Synthetic heparin and heparan sulfate oligosaccharides and their protein interactions

Synthetic

heparin

and

heparan

sulfate

oligosaccharides

and

their

protein

interactions

Medel

Manuel

L

Zulueta

,

Shu-Yi

Lin

,

Yu-Peng

Hu

and

Shang-Cheng

Hung

Heparinandheparansulfatebindahostofbasicproteinsthat takeadvantageofthesugar’sdensestructuralinformation.The significanceoftheseinteractionsinvariousaspectsof development,physiology,anddiseasestimulatedkeeninterest inevaluatingstructure–activityrelationships.Thewell-defined heparinandheparansulfateoligosaccharidesneededforthese studiescanbemainlyaccessedbychemicalsynthesisand, morerecentlybychemoenzymaticmeans.Thevarious syntheticstrategiesavailabletochemicalsynthesishave recentlyenabledtheacquisitionofseveralregularandirregular sequences,includinganumberofdodecasaccharides,through improvedcouplingmethodsandjudicialprotectinggroup manipulations.Controlledchainelongationandcritical applicationofmodificationenzymesallowedthegenerationof well-definedconstructsviachemoenzymaticsynthesis. Investigationsofvariousproteininteractionswiththesynthetic constructsdeliveredvaluableinformationthatcouldaidfuture drugdevelopmentendeavors.

Address

1_{GenomicsResearchCenter,AcademiaSinica,128,Section2,} AcademiaRoad,Taipei115,Taiwan

2_{DepartmentofAppliedChemistry,NationalChiaoTungUniversity,} 1001,Ta-HsuehRoad,Hsinchu300,Taiwan

Correspondingauthor:Hung,

Shang-Cheng([email protected])

CurrentOpinioninChemicalBiology2013,17:1023–1029 ThisreviewcomesfromathemedissueSyntheticbiomolecules EditedbyShang-ChengHungandDerekNWoolfson ForacompleteoverviewseetheIssueandtheEditorial Availableonline30thOctober2013

1367-5931/$–seefrontmatter,#2013ElsevierLtd.Allrights reserved.

http://dx.doi.org/10.1016/j.cbpa.2013.10.008

Introduction

Proteoglycans are vitalcomponents of cellsurfaces and

extracellular matrices of animal tissues [1,2]. They are

complex macromolecules that comprise a core protein

and one or more conjugated glycosaminoglycans

(GAGs)—linear polymers with repeating disaccharide

backbones.Whiletheproteinsegmentsdisplayednotable

activities[3,4],thevastmajorityofproteoglycanfunctions

areassociatedwithGAGsofwhichheparansulfate(HS)is

themostheterogeneousandmostwidespread[5].

Alter-nating1!4-linkeda-D-glucosamine(GlcN)andeither

b-D-glucuronic acid (GlcA) or a-L-iduronic acid (IdoA)

makeuptheextendedHSbackbone(Figure1).Potential

sulfationsmayoccuratC3andC6ofGlcNandatC2of

theuronicacid(UA),andtheGlcNaminefunctionmay

be sulfonated, acetylated or unsubstituted. These

vari-ationsaccountto48disaccharidepossibilitieswithinthe

chain.However,onlyabouthalfofthosewereobservedin

Nature, likely due to biological restrictions that also

grantedtissue-specificsulfonationpatternsand

intermit-tentswatchesofunsulfatedregions[6].Hundredsofbasic

proteins,implicatedinfertilization,growthand

develop-ment, bacterial and viral infections, wound healing,

immuneresponse,andcancerprogressionamongothers,

takeadvantageoftherichstructuraldiversityofHS[7].

HSgrantproteinslocalizedavailabilitynearthecell

sur-faceand facilitatevariousmeansof deliveringintended

functions. The biomedical significance of these

inter-actionspromptedintenseinvestigationsaimingto

deter-mine the structural features optimally required for

function.TheantithrombinactivationbytheHSanalog

heparinleadingtothedevelopmentoftheanticoagulant

fondaparinuxhaslonginspired thestudyof HS–protein

associations [8]. Sequestered in vivo by mastocytes,

heparin isgeneratedsimilartoHSandcarriesthesame

disaccharide variations. It is, however, more

homo-geneous with N-sulfonated and 6-O-sulfonated GlcN

(GlcNS6S)and2-O-sulfonatedIdoA(IdoA2S)occupying

mostofthechain[9].Thebindingofantithrombinwitha

distinct 3-O-sulfonated pentasaccharide sequence in

heparin triggers the exposure of the protease reactive

centerloopcapableofdeactivatingfactorsIIaandXaof

thecoagulationcascade [10].

The heparin and HS fragments isolated from natural

sources are typically unsuitable for structure–activity

relationship evaluations because of their polydispersity

andstructuralambiguity.Chemoenzymaticalterationsof

such fragments and that of heparosan, the N-acetyl-D

-glucosamine (GlcNAc)–GlcAcopolymerharvested from

EscherichiacolistrainK5[11,12],toaffordcertaindefined

features only supplied partial information onthe

struc-tural requirements for binding [13]. Here, the

non-uniform starting materials, the incomplete enzymatic

transformations,andthedifficultiesinreaction

monitor-ing and product purification are persistent concerns.

