)‐β‐D‐mannoseunits.9The innermannoseunit is mostly acetylated and the outer mannose can be pyruvylated (~65%),acetylated(~15%)orunsubstituted(~20%).10Removalofpyruvategroupsresultsinsolutionswith a lower butmore stable viscosity.11, 12 The removal of acetyl groupsresults in stronger interactions with galactomannans and gives rise to xanthansolutions with increased viscosity.13, 14 Furthermore, the viscosity stability of thesolution, especially at low pH, increases significantly due to deacetylation.8Consequently,theproductionofxanthanwithcontrolledacetyllevelscouldbeusefultobroadenxanthansapplicationsinindustry.


O

O

OH

OHO

HO

O

O

O

OH

CH2OH CH2OH

CH2OR1

O

HOOH

O

COO-

OR2O

HO

OH

O

CH2OR3

OH

n

R1= H or COCH3

R2= H; R3 = COCH3

R2R3 = H

R2R3 =

or

or

An acetyl esterase from B. subtilis strain 168 able to deacetylate xanthan

89

Acetylgroupscaneasilyberemovedfromxanthanbysaponification.15However,thisprocess is random and the removal of pyruvate groups and degradation of thexanthan backbone can also occur, which will result in a loss in viscosity. A morecontrolled method to influence the acetyl levels in xanthan is the production ofxanthanusingmutantstrains.14Adisadvantageoftheuseofmutantstrainstocontrolthe acetyl levels in xanthan is that partial removal of the acetyl groups is notpossible.14Furthermore,mutationsinonepartoftheXanthomonasgenomecanaffectother aspects of the xanthan biosynthesis as well, e.g. degree of pyruvylation,molecularweight.16Completecontrolofthexanthanstructureproducedis,therefore,not possible. Additionally the use of xanthans produced by genetically modifiedstrains is prohibited in Europe, that limits the possible application of the non‐acetylatedxanthansproduced(Regulation(EC)No1830/2003).Enzymaticremovaloftheacetylgroupscouldbeabettermethodfortheproductionofxanthanswithlowlevelsofacetylation.However,todateonlyoneenzymehasbeendescribed,thatcanpartlydeacetylatexanthan.17Thisenzymeexclusivelyremovestheacetylgroupsfromtheinnermannoseunitandisonlyactivetowardsxanthanintheabsence of salt, when xanthan appears in the random disordered xanthanconformation.Recently,apectinacetylesterase fromBacillussubtilisstrain168wasdescribed,whichwasfoundtohaveasideactivitytowardsxanthan.18Inthepresentstudythisnewenzymeisfurthercharacterizedforitsactivitytowardsxanthan.


Chemicals and substrates

Allchemicalsusedwere,ifnotmentionedotherwise,ofanalyticalgrade.ThexanthanusedwaskindlyprovidedbyDuPont (Melle,France)andcharacterized.10AdetailedoverviewofthechemicalcharacterizationisgiveninTable6.1.

Table 6.1. Chemical characterization of xanthan




RU‐1 RU‐2

RU‐3

RU‐4

RU‐5

RU‐6

Xanthan 1 : 0.88 : 0.41 5.6 4.4 19 11 2 62 2 4


Chapter 6

90

Enzyme assays

ThecodingsequenceofYesY19wasclonedinpET22b‐StrepIIc(Novagen,Merck,KGaA,Darmstadt, Germany), and heterologously expressed in E. coli Rosetta2(DE3)(pLysSRARE2) (Novagen). Enzyme purification was done as describedelsewhere.20Enzyme incubationswere initiallyperformedat twoxanthan concentrations to ruleout theeffectofahighsolutionviscosityon theenzymeactivityathigher substrateconcentrations. Xanthan solutions, 2 mg·mL‐1 and 5 mg·mL‐1, were incubated with4μgYesY·mgxanthan‐1.Incubationswereperformedin50mMcitricacidbufferpH6for24hat40°C.The effect of the enzyme to substrate ratio on the acetic acid release by YesYwasdetermined by incubation of 2mg·mL‐1 xanthan solutions in 50mM sodium citratebuffer pH 6.0, with enzyme concentrations ranging from 5.4 μg to 81 μg YesY·mgxanthan‐1Theinfluenceofthexanthanconformationontheenzymeactivitywasdeterminedbyincubating2mg·mL‐1xanthaninMilliporewater,1mM,2mM,5mMand10mMNaClsolutionswith54μgYesY·mgxanthan‐1at40˚Cfor24h.ThepHofthesolutionswas~5.6. As control, the influence of all the solution conditions on the YesY activitytowardspectinwasalsoinvestigated.The temperature optimumwas determined in 50mM sodium citrate buffer pH 6.0within the temperature range 30‐80°C after a 6 h incubation of 2mg·mL‐1 xanthansolutionswith54μgYesY·mgxanthan‐1.ThepHoptimumwasdeterminedin50mMMcIlvainbufferswithpHs3–8aftera6h incubationof2mg·mL‐1xanthansolutionswith54μgYesY·mgxanthan‐1at40˚C.TheaceticacidreleasewasdeterminedusingaMegazymeaceticacidkit(Megazyme,Wicklow, Ireland) and expressed as percentage of the total acetyl content in theparental xanthan. The company’s protocol was downscaled tomicrotiter scale. Thetotalacetylcontentwasdeterminedbytheanalysisof theaceticacidreleaseafterasaponification step with 1 M NaOH (18 h; 4°C). All enzyme incubations wereperformedinduplicates.


91

Determination of the relative abundance of the six constituent RUs in

YesY modified xanthan

The mode of action of YesY towards xanthan was determined by the structuralanalysis of unmodified and enzyme‐treated xanthan. Enzymatically deacetylatedxanthansolutionwasdialyzedagainstdemineralizedwaterfor24handlyophilized.ThexanthanobtainedwasredissolvedinMilliporewater(2mg·mL‐1)andincubatedwiththeexperimentalcellulasespreparationC1‐G1fromMyceliophthorathermophilaC1 (Dyadic Netherlands, Wageningen, The Netherlands) at 60˚C for 48 h.21 Thexanthandigestsobtainedwereanalyzedfortheirrepeatingunits(RUs)usingHILIC‐ELSD‐ESI‐IT‐MSnasdescribedelsewhere.10


InapreviousstudytheexpressionandpurificationofthepectinacetylesteraseYesYwasdescribed.18SubstratespecificityteststowardsvariousacetylatedpolysaccharidesubstratesshowedthatYesYwasalsoactive towardsxanthan(Table6.2.).Althoughthespecificactivity towardsxanthanwas~40 times lowercompared to theactivitytowardspectin,thissideactivityisveryinterestingasonlyoneotherenzymehasbeendescribedtodatethatcanremoveacetylgroupsfromxanthan.17InthisstudytheYesYactivitytowardsxanthanwasfurtherinvestigated.Table 6.2. Acetic acid release after a 10 min, 2 h or 24 h incubation of various substrates (5 mg∙mL‐1) with YesY in 50mm sodium citrate buffer pH 6.0 taken from Wagenknecht et al.18

Substrate Acetic acid release (μg∙mL‐1) 10 min 2h 24h

Xanthan n.d. 4 20 Konjac gluco mannan n.d. 0 2

Chitin n.d. n.d. n.d.

56% reacetylated chitosan n.d. n.d. n.d.

Xylan oligosachharides n.d. 3 49

Pectin 28 64 90

pNP‐acetate n.d. n.d. n.a.

n.d. = not detected n.a. = not analyzed

Chapter 6

92

Influence of the xanthan concentration on the YesY activity

ThespecificactivityofYesYtowardsxanthanasdeterminedbyWagenknechtetal.18wasdeterminedon5mg·mL‐1 substrate solutions. Inorder todetermine if thehighviscosityofa5mg/mlxanthansolutionhinderstheenzyme,theactivityofYesYwasalso determined towards a lower xanthan concentration. The enzyme to substrateratiowaskept constant compared to thepreviousstudy.Aftera24h incubation, inwhichthereactionisinthelinearrange,8.28μgaceticacid(6.9%ofallacetylgroupspresent)wasreleased.Thisequalsaspecificactivityof11.9mU∙mgprotein‐1,whichissimilar to the specific activity found previously (11.6 mU∙mg protein‐1). It was,therefore,concludedthattheviscosityofxanthansolutionswithconcentrationsupto5mg·mL‐1xanthandoesnotinfluencetheYesYactivity.Itwasdecidedtouse2mg·mL‐1xanthansolutionsforfurtherexperiments.

Influence of the enzyme:substrate ratio on the acetic acid release by YesY

Determination of the specific activity showed that, independent of the substrateconcentrationused,only~7%ofallaceticacidgroupsisreleasedfromxanthanafter24hofincubation.Extendingtheincubationtimedidnotresultinthereleaseofmoreacetic acid. Due to the repetitive structure of xanthan, it is not expected thatneighboring side chains within the xanthan molecule hinder YesY in its activitytowardsxanthan.

Figure 6.2. Influence of the enzyme:substrate‐ratio on the acetic acid release by YesY after 24 h of incubation at 40°C, pH 6.

0

5

10

15

20

25

30

35

40

0 20 40 60 80

Ace

tic

acid

rel

ease

(%

of

max

imu

m)

Enzyme: substrate ratio (μg/mg)


93

Thelowacetylreleasecouldbeduetoinactivationorinhibitionoftheenzymeduringthe incubation. In order to test this hypothesis the influence of theE:S‐ratioon theenzymeactivitywasdetermined.TheresultsareshowninFigure6.2.WithincreasingE:S‐ratio,anincreaseintheaceticacidreleaseisobserved,indicatingthattheenzymewas indeed hindered in its activity at low E:S‐ratio. Increasing the enzymeconcentrationtovalueshigher than54μgenzyme∙mgxanthan‐1doesnotresult inafurtherincreaseinaceticacidrelease.After24hofincubationamaximumof22%ofallacetylgroupswasreleasedfromxanthan.ItwasdecidedtouseanE:S‐ratio54μgenzyme∙mgxanthan‐1forfurtherexperiments.

