Chemical Approaches to Glycobiology...Chemical Approaches to Glycobiology Laura L. Kiessling1,2 and Rebecca A. Splain1 1Department of Chemistry, 2Department of Biochemistry, University

ANRV413-BI79-22 ARI 27 April 2010 21:25

Chemical Approachesto GlycobiologyLaura L. Kiessling1,2 and Rebecca A. Splain1

1Department of Chemistry, 2Department of Biochemistry, University ofWisconsin–Madison, Wisconsin 53706; email: [email protected]

Annu. Rev. Biochem. 2010. 79:619–53

First published online as a Review in Advance onApril 8, 2010

The Annual Review of Biochemistry is online atbiochem.annualreviews.org

This article’s doi:10.1146/annurev.biochem.77.070606.100917

Copyright c© 2010 by Annual Reviews.All rights reserved

0066-4154/10/0707-0619$20.00

Key Words

array, glycan, glycomimetic, glycosylation, lectin, multivalency

AbstractGlycans are ubiquitous components of all organisms. Efforts toelucidate glycan function and to understand how they are assembledand disassembled can reap benefits in fields ranging from bioenergyto human medicine. Significant advances in our knowledge of glycanbiosynthesis and function are emerging, and chemical biology ap-proaches are accelerating the pace of discovery. Novel strategies forassembling oligosaccharides, glycoproteins, and other glycoconjugatesare providing access to critical materials for interrogating glycanfunction. Chemoselective reactions that facilitate the synthesis ofglycan-substituted imaging agents, arrays, and materials are yieldingcompounds to interrogate and perturb glycan function and dysfunction.To complement these advances, small molecules are being generatedthat inhibit key glycan-binding proteins or biosynthetic enzymes.These examples illustrate how chemical glycobiology is providing newinsight into the functional roles of glycans and new opportunities tointerfere with or exploit these roles.

619

Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

Glycan: a genericterm referring to amonosaccharide,oligosaccharide,polysaccharide, or itsconjugate (e.g.,glycolipid,glycoprotein, or otherglycoconjugate)

Glycoconjugate: oneor more saccharideunits (glycone)covalently linked to anoncarbohydratemoiety (aglycone)

Contents

INTRODUCTION . . . . . . . . . . . . . . . . . . 620GLYCAN SYNTHESIS . . . . . . . . . . . . . . 622

Chemical Synthesisof Oligosaccharides . . . . . . . . . . . . . 623

Engineering Enzymesfor Glycan Synthesis . . . . . . . . . . . . 626

Glycoprotein andGlycopeptide Synthesis . . . . . . . . . 628

Chemical Glycobiology ofGlycolipids . . . . . . . . . . . . . . . . . . . . . 629

Chemoselective Reactionsto Modify Glycans . . . . . . . . . . . . . . 629

INTERROGATION OFGLYCAN RECOGNITION. . . . . . . 632Glycan Arrays . . . . . . . . . . . . . . . . . . . . . 632Lectin Arrays . . . . . . . . . . . . . . . . . . . . . . 634

PERTURBATION OFGLYCAN FUNCTION . . . . . . . . . . . 634Perturbation of Protein-Glycan

Recognition with MonovalentLigands . . . . . . . . . . . . . . . . . . . . . . . . 635

Perturbation of Protein-GlycanRecognition with MultivalentLigands . . . . . . . . . . . . . . . . . . . . . . . . 637

Perturbation of Glycan Assembly . . . 640Exploiting Alternative Substrates

in Glycan Biosynthesis . . . . . . . . . . 643Illuminating Glycan Biosynthesis . . . 644

CONCLUSION . . . . . . . . . . . . . . . . . . . . . 646

INTRODUCTION

Glycans, which are compounds that includemonosaccharides, oligosaccharides, polysac-charides, and their conjugates, are criticalconstituents of all organisms. Members of aglycan subset, the polysaccharides, are the mostabundant organic compounds on Earth. Gly-coconjugates (e.g., peptidoglycan, glycolipids,glycoproteins) also are prevalent. In humans,for example, half of all proteins are glycosy-lated (1). Consistent with glycan abundance innature, data from genomic sequencing indicate

that approximately 1% of each genome,from eubacteria to archea and eukaryotes, isdedicated to sugar-processing enzymes (2).Moreover, these genes can be highly conserved,as the components of few other biochemicalpathways are so invariant as those responsiblefor glycan biosynthesis (3). The importance ofthis conservation is underscored by data indi-cating that defects in the glycan biosyntheticmachinery in humans, known as congenital dis-orders of glycosylation, are rare and generallyhave severe deleterious consequences (4).

Genomic analysis is a powerful means toidentify enzymes that generate or degrade gly-cans and the proteins that recognize the gly-can products. Still, it does not reveal whatglycans are present in a cell or organism be-cause the synthesis of glycans is not templatedirected. As a result, elucidating the molecu-lar mechanisms that underlie glycan functionhas been a challenge. Nevertheless, researchershave uncovered numerous roles for glycans,including those in fertilization and develop-ment, hormone function, cell proliferation andorganization, host-pathogen interactions, andthe inflammatory and immune responses (3).These findings are providing additional impe-tus to devise new approaches that meet the chal-lenges of elucidating and manipulating glycanfunction.

The increased appreciation for the ubiquityof glycans and their importance to humanhealth has spawned the field of chemicalglycobiology. Because of the complexities ofglycans, their study has compelled researchersto pursue interdisciplinary approaches. Sincethe pioneering contributions of 1902 NobelLaureate Emil Fischer, it has been apparentthat our understanding of glycan functioncan be advanced using approaches that spanbiology and chemistry. The nucleation of thediscipline of chemical biology is yielding newand innovative strategies to probe glycan func-tion (5). Indeed, there has been an explosionof research in this area. As a result, this reviewcannot provide comprehensive coverage of thefield but rather offers an overview of select

620 Kiessling · Splain

Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

advances that illustrate the unique contribu-tions and exciting opportunities within thefield of chemical glycobiology.

The state of the art of chemical glycobiologyis focused on key questions: How are glycansmade and degraded, what are their biologicalroles once in place, and how can these rolesbe exploited? To address these questions,researchers have employed the complementarystrategies of interrogation and perturbation(Figure 1). The interrogation strategy strives tounderstand endogenous interactions betweennatural glycans and their cognate enzymes orbinding partners. Access to naturally occurringand novel glycans provides the means toexamine protein-glycan or enzyme-glycan in-teractions. Arrays composed of glycoconjugates(Figure 1a) or lectins (Figure 1b) are valuabletools for interrogating protein-binding speci-ficity or cellular glycosylation patterns. Withthe complementary perturbation approach, in-hibitors, analogs, or other nonnatural substratescan serve as probes of both the biosynthesis andthe biological roles of glycans. Indeed, novelnonnatural oligosaccharide mimics or syn-thetic glycoconjugates can inhibit or encouragespecific biomolecular interactions within cellsand organisms (Figure 1c,d ). Moreover,compounds have been identified that can blockkey steps within glycan biosynthetic pathways(Figure 1e). Finally, carbohydrate analogs canbe incorporated into glycans using the cellularbiosynthetic machinery (Figure 1f ). Such

−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→Figure 1Interrogation and perturbation in chemicalglycobiology. (a) Glycan arrays have been developedto interrogate the binding specificities of lectins(blue), antibodies (green), and other glycan-bindingproteins (orange). (b) Lectin arrays can be used tofingerprint cell-surface or pathogen glycosylationpatterns. Monovalent (c) and multivalent (d ) ligandsfor glycan-binding proteins can perturb protein-glycan interactions. (e) Inhibitors can prevent keysteps in glycan biosynthesis, thereby reducing theproduction of specific glycan structures.( f ) Nonnatural monosaccharides can serve assubstrates for biosynthetic enzymes and thereby beincorporated into glycans.

Glycobiology: thestudy of sugars inbiological systems,including theirstructures,biosynthesis, andphysiological roles

agents can be used for purposes ranging fromimaging glycans to cross-linking them to theirbinding partners. Together, these chemicalstrategies are illuminating the molecularmechanisms that underlie glycan function.

Glycan-bindingprotein Lectin

Antibody

Cell

Interrogation

Perturbation

InhibitorLectin

Surface

Glycanarray

Lectin

Multivalent ligand

Cell Inhibitor

Glycosyl-transferase

Normalsugar

Nonnaturalsugar

a

f

e

b

d

c

www.annualreviews.org • Chemical Approaches to Glycobiology 621

Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

Lectin: a glycan-binding protein ofnonimmune origin

GLYCAN SYNTHESISDefined oligosaccharides and glycoconjugatesare critical for unraveling the function of gly-cans. Obtaining these entities from naturalsources is difficult because their productiongenerally involves the participation of multipletransporters and enzymes (6). This complex-ity is illustrated by the pathway for eukaryoticglycoprotein synthesis (Figure 2). The saccha-ride building blocks (typically nucleotide sug-ars) must be generated and then transportedto the appropriate cellular location, where theycan be used by glycosyltransferases. The effi-ciency of producing any particular glycan de-pends on the concentration of building blocks,what glycosyltransferases and other biosyn-thetic enzymes are present, and the Km valuesof those building blocks for the glycosyltrans-ferases that use them. Pathways for the pro-duction of N-glycoproteins, O-glycoproteins,

glycolipids, glycosylphosphatidylinositol an-chors, proteoglycans, and polysaccharides areinfluenced by accessibility of the nucleotidedonors, but the mechanisms governing the reg-ulation of these pathways are still being eluci-dated. Thus, it is difficult to obtain sufficientquantities of glycans for study from biologicalsources.

Chemical strategies are addressing this de-ficiency by providing the means to generatean ever-increasing diversity of glycans. Natu-rally occurring glycans can be synthesized, ascan derivatives. In this way, critical structure-activity relationships can be elucidated. Thereare two general approaches for the synthesisof oligosaccharides: chemical and enzymatic.Here, we outline some of the major advancesthat have occurred on both fronts. More de-tailed information can be found in severalexcellent reviews (7–10).

Monosaccharidedonors

Cytosol

Cell-surfaceglycoconjugates

Monosaccharides

ER/Golgi

Metabolicinterconversions

Glycoconjugate assemblyin the secretory compartments

Extracellular milieu

Glycan-specific receptor

Figure 2Schematic depiction of glycoconjugate biosynthesis and cell-surface recognition of glycans. Mostexogenously supplied monosaccharides are taken up by cells and converted to monosaccharide donors in thecytosol. The donors are imported into the endoplasmic reticulum (ER) and Golgi compartments, where theyare used by glycosyltransferases to assemble glycoconjugates. In the case of N-linked glycoproteins, a coreoligosaccharide is assembled in the cytosol, transported into the ER where it is processed by glycosidases,and further elaborated by glycosyltransferases. Once displayed in fully mature forms on the cell surface, theglycoconjugates can serve as ligands for soluble lectins, cell-surface glycan-binding proteins, or glycan-binding proteins on other cells or pathogens. In principle, chemical glycobiology can yield molecules thatcan be used to inhibit or promote any stage of this process.


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

Chemical Synthesisof Oligosaccharides

The chemical synthesis of oligosaccharidesoffers tremendous flexibility. It can give riseto diverse glycans, including those availablein minute quantities from biological sourcesor those for which the biosynthetic enzymesare unknown. Moreover, chemical synthesisprovides the means to test the importance ofdifferent functional groups because nonnaturalsugars can be introduced. The power of chem-ical synthesis is illustrated by the developmentof potent and effective sulfated saccharides asanticoagulants (11). These defined compoundswere inspired by analysis of the properties ofheparin, an anionic glycosaminoglycan (GAG)that has long been used as an anticoagulant.Heparin can bind antithrombin III, therebyproducing a complex that blocks blood clot-ting. Pharmaceutical-grade heparin is typicallyisolated from porcine intestinal mucosa as amixture of sulfated polysaccharides. Chemicalsynthesis was important in verifying thatheparin’s activity resides in a critical pentasac-charide recognition sequence. These findingsindicate that proteins like antithrombin IIIcan recognize specific sulfated oligosaccharidesequences within GAGs. Additionally, they ledto development of the defined anticoagulantdrug Arixtra®. The recent contamination ofheparin isolated from biological sources, inwhich the presence of other sulfated polysac-charides led to over 100 deaths, highlightsthe utility of therapeutics that are based ondefined, synthetic glycans (12).

The chemical synthesis of oligosaccharidesappears deceptively simple. It involves theformation of glycosyl bonds, a reaction firstdescribed in 1893 (13). The approach that wasemployed then is similar to that used by natureand remains the preferred strategy for chemicalsynthesis: A donor monosaccharide, equippedwith a leaving group at the anomeric position,undergoes reaction with a nucleophilic groupon an acceptor (Figure 3a). In chemicalsynthesis, a promoter is added to the donormonosaccharide to facilitate the departure ofthe leaving group.

GAG:glycosaminoglycan

The simplicity of this approach beliesits complexity. Glycosylation reactions areregulated by the reactivity-selectivity principleof organic chemistry. The anomeric groupmust be sufficiently prone to leaving, such thata relatively poor nucleophile like a hydroxylgroup can engage in bond formation; however,the donor must not be so reactive that bondformation occurs without stereocontrol. Thus,whether the reaction occurs via an SN2-like(inversion) or SN1-like (oxocarbenium inter-mediate) pathway has a critical influence on thestereochemical outcome. For this reason, thedual problems of regiochemical control andstereoselectivity of glycosylation reactions areintertwined. Fraser-Reid and coworkers’ clas-sification of donors as “armed” (fast reacting)or “disarmed” (slow reacting) has led to majoradvances in the field (14) because it offersinsight into how the electronics of the glycosyldonor can be manipulated to control glyco-sylation reaction outcomes. The reactivity ofthe donor can be tuned by modifying severalfactors, including the electron-withdrawingability of the protecting groups, the labilityof the anomeric leaving group, the method ofleaving group activation, the conformation ofthe donor or acceptor, and the nature of thesolvent. These changes in reactivity impactglycosylation reaction stereochemistry becausethey influence the reaction mechanism (SN2-like versus SN1-like) (Figure 3a). It is possibleto tune a series of glycosylation reactions togain outstanding stereoselectivity.

A longstanding yet clever means to achiev-ing stereocontrol is to exploit protectinggroups that can influence glycosylation stereo-chemistry via neighboring group participation(Figure 3b). In the paradigmatic example, a 2-acyl group forms an acetoxonium ion by attackonto the anomeric carbon. With a protectedglucose derivative, the 1,2-acetoxonium ionblocks nucleophilic attack of the acceptor fromthe α-face and results in the formation of β-glucosides; for mannose, the β-face is blocked,and α-mannosides are produced. This strategyhas been used extensively to generate β-glucosides, β-galactosides, α-mannosides, and


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

α-rhamnosides. More recently, novel variationson the neighboring group participation strategyhave been employed to access a wider array of

glycosidic linkages, including α-glucosides andα-galactosides (7). Still, many key glycosidiclinkages cannot be formed via neighboring

O

PO

OPO

OP'

OP'

OR2. Protecting group

removal

1. Promoter

Glycosyl acceptorGlycosyl donor

P = Bn (armed)or Ac (disarmed)

LG

P = Bn

P = Ac+ or

a

b

Promoter ROH

Least reactive donor

cPromoter

Less reactive donorMost reactive donor

+ Promoter

1. Promoter, ROH

Promoter

2. Selective protectinggroup removal

LinkerHO LinkerO

1.