Despite theconsiderableeffort and resourcesinvolved,

chemical synthesis remains the most common and

reliable sourceofwell-definedheparinandHS

sophisticationbyadoptingadvancesinmainstream

carbo-hydratechemistryanddiscoveringnovelwaysindealing

withthechallengesassociatedwiththecomplexstructure

ofheparinandHS.Ontheotherhand,chemoenzymatic

approach progressed by dealing with the problem of

selectivity during enzymatic modifications [18]. The

recent methodologies in generating defined heparin

and HS oligosaccharides and the information obtained

from their biological evaluations are the subject of the

presentreview.

Chemical

synthesis

Asasynthetictargetformanyyears,numerousstrategies

weredisclosedaddressingthechallengesinthechemical

synthesis of heparin and HS oligosaccharides [17].

Orthogonal protecting groups at key positions played

centralrolesinthetransformations,ensuringthe

regios-electivityandstereoselectivityinglycosylationaswellas

the functional group pattern of the target constructs.

Additionally,unnaturalfunctionalgroups,suchaslinkers

that are tailor-made for assay purposes, can be

con-veniently installed on the sugar chain. Novel and

improved methodologies in recent reports contributed

inincreasingthevarietyandcomplexityofthe

synthes-izedstructures.

The repeating disaccharide nature of heparin and HS

motivatedthegenerationofdisaccharidebuildingblocks

toformlongerskeletons(Figure2).Asenzymatic

degra-dationofthenaturalcompounddeliversoligosaccharides

withrepeatingUA–GlcNbackbone,pastsynthesesoften

leaned toward disaccharidebuilding blocks

correspond-ingtothissequence.Notably,ahighernumberofrecent

effortsweredevelopedusingtheGlcN–UAdisaccharide

precursor.Thelatterapproachacknowledgesthegreater

difficultyina-glucosaminylationrelativeto1,2-trans

gly-cosylation involving the UA precursor. With azide

masking the 2-amine function, a-glucosaminylation

particularlyreliesonanomericeffect.a-Stereoselectivity

isfurtherenhancedbyanacceptorwithaxiallyoriented

hydroxyl nucleophile [19], the remote participation by

acylgroups,andthestericinfluenceofbulkygroups.In

particular,Hung showedthattert-butyldiphenylsilyland

p-bromobenzyl groups at the respective 6-O and 3-O

positions of a glucosaminyl donor confer full

a-stereo-selectivity regardless of leaving group, activator, and

acceptor [20]. tert-Butyldimethylsilyl group at4-O also

providedsimilareffectonstereoselectivity[21].TheUA

precursorcanbemadewiththecarboxylfunctionalready

present before chain assembly. Conversely, oxidation

maybe carried out, typically using

2,2,6,6-tetramethyl-1-piperidinyloxylfreeradical,untiltherelevant

glycosi-dicbondshaveformedtoavoidreactivityand

epimeriza-tion issues. The acquisition of the L-idose or IdoA

derivativeis anothermainconcern.Fueledby thehigh

price of the unprotected monosaccharide, several

syn-theticstrategies were developed using cheaper starting

materials[22].Therearerecentupdatesconcerningthe

formationof1,6-anhydro-L-idosebyHung[23]and

intro-duction of various protecting groups in the D

-xylose-derivedIdoAderivativebySeeberger[24].Alternatively,

Gardiner’sgroupdisclosedanewmethodviaD

-xylodial-doseinvolvingthestereoselectivecyanohydrinformation

atC5 intheL-ido configuration[25].

Glycosylations with the same disaccharide building

block is a typical route in generating heparin and HS

oligosaccharide. By this approach, differentlengths can

bereadily prepared leadingto compounds with regular

repeating patterns. Elongations were achieved using

Figure1

R = H or SO3–

R1 _{= H, Ac or SO3}–

Core protein

Linkage region

Heparan sulfate chain

Heparan sulfate proteoglycan

Major repeating unit of heparin Antithrombin-binding sequence

O O –_O 3SO OH –_O 2C O O NH HO OSO3– –_O 3S O O O –_O 3SO OH –_O 2C O O NH O OSO3– –_O 3S O O CO2– HO HO O O O NH HO OSO3– –_O 3S O NH HO OSO3– Ac O –_O 3S O O –_O 2C HO OH O OH O OH HO O OH O OH HO O O HO OH HN O NH O O –_O 2C HO OR O O R1HN OR RO O O –_O 2C HO OR O O R1_HN OR RO O O –_O 2C HO OR O O R1_HN OR RO O n ≈ 50–200

Current Opinion in Chemical Biology

iterative [2+n]-coupling toward the nonreducing end

[23,26,27,28,29,30] or by the initial generation and

further convergentuse of longchain donors[20,31,32].