Influence of the xanthan conformation on the YesY activity

Recently,we showed that the enzymatic removal of acetyl groups from xanthan byAXE3 strongly depends on the secondary xanthan structure.17 It was, therefore,investigatediftheactivityofYesYtowardsxanthanisalsoinfluencedbythexanthanconformation. The exact fraction of disordered xanthan (α) in solution can becontrolled by selecting specific solvent conditions and can be determined usingcircular dichroism. The exact xanthan conformations obtained in the solutioncondition chosen in this studywere previously determined.17 An overview of thesesolution conditions, the corresponding xanthan conformation and the acetic acidrelease by YesY after a 24 h incubation is given in Table 6.3. Enzyme incubationsperformed in thecompleteabsenceof salt, resulted inacleardecrease in theaceticacidrelease.Thepresenceofonly1mMNaClalreadyshowedtodoubletheaceticacidrelease compared to an incubation inMilliporewater.A further increaseof the saltconcentrationto10mMNaCldidnotresultinafurtherincreaseinaceticacidrelease.The clear decrease in enzyme activity in the absence of salts indicates that counterionsarenecessaryforenzymestabilization.Nosignificantdifferenceintheaceticacid Table 6.3. Incubation conditions, the corresponding ellipticity (θ) and fraction of disordered conformation (α) (taken from Kool et al.17) and the acetic acid release from xanthan after a 24 h incubation at the different incubation conditions

Sample condition θ219 α Acetic acid release

(% of maximum release)

Millipore water 40˚C ‐3.02 0.78 8

1 mM NaCl 40˚C ‐1.37 0.35 15

2 mM NaCl 40˚C ‐0.93 0.24 13

5 mM NaCl 40˚C ‐0.55 0.14 17

10 mM NaCl 40˚C ‐0.20 0.05 14

50 mM CA buffer 40 ˚C n.a. n.a. 22 n.a. = not analysed

Chapter 6

94

releasewasobservedwithinthedigestsobtainedatvarioussaltconcentrations,andthuswithin thedigestsobtainedatdifferent xanthanconformations. It is, therefore,concludedthattheYesYactivity,incontrasttotheAXE3activity,doesnotdependentonthexanthanconformation.Astheacetylgroupsontheinnermannosearebelievedto be directed to the inside of the ordered helical xanthan structure,17, 22 the YesYactivity towardsxanthan ishypothesizedtobespecific for theremovalof theacetylgroupsontheoutermannose.

Temperature optimum of YesY

The temperature profile of YesY was determined after a 6 h incubation in thetemperaturerange30‐80˚C.ApHof6.0waschosenasitshowedtobetheoptimalpH(seebelow).Theresults(Figure6.3a.) indicateatemperatureoptimumbetween50‐60˚C. A clear drop in activity was observed when the temperature was furtherincreased, indicating enzyme inactivation at temperatures above 65˚C for 6 h. Thetemperatureoptimumtowardspectindeterminedaftera10minuteincubation,18washighercomparedtotheoptimumobservedtowardsxanthan(Figure6.3a.).Also,theclear drop in activity that was observed for xanthan was not observed in thetemperatureprofileofYesYtowardspectin.

pH optimum of YesY

ThepHoptimumofYesYwasdeterminedaftera6hincubationinthepHrange3.5–8at40˚CandcomparedtothepHprofileofYesYtowardspectin.Figure6.3b.showsaclearpHoptimumtowardsxanthanatpH6.3.TwopHoptimawereobservedforYesYtowardspectin: one equal to thepHoptimum towards xanthan at pH~6.3 and theother,withamuchhigherspecificactivityatpH‐values≥7.5.18Becausecorrectionforautohydrolysis was necessary for xanthan digests obtained at pH≥7.5, the enzymeactivitytowardsxanthanathighpHvaluescouldnotbedeterminedaccurately.

Characterization of the enzymatically modified xanthan

YesY is especially active towards xanthan in the presence of salts, and therebytowardstheorderedxanthanconformation. InapreviousstudyweshowedthattheacetylgroupsontheinnermannosewasnotaccessibleforAXE3deacetylationwhenxanthanappearsinanorderedconformation.17Itis,therefore,hypothesizedthatYesYisspecificforthedeacetylationofacetylgroupspositionedattheoutermannoseunit.InordertoanalyzethespecificityofYesY,YesYmodifiedxanthanwasdegradedtoitsrepeatingunits(RUs)usingacellulasepreparation.Subsequently,therelative


95

Figure 6.3. A: Temperature optimum of YesY towards xanthan obtained after a 6 h incubation at pH 6. B: pH optimum of YesY towards xanthan obtained after a 6h incubation at 40°C. Xanthan ( ); pectin ( ) (data taken from Wagenknecht et al.18) abundance of the repeating units (RUs) present in cellulase digests of the YesYdeacetylated xanthan was analyzed using HILIC‐ELSD‐MS10 and compared to therelativeabundanceof theRUspresent in thecellulasedigestofunmodifiedxanthan(Figure 6.4.). As observed in earlier studies, five different peaks are present in theelutionprofile of unmodified xanthan.10 Co‐elutionof the two single acetylatedRUswasconfirmedbyMS‐fragmentationasdescribedpreviously.10Aftera24hincubationwithYesY,nodoubleacetylatedRUs(peak1)weredetectedinthemodifiedxanthan,the amount of single acetylated RUs (peak 2) increased, and no changes in theabundanceoccurredfortheotherRUs.AsnodecreaseintheamountofpyruvylatedandacetylatedRU(peak3),orincreaseinunsubstituted(peak4)and/orpyruvylated(peak5)RUswasobserved,itisconcludedthatYesYisindeedspecificfortheremovalof acetyl groups from the outer mannose. The increase in the abundance of singleacetylatedRUs,isthustheresultoftheconversionofdoubleacetylatedRUsintoRUswhicharesolelyacetylatedontheinnermannose.TodetermineifRUsthataresolelyacetylatedontheouterarealsodeacetylatedbyYesY,orthatacetylationattheinnermannoseisnecessaryforYesYdeacetylation,theremainingsingleacetylatedRUs

0

20

40

60

80

100

25 35 45 55 65 75 85

Rel

ati

ve a

cti

vity

(%

)

Temperature ( ̊C)

A

0

20

40

60

80

100

3 4 5 6 7 8

Rel

ativ

e a

ctiv

ity

(%)

pH

B

Chapter 6

96

8 10 12 14 16 18 20

Time (min)

EL

SD

Res

po

nse

YesY modified xanthan

unmodified xanthan

1 3

2 4

5 6

Figure 6.4. HILIC‐ELSD elution profiles of cellulase digests of unmodified xanthan and xanthanmodified with YesY for 24 h. Glucose ; mannose ; glucuronic acid ; acetyl groups ; pyruvicacid ketal .

m/z400 500 600 700 800 900 1000

Re

lati

veab

un

dan

ce

721

*703

*541 823

¤661 805*625¤499 ¤583 *643

559 ¤601

400 500 600 700 800 900 1000

Rel

ati

veab

un

dan

ce

721*703

*541

823

*625 805

*643559 ¤583 YesY modified xanthan

unmodified xanthan

¤ Fragments specific for the acetylation on the outer mannose* Fragments specific for the acetylation of the inner mannose

Figure 6.5. MS2‐fragmentation pattern of the single acetylated repeating unit (eluting at 13.2 min inFigure 4.4.) of unmodified and YesY modified xanthan


97

were further characterized. Changes in the abundance of the two single acetylatedRUs,inducedbytheYesYmodification,canbedeterminedusingtheMS‐fragmentationpattern since inner and outer mannose acetylated RUs have a different set ofdiagnosticfragmentions.10Theintensityofthesefragmentionsinthefragmentationpatternof thesingleacetylatedRUbeforeandafterYesYmodificationare shown inFigure6.5.Beforemodification~25%ofallsingleacetylatedRUswereacetylatedontheoutermannose.AfterYesYmodification100%oftheRUswereacetylatedat theinner mannose indicating that no RUs remain which are acetylated on the outermannose.Acetyl groupsat theoutermannoseunit can thusbehydrolyzedbyYesY,independentonthesubstitutionpatternoftheinnermannose.YesY also showed to be active towards xanthan when xanthan appears in a partlydisorderedconformation(Table6.3.).Atsuchconformationstheacetylgroupsattheinner mannose are better accessible to enzymatic hydrolysis. The YesY specificitytowards partly disordered xanthan and completely ordered xanthan could thus bedifferent. The YesY specificity towards partly disordered xanthan was, therefore,investigatedbythestructuralanalysisofxanthan,whichwasmodifiedwithYesYina2mMNaClsolution,whereα=0.24(resultsnotshown).AsimilarHILIC‐elutionprofilewasobtainedcomparedtotheelutionprofileobtainedforxanthanmodifiedwithYesYat a completely ordered conformation. It was, therefore, concluded that YesY isspecificforthedeacetylationoftheoutermannoseunit independentonthexanthanconformation.

CONCLUSIONS

YesY,apectinacetylesteraseoriginatingfromBacillussubtilisstrain168iscapableinremoving22%ofallacetylgroupsinxanthan.Optimalactivitywasobtainedat55°C,inthepresenceofsaltsatpH6.3.YesYisspecificfortheremovaloftheacetylgroupspositioned on the outer mannose unit, and its activity is independent on theacetylation on the inner mannose, all acetyl groups on the outer mannose can beremoved.Thereby the firstacetylesterase isdescribed that specificallydeacetylatesthe outer mannose units of xanthan, and which can be used for the production ofxanthanswithalteredandpossiblyimprovedfunctionalities.

ACKNOWLEDGEMENTS

This research was supported by the European Community within a consortiumPolyModEKBBE‐2007‐3‐3‐07andisgratefullyacknowledged.

Chapter 6

98

REFERENCES

[1] SwornG.XanthanGum. In: ImesonAed.FoodStabilisers,ThickenersandGellingAgents.Oxford,UK:Wiley‐Blackwell,2009;325‐42.

[2] Garcı́a‐OchoaF,SantosVE,CasasJA,GómezE.Xanthangum:Production,recovery,andproperties.BiotechnolAdv.2000;18:549‐79.



