Reversible sialylation: synthesis of cytidine 5'-monophospho-N-acetylneuraminic acid from cytidine 5'-monophosphate with alpha2,3-sialyl O-glycan-, glycolipid-, and macromolecule-based donors yields diverse sialylated products

trans-Sialylation: a strategy used to incorporate sialic acid into oligosaccharides

Sialic acid, as a component of cell surface glycoconjugates, plays a crucial role in recognition events. Efficient synthetic methods are necessary for the supply of sialosides in enough quantities for biochemical and immunological studies. Enzymatic glycosylations obviate the steps of protection and deprotection of the constituent monosaccharides required in a chemical synthesis. Sialyl transferases with CMP-Neu5Ac as an activated donor were used for the construction of α2-3 or α2-6 linkages to terminal galactose or N-acetylgalactosamine units. trans-Sialidases may transfer sialic acid from a sialyl glycoside to a suitable acceptor and specifically construct a Siaα2-3Galp linkage. The trans-sialidase of Trypanosoma cruzi (TcTS), which fulfills an important role in the pathogenicity of the parasite, is the most studied one. The recombinant enzyme was used for the sialylation of β-galactosyl oligosaccharides. One of the main advantages of trans-sialylation is that it circumvents the use of the high energy nucleotide. Easily available glycoproteins with a high content of sialic acid such as fetuin and bovine κ-casein-derived glycomacropeptide (GMP) have been used as donor substrates. Here we review the trans-sialidase from various microorganisms and describe their application for the synthesis of sialooligosaccharides.

One-pot multienzyme (OPME) chemoenzymatic synthesis of brain ganglioside glycans with human ST3GAL II expressed in E. coli

A human sialyltransferase ST3GAL II (hST3GAL II) was successfully expressed in Escherichia coli as an active soluble fusion protein with an N-terminal maltose-binding protein (MBP) and a C-terminal hexa-histidine tag. It was used as an efficient catalyst in a one-pot multienzyme (OPME) sialylation system for high-yield production of the glycans of ganglioside GM1b and highly sialylated brain gangliosides GD1a and GT1b. Further sialylation of GM1b and GD1a glycans using a bacterial α2-8-sialyltransferase in another OPME sialylation reaction led to the formation of the glycans of GD1c and brain ganglioside GT1a, respectively. The lower reverse glycosylation activity of the recombinant hST3GAL II compared to its bacterial sialyltransferase counterpart simplifies the handling of enzymatic synthetic reactions and has an advantage for future use in automated chemoenzymatic synthetic processes.

Cellular and Molecular Engineering of Glycan Sialylation in Heterologous Systems

Glycans have been shown to play a key role in many biological processes, such as signal transduction, immunogenicity, and disease progression. Among the various glycosylation modifications found on cell surfaces and in biomolecules, sialylation is especially important, because sialic acids are typically found at the terminus of glycans and have unique negatively charged moieties associated with cellular and molecular interactions. Sialic acids are also crucial for glycosylated biopharmaceutics, where they promote stability and activity. In this regard, heterogenous sialylation may produce variability in efficacy and limit therapeutic applications. Homogenous sialylation may be achieved through cellular and molecular engineering, both of which have gained traction in recent years. In this paper, we describe the engineering of intracellular glycosylation pathways through targeted disruption and the introduction of carbohydrate active enzyme genes. The focus of this review is on sialic acid-related genes and efforts to achieve homogenous, humanlike sialylation in model hosts. We also discuss the molecular engineering of sialyltransferases and their application in chemoenzymatic sialylation and sialic acid visualization on cell surfaces. The integration of these complementary engineering strategies will be useful for glycoscience to explore the biological significance of sialic acids on cell surfaces as well as the future development of advanced biopharmaceuticals.

Engineer P. multocida Heparosan Synthase 2 (PmHS2) for Size-Controlled Synthesis of Longer Heparosan Oligosaccharides