Thus far, chemical synthesis enabled the assembly of

several dodecasaccharides [26,27,29,31,32] on top of

manyshorterhomologues.Theensuingfunctionalgroup

transformations typically involve partial deblocking of

keyhydroxyls,oxidation,ifneeded,andsulfonationusing

an SO3-based reagent followedby completeremovalof

other O-protectinggroupsusuallythrough

hydrogenoly-sis.Theazidecanbeconvertedtotheamine,acetamide,

or sulfonamide at a synthetically convenient stage.

Becausemulti-O-sulfonationafterchainassemblyisoften

tricky,Huangdemonstratedthefeasibilityofpreinstalled

2,2,2-trichloroethylsulfateesters[33].Fluoroustagswere

also incorporated at either the reducing [30] or

non-reducing end [32] to allow purification via fluorous

solid-phase extraction.Boonsshowed[30]thatrepeated

reagent treatment to afford higher yields is possible in

fluorous-supported heparinandHSsynthesis.

Aside from length variations,heparin and HS diversity

can beemulatedby changingthe mannerof functional

group transformations from assembled skeletons—the

divergent approach—and by using differentially

functionalized building blocks in a modular fashion.

Understandably,representingtheheparinandHS

diver-sitybysynthesisbecomeslessfeasibleaslengthincreases

becauseofconcurrentexponentialincreaseinthenumber

of prospective structures. Complete synthesis was only

accomplished,sofar, onthedisaccharidelevelbyHung

and coworkers[34].Using twoorthogonally protected

disaccharidesprecursors, all48disaccharidepossibilities

in the heparin and HS chains were synthesized using

divergent functional group transformation. Hung [35]

andHuang[33]alsoapplieddivergenttransformationsof

Figure2 Bioassays O O OP PO TO O PO N3 OP _LG OP TO O O N3 OP PO O OP PO OP LG P = Protecting group P1 _{= Alkyl or aryl group}

T = Temporary protecting group for chain elongation or permanent protecting group for nonreducing end building blocks LG = Leaving group

Glycosylation ([2 + n], modular

and/or convergent)

Functional group transformation - deprotection (partial, global) - oxidation, if needed

- O-sulfonation and N-sulfonation - N-acetylation O O OP PO TO O P1_O 2C PO N3 OP _LG TO O O N3 OP PO O CO2P1 PO OP LG

Disaccharide Building Blocks

Various oligosaccharides that emulate the natural heparin and HS chains Chemical Synthesis O HO O –_O 2C HO OH O HO HO OH N NH O O O HO OH O P O P O O O O– _O– HO O HO –_O 2C HO N N N N NH2 O O OH O P O S O O O O– _O– P O O– O– UDP-GlcA UDP-GlcNAc/UDP-GlcNTFA GlcA-AnMan PAPS N NH O O O HO OH O P O P O O O O– _O– HO O AcHNorTFAHN OH HO Chemoenzymatic Synthesis Chain elongation with KfiA and pmHS2 Detrifluoroacetylation N-sulfonation C5-epimerization 2-O-sulfonation 6-O-sulfonation 3-O-sulfonation

Current Opinion in Chemical Biology

disaccharideprecursorstopreparemultipledisaccharide

buildingblocksforfurtheruseinmakingoligosaccharides

with irregularsequences. Proper selectionof protecting

groups,leavinggroups,and glycosylationconditionsled

tothesuccessfulsynthesisofexoticHSstructuressuchas

the 3-O-sulfonated octasaccharide generated by Hung

[35]notable for carryingamine, acetamide,and

sulfo-namide groups.Inseveral instances,commonfully

pro-tectedoligosaccharides(tetrasaccharideandlonger)were

variablytransformedtocreatemultiplefinalproductsthat

mainlydifferinoverallaminesubstitution[21,26,33,36]

or O-sulfonation patterns [29,32,37]. These strategies

enhancetheutilityoftheassembledskeletons.Modular

strategies coupled with divergent transformations were

recently applied to make small sugar libraries

[21,29,36,37]. In Huang’s case [21], several

hexasac-charide skeletons were assembled under

preactivation-based one-pot sequentialglycosylation from the

nonre-ducingtothereducingend.

Thefull-fledgedsynthesisofproteoglycanisstillbeyond

reach. Conjugation between a heparin-like

dodecasac-charidethroughanon-carbohydratelinkerandthelysine

residue of a CD4 mimetic peptide was, nonetheless,

achieved[31].Moreimportantly,aglycopeptidederived

fromsyndecan-1,awidespreadHSproteoglycan,

contain-ingaserine-attached tetrasaccharidelinkageregionand

includesanHS-based tetrasaccharidehasbeen

success-fully synthesized [38]. The synthesis encountered

gly-cosylationandfunctionalgrouptransformationissues,but

theknow-howgainedherecouldlaythegroundworkfor

thepreparationofmore complextargets.