[5] Frangou SA, Morris ER, Rees DA, Richardson RK, Ross‐Murphy SB. Molecular origin of xanthansolutionrheology:Effectofureaonchainconformationandinteraction.JPolymSci:PolymLettEd.1982;20:531‐8.

[6] Holzwarth G. Conformation of the extracellular polysaccharide of Xanthomonas campestris.Biochemistry.1976;15:4333‐9.

[7] Morris ER. Molecular origin of xanthan solution properties. In: Sandford PA, Laskin A, eds.ExtracellularMicrobioalPolysaccharides.Washington(DC),USA:ACS,1977;81‐9.

[8] Bejenariu A, PopaM, Picton L, Cerf DL. Effect of concentration, pH and temperature on xanthanconformation:Apreliminarystudybeforecrosslinking.RevRoumChim.2010;55:147‐52.








[16] RuijssenaarsHJ.Enzymaticmodificationofbacterialexopolysaccharides ‐Xanthan lyaseasa toolfor structural and functionalmodification of xanthan. PhD dissertation,Wageningen University,Wageningen,TheNetherlands.2001.

[17] KoolMM,ScholsHA,WagenknechtM,HinzSWA,MoerschbacherB,GruppenH.Characterizationofan acetyl estrease fromMyceliophthora thermophila C1 able to deacetylate xanthan. CarbohydrPolym.2013;submitted.Thisthesischapter5

[18] WagenknechtM,RemorozaC,SinghR,KoolMM,ScholsHA,MoerschbacherBM.YesY,aversatilecarbohydrateesterase‐Biochemicalandbioinformaticcharacterization.Manuscriptinpreparation2014.


99

[19] OchiaiA,ItohT,KawamataA,HashimotoW,MurataK.PlantcellwalldegradationbysaprophyticBacillussubtilisstrains:Geneclustersresponsibleforrhamnogalacturonandepolymerization.ApplEnvironMicrob.2007;73:3803‐13.

[20] Remoroza C,Wagenknecht M, Gu F, Buchholt HC, Moerschbacher BM, Gruppen H, Schols HA. ABacilluspectinacetylesteraseisspecificfordeacetylationofhomogalacturonan’sacetylatedatO‐3.Manuscriptinpreparation.



Chapter 6

100

Chapter 7

General discussion

Chapter 7

102

Multiplemodels exist in literature to describe the physical properties of a xanthansolution.Althoughstudiesonthebiosynthesisofxanthanshowthatvariationsinthexanthanstructuremayoccur,studiesdealingwiththephysicalpropertiesofxanthangenerally assume that the xanthan structure is rather repetitive with one singlerepeating unit (RU). This repeating unit is acetylated at the inner mannose andpyruvylated on the outer mannose (Figure 7.1; RU‐4). However, variations in thedegree of substitutions are known to occur, and in general only 90% of the innermannoseunitsarebelievedtobeacetylatedand50%oftheoutermannoseunitsarebelieved to be pyruvylated. Starting this study, we hypothesized that xanthansprimary structure is more complex than a polymer of this type of RU. As aconsequence,theconflictingmodelsproposedinliteraturewerehypothesizedtofindtheiroriginindifferencesbetweentheprimarystructuresofthexanthansused.Asnoevidenceforsuchdifferenceswasavailableatthestartofthisproject,themainaimofthis researchwas tocharacterize thexanthanprimarystructure inmoredetail thanwas usually done. According to the approach chosen, diagnostic xanthanoligosaccharides, preferable of intact RUs, should be produced as a first step toachieve this aim. To prevent random degradation or modification of the xanthanprimarystructureandconsequentlyoftheoligomersproduced,itwaschosentouseenzymes for the production of such oligosaccharides. Subsequently, the producedoligosaccharidesneeded tobecharacterizedandquantified inorder tocompare theprimary structure of multiple xanthan samples. Furthermore, enzymes thatspecificallymodify the side chains in order to study the distribution pattern of theacetyland/orpyruvategroupswithinthemoleculeweresearchedfor.

Figure 7.1. The six xanthan repeating units as revealed by HILIC‐ESI‐MS. The range of their relative abundance within 5 different xanthan samples, as determined in chapter 3 is given below each structure.

mannose

RU‐1

4 – 19%

RU‐2

3 – 12%

RU‐3

0– 2%

RU‐4

61 – 75%

RU‐5

2 – 3%

RU‐6

4 – 15%

acetyl pyruvateglucose glucuronic acid

General discussion

103

PRODUCTION AND CHARACTERIZATION OF XANTHAN REPEATING UNITS

Previous studies on the enzymatic backbone degradation of xanthan showed thatxanthan canonly bedegradedby cellulaseswhen incubations areperformed in theabsence of salts.1‐3 However, high molecular weight material rich in substituentsremainedattheendoftheenzymatichydrolysis.2,3Thisledtothehypothesis2,3thatxanthan is only susceptible to cellulase degradation when xanthan appears in thedisorderedconformationor that thesubstituents in thesidechainshindercompleteenzymatic xanthan degradation. A conclusive study to test either one of thehypotheseswasneverconducted.With the researchpresented inchapter2,we cannowconcludethatonlythexanthanconformationinfluencestheenzymaticbackbonedegradation of xanthan. The remaining highmolecularweightmaterial observed inother studies, therefore, is the result of the presence of ordered xanthan segmentsduringtheenzymatichydrolysis.Because the cellulase preparation used completely degraded xanthan to itsconstituentRUs,withoutundesiredadditionalmodificationofthexanthansidechains,itwaspossibletodeterminethexanthanstructureonRUlevel.StructureanalysisoftheRUsreleasedafterenzymaticbackbonedegradationbythecellulasesshowedthatsixdifferentxanthanRUsarepresentinaxanthansample(Figure7.1.).ThisledtotheconclusionthattherepetitivestructureofonesingletypeofRUasamodelforxanthanisincorrect.4Thexanthansanalyzed(intotal5differentxanthansinchapters3and4)in this studyall consist of the same6RUs,which shared the samepentasaccharidestructure, while having different acetyl and pyruvate substitution patterns. Theexistence of truncated side chains in unmodified xanthans, as hypothesized bySutherland2wasnotfoundalthoughexplicitlysearchedfor.ThemostinterestingvariationobservedwithinthesixRUs,isthesubstitutionoftheoutermannosewithanacetylgroup.Althoughthisvariationwashypothesizedbasedonxanthanbiosynthesis studies,5‐7 conclusiveevidencewasneverprovidedprior tothestartofthepresentstudy.Thepossibilityofthissubstitutionhas,therefore,alwaysbeenneglectedwhenstudiesonthestructure‐functionrelationshipofxanthanwereperformed. The influence of the acetyl groups on the functionality of xanthan hasalwaysbeenbasedonthetotallevelofacetylation.Asitisnowproventhat5‐20%oftheoutermannoseunitscanbeacetylated,covering5‐19%ofallacetylgroupswithinaxanthansample(chapter3),wesuggestthatpreviousoutspokenconclusionsmadeon structure function‐relationship of xanthan, especially regarding the influence ofacetylation,shouldbereconsidered.

Chapter 7

104

As it was found that the primary xanthan structure does not hinder the backbonedegradationofxanthanbycellulases(chapter2),itwasalsopossibletocomparethestructures of various xanthan samples. Quantification of the RUs present in thedifferentxanthancellulasedigestsenabled thecomparisonon theRU level (chapter3).Themajorconclusiondrawnfromthiscomparisonisthattherelativeabundanceof the RUs between xanthan samples can vary, even when their molecularcompositionsarethesame.In chapter 4was shown that particularly those segments enriched in RUs that areacetylated on the outer mannose stabilize the xanthan conformation through theformation of hydrogen bonds. In contrast to previous studies, which assumed thatonlytheinnermannoseunitisacetylated, itwasshownthatacetylationoftheinnermannose does not significantly contribute to the stability of the ordered xanthanconformation.Inconclusion:thepositionoftheacetylgroupwithinthexanthansidechaindeterminestheinfluenceoftheacetylgroupsonxanthan’stransitionalbehavior.TherelativeabundanceofthedifferentRUs,therefore,determinestheconformationalbehaviorofaxanthansampleandnotthemolarcomposition.Xanthansampleswithequalmolar compositions can, therefore, still vary in their conformational behaviorandthusintheirsolutionproperties.

DISTRIBUTION OF THE DIFFERENT REPEATING UNITS OVER THE XANTHAN BACKBONE

The correlation between the enzymatic release of individual types of RUs and thefraction of disordered conformation showed that the distribution pattern of thedifferently substituted RUs is non‐regular (chapter 4). It was proposed thatpyruvylatedRUsarerandomlydistributedinthexanthansamplesstudied.Incontrast,thesingleanddoubleacetylatedRUsareproposedtobedistributed inablockwisemanner. Based on this distribution pattern, a tentative model is proposed for thetransitional behavior of differently substituted xanthan segments (Figure 7.2.).Accordingtothismodel,segmentsrichinoutermannoseacetylatedRUsdissociatetoa random disordered conformation only at the end of the xanthan transition.Segments rich in RUs which are solely acetylated on the innermannose dissociaterathereasilyintoadisorderedconformation.Furthermore,itishypothesizedthattheintermolecular interactions involved within a xanthan network are induced by theacetylgroupsontheoutermannose.ThejunctionzonesofaxanthannetworkwouldthusberichinoutermannoseacetylatedRUs.

General discussion

105

Figure 7.2. Proposed conformational behavior of xanthan when 50% appears in a disordered conformation.8

TheproposeddistributionpatternoftheRUsisbasedonthereleaseofsingleRUsandis thusbasedon indirectevidence.Expandingthecellulase fingerprintingmethod inordertoobtainxanthanoligosaccharideslargerthan1RUis,therefore,importantinviewofunderstandingthestructure‐functionrelationshipofxanthan.