Pasteurella multocida heparosan synthase 2 (PmHS2) is a dual-function polysaccharide synthase having both α1-4-N-acetylglucosaminyltransferase (α1-4-GlcNAcT) and β1-4-glucuronyltransferase (β1-4-GlcAT) activities located in two separate catalytic domains. We found that removing PmHS2 N-terminal 80-amino acid residues improved enzyme stability and expression level while retaining its substrate promiscuity. We also identified the reverse glycosylation activities of PmHS2 which complicated its application in size-controlled synthesis of oligosaccharides longer than hexasaccharide. Engineered Δ80PmHS2 single-function-glycosyltransferase mutants Δ80PmHS2_D291N (α1-4-GlcNAcT lacking both forward and reverse β1-4-GlcAT activities) and Δ80PmHS2_D569N (β1-4-GlcAT lacking both forward and reverse α1-4-GlcNAcT activities) were designed and showed to minimize side product formation. They were successfully used in a sequential one-pot multienzyme (OPME) platform for size-controlled high-yield production of oligosaccharides up to decasaccharide. The study draws attention to the consideration of reverse glycosylation activities of glycosyltransferases, including polysaccharide synthases, when applying them in the synthesis of oligosaccharides and polysaccharides. The mutagenesis strategy has the potential to be extended to other multifunctional polysaccharide synthases with reverse glycosylation activities to generate catalysts with improved synthetic efficiency.

Enzymatic synthesis of N-acetyllactosamine from lactose enabled by recombinant β1,4-galactosyltransferases

Synthesis of LacNAc with reversible GalTs.

Active-Site His85 of Pasteurella dagmatis Sialyltransferase Facilitates Productive Sialyl Transfer and So Prevents Futile Hydrolysis of CMP-Neu5Ac

Sialyltransferases of the GT-80 glycosyltransferase family are considered multifunctional because of the array of activities detected. They exhibit glycosyl transfer, trans-sialylation, and hydrolysis activities. How these enzymes utilize their active-site residues in balancing the different enzymatic activities is not well understood. In this study of Pasteurella dagmatis α2,3sialyltransferase, we show that the conserved His85 controls efficiency and selectivity of the sialyl transfer. A His85→Asn variant was 200 times less efficient than wild-type for sialylation of lactose, and exhibited relaxed site selectivity to form not only the α2,3- but also the α2,6-sialylated product (21 %). The H85N variant was virtually inactive in trans-sialylation but showed almost the same CMP-Neu5Ac hydrolase activity as wild-type. The competition between sialyl transfer and hydrolysis in the conversion of CMP-Neu5Ac was dependent on the lactose concentration; this was characterized by a kinetic partition ratio of 85 m^-1 for the H85N variant, compared to 17 000 m^-1 for the wild-type enzyme. His85 promotes the productive sialyl transfer to lactose and so prevents hydrolysis of CMP-Neu5Ac in the reaction.

Novel interactions of complex carbohydrates with peanut (PNA), Ricinus communis (RCA-I), Sambucus nigra (SNA-I) and wheat germ (WGA) agglutinins as revealed by the binding specificities of these lectins towards mucin core-2 O-linked and N-linked glycans and related structures

Plant lectins through their multivalent quaternary structures bind intrinsically flexible oligosaccharides. They recognize fine structural differences in carbohydrates and interact with different sequences in mucin core 2 or complex-type N-glycan chain and also in healthy and malignant tissues. They are used in characterizing cellular and extracellular glycoconjugates modified in pathological processes. We study here, the complex carbohydrate-lectin interactions by determining the effects of substituents in mucin core 2 tetrasaccharide Galβ1-4GlcNAcβ1-6(Galβ1-3)GalNAcα-O-R and fetuin glycopeptides on their binding to agarose-immobilized lectins PNA, RCA-I, SNA-I and WGA. Briefly, in mucin core 2 tetrasaccharide (i) structures modified by α2-3/6-Sialyl LacNAc, LewisX and α1-3-Galactosyl LacNAc resulted in regular binding to PNA whereas compounds with 6-sulfo LacNAc displayed no-binding; (ii) strucures bearing α2-6-sialyl 6-sulfo LacNAc, or 6-sialyl LacdiNAc carbohydrates displayed strong binding to SNA-I; (iii) structures with α2-3/6-sialyl, α1-3Gal LacNAc or LewisX were non-binder to RCA-I and compounds with 6-sulfo LacNAc only displayed weak binding; (iv) structures containing LewisX, 6-Sulfo LewisX, α2-3/6-sialyl LacNAc, α2-3/6-sialyl 6-sulfo LacNAc and GalNAc Lewis-a were non-binding to WGA, those with α1-2Fucosyl, α1-3-Galactosyl LacNAc, α2-3-sialyl T-hapten plus 3'/6'sulfo LacNAc displayed weak binding, and compounds with α2-3-sialyl T-hapten, α2.6-Sialyl LacdiNAc, α2-3-sialyl D-Fucβ1-3 GalNAc and Fucα-1-2 D-Fucβ-1-3GalNAc displaying regular binding and GalNAc LewisX and LacdiNAc plus D-Fuc β-1-3 GalNAcα resulting in tight binding. RCA-I binds Fetuin triantennary asialoglycopeptide 100 % after α-2-3 and 25 % after α-2-6 sialylation, 30 % after α-1-2 and 100 % after α-1-3 fucosylation, and 50 % after α-1-3 galactosylation. WGA binds 3-but not 6-Fucosyl chitobiose core. Thus, information on the influence of complex carbohydrate chain constituents on lectin binding is apparently essential for the potential application of lectins in glycoconjugate research.