Chemoenzymatic

synthesis

Withsome exemptions,the enzymesthat give heparin

andHStheirremarkablestructurecanbeexpressedinE.

coli in adequate quantities [13,39]. Together with

heparosan’savailabilityasstartingmaterial[12],the

che-moenzymaticapproachattractedinterestsasamethodfor

generating heparin-like and HS-like compounds with

certain desired characteristics. The synthetic cost was

broughtdownsignificantlybyregeneration[40]or

enzy-maticsynthesis[41]ofthesulfotransferasecoenzyme30

-phosphoadenosine-50-phosphosulfate(PAPS),whichcan

also be radiolabeled with 35S to aid detection during

purificationandanalysis.Chemicalsynthesisalsooffered

accesstostartingmaterialsofdefinedlengthstoaddress

the polydispersity of heparosan chains [20,42]. While

particular modifications can be introduced with high

regioselectivity, however, controlling the extent and

locationof the modificationamidnumerous residues in

thesubstratetoaffordawell-definedstructureisdifficult.

Despite many concerns, chemoenzymatic treatment of

heparosansufficientlyprovidedlongoligosaccharidesfor

bioassaysthatareimpracticalbycurrentchemicalmeans.

Iteveneffectivelydelivereda13C/15N-labeledHS-based

octasaccharidefromapurposelylabeledheparosan

poly-mer[43].

Arecentshift inchemoenzymaticsynthesisallowedthe

preparationofdefinedheparin-basedandHS-based

com-pounds.Alternatingapplicationoftwobacterialenzymes,

KfiAfromE. colistrainK5and pmHS2fromPasteurella

multocida,permittedcontrolledpolymerelongationusing

uridinediphosphate (UDP) derivatives of GlcNAc and

GlcA,respectively(Figure2)[39].Aconvenientstarting

unitin this caseisGlcA–AnMan (AnMan:

2,5-anhydro-mannitol)obtainedbychemicaldegradationofheparosan

[44].GlcA–AnManisalsoamenabletofunctionalization,

such as adding a fluorous tag [45,46]. KfiA cannot add

GlcN and selective N-deacetylation is not feasible by

enzymaticorchemicalmeans.Fortunately,

N-trifluoroa-cetylglucosamine(GlcNTFA)canalsobeaddedbyKfiA,

providing selective access to the free amine and its

sulfonation with N-sulfotransferase (NST) and PAPS

[45].NSTisthetruncatedversionofthenatural

bifunc-tional enzyme N-deacetylase/N-sulfotransferase.

Con-versely, pmHS2 adds GlcA to substrates with

nonreducingGlcNAc,GlcNTFA,and theN-sulfonated

GlcN(GlcNS)residues,butnotGlcN[42].C5-epimerase

(C5-epi)acts onGlcAresidues flanked byGlcNS atits

nonreducingsideandGlcNS orGlcNAcatitsreducing

side[45].The GlcAresidue transformationintoIdoAis

knowntobereversible,butaGlcNActhreeresiduesaway

at the nonreducing side was found to influence the

irreversibleIdoAgeneration [47].Concurrent treatment

withC5-epiand2-O-sulfotransferasecausestheselective

formationofIdoA2S.Whilethe6-O-sulfotransferase

iso-forms1and3appeartoindiscriminatelyactondifferent

GlcN residues [46], the 3-O-sulfotransferase (3-OST)

isoform-1only sulfonateGlcNS6Swithanunsulfonated

UAatitsnonreducingside[39].Other3-OSTisoforms

provide differing substrate specificities [39,48,49]. By

proper order andselectionof enzymatictreatment, Liu

generated variouswell-defined compounds, includinga

heptasaccharide carrying the antithrombin-binding

sequenceatagoodyieldandscale[50].Hislaboratory

alsoachieved a 21-meroligosaccharidethathas anti-IIa

and anti-Xa activities [51]. Because contiguous GlcNS

residuesarepresentinthatconstruct,however,reversible

epimerizationoccurredonC5-epitreatmentresultingtoa

mixtureofcompounds.

Protein

interactions

with

synthetic

heparin

and

HS

oligosaccharides

Crucial structural features and parameters involved in

proteininteractionsare revealed byvarious techniques,

including surface plasmon resonance (SPR), isothermal

titration calorimetry (ITC), microarray analysis, affinity

electrophoresis, fluorescence resonance energy transfer,

various competition assays, NMR perturbation, X-ray

crystallography,andmanyothers(Figure2).Thekinetic

participatingresiduesandfunctionalgrouponboth

bind-ingentitiesarebeneficialingeneratingagoodpictureof

theencounter[7].

Growth factors are among themost important group of

proteins that bind HS. With 48 disaccharides, Hung

identified GlcNS–IdoA2S by ITC as the minimum

requirementforfibroblastgrowthfactor-1(FGF-1)

inter-actionthatoccursatashallowbindingsiteintheprotein

revealed by X-ray analysis [34]. Aside from the

N-sulfonateand2-O-sulfonate functionalities,the

3-O-sul-fonate,ifpresent,andthe3-hydroxylofIdoAcontribute

to the encounter. FGF-2 binds the disaccharide

GlcNS6S–IdoA2S,albeitweak [28].Usinglongersugars,

Huang and Liu determined the importance of IdoA2S

and GlcNSinbindingto FGF-2andanadditional

6-O-sulfonate group gives a twofold affinity enhancement

[42]. Jayson and coworkers showed that theinteraction

of a GlcNS–IdoA2S repeating sequence with vascular

endothelial growth factor165 is considerably weaker

than thatwithFGF-2and along sugar,suchas a

dode-casaccharide, is needed for effective protein inhibition

[26].