Structural analysis of xanthan fragments larger than 1 RUs

To obtain xanthan oligosaccharides larger than 1 RU (RUn), xanthan A (chapter 4)present in a completely disordered conformation (60°C; Millipore water) wasincubated with the cellulases for different time intervals (unpublished results).AnalysisofthedigestsobtainedwithHPAEC(accordingtochapter3)showedthatthesingleRUs, fromnowondenoted ‘RU1’,weredominant in thedigests (Figure7.3.).Other structureswere released aswell, whichwere proposed to be RUn structuresbased on the HPAEC elution patterns observed for other mono‐ andglucooligosaccharides.ThehighestvarietyintheseRUnstructureswasobtainedafter2hofincubation.The 2 h digestwas further analyzed byHILIC‐ELSD‐MS (according to chapter 3) toidentify thedifferentRUn structurespresent. Similar to theHPAECprofiles, theRU1structuresdominatedtheelutionprofile(resultsnotshown).QuantificationbasedonELSD peak area indicated that 74% of the xanthan molecule was released as RU1.BasedonthequantificationoftheindividualtypesofRU1obtained, itwasestimatedwhichpercentageofeachtypeofRUwaspresentasRUn.AnoverviewisgiveninTable7.1.TheoutermannoseacetylatedRUs(RU‐1andRU‐3)aremainlypresentasRUn.Morethan70%oftheseRUsarepresentaslargerxanthanfragments,whichrepresent

glucose

glucuronic acid

mannose

pyruvate

acetyl

Chapter 7

106

Figure 7.3. HPAEC elution pattern of xanthan digests obtained after incubation of xanthan with cellulases in Millipore water at 60°C for different time intervals. Annotation of peaks as described in chapter 3.

26%ofthetotalxanthandigest.Theseresults,thereby,confirmtheconclusionmadeinchapter4thatoutermannoseacetylatedRUsaredistributedinaratherblockwisemanner. Similarly, these results confirm that theRU that is solely acetylatedon theinnermannose is alsodistributed in a blockwisemanner, as only 6%of thisRU ispresentasRUnstructures,andthusin26%ofthexanthanstructure.Furthermore,theconclusion that RU‐4 is distributed randomly is confirmed as 23% of this RU ispresentin26%ofthemolecule.

Table 7.1. Appearance of the six different xanthan RUs in a 2 h cellulase digest (Millipore; 60°C) as RU1 or as/in RUn

Type of RUa Present as RU1

( % of total) Present as RUn

(% of total)

Total xanthan sample 74 26

RU‐1 28 72

RU‐2 94 6

RU‐3 26 74

RU‐4 77 23

RU‐5 85 15

RU‐6 60 40

a= The molecular structures of RU ‐1‐6 are given in Figure 7.1.

0 5 10 15 20 25 30 35 40 45

PA

D-R

esp

on

se (

NC

)

Time (min)

8 hours

4 hours

2 hours

1 hours

0.5 hours

0 hours

Glucose

Non‐pyruvylated (RU)1

Pyruvylated (RU)1

General discussion

107

Figure 7.4. Structures of xanthan oligosaccharides present in a 2 hr. xanthan digest as determined by HILIC‐MSn. The oligosaccharides consisting of only 1 repeating unit are excluded. orange box = most abundant; green box second most abundant

Inorder toverify that theRUnstructuresobtainedare indeedenriched inRU‐1andRU‐3,effortwasdirectedinobtaininggoodMS‐spectraoftheRUnstructures.TheRUnstructurespresent in the2hrdigestwere, therefore,separated fromtheRU1bysizeexclusionchromatography,usingthreeSuperdex30columnsinseriesandammoniumformate buffer (250 mM) elution.9 Three different fractions were obtained, whichwereanalyzedbyUPLC‐MSn.OneofthefractionsobtainedindeedcontainedtheRU1structures, another fraction was rich in RU2 structures and the third fractioncontainedRU3structures.Largerxanthanoligosaccharides(XaOS)werenotobserved.AllRU2andRU3structuresfoundarepresentedinFigure7.4.Intotal,21differentstructuresarepossibleforRU2and35forRU3whentheorderofthe individual RUswithin each XaOS is not considered. However, only 28 differentstructuresweredetectedbyMS‐analysis.AlthoughquantificationofthedifferentXaOSobservedwas difficult, an estimation on themost abundant structures presentwasmadebasedon theELSDelutionprofiles. These abundant structuresdetectedwereindeedenrichedwithsidechainsacetylatedontheoutermannose,as7outofthe12dominantstructurescontainedatleast1RUacetylatedontheoutermannoseunit.In

acetyl pyruvateglucose mannose glucuronic acid

Chapter 7

108

total12outofthe30individualRUswithinthedominantstructureswasacetylatedonteoutermannose.AsxanthanappearedinacompletedisorderedconformationduringtheproductionoftheRUn,thisindicatesthattheRUswhicharenotacetylatedattheoutermannosearepreferentiallyhydrolyzedbythecellulasesused.Structuresrichinpyruvylated side chainswerealsodetected in thedimerand trimer fractionsof thedigest.ThiswasexpectedastheseRUsarerandomlydistributedandrepresent~65%ofallsidechainswithinthestudiedxanthan.Therefore,approximately65%ofallsidechainspresentintheXaOSshouldbepyruvylated,whichwasindeedthecase(10outof12of thedominantstructures;or15outof the30 individualRUs). Inconclusion,thedistributionpatternproposedinchapter4wasconfirmedbystructuralanalysisoflargerXaOS.TheresultspresentedonlyprovideafirstindicationontheRUdistributionpatternofxanthan, since fragments of 2 or 3RUs arenot sufficiently long tomake conclusivestatementsontheRUdistributionpattern.Especiallywhenconsidering the fact that76%ofallRUswerereleasedasindividualRUs.Asthecellulasesusedinthisresearchseem to prefer the release of single xanthan RUs (Figure 7.3.), screening for othercellulases that prefer the release of larger XaOS could be useful for further studiestowardsthedistributionpatternofthesixRUs.

The xanthan structure revisited

Inchapters2‐4itwasshownthatxanthanhasmorevariationinitsprimarystructurethanwasgenerallyassumed.Especiallytheobservationthattheoutermannosecanbeacetylated showed to be important, as this type of substitution is concluded to beimportant for the conformationalbehaviorof xanthan.Furthermore, combinedwiththe data presented above, assumptions on the distribution pattern of the differentxanthan RUs can bemade. Based on the research presented in this thesis,we nowpropose a new model for the xanthan structure (Figure 7.5.). In this model thexanthan structure is divided into segments,which are either rich in outermannoseacetylatedRUs(RU‐1andRU‐3)orrich inRUswhicharesolelysubstitutedwithanacetylgroupontheinnermannose(RU‐2).TheremainingRUs4,5and6areproposedtoberandomlydividedoverthesedifferentsegments.Itshouldbeconsideredthattherelativeabundanceof theRUsdiffers fordifferent xanthansamples (chapter3)andconsequently the precise distribution will vary. The validity of this postulatedstructureshouldbetestedinfutureresearch.Theuseofspecificsidechainmodifyingenzymeswillbeofhelptoachievethis.

General discussion

109

Figure 7.5. Tentative structure of xanthan as reconstructed from cellulase fingerprinting. The molecular structures of RU 1‐6 are given in Figure 7.1.

XANTHAN SIDE CHAIN MODIFYING ENZYMES

Modification of the xanthan side chains, as extension to the introduced cellulasefingerprintingmethod,wouldbeusefultofurtherelucidatethedistributionpatternofthe six RUs. As a first step to achieve this, enzymes were searched for whichspecificallymodifythexanthansidechains.

Xanthan acetyl esterases

In chapters 5 and 6 the first two acetyl esterases able to act on xanthan evermentionedinliteraturewererecognized,describedandcharacterizedfortheiractiontowardsxanthan.Oneoftheenzymes,YesY,specificallyremovesacetylgroupsattheouter mannose unit and is capable of removing 100% of its target within a 24 hincubation. The other enzyme, AXE3, is specific for the removal of acetyl groupspositionedattheinnermannoseunitsandreleases75%ofthesetargetsaftera3dayincubation.Completeenzymatic removalof theacetyl groupson the innermannosewas not obtained using AXE3, which was not understood. Further screening forxanthanacetyl esterases that enable complete xanthandeacetylationwould thusbeuseful.Itwasshownthattheacetylgroupsattheinnermannoseareonlysusceptiblefor enzymatic removalwhen xanthan is in a disordered conformationFurthermore,most enzymes need counter ions for their stability. Future screening for putative

Chapter 7

110

xanthanacetyl esterases should, therefore,beperformed in solution conditions thatfavorboth,thedisorderedaswellastheorderedxanthanconformation,inordernottooverlookanyacetylesteraseactivity.Furtherresearchtowardsthemodeofactionof the two presented enzymes, by structural analysis of larger XaOS after cellulasedegradation, should be considered as well. Such studies will reveal whether theenzymes act randomly over the backbone or if they show a single chain‐multipleattack action,which couldhelp to further determine the distribution pattern of theacetylgroupsoverthexanthanmolecule.Inchapter4itwasdiscussedthatthepositionoftheacetylgroupwithinthexanthanside chain is important for the functionalpropertiesof xanthan.Due to the focusofthis thesis, no rheological experimentswereperformed to confirm this assumption.Future,analysisof therheologyofxanthan,whichwaseithermodifiedwithYesYorwith AXE3, would reveal the exact correlation between the position of the acetylgroups within the xanthan side chain and the solution properties of xanthan.Furthermore, the enzymes couldbeused for theproductionof xanthanswithnovelrheological properties.As the specific activity of both acetyl esterases is rather low(YesY=11.9 mU∙mg protein‐1 and AXE3=13 mU∙mg protein‐1), further researchtowardsxanthanacetylesteraseswithahigherspecificactivityisuseful,especiallyforthelargescaleproductionofmodifiedxanthanswithalteredfunctionalproperties.