A systematic analysis of acceptor specificity and reaction kinetics of five human α(2,3)sialyltransferases: Product inhibition studies illustrate reaction mechanism for ST3Gal-I

Sialyltransferases (STs) catalyze the addition of sialic acids to the non-reducing ends of glycoproteins and glycolipids. In this work, we examined the acceptor specificity of five human α(2,3)sialyltransferases, namely ST3Gal -I, -II, -III, -IV and -VI. KM values for each of these enzymes is presented using radioactivity for acceptors containing Type-I (Galβ1,3GlcNAc), Type-II (Galβ1,4GlcNAc), Type-III (Galβ1,3GalNAc) and Core-2 (Galβ1,3(GlcNAcβ1,6)GalNAc) reactive groups. Several variants of acceptors inhibited ST3Gal activity emphasizing structural role of acceptor in enzyme-catalyzed reactions. In some cases, mass spectrometry was performed for structural verification. The results demonstrate human ST3Gal-I catalysis towards Type-III and Core-2 acceptors with KM = 5-50 μM and high VMax values. The KM for ST3Gal-I and ST3Gal-II was 100 and 30-fold lower, respectively, for Type-III compared to Type-I acceptors. Variants of Type-I and Type-II structures characterized ST3Gal-III, -IV and -VI for their catalytic specificity. This manuscript also estimates KM for human ST3Gal-VI using Type-I and Type-II substrates. Together, these findings built a platform for designing inhibitors of STs having therapeutic potential.

Mechanisms of the sialidase and trans-sialidase activities of bacterial sialyltransferases from glycosyltransferase family 80

Many important biological functions are mediated by complex glycan structures containing the nine-carbon sugar sialic acid (Sia) at terminal, non-reducing positions. Sia are introduced onto glycan structures by enzymes known as sialyltransferases (STs). Bacterial STs from the glycosyltransferase family GT80 are a group of well-studied enzymes used for the synthesis of sialylated glycan structures. While highly efficient at sialyl transfer, these enzymes also demonstrate sialidase and trans-sialidase activities for which there is some debate surrounding the corresponding enzymatic mechanisms. Here we propose a mechanism for STs from the glycosyltransferase family GT80 in which sialidase and trans-sialidase activities occur through reverse sialylation of CMP. The resulting CMP-Sia is then enzymatically hydrolyzed or used as a donor in subsequent ST reactions resulting in sialidase and trans-sialidase activities, respectively. We provide evidence for this mechanism by demonstrating that CMP is required for sialidase and trans-sialidase activities and that its removal with phosphatase ablates activity. We also confirm the formation of CMP-Sia using a coupled enzyme assay. A clear understanding of the sialidase and trans-sialidase mechanisms for this class of enzymes allows for more effective use of these enzymes in the synthesis of glycoconjugates.

Critical role of evolutionarily conserved glycosylation at Asn211 in the intracellular trafficking and activity of sialyltransferase ST3Gal-II

ST3Gal-II is largely responsible for ganglioside terminal α2,3-sialylation in mammals. We demonstrated that ST3Gal-II mainly distributes in proximal Golgi compartments and that the inhibition of N-glycosylation and oligosaccharide trimming is critical for its enzymatic activity and intracellular distribution.