HSoftenassists bacterialandviralentryintohostcells.

SPR competitive assays indicated that the HS

dodeca-saccharide–CD4mimicconjugateinhibitsthebindingof

immobilized CD4 with gp120, a human

immunodefi-ciencyvirus(HIV)envelopeglycoproteinneededforcell

entry [31]. Binding of the CD4 mimic exposed the

adjacent coreceptordomain ofgp120,whichhasaffinity

totheHSdodecasaccharide.Thiscooperativebehaviorof

the conjugated components enabled the inhibition of

HIV entry onto peripheral blood mononuclear cells.

The dodecasaccharide carries the IdoA2S–GlcNS6S

repeating unit and is at an appropriate length to cover

thecoreceptordomainofgp120.Conversely,two

differ-ent3-O-sulfonated octasaccharidesinhibited theherpes

simplex virus-1 (HSV-1) infection of Vero cells in a

dosage dependent manner [35].A 3-O-sulfonated HS

is knownto interact withHSV-1 envelope glycoprotein

gDenablingviralentry.Thesimilarinhibitionprofilesof

thetwosugarssuggestminorcontributionofother

struc-turalfeaturesandthelocationofthe3-O-sulfonategroup

ontheextentofgDbinding.Heparin-binding

hemagglu-tinin (HBHA) is the virulence factor crucial for

extra-pulmonary dissemination of Mycobacterium tuberculosis.

ITC experiments identified a hexasaccharide with

GlcNS6S–IdoA2S repeating sequence as the shortest

sugarthatinteractswithHBHAinanentropicallydriven

manner [20].

Assayswithotherproteinsalsorevealedinterestingresults.

Competitive inhibition indicated that aheptasaccharide

withGlcNS6S–IdoA2Srepeatscaninhibitthebindingof

eosinophil-derived neurotoxin (EDN) to Beas-2B cells

[23]. Fluorescence-assisted carbohydrate electrophoresis

showed that the shorter pentasaccharide of the same

sequence possess theability to bindeosinophil cationic

protein(ECP),acloserelativeofEDN[52].UsingNMR

perturbation,thedissociationconstantofthetrisaccharide

GlcNS6S–IdoA2S–GlcNS6S and ECP was measured at

around15–30mMwiththenonreducingendpositionedat

theproteininterior[53].ConcerningtheAlzheimer’s

dis-ease-related protease BACE-1, interaction with both

sequences containing IdoA/GlcNS and GlcA/GlcNAc

suggests the probability of dual binding regions [36].

Octasaccharides and longer sugars with GlcNAc6S–

UA2S units showedpotent binding with BACE-1[29].

SPRandmicroarrayexperimentswiththenatural

cytotox-icityreceptorsNKp30,NKp44,andNKp46usinga

com-poundlibrarydenotebindingtothehighlychargedregions

ofthesugar,butwithdifferingindividualspecificitiesand

lengthdependencies[54].

Conclusion

InamoleculeascomplexasHS,onlyasmallsamplingofa

huge structural potential have been evaluated. This

reflectsonthemanydifficultiesassociatedwiththe

acqui-sition of well-defined oligosaccharides by chemical and

chemoenzymatic means.Nevertheless,current advances

keeppushingtheboundariesofstructuralcomplexityand

the effectivenessandefficiencyofthesyntheticprocess.

Recentevaluationsidentifytheimportantstructural

fea-turesinthesugarthatcouldformthefoundationsoffuture

studies. With the antithrombin-binding sequence that

startedarethinkingofHS–proteininteractions,the

motiv-ations foridentifying new candidatedrugs derivedfrom

heparinandHSarestrongerthanever.

Acknowledgements

ThiscontributionissupportedbytheNationalScienceCouncil,National HealthResearchInstitute,andAcademiaSinica.

References

and

recommended

reading

Papersofparticularinterest,publishedwithintheperiodofreview, havebeenhighlightedas:

 ofspecialinterest ofoutstandinginterest

1. Ha¨ckerU,NybakkenK,PerrimonN:Heparansulphate proteoglycans:thesweetsideofdevelopment.NatRevMol CellBiol2005,6:530-541.

2. CouchmanJR,PatakiCA:Anintroductiontoproteoglycansand theirlocalization.JHistochemCytochem2012,60:885-897. 3. ErikssonAS,SpillmannD:Themutualimpactofsyndecan-1and

itsglycosaminoglycanchains—amultivariablepuzzle.J HistochemCytochem2012,60:936-942.