Pyruvate specific xanthan lyases

Severalxanthanlyasesareknownthatspecificallyremovethepyruvylatedmannosefromthexanthansidechain.10,11Inadditiontotheresearchdescribedinthepreviouschapters,aneffortwasdirectedtowardstheproductionofsuchlyasesandthefurtheruseoftheseenzymesforstructureelucidation.TwoxanthanlyasesfromBacillussp.strain GL1 (BAB21059.1) and Paenibacillus alginolyticus XL‐1 (AAG24953.1) weresuccessfully cloned, expressed and purified as described elsewhere.12 They will bereferredtoasXalA_GL1andXalA_XL1,respectively.To verify the specificity towards pyruvylated outer mannose units, the producedlyaseswereincubatedwith‘normal’xanthan(xanthanAinchapter4),pyruvatefreexanthan(PFXinchapter2)andhighlypyruvylatedxanthan(xanthanCinchapter4).TheelutionprofilesofthedigestsafterincubationwithXal_GL1areshowninFigure7.6. Similar elution patterns were obtained for XalA_XL1 digests. These resultsindicatethatbothenzymeswereactivetowardsxanthanindependentonthelevelofpyruvylation. Furthermore, both enzymes released mannose and pyruvylatedmannoseasrevealedbyHPAECanalysis(Figure7.6).Thepresenceofpyruvylated

General discussion

111

Figure 7.6. HPAEC elution profiles of xanthans differing in pyruvate content after a 24h incubation Xanthan lyase Xal_GL1.

mannoseinthedigestswasverifiedbythedetectionofmannoseandpyruvicacidinamolar ratio 1:1 after acid hydrolysis of the degradation products. Analysis of themodifiedxanthanbyHILIC‐ELSD‐MSafterdegradationofthebackbonebycellulases,confirmed that all different xanthan side chains are modified by the lyases,independentonthepresenceofasubstituentontheoutermannose.Bothlyasesarethusactive towardspyruvylated,acetylatedandunsubstitutedoutermannoseunits.Thelyasesproducedaretherebydifferentfromthelyasesdescribedbefore,10,13eventhoughtheaminoacidsequenceoftheenzymeswereexactlythesame.The xanthan conformation showed to be important for both enzymatic backbonedegradation (chapter 2‐4) and side chainmodification (chapter 5) of xanthan. Theobserveddifferenceinactivitybetweentheclonedlyasesandthelyases,however,didnotoriginatefromdifferencesinthexanthanconformation,assimilarlyaseactivitieswereobservedtowardsxanthaninafullyorderedordisorderedconformation.In conclusion, the lyases produced in this study really have a different specificitycompared to the lyases described before, even though the amino acid sequence isexactly the same. Although the difference in activity is not fully understood, theymight origin from the expression system used for the enzyme production, as weclonedthelyasesinanE.colistrainandtheoriginallyaseswereproducedbyBacillusspp.Astheclonedlyaseswerenotspecificforanytypeofsubstitution,theycannotbeusedforenzymaticxanthanfingerprinting.Furtherresearchtowardsspecificxanthanlyasesis,therefore,necessary.

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

High pyruvylated xanthan

Normal xanthan

Pyruvate free xanthan

pyruvylated mannosemannose

PA

D R

esp

on

se

Time (min)

Chapter 7

112

Microbial sources for other xanthan modifying enzymes

Asdescribedinchapter2,P.alginolyticusXL1canproduceavarietyofxanthansidechain modifying enzymes.14 Therefore, P. alginolyticus strain (DSMZ_5050) wascultured as described elsewhere.10 The crude enzyme extract was analyzed for theenzymes present, by the characterization of degradation products present in axanthandigestsobtainedafter incubationwith thisextract (resultsnotshown).Theproduction of several xanthan modifying enzymes by P. alginolyticus was indeedconfirmedincludingtheproductionof:xanthanlyases;axanthanacetylesteraseandapyruvate removing enzyme (unpublished results). As, the analysis of the digest byHPAEConlyshowed thepresenceofmannoseandpyruvylatedmannose,andnotofoligosaccharidesstructures,itwasconcludedthatallenzymesareactivetowardsthexanthanpolymerandarethuspotentialxanthanmodifyingenzymes.Next to the cultivation of P. alginolyticus on xanthan medium, xanthan was alsofermentedbymicrobiotafromthehumangutasthehumangutflorashowedtobeagood ‘producer’ for several carbohydrate modifying and degrading enzymes.15Characterization,usingHPSEC,HPAECandMaldi‐TofMS(resultsnotshown),of thedegradationproductspresentinaxanthandigestsobtainedafterincubationwiththesupernatantofa10dayfermentationliquidrevealedthepresenceofseveralenzymes:anacetylesterase,abackbonedegradingenzymeandanenzymethatreleasesdimersofβ‐mannose‐(14)‐β‐D‐glucuronicacid(unpublishedresults).Inordertodeterminewhetherthesidechainmodifyingenzymeswereactivepriortobackbonedegradation,the enzymeactionof the enzyme cocktailwasmonitored in time. Itwas found thatbackbone degradation occurs prior to the removal of the glucuronic acid‐mannosedimers. The acetic acid was released directly from the polymer. The enzymeincubations were conducted in 50mM sodium citrate buffer, pH 5.5. A disorderedxanthan conformation is, thus, not required for this acetyl esterase to be active. Asmorethan40%ofallacetylgroupswerereleasedbytheenzymecocktail,partoftheacetylgroupsremovedmusthavebeenoriginatedfromtheinnermannose.Theacetylesterase produced by the human gut flora is, therefore, different from the twoenzymescharacterizedinchapters5and6.Despite theproblems inexpressing thepyruvate specific lyase fromP.alginolyticus,both microbial sources described show to be good potential sources for theproductionofnovelxanthanmodifyingenzymes.However,nofurtherattemptsweremadetowardsthepurificationoftheenzymesdetectedintheculturebroths.

General discussion

113

IMPACT OF OUR FINDINGS ON UNDERSTANDING THE STRUCTURE‐FUNCTION RELATIONSHIP OF XANTHAN

Using the approach introduced in the presented study, amore detailed correlationbetween the xanthan substituents and xanthan’s functionality can be obtainedcompared to previous studies. Because no rheological analysis is performed in thisresearch, our findings regarding the primary xanthan structure were correlated topublishedresultsonxanthan’sfunctionality.

Xanthan conformation

In chapter 4, it was concluded that the position of the acetyl groups within thexanthan repeating units is important for the stability of the xanthan conformation.Substituentsontheoutermannoseare,therefore,proposedtobemostimportantforxanthanssolutionbehavior.Althoughtheacetylationoftheoutermannosehasneverbeen consideredwhen studying the conformation of xanthans, the influence of thelength of the xanthan side chains has been discussed.16‐18 It was concluded thatxanthan lackingtheoutermannose,alsocalledthe ‘polytetramer’,conductsahelicalconformationintheexcessofsalts,whichissimilartothatofnormalxanthan.16Underaqueous conditions most of the polytetramer already appeared in a disorderedconformationat25°C.17Theseresultstherebyconfirmthatthestabilityofthexanthanconformationdependsontheoutermannose.

Viscosity of xanthan solutions

Theinfluenceof theprimarystructureonxanthansrheologyhasbeenstudiedmoreextensivelythanitsinfluenceonthexanthanconformation.Nevertheless,theviscosityof xanthan solutions is believed to depend on the xanthan conformation.19 Asespecially the substituents on the outer mannose unit determine xanthansconformation(Chapter4),weexpectthatthisalsoappliestotheviscosityofxanthan.Acetylationoftheoutermannoseis,therefore,hypothesizedtoreducetheviscosityofaxanthansolutiontoalargerextentthantheacetylationoftheinnermannose.Theinfluenceofthetotaldegreeofacetylationontheviscosityofaxanthansolutionhasmostlybeenstudiedbytheaspecificremovalofacetylgroups,whichresultedinanincreaseoftheviscosity.20, 21Theinfluenceoftheacetylpositionontheviscosity,however,wasonlystudiedonce.7 Inthatstudy itwasconcludedthatthepositionoftheacetylgroupswithinthexanthanmoleculeisnotimportantfortheviscosity.Thiswasbasedonthecomparisonoftheviscosityofxanthansproducedbymutantstrains.

Chapter 7

114

Thesexanthanswereeither solelyacetylatedon the innermannoseor on theoutermannose and were not pyruvylated at all.7 The influence of the pyruvate groupsand/or intramolecular interactions on xanthan’s viscosity was thus neglected. Asdiscussed in chapter 4, the distribution of the six RUs in a xanthan segment isimportant for the final transitional behavior of that segment. The use of mutantstrains to study the influenceof theprimarystructureonxanthan’sviscositymight,therefore,notberepresentativeforstandardxanthans.Thisisespeciallyplausibleastheproductionofxanthansbymutantstrainsmayalsoresultinchangesotherthanatthe level of acetylation, e.g. the molecular weight and the level of pyruvylation.14Furthermore, the viscosity of the xanthans produced by mutant strains wasdeterminedbeforexanthanwasexposedtoaheattreatment.Thefirstheattreatmentis known to change the xanthan conformation and xanthan’s functionality.22‐24 Thesolution behavior of the xanthan produced by mutant strains could thus differsignificantly to thatof standardxanthansasa resultofdifferent sample treatments.Indicationsonthestructure‐functionrelationshipobtainedbyxanthansfrommutantstrainsare,therefore,difficulttotranslatetostandardxanthanproducedbywild‐typeXanthomonasspp.We,therefore,concludethattheapproachintroducedinthisthesisenables amore precise correlation between the primary xanthan structure and theviscosityofaxanthansolution,thantheapproachesusedtilltodate.

Xanthan‐galactomannan interactions

Studies on the influence of xanthan acetylation on the interaction withgalactomannansshowedthatacetylfreexanthanformsstrongerinteractiongelsthannormal xanthan.18, 25, 26 This improved interaction is believed to result from anincreasedchainflexibilityofxanthanmoleculesupondeacetylation.20Asacetylationattheoutermannoseismostimportantforthestabilityofthexanthanconformation,wesuggestthatespeciallytheacetylgroupsattheoutermannosehindertheinteractionswithgalactomannans.Shatwell et al.18 used xanthanproducedbymutant strains to study the influenceofxanthanacetylationon thexanthan‐galactomannan interactions.Oneof the xanthansamples produced was particularly rich in acetyl groups (7.7% w/w), pointing toacetylationat both the inner and theoutermannose.Alkaline removal of theacetylgroupsincreasedthegelstrengthofthisparticularxanthanwithafactorfour,whilethegelstrengthofthestandardxanthanonlyslightlyincreased.18Takenintoaccountprecautions about the use of mutant strains, based on these results it can bespeculated that acetylation at the outermannose indeed reduced gel strengths to alargerextentthanacetylationattheinnermannose.