Enhancing MS(n) mass spectrometry strategy for carbohydrate analysis: A b2 ion spectral library

Searchable mass spectral libraries for glycans may be enhanced using a B2 ion library. Using a quadrupole ion-trap mass spectrometer, successive fragmentations of sodiated oligosaccharides were carried out in the positive ion mode. In B,Y-type fragmentation, disaccharide B2 ions are generated which correspond to specific glycosidic linkages using progressive MS stages. Fragmentation of "B2 ions" corresponding to glycosidic linkages such as Hex-Fuc, Hex-Hex, Hex-HexNAc, HexNAc-Hex and HexNAc-HexNAc, were systematically studied in low energy CID and collected to form a "B2 library". Linkages produce characteristic fragmentation patterns in the absence of cross-ring fragmentation. Patterns of "B2 ions" rely on relative stability of glycosidic bonds and carbohydrate-metal complexes in the gas phase. MS(n) studies of linear, branched trisaccharides and tetrasaccharides show that isomers for which B2 ion information is not available are rarely a problem in practice by their absence in an isomeric sequence or by their scarcity in nature. This MS strategy for linkage determination of carbohydrates aided by a "B2 library" was developed with a scope for expansion, providing an improved tool for glycomics. We validated this method examining levels of expressed activities of two glycosyl transferases in cancer cell lines: β3(B3GALNT2) and β4GalNAcT(B4GALNT3&4) that generate GalNAcβ3GlcNAcβ and GalNAcβ4GlcNAcβ.

BIOLOGICAL SIGNIFICANCE

Glycosylation is an important class of the "postranslationome", which includes manifold aspects of post-translational protein modification, affecting protein conformation, providing ligands for protein receptors [1-5], and encoding unique haptenic [6,7] or antigenic markers for oncology [8-11] and other applications. Identification of individual monomeric units, linkages, ring size, branching and anomerity has posed significant challenges to mass spectrometrists. MS(n) is a growing key instrumental method to differentiate among isomers [12]. While the potential isomers in oligosaccharides are impossibly large [12], likely possibilities can be limited by the biological system, including the expressed glycosyl transferases [13-20]. Mass spectra from sequential stages of collision activation (MS(n)) can supply structural details for precise characterization of linkage, monomer ID, substitutions, anomerity and branching [21-25]. There is a fundamental need for high throughput tools in glycomics to complement proteome studies. In that regard, nothing could be more important than searchable spectral library files for structural confirmation. The National Academy of Science (NAS) report (http://glyco.nas.edu) recommends the need of more than 10,000 synthetic structures of carbohydrates to advance the field of glycomics. This study demonstrates that the general reproducibility of ion trap spectra, and energy independence from modes of ionization and collisional activation, make compiling an MS(n) library for carbohydrate identification an achievable research target [26]. We intend to use the new B2 library for carbohydrate differences found on cancers, where we profile the glycosyltransferases to predict classes of potential structures, and use the library for MS identification of the expected cohort of altered structures.

Scaling down the size and increasing the throughput of glycosyltransferase assays: activity changes on stem cell differentiation

Glycosyltransferases (glycoTs) catalyze the transfer of monosaccharides from nucleotide-sugars to carbohydrate-, lipid-, and protein-based acceptors. We examined strategies to scale down and increase the throughput of glycoT enzymatic assays because traditional methods require large reaction volumes and complex chromatography. Approaches tested used (i) microarray pin printing, an appropriate method when glycoT activity was high; (ii) microwells and microcentrifuge tubes, a suitable method for studies with cell lysates when enzyme activity was moderate; and (iii) C(18) pipette tips and solvent extraction, a method that enriched reaction product when the extent of reaction was low. In all cases, reverse-phase thin layer chromatography (RP-TLC) coupled with phosphorimaging quantified the reaction rate. Studies with mouse embryonic stem cells (mESCs) demonstrated an increase in overall β(1,3)galactosyltransferase and α(2,3)sialyltransferase activity and a decrease in α(1,3)fucosyltransferases when these cells differentiate toward cardiomyocytes. Enzymatic and lectin binding data suggest a transition from Lewis(x)-type structures in mESCs to sialylated Galβ1,3GalNAc-type glycans on differentiation, with more prominent changes in enzyme activity occurring at later stages when embryoid bodies differentiated toward cardiomyocytes. Overall, simple, rapid, quantitative, and scalable glycoT activity analysis methods are presented. These use a range of natural and synthetic acceptors for the analysis of complex biological specimens that have limited availability.