Promoter

Cleavage fromsupport

Protecting groupremoval

d

POPO

HOP'O

O

OH

OH

HOHO

O

OH

OH

OROHO

O

HO

OH

HOHO

O

OH

OH

OROHO

O

NH

OP

POPO LG

O

O

HN+

OP

POPO

O

O

NH

OP

POPO OR

O

O

OP

OP

LGPO

POO

OP'

OP'

LG'HO

P'O

O

OP"

OP"

LG"HO

"PO

2. Protecting groupremoval

O

OH

OH

OHO

O

OH

OH

OROHO

O

OH

OH

HOHO

O

OP

OP

LGPO

O

OP

OP

HOPO

O

OP

OP

LGPO

PO

O

OH

OH

HOHO

O

OH

OH

OROHO

P1O


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

group participation strategies. Important ex-amples include β-mannosides, β-rhamnosides,and sialic acid derivatives. New insight into theinterplay of stereoelectronic and conforma-tional effects is providing strategies to assemblethese kinds of linkages (15). The synthesisof sialic acid derivatives has been especiallychallenging because the anomeric positionis more hindered and possesses an electron-withdrawing group, but the use of protectinggroups to alter thiosialoside conformation incombination with novel conditions for donoractivation has led to dramatic improvements inglycosylation yields (16–18). The value of thesenew approaches is illustrated by the synthesisof an α2,9-trisialic acid oligomer in a singlereaction vessel (one-pot) (19).

Efforts to streamline the chemical synthesisof oligosaccharides have focused on mini-mizing purification steps. One approach isto conduct the kind of one-pot glycosylationreactions described above, in which multipleglycosidic bonds are made without isolationor purification of intermediates (Figure 3c).There are three general strategies to achievethis end. First, the relative reactivities of theglycosyl donors can be varied by protectinggroup selection, such that the addition of a pro-moter triggers the most armed glycosyl donorfirst, and the most disarmed donor eventuallyengages in the final glycosylation reaction.Second, glycosyl donors can be preactivatedbefore exposure to a glycosyl acceptor, and theorder and timing of their addition determinethe reaction outcome. Third, orthogonalanomeric leaving groups can be selected thatare activated by different promoters. Impres-sive one-pot syntheses of biologically relevant

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−Figure 3In its most versatile form, the chemical synthesis of oligosaccharides depends upon glycosylation reactions that involve activation of aglycosyl donor. (a) Treatment with a promoter facilitates the departure of the anomeric leaving group (LG) such that a glycosidic bondcan be formed in a substitution reaction with a nucleophilic glycosyl acceptor. Depending on the electron-withdrawing potential of theprotecting groups (P) on the glycosyl donor, the reaction can proceed through an SN2-like mechanism (P = Ac, acetate) or an SN1-likemechanism in which the stereochemical outcome is controlled by the anomeric effect (P = Bn, benzyl). (b) Neighboring groupparticipation can dictate product stereochemistry. One-pot methods (c) and solid-phase synthesis (d ) are approaches that eliminatepurification steps and thereby facilitate glycan assembly. Abbreviation: ROH, an alcohol, including a hydroxyl group of a protectedsugar derivative.

oligosaccharides are becoming more common.Two notable examples include generation ofthe branched hexasaccharides GM1 and thetumor-associated carbohydrate antigen Globo-H (20). In the former, three components werejoined in a single reaction; while in the latter,four building blocks were linked (21).

Another strategy for oligosaccharide syn-thesis that circumvents the need for multi-ple purification steps is solid-supported syn-thesis (22). In this manifold, reactions canbe driven to completion by the addition ofan excess of one partner. In the typical con-figuration, a nucleophilic acceptor substrateis appended onto a solid support and ex-posed to an excess of activated donor in so-lution (Figure 3d ). Subsequent steps involvehydroxyl protecting group removal followedby glycosylation. The multiple sites of re-activity and branching found in oligosaccha-rides require monomers that possess orthogo-nal protecting groups, which can be masked andunmasked at appropriate stages of glycan con-struction. Thus, the solid-supported synthesisof oligosaccharides is complicated by the needfor diverse building blocks. Nevertheless, thepotential of solid-supported synthesis has con-tinued to spark advances, including methods toautomate the process. Automated synthesis nowcan be used to prepare even complex oligosac-charides (22), such as a branched β-glucan do-decasaccharide; blood group oligosaccharidesLewis x, Lewis y, and the Lewis x-Lewis ynonasaccharide; and tumor-associated carbo-hydrate antigens Gb-3 and Globo-H. Theselatter examples are compounds with multipletypes of glycoside linkages, and their success-ful synthesis demonstrates that glycosylation


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

reactions in solid-supported synthesis can oc-cur with excellent stereoselectivity.

There has yet to emerge a universal strat-egy to form glycosidic bonds with the requi-site regio- and stereocontrol. Suitable reactionconditions must be optimized for each glyco-sidic variation. Thus, current methods are fo-cused on developing sets of standard build-ing blocks that can be used to generate keybioactive oligosaccharides. To this end, manyof the targets assembled to date possess link-ages that can be formed reliably, such as α-mannosides, β-galactosides, and β-glucosides.It is estimated that approximately 500 orthog-onally protected monosaccharides would beneeded to synthesize the bioactive oligosaccha-rides found in mammals (23), although a re-cent analysis suggests that 36 building blockscould generate 75% of the known mammalianoligosaccharides (24). The need for glycans thatreflect the diversity of physiological systemsis driving efforts to develop methods to pre-pare all the relevant glycosidic bonds, includ-ing those that have been challenging (e.g., β-mannosides, sialic acid derivatives). Progresson this front has made accessible biologicallyimportant glycans, such as sulfated GAG se-quences and protein glycosylphosphatidylinos-itol anchors. Despite the rapid development ingenerating mammalian oligosaccharides, therehas been less emphasis on assembling glycansfound in microbes. Many of these contain non-canonical sugars (e.g., deoxysugars and fura-noses) that pose unique challenges. Methods toassemble these glycans are needed to elucidatetheir roles in microbes, to probe host-pathogeninteractions, and to investigate novel antimicro-bial strategies.

Engineering Enzymesfor Glycan Synthesis

Enzymatic and chemoenzymatic methods forglycan assembly complement those from chem-ical synthesis. These approaches harness thecomponents used by physiological systems togenerate glycans. Specifically, glycosyltrans-ferases transfer nucleotide-sugar donors onto

glycone or aglycone acceptors. The use of theseenzymes can facilitate the chemoenzymatic syn-thesis of glycans, and recent advances have in-creased the utility of this approach. Historically,nucleotide-sugar donors can only be generatedwith naturally occurring sugars, although ef-forts in enzyme engineering indicate that thisproblem can be overcome (25). Indeed, the util-ity of enzymes for generating oligosaccharideshas increased in the last decade (26, 27), owing,in part, to the ability to identify glycosyltrans-ferases from sequence data.

Bacterial glycosyltransferases and relatedbiosynthetic enzymes have proved especiallyuseful for glycan assembly. These enzymes andtheir variants can be produced and purifiedmore readily than their eukaryotic counter-parts, and they often act on a broad array ofsubstrates. In a powerful example, Chen andcoworkers (28) used three classes of bacterialsugar-processing enzymes (a sialic acid al-dolase, a cytidine 5′-monophosphate-sialic acidsynthetase, and a sialyltransferase) to producea library of 72 biotinylated sialosides in anarray format. By screening this array, detailedinformation about the binding preferencesof a key immunomodulatory protein, humanCD22, could be gleaned. Another examplein which the broad substrate specificity ofbacterial glycosyltransferases was exploited isin the production of 70 glycoforms of the nat-ural products calicheamicin and vancomycin(29). Because glycosylation can influencenatural product biological activity, specificity,and pharmacology, the ability to introducedifferent saccharide substituents is valuable.The enzymes can be engineered to increasetheir substrate tolerance even further (30).

Another chemoenzymatic approach tooligosaccharides is based upon glycosidases.Nature’s antipode to the glycosyltransferase isthe glycosidase, an enzyme that catalyzes thehydrolytic cleavage of glycosidic bonds. Re-placement of the active-site water nucleophilewith a glycosyl acceptor can result in trans-glycosylation (26, 27). Because glycosidases canreadily be produced, often are highly solubleand stable, and tend to be more promiscuous


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

than glycosyltransferases, they are attractive ascatalysts. Unfortunately, glycosidase-catalyzedtransglycosylation reactions suffer from lowyields and product hydrolysis because theproducts are themselves substrates. A majorbreakthrough in the field occurred with theinvention of nonhydrolyzing glycosidases, orglycosynthases.

Glycosynthases were developed by exploit-ing key features of glycosidase mechanisms.

Glycosynthase:engineered glycosidasecapable of catalyzingtransglycosylationreactions

Glycosidases come in two varieties, retainingand inverting (26, 27). In general, both pos-sess active sites in which catalytic carboxylicacid residues are proximal (Figure 4). Most gly-cosynthases are based on retaining glycosidases,which use a two-step, double-substitutionmechanism that proceeds through a covalentcarbohydrate-protein adduct (Figure 4b). Sub-stitution of the nucleophilic active site aspartateor glutamate with a small hydrophobic residue

OO

RHO

Enz

O O

Enz

O O

H

OHO

Enz

O O

Enz

O O

HO

HO

OHHO

Enz

O O

Enz

O O

H

– ROH

OHO

Enz

O O

Enz

O O

HO

HO

OHHO

Enz

O O

Enz

O O

H

ORH

– ROH

OHO

Enz

O O

CH 3

Enz

HO

RO

ORHO

Enz

O O

CH 3

Enz

H

F

– F–

a Inverting glycosidase mechanism

b Retaining glycosidase mechanism

c Glycosynthase mechanism

+ H2O

Figure 4Catalytic mechanisms for glycosidases and glycosynthases. (a) Inverting glycosidases use two catalyticcarboxylate residues positioned proximally. (b) Retaining glycosidases use a two-step, double-substitutionmechanism, with a covalent carbohydrate-enzyme adduct. (c) Substitution of the nucleophilic active-sitecarboxylate of retaining glycosidases with a nonpolar side chain affords glycosynthases capable oftransglycosylation. Abbreviations: Enz, enzyme; ROH, an alcohol, including a hydroxyl group of a protectedsugar derivative.


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

(i.e., alanine) (Figure 4c) renders the enzymeincapable of hydrolysis. These variants can cat-alyze the formation of glycosidic bonds be-tween an acceptor and an α-glycosyl fluoridedonor. In the engineered enzyme, the productis no longer susceptible to hydrolysis; there-fore, transglycosylation reactions can occur inhigh yields and with high levels of regio- andstereoselectivity.

Glycosynthases engineered to process al-ternative substrates can be generated by eitherrational mutagenesis or directed evolution.Because libraries of enzyme variants can bereadily prepared, the major roadblock in thediscovery of novel glycosynthase enzymes hasbeen the development of high-throughputscreens. The enzyme-catalyzed glycosylationreactions are not accompanied by the release ofa chromophore, so novel screens were needed.The approaches that have been devised fall intothree categories: a yeast three-hybrid chemicalcomplementation assay (31), an assay based onpH changes (32), and a fluorescence-activatingcell sorting (FACS) assay (33). In all threeapproaches, glycosynthase activity is evaluatedin whole cells; therefore, protein isolation isnot required.

Glycosynthases have been identified (26, 27)that act on a range of nucleophiles, includingglycone and aglycone acceptors. A notable fea-ture of glycosynthases is that they can produceoligosaccharides that are difficult to obtainusing chemical synthesis (e.g., β-mannosides)(34). In addition, disaccharide fluoride donorsand acceptors can be used in transglycosy-lation reactions, thereby enabling the rapidassembly of complex oligosaccharides (26).Glycosynthases also can be used to generateoligosaccharides that contain β-glucuronic acidor β-galacturonic acid residues, suggesting theycan be used for GAG assembly (26). Notably,retaining endoglycosidases have also been de-vised that catalyze the convergent assembly ofN-glycans. Specifically, these modified enzymespromote the reaction of oxazolines derivedfrom 2-deoxy N-acetyl glucosamine-containingsubstrates with asparagine-containing peptides(35). To complement the linkages that can be

formed by engineered retaining glycosidases,inverting glycosidases have been generatedthat afford α-linkages (36, 37). Thus, ad-vances in the identification and engineeringof both glycosyltransferases and glycosyn-thases are extending the range of accessibleglycans.

Glycoprotein andGlycopeptide Synthesis

In parallel with the development of new meth-ods for oligosaccharide assembly, there haveemerged new approaches for glycoconjugatesynthesis. The prevalence of glycosylated pro-teins and the benefits of access to single-proteinglycoforms have inspired the development ofmethods to generate N- and O-glycosylatedpeptides and proteins. Protein glycosylationcan influence the pharmacological propertiesof therapeutic proteins, including their serumhalf-lives, their ability to target specific cells ororgans, and their modes of clearance. Glyco-sylation can also exert an influence by playinga direct role in recognition, such that wholeglycosylated protein is more than the sum ofthe parts. A notable example of the latter in-volves P-selectin, a protein involved in the in-flammatory response, which binds to a highlyO-glycosylated protein (a mucin) bearing thetetrasaccharide sialyl Lewis x. The tightestcomplexes between P-selectin and its ligand areformed when specific tyrosine residues adjacentto a sialyl Lewis x motif are sulfated (38). Thus,the identity of the glycoconjugate as a wholeis important for recognition (39). Together, themode of P-selectin recognition and the require-ment for therapeutic glycoproteins with opti-mized properties underscore the need to obtaindefined glycoconjugates.

Several strategies have emerged that yielddefined glycoproteins and glycopeptides. Oneapproach is to employ engineered cell lines orrecombinant enzymes to obtain glycoproteins.For instance, complete heterogeneous gly-coproteins can be expressed, isolated, andtrimmed via glycosidases to bear individ-ual monosaccharide moieties (40). These


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

monosaccharides serve as starting points forembellishment by recombinant glycosyltrans-ferases or by transglycosylation using endogly-cosidases. The latter strategy is especially usefulfor the rapid assembly of complex glycoproteinsbecause a large glycan motif can be added in asingle step (as described above).

Synthetic chemists also have taken on thechallenge of glycopeptide and glycoproteinpreparation. In focusing on various complexglycopeptides and glycoproteins as targets forsynthesis, the Danishefsky research group (41)has pushed the limits of existing synthetic meth-ods. In pursuing their complex targets, includ-ing prostate-specific antigen, gp120 fragments,and erythropoietin, they have developed newstrategies for the assembly of multiple peptideprecursors.

The chemical synthesis of glycoproteins isfueled both by methods to construct complexglycopeptide fragments and the advent ofchemoselective ligation reactions to join them.Solid-phase peptide synthesis is generally lim-ited to glycopeptides <50 residues. Chemicalligation reactions, such as native chemical lig-ation (NCL), alleviate this limitation becausethey can be used to link peptide fragmentstogether (Figure 5a) (42). The NCL processinvolves the transthioesterification reactionof a C-terminal thioester with an N-terminalcysteine residue of a second peptide. Theresulting thioester intermediate subsequentlyundergoes an intramolecular transacylationreaction to produce a stable peptide bond.One valuable variation on NCL is expressedprotein ligation (EPL), in which the peptidecomponent bearing the C-terminal thioesteris produced using recombinant DNA methods(43). Although both NCL and EPL increasethe scope of glycopeptide synthesis, theyrequire a cysteine residue, a relatively rareamino acid, at the ligation junction.