General discussion

115

FUTURE PERSPECTIVES

Inconclusion,thisthesisshowsthattheprimaryxanthanstructureismorecomplexthan was generally assumed. Hence, characterization of xanthan samples based onmolarcompositionisinsufficient.Althoughnophysicalstudieswereperformedinthisstudy,itisconcludedthatthepositionoftheacetylgroupsisimportantforxanthansfunctionality. Conclusions on the influence of the primary xanthan structure onxanthans functionality made previously should, therefore, be reconsidered. Futureworkon thexanthan functionality shouldcharacterizexanthansamplesonRU levelandnotsolelyonmolarcomposition.Additionally,thetwoacetylesterasesdescribedin this researchcouldhelp toreveal theexact influenceof thepositionof theacetylgrouponxanthanfunctionalproperties.Although a method is introduced for a more detailed structure characterization,analyticalchallengesremainasthedistributionpatternofdifferentsubstituentsalongthe backbone of xanthan has not yet been established completely. Expanding theenzymaticanalyticaltoolboxintroducedinthisresearchis,therefore,necessary.Theavailabilityofotherbackbonedegradingenzymes,whichreleaseXaOSlargerthanRU3couldhelpstudyingthedistributionpattern.Furthermore,othersidechainmodifyingenzymes, e.g. an enzyme that specifically removes the pyruvate group or acetylesteraseswith a high specific activity, would be useful. In these future studies, theconformationduringenzymeincubationshastobeconsidered.

Chapter 7

116

REFERENCES





[5] Vorhölter F‐J, Schneiker S, Goesmann A, et al. The genome of Xanthomonas campestris pv.campestris B100 and its use for the reconstruction of metabolic pathways involved in xanthanbiosynthesis.JBiotechnol.2008;134:33‐45.

[6] Katzen F, Becker A, Zorreguieta A, Pühler A, Ielpi L. Promotor analysis of the Xanthomonascampestris pv campestris gum operon directing biosynthesis of the xanthan polysaccharide. JBacteriol.1996;178:4313‐8.



[9] LeijdekkersAGM,Bink JPM,GeutjesS,ScholsHA,GruppenH.Enzymaticsaccharificationof sugarbeet pulp for the production of galacturonic acid and arabinose; a study on the impact of theformationofrecalcitrantoligosaccharides.BioresourTechnol.2013;128:518‐25.


[11] HashimotoW,Miki H, Tsuchiya N, Nankai H,Murata K. Xanthan lyase ofBacillus sp. strain GL1liberatespyruvylatedmannosefromxanthansidechains.ApplEnvironMicrob.1998;64:3765‐8.

[12] Remoroza C,Wagenknecht M, Gu F, Buchholt HC, Moerschbacher BM, Gruppen H, Schols HA. ABacilluspectinacetylesteraseisspecificfordeacetylationofhomogalacturonan’sacetylatedatO‐3.Manuscriptinpreparation.

[13] HashimotoW,Miki H, Tsuchiya N, Nankai H,Murata K. Polysaccharide lyase:Molecular cloning,sequencing,andoverexpressionofthexanthanlyasegeneofBacillussp.strainGL1.ApplBiochemBiotechn.2001;67:713‐20.

[14] RuijssenaarsHJ.Enzymaticmodificationofbacterialexopolysaccharides ‐Xanthan lyaseasa toolfor structural and functionalmodification of xanthan. PhD dissertation,Wageningen University,Wageningen,TheNetherlands.2001.

[15] Jonathan MC. Monitoring the degradation of individual dietary fibers in pig models. F PhDdissertation,WageningenUniversity,Wageningen,TheNetherlands.2013.

[16] MillaneRP,NarasaiahTV.X‐RAYfiberdiffractionstudiesofavariantofxanthanguminwhichthesidechainterminalmannoseunitisabsent.CarbohydrPolym.1990;12:315‐21.

[17] TaitMI,SutherlandIW.Synthesisandpropertiesofamutanttypeofxanthan.JApplMicrob.1989;66:457‐60.


[19] Rochefort WE, Middleman S. Rheology of xanthan gum: Salt, temperature, and strain effects inoscillatoryandsteadyshearexperiments.JRheol.1987;31:337‐69.


General discussion

117





[25] ShatwellKP,Sutherland IW,Ross‐MurphySB,Dea ICM. Influenceof theacetyl substituenton theinteraction of xanthan with plant polysaccharides ‐ II. Xanthan‐guar gum systems. CarbohydrPolym.1991;14:115‐30.

[26] ShatwellKP,Sutherland IW,Ross‐MurphySB,Dea ICM. Influenceof theacetyl substituenton theinteractionofxanthanwithplantpolysaccharides‐III.Xanthan‐konjacmannansystems.CarbohydrPolym.1990;14:131‐47.

Chapter 7

118

Summary

Summary

120

SUMMARY

Xanthaniswidelyusedasrheologymodifierandemulsifierinfoodindustry.However,the influence of theprimary and secondary structureon, and the exact interactionsinvolvedintherheologypropertiesarenotfullyunderstood.

In chapter 1 an overview is given of different models proposed in literature thatexplainxanthansfunctionalproperties.Theexactinfluenceofthesecondaryxanthanconformation on xanthan’s functional properties remains uncertain, however thesecondaryconformation isbelieved tobean important factor forxanthan’ssolutionproperties. The xanthan primary structure is also known to influence xanthansfunctional properties, however theprimary structure affects the xanthan secondaryconformation as well. It was, therefore, hypothesized that the different modelspresentedinliteraturethatexplainxanthan’sfunctionality,originfromdifferencesinthe primary structure of the xanthan samples used. However, till to date no goodanalyticalmethod is available to study xanthan’s primary structure in detail. Hencethe aim of this thesis was to develop an enzyme basedmethod to study xanthan’sprimary structure, to enable a better understanding into the structure‐functionrelationshipofxanthan.

Tobeabletocomparetheprimarystructureofmultiplexanthans,diagnosticxanthanoligosaccharidesarenecessary. In this researchanenzymaticapproachwasused toproduce such oligosaccharides. As the enzyme activity might depend on both theprimaryandthesecondaryxanthanstructure, the influenceofboththesestructuresontheenzymaticdegradationofxanthanwasstudied(chapter2).ItwasshownthatthexanthanbackbonecanbedegradedbycellulasesfromMyceliophthorathermophilaC1independentontheprimaryxanthanstructure.Howeveronlydisorderedxanthansegmentscanbedegraded.Undegradedorderedxanthanstructuresremainpresentashigh molecular weight material, indicating that ordered xanthan structures arepresentinanetworkofmolecules.

Asxanthancanbeenzymaticallydegradedindependentontheprimarystructure,theprimarystructureofdifferentxanthanscanbecompared.Inchapter3fivexanthanswere enzymatically degraded to their repeating units (RUs). The type and relativeabundanceofthedifferentRUsweredeterminedusingUPLC‐HILIC‐ELSD‐MSn.ItwasshownthatsixdifferentRUsexistwithinaxanthanmolecule.Surprisingly theoutermannose could be unsubstituted, pyruvylated or acetylated, indicating that theprimary structure ismore complex thangenerally assumed.Comparisonof xanthansamplesonRUslevelshowedthatalthoughthemolecularcompositionoftwosamples

Summary

121

isexactlythesame,therelativeabundanceoftheRUsinthesesamplescouldstillbedifferent.

Asonlydisorderedxanthan segmentsare susceptible to enzymaticdegradation, theinfluence of the six different RUs on the transitional behavior of xanthan could bestudied(chapter4). Itwasconcludedthatespecially theacetylgroupsontheoutermannose stabilize thehelical xanthanconformation.The acetyl groupson the innermannose did not show to have a significant effect on the stability of the xanthanconformation.Thepositionof the acetyl groupwithin the xanthan side chain could,therefore,beimportantforxanthan’sfunctionalityaswell.Astheenzymaticreleaseofpyruvylated RUs increased linearlywith the fraction of disordered conformation, itwasconcludedthatthepyruvylatedRUsarerandomlydistributionoverthexanthanmolecules and that the distribution of the different RUs over the xanthanmoleculecouldbeimportantforthexanthan’stransitionalbehavior.

In chapter 5 several carbohydrate acetyl esterases were tested for their activitytowards xanthan. AXE3, an acetyl xylan esterase produced by MyceliophthorathermophilaC1,canremove~60%ofallacetylgroupswithinthexanthanmolecule.However,theenzymeisonlyactivetowardsdisorderedxanthansegments.Analysisofthe RUs present in the modified xanthan indicated that AXE3 is specific for theremovaloftheacetylgroupsontheinnermannose.

In chapter 6 YesY, a pectin acetyl esterase from Bacillus subtilis strain 168, wascharacterized for its activity towards xanthan. In contrast to AXE3, this enzymes isactivetowardstheorderedxanthanconformation.AnalysisoftheRUspresentinthemodifiedxanthanindicatedthatYesYisspecific fortheremovaloftheacetylgroupson the outermannose. Thereby two complementary acetyl esterases are described,which can be further used for the structure elucidation of xanthan and/or theproductionofnovelxanthanswithimprovedfunctionality.

Chapter7providesanoverviewoftheinformationdescribedinpreviouschapters,inorder to address important points for further studies on the structure‐functionrelationshipofxanthan.Inaddition,possiblestrategiesarediscussedtoimprovethe‘toolbox’ for the structural analysis of xanthan. The characterization of xanthanoligosaccharidesconsistingof2or3RUs,showedthatRUsthatareacetylatedontheouter mannose units are distributed in a rather block wise manner. Furthermore,otherpotentialsourcesforxanthanmodifyingenzymeswereintroducedinchapter7,toenablethecontinuationofstructureelucidationofxanthans

Summary

122

Samenvatting

Samenvatting

124

SAMENVATTING

Depolysacharidexanthaanwordt inde levensmiddelenindustrieveelvuldiggebruiktalsstabilisator,emulgatorenalsreologiecontroleur.Deexacteinteractiesbetrokkenbij de specifieke eigenschappen van een xanthaanoplossing en de invloed van deprimaireensecundairexanthaanstructuuropdezeeigenschappenzijntotophedenechteronbekend.