Characterization of cancer associated mucin type O-glycans using the exchange sialylation properties of mammalian sialyltransferase ST3Gal-II

Our previous studies suggest that the α2,3sialylated T-antigen (NeuAcα2,3Galβ1,3GalNac-) and associated glycan structures are likely to be elevated during cancer. An easy and reliable strategy to label mucinous glycans that contain such carbohydrates can enable the identification of novel glycoproteins that are cancer associated. To this end, the present study demonstrates that the exchange sialylation property of mammalian ST3Gal-II can facilitate the labeling of mucin glycoproteins in cancer cells, tumor specimens, and glycoproteins in cancer sera. Results show that (i) the radiolabeled mucin glycoproteins of each of the cancer cell lines studied (T47D, MCF7, LS180, LNCaP, SKOV3, HL60, DU4475, and HepG2) is distinct either in terms of the specific glycans presented or their relative distribution. While some cell lines like T47D had only one single sialylated O-glycan, others like LS180 and DU4475 contained a complex mixture of mucinous carbohydrates. (ii) [14C]sialyl labeling of primary tumor cells identified a 25-35 kDa mucin glycoprotein unique to pancreatic tumor. Labeled glycoproteins for other cancers had higher molecular weight. (iii) Studies of [14C] sialylated human sera showed larger mucin glycopeptides and >2-fold larger mucin-type chains in human serum compared to [14C]sialyl labeled glycans of fetuin. Overall, the exchange sialylation property of ST3Gal-II provides an efficient avenue to identify mucinous proteins for applications in glycoproteomics and cancer research.

Understanding glycomechanics using mathematical modeling: a review of current approaches to simulate cellular glycosylation reaction networks

Following the footsteps of genomics and proteomics, recent years have witnessed the growth of large-scale experimental methods in the field of glycomics. In parallel, there has also been growing interest in developing Systems Biology based methods to study the glycome. The combined goals of these endeavors is to identify glycosylation-dependent mechanisms regulating human physiology, check points that can control the progression of pathophysiology, and modifications to reaction pathways that can result in more uniform biopharmaceutical processes. In these efforts, mathematical models of N- and O-linked glycosylation have emerged as paradigms for the field. While these are relatively few in number, nevertheless, the existing models provide a basic framework that can be used to develop more sophisticated analysis strategies for glycosylation in the future. The current review surveys these computational models with focus on the underlying mathematics and assumptions, and with respect to their ability to generate experimentally testable hypotheses.

Mammalian sialyltransferase ST3Gal-II: its exchange sialylation catalytic properties allow labeling of sialyl residues in mucin-type sialylated glycoproteins and specific gangliosides

While glycosyltransferases are known to display unidirectional enzymatic activity, recent studies suggest that some can also catalyze readily reversible reactions. Recently, we found that mammalian sialyltransferase ST3Gal-II can catalyze the formation of CMP-NeuAc from 5'-CMP in the presence of a donor containing the NeuAcα2,3Galβ1,3GalNAc unit [Chandrasekaran, E. V., et al. (2008) Biochemistry 47, 320-330]. This study shows by using [9-(3)H]- or [(14)C]sialyl mucin core 2 compounds that ST3Gal-II exchanges sialyl residues between CMP-NeuAc and the NeuAcα2,3Galβ1,3GalNAc unit and also radiolabels sialyl residues in gangliosides GD1a and GT1b, but not GM1. Exchange sialylation proceeds with relative ease, which is evident from the following. (a) Radiolabeleling of fetuin was ~2-fold stronger than that of asialo fetuin when CMP- [9-(3)H]NeuAc was generated in situ from 5'-CMP and [9-(3)H]NeuAcα2,3Galβ1,3GalNAcβ1,3Galα-O-Me by ST3Gal-II. (b) ST3Gal-II exchanged radiolabels between [(14)C]sialyl fetuin and [9-(3)H]NeuAcα2,3Galβ1,3GalNAcβ1,3Galα-O-Me by generating CMP-[(14)C]- and -[9-(3)H]NeuAc through 5'-CMP; only 20.3% (14)C and 28.0% (3)H remained with the parent compounds after the sialyl exchange. The [9-(3)H]sialyl-tagged MN glycophorin A, human chorionic gonadotropin β subunit, GlyCAM-1, CD43, fetuin, porcine Cowper's gland mucin, bovine casein macroglycopeptide, human placental glycoproteins, and haptoglobin were analyzed by using Pronase digestion, mild alkaline borohydride treatment, Biogel P6, lectin agarose, and silica gel thin layer chromatography. Sulfated and sialylated O-glycans were found in GlyCAM-1 and human placental glycoproteins. This technique has the potential to serve as an important tool as it provides a natural tag for the chemical and functional characterization of O-glycan-bearing glycoproteins.