The Staudinger ligation of peptidethioesters circumvents the need for cysteine atthe junction, as the two peptides couple whena C-terminal phosphinothioester undergoesreaction with an N-terminal azide. The utilityof the Staudinger ligation for glycopeptide

synthesis is under investigation (44). Otherapproaches have been described that capi-talize on removable or transient auxiliaries(45–47). One of these, sugar-assisted ligationis particularly useful in the construction ofN-linked glycans (Figure 5b) (48). Thus, themeans to construct larger glycopeptides andglycoproteins are available and can be usedto examine the influence of glycosylation onprotein function.

Chemical Glycobiology of Glycolipids

Glycolipids have been implicated in many criti-cal processes, but identifying their precise phys-iological roles has been difficult. Recent discov-eries have revealed that glycolipids can serveas critical immunomodulators. Natural killer T(NKT) cells are a class of T cells that play acentral role regulating the immune response,and NKT cells can recognize glycolipids dis-played by CD1d-positive antigen-presentingcells. Both endogenous and exogenous glycol-ipids can serve as CD1d ligands and therebyactivate NKT cells. An endogenous glycolipidis presumably necessary for positive selectionof NKT cells in the thymus, and NKT cellscan recognize exogenous lipopolysaccharidesfrom bacterial pathogens (49). The syntheticglycolipid antigen KRN7000 (Figure 6) andrelated compounds are illuminating the criticalfeatures of the glycolipid that result in NKTcell activation. This understanding can lead tothe development of new immunomodulators.Additionally, because glycolipid trafficking anddegradation are involved in several diseases,glycolipid analogs serve as probes and thera-peutic leads (50).

Chemoselective Reactionsto Modify Glycans

Glycoconjugates are critical tools in the ex-amination of glycan function. They can beimmobilized for affinity isolation of glycan-binding proteins, used to generate glycan ar-rays, or converted to natural or nonnaturalprobes. Such probes can be generated from


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

SR

O

PeptideH2N

HS

S

PeptideH2N

PeptideNH

HSO

OHO

OHO

OHO

Peptide SR Peptide

OHO

NHHS

X

H2N

Peptide S

Peptide

OHO

NH X

H2N

Peptide NH

HS

Peptide

OHO

NH X

Peptide NH

Peptide

OHO

NH X

a Expressed protein and native chemical ligation

b Sugar-assisted ligation

Desulfurization

Recombinant DNA methods

or SPPS Peptide

Peptide

Peptide

O

O O

O O O OO

O

Figure 5Ligation strategies for glycopeptide and glycoprotein synthesis. (a) In native chemical ligation (NCL), thepeptide components are obtained by solid-phase peptide synthesis (SPPS). For expressed protein ligation(EPL), recombinant DNA methods are used to produce a peptide or protein fragment with a C-terminalthioester. For both EPL and NCL, the thioester is captured by an N-terminal cysteine residue, and theincipient thioester conjugate rearranges to the amide. (b) In sugar-assisted ligation, a carbohydrate bearing athiol substituent serves in the same capacity as the Cys side chain.

OOHHO

HO

ONH OH

OH

O

HO

Figure 6The structure of glycolipid antigen KRN7000, which functions as animmunomodulator that leads to natural killer T cell activation.

chemoselective reactions (i.e., reactions thatoccur among select functional groups in thepresence of others) of natural and syntheticoligosaccharides (51). One of the most commonstrategies is to use the intrinsic reactivityof oligosaccharides, which contain an elec-trophile at the reducing end, most commonly,an aldehyde. This masked carbonyl group is


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

susceptible to nucleophilic addition, which hasbeen exploited for conjugate production. Nu-cleophiles, such as alkoxylamine, hydrazine, oracylhydrazine derivatives, can be employed toafford glycoconjugates containing oxime, hy-drazone, or hydrazide linkages, respectively(Figure 7a). These functional groups vary intheir stability and in whether the reducing-endsugar exists in the open or closed form. Thus,the mode of conjugation can be chosen for aspecific purpose.

An alternative strategy is to generate aglycan that bears a linker that possess a func-tional group that can undergo reaction with acoupling partner in the presence of hydroxyl,acetamide, carboxylate, and other common car-bohydrate functional groups. Perhaps the mostcommonly used linker, and that used by theConsortium for Functional Glycomics (CFG,http://www.functionalglycomics.org), is anaminopropyl group appended to the anomericposition. The resulting amine-bearingoligosaccharides can undergo reaction withseveral types of partners, including thosebearing N-hydroxysuccinimidyl esters, alde-hydes (followed by reductive amination), ordimethyl squarate (Figure 7b) (52–55). Otherchemoselective linkage strategies rely on theunique reactivity of thiol-containing saccha-rides, which can undergo conjugate addition

−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→Figure 7Chemoselective reactions used to generateglycoconjugates. (a) Glycans with free reducing endscan undergo reaction with aminooxy- andhydrazine-bearing linkers to form oxime andhydrazone linkages. (b-e) Different reactant sets forthe general reaction shown at the top of the table.(b) Glycans bearing amino groups can attackN-hydroxysuccinimidyl esters, aldehydes, ordimethyl squarate to generate adducts. (c) Glycansthat possess azide functional groups can engage inazide-alkyne cycloaddition [Cu(I)-catalyzed orstrain-promoted reactions] and Staudinger ligationreactions. Diels-Alder cycloadditions (d ) and olefinmetathesis reactions (e) are other examples ofchemoselective methods for glycan attachment.

OHO OH

aOR OH

HO NO R

NH OHO

HN

NH

RR

OHO O

R OHO O

R

NH2

N3

CH3O

PPh2 R

H

O

O

R

OO

R

O RN

O

CH3OHN

O O

R

R

S RPh2P

RGrubbs’ catalyst R

N

N

NH

NH

NH

NH

R

OR

HN

R

OO

NN

R

O

O

R

PPh2O

R

O

O

R R

b

c

d

e

H2N

H2N

O

O

O


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

CFG: Consortium forFunctional Glycomics

to maleimide-bearing glycans (56, 57) or formdisulfide-linked conjugates (58). Cycloadditionreactions, including Cu(I)-catalyzed azide-alkyne (Figure 7c) and Diels-Alder reactions(Figure 7d ), also have been used (59, 60). Al-though olefin metathesis depends upon a metalcarbene catalyst, remarkably, it has been shownto be compatible with carbohydrate function-ality (Figure 7e), including sulfate groups (22,61). The Staudinger ligation reactions of azideswith phosphinoesters or phosphinothioestersalso are useful chemoselective reactions, andthe phosphinoester version has been used indiverse contexts (Figure 7c) (62, 63). Someexamples of how the different aforementionedchemoselective reaction processes have beenexploited to investigate glycan function aredescribed in subsequent sections.

INTERROGATION OFGLYCAN RECOGNITION

Glycans are present both inside and outside ofcells. Within cells, glycosylation is critical forprotein trafficking, and more recently, it hasbeen found to influence gene expression (64).Glycans on the surface of pathogens can serveboth as a protective shield and as a means forrecognizing and entering target cells. Similarly,protein-glycoconjugate interactions are amajor line of communication between cellsand their environment. Lectins, proteins ofnonimmune origin that bind to specific glycanstructural motifs, typically use solvent-exposedbinding sites to interact with their targetoligosaccharide ligands (65). As a result, theybind weakly to single carbohydrate residuesand even oligosaccharides. Indeed, monovalentprotein-glycan binding dissociation constantsare often in the range of 10−4 to 10−3 M(66). These low affinities might suggest thatprotein-glycan complexes are not important,yet weak binding is ideal for mediating celladhesion. When cells interact, glycans onone cell surface can bind to multiple copiesof a lectin on another, thereby increasingthe apparent binding constant (functionalaffinity). The advantage of using low-affinity

interactions for cell-cell recognition is that,when each individual receptor-ligand interac-tion is weak, binding will be kinetically labile.In this way, only cells with the correct combina-tion of receptor-ligand pairs will interact stably.

The involvement of multivalent binding inmany protein-glycan interactions complicatesthe identification of the relevant endogenousligands. Standard receptor-ligand assays lackthe necessary sensitivity to monitor low-affinitybinding. Accordingly, many methods to assessprotein-glycan interactions depend upon mul-tivalent display of one or both binding part-ners. Such assays have higher sensitivity andcan have even higher specificity (67, 68). More-over, they can minimize the amount of ma-terial required, which is especially importantconsidering the challenges associated with theacquisition of glycans. Still, they are best usedto compare compounds because determiningthe true equilibrium constant for a multivalentinteraction is complicated. Indeed, many dis-tinct types of binding modes can contribute tothe strength of a multivalent interaction (69,70). Glycan arrays are a technology that in-vokes multivalent binding and allows many dif-ferent samples to be compared simultaneously.New tools for array fabrication and analysishave been advanced that depend on a combi-nation of analytical, biochemical, and syntheticmethods. The topic of glycan arrays has beenreviewed extensively (71–75), and our goal is tohighlight relevant contributions of chemistry totheir development.

Glycan Arrays

Glycan arrays have been widely embraced asplatforms suited to rapid screening of proteinbinding to carbohydrates. On the basis of prin-ciples developed for DNA and protein microar-rays, glycan arrays have emerged as tools to as-sess the specificities of lectins, antibodies, andother glycan-binding proteins. There are manymethods for fabricating glycan arrays, yet allhave the same overall features. Specifically, nat-ural or synthetic glycans are immobilized onto asurface through either covalent or noncovalent


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

attachment. The resulting glycan surfaces aretreated with whole cells, complex biologicalsamples such as sera, or isolated glycan-bindingproteins. Binding can be assessed using fluores-cence or another type of reporter.

The first challenge to creating glycan arrayswas to develop a method to spatially patternvarious oligosaccharides or glycoconjugates ona surface. To date, the methods implementedfor array fabrication fall into three general cat-egories: (a) immobilization by physical adsorp-tion, (b) immobilization via high-affinity, spe-cific noncovalent interactions, or (c) covalentcapture, in which complementary reactants aredisplayed on the glycan and the surface. Physi-cal adsorption, which exploits the ability of gly-cosylated proteins or glycolipids to adhere tothe surface, helped to establish the utility of gly-can arrays as a multivalent assay platform (76–78). Nonnatural glycolipids can be generatedfrom oligosaccharides that possess a free reduc-ing end (i.e., a masked aldehyde) using lipidsthat bear nucleophiles, such as amines (79) oralkoxylamines (80). In an innovative variationof the adsorption approach, fluorous lipid tags,which can be used both for synthesis and im-mobilization, have been employed (81). An al-ternative approach is to immobilize a glycocon-jugate through specific noncovalent complexes,such as biotin-streptavidin binding or DNA hy-bridization (82). In general, however, in mostarrays, the glycan is linked to the surface by co-valent bond formation.

A common means for glycan array construc-tion is to exploit the unique reactivity of theanomeric position. For example, oligosaccha-rides can be appended to a surface that presentsnucleophiles via reaction with the reducing end(83). Similarly, the reducing end can undergoreductive amination with 2-aminobenzaminederivatives in solution (84, 85), and these flu-orescent saccharides can be subsequently at-tached to the surface. In the latter approach,the fluorescent tag serves multiple purposes: Itprovides a means to purify heterogeneous poolsof natural glycans, it can react with an elec-trophilic (e.g., succinimidyl ester- or epoxide-functionalized) surface, and it provides a means

of quantifying the immobilization efficiency.Reductive amination reactions of oligosaccha-rides with the lysine side chains of proteinsalso have been exploited to yield multivalentglycoconjugates that were subsequently immo-bilized; the presentation of these conjugateson the surface can mimic that of glycosylatedproteins (86–88).

When glycans are generated by chemi-cal synthesis, tailored functional groups forimmobilization can be introduced. As men-tioned previously, the mostly widely used strat-egy is to introduce a linker bearing an aminegroup and exploit its nucleophilicity withN-hydroxysuccinimidyl ester-coated slides. Acomplementary approach is to build oligosac-charides directly onto the array surface (56).This general strategy provides not only a meansto construct known glycan structures, but alsothe opportunity to interrogate the selectivitiesof glycosyltransferases (89).

Many relevant protein-glycan interactionscan be uncovered using glycan arrays as exper-iments focused on Tn antigen illustrate. TheTn antigen is rarely expressed in normal tissuesbut is associated with several cancers. Glycanarrays were used to reveal that only a subsetof prostate tumors display the Tn antigen, afinding that may have therapeutic implications(90). Another instructive feature of the Tn anti-gen studies is that variations in the specificitiesdetermined by individual research groups weredifferent. This unexpected outcome highlightsthe variability that can arise from glycan arraydata and the need for standardization.

Despite the challenges, glycan arrays areproviding new insight into the protein-bindingspecificities in complex systems, including thoseinvolving highly anionic sulfated GAG se-quences. GAGs such as heparin, heparan sul-fate, and chondroitin sulfate are involved inprocesses ranging from development, angio-genesis, cancer metastasis, wound healing, andviral invasion (91). GAGs can be composed ofheterogeneous sequences, but the idea that spe-cific sequences are recognized selectively hasbeen controversial owing to a paucity of sup-porting data. To investigate this hypothesis,


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

GAG arrays have been assembled from theattachment of di- to hexasaccharides bearingamine- and aminooxy-terminated linkers toelectrophilic surfaces (52, 53, 92). The resultingarrays are bolstering the hypothesis that spe-cific GAG sequences have defined physiologicalfunctions (93–95).

With all the advances in array fabrication,the most significant barrier to the widespreadadoption of glycan arrays is the limited avail-ability of oligosaccharide structures. The CFGprovides arrays for the nonspecialist that focuson human and mammalian glycans. The ver-sion currently available to researchers (Version4.0) displays 442 mammalian glycans, whereasthe pathogen array presents 96 glycans derivedfrom seven pathogen species. Researchers needto continue to identify and generate a broadarray of glycans to extend further the utilityof the array platform. Additionally, althoughthe array surface is suited to multivalent inter-actions, it is unclear how the mode of glycandisplay influences protein recognition. Newtechnologies to address the role of presentationinclude the immobilization of multivalentglycosylated scaffolds, such as proteins andpeptides (87, 88) or polymer ligands (86).Additionally, fluidic arrays have been intro-duced that are designed to mimic the mobilityof glycoconjugates imbedded in a lipid bilayer(96). As glycan array technology continues toevolve, standard methods, from fabrication tointerpretation, will undoubtedly emerge.

Lectin Arrays

Lectin arrays provide a means to assess lectinbinding to individual glycoconjugates, patternsof cell-surface glycosylation, and pathogen-lectin interactions (73). They can provideimportant structural information about un-characterized glycans and serve as multivalentand sensitive monitors of protein-glycan inter-actions. Lectin arrays are typically fabricatedusing commercially available carbohydrate-binding proteins of defined specificity. Mahaland coworkers (97) have pioneered a two-colortechnique for the analysis of glycans from

mixtures, similar to that employed for DNAarrays. The ratiometric data obtained from apair of dye-swapped arrays afford reproducibledata. This method was used to support thehypothesis that the human immunodeficiencyvirus (HIV) co-opts the microvesicular exo-cytic mechanism to exit T cells. Lectin arraytechnology also has been used to elucidatedifferences in sialic acid expression betweennontumorigenic and adenocarcinoma cells (98).