4. ChoiY,ChungH,JungH,CouchmanJR,OhE-S:Syndecansas cellsurfacereceptors:uniquestructureequateswith functionaldiversity.MatrixBiol2011,30:93-99. 5. BishopJR,SchukszM,EskoJD:Heparansulphate

proteoglycansfine-tunemammalianphysiology.Nature2007, 446:1030-1037.

6. EskoJD,SelleckSB:Orderoutofchaos:assemblyofligand bindingsitesinheparansulfate.AnnuRevBiochem2002, 71:435-471.

7. GandhiNS,ManceraRL:Thestructureofglycosaminoglycans andtheirinteractionswithproteins.ChemBiolDrugRes2008, 72:455-482.

8. PetitouM,vanBoeckelCAA:AsyntheticantithrombinIIIbinding pentasaccharideisnowadrug!Whatcomesnext? Angew ChemIntEd2004,43:3118-3133.

9. RabensteinDL:Heparinandheparansulfate:structureand function.NatProdRep2002,19:312-331.

10. vanBoeckelCAA,PetitouM:TheuniqueantithrombinIIIbinding domainofheparin:aleadtonewsyntheticantithrombotics. AngewChemIntEdEngl1993,32:1671-1690.

11. WangZ,LyM,ZhangF,ZhongW,SuenA,HickeyAM,DordickJS, LinhardtRJ:E.coliK5fermentationandthepreparationof heparosan,abioengineeredheparinprecursor.Biotechnol Bioeng2010,107:964-973.

12. LiP,ShengJ,LiuY,LiJ,LiuJ,WangF:Heparosan-derived heparansulfate/heparin-likecompounds:onekindof potentialtherapeuticagents.MedResRev2013,33:665-692. 13. PetersonS,FrickA,LiuJ:Designofbiologicallyactiveheparan

sulfateandheparinusinganenzyme-basedapproach.Nat ProdRep2009,26:610-627.

14. PolettiL,LayL:Chemicalcontributionstounderstanding heparinactivity:synthesisofrelatedsulfated

oligosaccharides.EurJOrgChem2003:2099-3024. 15. NotiC,SeebergerPH:Chemicalapproachestodefinethe

structure–activityrelationshipofheparin-like glycosaminoglycans.ChemBiol2005,12:731-756.

16. Bonnaffe´ D:Bioactivesyntheticheparansulfateandheparin derivatives:fromlongfragmentsmimeticstochimeras.CR Chim2011,14:59-73.

17. 

DulaneySB,HuangX:Strategiesinsynthesisofheparin/ heparansulfateoligosaccharides:2000–present.Adv CarbohydrChemBiochem2012,67:95-136.

This is an excellent review that summarizes the achievements and strategies in the chemical synthesis ofheparin and heparan sulfate oligosaccharidesinthepastdecade.

18. LinhardtRJ,LiuJ:Syntheticheparin.CurrOpinPharmacol2012, 12:217-219.

19. OrgueiraHA,BartolozziA,SchellP,SeebergerPH: Conformationallockingoftheglycosylacceptorfor stereocontrolinthekeystepinthesynthesisofheparin. AngewChemIntEd2002,41:2128-2131.

20. 

ZuluetaMML,LinS-Y,LinY-T,HuangC-J,WangC-C,KuC-C, ShiZ,ChyanC-L,IreneD,LimL-H:GlycosylationbyD -glucosamine-deriveddonors:synthesisofheparosanand heparinanaloguesthatinteractwithmycobacterial heparin-bindinghemagglutinin.JAmChemSoc2012,134:8988-8995. Theauthorsevaluatedtheeffectsofprotectinggroupson a-glucosami-nylationand prepareddisaccharidebuilding blocks that werefurther utilizedtogeneratedifferentlengthsofheparosanandheparinanalogues. Theinteractionoftheheparinanalogueswithheparin-binding hemag-glutininwasalsoinvestigated.

21. 

WangZ,XuY,YangB,TiruchinapallyG,SunB,LiuR, DulaneyS,LiuJ,HuangX:Preactivation-based,one-pot combinatorialsynthesisofheparin-likehexasaccharidesfor theanalysisofheparin–proteininteractions.ChemEurJ2010, 16:8365-8375.

Theauthorsapplied preactivationtoprepare severalhexasaccharide skeletonsinonepot.They alsoapplieddivergenttransformations to generatetwofinalproductsforeveryfullyprotectedprecursor. 22. ZuluetaMML,ZhongY-Q,HungS-C:SynthesisofL-hexosesand

theirrelatedbiomolecules.ChemCommun2013,49:3275-3287. 23. HungS-C,LuX-A,LeeJ-C,ChangMD-T,FangS-L,FanT-C,

ZuluetaMML,ZhongY-Q:Synthesisofheparin

oligosaccharidesandtheirinteractionwitheosinophil-derived neurotoxin.OrgBiomolChem2012,10:760-772.