Hoofdstuk 1 geeft een overzicht van de bestaande modellen om de functioneleeigenschappen van xanthaan te verklaren. Het wordt algemeen aangenomen datzoweldeprimairealsdesecundairexanthaanstructuurdefunctioneleeigenschappenvaneenxanthaanoplossingsterkbeïnvloeden.Deprecieze invloedvanmetnamedesecundaireblijftechteronzeker,aangeziendeprimairestructuurookbepalendisvoorde secundaire xanthaan structuur. In het begin van dit onderzoek is daarom dehypothese gesteld dat de in literatuur beschreven verschillende en somstegenstrijdige modellen om de functionaliteit van xanthaan te verklaren hunoorsprong vinden in verschillen in de primaire, ofwel chemische structuur van degebruiktexanthaanmonsters.Totophedenisergeengoedemethodebeschikbaaromde chemische structuur van xanthaan in detail te onderzoeken. Het doel van ditonderzoekwasdaaromomeengeschiktemethodeteontwikkelenomdechemischestructuur van xanthaan te onderzoeken, zodat de structuur‐functie relatie vanxanthaanindetoekomstbeterbegrepenkanworden.

Om de chemische structuur van verschillende xanthaan monsters met elkaar tekunnen vergelijken, zijn diagnostische xanthaan oligosachariden nodig. In ditonderzoek is gekozen voor een enzymatische afbraak van xanthaan om zulkeoligosacharidenteproduceren.Echter,deenzymactiviteitzouookafhankelijkkunnenzijnvandesecundairexanthaanstructuur,ofwelxanthaanvouwing.De invloedvanzoweldechemischestructuuralsdexanthaanvouwingopdeenzymatischeafbraakvanxanthaanisdaaromeerstonderzocht.Inhoofdstuk2latenweziendatcellulasesvan Myceliophthora thermophila C1 de xanthaan backbone kunnen afbrekenonafhankelijk van de chemische structuur van een xanthaanmonster. De xanthaanvouwing beïnvloed de enzymatische afbraak aanzienlijk, waarin alleen xanthaansegmentenafgebrokenwordendie aanwezig zijn in eenopenontvouwen structuur.Deonafgebrokenxanthaansegmentenblijvenaanwezig ineengeordendegevouwenstructuur en hebben een hoog molecuul gewicht. Dit toont aan dat geordendegevouwenxanthaansegmenteneennetwerkvormen.

De chemische structuur van xanthaan beïnvloedt de enzymatische afbreekbaarheidvanxanthaanniet.Dechemischestructuurvanverschillendexanthaanmonsterskan

Samenvatting

125

daarom met elkaar worden vergeleken. In hoofdstuk 3 wordt beschreven hoe 5verschillende xanthaan monsters enzymatisch afgebroken worden tot hunherhalingseenheid(HE).OmhettypeenderelatievehoeveelheidvandeverschillendeHEn aanwezig te onderzoeken, is gebruik gemaakt van UPLC‐HILIC‐ELSD‐MSn. Wijkonden concluderen dat een xanthaan molecuul is opgebouwd uit 6 verschillendeHEn.De acetylering vande buitenstemannosewas verrassend en toont aandat dechemische structuur van xanthaan complexer is dan algemeen wordt aangenomen.Hetvergelijkenvanxanthaanmonstersopbasisvanderelatievehoeveelheidvande6HEn aanwezig, toonde aan dat xanthaan monsters met exact dezelfde moleculairesamenstellingtocheenanderesamenstellinginHEnkunnenhebben.

Alleen die xanthaan segmenten die aanwezig zijn in de ontvouwen ongeordendestructuurkunnenafgebrokenwordendoorcellulases.Hetisdaardoormogelijkomdeinvloed van de 6 verschillende HEn op de stabiliteit van de gevouwen geordendexanthaan structuur teonderzoeken (hoofdstuk4).Wij hebbenaangetoonddatmetnamede acetylgroepenaandebuitenstemannosede gevouwenxanthaan structuurstabiliseren.Acetylgroepenopdebinnenstemannosehaddengeenduidelijkeffectopde xanthaan vouwing. Hieruit blijkt dat de positie van de acetylgroepen in eenxanthaanmolecuulbepalendisvoordesecundairexanthaanstructuur.Depositievandeacetylgroepenisdusvanwezenlijkbelangvoordefunctioneleeigenschappenvaneen xanthaan monster. HEn met een pyruvaat groep bleken zich geleidelijk teontvouwen,wataantoontdatdezeHEnwillekeurigverdeeldzijnovereenxanthaanmolecuul.

Inhoofdstuk5 werden verschillende koolhydraat acetyl esterases getest voor hunactiviteit op xanthaan. AXE3, een acetyl xylaan esterase geproduceerd doorMyceliophthora thermophila C1, verwijderde ongeveer 60% van alle aanwezigeacetylgroepen in xanthaan. Enzym activiteit was echter alleen mogelijk opongevouwen ongeordende xanthaan segmenten. Karakterisering van de chemischestructuurvanenzymatischgemodificeerdxanthaanbeweesdatAXE3specifiekisvoorhet verwijderen van de acetylgroepen die gepositioneerd zijn aan de binnenstemannose.

Hoofdstuk6beschrijftdegedetailleerdekarakteriseringvandeenzymactiviteitvanpectineacetylesteraseYesY,geproduceerddoorBacillussubtilisras168,opxanthaan.In tegenstelling tot AXE3, is dit enzym actief op gevouwen xanthaan segmenten.Analysevandechemischestructuurvanenzymatischgemodificeerdxanthaantoondeaan dat YesY specifiek is voor het verwijderen van de acetylgroepen diegepositioneerdaandebuitenstemannose.Gedurendeditonderzoekhebbenwedusdeeerste2enzymengevondenengekarakteriseerddiedeacetylgroepeninxanthaan

Samenvatting

126

kunnen verwijderen. Deze twee complementerende enzymen kunnen in vervolgonderzoeken gebruikt worden voor de opheldering van de chemische xanthaanstructuur en/of de productie van xanthanen met verbeterde functioneleeigenschappen.

TotslotwordtinHoofdstuk7eenoverzichtgegevenvandebehaalderesultaten,enworden de consequenties van deze resultaten voor verder onderzoek naar destructuur‐functierelatievanxanthaanbediscussieert.Verderwordeneraanvullendemogelijkhedengegevenomde chemische structuur van xanthaan inde toekomst inmeer detail te karakteriseren. Voorlopige resultaten tonen aan dat HEn met eenacetylgroepopdebuitenstemannoseinblokformatiesgeorganiseerdzijn.Daarnaastgeeft hoofdstuk 7 een overzicht van potentiële bronnen voor andere xanthaanmodificerende enzymen, die bij kunnen dragen aan de verdere opheldering van dexanthaanstructuur.

Acknowledgements

Acknowledgements

128

Opmijn 18ewist ik het heel zeker, ikwilde niet net als de rest van het gezin naarWageningenvoordestudie.Enhierzitikdan11jaarlaternogsteedsinWageningenen eindelijk ‘uitgestudeerd’. Gelukkig waren die 11 jaar helemaal super, en heb ikzeker geen spijt dat ik na mijn studie in Wageningen ben gebleven voor eenpromotieonderzoek.NatuurlijkwasdezetijdinWageningennooitzo’nsuccesgeweestzonder de hulp en gezelligheid van velen. Daaromwil ik graag iedereen diemij deafgelopenjarengeholpenengesteundheeftbijdezebedanken.Henk,samensprongenwehetgrijzegebiedindatxanthaan‐onderzoekheet.Noudat“grijze onderzoeksgebiedje” bleek behoorlijk zwart te zijn en het duurde erg langvoordat het licht gevonden werd, maar het is ons gelukt . Bedankt voor jevertrouwendathetuiteindelijkechtwelgoedzoukomenenjesteuntoenikhetzelfhelemaalnietmeerzagzitten.EnnatuurlijkeenbigTHenksvoorjeenthousiasmeeninteressevoorzowelmijnonderzoekalsinmijalspersoon.Harry,bedanktvoor jebetrokkenheidbinnende leerstoelgroepen jeaandachtvoorallekleinedetailsdiehetwerkenbijFCHzoprettigmaken.DatwijalsenigeinAXISeenechtekoffiekamerhebben laatzienhoebelangrijk jedesfeerbinnenFCHvindt.Natuurlijkookbedanktvooralleurendie jehebtgestopt inhet lezenencorrigerenvanmijnmanuscripten.Dekwaliteitvanmijnpublicaties ishierdoorzekernaareenhogerniveaugebracht.I’d like to thank all members of the EU‐project ‘PolyModE’, with whom I’ve spentmanynicetrips.Itwasagreatexperienceworkinginsuchamulti‐disciplinaryteam.Graham,thankyouforall fruitfuldiscussionsandyourpatiencetohelpthischemistunderstandallphysicalaspectsofxanthan.Jolanda, een betere secretaresse kan een leerstoelgroep zich niet wensen. Superbedanktvoor jouhulpbijalleadministratievezaken,bijhetboekenvanvluchtenenhotels(watjetochiederhalfjaarweeropnieuwaanmemoestuitleggen)ennatuurlijkvoordegezelligheid.Ikwilgraagalleanalistenbedankenvoorhunuitlegenvoorhetonderhoudvanalleapparatuur.Margaret,wathebbenwesamenveeluurtjesachterde HPLCs doorgebracht. Bedankt voor je steun toen bijna iedere kolom die ikgebruiktekapotging.Mededankzijjouwideeënishettochgeluktomalmijnsamplesteonderzoeken.Martine,ondanksdatjegeenofficiëlesupervisorwasvandeHPAEChebjememetaljekennisenormgeholpen.BedanktvoorhetdelenvanallefrustratiesrondomdeHPLCs,ennatuurlijkdefrustratiesomdeberuchteEU‐verslagen;).Ingrid,uiteindelijkishettochnietgeluktomeenenzymoptezuiverenuitmijnfermentatie‐vloeistoffen, maar zonder jou hulp had ik niet eens fermentatievloeistoffen gehad.Bedankt voor alle hulp met de microbiologische kant van mijn onderzoek en hetwegwijsmakenvanzowelmijzelfalsZiruopde4evanhetBiotechnion.