Although lectin array technology is morenascent than that of glycan arrays, some ofthe challenges are shared. Notably, just as gly-can arrays are limited by the availability ofoligosaccharides, lectin arrays are restricted bythe recognition specificities of known lectins.Most commercially available lectins are isolatedfrom plants, but pathogenic organisms oftencontain unique glycan structures that are notrecognized by the available and characterizedlectins. As more carbohydrate-binding proteinsbecome available, the value of these arrays andtheir ability to distinguish between different celltypes will increase.

Generally, glycan and lectin arrays use fluo-rescence to detect protein-glycan interactions.Methods of introducing probes include con-jugation of fluorescent tags to cells or lectins,cell staining, and incubation with labeled anti-bodies. Modification-free techniques for arrayanalysis are also under development, and theseinclude evanescent-field fluorescence detection(99, 100), surface plasmon resonance (101), andfluorescence interference contrast microscopy(102).

PERTURBATION OFGLYCAN FUNCTION

A traditional approach to perturb protein func-tion is to delete a protein or proteins of inter-est within a cell or organism. The applicationof RNA interference or gene knockouts canprovide insight into the importance of a lectinor an enzyme involved in glycan biosynthesis(103). Drawing conclusions from single genedeletions of enzymes involved in the biosyn-thesis or processing of glycans, however, can


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

be difficult, as studies of CD22 illustrate. Micethat lack CD22 are hypersensitive to B cell anti-gen receptor stimulation, but those that lackthe glycosyltransferase (ST6Gal-I) that gener-ates the CD22 ligand have compromised B cellresponses to antigen (104). To determine thatCD22-ligand interactions suppress B cell ac-tivation required further experimentation. An-other complication of genetic experiments isthe masking of phenotypic changes becauseof compensation by other enzymes, which cancamouflage the function of the protein of in-terest. Alternatively, a single protein may havemultiple roles, but a null mutant lacks all ofthem. Perhaps most significantly, genetic meth-ods were devised to examine protein function;even though they can be applied to a par-ticular protein that binds or generates a gly-can, they do not report on the function ofthe glycan itself. Thus, although genetic meth-ods are powerful, complementary strategies areneeded. One alternative is to perturb glycanfunction with compounds that disrupt or al-ter specific protein-glycan interactions or theproduction of specific glycans. Such perturba-tions can shed light on the physiological pro-cesses mediated by a glycan and also providetherapeutic leads. In the following sections, weoutline recent advances in the use of syntheticmolecules to probe glycan function.

Perturbation of Protein-GlycanRecognition with Monovalent Ligands

The recognition that protein-glycan complexesare critical in physiological and therapeuticallyimportant processes, which include inflamma-tion, immune system function, cancer, andhost-pathogen interactions, has fueled effortsto generate inhibitors. The aforementionedfeatures of the proteins that bind glycans—their low-affinity and solvent-exposed bindingsites—render the generation of effective lig-ands a formidable challenge. Still, inhibitorswith the requisite attributes are emerging froman enhanced understanding of glycan-lectininteractions coupled with advances in high-throughput methods for identifying them.

Glycomimetic:a small moleculedesigned to mimic thefunction of acarbohydrate withimprovedpharmacologicalproperties

Oligosaccharides based on endogenous gly-cans are an obvious starting point (105), butconverting these polar molecules into potentinhibitors has been difficult. The aim is to de-vise molecules with improved affinity and se-lectivity, reduced polarity, and greater stabilitythan naturally occurring glycans. One strategyto address these issues is to apply molecular de-sign principles. Although the rationale used tooptimize a glycomimetic generally is tailoredto the specific lectin target, analysis of the suc-cessful design efforts to date reveals some com-mon strategies (105). First, either structure-function relationship data or the structure ofthe complex is used to identify glycan func-tional groups that are critical for binding.Second, nonessential polar functional groups(e.g., hydroxyl and acetamido groups) are re-moved to increase lipophilicity. Third, confor-mational control elements are introduced topreorganize the oligosaccharide to adopt theactive, bound conformation. Fourth, the ob-servation that many glycan-binding sites arelined with aromatic residues can be exploitedby introducing aromatic substituents at key po-sitions to enhance binding affinity. Some exam-ples that illustrate successful implementation ofthese design elements follow.

The development of galectin inhibitorshighlights the value of the aforementionedstrategies. Galectins are a class of glycan-binding proteins found in multicellular or-ganisms; humans possess 12 genes encod-ing galectin family members (106). Theirname comes from their propensity to bindβ-galactose-containing oligosaccharides, al-though individual galectins can exhibit distinctselectivities. These proteins are involved ina range of physiological processes, includingregulation of cell growth, differentiation andapoptosis, cell adhesion, chemoattraction, andcell migration (107, 108). They also are impli-cated in the inflammatory response and tumorprogression. Unlike most mammalian lectins,galectins are not membrane bound but ratherare produced in the cytosol and then secreted.Consistent with their ability to occupy two dif-ferent cellular locations, galectins appear to


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

have important intracellular and extracellularfunctions. Still, their functional roles have beendifficult to discern, and the relevance of glycanbinding is not always clear. Thus, inhibitors thatcould interact selectively with different mem-bers of the galectin family could serve as valu-able biological probes.

Several of the galectins, including galectin-3, form complexes in which the 4- and 6-hydroxyl groups of galactose form hydrogen

OCH3

HOOH

OH

OO

NHAcOOO

OH

OHOHO

HO

O

HOHO

OHHO

AcHN

HO2C

OCH3

HOOH

OH

S

HNO

HO

OHHO

O

CH3CH2O2C

Sialyl Lewis x

NH

HN

O

S

N

O

N

Cl

F

N

Cl

CO2H

OH

OCH3

HOOH

OH

OOOO

OH

OHOHO

HO–O2C

b

c

d

e

a

Figure 8Monovalent ligands for perturbation of protein-glycan recognition.Glycomimetics that present key functional groups in specific orientations havebeen designed. The tetrasaccharide sialyl Lewis x (a) binds to the selectins, andcompounds b–d have been designed to mimic critical attributes of theoligosaccharide. The naturally occurring oligosaccharide ligands are boxed, andimportant functional groups that have been incorporated into the glycomimeticare highlighted in red. Compound e binds to another member of the C-type(Ca2+-dependent) lectin family, DC-SIGN. Abbreviation: Ac, acetate.

bonds to the protein, whereas the remain-ing hydroxyl groups do not make direct con-tact (109). These nonessential ligand hydroxylgroups serve as points for modification, and itwas postulated that aromatic substituents at the3-position would enhance binding. Inhibitorsof this type possess dissociation constants thatare 1000-fold more potent (Kd < 50 nM) thanN-acetyllactosamine (110). They also show se-lectivity (∼100-fold) for galectin-3 over othergalectins. A galectin-3 ligand of this type re-vealed that glycan binding by this lectin plays arole in alternative macrophage activation (111).Alternative macrophage activation is linked toprocesses ranging from asthma to wound re-pair and fibrosis; therefore, these studies sug-gest that galectin inhibitors could have benefi-cial therapeutic effects.

Many efforts to devise glycomimetic in-hibitors have focused on the C-type lectin fam-ily, whose members require Ca2+ for binding.Three lectins from this group, E-, L-, and P-selectin, have served as a major testing groundfor glycomimetic design. The selectins havebeen targets because of their participation inrecruiting leukocytes to inflamed tissue andtheir putative roles in tumor cell migration.Each selectin can bind to the structurally re-lated tetrasaccharides sialyl Lewis x and sialylLewis a, and extensive structure-activity stud-ies with oligosaccharide derivatives establishedthe key features that contribute to binding. ForE-selectin complexation, critical attributes in-clude the carboxylic acid group, the three hy-droxyl groups of fucose, and the 4- and 6-hydroxyl groups of galactose. This motif wasused to guide the design of a glycomimetic,in which the relevant groups were presentedon a scaffold that is preorganized for binding(Figure 8b) (105). Ligands with even less re-semblance to the saccharide residues they weredesigned to mimic also were effective. For ex-ample, peptide motifs appended to fucose bindto P-selectin. These also exhibit selectivity forP- over E-selectin, which is consistent with theability of the former to recognize an epitope en-compassing a carbohydrate and a peptide back-bone (Figure 8c) (112). As other examples of


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

noncarbohydrate ligands, two small-moleculeinhibitors of P-selectin were generated froma quinoline salicylic acid scaffold (Figure 8d ).These compounds have progressed into clinicaltrials for rheumatoid arthritis and atherothrom-botic vascular events (113).

To date, most glycomimetic designstrategies have focused on individual protein-saccharide complexes. In contrast, pep-tidomimetics often model structural elements(e.g., a β-turn) known to be critical for protein-protein contacts. There are common featuresof the C-type lectin complexes that might beexploited in inhibitor design. Many C-typelectin complexes use adjacent hydroxyl groupson fucose to coordinate the protein-boundcalcium ion, which suggests that scaffolds thatpossess key Ca2+-coordinating groups can bemodified to enhance affinity or specificity. Onesuch strategy has been described that employsfocused libraries of glycomimetics usingshikimic acid as a building block (114). Thesehave yielded inhibitors of the prototypicallectin mannose-binding protein A. The identi-fication of other approaches that can be appliedbroadly to other lectin classes could acceleratethe pace of glycomimetic generation.

An alternative to the design approach isto identify small-molecule inhibitors of lectinsthrough screening. This strategy could be valu-able if cell-permeable ligands could be found;to date, however, a limited number of suchcompounds have been described. Studies ofthe selectins have yielded some positive results(105), as have investigations focused on the C-type lectin DC-SIGN. DC-SIGN facilitatesseveral host-pathogen interactions, includingdissemination of HIV (115). Inhibitors ofDC-SIGN were identified from a 35,000-compound small-molecule library using a high-throughput fluorescence competition assay(116). Seven compounds were identified thatare ≥100-fold more potent than N-acetylman-nosamine for DC-SIGN (Figure 8e). Noneof the small molecules that bind the se-lectins or DC-SIGN resembles carbohy-drates, an observation that provides impe-tus to use high-throughput screens to search

for effective inhibitors of other carbohydrate-binding proteins.

Perturbation of Protein-GlycanRecognition with Multivalent Ligands

An alternative strategy to overcome low-affinityprotein-glycan interactions is to employ multi-valent ligands. This approach can be especiallyeffective for blocking protein-glycan engage-ment at the cell surface. Naturally occurring,multivalent glycan displays are widespread;representatives include glycosylated proteins,the glycan coats of bacteria, viruses, otherpathogens, and the surfaces of mammaliancells. Many carbohydrate-binding proteins areoligomeric and therefore are present in multi-ple copies on the cell surface. In this way, bothcell-surface glycans and lectins are poised toengage in multivalent binding.

Multivalent carbohydrate derivatives can ex-ploit unique modes of recognition not availableto their monovalent counterparts. Many lectinscontain more than one saccharide-binding siteor can oligomerize to form larger structureswith multiple binding sites. Multivalent ligandsthat can span the distance between bindingsites have an advantage over their monovalentcounterparts. This chelation mechanism isadvantageous because the translational entropycost is paid with the first receptor-ligand contact(70, 117). Nevertheless, the apparent affinityof a multivalent interaction often is less thanmight be expected, presumably because of theconformational entropy restrictions incurredby multipoint binding. Multivalent ligands alsocan exhibit functional affinity enhancementsby occupying secondary binding sites. Alterna-tively, glycan-binding proteins may cluster in amembrane microdomain either in response toa multivalent ligand or in response to cellularsignals. Although multivalent ligands can bepotent inhibitors, their ability to cluster glycan-binding receptors allows them also to serve asactivators of signaling pathways (69). Thus,depending on their binding modes, multivalentligands can exhibit a wide range of differentactivities.


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

Chemical synthesis can provide architec-turally diverse multivalent ligands, includinglow-molecular-weight displays, dendrimers,polymers, liposomes, and proteins (69, 118).This diversity can be used to optimize a syn-thetic ligand for a given application. For ex-ample, unlike naturally occurring, multivalentglycan ligands, the valency of a synthetic lig-and can be altered systematically by varyingthe length or size of the scaffold. Polymersof defined lengths or dendrimers of differentgenerations will possess different valencies andtherefore differing activities. Evaluating the im-pact of these changes on the biological re-sponse can illuminate the mechanisms under-lying the function of natural protein-glycaninteractions and lead to highly efficaciousinhibitors.

Potent multivalent inhibitors for severalmedically relevant protein-glycan interactionshave been identified. One target that has beenexplored is influenza virus hemagglutinin, anda number of multivalent sialic acid derivativeshave been generated that block the interac-tion of the virus with cells (119). Other host-pathogen interactions also can be inhibitedwith multivalent ligands, and representative ex-amples include compounds that prevent Pseu-domonas aeruginosa adhesion or the binding ofuropathogenic Escherichia coli (120). Althoughthe influence of scaffold structure is just begin-ning to be explored (121), the inhibitors of theAB5 bacterial toxin family highlight the benefitsof multivalent ligand design. This toxin familyis characterized by one active component (A)and a pentamer of subunits (B) that bind car-bohydrates displayed on cellular surfaces. TheAB5 toxins are responsible for diseases rang-ing from travelers’ diarrhea (heat-labile en-terotoxin), to acute kidney failure in children(Shiga-like toxins), to fatal cholera (Choleratoxin). Efforts from several groups have high-lighted the importance of distance betweenbinding motifs (118), a parameter that appearsto be at least as important as the identity ofthe ligands themselves. For example, a penta-cyclen core was used to display five galactoseresidues that were tethered using a range of

linker lengths (122). The activity of the mul-tivalent ligands against heat-labile enterotoxindepended on the linker. The most potent ligandpossessed the longest linker and was 105-foldmore active than the corresponding monova-lent galactose derivative. Dynamic light scatter-ing experiments indicated a 1:1 protein-ligandcomplex, suggesting the efficacy of the ligandis the result of the chelate effect. In anotherexample, Bundle and coworkers (123) used glu-cose as a core structure to display two trisaccha-rides per glucose oxygen on long spacer arms(Figure 9a). This STARFISH ligand was de-signed to occupy a Shiga-like toxin throughboth the primary binding site and a subsite. Itwas a highly effective inhibitor (IC50 0.24 nM).Unexpectedly, however, X-ray crystallographicanalysis revealed that the designed multivalentligand did not bind a single pentamer but ratherit dimerized two copies by occupying all fiveB subunits. Thus, for both AB5 toxin ligands,the spacing between the binding elements wascritical for activity, even though their modes ofmultivalent binding differ.