24. Bindscha¨dlerP,AdibekianA,Gru¨nsteinD,SeebergerPH:Denovo synthesisofdifferentiallyprotectedL-iduronicacid

glycosylatingagents.CarbohydrRes2010,345:948-955.

25. HansenSU,Bara´thM,SalamehBAB,PritchardRG,StimpsonWT, GardinerJM,JaysonGC:ScalablesynthesisofL-iduronicacid derivativesviastereocontrolledcyanohydrinreactionfor synthesisofheparin-relateddisaccharides.OrgLett2009, 11:4528-4531.

26. ColeCL,HansenSU,Bara´thM,RushtonG,GardinerJM, AvizienyteE,JaysonGC:Syntheticheparansulfate oligosaccharidesinhibitendothelialcellfunctionsessential forangiogenesis.PLoSONE2010,5:e11644.

27. 

HansenSU,MillerGJ,JaysonGC,GardinerJM:Firstgram-scale synthesisofaheparin-relateddodecasaccharide.OrgLett 2013,15:88-91.

Longchainheparinandheparansulfateoligosaccharidesaredifficultto synthesizeandtheauthorsofthispaperachievedadodecasaccharidein gram-scale.

28. MazaS,MacchioneG,OjedaR,Lopez-PradosJ,AnguloJ,de PazJL,NietoPM:Synthesisofamine-functionalizedheparin oligosaccharidesfortheinvestigationofcarbohydrate– proteininteractionsinmicrotiterplates.OrgBiomolChem 2012,10:2146-2163.

29. 

Schwo¨rerR,ZubkovaOV,TurnbullJE,TylerPC:Synthesisofa targetedlibraryofheparansulfatehexa-to

dodecasaccharidesasinhibitorsofb-secretase:potential therapeuticsforAlzheimer’sdisease.ChemEurJ2013, 19:6817-6823.

Theauthorsprepared16compoundsrangingfromhexasaccharideto dodecasaccharideswithregularandirregularstructures.Divergent func-tional group transformations were also applied to generate multiple productsfromcommonprecursors.

30. ZongC,VenotA,DhamaleO,BoonsG-J:Fluoroussupported modularsynthesisofheparansulfateoligosaccharides.Org Lett2013,15:342-345.

31. BaleuxF,Loureiro-MoraisL,HersantY,ClayetteP,

Arenzana-SeisdedosF,Bonnaffe´ D,Lortat-JacobH:Asynthetic CD4-heparansulfateglycoconjugateinhibitsCCR5and CXCR4HIV-1attachmentandentry.NatChemBiol2009, 5:743-748.

32. TanakaH,TatenoY,TakahashiT:Convergentstereoselective synthesisofmultiplesulfatedGlcNa(1,4)GlcAa(1,4) dodecasaccharides.OrgBiomolChem2012,10:9570-9582. 33. TiruchinapallyG,YinZ,El-DakdoukiM,WangZ,HuangX:

Divergentheparinoligosaccharidesynthesiswithpreinstalled sulfateesters.ChemEurJ2011,17:10106-10112.

34. 

HuY-P,ZhongY-Q,ChenZ-G,ChenC-Y,ShiZ,ZuluetaMML, KuC-C,LeeP-Y,WangC-C,HungS-C:Divergentsynthesisof 48heparansulfate-baseddisaccharidesandprobingthe specificsugar–fibroblastgrowthfactor-1interaction.JAm ChemSoc2012,134:20722-20727.

Theauthorsgenerated48disaccharidesbasedonallpossible disac-charidesfoundinheparinandheparansulfateusingdivergentsynthesis fromonlytwodisaccharideprecursors.Fourdisaccharideswerefoundto bindfibroblastgrowthfactor-1usingisothermaltitrationcalorimetry. X-ray cocrystal analysis further showed the structural features of the interaction.

35. 

HuY-P,LinS-Y,HuangC-Y,ZuluetaMML,LiuJ-Y,ChangW, HungS-C:Synthesisof3-O-sulfonatedheparansulfate octasaccharidesthatinhibittheherpessimplexvirustype1 host–cellinteraction.NatChem2011,3:557-563.

Twoirregular3-O-sulfonatedoctasaccharideswereprepared.The octa-saccharidesinhibitedtheinfectionofVerocellsbyherpessimplexvirus-1 inadosagedependentmanner.

36. ArungundramS,Al-MafrajiK,AsongJ,Leach FEIII,AmsterIJ, VenotA,TurnbullJE,BoonsG-J:Modularsynthesisofheparan sulfateoligosaccharidesforstructure–activityrelationship studies.JAmChemSoc2009,131:17394-17405.

37. Mun˜oz-Garcı´a JC,Lo´pez-PradosJ,AnguloJ,Dı´az-ContrerasI, ReichardtN,dePazJL,Martı´n-Lomas M,NietoPM:Effectofthe substituentsoftheneighboringringintheconformational equilibriumofiduronateinheparin-liketrisaccharides.Chem EurJ2012,18:16319-16331.