Trans-sialidase activity of Photobacterium damsela alpha2,6-sialyltransferase and its application in the synthesis of sialosides

Trans-sialidases catalyze the transfer of a sialic acid from one sialoside to an acceptor to form a new sialoside. alpha2,3-Trans-sialidase activity was initially discovered in the parasitic protozoan Trypanosoma cruzi, and more recently was found in a multifunctional Pasteurella multocida sialyltransferase PmST1. alpha2,8-Trans-sialidase activity was also described for a multifunctional Campylobacter jejuni sialyltransferase CstII. We report here the discovery of the alpha2,6-trans-sialidase activity of a previously reported recombinant truncated bacterial alpha2,6-sialyltransferase from Photobacterium damsela (Delta15Pd2,6ST). This is the first time that the alpha2,6-trans-sialidase activity has ever been identified. Kinetic studies indicate that Delta15Pd2,6ST-catalyzed trans-sialidase reaction follows a ping-pong bi-bi reaction mechanism. Cytidine 5'-monophosphate, the product of sialyltransferase reactions, is not required by the trans-sialidase activity of the enzyme but enhances the trans-sialidase activity modestly as a non-essential activator. Using chemically synthesized Neu5AcalphapNP and LacbetaMU, alpha2,6-linked sialoside Neu5Acalpha2,6LacbetaMU has been obtained in one-step in high yield using the trans-sialidase activity of Delta15Pd2,6ST. In addition to the alpha2,6-trans-sialidase activity, Delta15Pd2,6ST also has alpha2,6-sialidase activity. The multifunctionality is thus a common feature of many bacterial sialyltransferases.

De-N-acetylated sialic acid is immunogenic and elicits antibodies that are protective against Neisseria meningitidis

Recently, we showed that monoclonal antibodies (mAbs) that are reactive with derivatives of polysialic acid containing de-N-acetylated neuraminic acid (Neu) residues are protective against N. meningitidis group B strains (Moe et al. 2005, Infect Immun73: 2123; Flitter et al., in preparation). In addition, we found that fully de-N-acetylated PSA (i.e. poly alpha 2,8 Neu) conjugated to tetanus toxoid (DeNAc) elicits IgM and IgG antibodies of all subclasses in mice that bind to group B strains, activate human complement deposition, are protective in an infant rat model of meningococcal bacteremia and are bactericidal against group C strains (Moe et al, in press). We show here that anti-DeNAc mAbs, DA1 and DA2 (both IgM), are reactive with polysaccharides containing Neu, bind to group B, C, W135 and Y but not X strains grown in chemically defined media (CDM). However, when the group X strain is grown in CDM supplemented with human plasma, DA2 binds. Also both mAbs mediate bactericidal activity against B, C, W135, and X strains with human complement. The results suggests that N. meningitidis express and/or acquire zwitterionic de-N-acetyl sialic acid antigens that can be the target of protective antibodies.

Bidirectional N-acetylglucosamine transfer mediated by beta-1,4-N-acetylglucosaminyltransferase III

beta-1,4-N-Acetylglucosaminyltransferase III (GnT-III) catalyzes the formation of the bisecting GlcNAc and plays a regulatory role in the biosynthesis of the N-linked oligosaccharide. In this study, we examined whether the glycosyl transfer catalyzed by GnT-III is reversible, and, in addition, investigated the equilibrium of the GnT-III-catalyzed reaction. Incubation of the agalactosyl-bisected biantennary oligosaccharide with GnT-III in the presence of the sufficiently high concentration of uridine diphosphate (UDP) resulted in conversion of the bisected oligosaccharide into the nonbisected one. This reaction was accompanied by the stoichiometric formation of UDP-GlcNAc, which appeared to result from the transfer of GlcNAc from the oligosaccharide to UDP. Thus, these results indicate that GnT-III is capable of perceivably catalyzing the reverse reaction in vitro, as found in some glycosyltransferases. When the equilibrium of the reaction was kinetically analyzed, it was found that the state of the equilibrium is greatly displaced toward the formation of the bisecting GlcNAc. In terms of free energy change, as estimated, the reaction by GnT-III can be comparable to the hydrolysis of ATP. Although GnT-III catalyzes bidirectional transfer of GlcNAc between the oligosaccharide and UDP, the removal of the bisecting GlcNAc is unlikely in vivo, due to the displacement of the equilibrium. It is known that equilibria of certain glycosyltransferase reactions are not biased as greatly as the case of GnT-III, and thus it seems likely that there are a variety of equilibrium states in glycosyltransferase reactions. In living cells, the assembly of oligosaccharides could be regulated by not only rate control but also equilibrium control.