Most applications of multivalency in gly-cobiology involve the use of multivalent lig-ands as inhibitors, but multivalency also canbe used to activate particular cellular processes.An example involves blocking the action ofL-selectin, which mediates leukocyte migra-tion and recruitment from the blood to lym-phatic tissues and sites of inflammation (124).The natural ligands for L-selectin are mucinsthat present a multivalent display of sialylLewis x derivatives. Experiments using mucinmimics highlight the importance of multiva-lency for L-selectin recognition (Figure 9b).In this study, synthetic multivalent ligands weregenerated using the ring-opening metathesispolymerization (ROMP). This polymerizationis especially valuable for multivalent ligand syn-thesis because the length, and therefore valencyof the ligands, can be controlled. Interestingly,the ROMP-derived polymers not only bindL-selectin, but also promote its proteolytic re-lease, or shedding, from the cell surface (125).These results suggest that clustering L-selectinmay signal for its cleavage.


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

Multivalent displays of oligosaccharides alsoare being tested as vaccines against bacteria,parasites, and cancers (7, 9, 126, 127). Althougholigosaccharides typically do not elicit robustimmune responses, they are effective when ap-pended to an immunogenic carrier protein. Toboost immunity further, novel multivalent con-jugates are being developed. An innovative ex-ample is a conjugate that simultaneously dis-plays multiple groups: a B cell epitope, a Thelper epitope, and a toll receptor ligand (128).These groups can synergistically augment im-mune responses by recruiting different aspectsof the innate and adaptive immune responses.

Multivalent ligands also can be used torecruit lectins to signaling complexes. Suchassemblies are useful in controlling cellular re-sponses because some glycan-binding proteinsenhance signaling and others diminish it. Al-though vaccine designs are necessarily focusedon augmenting the immune response, com-pounds that suppress autoimmune responsesalso are needed. The aforementioned CD22,which dampens immune activation, is a sialicacid–binding lectin from the Siglec family. ACD22 ligand [i.e., N-acetylneuraminic acid-α(2,6)-galactose-β(1,4)-glucose] was attachedto a multivalent antigen, such that the result-ing polymer engaged both CD22 and the B cellantigen receptor (Figure 9b) (129). This sialy-lated antigen inhibited B cell activation. Theseresults identify a mechanism by which antigenglycosylation can suppress immune activation.Moreover, they highlight how multivalent lig-ands can be used to perturb the assembly ofglycan-binding proteins on the cell surface.

−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→Figure 9Selected examples of multivalent ligands that can beused to perturb protein-glycan recognition. (a) Thepentameric STARFISH core has been used to devisepotent inhibitors of the AB5 toxin family.(b) Polymers are useful scaffolds for producingmultivalent ligands. The structures depicted weregenerated using the ring-opening metathesispolymerization (ROMP), which differs from mostpolymerization reactions in that it can afford definedligands. Abbreviation: Ac, acetate.

A new but related aspect of multivalentprotein-glycan interactions involves exploitingnoncovalent interactions to create functionalsupramolecular protein-glycan assemblies. Twodifferent general strategies have emerged; both

OO

O

O OH

HOO

HOOH

OCH3

OH

OOHHO

HOHO

OO

O

O OH

HOO

HOOH

OCH3

OH

OOHHO

HOHO

O

O

NH O

NH

HN

O

NH

O

O

O

ONH

HN

OSR =

OO

OO

OO

R

RR

R

R

STARFISH

ROMP

Ph

HN

O

R

R'n

O

6

L-selectin ligand

R =

OCH3

HOOHOH

OO

OHOOO

OH

OHOHO

HO

O

HOHO

OHHO

AcHN

R' = H

CD22/B cell antigen receptor ligand

R =

OO

HOHO

OHHO

AcHN

R' =

O

OH

OOH

HO OOH

HOOH

O

a

b

and NO2

O2N

–O2C

HN

CO2H

(CH2)8

–O2C


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

Ligand Template

Target

Supramolecularcomplex

Figure 10Polymeric ligands possessing a preorganized structure can form supramolecularcomplexes that display increased functional binding affinity or increased serumhalf-life.

rely on bifunctional ligands. In the first, poly-meric ligands that display multiple copies oftwo distinct recognition elements, such as thesialylated antigens described above, promotemacromolecular protein assemblies. Multiva-lent assemblies of this type were used to ad-dress the disappointing in vivo activity of theaforementioned pentavalent inhibitors of Shigatoxin. Polymeric ligands that possess a preor-ganized architecture promote the formation ofcomplexes of Shiga toxin 1 and an endogenouscirculating pentavalent receptor human serumamyloid P (Figure 10) (130). When tested inmice, a stable macromolecular complex wasgenerated with a prolonged half-life in circula-tion. Most importantly, the animals were pro-tected from the toxin.

A second approach is to employ compoundsthat are not polymeric yet possess two dif-ferent epitopes (54, 131). A bifunctional lig-and was used to selectively kill tumor cellsvia recruitment of a glycan-binding antibody,which subsequently led to complement activa-tion. The bifunctional ligand was designed suchthat, upon binding to a cell-surface receptor as-sociated with cancer, it could present the anti-genic epitope galactosyl-α(1,3)-galactose (α-Gal), which would allow cells to be recognizedas foreign by the naturally occurring anti-α-Galantibody (54). The α-Gal epitope is not foundin humans (132), but human serum contains asignificant level of anti-α-Gal antibody (about3% of circulating antibody). Cells displayingα-Gal epitopes are subject to complement-mediated cell killing. Importantly, the recruit-ment of anti-α-Gal to the cell surface requiresa multivalent presentation of glycan residues.The final conjugate consists of an α-Gal epi-tope linked to an integrin αvβ3-binding ligand,as this integrin is upregulated on tumor cells.When cells are treated with the bifunctional lig-and, complement-mediated lysis occurs. Cellswith low levels of the integrin receptor werespared, whereas tumor cells displaying high lev-els of integrin were killed selectively (54), andthis selectivity underscores the advantage of us-ing multivalent interactions for cell targeting.Thus, ligands that promote macromolecular as-semblies of carbohydrate binding-proteins canco-opt the function of glycans and their recep-tors for new purposes.

Perturbation of Glycan Assembly

An emerging strategy to understand glycanfunction is to block selected steps in glycocon-jugate assembly or disassembly. To this end, ef-forts have been launched to develop inhibitorsof specific glycosyltransferases, glycosidases,and other sugar-processing enzymes. Inhibitorscan illuminate both the biological roles of gly-cans and the mechanisms behind their turnoverand assembly. Indeed, compounds that selec-tively target major pathways (including the pro-duction of N-glycoproteins and O-glycosylated


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

mucins, as well as the attachment of O-GlcNAcresidues) could be used to probe the roles ofdifferent glycan classes. Moreover, inhibitorsof the production of specific glycans in tu-mors or pathogens could serve as therapeuticleads.

Nature has provided some design strate-gies for generating compounds that inhibit N-glycosylation in the form of natural products.For example, tunicamycin (Figure 11a) blocksa crucial transphosphorylation reaction be-tween UDP-GlcNAc and dolichol-phosphatethat generates dolichol-PP-GlcNAc. This stepis required for the synthesis of N-glycoproteins,and tunicamycin has therefore been used to illu-minate the consequences of deficiencies in N-glycan production (133). Another strategy toinhibit N-glycosylation is to block the activityof the oligosaccharyltransferase complex, andthis mode of inhibition is exhibited by a cyclicpeptide that adopts the conformation of thepeptide acceptor (134). Cell-permeable com-pounds that block N-glycosylation selectivelyin either bacteria or eukaryotes would be usefulprobes (135).

Inhibitors of glycosyltransferases involvedin O-glycan biosynthesis also have been sought(136). Given that O-glycans are involved inprocesses from pathogen binding to cancer,inhibitors will be useful for investigating thephysiological and pathophysiological roles ofthis important class of glycans. For these tar-gets, there are no obvious starting points be-cause natural product inhibitors have not beenidentified, and the peptide sequence require-ments for glycosylation have not been fully de-lineated. Thus, high-throughput screens wereemployed. One innovative assay exploits the ob-servation that the rate of proteolysis of a glyco-sylated peptide is diminished relative to that ofits unmodified counterpart (136). By append-ing N- and C-terminal fluorophores that canengage in Forster resonance energy transfer,a peptide’s susceptibility to proteolytic cleav-age can be assessed. When a glycosyltransferaseis present, cleavage is decreased, whereas thepresence of a glycosyltransferase inhibitor re-stores rapid proteolysis. This assay format has

O CO2H

HN

H2NNH

HOOH

O CO2CH2CH3OAcHN

d

Zanamivir (Relenza®)

e

Oseltamivir (Tamiflu®)

ON

OHHO

NH

O

OO

OH

OHHO

HN O

O

OOH

HOHO

AcHN

a

Tunicamycin

b

Deoxynojirimycin

N

OHHHO

HO

HOc

Castanospermine

NHHOHO

OH

OH

HOAcHN

(CH2)9

H3N+

H2PO4–

Figure 11Natural products (compounds 11a–11c) have inspired the generation of drugs(compounds d and e) that act as transition-state analogs and thereby inhibitinfluenza virus neuraminidase. Inhibitors of this type can be used to perturbglycan assembly.

UDP-GlcNAc:uridine5′-diphosphate-N-acetylglucosamine

been employed to identify compounds that tar-get O-GlcNAc transferase (OGT) (Figure 12).OGT is an essential enzyme that regulates sig-naling through its ability to mediate intracel-lular O-glycosylation. Compounds that blockthis enzyme can serve as valuable probes thatcomplement those known for the correspond-ing O-GlcNAc residue hydrolyzing glycosi-dase. These studies illustrate the value of high-throughput screens to identify inhibitors, anda number of strategies have been implementedincluding those that rely on glycan arrays (72,137), activity assays (138), and binding assays(139).

The enzymes responsible for glycan assem-bly or modification can be critical for microbesand therefore represent attractive targets. Onesuch target is the influenza viral coat proteinneuraminidase, whose function is to catalyze


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

FNH2

O O

S

HO2C

NH

O

S

Figure 12Compounds identified in a high-throughput screenas inhibitors of the essential glycosyltransferaseO-GlcNAc transferase (OGT).

the cleavage of sialic acids on the host cellsurface to enable viral infection. The design ofthese compounds was inspired by natural prod-uct glycosidase inhibitors, such as deoxyno-jirimycin (Figure 11b) and castanospermine(Figure 11c). These compounds are aminesthat are protonated at neutral pH, and theyare thought to mimic an oxocarbenium tran-sition state (140). Design strategies inspiredby transition state models have led to valuabledrugs; these include zanamivir (marketed asRelenza®) (Figure 11d ) and oseltamivir (mar-keted as Tamiflu®) (Figure 11e) for the treat-ment of influenza (141). These compounds,which block viral replication and infection ofnew host cells, underscore the value of target-ing enzymes that operate on glycans. More-over, this strategy has been applied to affordinhibitors of other glycosyltransferases or gly-cosidases (142).

Carbohydrate-modifying enzymes criticalfor cell wall biosynthesis in microbes alsohave been the objects of inhibition studies.Several investigations in this area have em-ployed ligand displacement assays that relyupon fluorescence polarization (FP). FP as-says serve as a general method to study en-zymes that use nucleotide sugar substrates,which are prevalent yet poorly understood. FPassays have been used to identify inhibitorsof two different enzymes essential for cell

wall biosynthesis: UDP-galactopyranose mu-tase (UGM), which is found in mycobacteriaand catalyzes the interconversion of the iso-meric compounds UDP-galactopyranose andUDP-galactofuranose, and the glycosyltrans-ferase MurG, which is a bacterial glucosamino-transferase involved in the biosynthesis ofthe crucial bacterial cell wall componentpeptidoglycan.

In the case of MurG, a fluorescent probewas designed based on the structure of thedonor, UDP-GlcNAc, complexed to the en-zyme (Figure 13) (139). A library of 64,000compounds was screened, and subsequentanalysis revealed many of the most potentbinders possessed a common scaffold: a 1,3-disubstituted heterocyclic core. Because gly-cosaminotransferases are present in all organ-isms, probes or lead compounds must bind thebacterial enzyme selectively. Importantly, theinhibitors are selective over other nucleotide-sugar-processing enzymes (143).

In another example, a UDP-fluoresceinprobe was used in a 16,000-compound, high-throughput inhibitor screen of the UGM fromMycobacterium tuberculosis (144). The activecompounds identified share structural featureswith those found to block MurG, includinga five-membered thiazolidinone heterocycle,substituents at the 1- and 3-positions, and atleast one aromatic group that might mimic theuracil (Figure 13). This class of thiazolidinonederivatives is subject to reaction with nucle-ophiles in a biological milieu (144), which lim-its their utility as biological probes or thera-peutic leads. Still, their identification in twoindependent screens suggests that the shapeof these compounds is well suited for target-ing nucleotide-sugar-utilizing enzymes. Fur-ther design and optimization based on thishypothesis resulted in the identification ofcompounds that block mycobacterial growth(Figure 13) (145). These studies underscorethat high-throughput screens can be used astools to aid in identifying new therapeutic tar-gets and, more generally, as probes of glycanbiosynthesis.


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

UDPO

OOH

AcHN

UDPO

OOH

HOAcHN

HO

UDPO

OOH

HN

O

UDPONH(CH2)6

S

NHO

HO

O

COO–

NHO

O

O

HO

–OOC

HN

I

N

S

Cl

Cl

N

S

O

S Cl

NN

SS

O

Nucleotide-sugarsubstrate

FP probe InhibitorEnzymetarget

MurG

UGM

HOHO HO

HO

CO2H

CO2H

Figure 13Fluorescence polarization (FP) has been used to identify inhibitors for sugar-processing enzymes with essential roles in bacterial cellwall biosynthesis. These enzymes include the glycosyltransferase MurG and the isomerase UDP-galactopyranose mutase (UGM).

Exploiting Alternative Substratesin Glycan Biosynthesis

Modified glycans can be used to investigate gly-can recognition, biosynthesis, cellular or or-ganismal localization, and turnover. For exam-ple, primer saccharides can be introduced thatcompete with endogenous substrates for theglycosyltransferases (146). These compoundsserve as decoys by preventing the produc-tion of physiological glycoconjugates. This ap-proach can block the generation of specific cell-surface glycoconjugates, but the consequencesof harboring the resulting glycan chains withincells have not been elucidated fully. Glycanbiosynthetic pathways also can be co-opted toincorporate monosaccharide analogs, therebygenerating modified glycoconjugates in cellsor organisms. The feasibility of this strat-egy was demonstrated by exposing cells toN-propanoylmannosamine, an analog of N-acetylmannosamine, a key intermediate in thebiosynthesis of sialylated glycans. With thistreatment, the cell-surface glycans generated

bear sialic acid residues with N-propionamidegroups (147). This general strategy has subse-quently been exploited for diverse applications.

The incorporation of a modified buildingblock can interfere with subsequent glycosyla-tion reactions, thereby blocking the productionof specific glycan structures. In this way, glycanfunctional roles can be interrogated. Forexample, when cells are treated with N-butanoylmannosamine, they display truncatedpolysialic acid chains and provide a means toassess the influence of the length of polysialicacid on neuronal plasticity (148). Anotherexample of altering glycan production employs2-deoxygalactose, which precludes the gener-ation of the fucosyl-α(1,2)-galactose epitope.Investigations using this deoxysugar revealedthat fucosylation prevents the proteolyticdegradation of synapsin, a critical regulator ofneuronal function (149). The incorporationof modified saccharide residues also has beenused to investigate the requirements for glycanrecognition. For example, the use of metabolicengineering to generate N-glycosyl-substituted


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

sialylglycoconjugates disrupts the interactionof neural glycoproteins with myelin-associatedglycoprotein (150). N-Acetylmannosaminederivatives also have been used to decoratetumor cell surfaces with immunogenic glycans(151). Thus, altering glycan biosynthetic path-ways can provide insight into the functionalroles of specific carbohydrate epitopes.