38. YangB,YoshidaK,YinZ,DaiH,KavunjaH,El-DakdoukiMH, SungsuwanS,DulaneySB,HuangX:Chemicalsynthesisofa

heparansulfateglycopeptide:syndecan-1.AngewChemIntEd 2012,51:10185-10189.

39. 

ZhouX,O’LearyTR,XuY,ShengJ,LiuJ:Chemoenzymatic synthesisofheparansulfateandheparin.BiocatalBiotransform 2012,30:296-308.

An excellentreviewofthechemoenzymatic synthesisofheparin and heparansulfatethatbroadlycoversrecentadvances.

40. BurkartMD,IzumiM,ChapmanE,LinC-H,WongC-H: RegenerationofPAPSfortheenzymaticsynthesisofsulfated oligosaccharides.JOrgChem2000,65:5565-5574.

41. ZhouX,ChandarajotiK,PhamTQ,LiuR,LiuJ:Expressionof heparansulfatesulfotransferasesinKluyveromyceslactisand preparationof30_{-phosphoadenosine-5}0_{-phosphosulfate.}

Glycobiology2011,21:771-780.

42. XuY,WangZ,LiuR,BridgesAS,HuangX,LiuJ: Directingthebiologicalactivitiesofheparansulfate oligosaccharidesusingachemoenzymaticapproach. Glycobiology2012,22:96-106.

43. LaguriC,SapayN,SimorreJ-P,BrutscherB,ImbertyA,GansP, Lortat-JacobH:13C-labeledheparansulfateanalogueasatool tostudyprotein/heparansulfateinteractionsbyNMR spectroscopy:applicationtotheCXCL12achemokine.JAm ChemSoc2011,133:9642-9645.

44. ChenM,BridgesA,LiuJ:Determinationofthesubstrate specificitiesofN-acetyl-D-glucosaminyltransferase.

Biochemistry2006,45:12358-12365.

45. LiuR,XuY,ChenM,Weı¨wer M,ZhouX,BridgesAS,DeAngelisPL, ZhangQ,LinhardtRJ,LiuJ:Chemoenzymaticdesignof heparansulfateoligosaccharides.JBiolChem2010, 285:34240-34249.

46. LiuR,LiuJ:Enzymaticplacementof6-O-sulfogroupsin heparansulfate.Biochemistry2011,50:4382-4391.

47. ShengJ,XuY,DulaneySB,HuangX,LiuJ:Uncoveringbiphasic catalyticmodeofC5-epimeraseinheparansulfate

biosynthesis.JBiolChem2012,287:20996-21002.

48. XuD,MoonAF,SongD,PedersenLC,LiuJ:Engineering sulfotransferasestomodifyheparansulfate.NatChemBiol 2008,4:200-202.

49. NguyenTKN,ArungundramS,TranVM,RamanK,Al-MafrajiK, VenotA,BoonsG-J,KuberanB:Asyntheticheparansulfate oligosaccharidelibraryrevealsthenovelenzymaticaction ofD-glucosaminyl3-O-sulfotransferase-3a.MolBiosyst2012, 8:609-614.

50. 

XuY,MasukoS,TakieddinM,XuH,LiuR,JingJ,MousaSA, LinhardtRJ,LiuJ:Chemoenzymaticsynthesisof

homogeneousultralowmolecularweightheparins.Science 2011,334:498-501.

Usingthechemoenzymaticapproach,theauthorseffectivelyprepared two heptasaccharides carrying theantithrombin-binding sequence in goodscaleandyield, demonstratingeffectivecontrolofselectivityof theappliedenzymes.Theapproachallowedthepreparationofavariant offondaparinuxwithconsiderablyfewerstepsthanchemicalsynthesis. 51.



XuY,PempeEH,LiuJ:Chemoenzymaticsynthesisofheparin oligosaccharideswithbothanti-factorXaandanti-factorIIa activities.JBiolChem2012,287:29054-29061.

Longoligosaccharidesofupto21-merwerepreparedby chemoenzy-maticapproach.Thesugarsincludeanantithrombin-bindingregionatthe reducingendandafactorIIa-bindingregionatthenonreducingend.

52. FanT-C,FangS-L,HwangC-S,HsuC-Y,LuX-A,HungS-C,Lin S-C,ChangMD-T:Characterizationofmolecularinteractions betweeneosinophilcationicproteinandheparin.JBiolChem 2008,283:25468-25474.

53. Garcı´a-MayoralMF,CanalesA´,Dı´azD,Lo´pez-PradosJ, MoussaouiM,dePazJL,AnguloJ,NietoPM,Jime´nez-BarberoJ, BoixEetal.:Insightsintotheglycosaminoglycan-mediated cytotoxicmechanismofeosinophilcationicproteinrevealed byNMR.ACSChemBiol2012,8:144-151.

54. HechtM-L,RosentalB,HorlacherT,HershkovitzO,DePazJL, NotiC,SchauerS,PorgadorA,SeebergerPH:Natural cytotoxicityreceptorsNKp30,NKp44andNKp46bindto differentheparansulfate/heparinsequences.JProteomeRes 2009,8:712-720.