Acknowledgements

129

Duringmy project I also had the opportunity to supervise some students. Thomas,Ziru,BastiaanandCatalina,thanksforyourcontributionstoandyourenthusiasmformyresearch.OfcourseI’dliketothankallmyFCHcolleaguesforthemanylabactivities,monthlydrinks, PhD trips and the daily coffee breaks. All together you really created theperfectworkatmosphere.Toeveryone I sharedanofficewithover the last5years:Thanksforallthefun,yourpatiencewithmetalkingallthetimeandforsharingtheupsanddownsofPhD‐life.Girls, thanksforkeepingthe ‘ChocoLA’filledatall times!Rudy, jebentenblijftmijn seniorPhDen ikhadmegeenbetere ‘buurman’bij FCHkunnenwensen.Bedanktvoordegezelligegesprekkenenallepeptalksopdefietsofindetrein.Connie, the longer our ‘PolyModE adventure’ lasted, the betterwe connected (eventhoughpectinandxanthanarenotsosimilarasexpected).EspeciallythelastyearwereallygottoknoweachotherandI’mhappyyouweretheretosharethefrustrationsregardingthewritingofallourpapersinsuchashorttime.Thankyou!Roy,zonderjouzouiknunogsteedseennatuurkundenitwitzijnenikkanjedanooknietgenoegbedankenvooraljehulpengeduldbijhetuitleggenvaniederartikeloverviscositeit, conformaties of interacties. Onze discussies, werk gerelateerd ofpersoonlijk, hebben me enorm geholpen en ik ben blij dat je me tijdens mijnverdedigingnogeenkeertjebijwiltstaanalsparanimf.Maxime,you’retheproofthatcolleaguescanbefriendsandI’msohappyyou’reuponthatstagewithmeduringmydefence.Thanksforallyoursupport,the(pep)talksandforallthefunduringourout‐of‐officeactivitiesoverthelastyears.I’llneverforgetthestayoversatyourplace,ourlate‐nighttalksandtheperfectbreakfastsyouserve:Youreallyownthebesthotelintown;).Thatmanymorestayoversmayfollow!GaafstekorfbalteamvanUtrecht,julliekondenhetopdetrainingmeestalwelmerkenalshetnietzosoepelgingmetmijnonderzoek;).Superbedanktdatikbijjullie2keerperweek stoomkonafblazen ende frustraties vanme af kongooien. Sportenblijfttochdebesteuitlaatklep.ChiQQa’s,allevakanties,weekendjes,festivalsenvrijdagavondensamenhebbenzekergezorgdvoordebroodnodigeafleidingengezelligheidtussendoor.Julliezijntop!Opnaardevolgende10jaar.Pap, Mam, Dorien en Ruud, wat waren het 5 bewogen jaren met veel ups en veeldowns. Jullieeeuwigevertrouwendathetheuswelgoedzoukomen, jullieoprechteverontwaardigingalsanderenhunbeloftesnietnakwamen,hetdelenvandevreudge

Acknowledgements

130

bij elke kleine overwinning, en jullie onvoorwaardelijke steun en trots. Dat alles ennogveelmeerheeftmeerdoorheengesleept.Bedankt!Adriaan,doorjoukanikmijnwerkopmijnwerklatenenthuisookechtontspannen.Dankjeweldatikbijthuiskomstallefrustraties,blijdschap,woedeofonmachtvanmeafmaggooien.Ofjenubegrijptwaarikhetoverhebofniet,jeweetaltijdprecieshetjuiste te zeggen ommeweer gewoonmet 2 benen op de grond te zetten.Met zo’nthuisfrontkaniedereendewereldaanenikbenblijdatjijmijnthuisbent!

About the author

About the author

132

CURRICULUM VITAE

MarijnM.KoolwasbornonJanuary16th1984inOosterhout,The Netherlands. After graduating from high school (Mgr.FrenckenCollege,Oosterhout)in2002,shestartedherstudiesFoodTechnologyatWageningenUniversity in2002.HerBScdegree was completed with a thesis on the acidification ofliquid coffee at the Laboratory of Food Chemistry. Her MScdegree in Food Technology, with a specialization in ProductFunctionality, was completed with a thesis on the isolationandcharacterizationofapotentialantimicrobialcompoundfromfermentedsoybeansattheLaboratoriesofFoodMicrobiologyandFoodChemistry.Marijnspentthelast5monthsofherstudyinPalmerstonNorth,NewZealand,whereshestudiedthemainellagitanninspresentinboysenberriesattheNewZealandInstituteforPlantandFoodResearch.

AftergraduatinginJune2008,sheworkedasaresearcherattheLaboratoryofFoodChemistryfor6months,beforestartingtheworkdescribedinthisPhDthesisinMay2009.Marijn isnowtemporarilyemployedaspost‐docresearcherattheLaboratoryofFoodChemistryofWageningenUniversity.

About the author

133

LIST OF PUBLICATIONS

Kool,M.M.;Comeskey,D.J.;Cooney, J.M.;McGhie,T.K.Structural identificationofthemainellagitanninsofaboysenberry(RubusloganbaccusxbaileyanusBritt.)extractbyLC‐ESI‐MS/MS,MALDI‐TOF‐MS and NMR spectroscopy. Food Chemistry,2010, 119(4),1535‐1543.

Kool,M.M.;Schols,H.A.,Delahaije,R.J.B.M.;Sworn,G.,Wierenga,P.A.;Gruppen,H.Theinfluenceoftheprimaryandsecondaryxanthanstructureontheenzymatichydrolysisofthexanthanbackbone.CarbohydratePolymers,2013,97(2),368‐375.

Kool, M.M.; Gruppen, H.; Sworn, G.; Schols, H.A. Comparison of xanthans by therelative abundance of its six constituent repeating units. Carbohydrate Polymers,2013,98(1),914‐921.

Kool,M.M.;Gruppen,H.; Sworn,G.; Schols,H.A.The influenceof the six constituentxanthan repeating units on the order‐disorder transition of xanthan. Accepted forpublicationinCarbohydratePolymers.

Kool, M.M.; Schols, H.A.; Wagenknecht, M.; Hinz, S.W.A.; Moerschbacher, B.M.;Gruppen,H.CharacterizationofanacetylesterasefromMyceliophtorathermophilaC1abletodeacetylatedxanthan.SubmittedforpublicationinCarbohydratePolymers.

Wagenknecht, M.; Remoroza, C.; Singh, R.;Kool,M.M.; Schols, H.A.; MoerschbacherB.M. YesY a versatile carbohydrate esterase – Biochemical and bioinformaticscharacterization.Tobesubmitted.

Kool, M.M.; Wagenknecht M.; Moerschbacher, B.M.; Gruppen, H.A.; Schols, H.A.Characterization of an acetyl esterase from Bacillus subtilis strain 168 able todeacetylatedtheoutermannoseofxanthan.Tobesubmitted.

PATENTS

Kool,M.M.;Schols,H.A.,Moerschbacher,B.M.;WagenknechtM.;Meansandmethodsforproducingdeacetylatedxanthan.Europepatentapplicationnumber:13003212.1,filed:24thofJune2013.

About the author

134

OVERVIEW OF COMPLETED TRAINING ACTIVITIES

Discipline specific activities

Courses

SummerCourseGlycosciences(VLAG),Wageningen,TheNetherlands,2010† AdvancedFoodAnalysis(VLAG),Wageningen,TheNetherlands,2010† Chromeleoncourse(Dionex),Heeze,TheNetherlands,2011 FoodandBiorefineryEnzymology(VLAG),Wageningen,TheNetherlands,2011†Conferences

8thCarbohydrateengineeringmeeting(Instituteofproteinbiochemistry),Ischia,Italy,2009

EPNOEmeeting(VLAG),Wageningen,TheNetherlands,2011† 12th International Hydrocolloid Conference (Whistler Centre for Carbohydrate

research),West‐Lafayette(IN),USA,2012† Gums and Stabilizers for Food Industry (Food Hydrocolloids Trust), Wrexham, UK,

2013‡

General courses

VLAGPhDintroductionweek(VLAG),2009 Projectandtimemanagement(WGS),2009 Techniquesforwritingandpresentingascientificpaper(WGS),2010 Scientificwriting(VLAG),2012 CareerPerspectives(WGS),2012

Additional activities

PreparationPhDresearchproposal FoodChemistrystudytriptoGhent,Belgium,2009 PhDtripFCHtoSwitzerland/Italy,2010‡ PhDtripFCHtoSingapore/Malaysia,2012‡ PolyModEprojectmeetings2009/2013 BSc/MScthesisstudentpresentationsandcolloquia,2009/2013 PhDpresentationsFoodChemistry,2009/2013

†Poster;‡PosterandoralpresentationVLAG:GraduateSchoolforNutrition,FoodTechnology,AgrobiotechnologyandHealthSciences

WGS;WageningenGraduateSchoolFCH:FoodChemistry

136

TheworkdescribedinthisthesiswasperformedattheLaboratoryofFoodChemistry,WageningenUniversity,TheNetherlands.Thisresearchwasfinanciallysupportedbythe Commission of the European Communities within the Seventh FrameworkProgrammeforresearchandtechnologicaldevelopment(FP7),GrantagreementNo.222628.

ThisthesiswasprintedbyGVODrukkersenVormgeversB.V./Ponsen&Looijen,Ede,TheNetherlandsEdition:375copies

Coverdesign:AdriaanJ.Gijsberts

MarijnM.Kool,2014

Marijn M. Kool - WUR

Documents

Transcript of Marijn M. Kool - WUR