Oligosaccharide biosynthesis also canbe co-opted to introduce unique functionalgroups into cellular glycans. The Bertozzigroup (152) has investigated different aspectsof glycobiology using metabolic labeling com-bined with chemoselective functionalizationreactions. In one example, they showed thatN-azidoacetylglucosamine (GlcNAz) couldbe incorporated at sites subject to O-GlcNAcmodification. Azides are valuable handles be-cause they are absent from biological systemsand are relatively unreactive toward most bio-logical function groups. As stated above, theycan undergo transformations in the presenceof other nonphysiological function groups.Specifically, an azide-bearing compound canbe coupled using a Staudinger ligation reactionwith a phosphinoester bearing a reporter(e.g., biotin) to generate a conjugate. Thisreaction scheme was utilized to profile proteinsmodified with O-GlcNAc. Cells were treatedwith GlcNAz to afford azide-substitutedglycoproteins that were subsequently modifiedfor isolation and characterization (153, 154).A similar strategy has been used with N-azidoacetylgalactosamine (GalNAz) to detectmucins. Metabolic labeling also can be usedto introduce photoaffinity labels, such thatcross-linking to glycan-binding proteins canbe carried out. For example, aryl-azide (155)or diazirine (156) moieties can be incorporatedinto sialic acid-bearing glycans at the C9 orC5 positions of sialic acid, respectively. Whenthese photoactivable groups are displayed oncell-surface glycans, they can be used to identifyglycan-binding proteins by covalent trapping.The ability to modify glycans by metabolicincorporation also can be used to introducegroups to alter cell adhesion. The display ofN-thioglycolylneuraminic acid in cell-surface

glycans causes self-assembly into large clus-tered cell aggregates, which may prove valuablefor tissue engineering purposes (157).

Illuminating Glycan Biosynthesis

Our understanding of glycans would beadvanced by strategies to visualize and tracktheir cellular and organismal distribution indevelopment or disease. The biosynthesis ofglycans containing specific residues can bevisualized by the incorporation of functionalgroups that can be used to append reportergroups. Both the Staudinger ligation and theHuisgen 1,3-dipolar cycloaddition reaction ofazides with alkynes (often referred to as clickchemistry) have been employed in biologicalenvironments (Figure 7c) (51). A major issuein these transformations is the rate of reaction,which determines the labeling time and sensi-tivity. Attempts to increase reaction efficiencyhave focused on the azide-alkyne 1,3-dipolarcycloaddition. Although this reaction can becatalyzed by copper(I), these conditions aredeleterious to cells. Using a more reactive,strained difluorinated cyclooctyne partnercircumvents the requirement for copper catal-ysis (Figure 14a). A variety of sugar buildingblocks bearing azido groups can be incorpo-rated, including derivatives corresponding tosialic acid, N-acetylgalactosamine (GalNAc),GlcNAc, and fucose (152). The azide moietycan then undergo reaction with an alkyne bear-ing a fluorophore or other reporter group. Theadvantages of this labeling strategy are illus-trated by its use in visualizing glycans in wholeanimals. Specifically, zebrafish embryos wereincubated with GalNAz to generate glycansthat bear azide groups. The GalNAz residuescould be visualized by their ability to undergoa cycloaddition reaction with fluorophore-labeled cyclooctynes. By using two differentfluorescent tags, changes in O-glycosylationduring the course of zebrafish developmentcould be observed. An alternative approach forglycan tagging in cells takes advantage of mildoxidation, which occurs at endogenous sialicacid residues, to afford an aldehyde at the C7


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

Reporter

Cell

Glycosyltransferase

N3

N3

N3

FF

NN

NF

F

a

OHO

O

OHO

NO

O

b

OOHHO

HONH

O

N3

OH

OHO

OH

OH

HOAcHN

CO2H

AcHN AcHN

NaIO4

CO2H

NH2

CO2H

Figure 14Strategies for visualizing glycans can be achieved by (a) metabolic incorporation of a sugar bearing a latentreactive group (e.g., azide) or (b) chemical modification of endogenous glycans to introduce a chemicalhandle that can be functionalized with fluorophores or other reporter molecules. Abbreviation: Ac, acetate.


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

position (158). Aniline-catalyzed oxime ligationwith an aminooxy-modified molecular tag pro-vides a chemical handle that can be used for vi-sualization or tracking of glycans (Figure 14b).This method is complementary to metaboliclabeling; not all systems will tolerate the mildoxidation step, but the labeling can be done on afaster timescale. Together, these results empha-size the advances that have been made in strate-gies to observe glycan production and turnover.

CONCLUSION

Emil Fischer, a pioneer in recognizing thecritical role of chemistry in understanding

carbohydrates, stated in his Nobel Lecture(159), “. . . the chemical enigma of Life willnot be solved until organic chemistry has mas-tered another, even more difficult subject, theproteins, in the same way as it has mas-tered the carbohydrates.” Though one canonly admire Fischer, today his statement ap-pears ironic, especially given the formidablechallenges of elucidating glycan function thatremain. These challenges are being met byan assemblage of interdisciplinary approaches.Research in chemical glycobiology is driv-ing discovery by offering new approaches andtools to explore and exploit the functions ofglycans.

SUMMARY POINTS

1. Glycans are involved in myriad specific molecular and cellular events from developmentto disease.

2. The field of glycobiology demands the development of new approaches to elucidateglycan function. Chemical biology approaches provide critical tools and strategies thatcan be used to probe or perturb glycan function.

3. The synthesis of glycans and the development of chemoselective reactions have greatlyexpanded the repertoire of oligosaccharides and glycoconjugates available.

4. Glycan and lectin arrays have emerged as valuable platforms to interrogate protein-glycaninteractions.

5. The biosynthesis of glycans can be exploited to introduce new functionality that can beused for analysis or imaging.

6. Inhibitors, both rationally designed glycomimetics and small molecules, are promisingnew tools with which to investigate and interfere with the biological roles of glycans.

7. Synthetic multivalent ligands are valuable tools for investigating the low-affinity inter-actions characteristic of glycan-binding events.

FUTURE ISSUES

1. The continued development of new strategies and approaches in both chemical andchemoenzymatic syntheses will provide increased access to complex glycans. Methodsare needed to assemble less explored glycans, including sulfated oligosaccharides and theunique glycans found in microbes and other pathogens.

2. Glycans lack functional group handles that can be used to install fluorophores or otherreporters to directly assess binding or enzyme-catalyzed modification reactions. Thus,new strategies for devising high throughput are needed to identify and engineer enzymesfor the synthesis of glycans and to promote the discovery of probes of glycan function.


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

3. Methods for monitoring glycan degradation are needed, as this process is critical forhuman health and has ramifications for harvesting energy from plant cell walls.

4. Although great strides have been made in the development of small-molecule ligands forglycan-binding proteins, inhibitors exist for only a handful of lectins. Selective inhibitorsfor a wide range of lectins are needed.

5. Small molecules that interfere with glycan processing are valuable, as the drugs that blockinfluenza virus neuraminidase illustrate. Recent studies indicate that small-molecule in-hibitors of key enzymes in glycan biosynthesis can be found, but more probes couldexpedite the elucidation of glycan function.

6. Great strides are being made in devising glycoconjugates that can elicit or inhibit immuneresponses. Promising glycolipids and synthetic glycoconjugates are being designed asvaccines, adjuvants, and immunomodulators.

7. With an increased understanding of how glycans function, chemistry can provide newtools to co-opt these functions for new purposes. From the metabolic incorporation ofunique functional groups to the use of ligands that recruit anticarbohydrate antibodiesto kill tumor cells, glycan function can be used to elicit new and valuable biologicalresponses in cells and organisms.

DISCLOSURE STATEMENT

The authors are not aware of any affiliations, memberships, funding, or financial holdings thatmight be perceived as affecting the objectivity of this review.

ACKNOWLEDGMENTS

We apologize to authors whose contributions were omitted from this review owing to limitationsof space and limitations in the number of references. This research was supported by the NationalInstitutes of Health (GM49974, GM55984, AI063596, and AI055258) to L.L.K. R.A.S. acknowl-edges the American Chemical Society Division of Medicinal Chemistry for a fellowship. We thankC.D. Brown, S.L. Mangold, L. Li, and M.R. Levengood for their comments on the manuscript.

LITERATURE CITED

1. Apweiler R, Hermjakob H, Sharon N. 1999. On the frequency of protein glycosylation, as deduced fromanalysis of the SWISS-PROT database. Biochim. Biophys. Acta 1473:4–8

2. Coutinho PM, Deleury E, Davies GJ, Henrissat B. 2003. An evolving hierarchical family classificationfor glycosyltransferases. J. Mol. Biol. 328:307–17

3. Varki A. 1993. Biological roles of oligosaccharides: All of the theories are correct. Glycobiology 3:97–1304. Jaeken J, Matthijs G. 2007. Congenital disorders of glycosylation: a rapidly expanding disease family.

Annu. Rev. Genomics Hum. Genet. 8:261–785. Bertozzi CR, Kiessling LL. 2001. Chemical glycobiology. Science 291:2357–646. Baum LG. 2002. Developing a taste for sweets. Immunity 16:5–87. Boltje TJ, Buskas T, Boons G-J. 2009. Opportunities and challenges in synthetic oligosaccharide and

glycoconjugate research. Nat. Chem. 1:611–228. Bernardes GJ, Castagner B, Seeberger PH. 2009. Combined approaches to the synthesis and study of

glycoproteins. ACS Chem. Biol. 4:703–13


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

9. Galonic DP, Gin DY. 2007. Chemical glycosylation in the synthesis of glycoconjugate antitumourvaccines. Nature 446:1000–7

10. Zhu X, Schmidt RR. 2009. New principles for glycoside-bond formation. Angew. Chem. Int. Ed. 48:1900–34

11. Petitou M, Duchaussoy P, Driguez P-A, Jaurand G, Herault J-P, et al. 1998. First synthetic carbohydrateswith the full anticoagulant properties of heparin. Angew. Chem. Int. Ed. 37:3009–14

12. Guerrini M, Zhang Z, Shriver Z, Naggi A, Masuko S, et al. 2009. Orthogonal analytical approaches todetect potential contaminants in heparin. Proc. Natl. Acad. Sci. USA 106:16956–61

13. Fischer E. 1893. Ueber die Gluooside der Alkohole. Ber. Dtsch. Chem. Ges. 26:2400–1214. Mootoo DR, Konradsson P, Udodong U, Fraser-Reid B. 1988. Armed and disarmed n-pentenyl glyco-

sides in saccharide couplings leading to oligosaccharides. J. Am. Chem. Soc. 110:5583–8415. Crich D, Smith M. 2001. 1-Benzenesulfinyl piperidine/trifluoromethanesulfonic anhydride: a potent

combination of shelf-stable reagents for the low-temperature conversion of thioglycosides to glycosyltriflates and for the formation of diverse glycosidic linkages. J. Am. Chem. Soc. 123:9015–20

16. Ikeda K, Aizawa M, Sato K, Sato M. 2006. Novel glycosylation reactions using glycosyl thioimidates ofN-acetylneuraminic acid as sialyl donors. Bioorg. Med. Chem. Lett. 16:2618–20

17. Crich D, Li W. 2006. Efficient glycosidation of a phenyl thiosialoside donor with diphenyl sulfoxide andtriflic anhydride in dichloromethane. Org. Lett. 8:959–62

18. Meijer A, Ellervik U. 2004. Interhalogens (ICl/IBr) and AgOTf in thioglycoside activation; synthesis ofbislactam analogues of ganglioside GD3. J. Org. Chem. 69:6249–56

19. Tanaka H, Tateno Y, Nishiura Y, Takahashi T. 2008. Efficient synthesis of an α(2,9) trisialic acid byone-pot glycosylation and polymer-assisted deprotection. Org. Lett. 10:5597–600

20. Mong TKK, Lee HK, Duron SG, Wong CH. 2003. Reactivity-based one-pot total synthesis of fucoseGM1 oligosaccharide. Proc. Natl. Acad. Sci. USA 100:797–802

21. Wang Z, Zhou LY, El-Boubbou K, Ye XS, Huang XF. 2007. Multi-component one-pot synthesis of thetumor-associated carbohydrate antigen Globo-H based on preactivation of thioglycosyl donors. J. Org.Chem. 72:6409–20

22. Seeberger PH. 2008. Automated oligosaccharide synthesis. Chem. Soc. Rev. 37:19–2823. Sears P, Wong CH. 2001. Toward automated synthesis of oligosaccharides and glycoproteins. Science

291:2344–5024. Werz DB, Ranzinger R, Herget S, Adibekian A, von der Lieth C-W, Seeberger PH. 2007. Exploring

the structural diversity of mammalian carbohydrates (“glycospace”) by statistical databank analysis. ACSChem. Biol. 2:685–91

25. Koeller KM, Wong CH. 2000. Synthesis of complex carbohydrates and glycoconjugates: enzyme-basedand programmable one-pot strategies. Chem. Rev. 100:4465–94

26. Shaikh FA, Withers SG. 2008. Teaching old enzymes new tricks: engineering and evolution of glycosi-dases and glycosyl transferases for improved glycoside synthesis. Biochem. Cell Biol. 86:169–77

27. Wang LX, Huang W. 2009. Enzymatic transglycosylation for glycoconjugate synthesis. Curr. Opin.Chem. Biol. 13:592–600

28. Chokhawala HA, Huang S, Lau K, Yu H, Cheng J, et al. 2008. Combinatorial chemoenzymatic synthesisand high-throughput screening of sialosides. ACS Chem. Biol. 3:567–76

29. Zhang C, Griffith BR, Fu Q, Albermann C, Fu X, et al. 2006. Exploiting the reversibility of naturalproduct glycosyltransferase-catalyzed reactions. Science 313:1291–94

30. Williams GJ, Gantt RW, Thorson JS. 2008. The impact of enzyme engineering upon natural productglycodiversification. Curr. Opin. Chem. Biol. 12:556–64

31. Tao H, Peralta-Yahya P, Decatur J, Cornish VW. 2008. Characterization of a new glycosynthase clonedby using chemical complementation. ChemBioChem 9:681–84

32. Ben-David A, Shoham G, Shoham Y. 2008. A universal screening assay for glycosynthases: directed evo-lution of glycosynthase XynB2(E335G) suggests a general path to enhance activity. Chem. Biol. 15:546–51

33. Aharoni A, Thieme K, Chiu CP, Buchini S, Lairson LL, et al. 2006. High-throughput screening method-ology for the directed evolution of glycosyltransferases. Nat. Methods 3:609–14

34. Sasaki A, Ishimizu T, Geyer R, Hase S. 2005. Synthesis of β-mannosides using the transglycosylationactivity of endo-β-mannosidase from Lilium longiflorum. FEBS J. 272:1660–68


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

35. Huang W, Li C, Li B, Umekawa M, Yamamoto K, et al. 2009. Glycosynthases enable a highly effi-cient chemoenzymatic synthesis of N-glycoproteins carrying intact natural N-glycans. J. Am. Chem. Soc.131:2214–23

36. Wada J, Honda Y, Nagae M, Kato R, Wakatsuki S, et al. 2008. 1,2-α-L-Fucosynthase: a glycosynthasederived from an inverting α-glycosidase with an unusual reaction mechanism. FEBS Lett. 582:3739–43

37. Honda Y, Fushinobu S, Hidaka M, Wakagi T, Shoun H, et al. 2008. Alternative strategy for convertingan inverting glycoside hydrolase into a glycosynthase. Glycobiology 18:325–30

38. Leppanen A, White SP, Helin J, McEver RP, Cummings RD. 2000. Binding of glycosulfopeptides toP-selectin requires stereospecific contributions of individual tyrosine sulfate and sugar residues. J. Biol.Chem. 275:39569–78

39. Herzner H, Reipen T, Schultz M, Kunz H. 2000. Synthesis of glycopeptides containing carbohydrateand peptide recognition motifs. Chem. Rev. 100:4495–538

40. Gamblin DP, Scanlan EM, Davis BG. 2009. Glycoprotein synthesis: an update. Chem. Rev. 109:131–6341. Kan C, Danishefsky SJ. 2009. Recent departures in the synthesis of peptides and glycopeptides.

Tetrahedron 65:9047–6542. Dawson PE, Kent SBH. 2000. Synthesis of native proteins by chemical ligation. Annu. Rev. Biochem.

69:923–6043. Muir TW. 2003. Semisynthesis of proteins by expressed protein ligation. Annu. Rev. Biochem. 72:249–8944. Liu L, Hong ZY, Wong CH. 2006. Convergent glycopeptide synthesis by traceless Staudinger ligation

and enzymatic coupling. ChemBioChem 7:429–3245. Macmillan D, Anderson DW. 2004. Rapid synthesis of acyl transfer auxiliaries for cysteine-free native

glycopeptide ligation. Org. Lett. 6:4659–6246. Crich D, Banerjee A. 2007. Native chemical ligation at phenylalanine. J. Am. Chem. Soc. 129:10064–6547. Chen J, Wan Q, Yuan Y, Zhu J, Danishefsky SJ. 2008. Native chemical ligation at valine: a contribution

to peptide and glycopeptide synthesis. Angew. Chem. Int. Ed. 47:8521–2448. Bennett CS, Dean SM, Payne RJ, Ficht S, Brik A, Wong CH. 2008. Sugar-assisted glycopeptide ligation

with complex oligosaccharides: scope and limitations. J. Am. Chem. Soc. 130:11945–5249. Savage PB, Teyton L, Bendelac A. 2006. Glycolipids for natural killer T cells. Chem. Soc. Rev. 35:771–7950. Lahiri S, Futerman AH. 2007. The metabolism and function of sphingolipids and glycosphingolipids.

Cell. Mol. Life Sci. 64:2270–8451. Sletten EM, Bertozzi CR. 2009. Bioorthogonal chemistry: fishing for selectivity in a sea of functionality.

Angew. Chem. Int. Ed. 48:6974–9852. Gama CI, Tully SE, Sotogaku N, Clark PM, Rawat M, et al. 2006. Sulfation patterns of glycosamino-

glycans encode molecular recognition and activity. Nat. Chem. Biol. 2:467–7353. Noti C, de Paz JL, Polito L, Seeberger PH. 2006. Preparation and use of microarrays containing synthetic

heparin oligosaccharides for the rapid analysis of heparin-protein interactions. Chem. Eur. J. 12:8664–8654. Carlson CB, Mowery P, Owen RM, Dykhuizen EC, Kiessling LL. 2007. Selective tumor cell targeting

using low-affinity, multivalent interactions. ACS Chem. Biol. 2:119–2755. Kitov PI, Shimizu H, Homans SW, Bundle DR. 2003. Optimization of tether length in nonglycosidically

linked bivalent ligands that target sites 2 and 1 of a Shiga-like toxin. J. Am. Chem. Soc. 125:3284–9456. Park S, Lee MR, Pyo SJ, Shin I. 2004. Carbohydrate chips for studying high-throughput carbohydrate-

protein interactions. J. Am. Chem. Soc. 126:4812–1957. Verez-Bencomo V, Fernandez-Santana V, Hardy E, Toledo M, Rodriguez M, et al. 2004. A synthetic

conjugate polysaccharide vaccine against Haemophilus influenzae type b. Science 305:522–2558. Gamblin DP, Garnier P, van Kasteren S, Oldham NJ, Fairbanks AJ, Davis BG. 2004. Glyco-SeS:

selenenylsulfide-mediated protein glycoconjugation—a new strategy in post-translational modification.Angew. Chem. Int. Ed. 43:828–33

59. Bryan MC, Fazio F, Lee HK, Huang CY, Chang A, et al. 2004. Covalent display of oligosaccharidearrays in microtiter plates. J. Am. Chem. Soc. 126:8640–41

60. Fazio F, Bryan MC, Blixt O, Paulson JC, Wong CH. 2002. Synthesis of sugar arrays in microtiter plate.J. Am. Chem. Soc. 124:14397–402

61. Mortell KH, Gingras M, Kiessling LL. 1994. Synthesis of cell agglutination inhibitors by ring-openingmetathesis polymerization. J. Am. Chem. Soc. 116:10253–54


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

62. Nilsson BL, Kiessling LL, Raines RT. 2001. High-yielding Staudinger ligation of a phosphinothioesterand azide to form a peptide. Org. Lett. 3:9–12

63. Saxon E, Armstrong JI, Bertozzi CR. 2000. A “traceless” Staudinger ligation for the chemoselectivesynthesis of amide bonds. Org. Lett. 2:2141–43

64. Hart GW, Housley MP, Slawson C. 2007. Cycling of O-linked β-N-acetylglucosamine on nucleocyto-plasmic proteins. Nature 446:1017–22

65. Weis WI, Drickamer K. 1996. Structural basis of lectin-carbohydrate recognition. Annu. Rev. Biochem.65:441–73

66. Lundquist JJ, Toone EJ. 2002. The cluster glycoside effect. Chem. Rev. 102:555–7867. Mortell KH, Weatherman RV, Kiessling LL. 1996. Recognition specificity of neoglycopolymers pre-

pared by ring-opening metathesis polymerization. J. Am. Chem. Soc. 118:2297–9868. Lee YC, Lee RT. 1995. Carbohydrate-protein interactions: basis of glycobiology. Acc. Chem. Res. 28:321–

2769. Kiessling LL, Gestwicki JE, Strong LE. 2006. Synthetic multivalent ligands as probes of signal trans-

duction. Angew. Chem. Int. Ed. 45:2348–6870. Mammen M, Choi S-K, Whitesides GM. 1998. Polyvalent interactions in biological systems: Implica-

tions for design and use of multivalent ligands and inhibitors. Angew. Chem. Int. Ed. 37:2754–9471. Horlacher T, Seeberger PH. 2008. Carbohydrate arrays as tools for research and diagnostics. Chem. Soc.

Rev. 37:1414–2272. Liang PH, Wu CY, Greenberg WA, Wong CH. 2008. Glycan arrays: biological and medical applications.

Curr. Opin. Chem. Biol. 12:86–9273. Krishnamoorthy L, Mahal LK. 2009. Glycomic analysis: an array of technologies. ACS Chem. Biol.

4:715–3274. Park S, Sung JW, Shin I. 2009. Fluorescent glycan derivatives: their use for natural glycan microarrays.

ACS Chem. Biol. 4:699–70175. Oyelaran O, Gildersleeve JC. 2009. Glycan arrays: recent advances and future challenges. Curr. Opin.

Chem. Biol. 13:406–1376. Willats WG, Rasmussen SE, Kristensen T, Mikkelsen JD, Knox JP. 2002. Sugar-coated microarrays: a

novel slide surface for the high-throughput analysis of glycans. Proteomics 2:1666–7177. Bryan MC, Plettenburg O, Sears P, Rabuka D, Wacowich-Sgarbi S, Wong CH. 2002. Saccharide display

on microtiter plates. Chem. Biol. 9:713–2078. Wang DN, Liu SY, Trummer BJ, Deng C, Wang AL. 2002. Carbohydrate microarrays for the recognition

of cross-reactive molecular markers of microbes and host cells. Nat. Biotechnol. 20:275–8179. Fukui S, Feizi T, Galustian C, Lawson AM, Chai W. 2002. Oligosaccharide microarrays for high-

throughput detection and specificity assignments of carbohydrate-protein interactions. Nat. Biotechnol.20:1011–17

80. Liu Y, Feizi T, Carnpanero-Rhodes MA, Childs RA, Zhang Y, et al. 2007. Neoglycolipid probes preparedvia oxime ligation for microarray analysis of oligosaccharide-protein interactions. Chem. Biol. 14:847–59

81. Mamidyala SK, Ko KS, Jaipuri FA, Park G, Pohl NL. 2006. Noncovalent fluorous interactions for thesynthesis of carbohydrate microarrays. J. Fluor. Chem. 127:571–79

82. Zhang J, Pourceau G, Meyer A, Vidal S, Praly J, et al. 2009. DNA-directed immobilisation of gly-comimetics for glycoarrays application: comparison with covalent immobilisation, and development ofan on-chip IC50 measurement assay. Biosens. Bioelectron. 24:2515–21

83. Zhi ZL, Powell AK, Turnbull JE. 2006. Fabrication of carbohydrate microarrays on gold surfaces: directattachment of nonderivatized oligosaccharides to hydrazide-modified self-assembled monolayers. Anal.Chem. 78:4786–93

84. Song X, Lasanajak Y, Xia B, Smith DF, Cummings RD. 2009. Fluorescent glycosylamides produced bymicroscale derivatization of free glycans for natural glycan microarrays. ACS Chem. Biol. 4:741–50

85. de Boer AR, Hokke CH, Deelder AM, Wuhrer M. 2007. General microarray technique for immobiliza-tion and screening of natural glycans. Anal. Chem. 79:8107–13

86. Gestwicki JE, Cairo CW, Mann DA, Owen RM, Kiessling LL. 2002. Selective immobilization of mul-tivalent ligands for surface plasmon resonance and fluorescence microscopy. Anal. Biochem. 305:149–55


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

87. Oyelaran O, Li Q, Farnsworth D, Gildersleeve JC. 2009. Microarrays with varying carbohydrate densityreveal distinct subpopulations of serum antibodies. J. Proteome Res. 8:3529–38

88. Godula K, Rabuka D, Nam KT, Bertozzi CR. 2009. Synthesis and microcontact printing of dual end-functionalized mucin-like glycopolymers for microarray applications. Angew. Chem. Int. Ed. 48:4973–76

89. Park S, Shin I. 2007. Carbohydrate microarrays for assaying galactosyltransferase activity. Org. Lett.9:1675–78

90. Manimala JC, Li Z, Jain A, VedBrat S, Gildersleeve JC. 2005. Carbohydrate array analysis of anti-Tn an-tibodies and lectins reveals unexpected specificities: implications for diagnostic and vaccine development.ChemBioChem 6:2229–41

91. Fuster MM, Esko JD. 2005. The sweet and sour of cancer: glycans as novel therapeutic targets. Nat. Rev.Cancer 5:526–42

92. de Paz JL, Noti C, Seeberger PH. 2006. Microarrays of synthetic heparin oligosaccharides. J. Am. Chem.Soc. 128:2766–67

93. Tully SE, Rawat M, Hsieh-Wilson LC. 2006. Discovery of a TNF-α antagonist using chondroitin sulfatemicroarrays. J. Am. Chem. Soc. 128:7740–41

94. Shipp EL, Hsieh-Wilson LC. 2007. Profiling the sulfation specificities of glycosaminoglycan interactionswith growth factors and chemotactic proteins using microarrays. Chem. Biol. 14:195–208

95. de Paz JL, Moseman EA, Noti C, Polito L, von Andrian UH, Seeberger PH. 2007. Profiling heparin-chemokine interactions using synthetic tools. ACS Chem. Biol. 2:735–44

96. Zhu XY, Holtz B, Wang Y, Wang L, Orndorff PE, Guo A. 2009. Quantitative glycomics from fluidicglycan microarrays. J. Am. Chem. Soc. 131:13646–50

97. Pilobello KT, Slawek DE, Mahal LK. 2007. A ratiometric lectin microarray approach to analysis of thedynamic mammalian glycome. Proc. Natl. Acad. Sci. USA 104:11534–39

98. Chen S, Zheng T, Shortreed MR, Alexander C, Smith LM. 2007. Analysis of cell surface carbohydrateexpression patterns in normal and tumorigenic human breast cell lines using lectin arrays. Anal. Chem.79:5698–702

99. Ebe Y, Kuno A, Uchiyama N, Koseki-Kuno S, Yamada M, et al. 2006. Application of lectin microarrayto crude samples: differential glycan profiling of Lec mutants. J. Biochem. 139:323–27

100. Kuno A, Uchiyama N, Koseki-Kuno S, Ebe Y, Takashima S, et al. 2005. Evanescent-field fluorescence-assisted lectin microarray: a new strategy for glycan profiling. Nat. Methods 2:851–56

101. Smith EA, Thomas WD, Kiessling LL, Corn RM. 2003. Surface plasmon resonance imaging studies ofprotein-carbohydrate interactions. J. Am. Chem. Soc. 125:6140–48

102. Godula K, Umbel ML, Rabuka D, Botyanszki Z, Bertozzi CR, Parthasarathy R. 2009. Control of themolecular orientation of membrane-anchored biomimetic glycopolymers. J. Am. Chem. Soc. 131:10263–68

103. Ohtsubo K, Marth JD. 2006. Glycosylation in cellular mechanisms of health and disease. Cell 126:855–67104. Collins BE, Blixt O, Bovin NV, Danzer CP, Chui D, et al. 2002. Constitutively unmasked CD22 on B

cells of ST6Gal I knockout mice: novel sialoside probe for murine CD22. Glycobiology 12:563–71105. Ernst B, Magnani JL. 2009. From carbohydrate leads to glycomimetic drugs. Nat. Rev. Drug Discov.

8:661–77106. Paulson JC, Blixt O, Collins BE. 2006. Sweet spots in functional glycomics. Nat. Chem. Biol. 2:238–48107. Yang R-Y, Liu F-T. 2003. Galectins in cell growth and apoptosis. Cell. Mol. Life Sci. 60:267–76108. Liu F-T, Rabinovich GA. 2005. Galectins as modulators of tumour progression. Nat. Rev. Cancer 5:29–41109. Kiessling LL, Carlson EE. 2007. The search for chemical probes to illuminate carbohydrate function.

In Chemical Biology. From Small Molecules to System Biology and Drug Design, ed. SL Schreiber, T Kapoor,G Wess, 2:635–67. Weinheim: Wiley-VCH

110. Cumpstey I, Salomonsson E, Sundin A, Leffler H, Nilsson UJ. 2008. Double affinity amplification ofgalectin-ligand interactions through arginine-arene interactions: synthetic, thermodynamic, and com-putational studies with aromatic diamido thiodigalactosides. Chem. Eur. J. 14:4233–45

111. MacKinnon AC, Farnworth SL, Hodkinson PS, Henderson NC, Atkinson KM, et al. 2008. Regulationof alternative macrophage activation by galectin-3. J. Immunol. 180:2650–58

112. Moreno-Vargas AJ, Molina L, Carmona AT, Ferrali A, Lambelet M, et al. 2008. Synthesis and biologicalevaluation of S-neofucopeptides as E- and P-selectin inhibitors. Eur. J. Org. Chem. 2008:2973–82


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

113. Kaila N, Janz K, Huang A, Moretto A, DeBernardo S, et al. 2007. 2-(4-Chlorobenzyl)-3-hydroxy-7,8,9,10-tetrahydrobenzo[H]quinoline-4-carboxylic acid (PSI-697): identification of a clinical candidatefrom the quinoline salicylic acid series of P-selectin antagonists. J. Med. Chem. 50:40–64

114. Schuster MC, Mann DA, Buchholz TJ, Johnson KM, Thomas WD, Kiessling LL. 2003. Parallel synthesisof glycomimetic libraries: targeting a C-type lectin. Org. Lett. 5:1407–10

115. van Kyook Y, Geijtenbeek TBH. 2003. DC-SIGN: escape mechanisms for pathogens. Nat. Rev. Immunol.3:697–709

116. Borrok MJ, Kiessling LL. 2007. Non-carbohydrate inhibitors of the lectin DC-SIGN. J. Am. Chem. Soc.129:12780–85

117. Page MI, Jencks WP. 1971. Entropic contributions to rate accelerations in enzymic and intramolecularreactions and the chelate effect. Proc. Natl. Acad. Sci. USA 68:1678–83

118. Pieters RJ. 2009. Maximising multivalency effects in protein-carbohydrate interactions. Org. Biomol.Chem. 7:2013–25

119. Kiessling LL, Gestwicki JE, Strong LE. 2000. Synthetic multivalent ligands in the exploration of cell-surface interactions. Curr. Opin. Chem. Biol. 4:696–703

120. Imberty A, Chabre YM, Roy R. 2008. Glycomimetics and glycodendrimers as high affinity microbialanti-adhesins. Chem. Eur. J. 14:7490–99

121. Gestwicki JE, Cairo CW, Strong LE, Oetjen KA, Kiessling LL. 2002. Influencing receptor-ligandbinding mechanisms with multivalent ligand architecture. J. Am. Chem. Soc. 124:14922–33

122. Fan EK, Zhang ZS, Minke WE, Hou Z, Verlinde C, Hol WGJ. 2000. High-affinity pentavalent ligands ofEscherichia coli heat-labile enterotoxin by modular structure-based design. J. Am. Chem. Soc. 122:2663–64

123. Kitov PI, Sadowska JM, Mulvey G, Armstrong GD, Ling H, et al. 2000. Shiga-like toxins are neutralizedby tailored multivalent carbohydrate ligands. Nature 403:669–72

124. McEver RP. 2002. Selectins: lectins that initiate cell adhesion under flow. Curr. Opin. Cell Biol. 14:581–86125. Mowery P, Yang ZQ, Gordon EJ, Dwir O, Spencer AG, et al. 2004. Synthetic glycoprotein mimics

inhibit L-selectin-mediated rolling and promote L-selectin shedding. Chem. Biol. 11:725–32126. Ouerfelli O, Warren JD, Wilson R, Danishefsky SJ. 2005. Synthetic carbohydrate-based antitumor

vaccines: challenges and opportunities. Expert Rev. Vaccines 4:677–85127. Seeberger PH, Werz DB. 2007. Synthesis and medical applications of oligosaccharides. Nature 446:1046–

51128. Ingale S, Wolfert MA, Gaekwad J, Buskas T, Boons G-J. 2007. Robust immune responses elicited by a

fully synthetic three-component vaccine. Nat. Chem. Biol. 3:663–67129. Courtney AH, Puffer EB, Pontrello JK, Yang Z-Q, Kiessling LL. 2009. Sialylated multivalent antigens

engage CD22 in trans and inhibit B cell activation. Proc. Natl. Acad. Sci. USA 106:2500–5130. Kitov PI, Mulvey GL, Griener TP, Lipinski T, Solomon D, et al. 2008. In vivo supramolecular templating

enhances the activity of multivalent ligands: a potential therapeutic against the Escherichia coli O157 AB5

toxins. Proc. Natl. Acad. Sci. USA 105:16837–42131. O’Reilly MK, Collins BE, Han S, Liao L, Rillahan C, et al. 2008. Bifunctional CD22 ligands use

multimeric immunoglobulins as protein scaffolds in assembly of immune complexes on B cells. J. Am.Chem. Soc. 130:7736–45

132. Galili U, Shohet SB, Kobrin E, Stults CLM, Macher BA. 1988. Man, apes, and Old World monkeysdiffer from other mammals in the expression of α-galactosyl epitopes on nucleated cells. J. Biol. Chem.263:17755–62

133. Tkacz JS, Lampen O. 1975. Tunicamycin inhibition of polyisoprenyl N-acetylglucosaminyl pyrophos-phate formation in calf-liver microsomes. Biochem. Biophys. Res. Commun. 65:248–57

134. Eason PD, Imperiali B. 1999. A potent oligosaccharyl transferase inhibitor that crosses the intracellularendoplasmic reticulum membrane. Biochemistry 38:5430–37

135. Elbein AD. 1987. Inhibitors of the biosynthesis and processing of N-linked oligosaccharide chains. Annu.Rev. Biochem. 56:497–534

136. Gross BJ, Swoboda JG, Walker S. 2008. A strategy to discover inhibitors of O-linked glycosylation. J.Am. Chem. Soc. 130:440–41

137. Laurent N, Voglmeir J, Flitsch SL. 2008. Glycoarrays—tools for determining protein-carbohydrateinteractions and glycoenzyme specificity. Chem. Commun. 37:4400–12


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

ANRV413-BI79-22 ARI 27 April 2010 21:25

138. Rose NL, Zheng RB, Pearcey J, Zhou R, Completo GC, Lowary TL. 2008. Development of a coupledspectrophotometric assay for GlfT2, a bifunctional mycobacterial galactofuranosyltransferase. Carbohydr.Res. 343:2130–39

139. Helm JS, Hu Y, Chen L, Gross B, Walker S. 2003. Identification of active-site inhibitors of MurG usinga generalizable, high-throughput glycosyltransferase screen. J. Am. Chem. Soc. 125:11168–69

140. Lillelund VH, Jensen HH, Liang X, Bols M. 2002. Recent developments of transition-state analogueglycosidase inhibitors of non-natural product origin. Chem. Rev. 102:515–53

141. Asano N. 2003. Glycosidase inhibitors: update and perspectives on practical use. Glycobiology 13:R93–104142. Compain P, Martin OR. 2003. Design, synthesis and biological evaluation of iminosugar-based glyco-

syltransferase inhibitors. Curr. Top. Med. Chem. 3:541–60143. Hu Y, Helm JS, Chen L, Ginsberg C, Gross B, et al. 2004. Identification of selective inhibitors for the

glycosyltransferase MurG via high-throughput screening. Chem. Biol. 11:703–11144. Carlson EE, May JF, Kiessling LL. 2006. Chemical probes of UDP-galactopyranose mutase. Chem. Biol.

13:825–37145. Dykhuizen EC, May JF, Tongpenyai A, Kiessling LL. 2008. Inhibitors of UDP-galactopyranose mutase

thwart mycobacterial growth. J. Am. Chem. Soc. 130:6706–7146. Brown JR, Crawford BE, Esko JD. 2007. Glycan antagonists and inhibitors: a fount for drug discovery.

Crit. Rev. Biochem. Mol. Biol. 42:481–515147. Kayser H, Zeitler R, Kannicht C, Grunow D, Nuck R, Reutter W. 1992. Biosynthesis of a nonphysio-

logical sialic acid in different rat organs, using N-propanoyl-D-hexosamines as precursors. J. Biol. Chem.267:16934–38

148. Mahal LK, Charter NW, Angata K, Fukuda M, Koshland DE, Bertozzi CR. 2001. A small-moleculemodulator of poly-α2,8-sialic acid expression on cultured neurons and tumor cells. Science 294:380–82

149. Murrey H, Gama C, Kalovidouris S, Luo W, Driggers E, et al. 2006. Protein fucosylation regulatessynapsin Ia/Ib expression and neuronal morphology in primary hippocampal neurons. Proc. Natl. Acad.Sci. USA 103:21–26

150. Collins B, Yang L, Mukhopadhyay G, Filbin M, Kiso M, et al. 1997. Sialic acid specificity of myelin-associated glycoprotein binding. J. Biol. Chem. 272:1248–55

151. Liu T, Guo Z, Yang Q, Sad S, Jennings H. 2000. Biochemical engineering of surface α2–8 polysialicacid for immunotargeting tumor cells. J. Biol. Chem. 275:32832–36

152. Agard NJ, Bertozzi CR. 2009. Chemical approaches to perturb, profile, and perceive glycans. Acc. Chem.Res. 42:788–97

153. Vocadlo DJ, Hang HC, Kim E-J, Hanover JA, Bertozzi CR. 2003. A chemical approach for identifyingO-GlcNAc-modified proteins in cells. Proc. Natl. Acad. Sci. USA 100:9116–21

154. Hang H, Yu C, Kato D, Bertozzi C. 2003. A metabolic labeling approach toward proteomic analysis ofmucin-type O-linked glycosylation. Proc. Natl. Acad. Sci. USA 100:14846–51

155. Han S, Collins BE, Bengtson P, Paulson JC. 2005. Homomultimeric complexes of CD22 in B cellsrevealed by protein-glycan cross-linking. Nat. Chem. Biol. 1:93–97

156. Tanaka Y, Kohler JJ. 2008. Photoactivatable crosslinking sugars for capturing glycoprotein interactions.J. Am. Chem. Soc. 130:3278–79

157. Sampathkumar S-G, Li AV, Jones MB, Sun Z, Yarema KJ. 2006. Metabolic installation of thiols intosialic acid modulates adhesion and stem cell biology. Nat. Chem. Biol. 2:149–52

158. Zeng Y, Ramya TNC, Dirksen A, Dawson PE, Paulson JC. 2009. High-efficiency labeling of sialylatedglycoproteins on living cells. Nat. Methods 6:207–9

159. Fisher E. 1902. Syntheses in the purine and sugar group. Nobel lect. http://nobelprize.org/nobelprizes/chemistry/laureates/1902/fischer-lecture.pdf


Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

AR413-FM ARI 28 April 2010 16:43

Annual Review ofBiochemistry

Volume 79, 2010Contents

Preface

The Power of OneJames E. Rothman � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �v

Prefatory Article

FrontispieceAaron Klug � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � xiv

From Virus Structure to Chromatin: X-ray Diffractionto Three-Dimensional Electron MicroscopyAaron Klug � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 1

Recent Advances in Biochemistry

Genomic Screening with RNAi: Results and ChallengesStephanie Mohr, Chris Bakal, and Norbert Perrimon � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �37

Nanomaterials Based on DNANadrian C. Seeman � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �65

Eukaryotic Chromosome DNA Replication: Where, When, and How?Hisao Masai, Seiji Matsumoto, Zhiying You, Naoko Yoshizawa-Sugata,

and Masako Oda � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �89

Regulators of the Cohesin NetworkBo Xiong and Jennifer L. Gerton � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 131

Reversal of Histone Methylation: Biochemical and MolecularMechanisms of Histone DemethylasesNima Mosammaparast and Yang Shi � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 155

The Mechanism of Double-Strand DNA Break Repair by theNonhomologous DNA End-Joining PathwayMichael R. Lieber � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 181

The Discovery of Zinc Fingers and Their Applications in GeneRegulation and Genome ManipulationAaron Klug � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 213

vii

Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

AR413-FM ARI 28 April 2010 16:43

Origins of Specificity in Protein-DNA RecognitionRemo Rohs, Xiangshu Jin, Sean M. West, Rohit Joshi, Barry Honig,

and Richard S. Mann � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 233

Transcript Elongation by RNA Polymerase IILuke A. Selth, Stefan Sigurdsson, and Jesper Q. Svejstrup � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 271

Biochemical Principles of Small RNA PathwaysQinghua Liu and Zain Paroo � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 295

Functions and Regulation of RNA Editing by ADAR DeaminasesKazuko Nishikura � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 321

Regulation of mRNA Translation and Stability by microRNAsMarc Robert Fabian, Nahum Sonenberg, and Witold Filipowicz � � � � � � � � � � � � � � � � � � � � � � � � 351

Structure and Dynamics of a Processive Brownian Motor:The Translating RibosomeJoachim Frank and Ruben L. Gonzalez, Jr. � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 381

Adding New Chemistries to the Genetic CodeChang C. Liu and Peter G. Schultz � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 413

Bacterial Nitric Oxide SynthasesBrian R. Crane, Jawahar Sudhamsu, and Bhumit A. Patel � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 445

Enzyme Promiscuity: A Mechanistic and Evolutionary PerspectiveOlga Khersonsky and Dan S. Tawfik � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 471

Hydrogenases from Methanogenic Archaea, Nickel, a Novel Cofactor,and H2 StorageRudolf K. Thauer, Anne-Kristin Kaster, Meike Goenrich, Michael Schick,

Takeshi Hiromoto, and Seigo Shima � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 507

Copper MetallochaperonesNigel J. Robinson and Dennis R. Winge � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 537

High-Throughput Metabolic Engineering: Advances inSmall-Molecule Screening and SelectionJeffrey A. Dietrich, Adrienne E. McKee, and Jay D. Keasling � � � � � � � � � � � � � � � � � � � � � � � � � � 563

Botulinum Neurotoxin: A Marvel of Protein DesignMauricio Montal � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 591

Chemical Approaches to GlycobiologyLaura L. Kiessling and Rebecca A. Splain � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 619

Cellulosomes: Highly Efficient Nanomachines Designed toDeconstruct Plant Cell Wall Complex CarbohydratesCarlos M.G.A. Fontes and Harry J. Gilbert � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 655

viii Contents

Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

AR413-FM ARI 28 April 2010 16:43

Somatic Mitochondrial DNA Mutations in Mammalian AgingNils-Goran Larsson � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 683

Physical Mechanisms of Signal Integration by WASP Family ProteinsShae B. Padrick and Michael K. Rosen � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 707

Amphipols, Nanodiscs, and Fluorinated Surfactants: ThreeNonconventional Approaches to Studying Membrane Proteins inAqueous SolutionsJean-Luc Popot � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 737

Protein Sorting Receptors in the Early Secretory PathwayJulia Dancourt and Charles Barlowe � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 777

Virus Entry by EndocytosisJason Mercer, Mario Schelhaas, and Ari Helenius � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 803

Indexes

Cumulative Index of Contributing Authors, Volumes 75–79 � � � � � � � � � � � � � � � � � � � � � � � � � � � 835

Cumulative Index of Chapter Titles, Volumes 75–79 � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 839

Errata

An online log of corrections to Annual Review of Biochemistry articles may be found athttp://biochem.annualreviews.org

Contents ix

Ann

u. R

ev. B

ioch

em. 2

010.

79:6

19-6

53. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

nive

rsity

of

Wis

cons

in -

Mad

ison

on

07/2

9/10

. For

per

sona

l use

onl

y.

Documents

Chemical Approaches to Glycobiology...Chemical Approaches to Glycobiology Laura L. Kiessling1,2 and Rebecca A. Splain1 1Department of Chemistry, 2Department of Biochemistry, University